摘要:

Russian Chemical Reviews 68 (10) 801 ± 820 (1999) Ternary and quaternary chalcogenides of Group IB elements V A Starodub Contents I. Introduction II. Molecular copper, silver and gold polychalcogenides III. Coordination polymers based on polychalcogenides of Group IB elements IV. Chalcogenides containing only X2¡ and/or X2¡ 2 ions V. Conclusion Abstract. Methods of synthesis, molecular and electronic struc- tures and electrophysical, optical and magnetic properties of ternary and quaternary chalcogenides based on copper, silver and gold are considered. The possibility of the synthesis of chalcogenide analogues of high-temperature superconductors is discussed. The prospects of searching for new high-temperature superconductors in Ln7Cu7X systems, where Ln is a rare-earth element and X is S, Se or Te, are shown.The bibliography includes 148 references. I. Introduction The discovery of the high-temperature superconductivity in 1986 1 has given powerful impetus to the development of solid-state chemistry. This has awakened great interest in metal oxide systems with the perovskite-like structures. High-temperature supercon- ductors (HTSCs) with the superconducting transition temper- atures (Tc) of up to 150 K (see Ref. 2) were synthesised (the data, while unconfirmed, on the synthesis of HTSCs with Tc of up to 250 K are also available in the literature 3, 4). It has since become clear that the properties of new superconducting materials cannot be explained within the framework of the BCS theory (Bardeen ± Cooper ±Schrieffer).Apparently, the RVB (resonance valence bond) model proposed by Anderson 5 is most realistic.5 ±9 This model assumes the occurrence of mixed-valence states in the system of chemically bonded atoms as an essential prerequisite for the high-temperature superconductivity. The mixed valence is necessary not only for the provision of the required carrier concentration and the equivalent positions for the carrier motion but also provides the basis for the coupling mechanism. In this connection, it becomes apparent that solid-state chemistry is of paramount importance in the solution of both theoretical and applied problems of physics of superconductivity. In spite of the ten-year history of HTSCs, much in the chemistry of these materials remains unclear.Suffice it to mention that the problem associated with the estimation of the degree of oxidation of copper in metal-oxide samples, which is of importance for constructing the theory of HTSCs, is still debated. Leaving aside the details of this problem, let us mention a study,10 V A Starodub Department of Chemistry, Kharkov State University, pl. Svobody 4, 310077 Kharkov, Ukraine. Fax (38-057) 247 18 16. Tel. (38-057) 245 73 52 Received 19 January 1999 Uspekhi Khimii 67 (10) 883 ± 903 (1999); translated by T N Safonova #1999 Russian Academy of Sciences and Turpion Ltd UDC 548.3 : 549.331 801 802 805 807 818 in which the authors used X-ray emission spectroscopy for estimating the degree of oxidation of copper atoms in HTSCs.This method is free from the drawbacks of X-ray photoelectron spectroscopy (XPS), which is most often used for these purposes. The role of oxygen atoms in HTSCs also remains unclear. In the available chemical models of superconductivity of HTSCs, which are based on Anderson's ideas,11 ± 13 the possibility of formation of weakly bound O4 complexes upon transformation of HTSCs into the superconducting state is considered. The bond strength in these complexes is determined, on the one hand, by the sizes of the cations and, on the other hand, by the concentration of holes, which disturb the antiferromagnetic order in the initial dielectric system. Taking into account the greater ability of sulfur and selenium atoms to form polychalcogenide ions, chalcogenide analogues of HTSCs would be expected to possess stronger electrophysical characteristics compared to those of oxide HTSCs.This idea was proposed, for example, in the study;14 simultaneously, it was assumed that the structural characteristics remain unchanged. The prospects for synthesising HTSCs based on thio-, seleno- and tellurocuprates are very attractive because chalcogenides of transition metals, in particular, copper chalco- genides, provide substantially greater possibilities for the prepa- ration of compounds possessing various physicochemical properties compared to those of the corresponding oxides. The most essential differences consist in the higher covalence of the metal ± chalcogen bonds, the ability of chalcogen atoms to form bonds with each other resulting in molecular anions and the higher ability of metal atoms to form metal ± metal bonds (due to the lower formal charge).Owing to these facts, chalcogenides of Group IB metals can exhibit a broad spectrum of electrical (from dielectric to superconducting 15) and magnetic properties (from diamagnetism to temperature-independent paramagnetism). Table 1 gives the structural and electrophysical characteristics of selected superconductors based on chalcogenides of transition metals, viz., chalcogenides of copper and rare-earth elements, which are structural elements of HTSCs, as well on some ternary chalcogenides characterised by high Tc values.Analysis of these data suggests that a search for HTSCs in M7Ln7X systems, where M is IB Group metal, Ln is a rare-earth element and X is chalcogen, may have promise. Thus, in the studies of the Cu2S7La2S3 system performed in 1968 ± 1972,19 ± 21 ternary sulfides, viz., the hexagonal Cu2La4S7 and monoclinic CuLaS2 phases, were detected. However, Andreev 22 studied the same system in 1988 and found only the CuLaS2 phase. Later on, Andreev published the paper entitled `The Cu2S ± BaS system as a possible HTSC', 23 where he demonstrated that two phases, viz., the tetragonal BaCu2S2 and orthorhombic BaCu4S3 phases, exist802 Table 1. Structural and electrophysical characteristics of selected chalco- genide superconductors.16 ¡À 18 Composition Compound Tc /K Lattice type Density of states /cm73 hexagonal 1.65 CuS Cu�¢ *1020 .(S2¡¦ Cu2+(S2¡¦ 4 Cu22�¢ . 2 )2(S2¡¦)2 2 ) CuS2 1.48 ¡À 1.53 *1021 Cu2+(Se2¡¦ 2 ) 2 ) CuSe2 CuTe2 YS 2.30 ¡À 2.43 *1021 1.25 ¡À 1.3 1.9 Y2+/Y3+ *1021 *1022 Y2+/Y3+ 2+2+ 2+ 2+ pyrite, cubic the same " Cu2+(Te2¡¦ NaCl, cubic the same " Y/Y3+ " La/La3+ " La/La3+ " La/La3+ cubic 2.5 2.05 0.87 1.02 1.48 1 YSe YTe LaS LaSe LaTe La2Se3 La2+/La3+ La2+/La3+ 2+ 8 ¡À 11.5 7.63 ¡À 9.4 4.61 8.9 La3S4 La3Se4 La3Te4 AgMo4S5 the same " La/La3+ ¡À¡À¡À *1022 *1022 *1022 *1022 *1022 semicon- ductor 661021 1021 ¡À1022 ¡À¡À Cu1 ¡À 1.9Mo3S4 10.9 ¡À 10.5 ¡À 10 ¡À 13 ¡À¡À¡Àhexagonal Ti3+/Ti4+ ¡À LixTi1.1S2 (0.1< x<0.3) in this system.In 1994, Chinese researchers 24 at the National Laboratory of Superconductivity synthesised a series of oxy- selenides LnCuSeO (Ln=La, Sm, Gd or Y). The structures of the latter contain the alternating [Ln2O2] and [Cu2Se2] layers. Analogous layers have been found previously by Popovkin 25 in oxyselenide BiCuSeO. Attempts were undertaken to perform partial replacement of oxygen atoms in HTSCs by halogen or chalcogen atoms. These attempts often cast doubt from the chemical standpoint 14, 26, 27 and do not solve the problem because they allow only an insignificant amount of oxygen to be replaced by chalcogen. The effect of the latter in these systems can be considered only as a weak perturbation, which was in essence found in the above-mentioned studies.As early as 1991, we attempted to synthesise chalcogenide analogues of HTSCs, in particular, La2CuS4+d and YBa2Cu3S6+d.28 However, because of difficulties associated with the preparation of a single phase and the lack of data on simpler systems (for example, Ba7Cu7S and Cu7Ln7S), we turned our attention primarto ternary systems consisting of chalcoge- nides of Group IB metals and chalcogenides of alkali, alkaline- earth or rare-earth metals. Nevertheless, even regardless of the problem of HTSCs, the success achieved in the chemistry of ternary and quaternary chalcogenides based on Group IB elements within the last decade is so impressive that it makes sense to expose not only chemists and physicist involved in studies of HTSCs but also researchers working in other fields to the results obtained.The present review is devoted to analysis of methods of synthesis and molecular and electronic structures of chalcogenides in relation to their magnetic and electrophysical properties. 2 Molecular polychalcogenides are considered first because it is in this class of compounds that sulfides and selenides differ most sharply from the corresponding oxides. We next consider the coordination polymers based on polychalcogenides of Group IB metals. Finally chalcogenides containing only X2¡¦ and/or X2¡¦ ions, i.e., the structural elements identical to those in oxide HTSCs, are considered.Tellurides are considered at the end of each chapter for two reasons. First, tellurium has peculiarities and the properties of tellurides differ from those of the corresponding V A Starodub sulfides and selenides and second, these compounds are poorly known. II. Molecular copper, silver and gold polychalcogenides 6 8 Molecular polychalcogenides of Group IB metals containing isolated [MnXm]q7 clusters (M=Cu, Ag or Au; X=S, Se or Te) were virtually unknown up to the early 1980s.29 (The synthesis of polynuclear copper sulfide complexes was reported for the first time in 1983.) The (Et4N)3[Cu3(S6)3] complex was prepared by the reaction of copper(II) acetate with an ethanolic solution of polysulfides.30 The complex contains the complex [Cu3(S6)3]37 ion in which the Cu+ ions are chelated by the hexasulfide ions (Fig.1). One of the terminal sulfur atoms of each S2¡¦ ion acts as a bridge and is involved in formation of the central six-membered Cu3S3 ring. In the following year,31 researchers succeeded in isolating salts, which contain [Cu3(S4)3]37 ions and are structurally similar to [Cu3(S6)3]37. In this study, the (Ph4P)4[Cu2S20] salt containing the [S6CuS8CuS6]47 ions in which two CuS6 rings are bound to each other through the bridging S2¡¦ anion was also described. Analo- gous silver salts containing the [Ag2S20]47 ions were also obtained.32 Note that in all these salts, the copper and silver atoms adopt the coordination number of 3, which is rarely observed in the chemistry of coordination compounds.Cu S Figure 1. Structure of the [Cu3(S6)3]37 anion.29 Mu�� ller and coworkers synthesised a series of four-nuclear clusters of the general formula [Cu4(S4)37x(S5)x]27. These clusters contain the tetrahedral Cu4 core in which the copper atoms are chelated by the bridging S2¡¦ 4 and/or S25 ¡¦ ions. All members of this series were isolated, viz., salts with x=0,31 1, 2 33 and 3.34 The structure of the [Cu4S4(S5)2]27 cluster is shown in Fig. 2. It should be noted that these clusters provide rare examples of coexistence of the MS4 and MS5 chelates. Cu S Figure 2. Structure of the [Cu4S4(S5)2]27 anion.33 In 1984, the (Ph4P)2[Cu6(S4)3S5] salt was synthesised virtually simultaneously at the laboratories headed by Henkel 35 and Mu�� ller.36 This salt can be prepared either by the reaction of CuCl2 .2H2O with a threefold excess of NaSR (R=Et or CH2Ph) in methanol 35 or by the reaction of Cu(acac)2 with an ethanolic solution of polysulfides.36 In both cases, the salt was isolated as orange-red crystals upon addition of a solution of Ph4PBr to theTernary and quaternary chalcogenides of Group IB elements 1 20 2 30 Cu S 3 30 6 40 3 4 5 50 20 2 60 10 7 1 70 80 890 9 Figure 3. Structure of the [Cu6(S4)3S5]27 anion.36 reaction mixture. In this complex, a new coordination mode was found (Fig. 3). Six copper atoms form two distorted tetrahedra, whose vertices are occupied by the Cu(1), Cu(2), Cu(3), Cu(10), Cu(20) and Cu(30) atoms.These tetrahedra are linked via the shared Cu(3)7Cu(30) edge. The metal atom is coordinated by the tetrasulfide ions only through the terminal atoms, while the pentasulfide ions are coordinated to the metal atoms through four sulfur atoms [S(2), S(20), S(3) and S(30)]. In the laboratory headed by Henkel,37 the (Ph4P)4[Cu12S8] salt containing the [Cu12S8]47 cluster ions was synthesised. This salt was prepared as wine-red needles by the reaction of CuCl2 . 2H2O with a threefold excess of NaSEt in methanol (*778 8C) followed by addition of Li2S and (after completion of the reaction) of a solution of Ph4PBr. The [Cu12S8]47 anion has the ideal Oh symmetry. The sulfur atoms occupy the vertices of the cube and the copper atoms occupy the centres of the edges.Later on, Marbach and Stra�� hle 38 isolated an analogous gold complex as the tetraphenylarsonium (Ph4As)4[Au12S8] salt. The [Au12S8]47 and [Cu12S8]47 anions have identical structures. In 1992, Kanatzidis and Huang 39 prepared the (Et4N)3[NaAu12Se8] complex, which is the selenium analogue of the (Ph4As)4[Au12S8] salt. The larger size of the selenium atom compared to that of the sulfur atom allows Na+ ions to occupy the centres of the [Au12Se8]47 cuboctahedra. Hence, this complex is one of a few examples of inorganic cryptands. The Na+ ion occupies the inversion centre. The distances from the sodium ion to the gold ions are shorter than those to the selenium atoms. Apparently, the interactions between the sodium ion and the twelve Au+ ions are stronger.The authors 39 believed that this fact reflects the general tendency of the Au+ ions to interact with alkali-metal ions. Note that the Au+ or Cu+ ions in the [M12X8]47 complexes (M=Au or Cu) have a typical linear coordination. An example of a supposedly extremely large copper chalcoge- nide cluster was described in the study.40 The authors succeeded not only in isolating the [Cu146Se73(Ph3P)30] cluster but also in characterising it structurally. Although this compound is not a purely chalcogenide cluster because it contains the triphenylphos- phine group, it corresponds to the subject of this review because this compound is formally a derivative of copper selenide Cu2Se7{[Cu2Se]73 .30(Ph3P)}. This cluster is also of interest because it is a semiconductor with the conductivity sRT=1.561072 O71 cm71. The selenium atoms in this cluster form three planar layers consisting of 21, 31 and 21 atoms (ABA packing) arranged in a close packed fashion. The copper atoms occupy the cavities in this packing. In the crystal, the clusters are arranged in such a way as to form cavities (channels) with a diameter of 1.6 nm. When crystallisation is performed from organic solvents, channels are occupied by the solvent molecules, which can be, however, removed under mild conditions without destruction of the crystal lattice. For example, 42 tetrahydrofuran molecules can add per formula unit of the cluster. In the above-considered examples, the specificity of the Cu+, Ag+ and Au+ ions is not manifested.Let us consider examples of chalcogenides, in which this specificity is apparent. The Au+ ion, like the Cu+ ion, forms a complex with the tetrasulfide ion with 803 Au S Figure 4. Structure of the [Au2(S4)2]27 anion.34 composition [Au2(S4)2]27.34 Unlike analogous copper complexes, the gold(I) tetrasulfide complex can be described as a ten- membered ring, in which the Au+ ion has a nearly linear coordination (Fig. 4). The gold(I) pentaselenide complex was isolated as the 4 5 [(Ph2P)2NPh2]2[Au2Se10] salt. However, according to the data of X-ray diffraction analysis,41 this complex is in fact an Au3+ derivative and contains coordinated selenide and tetraselenide ions.It was suggested that the Au+ complex with the Se2¡¦ 5 ions in which the Au+ion has linear coordination, was actually formed at the initial stage. However, the subsequent intramolecular two- electron transfer from the Au+ ion to the terminal Se7Se bond afforded the Au3+ and Se27 ions and the Se2¡¦ ions were formed from the Se2¡&. The complex [Se4AuSe2AuSe4]27 ion is planar, like the majority of Au(III) complexes. Se 2¡¦ Se Se 2¡¦ Se Se Se Se Se Se Se Au Au Au Se Au Se Se Se Se Se Se Au3+ complex Se Se Se Au+ complex Oxidation of Au(I) with the pentaselenide ion proceeded unusually. The reaction of AuCl3 with Na2Se5 upon addition of a Ph4PBr solution to the reaction mixture afforded gold and the product of oxidation of the pentaselenide ion, viz., (Ph4P)2Se11, i.e., Au3+ was reduced with the pentaselenide ion.Apparently, the only analogous reaction is the reaction in the Tl+7I¡¦3 system in which Tl3+ ions were completely reduced with I7 ions to Tl+. However, Tl+ ions reacted with an excess of I¡¦3 ions to form the Tl3+ complex with composition [TlI4]7.41 n Polyselenide Au(I) complexes were prepared with the use of polyselenides Se2¡¦ with n=1 ¡À 4.42 Thus, the reaction of AuCN with a Na2Se2 or Na2Se3 solution in the presence of Ph4PCl yielded the (Ph4P)2[Au2Se2Se3] salt containing the seven-mem- bered cyclic [Au2Se2Se3]27 anions (Fig. 5). The complex anion adopts an envelope conformation with the linear coordination of the Au+ ions.The reaction performed under analogous condi- tions with the use of K2Se4 instead of Na2Se2 or Na2Se3 gave the (Ph4P)2[Au2Se2Se4] salt. This salt contains the eight-membered cyclic anion in which two Au+ions are symmetrically coordinated by the diselenide and tetraselenide ions. Au Se Figure 5. Structure of the [Au2Se2Se3]27 anion.42 Finally, the reaction with the use of pentaselenide K2Se5 gave the K2[Au2Se2(Se4)2] complex. Unlike the two above-described complexes, the latter complex contains the Au3+ ion as a result of the higher oxidising capacity of the Se2¡¦ 5 ion compared to those of the Se2¡¦ 4 and Se23 ¡¦ ions.43804 2Se2¡¦ 4 +Se22 ¡¦, 2Se2¡¦ 5 +2e Au3+. Au+72e In addition to the above-mentioned [Ag2S20]47 anion, the Ag+ complex with hexasulfide ions with composition [Ag2(S6)2]27 is known. In this complex, the coordination of the S26 ¡¦ ligands is analogous to that observed in the [Cu3(S6)3]37 anion, with the only difference that the binuclear [Ag2(S6)2]27 anion contains the four-membered Ag2S2 rings.44 4 4 The coordination of the tetraselenide anions by the Ag+ cations is also very similar to that observed in Cu+ salts.Thus, four-nuclear silver tetraselenides (Et4N)4[Ag4(Se4)4] and (Pr4N)2[Ag4(Se4)3] were isolated.45 The [Ag4(Se4)4]47 anion is formed by four Ag+ ions, which occupy the vertices of the planar rhombus and which are linked via the bridging Se2¡¦ ligands (Fig. 6). The silver atoms in this cluster are nonequivalent. The coordination environment about two of them [Ag(1) and Ag(10), see.Fig. 6] is a tetrahedron, while two other atoms [Ag(2) and Ag(20)] have a planar-trigonal coordination. Analogously, the Se2¡¦ ions are also nonequivalent. Two of them link three silver ions each and the remaining two complex ions link two silver ions each. 3 2 7 1 6 4 Ag Se 1 8 5 2 2 0 8 0 5 0 1 0 4 0 6 0 1 0 2 0 7 0 3 0 Figure 6. Structure of the [Ag4(Se4)4]47 cluster.45 4 The structure of the cluster [Ag4(Se4)3]27 ion (Fig. 7) is based on the tetrahedral cage formed by the Ag+ ions, which are linked via three Se2¡¦ ligands. This cage, in turn, forms the distorted adamantane-like central Ag4Se6 core in which all silver atoms adopt a planar-trigonal coordination. Ag Se Figure 7.Structure of the [Ag4(Se4)3]27 cluster.45 The complex copper and silver selenides, viz., [Cu2Se4(Se5)2]47, [Cu4(Se4)2.4(Se5)0.6]27 and (Ph4P)2. .[Ag4(Se4)37x(Se5)x] , which are related to the above-described copper sulfides [Cu4(S4)37x(S5)x]27, were also described.46, 47 These salts were prepared by the reaction of CuCl with a melt of polyselenide Li2Sen or by fusing a mixture of CuCl2, sodium and selenium. Their structures are similar to those of the correspond- ing sulfides. The mononuclear complexes with the nonasulfide ion are known only for gold and silver. Marbach and Stra�� hle isolated cyclic nonasulfidoaurate(I) (Ph4As)[AuS9].48 Its structure is shown in Fig. 8. The Au+ ion binds the terminal sulfur atoms of the ligand and has a nearly linear coordination [the S(1)AuS(9) angle is 176 8].The analogous silver salt (Ph4P)[AgS9] is also V A Starodub 8 4 9 3 1 5 Au S 7 2 6 Figure 8. Structure of the [AuS9]7 anion.48 known.29 However, the anions in the crystal of this salt are substantially disordered. Finally, let us consider a few examples of molecular poly- tellurides, which are presently known only for gold. In 1985, the first gold polytelluride (Ph4P)4[KAu9Te7] containing isolated cluster ions was described.49 The reaction of a methanolic solution of K2AuTe2 with a solution of Ph4PBr afforded an orange precipitate. Additional treatment of the latter with DMF and a solution of Ph4PBr in MeOH gave two products.One of them, which was isolated as triangular red platelet-like crystals, has the composition (Ph4P)4[KAu9Te7]. The structure of the [KAu9Te7]47 anion (Fig. 9) can be considered as that derived from the cubic structure of the above-described [M12S8]47 ions (M=Cu or Au) in which one vertex and three edges are removed. The ideal cubic structure is distorted due to the contraction along the threefold axis passing through the absent vertex and the centre of the ion. This contraction is responsible for the pronounced Au7Au interaction. The Au7Au distances in the triangles (see Fig. 9) vary from 0.3039 to 0.3221 nm. The average distance (0.313 nm) is close to the Au7Au distance in metallic gold (0.288 nm) and is substantially shorter than that in the hypo- thetical [Au12Te8]47 ion (0.363 nm).Like [NaAu12Se8]37, the [Au9Te7] framework incorporates the encapsulated potassium ion, the Au7K distances being shorter than the Au7Te dis- tances, which reflects the higher electronegativity of gold com- pared to that of tellurium (2.4 and 2.1, respectively). It is believed that the above-described encapsulation is responsible for the formation of a cryptand with this composition because the K+ ion is too large to be encapsulated into the hypothetical [Au12Te8]47 ion. KAu Te Figure 9. Structure of the [KAu9Te7]47 anion.49 The electrochemical preparation of the (Ph4P)2[Au2Te4] and (Bun4 N)3[Au3Te4] salts with the use of AuTe2 as the electrode was reported in a study.50 The [Au3Te4]37 anion is gold(I) tellurodi- telluride (Fig.10 a) and the [Au2Te4]27 anion is gold(I) bistellur- ide (Fig. 10 b). In both cases, the gold atoms have a linear coordination typical of Au+ ions. However, the tendency for the trigonal coordination through the Au7Au bonds is also pro- nounced. The lengths of these bonds (see Fig. 10 a, b) are only slightly larger than the Au7Au distances in metallic gold. Treatment of gold tellurides KAuTe and K3AuTe2 with ethylenediamine or DMF afforded salts containing the [Au2(Te2)2]27 and [KAu9Te7]47 anions and the [K2Au4Te4]27 anion in which the [Au4Te4]47 rhombi are linked in infinite stacks through the K+ ions.49, 51Ternary and quaternary chalcogenides of Group IB elements a b2.960 nm Au Te 3.049 nm Figure 10.Structures of the [Au3Te4]37 (a) and [Au2Te4]27 (b) anions.50 The reaction of AuCN with polytellurides Te2¡¦ n inDMFin the presence of PEt3 and NEt4Cl yielded the (NEt4)3[AuTe7] salt.52 The [AuTe7]37 anion is planar, which is typical of Au(III) complexes (Fig. 11 a). The fact that the degree of oxidation of gold is equal to +3 is indirectly supported by the Au7Te bond lengths (0.2638 ¡À 0.2664 nm), which are comparable with those in the known Au(III) compounds, for example, in AuTeI (0.2642 ¡À 0.2684 nm), and is also confirmed by the data of Mo�� ssbauer spectroscopy on 197Au nuclei. Hence, the anion under consider- ation is bisditelluridotritelluridoaurate(III). a b4 3 5 2 6 Au Te Ag Te 7 1 Figure 11.Structures of the [AuTe7]37 (see Ref. 52) (a) and [AgTe7]37 (b) anions.54 7 The structure of the analogous silver complex [AgTe7]37 isolated as the (Ph4P)2(Et4N)[AgTe7] salt (Fig. 11 b) 53, 54 differhat of the [AuTe7]37 anion. In this silver complex, the Ag+ ion adopts a trigonal coordination. Based on the charge of the [AgTe7]37 anion, the charge of 74 should be assigned to the ligand and the ligand can be considered as Z3-Te4¡¦ 7 . The Te(4)7Te(5) and Te(4)7Te(3) bond lengths (0.2866 and 0.3230 nm, respectively) are substantially different, whereas this difference in the corresponding bonds in the analogous [HgTe7]27 complex is insignificant (0.3050 and 0.2997 nm, respectively 53). Therefore, the nature of the bonds in the new polytelluride Z3-Te4¡¦ ligand is still unknown.Polyhalide ions, in particular, such as I2¡¦ 8 , are the closest analogues of this ligand. The concept of hypervalence is used for describing the nature of bonds in these compounds, in particular, in S2N2, S4N4, polyhalides, xenon halides, etc.55 In accordance with this concept, the Te(3)7Te(4)7Te(5) bond can be considered as a three-centre four-electron bond, whereas the Te(2)7Te(3) and Te(5)7Te(6) bonds are usual two-centre two-electron bonds. Copper and silver tetratellurides containing the [Te4M(m- Te4)MTe4]47 ions (M=Cu or Ag) were also prepared.56 The structures of these ions are similar to that of the [S6Cu(m- S8)CuS6]47 ion. In these tellurides, the copper(I) and silver(I) atoms are trigonally coordinated through the chelating and bridging Te2¡¦ 4 ions.III. Coordination polymers based on polychalcogenides of Group IB elements 2 Coordination polymeric polychalcogenides are intermediate between low-molecular-weight polychalcogenides and chalco- genides containing only X2¡¦ and/or X2¡¦ ions. These ions are bound to metal ions through substantially covalent bonds due to which these compounds are virtually insoluble in the usual organic solvents. 805 Polymeric polychalcogenides containing the [CuS4]7 anion have been known since the early 20th century.57 However, their structures were established only late in the 1980s. Generally, polymeric polychalcogenides were prepared from solutions. However, the procedure for the synthesis of these compounds from melts of alkali metal polychalcogenides, which act simulta- neously as the solvent and the reagent, was proposed in 1989.58 This method appears to be more promising and allows one to prepare more perfect single crystals suitable for X-ray diffraction study. Coordination polymeric polychalcogenides can be synthes- ised in the temperature range of 150 ¡À 350 8C.At lower temper- atures, molecular polychalcogenides are obtained. At higher temperatures, mono- and dichalcogenides of Group IB metals are formed. Oxidation of copper in a melt of K2S4 at 215 and 250 8C affords two modifications (a and b, respectively) of the K[CuS4] salt. These modifications are insoluble in usual organic solvents and stable to hydrolysis and oxidation with atmospheric oxygen.Both phases have quasi-one-dimensional structures (Fig. 12). a Cu S b Cu S Figure 12. Infinite a-[Cu(S4)]n¡¦ n (a) and b-[Cu(S4)]nn ¡¦ (b) chains.58 4 The a-K[CuS4] salt is isostructural to the (NH4)[CuS4] salt described previously 59 and contains the tetrasulfide ligands chelating the copper(I) ions (Fig. 12 a). The [CuS4]7 ions form infinite noncentrosymmetrical chains extended along the a axis. These chains can be considered as a sequence of fused five- membered CuS4 and Cu2S3 rings. Therefore, the S2¡¦ ion in this structure plays the unprecedented role of a bridge that links three Cu+ ions. n 4 The structure of b-K[CuS4] contains centrosymmetrical one- dimensional b-[CuS4] n¡¦ chains with a nearly linear (the CuCuCu angle is 173.5 8) arrangement of the copper atoms linked through the bridging S2¡¦ ligands (Fig.12 b). Unlike the a-phase, the b-[CuS4] n¡¦ n chains consist of the five-membered (CuS4) and four- membered (Cu2S2) rings. Since the b-modification is formed at higher temperature, it can be suggested that this phase is thermodynamically more stable. Heating of crystals of a-K[CuS4] above 250 8C results in transition to the b-phase. However, cooling of crystals of the b-phase do not lead to the reverse transition. TheK2S4 solvent is of considerable importance in the a?b-transition. Heating of crystals of a-K[CuS4] in the absence of the solvent in vacuo resulted in their decomposition to CuS and K2Sx.58 Treatment of an alkaline solution of the (NH4)[CuS4] salt with solutions of salts containing bulky cations afforded low-molec- ular-weight polysulfides, for example, (NMe4)2[Cu4(S6)3] .4H2O, (Et3NBn)2[Cu4(S4)(S5)2] . 4H2O and (Ph3PBn)2[Cu6(S6)4] . . 12H2O.60 The authors failed to prepare single crystals suitable for X-ray diffraction study and hence, the formulas were proposed by analogy with the known [Cu4(S4)x(S5)37x]27 and [Cu6(S4)3(S5)]27 salts.31, 35, 36 Metal cations bulkier than alkali806 metal cations were used, apparently, with the aim of achieving more complete isolation of depolymerisation products from solutions. n 5 n The [(NH3)2C2H4][CuS5]2 . 4H2O salt was synthesised as orange-red single crystals.61 In the [CuS5] n¡ anion formed as an infinite sequence of the 7Cu7S7Cu7S7 bonds, each S2¡ ion links three Cu+ ions (Fig.13). Therefore, the [CuS5]n¡ chains consist of the five-membered CuS4 and Cu2S3 rings. Analogous anionic chains were also observed in the crystals of the (NMe4)[AgS5] 62 and (NMe4)[AgSe5] salts.63 These examples demonstrate that the properties of the above- described polysulfides are relatively independent of the nature of the outer-sphere cations. Cu Sb c Figure 13. Infinite [CuS5]n¡ n chains.61 n 6 6 Treatment of copper with a melt of cesium polysulfide with composition Cs2S11 at 270 8C afforded the Cs[CuS6] salt.64 In the crystal, the [CuS6]7 anions form one-dimensional centrosymmet- rical [CuS6] n¡ chains (Fig.14) separated by Cs+ ions. The Cu+ ions are chelated by the terminal sulfur atoms of the S2¡ ions to form the seven-membered CuS6 rings. In this case, the terminal sulfur atoms of the S2¡ ion each act also as a bridge between the adjacent copper atoms resulting in the formation of the four- membered Cu2S2 rings. The Cs[CuS6] salt is one of a few examples of coordination polymers containing such long polychalcogenide ions. Cu S Figure 14. Infinite [CuS6]n¡ n chains.64 4 In the crystals of the {(Ph4P)[AgSe4]}n salt, the trigonal coordination about the silver(I) atom is formed by the Se2¡ ions. However, only five-membered rings occur in this case (Fig. 15). Only one of the terminal selenium atoms of the Se2¡ 4 ion acts as a bridge.65 Polymeric silver selenides with stack structures were prepared with the use of ethylenediamine in the supercritical state as the solvent:66, 67 350 8C K2Ag12Se7.K2Se4+3Se+12Ag Ag Se Figure 15. Infinite [Ag(Se4)]n¡ n chains.65 V A Starodub The crystals of K2Ag12Se7 contain the 18-membered Ag12Se6 rings. The seventh Se27 ion occupies the centre of this ring. In the crystal structure, the rings are packed in stacks resulting in the formation of channels. These channels are occupied by K+ ions. The sulfide with the same composition, which was prepared under analogous conditions, has quite a different structure.68 Oxidation of gold with a melt of potassium polyselenides at 250 8C afforded polymeric gold polyselenides KAuSe5 and K3AuSe13 containing the [AuSe5] n¡ n and [AuSe13] n3n¡ ions, respec- tively.69 5 n n 5 n Unlike the analogous copper and silver pentachalcogenides, the Au+ions in the [AuSe5] n¡ n anions are linearly coordinated (the SeAuSe angle is 179 8) by the bridging Se2¡ ions (Fig.16 a), the latter being coordinated only through the terminal selenium atoms. The structure of the [AuSe5] n¡ ions is analogous to that of the [AuSe] n¡ n ions with the formal replacement of the Se27 ions by the Se2¡ ions. The [AuSe5] n¡ chains are linked through Au7Au interactions. The Au7Au distance (0.295 nm) is close to that in metallic gold (0.288 nm) and is substantially shorter than the corresponding distances in KAuSe 70 and in a- and b-AuSe.69 Therefore, the presence of Au7Au contacts leads to dimerisation of polymeric [AuSe5] n¡ ions.In this case, the gold atom has the d 10 configuration and this dimerisation is one of a few examples of d 10 ± d10 interactions. Formally, a combination of two filled d-orbitals of the gold atom gives bonding and antibond- ing orbitals resulting in the absence of a bond. However, it is known that the relativistic effects are essential in the case of heavy elements, such as gold,71 and the reasoning regarding the electronic structures of silver complexes, which is based on the nonrelativistic approach, is incorrect. Pekka 71 examined the role of relativistic effects in studies of the structural molecular characteristics. If these effects are taken into account, the bond lengths are decreased, the value of this decrease being propor- tional to the square of the charge on the atomic nucleus and varying from 0.250 ± 0.300 nm for AuH to 0.450 ± 0.500 nm for Au2.Correspondingly, the force constant and the dissociation energy of the Au2 molecule increase by factors of more than four and two, respectively. The role of the relativistic effects decreases as the bond ionicity increases. Therefore, this role should be noticeable in gold chalcogenides in which the degree of covalence of the Au7Xbonds is large. Dimerisation of the [AuSe5] n¡ n ions is a prominent example of this phenomenon. The presence of these bonds between ions with closed shells was also described for other gold compounds, in particular, for the complex [Au2Se5]27 and [Au2Se6]27 anions,72 and for the sulfide Tl2Au4S3.73 Since this phenomenon is unknown in the chemistry of other elements, it was suggested 72 that the term `aurophilic attraction' to be used for the Au+7Au+ interaction.The authors believed 72 that the aurophilic attraction is a result of correlation effects, which are ignored in the Hartree ± Fock approximation. a Au Se b Au Se Figure 16. Structure of the [AuSe5]n¡ n (a) and [AuSe13]3n n¡ chains (b).69Ternary and quaternary chalcogenides of Group IB elements n n 5 3 However, the authors restricted themselves only to nonrelativistic calculations. Hence, the above-considered interpretation, in which the importance of the relativistic effects in chemistry was emphasised, is, admittedly, more reliable (see also the study 74).The [AuSe13] 3n¡ ion also has an unusual structure. This ion can be described as the catenabis(pentaselenido-m-triselenido- aurate)(III) ion [Se5Au(m-Se3)AuSe5] 3n¡. This is the only known complex containing simultaneously the monodentate Se2¡ ions and the bridging Se2¡ ions coordinated to the Au3+ ions (Fig. 16 b). 3 The reaction performed under the same conditions (ethyl- enediamine in the supercritical state) with the use of sodium polyselenides instead of potassium polyselenides afforded Na3AuSe8.75 The crystals of this complex contain polymeric chains formed by the planar-square Au3+ complexes. Two Se2¡ ions are coordinated in a monodentate fashion.Se Se Se Au Se Au Se Se Se Au Se Se Se [AuSe8]37 4 The polymeric complexes (Me4N)[MTe4] (M=Cu or Ag) containing tetratelluride ions were prepared for the first time in 1993.43 These complexes, unlike the analogous tetrasulfides and tetraselenides, have layered structures (see Fig. 17). The layers consist of the five-membered MTe4 rings. Each Te2¡ ligand links three copper atoms. An interesting feature of this structure is the presence of cavities formed by 14-membered rings (see Fig. 17). The size of the cavity is determined by the shortest distances between the opposite tellurium atoms (0.6034 and 0.6195 nm). The (Me4N)[MTe4] salts are diamagnetic semiconductors with the width of the forbidden band of 0.80 and 0.90 eV forM=Cu and Ag, respectively.6 3 3 In attempting to prepare mixed copper sulfidotelluride com- plexes from melts of Cs2SxTey, the authors 64 isolated the Cs6Cu2(S6)2(TeS3)2 complex containing the dimeric [Cu2(S6)2(TeS3)2]67 anion (Fig. 18). In this anion, two Cu+ ions are chelated by two Se2¡ ions (the seven-membered CuS6 rings) and are linked to each other through two bridging TeS2¡ ions. Only one sulfur atom of each TeS2¡ ion is involved in coordi- nation, i.e., the four-membered Cu2S2 rings are formed. The trithiotellurite ion differs sharply in this property from the isoelectronic SO2¡ 3 ion. Te Cu c b Figure 17. Structure of the [CuTe4]7 layers.43 One of the 14-membered rings is hatched. 807 Cu Te S Figure 18.Structure of the [Cu2(S6)2(TeS3)2]67 ion.64 Unfortunately, successful attempts to prepare mixed seleni- dotelluride complexes are as yet unknown although their synthe- ses are quite possible because selenium and tellurium differ in properties to a lesser degree than sulfur and tellurium. For example, mixed sulfidoselenide complexes were formed in the a-CsCu(SxSe47x) and CsCu(SxSe67x) systems.76 2 IV. Chalcogenides containing only X2¡ and/or X2¡ ions The majority of compounds of this type give crystals with essentially covalent M7X bonds. Many of these compounds possess metallic properties and exhibit superconducting proper- ties at low temperatures (Table 1). According to the RVB theory,5±9 the metallic state in these crystals can occur on condition that the metal atoms exhibit a mixed valence.It is well known that the simplicity of the electronic structures of chalcogenides, such, for example, as CuS or CuS2, is more apparent than real. Thus, according to the X-ray diffraction data, the formula Cuá4 Cu22á(S22 ¡)2(S2¡)2 should be assigned to covelline CuS, i.e., both the metal and chalcogen atoms in this compound exhibit mixed valences.17 However, after the results of the study performed by Jellinek were reported,77 the question about the electronic structures of copper chalcogenides became a matter of discussion. Based on the results of X-ray photoelectron spectro- scopy, Jellinek assigned the degree of oxidation of +1 to the copper atoms in all chalcogenides (for example, in CuS, CuSe, CuTe, CuS2, KCu4S3, etc.).To satisfy the principle of electro- negativity of the salts, Jellinek had to represent the structures of the copper chalcogenides by formulas, which are absurd from the chemical standpoint, such as, for example, Cu+Te7 for CuTe, Cuá3 (S22 ¡)S7 for CuS or K+Cuá4 S22 ¡S7, where S7 and Te7 are nothing but radical ions, which should be dimerised to the S2¡ 2 and Te2¡ 2 ions, respectively. I do not share the opinion of Jellinek and believe that one of the reasons for the absence of lines from the Cu2+ ions in the X-ray photoelectron spectra 77 is a mistake in the procedure for the preparation of samples. Thus, ground samples of the dichalcogenides were applied to the freshly purified surface of indium. In this case, it is quite possible that the chalcogenide reacted with indium as a result of which Cu2+ was reduced to Cu+ on the surface: Cu+ + In+.Cu2+ + In 2 The authors 77 did not perform subsequent treatment of the surface of the samples under study with Ar+ ions (to remove the surface layer containing the adsorbed molecules) until this treat- ment ceased to affect the spectral patterns. Note also that the formulas proposed by Jellinek do not allow one to explain the fact that the above-mentioned chalcogenides possess the metallic conductivity. Since both the Cu+ and S2¡ ions have filled electron shells, only the S7 ions (actually, these ions are the S7. radical anions) can be responsible for the charge transfer.Interactions between the orbitals of the adjacent S7. species would necessarily lead to the formation of the covalent bond 2S7. S2¡ 2 , and hence, the formulas proposed by Jellinek would be more properly written as Cuá2 (Te22 ¡) or Cuá6 (S22 ¡)3. However, in this808 case, the copper chalcogenides would exhibit dielectric properties and, moreover, the above-considered formulas are contradictory to the structural data.17 The fact that the approach proposed by Jellinek is inadequate gave impetus to the more accurate determination of the structure of CuS. Thus, Gotsis et al.78 attempted to redetermine the structure of CuS by the high-resolution X-ray powder diffraction method and calculated the electronic structure using density functional theory.Based on the results obtained, the authors reasoned that the Cu+7S2¡¦ 2 7Cu+ formula is more realistic for CuS (the authors believed that this formula is also appropriate for CuSe). However, this interpretation of the charge state of CuS suffers from drawbacks analogous to those of Jellinek's interpre- tation. Liang and Whangbo 79 believed that the Cu�¢3 (S¡¦2 )S27 formula is more adequate than the formula suggested by Jellinek [Cu�¢3 (S22 ¡¦)S7]. Although the authors noted the substantially covalent character of the Cu7S bond, they concluded that holes in the valence band are associated with oxidation of only sulfur, i.e., the authors did not assume the mixed-valence state for copper. A study 80 was devoted to the electronic structures of copper chalcogenides and their relationship with the conductivity.The authors calculated the electronic structures of copper sulfides with the use of the extended Hu�� ckel method and demonstrated that holes are approximately equally distributed over copper and sulfur atoms due to substantial overlap between the 3p orbitals of the sulfur atoms and the 3d orbitals of the copper atoms. This indicates that the classical formula for CuS is more realistic and, as will be seen from the data given below, agrees with the existence of charge density waves (CDWs) in conducting chalcogenides based on copper. The X-ray photoelectron spectra of samples of highly pure CuS, which was synthesised according to a procedure reported in the handbook,81 have lines at 932.2 and 933.7 eV with the intensity ratio of 1.3 : 1.10 These lines correspond to the energies of the 2p3/2 levels of copper [ECu(2p3/2)].The first and second lines can be assigned to the Cu+ and Cu2+ ions, respectively. Note that the ECu(2p3/2) value given for CuS (932.8 eV) in the handbook 82 is the average of the values determined in the study. 10 Jellinek 77 reported an obviously unrealistic value of 931.4 eV, which is even smaller than that for Cu2S (cf. 932.0 eV 77 and 932.9 eV 82). To summarise, it can be noted that the problem of the determination of the charge state of copper in chalcogenides is in essence identical to that for oxide HTSCs based on copper. However, in the case of chalcogenides the situation is more complex because the Cu7X bonds in these compounds are more covalent that those in the corresponding oxides. Taking into account the complexity and ambiguity of the determination of charge states of the copper and chalcogen atoms in chalcogenides, hereinafter the following approach will be used.Chalcogenides are considered to be compounds of monova- lent copper if the sum of the degrees of oxidation of the cations is equal to twice the number of the chalcogen atoms (i.e., on the assumption that the degrees of oxidation of the chalcogen and copper atoms are72 and +1, respectively). If this sum is smaller than twice the number of the chalcogen atoms, chalcogenides are considered as mixed-valence compounds in which the degree of oxidation of copper is larger than +1 and the degree of oxidation of chalcogen is larger than72.Thus, according to this approach, sulfide NaCu5S3 contains the Cu+ and S27 ions, whereas sulfide NaCu4S3 is a mixed-valence compound. This approach is known in the literature as Zintl's concept (according to Ref. 83). Ternary copper chalcogenides can be divided into two groups: compounds with chain structures and compounds with layered structures. At the same time, according to the above-proposed approach, these compounds can be separated into copper(I) chalcogenides and mixed-valence copper chalcogenides (the degree of oxidation of copper is larger than unity). The fruitful- ness of this approach is apparent. Thus, virtually all copper(I) chalcogenides are semiconductors, whereas mixed-valence copper chalcogenides possess the metallic conductivity.V A Starodub All chalcogenides of the general formula MCu2nXn+1 (where M is alkali metal, X is chalcogen, n=1 ¡À 3) have layered struc- tures. The following layered ternary copper chalcogenides are known: TlCu2Te2 ,80 TlCu2Se2 ,84 TlCu2S2 , TlCu2SeS,85 TlCu4S3, TlCu4Se3,80, 86 KCu4S3,87 KCu4Se3,86 RbCu4S3,88 RbCu4Se3, CsCu4S3, CsCu4Se3 86 and TlCu6S4.89 Since the total charge of the cations in the compounds of this series is one less than twice the number of the chalcogen atoms, these chalcogenides should be considered as mixed-valence compounds. It is therefore concluded that the valence band contains holes.Hence, these chalcogenides should exhibit metallic properties. Selenide TlCu2Se2 was prepared 84 by heating (<1050 8C) stoichiometric amounts of TlSe, copper and selenium in vacuo followed by slow cooling. The crystals of TlCu2Se2 consist of the Cu7Se layers. The Tl+ ions are located above and below the layers. The Cu7Se bonds within the layers are substantially covalent. The copper atoms are located inside the tetrahedra formed by the selenium atoms, only three-fourths of positions being occupied (the ThCr2Si2 structural type) (Fig. 19). Tl Se Cu a c Figure 19. Crystal structure of TlCu2Se2.90 It was found 90 that the TlCu2Se2 salt exhibits metallic properties. To account for this fact, Brun et al.90 assumed the coexistence of Tl(I) and Tl(III) selenides in the lattice.However, this assumption is contrary to the structural data and the results of electrophysical measurements.91 Selenide TlCu2Se2 possesses the rather high conductivity (rRT=561075 O cm). The temper- ature dependence of the conductivity, which was measured down to liquid helium temperature, is typical of metals. No substantial difference between the conductivities measured for single crystals and for pressed powders of TlCu2Se2 was observed, which is indicative of the weak anisotropy of the resistance of crystals. For TlCu2Se2, the Hall effect was measured and it was demonstrated that this selenide is a metal with hole conductivity (one hole per formula unit, like in the isostructural TlCu2Te2 compound 91).This fact rules out the existence of two forms of thallium suggested in the study 90 because if this were the case, electrons, which fill the conduction band, would act as carriers. Isostructural sulfide TlCu2S2 was prepared by Berger.85 Thus, heating a one-phase sample of Tl4S3, which has been synthesised beforehand, with powdered copper and sulfur afforded TlCu3S2 the subsequent treatment of which with a concentrated ammonia solution and a 30% H2O2 solution yielded TlCu2S2. This com- pound is unstable 85 and reversibly decomposes to three phases, viz., to TlCuS, TlCuS2 and TlCu4S3, on heating to 330 K. Sulfide TlCu2S2, like the selenide and telluride, exhibits metallic properties. Later on, chalcogenides TlCu2X2 (X=S or Se) and mixed chalcogenide TlCu2SSe were prepared 92 and studied by X-ray photoelectron spectroscopy. According to Jellinek's concept, the authors 92 assigned the degree of oxidation of +1 to the thallium and copper atoms in TlCu2SSe.The X-ray photoelectron spec- trum of this compound has two lines in the ECu(2p3/2) region. The authors attributed these lines to the difference in the environment about the copper atoms and suggested that one line belongs to the copper atom bound to the selenium atom and the second line belongs to the copper atom bound to the sulfur atom. However, this assignment of the lines implies the structural ordering in theTernary and quaternary chalcogenides of Group IB elements crystals of the mixed chalcogenide and the equality of the two lines, which was not observed experimentally.The metal ¡À nonmetal transition should occur on addition of one electron to the valence band of chalcogenide MCu2X2 (M is alkali metal) because the valence band becomes completely filled. Actually, the replacement of alkali metal by alkaline-earth metal causes this transition and chalcogenides BaCu2S2 and BaCu2Se2 exhibit dielectric properties.83, 93 ¡&AgrSingle crystals of BaCu2X2 develop an orange colour in transmitted light and a metallic-black colour in reflected light, which is characteristic of semiconductors. In the study,93 chalco- genides BaCu2X2 were prepared by heating stoichiometric amounts of barium and copper chalcogenides with metallic copper.In this case, orthorhombic crystals with a chain structure were formed. Sulfide BaCu2S2 was prepared by the reaction of BaCO3 and CuO with sulfur at 700 8C under a stream of CS2.94 However, this procedure afforded the tetragonal phase of BaCu2S2,95 which has the layered structure 83, 95 analogous to the structure of TlCu2X2 (the ThCr2Si2 structural type). The conductivity of sulfide BaCu2S2 was measured 96 and this 2 S22 ¡¦. At the compound was found to exhibit metallic properties in the range of 40 ¡À 293 K. This result is difficult to explain on the basis of the accepted notion of the electronic structure of chalcogenides because sulfide BaCu2S2 does not contain holes in the valence band and can be represented by the formula Ba2�¢Cu�¢ same time, it was demonstrated 83 that this compound exhibits semiconducting properties at 77 ¡À 293 K.This discrepancy in the results was accounted 83 for by the procedure for the synthesis used in the study,96 which afforded a mixture of Ba0.96Cu2.01S1.94 and Ba0.95Cu3.91S2.97 in a ratio of 17 : 3. The resistance of the sulfide increases insignificantly (by a factor of *2.7) as the temperature decreases from 300 to 77 K, from which an approx- imate estimate of the activation energy can be obtained (Ea*0.01 eV). This estimate was not made in the study.83 For BaCu2S2, the magnetic susceptibility was measured and the superconductivity was found to be absent to 2 K. The temper- ature dependence of the magnetic susceptibility is described by the following equation wexp=CT+wdia+w0 , where C is the Curie constant, wdia is the diamagnetism of the core and w0 is the temperature-independent paramagnetism.Judging from the value of the Curie constant (C= 5.8361073 emu K mol71), the paramagnetic contribution results from admixtures whose concentration was *1.5% and BaCu2S2 possesses diamagnetic properties. Hence, Zintl's concept appears to be true for both polymorphous modifications of BaCu2S2, which justifies its use in the present review. Chalcogenides NaCuSe, NaCuTe, KCuSe, KCuTe 97 and KCuS 98 in which the degrees of oxidation of the copper and chalcogen atoms are +1 and 72, respectively, are similar in electronic properties to BaCu2X2. In spite of the formal similarity, these compounds have different structures.Thus, chalcogenides NaCuSe and NaCuTe belong to the PbFCl structural type, whereas KCuS, KCuSe and KCuTe belong to the Ni2In struc- tural type (contain zigzag Cu7S chains). The authors related this change in the structural type caused by the replacement of sodium by potassium to the difference in the ionic radii and noted that this change is unusual because only intermetallic compounds rather than salts belong to the Ni2In structural type. These chalcoge- nides, judging from their colour, are narrow-band semiconduc- tors. All these compounds were isolated as bright black crystals. Chalcogenides of the general formula MCu4X3 (M is alkali metal and X=S or Se) were described for the first time by Ru�� dorff.99 He isolated the KCu4S3 and RbCu4S3 salts.The CsCu4S3 88 and TlCu4S3 86 salts and the selenium analogues MCu4Se3 (M=Tl, K, Rb or Cs) 86 were described. The majority of these compounds were synthesised according to Ru�� dorff's 809 carbonate method. Copper and chalcogen were mixed with carbonate of the corresponding metal and the mixture was heated to 650 ¡À 700 8C. The resulting product was isolated as platelet-like black crystals. The thallium salts were synthesised with the use of Tl2S or TlSe. Unlike the sulfide, the selenide CsCu4Se3, which was prepared according to the carbonate method, was not a one-phase compound.86 The KCu4Se3 salt was prepared 86 also by fusing stoichiometric amounts of K2Se, Cu and Se in a sealed evacuated quartz tube at 800 8C.All the above-mentioned chalcogenides have layered struc- tures. The double layers consist of sulfur atoms, which form tetrahedra linked via shared edges. The copper atoms are located inside the tetrahedra and the alkali metal or thallium atoms are located between layers. Due to mutual repulsions between the like-charged [Cu4X3] n¡¦ n layers, it is required by the sizes of theM+ ions that the M7X distances will be larger than the sum of their ionic radii. Thus, the Na+ ion is too small to screen mutual repulsions between the layers and hence, NaCu4X3 salts do not exist. Goodenough (according to Ref. 86) suggested the D0(n, l) criterion, which depends on the electronic configuration of the metal atom and the nature of the bridging ligand.This criterion determines the possibility of the nonmetal ¡À metal transition. According to this criterion, Cu2+ sulfide complexes with the d9 configuration can exhibit metallic properties if the internuclear distances Cu7Cu<0.31 nm. In sulfide KCu4S3, four Cu7Cu distances are 0.276 nm and one Cu7Cu distance is 0.297 nm. These data satisfy Goodenough's criterion well and, conse- quently, KCu4S3 can exist in the metallic state. Hence, taking into account also the mixed-valence state of the copper and chalcogen atoms (the formal degree of oxidation of copper is +1.25), it can be suggested that chalcogenides MCu4X3 are metals. Actually, as early as 1952 Ru�� dorff found 99 that KCu4S3 and RbCu4S3 exhibit metallic properties. According to Ru�� dorff, the specific resistance rRT of KCu4S3 and RbCu4S3 at*20 8C is 2.561072 and 1.6761072 O cm, respectively.Later on, the major metallic state of KCu4S3 was confirmed: 87 r=2.561074 and 1.6761075 O cm at 293 and 20 K, respec- tively. The specific resistance changes linearly with temperature throughout the temperature range. Sulfide KCu4S3 exhibits temperature-independent paramagnetism (Pauli paramagnetism) and shows metallic reflection in the visible and near UV regions. Ru�� dorff obtained higher values of the electrical resistance because he performed measurements with the use of the double-contact method, which ignored the thermoelectromotive force. In 1983, it was established 100 that other alkali thiocuprates MCu4S3 also exhibit metallic properties.Their electrical resistance changes linearly with temperature in the range of 150 ¡À 300 K. At *293 K, the resistance is 6.2561074, 7.1461074 and 861074 O cm for KCu4S3, RbCu4S3 and CsCu4S3, respectively. All three salts show temperature-independent paramagnetism. Selenide CsCu4Se3 was prepared 101 by the hydrothermal synthesis. Alkali metal polyselenide M2Sex (M=Na or K; x=2, 4 or 5), CsCl and copper were placed in a tube with water. The tube was evacuated, sealed and heated at *120 8C for five days. Needle-like crystals of a-CsCuSe4 and platelet-like crystals of CsCu4Se3 were isolated from the reaction mixture. The increase in the temperature to 170 8C resulted in an increase in the yield of CsCu4Se3 to 80% but the compound was formed as polycrystals.The Cu7Cu distances in selenides MCu4Se3 (0.2895 and 0.3071 nm101) satisfy Goodenough's criterion. Hence, taking into account also the mixed-valence states of the copper and selenium atoms, these selenides can be considered as metals. n The crystal structures of the MCu4X3 salts 102 are character- ised by the presence of octahedral cavities (one cavity per formula unit) between the [Cu4X3] n¡¦ layers. The size of the cavity is large enough for an additional copper atom to occupy this cavity. The insertion of copper atoms into the octahedral cavities should lead to annihilation of holes in the valence band and, consequently, should cause the metal ¡À nonmetal transition.According to Zintl's810 criterion, chalcogenide MCu5X3 contains the M+Cu�¢5 X23 ¡¦ ions with the closed shells. The electrochemical reaction annihilation of holes (1) K+[(Cut)4S3]7+Cu+ + e K+[(Cu�¢t )4Cu�¢o n of holes semiconductor (diamagnetic) metal (Pauli paramagnetism) afforded 103 chalcogenide KCu5S3 (Cut and Cuo are the copper atoms occupying the tetrahedral and octahedral positions, respectively); the back reaction, viz., anodic oxidation of KCu5S3, gave the initial sulfide KCu4S3. According to the X-ray diffraction data,103 the tetragonal symmetry of the crystals is retained in the course of the reaction. The unit-cell volume decreases from 140.46109 nm3 for KCu4S3 to 129.66109 nm3 for KCu5S3 due to additional attractions between the charged layers caused by the copper ions occupying the octahedral cavities between the layers.The specific resistance increases from 7.3461074 O cm for KCu4S3 to 1.9061073 O cm for KCu5S3, the sign of the temperature resistance coefficient changing to minus, i.e., KCu5S3 is a semiconductor. As mentioned above, the NaCu4S3 salt does not exist. However, the Na2Cu4S3 and NaCu5S3 salts are known.104, 105 Both these salts can be considered as analogues of theKCu5S3 salt, in which octahedral cavities are occupied either by Na+or by Cu+ ions. The NaCu5S3 salt was prepared by the hydrothermal synthesis and the Na2Cu4S3 salt was synthesised from a carbo- nate melt. Crystals of NaCu5S3 belong to the hexagonal system (unlike tetragonal crystals of KCu5S3) (Fig.20). There are two crystallo- graphically nonequivalent copper atoms per asymmetric unit. The Cu(1) atoms are almost linearly coordinated by two sulfur atoms. The Cu(2) atoms are coordinated by three sulfur atoms (as in Na2Cu4S3) to form a trigonal pyramid. The Cu(1) atoms each have four Cu(2) neighbouring atoms, whereas the Cu(2) atoms each are surrounded by six Cu(1) atoms. The Cu7Cu distances (0.27 nm) satisfy Goodenough's criterion. However, according to Zintl's criterion, both salts contain ions with closed shells and, consequently, the NaCu5S3 and Na2Cu4S3 salts are, apparently, semiconductors. This is also true for selenide TlCu5Se3.85 Annihilation of holes in the valence band of MCu4X3 can be also induced by the replacement of alkali metal by alkaline-earth metal.Sulfide BaCu4S3 was prepared 106 by heating a mixture of BaS, copper and sulfur in a sealed evacuated tube at 800 8C as crystals, which develop a red colour in transmitted light and exhibit metallic lustre in reflected light. According to the X-ray diffraction data, these are two different phases, viz., the a and b phases. The a phase, which is stable at room temperature, is transformed into the b phase at 640 8C. Both phases belong to the Cu(1)S2 b NaS6 a Cu(2)S3 0 1 2 3 A Figure 20. Crystal structure of NaCu5S3 projected onto (001). The NaS6 octahedra (the Na atoms at z/c=1/4 and 3/4), the linear coordination in Cu(1)S2 (the Cu(1) atoms at z/c=1/4 and 3/4) and the Cu(2)S3 pyramids [the Cu(2) atoms at z/c*0 and 1/2].V A Starodub orthorhombic system. Their structures are similar and consist of chains formed by the BaS6 trigonal prisms linked to each other via shared triangular faces. The copper atoms form infinite zigzag chains in the a phase (Cu7Cu, 0.275 nm) and infinite linear chains in the b phase (Cu7Cu, 0.29 nm). In spite of the filled valence band (as follows from the formula Ba2+Cu�¢4 S23 ¡¦), the short Cu7Cu contacts are responsible for the noticeable con- ductivity (rRT=0.2 O cm) of the crystals of BaCu4S3, which is virtually independent of the temperature. Only one salt belonging to the MCu2nXn+1 series with n=3is known, viz., TlCu6S4.89 This salt was prepared by fusing a mixture of Tl2S, copper and sulfur in a quartz tube.Its structure is similar to those of the preceding members of the series with n=1 or 2. The crystal contains layers consisting of the CuS4 tetrahedra alternating with layers of the thallium atoms. The TlCu6S4 salt should exhibit metallic properties because it is a mixed-valence compound with the Cu7Cu distances of 0.2783 and 0.2917 nm. Actually, the electrical resistance of single crystals of TlCu6S4 varies (although not quite linearly) proportional with temper- ature. The crystals are characterised by a plus sign of the Hall constant. The concentration of holes was determined from this constant (+1.29 per formula unit). Berger and Eriksson 89 noted that the effective radius of sulfur increases in the series TlCu2nSn+1 as n increases and related this effect to a change in the concen- tration of holes in the valence band per sulfur atom.A decrease in the ratio from +0.5 at n=1 to +0.25 at n=3 correlates with an increase in the radius (the closer the charge of sulfur to 72, the larger the radius). This interpretation is reasonable. However, from this interpretation it does not follow that holes are localised exclusively on the sulfur atoms. In chalcogenides TlCu7X4 (X=S or Se), which are derivatives of the TlCu6S4 salts, additional copper atoms occupy cavities, which leads to filling of the valence band and causes the metal ¡À nonmetal transition.85 The mineral crookesite with com- position TlCu6.8Se4, in which 80% of vacancies are occupied, possesses metallic properties 89 because the valence band is only partially filled (there are +0.2 holes per formula unit in the crystal).The members of the MCu6X4 series are represented only by thallium salts. However, the ammonium analogue of TlCu7X4, viz., the NH4Cu7S4 salt, is known. This salt satisfies Zintl's criterion and, consequently, has a filled valence band.87 Judging from the Cu7Cu bond lengths (0.283 and 0.294 nm), this salt possesses semiconducting properties. Since no salts of the MCu2nXn+1 series with n>3 are available, the question of whether these salts can exist was considered theoretically.80 The formal analogy of the reactions of the type Cu2nXn+1+Cu2X Cu2(n+1)Xn+2 was drawn to the so-called isodesmic reactions, which are well known in theoretical organic chemistry.In these reactions, the number of bonds of a particular type remains invariant, but their relative arrangement changes. For example, 2CH3CH3. CH3CH2CH3+CH4 It was demonstrated that this concept is applicable to two layers of the Cu2nXq¡¦ n�¢1 anions (q=1 or 2) and the energy effects of the isodesmic reactions with the participation of these anions were calculated (the extended Hu�� ckel method).80 (2) Cu2Sq¡¦2 +Cu2S Cu4 Sq3¡¦, (3) Cu4Sq¡¦3 +Cu2S Cu6 Sq4¡¦, (4) Cu6Sq¡¦4 +Cu2S Cu8 Sq5¡¦. (4) (3) (2) Reaction 70.061 70.159 70.320 DE /eV (q=1) 0.000 70.002 70.023 DE /eV (q=2)Ternary and quaternary chalcogenides of Group IB elements The results of calculations are in complete agreement with the experimental data.Thus, these reactions are most favourable for compounds with q=1, viz., MCu2(n+1)Xn+2 , MCu2nXn+1+Cu2X where M is a single-charged ion (alkali metal or thallium). These reactions are less favourable for analogous barium salts (q=2). Thus, the absence of the BaCu6S4 salt is attributable to fact that the gain in energy is virtually absent for the reaction BaCu6S4 . BaCu4S3+Cu2S Judging from the very low energy of reaction (4), MCu8S5 salts, most probably, do not exist. The absence of compounds of the MAg2nXn+1 series lends support to the validity of the approach, which we applied to the determination of the electronic structures of ternary copper chalcogenides.In these compounds, silver should possess mixed valence because the oxidation state of this element in chalcoge- nides is no greater than +1. However, when hydrogen sulfide was passed through a melt of Ba[Ag(CN)2] in a mixture with KNCS (500 8C, 30 min), the sulfide BaAg2S2 (CaAl2Si2 structural type), which meets Zintl's concept, was obtained.107 The structure of CaAl2Si2 is characterised by the presence of the [Al2Si2]27 structural elements in which three crystallographically equivalent Si7Al bonds are shorter than the fourth bond directed along the c axis. An analogous situation was observed in the crystals of BaAg2S2, with the only difference that three equivalent S7Ag bonds (0.2686 nm) are longer than the fourth bond (0.2659 nm).The data on the conductivity were not reported in the study.107 However, based on Zintl's conceaking into account the dark-red colour of the crystals, it is believed that sulfide BaAg2S2 is a semiconductor. Chalcogenides K2Ag4S3 87, 110 (Fig. 21 a) and K2Ag4Se3 109 (analogues of Na2Cu4S3) as well as RbAg5S3 108 (an analogue of NaCu5S3) and CsAg7S4 108 (an analogue of TlCu7S4) satisfy Zintl's criterion. The M2Ag6S4 salts (M=Na or K) containing an alkali metal atom instead of the seventh silver atom were reported.66 These chalcogenides were prepared according to the carbonate procedure by heating a mixture of powdered silver, chalcogen and the corresponding alkali metal carbonate to 1000 8C for *6 h. Since the resulting anthracite-like bright crystals are unstable in air, the authors 109 failed to isolate the K2Ag4Se3 salt as single crystals suitable for X-ray diffraction analysis.a 1/2 1/2 0 0 1/20 0 1/2 1/2 1/2 0 0 1/2 1/2 0 1/20 1/2 0 0 0 1/2 0 1/2 1/2 0 1/2 0 1/2 0 0 1/2 1/2 1/2 1/2 0 0 0 1/2 1/2 1/2 0 0 0 1/2 SKAg 1/2 0 1/2 1/2 0 0 b 1/2 0 1/2 0 1/2 1/2 0 1/2 0 1/2 0 0 1/2 0 0 1/2 0 1/2 0 0 1/21/2 0 1/2 1/2 1/2 0 0 1/2 0 0 1/2 0 0 1/2 1/2 0 1/2 0 1/2 0 1/2 0 0 1/2 0 1/2 1/2 0 0 0 SCs Ag 1/2 0 0 1/2 1/2 1/2 1/2 0 0 1/2 Figure 21. Projections of the structures K2Ag4S3 (a) and CsAg3S2 (b) along the b axis. 811 The RbAg5S3 and CsAg7S4 salts were prepared as dark-red needle-like crystals according to the carbonate procedure by heating in an autoclave (*300 8C, 84 h) with the use of ethyl- enediamine in the supercritical state as the solvent.The hexagonal crystals of RbAg5S3 contain the Ag6S6 twelve- membered rings in which the silver and sulfur atoms adopt the coordination number of 3. The rings are linked to each other through the silver atoms (the coordination number is 2) with the nearly linear coordination and are packed in stacks. The structure contains channels with a diameter of *0.7 nm. These channels are occupied by the Rb+ ions (Fig. 22). The Ag7Ag distances vary from 0.296 to 0.307 nm, which may indicate the presence of covalent interactions between the Ag+ ions with the d 10 config- uration.It was noted 108 that the potassium analogue KAg5S3 has an analogous structure. Rb Ag S Figure 22. Projection of the structure of RbAg5S3 along the c axis;108 columns containing Rb+ ions are shown. The structure of the cesium salt CsAg7S4 is unique. The crystals contain the eight-membered Ag4S4 rings in which the silver and sulfur atoms adopt the coordination numbers of 2 and 3, respectively. These rings, as in the structure of RbAg5S3, are linked to each other through the silver atoms, which have a nearly linear coordination. The Ag+7Ag+ contacts, as in the rubidium salt, are in the range of 0.2961 ± 0.3087 nm.99 The rings are packed in infinite stacks. The channels (with a diameter of *0.6 nm) formed in the structure are occupied by the cesium ions (Fig.23). Analysis of these structures allows the conclusion to be drawn about the similarity of the structural motifs of ternary copper and silver chalcogenides satisfying Zintl's concept. The ions with the d 10 configuration have the linear or trigonal coordination, whereas the copper ions, which exist in a formal oxidation state greater than +1, are tetrahedrally coordinated. The compounds of the series of chalcogenides of the general formula MB3X2 (M is alkali metal or thallium; B=Cu or Ag; X=S or Se) can be considered as chalcogenide derivatives MCu2X2 (5) MCu3X2 . MCu2X2+Cu As in the case of reaction (1), the insertion of an additional copper atom leads to annihilation of holes in the valence band and causes the metal ± semiconductor transition.Since the compounds of the MB3X2 series meet Zintl's criterion, i.e., contain only B+ ions, these compounds may be thought to possess a structural motif identical to that of the above-considered silver salts. The first compounds of this series to be prepared by the carbonate procedure were the silver salts RbAg3S2 and CsAg3S2.111 Their structures are isotypic and are in many respects similar to the structure of K2Ag4S3. In both cases, the sulfur and silver atoms form layers between which the alkali metal atoms are812 Cs Ag S Figure 23. Projection of the structure of CsAg7S4 along the c axis;108 channels formed by eight-membered rings are shown. located, with the only difference that these layers in MAg3S2 are parallel to the (001) plane, whereas these layers in K2Ag4S3 are parallel to the (100) plane (see Fig.21 a, b). However, the structures of the layers are different. Thus in the structure of K2Ag4S3, the eight-membered Ag4S4 rings are linked to each other via shared vertices through the bridging sulfur atoms, whereas the analogous rings in MAg3S2 are linked via shared edges. In Fig. 21 a, b, the fused Ag4S4 rings, which are similar in appear- ance but crystallographically nonequivalent, are hatched. The first compound of the series of copper chalcogenides to be synthesised was sulfide KCu3S2, which was prepared by heating a mixture of K2CO3, copper and sulfur to 850 8C under an Ar atmosphere.112 When the reaction was performed at a temper- ature lower than 780 8C, sulfides KCu4S3 and K3Cu8S6 were obtained 99 (see below).The KCu3S2 salt exists as bright black needle-like crystals insoluble in water. These crystals as well as the crystals of MAg3S2 (M=Rb or Cs) belong to the same structural type (X-ray diffraction data). Later on,88 the cesium salt CsCu3S2 was prepared by the carbonate procedure. It was found that brief heating (up to 1.5 h) at 700 8C afforded CsCu4S3. However, an increase in the duration of annealing and in the temperature (4 h, 800 8C) resulted in the CsCu3S2 salt, which was free from admixtures. Unlike the blue- black crystals of CsCu4S3, the crystals of CsCu3S2 are transparent and almost colourless. Since these crystals are oxidised in air and develop a dark-brown colour, they are stored under an inert atmosphere.From the colour of the crystals and their composi- tion, which satisfies Zintl's criterion, CsCu3S2 can be unambigu- ously assigned to compounds possessing dielectric properties. In spite of the formal similarity to KCu3S2, the crystal structure of CsCu3S2 is quite different (the CdI2 structural type). The sulfur atoms form hexagonal close packing. Half of octahe- dral positions are occupied by layers of Cs+ ions. The layers with unoccupied octahedral positions, which remain unoccupied in the structure of CdI2, are filled with copper atoms located at the centres of the edges of the octahedra (Fig. 24). The sulfur (the coordination number is 3) and copper (the coordination number is 2) atoms form layers perpendicular to the c axis.The copper ions have a linear coordination, which is uncommon in the chemistry of alkali chalcogenides of copper. It is evident from the above-considered examples of chalcogenides, which meet Zintl's criterion, that trigonal coordination or trigonal coordina- tion combined with the linear coordination is more typical of these compounds. V A Starodub +z +z 7z 7z 7z 0 1/2 1/2 +z +z +z 0 0 0 Cu Cs S 7z 7z 7z 1/2 0 1/2 +z +z +z b a 7z 7z Figure 24. Projection of the structure of CsCu3S2 along the c axis.88 In the above-considered sulfide KCuS, the copper atoms have a linear coordination (the SCuS angle is 179 8). However, the crystal structure consists only of chains, which impart the quasi- one-dimensional structure to this salt.The Cu7S distances in KCuS (0.213 and 0.216 nm) and in CsCu3S2 (0.217 nm) are virtually identical and are typical of copper sulfides with covalent bonds. The thallium chalcogenides TlCu3X2 (X=S or Se) described previously 85, 113 ± 117 were synthesised in two stages. Initially, Tl4S3 was prepared from the elements. Then, the fusion of the latter with copper and sulfur afforded sulfide TlCu3S2. These chalcogenides are very sensitive to oxidising agents. Thus, the selenide is oxidised in air, whereas active oxygen formed upon decomposition of hydrogen peroxide is required to oxidise the sulfide. The reaction (6) TlCu2X2+Cu TlCu3X2 can be performed by oxidation (including electrochemical oxida- tion 116, 117).Berger called this reaction `extraction' of copper from TlCu3X2. This reaction is analogous to the above-described process (1) for the KCu5S3 ±KCu4S3 system and can be formally represented by Scheme (5). To perform chemical oxidation, powdered TlCu3S2 was deposited on a glass filter (with the aim of performing the process in a dynamic mode) and an ammonia solution of hydrogen peroxide, which decomposed under the action of the sulfide, was added dropwise. Ammonia that was released coordinated the Cu2+ ions, which were formed as result of oxidation, thus facilitating their extraction into a solution. Selenide KCu3Se2 was synthesised by heating (910 8C, 3 h) a mixture of powdered copper, potassium and copper selenide in a quartz tube.118 The KCu3Se2 salt was obtained as well-faceted black platelet-like crystals.Its crystal structure is analogous to that of TlCu3Se2 (X-ray diffraction data). The data on other copper selenides MCu3Se2 are unavailable in the literatute. However, analogous silver selenides RbAg3Se2 and CsAg3Se2 were reported.109 Both salts crystallise as dark anthracite-like crystals and are readily oxidised in air. Their structures are similar to those of the analogous sulfides. The selenium and silver atoms form layers separated by layers of the M+ ions. n Sulfide Na2Cu3S3, which crystallised as needles, was described in a pioneering study.99 More recently, it was demonstrated 119 that this sulfide has in fact the composition Na3Cu4S4 and its structure contains the [Cu4S4] 3n¡ stacks.Below is schematically shown the formation of these stacks from the cis-CuS chains. The coordination about the copper atoms is planar-trigonal. S S Cu Cu Cu S S S Cu Cu The stacks are separated by the Na+ ions, thus imparting the quasi-one-dimensional electronic structure to the crystals. The shortest Cu7Cu distances (0.2619 ± 0.306 nm) satisfy Goode- nough's criterion. Besides, the Na3Cu4S4 salt is a mixed-valence compound. As was noted in the study,119 all copper atoms in theTernary and quaternary chalcogenides of Group IB elements n structure are equivalent, which is typical of metallic bonds (electrons delocalised along the [Cu4S4]3n¡¦ stacks), and sulfide Na3Cu4S4 should exhibit metallic properties. The electrophysical properties of this sulfide were exam- ined.100, 120 The electrical resistance of single crystals of Na3Cu4S4 was found 120 to change linearly as the temperature decreased over the range from 300 to 150 K. Below 150 K, the r(T) dependence becomes weaker, tends to saturation and reaches 3.361076 O cm at 15 K.100 The electrical resistances for tableted samples at 293 K (rRT=3.361073 O cm) and 15 K (r15=6.761074 O cm) were reported.100 For single crystals, the electrical resistance along the crystal axis, i.e., along the stacks, was measured (rRT=6.561075 O cm).This r(T) dependence is characteristic of metals. Although Ghosh and Chaudhury 100 failed to measure r\ (the resistance perpendicular to the stacks), the noticeable difference (by a factor of *50) in the resistance of tableted samples and single crystals is indicative of the substantial anisotropy of the electrical resistance.The temperature-independent paramagnet- ism, which was found in the studies 100, 120 (15061076 and 17061076 emu mol71, respectively), is also evidence in favour of the metallic state of Na3Cu4S4. It should be noted that the change in the specific electrical resistance according to the law (7) r=r0+r 0T is typical of normal metals, unlike quasi-one-dimensional molec- ular metals, whose specific resistance changes quadratically. The authors 100, 120 used the approach proposed by Jellinek in studies of sulfide Na3Cu4S4 as a salt containing only Cu+ ions and mentioned that the oxidation state of +2 for copper atoms was not revealed by X-ray photoelectron spectroscopy. Curiously, the authors 120 considered all sulfur atoms as being also identical, whereas Jellinek's concept requires the presence of S27 and S7 ions in the structure.However, this is quite reasonable when it is considered that the compound under study is a metal and the electrons are delocalised along the stacks. In this case, the valence fluctuations Cu2+?Cu+within the stack occurred in*10716 s. The absence of the metal ¡À nonmetal transition up to 13 Kand the peculiarities in the r(T ) dependence associated with the formation of charge density waves (CDWs) are not understood. It is known 121 that the so-called Peierls instability, which is manifested in the structural distortion of the one-dimensional chain associated with the increase in its period, is characteristic of quasi-one-dimensional metals.If the degree of filling of the band with electrons or holes is n71, a decrease in the temperature leads to n-merisation of the chain accompanied by splitting of the conduction band in such a way that the metal ¡À nonmetal transition occurs. In this case, the larger is n, the smaller the gain in energy. n The reasons for the absence of the Peierls transition in the crystals of Na3Cu4S4 up to 13 K were theoretically examined.122 In particular, it was noted that Na3Cu4S4 is a one-dimensional metal because the shortest distance between the adjacent one- dimensional [Cu4S4] 3n¡¦ stacks (0.423 nm) in this salt is substan- tially larger that the sum of the van der Walls radii of sulfur atoms (0.360 nm).Calculations within the framework of the extended Hu�� ckel method demonstrated that the Fermi level intersects two bands, each being formed half by the 3d orbitals of the copper atoms and half by the 3p orbitals of the sulfur atoms. Hence, Whangbo and Canadell 122 believed that the hybrid formula including both the Cu2+ and S7 ions, viz., Cu�¢3 Cu2�¢S24 ¡¦ and Cu�¢4 S23 ¡¦S¡¦, is more adequate than Jellinek's formalism. This view is similar to our approach to the interpretation of the electronic structures of copper chalcogenides, which do not meet Zintl's concept.Whangbo and Canadell 122 believed that the state with CDWs preceding the Peierls transition is not formed due, probably, to the fact that the stabilisation of the electronic energy is exactly 813 compensated by the deformation energy of the stack. In the case of one-dimensional metals in the absence of states with CDWs, an alternative possibility consists in the formation of spin density waves (SDWs), which is generally observed at lower temperatures. In a review,121 we noted that the spin ¡À lattice interaction respon- sible for the spin-Peierls transition is more efficient than the electron ¡À lattice interaction at t&U, where t is the integral of the electron transition between the nearest electronic states and U is the interaction energy of two electrons located at one node.In one-dimensional conductors [MS4C4(CF3)4]TTF (M=Cu or Au; TTF is tetrathiafulvalene), the spin-Peierls transition is observed at 12 (Cu) and 2.1 K (Au). I believe that the deviation of the r(T ) dependence from the linearity for Na3Cu4S4 at T<80 K is, probably, caused by fluctuations preceding the spin-Peierls transition at T<13 K. A series of ternary copper sulfidesMnCuS with the Cu : S ratio of 1 : 1 can be distinguished based on formal attributes. In this series, all members with n from 0 (CuS) to 1 (KCuS) are known, the n value varying with a step of 0.25. Thus, the above-considered sulfide Na3Cu4S4 corresponds to n=0.75, whereas the above- considered layered sulfides MCu2S2 (the first members of the MCu2nXn+1 series) are characterised by n=0.5.Finally, sulfide NaCu4S4 synthesised in 1996 123 corresponds to n=0.25. It should be noted that the MnCuS series can be distinguished only formally because the insertion of alkali metal into the lattice of CuS leads not only to a change in the electronic properties but also to a change in the structure. Thus, unlike Na3Cu4S4, which has a quasi-one-dimensional stack structure, NaCu4S4 has a layered structure, though formally contains the same Cu4S4 structural elements. Heating of a mixture of copper, Na2S and BaS with sulfur (taken in a ratio of 1 : 3 : 0.5 : 6, respectively) in an evacuated glass tube (500 8C, 4 days) followed by coolite of 4 8C/h afforded a melt, which was freed off from polysulfides with water and DMF under a nitrogen atmosphere.As a result, platelet-like black crystals of NaCu4S4 were obtained in 79% yield. The crystals consist of trigonal-symmetrical layers of CuS alternating with layers of Na+ ions (Fig. 25). The structure of Na0.25CuS, like CuS, contains both sulfide and disulfide ions. The [Cu4S2(S)2]7 layers have the sandwich structure: the Cu2S2 layers (the anti-GaS structural type) are located between two CuS layers (the BN structural type). These layers are linked through the copper atoms from the anti-GaS layer and through the sulfur atoms from the BN layer to form the Cu7S bond resulting in four-coordinate S2¡¦ ions with the unusual inverted tetrahedral geometry.Unlike the structurally similar CuS salt in which the Cu2S2 layers of the anti-GaS type alternate with the CuS layers of the BN type, half of the anti-GaS layers in NaCu4S4 are replaced by layers of Na+ ions. Therefore, the structure of NaCu4S4 contains two crystallographically nonequivalent copper atoms, b a Cu SNac a b b Figure 25. Packings of the atoms in NaCu4S4 (a) and in the [Cu4S4]7 ions (b) along the c axis (large and small empty circles denote S atoms from two different layers).814 2 ions and the trigonally coordinated copper atom. viz., the copper atom tetrahedrally coordinated by one S2¡ ion and three S2¡ Sulfide NaCu4S4 is a metal with the hole conductivity. Its electrical resistance increases linearly with temperature in the range from 4.2 to 250 K. From measurements of the thermo- electromotive force it follows that holes act as charge carriers.This salt wM= possesses the Pauli paramagnetism, 6.261075 emu mol71. Sulfide BaCuS3, which we prepared 124 by heating a mixture of BaS, CuS and sulfur taken in stoichiometric amounts (800 8C, 2 h), has no analogues. This sulfide was formed as black bright platelet-like crystals belonging to the monoclinic system. Since the composition of this sulfide does not satisfy Zintl's criterion, it is believed that this compound is a metal. Actually, the specific resistance (rRT=9.461072 O cm) of the samples, though large, decreases as the temperature decreases and reaches 0.037 O cm at 15 K. Unlike other ternary copper sulfides, the r(T ) dependence is described by a fractional-rational rather than by a linear function (8) r(T) = 1 á aT b á gT , where b71=r(0), i.e., b=s(0), a=g=r(?).Equation (8) can be rewritten in a more customary form (9) r(T) = r?+b á T a with the `Curie-like' second term. The magnetic susceptibility of metals containing paramagnetic centres is described by this formula. In the case of BaCuS3, the following parameters were obtained for Eqn (9): r?=0.136 O cm, a=718.98 O cm K71 and b=178.3 K. Note that the temperature dependence of the electrical resistance described by Eqn (9) is untypical of all conducting materials, and not just of ternary sulfides. Analysis of the published data makes it possible to reveal analogous r(T) dependences for Cu1.5Co1.5S4,125 which has the spinel structure, for Rb4Mo18Se20 126 consisting of Mo6Se8 clusters and for spinel- like CuRh2S4 and CuRh2Se4.127 The dependences reported in these studies are well described by Eqn (9).Apparently, the authors 127 failed to approximate the r(T ) dependence through- out the temperature range because the high-temperature region was described by the linear function (7), the temperature range with T<20 K was described by the function (10) r(T) = r0+ATn with n=3, whereas the temperature range from 20 to 250 K was not interpreted. The authors considered this dependence as the supporting evidence for the applicability of the Wilson model. However, the r(T) dependence for single crystals of BaCuS3 as well as that for chalcogenides studied in Refs 125 ± 127 are described by Eqn (9) throughout the temperature range (from 300 K up to the superconducting transition).This result should give impetus to theoretical studies in the field of solid-state physics. b a 1/2 1/2 0 1/2 1/2 1/2 1/2 0 1/2 1/2 1/2 1/2 1/2 0 0 0 0 0 0 1/2 0 1/2 0 1/2 0 0 1/2 1/2 0 0 1/2 1/2 0 1/2 0 0 0 0 1/2 0 0 0 0 1/2 1/2 1/2 0 a 1/2 0 1/2 1/2 1/2 1/2 0 1/2 0 c Figure 26. Structures of sulfides KCu4S3 (a), K3Cu8S6 (b) and KCu3S2 (c).130 In all the structures, the K+ ions separate layers. V A Starodub Sulfide K3Cu8S6, which was the first representative of ternary copper chalcogenides of the general formulaM3Cu8X6 (M is alkali metal and X=S or Se), was synthesised as early as the mid- nineteenth century.128 Later on, this sulfide, along with other salts described above, was prepared according to the carbonate procedure and studied by the X-ray diffraction method.99 How- ever, the structure of this compounds remained unknown.In 1979, structural studies demonstrated 129 that this sulfide as well as the rubidium analogue Rb3Cu8S6 have layered structures analogous to that of KCu3S2. The structures of three sulfides, viz., KCu3S2 (K3Cu9S6), K3Cu8S6 and KCu4S3 (K2Cu8S6), are shown in Fig. 26 a ± c. This comparison, which is not just formal, confirms that K3Cu8S6 is intermediate between sulfides KCu3S2 and KCu4S3. It was demonstrated 130 that KCu4S3 occurs as a low- temperature phase, which is stable up to 820 8C.At higher temperatures, the KCu3S2 phase is stable. Sulfide K3Cu8S6 occurs as a metastable phase. The optimum conditions for the synthesis of the latter consist in annealing at 875 8C for 1 ± 2 h. n As can be seen from Fig. 26 a, the low-temperature KCu4S3 phase contains close-packed layers in which all copper atoms are tetrahedrally coordinated. In the high-temperature KCu3S2 phase (Fig. 26 c), the layers are corrugated because the copper atoms adopt both the tetrahedral and trigonal coordinations. The structure of K3Cu8S6 is intermediate between the above- described structures. The [Cu4S4] n¡ chains are linked in layers via shared edges of the tetrahedra so that the semisegment of the double layer, which is observed in KCu4S3, is formed.The Rb3Cu8S6 salt has the analogous structure.129 In the same year, selenides Rb3Cu8Se6 and Cs3Cu8Se6 were synthesised by the carbonate procedure from powdered copper and selenium sulfide.131 After annealing at 900 ± 1050 8C (3 ± 4 h), the melt was slowly cooled and washed with water and alcohol to obtain dark anthracite-like crystals. The structure of Rb3Cu8Se6 is similar to those of sulfides K3Cu8S6 and Rb3Cu8S6. Apparently, the cesium salt crystallises in two modifications one of which is analogous to Rb3Cu8Se6, whereas the second modification was not identified. The results of electrophysical measurements for selenides are unavailable in the literature. However, based on their compositions and on the shortest Cu7Cu distances (0.2488 ± 0.2893 nm) in these compounds, it can be said with assurance that these compounds are metals. The electrophysical and magnetic studies of the crystals of K3Cu8S6 were performed.130 The magnetic susceptibility depends only slightly on the temperature at T>180 K.However, this dependence becomes pronounced at temperatures lower than 180 K, which is indicative of a phase transition of the second kind. This transition is reversible. The transition temperature can be determined more accurately from the temperature dependence of the (qw=qT) derivative (150 K). At approximately 50 K, the second transition (of the first kind) with the noticeable hysteresis was observed.The electrical resistance of single crystals of K3Cu8S6 is 561074 O cm; the r(T ) dependence was measured in the range of 350 ± 5 K.131 The resistance decreases linearly as the temper- ature decreases, which is indicative of the metallic behaviour of c 0 1/2 0 1/2 0 1/2 0 1/2 1/2 0 0 1/2 0 1/2 1/2 1/2 1/2 0 0 0 1/2 0 1/2 0 0 1/2 1/2 0 0 1/2 1/2 0 1/2 1/2 1/2 Cu KS 1/2 1/20 1/2 0 0 1/2 1/2 0 0 0 0 1/2 0 0 0 0 a 0 0 1/2 1/2 1/2 0 1/2 1/2 1/2 0 c 1/2 1/21/2 1/2 1/2 0 0 0 1/2 a cTernary and quaternary chalcogenides of Group IB elements single crystals of K3Cu8S6 down to 160 K. At about 150 K, a reversible phase transition occurs, which is manifested in an increase in the resistance in the course of subsequent cooling to 63 K.At this temperature, the resistance decreases sharply and from this point on the r(T) dependence again corresponds to the metallic behaviour in the region of 50 ¡¾ 5 K. The anomaly at about 160 K is also present in the temperature dependence of the thermoelectromotive force. Its plus sign confirms that K3Cu8S6 is a metal with hole conductivity. In a study,131 all these anomalies were attributed to the occurrence of the state with CDWs, which is typical of low-dimensional metals, viz., of quasi-one-dimensional salts {for example, of K2[Pt(CN)4]Cl0.3 . xH2O and NbSe3} and of quasi-two-dimensional salts (TaSe2). However, this state can also occur in three-dimensional metals (for example, in CuV2S4). Sulfide K3Cu8S6 can be considered (as is evident from its structure) as a quasi-two-dimensional metal.However, unlike quasi-two-dimensional metals, such as, for example, NbSe2 or TaS2, the layers in K3Cu8S6 are noticeably anisotropic, as is evident from the mode of linkage of the Cu4S4 columns in layers. The fact that the metallic properties are retained at temperatures lower than the phase transition temperature indicates that this anisotropy is yet inadequate to consider the crystals ofK3Cu8S6 as quasi-one-dimensional. An analogous ansisotropy of the layers occurs, for example, in molybdenum bronzeK0.3MoO3, which is a structurally two-dimensional metal with a essentially monomeric CDWs. X-Ray scattering was studied at different temperatures 132 and it was found that the phase transition at 153 K generates states, which are incommensurate with the lattice period, with the wave vector q a O1 ¢§ dUb , 2 where d is the temperature-dependent variable (d&0.1 at 150 K) and b* is the reciprocal lattice vector. At 55 K, the wave vector of CDW becomes commensurate with the lattice period, which leads to a structural phase transition resulting in the superstructure with the commensurate period at q=(12, 12, 0).Sulfide K3Cu8S6 is the first example of a metal with the hole conductivity in whichCDWs with pronounced p character appear. In conclusion, let us consider a few available examples of tellur- ides of Group IB elements containing discrete Te2¢§ or Te2¢§ 2 ions. Thus, ternary copper tellurides with compositions MCuTe 97 and MCu3Te2 104, 133 (M=Na or K) satisfying Zintl's criterion were described. In 1976, telluride TlCu2Te2 was synthesised from the elements (600 8C, 48 h).134 This telluride can be assigned to the series of MCu2nXn+1 chalcogenides. According to the data of X-ray diffraction analysis, this telluride is isostructural to selenide TlCu2Se2.The electrophysical properties of the telluride under consideration were studied in Ref. 90. The electrical resistance (293 K) of single crystals of TlCu2Te2 is 2.361075 O cm. The r(T) dependence is analogous to that corresponding to metals, viz., the resistance decreases linearly (by a factor of 3.5) as the temperature decreases and reaches 6.561076 O cm at 77 K.Holes act as charge carriers (the results of measurements of the Hall effect). Although the electrical resistance decreases as the samples are cooled, the r300/r77 ratio is rather small. This is attributable to the fact that the concentration of the carriers in the sample decreases substantially (by a factor of 21) in the range of 300 ¡¾ 77 K simultaneously with an increase (by a factor of 76) in the carrier mobility (from 34 cm2 V71 at 300 K to 2600 cm2 V71 at 77 K), i.e., the effect of an increase in the carrier mobility is more pronounced than the effect of a decrease in their concen- tration. This leads to an increase in the conductivity upon cooling. To put it differently, TlCu2Te2 would be more properly charac- terised as a semimetal.Layered mixed-valence tellurideK2Cu5Te5 was synthesised 135 by heating the telluride K2Te with copper and tellurium in an evacuated tube (350 8C, 3 days). The telluride was obtained as black parallelepiped-shaped crystals. The structure of this salt 815 n n consists of the corrugated [Cu5Te5] 2n¢§ layers alternating with the layers of the potassium ions. The [Cu5Te5] 2n¢§ layers are formed by the Cu2Te2 rhombi linked via shared edges so that their structure corresponds to the distorted anti-PbO type. A comparison with the structure of CuTe demonstrates that the addition of 0.4 e per CuTe molecule leads to the cleavage of some of Te7Te bonds in K2Cu5Te5, whereas the addition of one electron (as in NaCuTe) leads to the cleavage of all Te7Te bonds and generates the ideal structure of the anti-PbO type.Therefore, K2Cu5Te5 is inter- mediate between CuTe and NaCuTe. The electrical resistance and the thermoelectromotive force were measured forK2Cu5Te5 and it was revealed that this telluride is a p-type metal in the range of 300 ¡¾ 5 K. The electrical resistance is 6.6761075 O cm at 293 K and reaches 3.161076 O cm at 5 K. The telluride possesses temperature-independent paramag- netism (2.5661074 emu mol71).135 The electronic structure of the K2Cu5Te5 salt was studied by quantum-chemical calculations using the extended HuE ckel method.136 Since the conductivity increases only insignificantly as the temperature decreases in the range of 293 ¡¾ 5 K, the nature of the ground state remains unclear, i.e., the question of whether this telluride is a metal or a semimetal remains open.Another problem studied in Ref. 136 is associated with the possibility of the synthesis of tellurides K3Cu5Te5 and KCu5Te5. In particular, the question was studied of whether the addition of one electron to the [Cu5Te5]27layer or removal of one electron from this layer can lead to the metal ¡¾ nonmetal transition. According to the data obtained, the pre-Fermi region (eF=710.5 eV) from 710.1 to 712.6 eV is formed predom- inantly by the 5p orbitals of tellurium atoms, whereas the region n n from 712.6 to 716 eV is formed by the 3d orbitals of copper atoms. A broad gap extends from710.1 to77.5 eV. However, a narrow peak corresponding to s* antibonding orbitals of the Te(2)7Te(2) bond is present at78.3 eV (Fig.27). The scheme of the structure of the [Cu5Te5] 2n¢§ layer is shown in Fig. 27. It can be seen that this layer contains three types of tellurium atoms. Only the Te(2) atoms are involved in formation of the Te7Te bonds. The populations of the 5py states of the tellurium atoms are approximately equal [1.43, 1.45 and 1.49 e for Te(1), Te(2) and Te(3), respectively], i.e., holes are to a large extent delocalised in the layer. This fact as well as the approximately equal Cu(2)7Te(1) and Cu(3)7Te(3) distances allows one to consider K2Cu5Te5 as a metal. From the results of calculations it follows that the crystals of this telluride should exhibit the noticeable anisotropic conductivity.Unfortunately, it was impossible to measure this anisotropy because of small dimensions of the crystals. The addition of one electron to the [Cu5Te5] 2n¢§ layer, which corresponds to the tranformation to K3Cu5Te5, leads to filling of the p band formed by the 5p orbitals of the tellurium atoms. One would expect that this will result in the metal ¡¾ nonmetal tran- sition. However, it was demonstrated 136 that the insertion of potassium leads not only to an increase in the degree of filling of the p band but also to structural changes. The structure of K2Cu5Te5 contains vacancies sufficient for location of potassium ions above the Te(2)7Te(2) bonds. The insertion of potassium leads to the cleavage of the bonds. In this case, the peak localised at78.3 eV is broadened and merges with the p band.As a result, K3Cu5Te5 retains the metallic properties, possesses two holes per formula unit and is a structural analogue of the NaCuTe salt. 3 1 1 2 2 Cu Te 2 3 2 1 2 1 1 1 2 3 c 2 2 1 1 3 a Figure 27. Scheme of the [Cu5Te5]2n¢§ layer.136 n816 2 Upon the removal of one electron, which corresponds to the transformation to KCu5Te5, the Te(1)7Te(3) bond could be formed, i.e., the Te2¡ ion could be formed if the potassium atoms located above this bond are removed. This would result in the formation of the energy gap, i.e., in the metal ± nonmetal transition. However, according to the results of the study,136 the metallic state is retained in this case because there are three Te7Te bonds per two holes with respect to K2Cu10Te10.As part of continuing studies of reactions proceeding in melts 2 of polytellurides, Zhang et al.137 prepared a new class of polytellurides containing dodecahedral [Cu8Te12] clusters, viz., K4Cu8Te11, M3Cu8Te10 (M=Rb or Cs), M,M02Cu8Te10 (M, M0=K, Rb or Cs) and M2BaCu8Te10 (M=K, Rb or Cs). Telluride K4Cu8Te11 has a three-dimensional framework struc- ture formed by the [Cu8Te12] clusters linked to each other. Each layer contains an encapsulated K+ ion (Fig. 28). The structure contains channels parallel to the b axis. These channels are occupied by potassium ions. The [KCu8Te12]37 clusters are ideal platon bodies (the point symmetry group Th ) formed by the Cu2Te3 planar pentagons linked to each other.Each tellurium atom belongs to a Te2¡ ion so that the clusters are in fact formed by the copper and ditelluride ions [Cu8(Te2)6 ]. KCu Te Figure 28. Structure of the [Cu8Te12] cluster inside which the K+ ion is located.137 n Unlike the K4Cu8Te11 salt containing three-dimensional net- works of the clusters, the Rb3Cu8Te10 and Cs3Cu8Te10 com- pounds have layered structures. The Rb3Cu8Te10 salt consists of [RbCu8Te10] 2n¡ layers alternating with the layers of Rb+ ions. In spite of a substantial structural difference, the major structural blocks in the latter salt are identical to those observed in K4Cu8Te11. The layers are formed by the [Cu8Te12] pentagonal dodecahedra linked to each other.The Rb+ ions are encapsulated into these layers. The structure of the cesium salt is analogous to that of the rubidium salt. The structural studies of the salts with the mixed composition demonstrated that incapsulation is favourable for ions of smaller radii. Thus, in the KCs2Cu8Te10 and KRb2Cu8Te10 salts, the potassium ions are located inside the dodecahedra. The K2BaCu8Te10 and Rb2BaCu8Te10 salts are isomorphous with the Rb3Cu8Te10 salt. The centre of the dodecahedral cavity is occupied by the Ba2+ ion, whose ionic radius is close to that of the K+ ion. In the case of the potassium salt, the K+ ions are located between the layers, which implies that encapsulation of the ion with the larger charge is more favourable in the case of the equal ionic radii.Zhang et al.137 studied the electrophysical and optical proper- ties of the tellurides synthesised. According to the structural data, the formula of the K4Cu8Te11 salt can be written as K4Cu8(Te2)5Te. Taking the charges of the ditelluride and tellur- ide ions to be equal to72 and the charge of the copper ions to be equal to +1, it can be concluded that this salt satisfies Zintl's criterion and, consequently, should exhibit semiconducting prop- erties. This suggestion is also confirmed by the fact that the electrical resistance of the crystals of K4Cu8Te11 (293 K) is rather small (6.2561073 O cm) but it increases according to the activa- tion law as the temperature decreases. The M3Cu8Te10 and MM02Cu8Te10 salts (which do not meet Zintl's criterion) contain the [Cu8(Te2)4Te2]37 anions and, con- sequently, these salts are mixed-valence compounds. According to V A Starodub the results of quantum-chemical calculations,137 the degree of oxidation of copper in these compounds is equal to +1 (the 3d orbitals of the copper atoms are located substantially lower than the 5p orbitals of the tellurium atoms).Therefore, the p band formed by the orbitals of the tellurium atoms contains one hole per formula unit. It is believed that these salts exhibit metallic properties. The data of electrophysical measurements confirmed that the Rb3Cu8Te10, Cs3Cu8Te10 and KRb2Cu8Te10 salts exhibit metallic properties in the range of 5 ± 300 K. Thus, the specific resistance of the samples of Cs3Cu8Te10 decreases by almost three orders of magnitude from 1.161073 O cm at 293 K to 1.261076 O cm at 5 K.The r5 value for this salt is the smallest among all chalcogenides based on copper. The thermoelectromotive forces of the M3Cu8Te10 salts are positive and increase linearly with temperature in the range of 90 ± 300 K, which indicates that these compounds are metals with hole conductivity. The magnetic properties of these salts are also typical of metals. These salts exhibit temperature-independent paramagnetism. The M2BaCu8Te10 salts satisfying Zintl's criterion should be analogous to K4Cu8Te11. It was found that K2BaCu8Te10 and Rb2BaCu8Te10 possess the low electrical resistances at 293 K. Their resistances increase only slightly as the temperature decreases down to *150 K. Below 150 K, the electrical resist- ance increases more rapidly.This means that either these salts are degenerate semiconductors or they have a very narrow forbidden band. The optical measurements in the visible and UV regions revealed no gaps. However, the gap is observed in the IR region and it corresponds to 0.1 eV for the Rb2BaCu8Te10 salt. The barium salts exhibit the temperature-independent paramagnetism typical of dielectrics. The scarce data on copper tellurides allow one to make n definite conclusions. Thus, the Te . . .Te contacts, which are shortened compared to the sum of the van der Walls radii, were found in tellurides Na6Te(Te5), Na2Te6, M2Te3 (M=K, Rb or Cs), M2Te5 (M=Rb or Cs) and CsTe4.138 However, these contacts are absent in tellurides containing bulky cations, for example, in (Ph4P)2Te4 .2MeOH and Me4NMTe4 (M=Cu or Ag).Apparently, cations characterised by a high charge to radius ratio assist in decreasing Coulomb repulsions between polytellur- ide ions and in reducing the possibility of Te . . .Te interactions. In this connection, melts of polytellurides containing barium ions along with alkali metal cations were studied further. Thus, isomorphous polytellurides NaBa6Cu3Te14 and (K0.60Ba0.40). .Ba6Cu2.58Te14 were isolated.138 The structure of the NaBa6Cu3Te14 salt is based on the [Cu3Te3(Te3)3]97 clusters of a new type. These clusters are six-membered rings formed by three copper atoms and three tellurium atoms.Each copper atom is coordinated in a bidentate fashion to the Te2¡ 3 ion, thus acquiring trigonal coordination (Fig. 29). Hence, the cluster ions each contain one six-membered ring and three four-membered rings. The [Cu3Te3(Te3)3]97 clusters are packed in stacks. The Na+ ions are encapsulated into the cavities between these stacks. The composition of the stacks can be represented by the formula [NaCu3Te3(Te3)3] 8n¡. Each stack is surrounded by six stacks formed by the Ba2+ ions. Each telluride ion is surrounded by six Ba2+ ions, thus adopting a trigonal-prismatic coordination environment. Therefore, the structure of this telluride is best described by the formula Ba6[NaCu3Te3(Te3)3]Te2, which satis- fies Zintl's criterion. Because of this and taking into account the existence of ionic Te .. . Te contacts, which are shortened com- pared to the sum of the van der Waals radii, this telluride would be expected to be a narrow-band semiconductor or a semimetal. This conclusion was confirmed by both electrophysical and optical measurements. Thus, rRT is*10 O cm, increases accord- ing to the activation law as the temperature decreases and reaches 2.56103 O cm at 20 K, which corresponds to the width of the forbidden band of about 0.01 eV and to the absorption in the far- IR region (at about 40 cm71). According to the data of measure-Ternary and quaternary chalcogenides of Group IB elements Te(4) Na b Te(1) Ba CuTe(2) a Te(3) Figure 29. Packing of NaBa6Cu3Te14 along the c axis.ments of the thermoelectromotive force, telluride NaBa6Cu3Te14 is a p-type semimetal (the Seebeck coefficient is*100 mV K71). The structure of telluride (K0.60Ba0.40)Ba6Cu2.58Te14 is very similar to that described above, with the difference that the cavities in the [Cu3Te3(Te3)3]97 stacks are statistically occupied by the K+ and Ba2+ ions. The retention of the electroneutrality is provided by the presence of an equivalent number of vacancies in positions occupied by the copper atoms. This telluride is a p-type semiconductor. Its rRT is 105 O cm, increases as the temperature decreases and reaches 1012 O cm at 130 K, which corresponds to the width of the forbidden band of 0.69 eV. Both tellurides possess diamagnetic properties.Telluride K0.33Ba0.67AgTe2 was described.139 Unlike analo- gous tellurides based on copper, this telluride has a layered structure (Fig. 30), which is characterised by the presence of two type of monolayers separated by the K+ and Ba2+ ions. One of the monolayers is a corrugated [Ag2Te2]27 layer and the second monolayer is a planar-square layer formed by the tellurium atoms. The degree of oxidation of the tellurium ions in the [Te2]4/37 monolayer is 72/3 on the assumption that the degrees of oxidation of the silver and tellurium ions in the [Ag2Te2]27 layer are +1 and72, respectively. It should be noted that the existence of the ideal planar-square layers formed only by the tellurium atoms is the unique peculiarity of telluride KBa2[Ag3Te3]Te3. The crystals of the RbTe6 and Cs3Te22 salts 140 also contain planar layers of tellurium atoms.However, these layers can be considered as derivatives of the structure shown in Fig. 30 in which 2/5 of all tellurium atoms are removed. The double planar layers of tellurium atoms were found in the unusual (Te2)2(I2) compound,141 which can be considered as a product of intercalation of tellurium by iodine. a b K, Ba Te Ag c Figure 30. Structure of K0.33Ba0.67AgTe2; (a) the unit cell; (b) the [Ag2Te2]2n¡¦ layers; (c) the [Te2]4n¡¦=3 layers.139 n n 817 The covalent interactions between the [Ag2Te2] and [Te2] layers are negligible. According to the quantum-chemical calcu- lations within the framework of the extended Hu�� ckel method, the Fermi level is located within the band formed by atomic orbitals of the tellurium atoms of the [Te2] layer.This corresponds to a gap in the band structure of the [Ag2Te2] layers. Consequently, this compound can exhibit metallic properties due to the planar- square [Te2] layers. The measurements of the electrical resistance and the thermo- electromotive force demonstrated that this telluride exhibits semiconducting properties. Thus, rRT is *5 O cm and increases according to the activation law as the temperature decreases. The width of the forbidden band, which was determined from the temperature dependence of r, is 0.06 eV to which the optical absorption at energies lower than 0.15 eV corresponds. The semiconducting properties of K0.33Ba0.67AgTe2 are also con- firmed by the results of measurements of the thermoelectromo- tive force.It has a positive value and the Seebeck coefficient is 200 ¡À 300 mV K71. This discrepancy between theory and experiment may be accounted for by different reasons. Thus, it is quite apparent that nonrelativistic calculations for a salt containing the heavy atom, such as tellurium, are inadequate. Another possible reason is that the mixed-valence state of the silver ions in the telluride is absent because silver, unlike copper, obviously cannot exist in chalcogenides in the form of Ag2+. The structural distortion of the [Te2]4/37 layers due to the appearance of CDWs may be also responsible for the manifestation of semiconducting properties.The latter possibility was considered in Ref. 139. Since X-ray diffraction studies did not reveal structural distortions resulting in the deviation from the ideal I4/mmm symmetry, the authors used the electron diffraction method, which is more sensitive to weak structural distortions of the [Te2]4/37 layers. The electron diffrac- tion data are unambiguously indicative of the presence of two superlattices. In one type of domain, distortions generate an incommensurate superlattice with asuper=0.352asub; bsuper= bsub, csuper=csub. The domains of the second type contain a commensurate superstructure with asuper=3asub; bsuper=3bsub, csuper=csub. The results obtained confirm the suggestion that CDWs are responsible for localisation of electrons in the [Te2]4/37 layers. The formal charge of the tellurium ions in the layers is equal to72/3.Hence, the appearance of CDWs is associated with the formation of the Te2¡¦ 3 ions in the [Te2] layers in the case of the occurrence of the superstructures 363 and 361 because the charge, which fluctuates in the ideal square lattice, is localised on going to the tritelluride ion. However, the available data on mixed copper tellurides, while rather scarce, demonstrate that tellurides differ substantially from sulfides and selenides. Although simple tellurides with composi- tions identical to those of sulfides and selenides exist, the majority of tellurides have no analogues among the corresponding sulfides and selenides.The same is true for their structures. Thus, the structures of CuS and CuSe are identical, whereas the CuTe structure is quite different. Unlike the widespread copper poly- sulfides, examples of the presence of polytelluride ions in copper salts are few in number. Thus, the isostructural a-KCuX4 and M3Cu8X6 salts (X=S or Se) are known, whereas the correspond- ing tellurides do not exist. At the same time, there are no sulfide analogues of K2Cu5Te5 or K4Cu8Te11. 3 As was mentioned in Chapter III, attempts to prepare mixed sulfide ¡À telluride copper complexes from a melt of M2SxTe17x afforded salts containing sulfide and trithiotelluride ions. Later on, the MM0TeS3 salts (M=K, Rb or Cs; M0=Cu or Ag) containing only TeS2¡¦ ions, viz., CsCuTeS3, a- and b-KAgTeS3, RbCuTeS3, RbAgTeS3 and CsAgTeS3, were isolated.142 The last- mentioned four salts are quasi-two-dimensional and isomor- phous.These salts contain the anionic [M0TeS3] n¡¦ n layers between which the alkali metal ions are located (Fig. 31). The Cu+ions are coordinated by four sulfur atoms of the TeS2¡¦ 3 ions, thus adopting a distorted tetrahedral coordination. Each TeS2¡¦ ion acts as a 3818 Rb Cu Te S a b Figure 31. Two-dimensional structure of RbCuTeS3 along the c axis. The rubidium ions form monoatomic layers between the [CuTeS3]n¡ n layers.142 bridge and links four Cu+ ions. Two sulfur atoms each link one Cu+ ion and the third sulfur atom acts as a bridge between two Cu+ ions. 3 The a-KAgTeS3 compound has a centrosymmetrical layered structure.The most substantial difference in the structures of the a and b phases consists in the difference in the arrangement of the pyramidal TeS2¡ ions. In the structure of a-KAgTeS3, the tellurium atoms are located on the surface of the layers so that their lone electron pairs are directed away from the layer. Neither silver ions nor alkali metals interact with these lone pairs. On the contrary, the K+ ions are surrounded by seven sulfur atoms. 3 The crystals of CsCuTeS3 belong to the cubic system. In the crystal structure, the CuTeS¡3 ions are linked to each other to form a three-dimensional non-centrosymmetrical chain. The Cuá ions have a planar-trigonal coordination. Each CuS3 triangle is linked to three adjacent TeS2¡ pyramids via shared vertices and vice versa. The crystallographically equivalent sulfur atoms are linked through the copper and tellurium atoms at an angle of*90 8.All compounds of the trithiotellurides series satisfy Zintl's criterion, which accounts for their dielectric properties. The optical measurements revealed the presence of the forbidden band in the region from 1.95 eV (RbCuTeS3) to 2.2 eV (a-KAgTeS3). Kanatzidis 140 summarised a rare experience in the field of inorganic chemistry of tellurium and noted that this division of synthetic and structural chemistry of chalcogenides is character- ised by the fact that both the compositions and structures of tellurides are unpredictable. However, even from the available data it is clear that ternary copper and silver tellurides are very promising material because they show diverse physical properties (from dielectric to semiconducting), possess interesting magnetic properties and exhibit transitions to states with CDWs.These transitions, which attracted the attention of ists only early in the 1970s, comprise a new exciting field of solid-state chemistry because they are closely associated with the problem of the variable valence and charge ordering in crystals. The detailed consideration of this problem is beyond the scope of this review. Let us only mention the small but rather informative review devoted to this problem. 143 V. Conclusion As is evident from the present review, the chemistry of ternary copper chalcogenides, which has progressed in the last decade, shows promise from the viewpoint of the development of new materials with a broad spectrum of electrophysical properties, including new high-temperature superconductors. In essence, only ternary copper and silver chalcogenides containing ions of alkali or alkaline-earth metals as well as thallium have been systemati- cally studied.Data on physicochemical properties of ternary Ln7Cu7X chalcogenides (Ln is a rare-earth element), which are analogues of high-temperature superconductors, are scarce. This field of research is in its infancy. In 1994, the first mixed copper chalcogenides containing cerium, viz., KCuCe2S6 and V A Starodub KCu2CeS4, were synthesised.144, 145 Both sulfides were prepared from K2S7Cu7Ce7S melts and have layered structures.Their compositions do not satisfy Zintl's criterion (it is highly improb- able that cerium in sulfides has the degree of oxidation of +4). However, both salts exhibit dielectric properties; the width of the forbidden band in KCuCe2S6 is 1.8 eV. In K2Cu2CeS4, the gap was not found in the range of 250 ± 2500 nm. The conductivity of this salt at 293 K (sRT*1073 O71 cm71) is five orders of magnitude larger than that of the KCuCe2S6 salt. However, the temperature dependence of the electrical resistance is typical of semiconductors. The thermoelectromotive force of crystals of the K2Cu2CeS4 salt has a plus sign and the thermoelectromotive force decreases as the temperature decreases, which may be indicative of the metallic state.However, the thermoelectromotive force is much larger than the expected value for a metal. Based on these facts as well as on the results of magnetic measurements, the authors 144 proposed the following formulas for mixed sulfides: K+Cu+(Ce3+)2(S27)2(S2¡ 2 )2 for KCuCe2S6 and Ká2 (Cu+)2Ce3+(S27)3(S¡) for K2Cu2CeS4. The magnetic susceptibility of the KCuCe2S6 salt obeys the Curie ± Weiss law (the Weiss constant y=767 K), meff=3.1mB, which is close to the purely spin value of 2.83mB for two unpaired electrons (one electron from each Ce3+ ions). A more complex situation arises with sulfide K2Cu2CeS4. For this compound, either the Ká2 Cu+Cu2+Ce3+S24 ¡ structure with the mixed-valence state for copper or the above-mentioned structure containing `holes' in the p band formed by the 3p orbitals of the sulfur atoms has to be postulated.According to the approach accepted in our works, the mixed-valence states both of sulfur and copper atoms should be considered, as in the case of the above-considered chalcogenides, taking into account substantially covalent interactions between copper and sulfur atoms. However, the absence of metallic properties in sulfide K2Cu2CeS4 seems to be strange. Sutorik et al.144 attributed this situation to the fact that the peculiarities of the crystal structure of this compound lead to the formation of a very narrow band, which, in turn, results in the formation of low-lability polarons in the presence of highly charged ions, such as Ce3+ ions in the case under consideration.In the case of sulfides KCuCe2S6 and K2Cu2CeS4, the combination of rare-earth elements and the alkali metal does not lead to an increase in the degree of oxidation of copper, unlike high-temperature superconductors. The reason is that the electro- negativity decreases in going from oxygen to sulfur. Besides, as is evident from the present review, no copper sulfides (neither binary nor mixed) containing only Cu2+ ions are known. Two years later,146 sulfide KCuEu2S6 was synthesised. This salt contains double layers formed by the EuS8 polyhedra (bicapped trigonal prisms) and the CuS4 polyhedra (tetrahedra) linked to each other. The authors did not perform electrophysical measurements. However, judging from the dark-brown colour of the crystals, it is believed that this sulfide is a narrow-band semiconductor.The manifestation of the effect of the chalcogen content on the physical properties of mixed copper oxides and sulfides is also different. It is known 10 that in oxides, such as, for example, La2CuO47x or YBa2Cu3O6.5+x, the degree of oxidation of copper increases as the oxygen content increases. Sulfide analogues of oxides La2CuO47x and Y2Cu2O5 were synthesised 147 with the aim of elucidating the effect of the anionic subsystem on the electrophysical properties of chalcogenides based on copper. Sulfides La2CuS4+x and Y2Cu2S5+x were prepared by fusing a mixture of Ln2S3, CuS and S at 800 ± 900 8C.Their individuality was confirmed by X-ray diffraction studies. The X-ray photo- electron spectra have doublets in the region of the ES(2p) values. One of these doublets at 161.8 eV corresponds to the S2¡ ions 82 and the second doublet at 162.8 eV can be assigned to the S2¡ 2 ions. In the ECu(2p3/2) region, the lines at 932.6 eV correspond to the Cu+ ions. According to the data of X-ray photoelectron spectro- scopy, a decrease in the sulfur content in samples upon annealingTernary and quaternary chalcogenides of Group IB elements in vacuo occurs as follows: lines at 934.5 eV, which correspond to the Cu2�¢ ions, appear and intensities of the lines corresponding to the disulfide ions decrease. Samples of La2CuS4+x saturated with sulfur have the compo- sition La2CuS5.Based on the results of X-ray photoelectron spectroscopy, their structures can be represented by the formula La3�¢ 2 Cu�¢(S22 ¡¦)1:5(S2¡¦)2 . Sulfide La2CuS5 exhibits dielectric prop- erties. When sulfur is removed until the sample acquires the composition La2CuS4, the degree of oxidation of copper increases and the content of disulfide ions decreases. The appear- ance of the mixed-valence state of copper leads to the nonmetal ¡À metal transition. The temperature dependence of the electrical resistance of samples of La2CuS4 is typical of metals. Below 100 K, the temperature dependence of the resistance is nonlinear. 2 Sulfide Y2Cu2S5+x, which is maximum saturated with sulfur, has the composition Y2Cu2S5.Its X-ray photoelectron spectrum has lines at 932.8 and 934.5 eV belonging to the Cu�¢ and Cu2�¢ ions.82 Sulfur is present as the S2¡¦ and S2¡¦ ions [ES(2p)=161.8 and 162.8 eV, respectively]. The r(T) dependence is typical of metals (Fig. 32). When sulfur is removed until the compound acquires the composition Y2Cu2S4, this dependence becomes more complex and goes through a weak maximum at T*240 K. According to the results of X-ray photoelectron spectroscopy, this sample contains sulfur only as sulfide ions and the Cu�¢ :Cu2�¢ ratio decreases from 3 : 1 in the initial sample to 1 : 1 in the final product.148 R/R300 1 1 2 0.5 300 T /K 100 200 Figure 32. Temperature dependence of the electrical resistance of samples of Y2Cu2S4 (1) and Y2Cu2S5 (2).148 Hence, in the case of sulfides, the situation is opposite to that observed in oxides.The possible reasons for this fact are the lower electronegativity of sulfur responsible for reducing properties of the sulfide ion, 2S2¡¦ 72e S22 ¡¦ as well as the reaction S + S2¡¦ S2¡¦ 2 , in the course of which excess sulfur is bound by sulfide ions rather than oxidises Cu�¢ ions. Apparently, these processes are hindered when the sulfur content is decreased and, as a consequence, the degree of oxidation of sulfur decreases as its content decreases. The review was written with the financial support of the ISSEP Foundation (Grant No APU073103). References 1. J G Bednorz, K A Mu�� ller Z. Phys. B64 187 (1986) 2.C W Chu, L Gao, F Chen, Z J Huang, R L Meng, Y Y Xue Lett. Na3. M Lagues, Xiao Ming Xie, H Tebbi, Xiang Zhen Xu, V Mairet, Ch Hatterer, C F Beuran, C Deville-Cavellin Science 262 1850 (1993) 4. R S Burrows, D H McDaniel Angew. Chem., Int. Ed. Engl. 33 897 (1994) 5. P W Anderson Science 235 1196 (1987) 819 6. M J S Dewar Angew. Chem, Int. Ed. Engl. 26 1273 (1987) 7. W Urland Angew. Chem., Int. Ed. Engl. 26 1275 (1987) 8. S R Ovshinsky Chem. Phys. Lett. 195 455 (1992) 9. A Villesuzanne, M Pouchard C. R. Hebd. Seances Acad. Sci., Ser. 2 322 155 (1996) 10. V V Nemoshkalenko, V N Uvarov, A A Ereshchenko, S K Tolpygo, V A Starodub,M A Obolenskii, A V Zhalko-Titarenko, L M Sheludchenko Sverkhprovod. Fiz.Khim. Tekhn. 3 856 (1990) 11. I Panas, R Gatt Chem. Phys. Lett. 270 178 (1997) 12. A Simon Angew. Chem. 109 1872 (1997) 13. I Panas, R Gatt Chem. Phys. Lett. 266 410 (1997) 14. R Cloots, A Rulmont, P A Godelaine, C Hannau, H W Vanderschueren, M Auslos Solid State Commun. 79 615 (1991) 15. C N R Rao, J Gopalakrishnan New Directions in Solid State Chemistry (Cambridge: Cambridge University Press, 1986) 16. B W Roberts Properties of Selected Superconductive Materials. 1978. Supplement National Bureau of Standards (Ed. BWRoberts) (Washington, DC: US Government Printing Office, 1978) p. 983 17. A F Wells Structural Inorganic Chemistry (Oxford: Clarendon Press, 1986) 18. S Robashkiewicz, M Zimpel, I Onyshriewicz, M A Obolensky Acta Magnetica 8 58 (1991) 19. M G Kollin, F Rouyer, J Loriers C.R. Hebd. Seances Acad. Sci., Ser. C 266 689 (1968) 20. M Julien-Pouzol, M Guittard, M A Mazurier Acad. Sci., Paris 271 C1317 (1970) 21. M Julien-Pouzol, M Guittard Ann. Chim. 7 253 (1972) 22. O V Andreev Zh. Neorg. Khim. 33 991 (1988) a 23. O V Andreev, N N Parshukov Zh. Neorg. Khim. 36 2106 (1991) a 24. WJ Zhu, Y Z Huang, C Dong, Z X Zhao Mater. Res. Bull. 29 143 (1994) 25. B A Popovkin Zh. Neorg. Khim. 38 2020 (1993) a 26. Yu M Yarmoshenko, V A Trofimova, V E Dolgih, M A Korotin, E Z Kurmaev, J A Aguiar, J M Ferreira, A C Pavao J. Phys., Condens. Matter 7 213 (1995) 27. D N Argiriou, J D Jorgensen, R L Hitterman, Z Hiroi, N Kobayashi, M Takano Phys. Rev. B, Condens. Matter 51 8434 (1995) 28.W A Starodub,M A Obolensky, Kh B Chashka, K V Krikunov, Yu B Poltoratsky, S Robashkiewicz, in Proceedings of International Conference on High Temperature Superconductivity and Localisation Phenomena (Abstracts of Reports), Moscow, 1991 p. C14 29. MDraganjac, T B Rauchfuss Angew. Chem., Int. Ed. Engl. 24 742 (1985) 30. A Mu�� ller, U Schimanski Inorg. Chim. Acta 77 L187 (1983) 31. A Mu�� ller, F-W Baumann, H Bo�� gge,M Ro�� mer, E Krickemeyer, 32. A Mu�� ller, M Ro�� mer, E Krickemeyer, F-W Baumann, K Schmitz 33. A Mu�� ller, M Ro�� mer, E Krickemeyer, H Bo�� gge Naturwissenschaften K Schmitz Angew. Chem. 96 607 (1984) Inorg. Chim. Acta 89 L7 (1984) 71 43 (1984) 34. A Mu�� ller, M Ro�� mer, H Bo�� gge, E Krickemeyer, K Schmitz Inorg. Chim.Acta 85 L39 (1984) 35. G Henkel, P Betz, B Krebs J. Chem. Soc., Chem. Commun. 314 (1984) 36. A Mu�� ller, M Ro�� mer, H Bo�� gge, E Krickemeyer, D Bergmann J. Chem. Soc., Chem. Commun. 348 (1984) 37. P Betz, B Krebs,G Henkel Angew. Chem., Int. Ed. Engl. 23 311 (1984) 38. G Marbach, J Stra�� hle Angew. Chem., Int. Ed. Engl. 23 715 (1984) 39. S-PHuang,MGKanatzidis Angew. Chem., Int. Ed. Engl. 31 787 (1992) 40. H Krautscheid,D Fenske,G Baum,M Semmelmann Angew. Chem., Int. Ed. Engl. 32 1303 (1993) 41. M G Kanatzidis, S-P Huang Inorg. Chem. 28 4667 (1989) 42. S-P Huang, M G Kanatzidis Inorg. Chem. 30 3572 (1991) 43. K-W Kim, M G Kanatzidis J. Am. Chem. Soc. 115 5871 (1993) 44. A Mu�� ller, E Krickemeyer,M Zimmermann,M Ro�� mer, H Bo�� gge, M Penk, K Schmitz Inorg.Chim. Acta 90 L69 (1984) 45. S-P Huang, M G Kanatzidis Inorg. Chem. 30 1455 (1991) 46. JCusick,ML Scudder,DGCraig, IGDance Polyhedron 8 1139 (1989) 47. U Mu�� ller, M L Ha-Eierdanz, G Kriuter, K Z Dehnicke Z. Naturforsch., B Chem. Sci., 45 1128 (1990) 48. G Marbach, J Stra�� hle Angew. Chem., Int. Ed. Engl. 23 246 (1984) 49. R C Haushalter Angew. Chem., Int. Ed. Engl. 24 432 (1985) 50. C J Warren, D M Ho, A B Bocarsly, R C Haushalter J. Am. Chem. Soc. 115 6416 (1993) 51. R C Haushalter Inorg. Chim. Acta 102 L37 (1985)820 52. MA Ansari, J C Bollinger, J A Ibers J. Am. Chem. Soc. 115 3838 (1993) 53. J M McConnachie,M A Ansari, J C Bollinger, R J Salm, J A Ib- ers Inorg.Chem. 32 3201 (1993) 54. M A Ansari, J M McConnachie, J A Ibers Acc. Chem. Res. 26 574 (1993) 55. R D Harcourt J. Inorg. Nucl. Chem. 39 237 (1977) 56. M A Ansari, J C Bollinger, J A Ibers Inorg. Chem. 32 1746 (1993) 57. K A Hofmann, F Hochtlen Ber. Dtsch. Chem. Ges. 36 3090 (1903) 58. M G Kanatzidis, Y Park J. Am. Chem. Soc. 111 3767 (1989) 59. Ch Z Burschka Z. Naturforsch., B Chem. Sci. 35 1511 (1980) 60. G Gattow, T Dingeldein Z. Anorg. Allg. Chem. 596 73 (1991) 61. GKiel,GGattow,TDingeldeinZ. Anorg. Allg.Chem. 596 111 (1991) 62. R M H Banda,D C Craig, J G Dance,M L Scudder Polyhedron 8 2379 (1989) 63. M G Kanatzidis, S-P Huang Angew. Chem., Int. Ed. Engl. 28 1556 (1989) 64. T J McCarthy, X Zhang,MG Kanatzidis Inorg. Chem. 32 2944 (1993) 65.M G Kanatzidis, S-P Huang J. Am. Chem. Soc. 111 760 (1989) 66. P T Wood, W T Pennington, J W Kolis J. Am. Chem. Soc. 114 9233 (1992) 67. L C Roof, J W Kolis Chem. Rev. 93 1037 (1993) 68. P T Wood, W T Pennington, J W Kolis J. Chem. Soc., Chem. Commun. 235 (1993) 69. Y Park,MG Kanatzidis Angew. Chem., Int. Ed. Engl., 29 914 (1990) 70. K O Klepp, W Bronger J. Less-Common Met. 127 65 (1987) 71. P Pekka Kem.-Kemi 8 498 (1981) 72. DE Bacelo, SDHuang,YIshikawa Chem. Phys. Lett. 277 215 (1997) 73. S Loeken, C Felser,W Tremel J. Chem. Soc., Chem. Commun. 1603 (1998) 74. Pyykko Pekka, J Li, N Runeberg Chem. Phys. Lett. 218 133 (1994) 75. M G Kanatzidis Chem. Matter. 2 353 (1990) 76. C C Raymond, P K Dorhout Inorg. Chem. 35 5634 (1996) 77.J C Folmer, F Jellinek J. Less-Common Met. 76 153 (1980) 78. H J Gotsis, A C Barnes, P Strange J. Phys., Condens. Matter 4 10 461 (1992) 79. W Liang,M H Whangbo Solid State Commun. 85 405 (1993) 80. G V Vajenine, R Hoffmann Inorg. Chem. 35 451 (1996) 81. Georg Brauer (Ed.) Handbuch der Preparativen Anorganischen Chemie Bd. 2 (Stuttdart: Ferdinand Enke Verlag, 1978) 82. V I Nefedov Rentgenoelektronnaya Spektroskopiya Khimicheskikh Soedinenii (X-Ray Electron Spectroscopy of Chemical Compounds) (Moscow: Khimiya, 1984) 83. A Ouammou, M Mouallem-Bahout, O Pena, J F Halet, J Y Saillard, C Carel J. Solid State Chem. 117 73 (1995) 84. K Klepp, H Boller Monatsh. Chem. 109 1049 (1978) 85. R Berger J. Less-Common Met. 147 141 (1989) 86. K Klepp, H Boller, H VoÈ llenkle Monatsh. Chem. 111 727 (1980) 87. D B Brown, J A Zubieta, P A Vella, J T Wrobleski, T Watt, W E Hatfield, P Day Inorg. Chem. 19 1945 (1980) 88. C Burschka Z. Anorg. Allg. Chem. 463 65 (1980) 89. R Berger, L Eriksson J. Less-Common Met. 161 165 (1990) 90. G Brun, B Gardes, J C Tedenac, A Raymond,M Maurin Mater. Res. Bull. 14 743 (1979) 91. R Berger, C F Van Bruggen J. Less-Common Met. 99 113 (1984) 92. L Karlsson,M P Keane, R Berger J. Less-Common Met. 166 353 (1990) 93. J E Iglesias, K E Pachali, H Steinfink J. Solid State Chem., 9 6 (1974) 94. M Saeki, M Onoda, H Kozaki Mater. Res. Bull. 23 603 (1988) 95. M Onoda,M Saeki Mater. Res. Bull. 24 1337 (1989) 96. M V Savel'eva, I N Trushnikova, A A Kamarzin, V I Alekseev, I A Baidina, S V Borisov, S A Gromilov, A G Blinov Izv. Sib. Otd. Akad. Nauk, Ser. Khim. Nauk (3) 157 (1989) 97. GSavelsberg,HSchaefer Z. Naturforsch., B Chem. Sci. 33 370 (1978) 98. GSavelsberg,HSchaefer Z. Naturforsch., B Chem. Sci. 33 711 (1978) 99. W RuÈ dorff, H G Schwarz,M Walter Z. Anorg. Allg. Chem. 269 141 (1952) 100. B P Ghosh,MChaudhury,KNag J. Solid State Chem. 47 307 (1983) 101. N S Hartig, P K Dorhout, S M Miller J. Solid State Chem. 113 88 (1994) 102. W Bronger Angew. Chem., Int. Ed. Engl. 20 52 (1981) 103. F Viola, R Schollhorn J. Chem. Soc., Chem. Commun. 907 (1992) V A Starodub 104. G Savelsberg, H Schaefer Mater. Res. Bull. 16 1291 (1981) 105. H Effenberger, F Pertlik Monatsh. Chem. 116 921 (1985) 106. J E Iglesias, K E Pachali, H Steinlg. Chem. 623 1357 (1997) 108. P T Wood,WT Pennington, JWKolis Inorg. Chem. 33 1556 (1994) 109. W Bronger, J Eyck, H Schils J. Less-Common Met. 60 5 (1978) 110. W Bronger, C Burschka Z. Anorg. Allg. Chem. 425 109 (1976) 111. C Burschka,W Bronger Z. Anorg. Allg. Chem. 430 61 (1977) 112. C Burschka,W Bronger Z. Naturforsch., B Chem. Sci. 32 11 (1077) 113. K Klepp, K Yvon Acta Crystallogr., Sect. B 36 2389 (1980) 114. R Berger J. Solid State Chem. 70 65 (1987) 115. R Berger Chem. Scr. 28 41 (1978) 116. R Berger, R V Bucur Mater. Res. Bull. 27 439 (1992) 117. R Berger, R Dronkowski, L Noren J. Solid State Chem. 112 120 (1994) 118. I Jacyna-Onyszkiewicz, M Sidowski,W Starodub, A Ratuszna, S Robaszkiewicz, in Proceedings of the European Conference on Physics of Magnetism, Wroclaw, 1996 p. 105 119. C Burschka Z. Naturforsch., B Chem. Sci. 34 396 (1979) 120. Z Perlinski, D B Brown, T Watt, W E Hatfield, P Day Inorg. Chem. 211752 (1982) 121. V A Starodub, I V Krivoshei Usp. Khim. 51 764 (1982) [Russ. Chem. Rev. 51 439 (1982)] 122. M-H Whangbo, E Canadell Inorg. Chem. 29 1395 (1990) 123. X Zhang,M G Kanatzidis J. Am. Chem. Soc. 118 693 (1996) 124. I Jacyna-Onyszkiewicz, M Sidowski,W Starodub, A Ratushna, S Robaszkiewicz, in Proceedings of the European Conference on Physics of Magnetism, Wroclaw, 1996 p. 79 125. Y Furukawa, S Wada, K Miyatani, T Tanaka,M Fukugauchi, M Ishikawa Phys. Rev. B, Condens. Matter 51 6159 (1995) 126. R Brusetti, O Laborde, A Sulpice, R Calemczuk, M Potel, P Cougeon Phys. Rev. B, Condens. Matter 52 4481 (1995) 127. T Hagino,Y Seki,N Wada, S Tsuji, T Shirane, Ken-ichi Kumagai, S Nagata Phys. Rev. B, Condens. Matter 51 12673 (1995) 128. R Schneider J. Prakt. Chem. 108 16 (1865) 129. C Burschka Z. Naturforsch., B Chem. Sci. 34 675 (1979) 130. L W Ter Haar, F J DiSalvo, H E Bair, R M Fleming, J V Waszczak Phys. Rev. B, Condens. Matter 35 1932 (1987) 131. H Schils,W Bronger Z. Anorg. Allg. Chem. 456 187 (1979) 132. R M Fleming, L WTer Haar, F J DiSalvo Phys. Rev. B, Condens. Matter 35 5388 (1987) 133. K O Klepp J. Less-Common Met. 128 79 (1987) 134. B Gardes, S Peytavin, M Maurin C. R. Hebd. Seances Acad. Sci., Ser. C 283 529 (1976) 135. Y Park, D C Degroot, J Schindler, C R Kannewurf, M G Kanatzidis Angew. Chem., Int. Ed. Engl. 30 1325 (1991) 136. S Seong, T A Albright, X Zhang,M Kanatzidis J. Am. Chem. Soc. 116 7287 (1994) 137. X Zhang, Y Park, T Hogan, J L Schindler, C R Kannewurf, S Seong, T Albright,M G Kanatzidis J. Am. Chem. Soc. 117 10 300 (1995) 138. X Zhang, J L Schindler, T Hogan, J Albritton-Thomas, C R Kannewurf,M G Kanatzidis Angew. Chem., Int. Ed. Engl. 34 68 (1995) 139. X Zhang, J Li, B Foran, S Lee, Hong-You Guo, T Hogan, C R Kannewurf,M G Kanatzidis J. Am. Chem. Soc. 117 10 513 (1995) 140. M G Kanatzidis Angew. Chem., Int. Ed. Engl. 34 2109 (1995) 141. R Kniep, H-J Beister Angew. Chem., Int. Ed. Engl. 24 393 (1985) 142. X Zhang,M G Kanatzidis J. Am. Chem. Soc. 116 1890 (1994) 143. S P Ionov Zh. Fiz. Khim. 56 2385 (1982) b 144. A C Sutorik, J Albritton-Thomas, C R Kannewurf, M G Kanatzidis J. Am. Chem. Soc. 116 7706 (1994) 145. M A Pell, J A Ibers Chem. Ber. 130 1 (1997) 146. W Bensch, P Dmrichen Chem. Ber. 129 1489 (1996) 147. I Jacyna-Onyszkiewicz, M Zimpel, V Starodub, S Robaszkiewicz Mol. Phys. Rep. 12 115 (1995) 148. V A Starodub, Doctoral Thesis in Chemical Sciences, Khar'kov State University, Khar'kov, 1994 a�Russ. J. Inorg. Chem. (Engl. Transl.) b�Russ. J. Phys. Chem. (Engl. Tran

ISSN:0036-021X    

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Russian Chemical Reviews 68 (10) 821 ± 857 (1999) #1999 Russian Academy of Sciences and Turpion Ltd UDC 541(15-64) : 537.5 Ferroelectricity of polymers based on vinylidene fluoride V V Kochervinskii Contents I. Introduction II. General characteristics III. Characteristic features of the polarisation switching processes IV. Curie transition V. The mechanism of the generation of ferroelectricity based on NMR studies VI. The influence of pressure on the characteristics of the ferroelectric transition VII. The influence of the structural features of polymers on their ferroelectric characteristics VIII. Ferroelectricity of ultrathin films IX. Conclusion elucidate the influence of the polymeric state of substance on its electrophysical characteristics and would thus provide the under- standing of the fundamental features of the formation of ferroe- lectricity and piezo- and pyroelectric properties in other polymers, for example, in poly(vinylidene cyanide),4±13 odd nylons,4 ±16 polyacrylonitrile,17, 18 polyurethanes,19, 20 liquid-crystal poly- mers 21 ± 26 and biopolymers.27 PVDF is known to exhibit clear-cut polymorphism.28 Three of the four crystallographic modifications of PVDF usually have non-zero cell dipole moments; therefore, there are prerequisites for the formation of a ferroelectric structure.Crystallites (or their sections) can be compared with domains characterised by a particular value of spontaneous polarisation. II. General characteristics Abstract. Characteristics of the ferroelectricity in polyvinylidene fluoride (PVDF) and copolymers based on vinylidene fluoride (VDF) with different previous thermomechanical histories are analysed.Switching of polarisation and the fluctuation mecha- nism of formation of new nuclei upon the appearance of conformation defects such as kink bonds are considered. The role of the space charge, which stabilises the new polarisation position, is noted. It is shown that the Curie transition can be either a first- or second-order phase transition depending on the copolymer composition. The influence of the morphology of PVDF on some ferroelectric characteristics is found. In partic- ular, the morphology of thick lamellar crystals is favourable for enhanced spontaneous polarisation. It is shown that a ferro- electric structure approaching the texture of an ideal single crystal can be created in Langmuir ± Blodgett films.As the number of unimolecular layers decreases, the concentration of conforma- tional defects in these films increases and the crystallographic order in the lattice is violated. This intensifies the low-temperature Curie transition. The bibliography includes 247 references. I. Introduction The ferroelectric character of PVDF and VDF copolymers is confirmed, among other facts, by the occurrence of dielectric hysteresis for these materials. Figure 1 presents the variation of the electric displacement (D) as a function of the electric field strength E for a uniaxially stretched PVDF film.29 It can be seen that hysteresis is displayed only in fields with a strength greater D /mC m72 80 1234 80 120 E /MV m71 7120 780 Poly(vinylidene fluoride) (PVDF) and copolymers based on vinylidene fluoride (VDF) present interest for both fundamental science and applied research.This is due to the fact that these materials were found to exhibit ferroelectric properties normally atypical of polymers. Owing to the high piezo- and pyro-electric activities found in PVDF films,1 this polymer has been recom- mended in some cases as a competitive material for energy converters of various types.2, 3 The results of experimental and theoretical studies in this field need to be surveyed at intervals for two reasons. Firstly, this could prompt the ways for optimisation of characteristics of the materials studied, which is significant for the solution of applied problems. Secondly, this would help to 740 780 V V Kochervinskii Troitsk Institute for Innovation and Fusion Research, 142092 Troitsk, Moscow Region, Russian Federation.Fax (7-095) 334 57 76. Tel. (7-095) 165 14 74 Figure 1. Electric displacement in a uniaxially stretched PVDF film at 20 8C vs strength of the external electric field.29 Frequency of the electric field 1073 Hz; strength /MV m71: (1) 40, (2) 80, (3) 100, (4) 120. Received 23 November 1998 Uspekhi Khimii 68 (10) 904 ± 943 (1999); translated by Z P Bobkova7240 30 1 2 4 3 4 2 7 0 120 0 Figure 2. Temperature dependences of the electric displacement (a) and the polarisation current density (b) vs the electric field strength in the initial non-polarised PVDF film.29 T /8C: 20 (1), 720 (2), 760 (3), 7100 (4) (a); 20 (1), 0 (2), 720 (3), 740 (4), 760 (5),780 (6), 7100 (7) (b); t is the current time.than 40 MV m71 and that an increase in the field strength results in an increase in both the remanent (for E=0) and spontaneous polarisation, Pr and Ps , respectively. In relatively low fields (80 MV m71), the curve shows double hysteresis. This can be seen in more detail in Fig. 2, which shows the variation of properties of a non-polarised film placed in a 240 MV m71 field for temperatures varying over wide limits. Temperature plays an important role for the manifestation of ferroelectricity in PVDF and in crystallisable VDF copolymers. Only a fraction of the sample bulk is occupied by crystals; the rest (which makes 50% to 60% in PVDF) is filled by a disordered phase. The role of this phase in the formation of ferroelectricity is not entirely clear.The dynamics of chains in the disordered phase can have a substantial influence on the characteristics of the electric hysteresis (see Fig. 2). In temperature regions below the glass transition temperature of PVDF (Tg^740 8C), large- amplitude motions of chain sections are frozen. Some researchers (e.g., Tamura et al.30) have observed even the disappearance of the hysteresis in low fields. However, in high fields, disappearance was not noted but the D± E dependences were characterised by double hysteresis (see Fig. 2a). This can be seen more clearly in the field dependences of the polarisation current density (see Fig. 2b), which have two peaks at low temperatures. The position of one (lower-field) peak remains virtually unchanged, while the other peak shifts to lower fields as the temperature rises.The low-field peak may be related to the response of polar crystals to the application of the field and the peak at higher fields may be due to the response of dipoles in the disordered phase. Evidently, an increase in the temperature is favourable for the activation of the latter process because the second peak markedly shifts to lower fields. In the region below 740 8C, this shift can occur due to a j0 / mA m72 822 43 2 1 a D /mC m72 1 100 3 50 4 120 240 E /MV m71 750 7100 b 120 90 60 5 6 120 240 E/MV m71 2 t /s0 V V Kochervinskii decrease in the relaxation times of the local forms of mobility upon an increase in temperature; above 740 8C, this can be due to the generation of cooperative forms of mobility with high ampli- tudes.28 Thus, the amorphous phase makes a certain contribution to the formation of ferroelectricity and cannot be regarded as an `inert matrix'. This conclusion appears sufficiently justified because the double hysteresis has been noted both for PVDF31 and for VDF copolymers with tetrafluoroethylene (TFE)32, 33 or trifluoroethylene (TrFE).34 It has been shown for the latter copolymer 34 that the probability of double hysteresis depends on the method used to prepare the sample.In particular, double hysteresis in the VDF copolymer with TrFE of the 52.8/47.2 composition is manifested only for samples crystallised by slow cooling of the melt. In the several initial polarisation cycles, these samples exhibit a typical double hysteresis, which is transformed into a single hysteresis in the subsequent cycles (Fig.3). A similar result has been obtained for uniaxially stretched PVDF samples (Fig. 4a); no double hysteresis is observed for polarised samples at any temperature. This transformation in the pattern of the D± E hysteresis should point to fundamental changes in the material structure after the generation of non-zero remanent polarisation. One process which might result in structural changes is the irreversible increase in the degree of crystallinity after polar- isation.35 The kinetics of transition of the film subjected to polarisation into a new structural state depends on the frequency of the electric field applied. Thus at room temperature and a frequency of *1073 Hz, this requires several poling cycles;29, 34 when the frequency is 50 Hz, the double hysteresis passes into single hysteresis in one cycle of the increase in the intensity of the polarising field.31, 32 Analysis of the time dependences of the polarisation current and thermodepolarisation curves for the 75/25 VDF/TrFE copolymer suggests that polarisation involves two types of dipoles (in the crystal and in the amorphous phase).36 It can be seen from comparison of Figs.2 and 4 that the structural changes in the film induced by poling have a substantial effect on the temperature dependences of both the coercive field and the remanent polarisation. Whereas in the initial sample, Ec changes slightly with the temperature, in the polarised state, this value markedly increases as the temperature diminishes.The state of chain dynamics in the disordered phase affects precisely the magnitude of the coercive field; it can be seen that the Ec ± T plot is broken at about the glass transition temperature (see Fig. 4b). The sharper growth in the Ec in the region of Tg indicates that freezing of the high-amplitude cooperative motion of chain segments retards substantially the processes of polarisation switching. The remanent polarisation depends on the temperature to a lesser extent, although an increase in Pr has been noted below 50 8C. j /mA m72 I /mA 1 2 1 10 0.5 5 U /kV 3 1 71 73 100 E /MV m71 50 750 Figure 3. Reversible polarisation current vs the intensity and voltage (U) of the electric field (561073 Hz/ 27 8C) in the several initial (1) and subsequent (2 ) cycles of poling of VDF/TrFE samples.34Ferroelectricity of polymers based on vinylidene fluoride a D /mC m72 3 4 100 2 1 50 0 120 E /MV m71 7240 7120 750 1 2 3 7100 4 b Pr /mC m72 Ec /MV m71 150 75 100 50 25 50 0 50 T / 8C 7100 750 Figure 4.Influence of the temperature on the D¡¾E hysteresis curves (a) and on the hysteresis parameters (coercive field Ec and the remanent polarisation Pr) (b) in uniaxially stretched polarised PVDF films.29 T / 8C: 7100 (1), 760 (2),720 (3), 20 (4). Measurements of the hysteresis during the variation of the polarising field frequency from 1072 to 1074 Hz demonstrated that neither Pr nor the piezoelectric constant e31 depend on the frequency. This finding permitted the conclusion29 that the ferroelectricity in PVDF has a purely dipole origin.A similar conclusion has been drawn for the 73/27 VDF/TFE copolymer, for which experimental Pr and Ps values are in good agreement with those calculated using a simple dipole model.37 It has been assumed 38 that free charge carriers in the polymer get frozen in deep traps and this compensates to an extent the intrinsic polarisation of crystals. This hypothesis was confirmed38 by the data on the influence of annealing of a polarised PVDF film on the shape of the D¡¾E hysteresis curve. The curve obtained in the first cycle of measurements after annealing was distorted. This was explained by assuming that immediately after the annealing, the charge carriers are characterised by a non-equilibrium distribution, which changes during the subsequent poling cycles.The dipole nature of the ferroelectricity is also confirmed by the hysteresis type of variation of the intensities of some absorp- tion bands in the IR spectra.28, 39, 40 This conclusion is also supported by the qualitatively consistent results of studies 41, 42 devoted to X-ray recording of the change in the distribution of the b axes (direction of the dipole moment) of the b-phase lattice after polarisation. It was shown that the intensity of the 110, 200 reflections of the crystalline b-phase changes by more than 20% following polarisation along particular directions. When develop- ing the model of polarisation switching, Kepler and Andersen 42 proceeded from the fact that the orthorhombic cell of the b-phase can be transformed into a hexagonal cell upon 1% distortion.The chains in the planar zigzag conformation can occupy one of six equivalent positions (Fig. 5). When a field E is applied, the b axis of the cell occupies one of these positions with the probability 42 4.89 A Figure 5. Packing of the PVDF chains in the crystal lattice of the b-phase. The dashed line denotes the boundary between the domains.43 expOa cos ciU fOcU a expOa cos cUia1 X6 ¢§1a fOcUdc where c is the angle between the domain dipole moment (m) and the direction of the field (E ); in addition, ci = c+(1/3)(i71)p; a = mE/kT. After polarisation, the orientation of the dipole moments can be described by the normalised distribution function # nOcU a fOcU n0 0 " 2Op a 3 expOa cos cU p ia1 X6 P a 3P0 p O2p 0 (n is the number of oriented axes, n0 is the total number of axes), and polarisation is described by the equation X6 expOa cos cU cosc where P0 is the overall polarisation under the assumption that all dipoles are completely oriented.The intensity of X-ray diffraction for the case of variation of the angle between the direction of incidence of the beam and the perpendicular to the film surface has been calculated.42 The intensity curves for various values of the parameter exp(2mE/kT ) were constructed. The experimental points obtained by averaging over seven samples are in the best agreement with the calculation based on the assumption that exp(2mE/kT )^103. Similar experiments have been carried out in another study. 44 The researchers monitored the variation in the intensity of the 110, 200 reflections after poling of a PVDF film shaped like a cylinder rotating around its axis, which coincided with the stretching direction.These experiments also confirmed the hypothesis that the dipole orientation can change by 60 8 steps during polarisation (see Fig. 5). This polarisation model is also confirmed by calculations of the potential energy for the chain rotation, which show that two neighbouring equilibrium positions of the CF2 group are separated by 60 8.42 Finally, the data on the hysteresis of the IR bands upon application of a field also point to 60-degree reorientations. Indeed, in terms of the model assuming 180-degree steps of rotation, no changes in the IR absorption intensity should be expected because the intensity is proportional to cos2y (see Ref.40).The values of polarisation calculated from Eqn (3) are presented in Table 1. The most probable exp(2mE/kT ) value for PVDF amounts to 103; it is matched by P^111 mC m72. The crystallinity of the sample is normally*50%; hence, polarisation should be 50 mC m72, which is consistent with the PVDF remanent polarisation found experimentally. The known exp(2mE/kT ) value can be used to determine the dipole moment of the domain. 42 For the field E = 600 kV cm71 at 100 8C, it amounts to 3610728 C m. If the dipole moment of a monomer 4.89A 823 E ¢§1 (1) , (2) ¢§1 expOa cos ciU expOa cos ciU¢§1 dc, (3) ia1824 Table 1. Polarisation of crystalline PVDF calculated for various magni- tudes of the mE/kT parameter.42 P /mC m72 exp(2mE/kT ) mE/kT 0 66 98 1 10 102 103 104 105 106 0.00 1.15 2.30 3.45 4.61 5.76 6.91 111 116 120 121 unit in a planar zigzag chain is taken to be m0 = 7610730 C m, then a domain should comprise *40 monomer units. The polar b-phase crystals with normal sizes (*100A) contain many more monomer units. This means that field-induced orientation occurs in an area the size of which is much smaller than the size of the b-phase crystal. This conclusion is probably confirmed by the data of analysis of the ferroelectric and pyroelectric characteristics of thin films of the 70/30 VDF/TrFE copolymer prepared by the Langmuir ¡¾ Blodgett method.45 In the films containing several monolayers, the formation of a domain structure was noted; hysteresis of the pyroelectric current and a first-order Curie transition of the ferroelectric ¡¾ paraelectric type was clearly observed for these films.The regions of spontaneous polarisation, typical of ferro- electrics, can be probed using the electrochromism of optical labels. The optical labels used are normally azo dyes that efficiently absorb in the visible region. For instance, 4-dimethyl- amino-40-nitrostilbene (DANS), for which the dipole moment in the ground state and the difference between the dipole moments in the ground and excited states are known, has been used as the optical label for PVDF and systems based on it.46, 47 To estimate the strength of the local (internal) field El which had appeared in the sample after polarisation, the researchers made use of the fact that the absorption spectra of DANS in the perturbed (E 6a 0) and non-perturbed (E=0) states differ in the absorbance and in the frequency of the main absorption band.48 The local field strength in polarised samples of a blend of PVDF and poly(methyl methacrylate) (PMMA) proved to be more than twice as high as the strength of the external field during poling.47 A similar result has been obtained for the VDF/TrFE copolymer.46 Apparently, for these polymers, the influence of the charge carriers injected from the electrode material should be taken into account.This hypothesis is supported by the results of analysis of the polarisation profile in these samples by the thermal pulse method.49, 50 The polarisation profile is formed not only by dipoles but also by the space charge.The role of the latter is to stabilise the areas of oriented dipoles, which ensures the invari- ability of El in the temperature range of annealing (60 ¡¾ 90 8C).47 Some decrease in El near 60 8C (glass transition region) implies, apparently, that the fields of the disordered phase areas should also make a contribution to El. This contribution is relatively low because the most pronounced decrease in El occurs at about *120 8C, where annealing is accompanied by depolarisation processes due to disorientation of the b-phase crystals. Jimbo et al.51 have attempted to determine the above-noted substantial changes in the coercive field observed in the polarised PVDF upon the temperature variation (see Fig. 4) using a simple phenomenological relaxation theory. In accordance with this theory, they calculated the Ec values, which corresponded to the experimental data in the temperature range chosen.As noted above, PVDF and VDF-based copolymers contain at least two phases (ordered crystalline phase and disordered amorphous phase) having substantially different electric parame- ters, in particular, dielectric permittivity. This is attributable to the fact that the disordered phase occurs at room temperature in the liquid-like state and its chains execute cooperative motion with V V Kochervinskii large amplitudes.28 Therefore, depolarisation field Ed , the direc- tion of which is opposite to that of the applied field E, can arise at the boundaries of ferroelectric crystals. The field El which really acts on the crystal can be described by the equation 52 E (4) a f 1 af ¢§ 1 P , E ¢§ L ee0 l a E ¢§ Ed a f OPc ¢§ PU = E ¢§ L ee0 where f is a parameter related to the depolarisation factor L and to the difference between the dielectric constant of the sample as a whole e and that of the crystalline region ec (their polarisations are P and Pc , respectively), e0 is the vacuum dielectric constant and a is the degree of crystallinity.The fact that El is lower than E means that the experimental hysteresis curves do not reflect characteristics of polar crystals. By processing experimental P¡¾ E curves, Ikeda et al.52 elucidated the dependences for the crystallites, i.e., those in the Pc ¡¾El coordi- nates. The theoretical Pc ¡¾ El curves were constructed for L = 0.04 and 0.08; provided that e = 12 and ec = 3, this is matched by f = 1.031 and 1.064 and the draw ratios for an ellipsoidal crystal are 6.5 and 3.8.Since f differs insignificantly from unity, then, taking into account that El = f Ec (see Ref. 51), the coercive field values for the calculated and experimental curves are virtually identical. Spontaneous polarisation of the crystal Ps c and that for a sample with the degree of crystallinity a are related as follows:51 Ps c = Pfa . When a = 0.45, spontaneous polarisation is substantially higher than the overall polarisation. The value Ps c^170 mC m72 obtained by Ikeda et al. 52 is consistent in order of magnitude with the spontaneous polarisation (130 mC m72 ) calculated for a perfect crystal of the b-phase. III. Characteristic features of the polarisation switching processes The analysis of the features of polarisation switching processes is conveniently started with the data on PVDF; this homopolymer has been studied in the uniaxially stretched state, in which the crystalline phase occurred in the ferroelectric b-modification.Switching experiments usually consist in the application to the film of a stepwise field and recording of the time dependence of the electric displacement or polarisation current. The latter operation is often replaced by differentiation of the D(t) curve (t is the current time) with respect to time. These curves are shown in Fig. 6. It can be seen that, as the electric field strength increases, the inflection point in the D(t) curve and, correspondingly, the a b D/ mC m72 qD q log t 10 120 120 10 9 6 7 8 9 5 6 7 8 4 5 60 60 3 4 3 2 1 2 1 0 log t (s) 0 log t (s) 7 06 72 74 76 72 74 Figure 6.Time dependences of the electric displacement (a) and its derivative with respect to time (b) for the ferroelectric switching in a polarised PVDF film at 20 8C and at various external electric field strengths.53 E/ MVm71: 20(1), 40 (2), 60 (3), 80 (4), 100 (5), 120 (6), 140 (7), 160 (8), 180 (9), 200 (10).Ferroelectricity of polymers based on vinylidene fluoride (5) E ts a t0 exp maximum of the derivative shift to shorter times. This type of behaviour has also been noted for classical inorganic ferroelectrics and is usually described by the relations Ea or ¢§n (6) , ts a t0 E E0 n, (7) t ts D=D0+2Ps 1 ¢§ exp where ts is the time of polarisation switching [corresponding to the maximum of the derivative of D(t) or to the switching point], t0 is the pre-exponential factor, Ea is the activation field, and E0 and n are constants.The D(t) function can be represented as follows where D0 is the instant electric displacement and 2Ps is the reversible polarisation. The exponent n corresponds approxi- mately to the half-width of the qD/qlogt peak. When the field strength is 60 MV m71, the exponent n for the 65/35 VDF/TrFE copolymer is greater than 5. It has been noted above that the specific character of the polymeric state is due, on the one hand, to the strong anisotropy of bonds along and across the chain and, on the other hand, to the presence of an amorphous phase in addition to the crystalline phase.The chain dynamics of the amorphous phase below and above glass transition temperature are substantially different;28 therefore, to understand the mechanism of generation of ferro- electricity in polymers, it is important to find out how this can influence the switching characteristics. Furukawa et al. 53 have studied the processes of switching in a polarised PVDF film in which the amorphous phase occurred in the liquid-like or vitrified state (Fig. 7). It can be seen that, when the field is so high (200 MV m71), switching can also take place below Tg (740 8C). Below the glass transition temperature, the equili- brium electric displacement regularly decreases, the maximum switching rate D diminishes, and the peaks themselves become markedly broader as the temperature decreases.Figure 8 presents the field dependences of the switching times in a polarised PVDF film at various temperatures. These plots are linear both above and below the glass transition temperature, indicating that relations (5) and (6) hold. Transition of the amorphous phase into the glassy state noticeably changes the switching parameters (Table 2). In fact, as the temperature diminishes from room temperature to Tg, the exponent n and the constant t0 in Eqn (6) substantially increase (see Table 2). If the field dependences are represented as Eqn (5), then, in accordance with the results obtained by Furukawa and Johnson,54 the transition of the disordered phase in the glassy state is also accompanied by a substantial increase in the activation field; Ea is 1.1 GV m71 at 20 8C and 1.9 GV m71 at760 8C.PVDF is characterised by higher activation fields and by higher exponents n in relations (6) or (7) than inorganic ferro- b a D/ mC m72 qD q log t 4 3 150 5 150 1 2 6 7 6 1 2 3 4 5 90 90 7 30 30 0 log t (s) 0 log t (s) 72 74 72 76 74 76 Figure 7. Time dependences of the electric displacement (a) and its derivative with respect to time (b) for the ferroelectric switching in a polarised PVDF film at E=200 MV m71 and various temperatures.53 T/8C: 20 (1), 0 (2),720 (3), 740 (4),760 (5), 780 (6),7100 (7). 825 log ts (s) 71 1 73 23 4 75 5 logE (MV m71) 2.1 1.9 Figure 8. Field dependences of the switching times in a polarised PVDF film at various temperatures.53 T/ 8C: 760 (1),740 (2), 720 (3), 0 (4), 20 (5).Table 2. Values n and t0 for PVDF and the 73/27 VDF/TrFE copolymer determined in the switching experiments. n T/ 8C log t0 (s) PVDF53, 54 200 7.6 8.0 9.7 10.7 10.6 12.02 13.63 18.35 21.19 22.15 720 740 760 VDF/TrFE 51 16.48 10.2 20 (8) s D a 2P electrics. This fact was explained 53 in terms of the model of nucleation and growth of domains. Within the framework of this model, the switching curve D(t) is described by the expression Ot 0Re¢§RsFOuOt ¢§ sUmUds , #) 1 ¢§ exp ¢§ N " ( 0 where N0 is the number of nucleation sites, R is the probability of nucleation, s is the current time, F is a factor depending on the shape of the growing domain, u is the growth rate, t is the total time, and m is the growth dimensionality. If switching is assumed to occur in very thin lamellae, consideration can be restricted to two-dimensional growth, i.e.m = 2. In this case, it follows from . (9) s 1 ¢§ exp Eqn (8) that

ISSN:0036-021X    

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Russian Chemical Reviews 68 (10) 859 ± 879 (1999) Methods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids V P Shevchenko, I Yu Nagaev, N F Myasoedov Contents I. Introduction II. Synthesis of tritium-labelled fatty acids and their derivatives III. Oxylipins IV. Steroids V. Conclusion Abstract. The achievements in the field of synthesis and applica- tion of tritium-labelled oxylipins, steroids, fatty acids, phospho-, sphingo- and other lipids are reviewed. The importance of these studies for the solution of current problems of biochemistry, biology and pharmacology is exemplified in the application of labelled compounds. The bibliography includes 148 references. I. Introduction Methods used for the synthesis of tritium-labelled classical representatives of lipids and oxylipins (fatty acids, steroids, prostaglandins, sphingo- and phospholipids, etc.) and their bio- logically active analogues close in physicochemical properties to lipids have gained considerable popularity.The choice of lipids for investigation is based on the fact that apart from the ability of these compounds to accumulate metabolic fuel, which for many decades was considered to be their sole function,1 two other important functions have been recently assigned to them. First, they are the principal structural components of cell membranes and, second, they are bioeffectors, which control intracellular biochemical reactions, intercellular interactions and other phys- iological processes occurring in the organism.Evidently, the interaction of bioeffectors with their targets and, consequently, the specificity of their action are largely determined by their structure.1 The advantages of tritium as a radioactive label, viz., low radiotoxicity, long half-life (12.26 years), high molar radioactivity (1.08 PBq g-atom71), etc., are universally known.2±5 The meth- ods used for the synthesis of tritium-labelled compounds include different versions of reactions of isotope exchange, chemical and enzymic methods. The mechanisms of labelling by isotope exchange { have been considered.5± 15 The labelling by a liquid-phase method is accompanied by reactions of isotope exchange; migration, isomerisation and hydrogenation of double bonds; selective hydrogenation of one of several double or triple bonds; dehalogenation and reduction of V P Shevchenko, I Yu Nagaev, N F Myasoedov Institute of Molecular Genetics, Russian Academy of Sciences, 123182 Moscow, pl.Kurchatova 46, Russian Federation. Fax: (7-095) 196 02 21. Tel. (7-095) 196 02 12 (V P Shevchenko), (7-095) 196 00 01. E-mail: img@glas.apc.org (N F Myasoedov) Received 13 May 1999 Uspekhi Khimii 67 (10) 944 ± 966 (1999); translated by R L Birnova #1999 Russian Academy of Sciences and Turpion Ltd UDC 546.110.23 : 15/17 859 860 868 874 877 the corresponding functional groups. Only a careful choice of reaction conditions can result in satisfactory yields of labelled target products. An important role is ascribed to the choice of a catalyst and a solvent and determination of optimum experimen- tal parameters (the pressure of gaseous tritium, the reaction time, the catalyst : substrate : modifying additives ratio, etc.).Classical methods have limited application, since only milli- gram and submilligram quantities of precursors are normally used in the synthesis of tritium-labelled compounds. As a result, the catalyst : precursor ratio should markedly exceed that prescribed for selective hydrogenation using classical procedures on purely technological grounds. In contrast, the proportional increase in catalyst-modifying additives (quinoline, pyridine, etc.) signifi- cantly decreases the yield of the labelled target product and considerably increases the total radioactivity of the reaction mass due to incorporation of the label into side products and modifying additives, which complicates the isolation and purifi- cation of the target compound.If the components used in biological studies are sufficiently stable at elevated temperatures, a solid-phase method where the substance applied onto the catalyst is heated in the atmosphere of gaseous tritium is the most promising approach.9, 12, 13 The method of isotope exchange with tritiated water in inert solvents is used for labelling compounds that are labile in hydro- genation conditions; the use of inert solvents prevents self- radiolysis of highly enriched tritiated water.2 The latter is usually obtained by reduction of PdO with gaseous tritium. Palladium black and palladium catalysts immobilised on inert supports initiate the isotope exchange.Obviously, the more active the catalyst the higher the degree of isotope exchange with tritiated water, but the probability of degradation of the labile compound increases also. As a rule, chemical syntheses of tritium-labelled lipids are preceded by the synthesis of the corresponding precur- sors 4, 5, 16 ± 22 containing triple or additional double bonds, halo- gen-substituted fragments, etc. Treatment of these compounds with gaseous tritium in the presence of a catalyst results in the formation of a labelled product. In some cases, tritium-labelled { Modern methods of isotope exchange can be conventionally divided into three main group, viz., heterogenous catalytic isotope exchange of gaseous tritium with an organic compound dissolved in a solvent (liquid-phase methods), heterogenous catalytic isotope exchange of gaseous tritium with a solid organic compound at elevated temperatures (solid-phase methods) and heterogenous catalytic isotope exchange of an organic compound dissolved in an aprotic solvent with tritiated water.860 lipids are obtained by a multistep synthesis which utilises a labelled preparation as a starting compound.This approach requires the use of methods which not only ensure high yields of target products but also make the process technologically safe. II. Synthesis of tritium-labelled fatty acids and their derivatives 1.Fatty acids Fatty acids differ in the structure of the carbon backbone and the number of double bonds as well as in the presence and number of other functional groups. They are present in natural subjects, predominantly as esters. Fatty acids are constituents of phospho- lipids, sphingolipids, glycolipids, gangliosides, glycerides, steroid esters and vitamins. It had long been thought that the sole function of fatty acids was to provide membrane integrity and the optimum level of unsaturation of tissue lipids;23 however, it has recently been established that these compounds also regulate the enzyme activity, modulate binding of steroid hormones to their specific receptors and control gene transcription.24 ± 26 Indeed, a change in the fatty acid composition of membrane lipids changes the activity of membrane-bound enzymes and expression of receptors.27 Isotopically labelled fatty acids can be synthesised by isotope exchange, selective hydrogenation of the corresponding acetylenic precursors with gaseous tritium as well as by chemical, micro- biological and other methods.5 The liquid-phase method of isotope exchange is the simplest procedure for obtaining unsaturated fatty acids, since it allows direct labelling of the target compound.Besides, this procedure is devoid of many drawbacks of microbiological methods, e.g., difficulties in isolation of labelled preparations from homoge- nates, isotopic dilution of the labelled target product with endogenous products derived from microbial cultures and the loss of the label in the course of biological conversions.4, 5 For example, the molar radioactivity of labelled (5Z,8Z,11Z,14Z,17Z)-eicosapentaenoic (thymnodonic) acid syn- thesised from [8-3H]stearic acid 5 using a culture of Saprolegnia parasitica (ATC 22284) decreased 250 ± 400-fold in comparison with that of the original fatty acid.The lability of alkenes in liquid-phase isotope exchange is the main property which restricts the application of this method for introducing the label into unsaturated fatty acids. It is known 3±5 that hydrogenation, cis ± trans-isomerisation and double bond migration occur extensively even with the Lindlar catalyst. The isomerisation reactions are equally probable, while the double bond migrates by 85%± 88% towards the carboxy group. How- ever, these restrictions did not prevent the labelling of both monoenoic and polyenoic acids by varying the reaction condi- tions (Table 1).Some data on the distribution of the label in molecules of these compounds will be given below. Despite the relative simplicity of the liquid-phase isotope exchange, it is of limited use, since the molar radioactivity of target products seldom exceeds 0.1 PBq mol71. Molar radio- activities of polyenoic fatty acids are much higher, if chemical methods are used. Acetylenic precursors can be prepared either from the corresponding fatty acids or by total chemical synthesis. The former method was used in the synthesis of 5-acetylenic analogues of methyl arachidonate, methyl thymnodonate, the 14-acetylenic analogue of methyl arachidonate and the 4-acety- lenic analogue of methyl (4Z,7Z,10Z,13Z,26Z,19Z)docosa- hexaenoate (methyl docosahexaenoate).This type of synthesis is illustrated in the example of 5-acetylenic analogues of eicosa- polyenoic acids and the 14-acetylenic analogue of methyl arachinodate (Schemes 1 and 2).16, 17, 30 The major difference in these schemes is that in the latter the epoxy acid is formed upon treatment of the starting compound with m-chloroperbenzoic acid in dichloromethane in the presence V P Shevchenko, I Yu Nagaev, N F Myasoedov Table 1. The values of molar radioactivities (a) of labelled unsaturated fatty acids.a Ref. Starting compound a /PBq mol71 Liquid phase isotope exchange 28 28 28 28 0.146 0.025 0.034 0.092 Methyl oleate Methyl linoleate Methyl dihomo-g-linolenate Methyl arachidonate Selective hydrogenation with gaseous tritium 29 29 29 16 16 16 17 8,11,14-Triyne analogue of methyl dihomo-g-linolenate 5,8,11,14-Tetrayne analogue of methyl arachidonate 5,8,11,14,17-Pentayne analogue of methyl thymnodonate 5-Acetylenic analogue of methyl arachidonate 14-Acetylenic analogue of methyl arachidonate 5-Acetylenic analogue of methyl thymnodonate 4-Acetylenic analogue of methyl docosahexaenoate 5.6 7.4 9.2 1.3 1.2 1.9 1.8 Note: In liquid-phase isotope exchange, the label is incorporated by isotope exchange of hydrogen atoms for tritium atoms; in hydrogenation with gaseous tritium, the label is incorporated owing to the reduction of triple bonds to double bonds.a Free fatty acids were obtained by alkaline hydrolysis of their methyl esters. Scheme 1 O O Et3N, MeOH I2 RCH CH(CH2)3CO2H RHCI O 1. HBr 2. CrO3 RCH CH(CH2)3CO2Me TosNHNH2 RCHBrC(CH2)3CO2Me O RC C(CH2)3CO2Me , R=C5H11CH=CH(CH2CH=CH)nCH2 (n=2, 3). Scheme 2 m-ClC6H4CO3H C5H11CH CH(CH2CH CH)3(CH2)3CO2Me 1. HBr C5H11CH CH(CH2CH CH)3(CH2)3CO2Me 2. CrO3 O TosNHNH2 CH)3(CH2)3CO2Me C5H11CBrHC(CH2CH O C5H11C C(CH2CH CH)3(CH2)3CO2Me. of KHCO3.16 It is known that the use of nonpolar solvents for epoxidation results in the formation of substantial amounts of diepoxides due to the high reactivity of the peroxy acid under the given experimental conditions; therefore, ethanol was used to increase the yield of monoepoxides.30 This procedure was appliedMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids Scheme 3 TA 5,6-epoxy-TA (10) 14,15-epoxy-TA (21) 17,18-epoxy-AA (38) 8,9- and 11,12-epoxy-AA (31) AA 5,6-epoxy-AA (12) 8,9-epoxy-AA (17) 11,12-epoxy-AA (29) 14,15-epoxy-AA (42) DA 4,5-epoxy-DA (4) and 10,11-epoxy-DA (11) 7,8-epoxy-DA (7) and 13,14-epoxy-DA (15) 16,17-epoxy-DA (18) 19,20-epoxy-DA (45) The figures in parentheses indicate the percentage of the monoepoxide with respect to the total monoepoxide content in the reaction mixture.for the preparation of monoepoxides from arachidonic (AA), thymnodonic (TA) and docosahexaenoic acids (DA) (Scheme 3). Acetylenic analogues of eicosapolyenoic acids, viz., the 8,11,14-triyne analogue of methyl homo-g-linolenate, the 5,8,11,14-tetrayne analogue of methyl arachidonate and the 5,8,11,14,17-pentayne analogue of methyl thymnodonate were synthesised by purely chemical methods (Scheme 4). The reduction of triple bonds to double bonds in the atmosphere of gaseous tritium was used to prepare highly labelled polyenoic fatty acids (see Table 1). In order to increase the selectivity of hydrogenation of the corresponding acetylenic analogues, one had to modify palladium catalysts with lead diacetate, pyridine and quinoline.However, the yield of tritium- labelled alkenes did not exceed 10%± 40% even when the hydro- genation of one triple bond was performed according to the previously described procedures (cf. Ref. 5). Satisfactory yields C5H11C CCH2C CH 1. (CH2O)n 2. PhSO2Cl CH C(CH2)6CO2H C5H11(C CCH2)2OSO2Ph CH CCH2C C(CH2)3CO2H CCH2)3(CH2)6CO2H C5H11(C C5H11(C CCH2)4(CH2)3CO2H Table 2. Acetylenic precursors used in the synthesis of the pheromones of Lepidoptera.31 Pheromone component Acetylenic precursor (8Z)-dodecenyl acetate (7Z,8Z)-epoxy-2-methyloctadecane (6Z)-heneicosen-11-one (7Z)-eicosen-11-one (5Z,7E)-dodecadien-1-ol (7E,9Z)-dodecadienyl acetate (9Z,12E)-tetradecadienyl acetate (4E,7Z)-tridecadienyl acetate (7Z,11E)-hexadecadienyl acetate Me(CH2)2C:C(CH2)6CH2OH Me(CH2)9C:C(CH2)4CHMe2 Me(CH2)4C:C(CH2)3CH(OH)(CH2)9Me Me(CH2)5C:C(CH2)3C(O)(CH2)3Me Me(CH2)3CH=CHC:C(CH2)3CH2OH MeCH2C:CCH=CH(CH2)5CH2OH MeCH=CHCH2C:C(CH2)7CH2OH Me(CH2)4C:CCH2CH=CH(CH2)2CH2OH Me(CH2)3CH=CH(CH2)2C:C(CH2)6OH 861 were obtained with palladium on barium sulfate as a catalyst, which had been treated consecutively with methanolic solutions of lead diacetate [Pd : Pb=1 : (0.5 ± 1.0) mg mg71] and sodium borohydride [Pd :NaBH4=50 : (0.4 ± 4.0) mg mg71].A 4 : 1 hep- tane ± dioxane mixture was used as a solvent. The cata- lyst : substrate ratio was (2 ± 6) : 1.29 Under these conditions, the yield of the target products with a molar radioactivity of 1.2 ± 1.8 PBq mol71 per selectively reduced triple bond was 89%± 99% after 25 ± 35 min (see Table 1).It is necessary to emphasise that the reaction conditions were optimised not only for the labelling of unsaturated fatty acids, but also for pheromones, attractants and lipid hormones the complete chemical synthesis of which includes selective hydrogenation of acetylenic precursors in the final stage.31 This can be exemplified in the use of acetylenic precursors in the synthesis of the Lepidoptera pheromones (Table 2). Thus, bombykol (molar radioactivity 1.85 PBq mol71) was synthesised from hexadec-(10E)-en-12-yn-1-ol (see Ref. 29); the abscisic acid analogue (molar radioactivity 1.21 PBq mol71) (see Ref. 29) was prepared from (2Z)-5-(4-tert-butyl-trans-1-hydroxy- cyclohexyl)-3-methylpent-2-en-4-ynoic acid; retinoic acid (molar radioactivity 1.75 PBq mol71) (see Ref.20) was obtained from 11,12-dehydroretinoic acid. (Z)-[11,12-3H]Hexadeca-1,11-dien-3- one, (Z)-[6,7-3H]eicos-6-en-11-one and some other labelled pher- omones 32 with a molar radioactivity of *1.5 PBq mol71 were obtained by selective hydrogenation of the acetylene bonds. The use of the liquid-phase isotope exchange for labelling saturated fatty acids is not promising, since molar radioactivities of reaction products were of the order of several hundredths of PBq mol71 (Table 3). However, in contrast with alkenes, the isotope exchange in saturated compounds is not accompanied by side reactions; therefore, investigations into the use of this method were carried out and some regularities were established. Thus it was found that the kinetic curves of the isotope exchange reaction reach the plateau rather rapidly.In this case, the degree of substitution of tritium for hydrogen is much smaller than could Scheme 4 CH CCH2Br 1. MeMgCl (2 equiv.) 2. BrCH2C CCH2Br C2H5(C CCH2)2Br CH C(CH2)3CO2H CH CCH2OSO2Ph 1. BrMgC:CCH2OTHP 2. TsOH, MeOH 3. PhSO2Cl C2H5(C CCH2)3OSO2Ph C2H5(C CCH2)5(CH2)3CO2H Species Grapholitha molesta, Cydia molesta Porthetria dispar Orgyia pseudotsugata Carposina niponensis Dendrolimus spectabilis Lobesia botrana Plodia interpunctella, Cadra cautella Phthorimaea operculella Sitotroga cerealella862 Table 3. The values of molar radioactivities of saturated fatty acids.Starting compound Liquid-phase isotope exchange Methyl stearate Methyl palmitate Methyl myristate Methyl laurate Solid-phase isotope exchange Stearic acid Palmitic acid Myristic acid Lauric acid 3-Hydroxypalmitic acid 6-Ketopalmitic acid Methyl stearate 11-Chloroundecanoic acid Cetyl alcohol Hexadecane Solid-phase dehalogenation 11-Chloroundecanoic acid Hydrogenation with gaseous tritium Palmitoleic acid Undec-10-enoic acid 5-(3-Octynyloxy)pentanoic acid 12-Methoxydodec-1-ynoic acid be expected from the equilibrium distribution of the tritium label. Besides, the kinetic curves of molar radioactivity have a complex shape, which is attributed to the different saturation of palladium with the hydrogen isotopes during the formation of the a-phase of palladium hydride and its subsequent transition into the b-phase.28 The high thermal stability of alkanes has made it possible to use solid-phase isotope exchange for introducing the label into saturated compounds (see Table 3). Studies of solid-phase isotope exchange with stearic acid as a model compound in the presence of palladium and platinum catalysts supported on different carriers showed that this reaction proceeded with the greatest efficiency on 5% Pt/C at 210 8C (the catalyst : substrate ratio=10 : 1) (Figs 1 ± 4).{ Similar investigations were carried out with stearic acid homologues.It was demonstrated that in this case, too, the optimal conditions for a solid-phase isotope exchange were as follows: 5% Pt/C, the catalyst : substrate ratio=10 : 1, reaction time=15 min.Other regularities were also established. First, the dependence of changes in the molar radioactivities of fatty acid homologues on the length of the carbon chain was practically linear, which points to the random labelling by this method. Second, the temperature at which the degree of isotope substitu- tion was a maximum (Tmax), was different for different analogues. Myristic Lauric Acid 58 200 44.3 190 M.p. /8C Tmax /8C In all probability, this temperature depends on the phase transition temperature for representatives of the same homolo- gous series (e.g., fatty acids). { In the study aimed at establishing the relative degree of the label incorporation, arel, the maximum molar radioactivity of the preparation was taken as 100%.Ref. a /PBq mol71 28 28 28 28 0.036 0.021 0.014 0.010 33 33 33 33 34 29 29 34 33 33 13.50 12.90 12.10 11.50 9.00 27.80 38.00 6.30 5.50 17.50 34 4.20 35 14 36 36 1.47 2.96 3.89 5.07 Stearic Palmitic 70 210 64 205 ± 210 V P Shevchenko, I Yu Nagaev, N F Myasoedov arel (%) 100 80 60 40 200 1 2 3 4 5 6 Catalyst Figure 1. The dependence of the relative degree of the label incorpora- tion into stearic acid on the nature of the catalyst. Catalyst: (1 ) 5% Pt/C, (2) 10% Pd/C, (3) 10% Pd/BaSO4, (4) 5% Pd/C, (5) 5%Pd/BaSO4, (6) 5%Rh/C. arel (%) 100 80 60 40 20 0 4 : 1 6 : 1 8 : 1 10 : 1 12 : 1 15 : 1 20 : 1 50 : 1 60 : 1 Catalyst : substrate ratio Figure 2.The dependence of the relative degree of the label incorpora- tion into stearic acid on the catalyst : substrate (5% Pt/C) ratio. arel (%) 100 80 60 40 200 220 140 180 T /8C Figure 3. The dependence of the relative degree of the label incorpora- tion into stearic acid on temperature (the catalyst is 5% Pt/C, the catalyst : substrate ratio is 10 : 1). arel (%) 100 80 60 40 20 60 40 20 80 t /min Figure 4. The kinetics of the label incorporation into stearic acid (the catalyst is 5% Pt/C, the catalyst : substrate ratio is 10 : 1, 210 8C). It is evident that at certain temperatures the diffusion of these lipid molecules increases to such an extent that the rate of diffusion of the spill-over hydrogen no longer affects the rate of the overall reaction of isotope exchange.The efficiency of the isotope exchange decreases drastically after 10 ± 15 min. The latterMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids circumstance can be rationalised as follows. The study of the valence state of hydrogen in hydrides of intermetallic compounds carried out by Semenenko et al.37 has offered a new outlook on the problem of the valence state of hydrogen in systems of the Pd7H2 type. Debates on this problem, which is of particular importance for hydride chemistry, have begun since the appearance of classical works of Ubbelhode,38 who put forward a concept of a positively charged hydrogen atom on palladium, and of Gibb,39 who proposed a model with a negatively charged hydride ion.Semenenko et al.37 proposed and rationalised the concept accord- ing to which hydride dispersion is a result of a redox reaction associated with the formation of hydride ions,Hd7, at a hydrogen concentration (cH) in the solid hydride exceeding a critical concentration (cHcrit), which, in turn, depends on the redox potential of the reaction: Hd+ + Md7 Hd 07 + Md 0+, H2 + M where M is the metal template of the catalyst. According to this model, the hydrogen chemisorbed on the active centres on the catalyst surface donates some of its s-electrons to the metal conduction band.In this context, at low values of cH the behaviour of the absorbed hydrogen is well described in terms of the Ubbelhode model38 which entails high diffusion mobility of theHd+ ions and an increase (or, at least, the lack of a decrease) in electric conductivity of the hydride phase in comparison with the original metal. The value of cHcrit corresponds to the concentration of the oxidising agent (Hd+ species) at which the latter oxidise the metal template and become electron density acceptors being converted into bulky, practically immobile Hd0¡ ions. As a result, the metal conduction band is depleted in electrons. Obviously, at higher degrees of filling of cavities with Hd0¡ ions having a fairly large effective volume, the electron deficiency on the metal ions forming a metal sublattice in the crystalline structure of the catalyst should increase which probably restricts its ability for reversible hydrogen absorption.In this connection, the concen- tration of the species (e.g., of tritium in the case of labelled preparations) which can enter into exchange reactions drops drastically, eventually resulting in a practically complete cessa- tion of the isotope exchange. Approaches to the synthesis of fatty acids with high molar radioactivity were successfully used for the preparation of other highly labelled saturated lipids, viz., fatty acid derivatives, fatty alcohols and alkanes (e.g., methyl stearate, 3-hydroxypalmitic, 6-ketopalmitic and chloroundecanoic acids, cetyl alcohol and hexadecane) (Tables 4 ± 6).34 It was found that the isotope exchange in halogen-containing saturated acids occurs much easier than dehalogenation.The experimental values of molar radioactivities reached tens of PBq mol71, which makes it possible to conduct biological experiments of any kind including ligand ± receptor binding studies. The nonspecific label distribu- Table 4. The values of the relative degree of isotope exchange in 11-chloroundecanoic acid at various temperatures and on different catalysts (the catalyst : substrate ratio=5 : 1, reaction time=20 min).34 5%Pt/C 10% Pd/C 5% Pd/C 10% Pd/BaSO4 5% Rh/C T /8C 19 16 40 55 51 28 12 25 710 7 7 48 50 75 38 32 29 721 719 50 70 90 120 140 160 170 180 190 200 210 220 7 7 7 7 7 7 743 64 59 7 7 7 100 50 43 19 18 14 22 710 79 61 90 48 41 35 16 20 79 79 863 Table 5.The values of the relative degree of isotope exchange in cetyl alcohol at various temperatures and on different catalysts (the cata- lyst : substrate ratio=6 : 1, reaction time=15 min).34 5% Rh/C 5% Pd/C 10% Pd/C 5%Pt/C T /8C 7 7 1 1 4 4 4 4 5 5 9 9 7 7 15 15 19 19 20 20 9 8 1 6 90 110 120 130 140 150 160 170 180 190 200 210 220 16 25 46 49 7 100 765 61 36 724 7 7 726 16 20 21 724 33 40 256 7 Table 6. The values of the relative degree of isotope exchange in hexadecane at various temperatures [the ratio 5% Pt/C : substrate= (6 ± 10) : 1, reaction time=15 min].34 Compound Temperature /8C 90 140 160 180 190 200 210 230 30 33 7 7 7 7 44 100 100 12 72 81 7 7 7 100 7 7 7 84 89 2 171 76 77 86 100 785 100 96 Methyl stearate 6-Oxopalmitic acid 3-Hydroxypalmitic acid Palmitic acid Myristic acid Lauric acid Hexadecane 30 61 69 43 80 100 62 78 7 7 7 44 60 7761 2 44 7 41 7 52 2 100 tion in the products obtained upon the isotope exchange is the only disadvantage of this method.Hydrogenation of double bonds (see Table 3) is used in the synthesis of saturated fatty acids with a definite position of the label (its precursor is an accessible unsaturated acid).4, 29, 35 Studies with [10,11-3H]undecanoic acid obtained by this method, which is a component of a termite pheromone, have shown that it comes to the pheromone glands with food or is biosynthesised.Among other methods used for the synthesis of labelled fatty acids, thermal activation of tritium deserves special mention. The use of this procedure for labelling of methyl 12-hydroxystearate (0.56 TBq mol71) was described by Shevchenko and Myasoe- dov.4 The method consists in the application of a substance as a thin film onto the inner surface of a cylindrical vessel containing a tungsten filament placed along its axis. The system is evacuated, filled with tritium up to 0.67 Pa, cooled with liquid nitrogen and the tungsten filament is heated to 2000 K for 30 s.Tritium molecules dissociate, which significantly increases the degree of the isotope exchange. [8L-3H]- and [8D-3H]Stearic acids 5 were obtained by conver- sion of methyl 8D(8L)-hydroxystearate into the corresponding tosylate, which was further reduced with lithium aluminium tritide (6.6 TBq mol71). Stearic alcohols formed were oxidised with CrO3 to yield the target product (3.3 TBq mol71). Labelled fatty acids find application in biological studies and in the synthesis of various derivatives and labelled oxylipins based on them. Thus, they can be used in studies of cholesterol esters biosynthesis 40 or in the synthesis of monoepoxides (see Schemes 1 ± 3), which are the products of the epoxygenase path- way of the arachidonic acid cascade or other polyunsaturated fatty acids.The biological activity of these epoxides is widely known. They stimulate the in vitro release of somatostatin from864 the median tubercle of hypothalamus and the secretion of insulin and glucagon from pancreatic islets, inhibit the vasopressin- stimulated release of H2O in bladder tumours, suppress aggrega- tion of platelets and neutrophils, enhance the release of calcium from dog aortic smooth muscle microsomes and increase the cytosolic levels of free calcium. Under the action of various enzymes, these epoxides are converted into other, more oxidised products, e.g., vicinal diols, 20- and 19-hydroxy derivatives.2. Fatty acid derivatives a. Fatty acid esters Glycerides. The high biological activity of glycerides can be exemplified in the properties of 2-arachidonoylglycerol, which is an intermediate in tri- and diglyceride metabolism and an alternative precursor of arachidonic acid manifesting cannabimi- metic activity.41 The use of a tritium-labelled analogue of this glyceride provided additional information about its pharmaco- logical activity. Labelled glycerides are usually obtained by condensation of the corresponding labelled fatty acids with glycerol.4 Different approaches to the synthesis of triglycerides (TG) have been considered.42 These lipids are formed upon heating of a mixture of glycerol and a fatty acid to 160 ± 190 8C; the reaction proceeds more efficiently in the presence of toluenesulfonic acid with removal of water formed by distillation with xylene.Higher yields of TG are obtained in reactions of silver salts of fatty acids with 1,2,3-tribromopropane in boiling xylene and of acid chlor- ides or mixed anhydrides with glycerol.TGcan also be synthesised under very mild conditions in the presence of dicyclohexylcarbo- diimide and 4-dimethylaminopyridine. This procedure can also be used in the synthesis of the corresponding steroid esters. The use of this approach (the reaction was carried out in 1,2-dichloroethane; a solution of dicyclohexylcarbodiimide in benzene was added in several portions) resulted in labelled TG in more than 90% yield. Di- and monoglycerides can be prepared from triglycerides under the action of the corresponding lipases.43 Thus 2-arachidonoyl- glycerol was prepared from triarachidonoylglycerol by treating it with Rhizopus delemar (`Seikagaku Kogyo', Japan) in 0.05 M AcONa7AcOH buffer (pH 5.6) containing 0.1 mmol litre71 NaCl and 10 mmol litre71 CaCl2 with vigorous stirring for 60 min.44 Among other methods used for glyceride synthesis, special mention should be made of the reduction of dihydroxyacetone 1,3-derivatives with sodium borotritide and the isotope exchange with tritiated water.Molar radioactivities of TG obtained by this method were by one order of magnitude lower than those of the tritiated water employed.4 Studies with 2-[3H]arachidonoylglycerol revealed that neuro- blastoma N18TG2 cells contain nonspecific esterases.Hydrolysis of this monoglyceride was only partly (by less than 40%) inhibited by an excess of the unlabelled substrate, whereas the degree of selectivity of the esterase isolated from peripheral cells of the basophilic rat leukemia RBL-2H3 was much higher (74%). It was shown also that in contrast with 2-palmitoylglycerol, 2-linoleoyl- glycerol inhibited hydrolysis of 2-arachidonoylglycerol by intact cells as a result of which the concentration of 2-arachidonoylgly- cerol capable to activate the cannabinoid receptor increased considerably. The experimental results described by different authors 41, 45 ± 49 suggest that non-cannabimimetic monoglycer- ides enhance the effect of 2-arachidonoylglycerol by increasing the half-time of its conversion and increase binding of 2-arach- idonoylglycerol to the cannabinoid receptor (4 ± 8-fold).Phospholipids. Formerly, phospholipids were considered exclusively as structural components of biological membranes which fulfilled the barrier function in respect of electrolytes and served as a matrix for membrane-bound enzymes and receptors. However, it has presently been recognised that phospholipids and the products of their enzymic and free-radical conversions also perform important cell regulatory functions, being substantial for the manifestation of catalytic activities of many enzymes. Besides, V P Shevchenko, I Yu Nagaev, N F Myasoedov Table 7. The values of molar radioactivities of various phospholipids.Compound Hydrogenation with gaseous tritium 1-O-[3H]Alkyl-2-O-methyl-sn- glycero-3-phosphocholine Lyso[3H]phosphatidylcholine [3H]Phosphatidylinositol 1-O-[3H]Palmitoyl-2-acetyl-sn- glycero-3-phosphocholine 1-O-[3H]Stearoyl-sn-glycero-3- phosphocholine [3H]Phosphatidylcholine 1-{10-Palmitoyl-20-[9,10-3H]stearoyl- sn-glycero-30-phosphoryl}-myo- inositol-4-phosphate Liquid-phase isotope exchange 1,2-[3H]Dipalmitoyl-sn-glycero-3- phosphocholine Combination of chemical and enzymatic methods 1,2-Dioleoyl-sn-[2-3H]glycero- 3-phosphoserine some phospholipids are endowed with hormone-like activities, play the role of platelet activation factors (PAF) (1-alkyl-2-acetyl- sn-glycero-3-phosphocholine), accelerate wound healing (lyso- phosphatidic acid), etc.Hydrolysis of phosphoinositides and methylation of phosphatidylethanolamine are the key reactions in the transmission of intercellular signals.50 ± 52 As a rule, the tritium label is introduced into phospholipids by hydrogenation of natural or synthetic unsaturated precursors (Table 7). The synthesis of tritium-labelled phosphatidylinositol and diphosphoinositide is shown in Scheme 5. Isotope exchange (Tables 7 and 8) and condensation of labelled fatty acids with the corresponding precursors as well as biological and enzymic methods and their combination are the most promising approaches.4, 55 Unsaturated precursors were synthesised in order to prepare labelled phosphatidylinositol and diphosphoinositide.56 Several steps of their synthesis including deacetylation with hydrazine hydrate, were identical. If the phosphatidylinositol was the target product, the protective groups were removed by anionic debenzy- lation and treatment with a cation-exchange resin (H+-form).In Table 8. The kinetics of liquid-phase isotope exchange between gaseous tritium and dipalmitoylphosphatidylcholine.28 Time /min Solvent 30 15 Catalyst�5%Pd/BaSO4 41 50 19 33 Chloroform Chloroform: : methanol (2 : 1) Methanol 28 11 Lindlar catalyst Chloroform Methanol 22 11 (Ph3P)3RhCl Chloroform 4 3 Ref. a /PBq mol71 53 2.30 28 28 54 0.14 0.50 1.85 11 0.48 28 29 0.54 0.39 29 0.02 55 0.93 150 180 90 60 45 88 100 80 97 65 78 54 73 47 56 70 67 35 34 30 23 22 22 44 34 11 11 7 6 6Methods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids Ch O O OH (PhO)2POCl O HO O Ch Ch O O O OP(OBzl)2 O AcO O Ch Ch O O O O P O O OBzl HO O Ch 1.NaI 2. H3O+ O HO OHO P O OH OH HO OH 3H2, 5%Pd/BaSO4 O HO OHO P O OH OH HO OH Ch is cyclohexylidene, Pl�C15H31CO, Ol�C17H33CO, St�C17H35CO, Bzl is benzyl. the synthesis of diphosphoinositide, inositol was phosphorylated at position 4 with an excess of phosphorus oxychloride in pyridine with subsequent deprotection. Hydrogenation of the unsaturated fatty acid residues in the myo-inositol derivatives in an atmosphere of gaseous tritium in the presence of a catalyst has led to highly labelled phospholipids.29 These conditions proved to be versatile and can be used in the synthesis of labelled sterol esters.Thus, the hydrogenation of cholesteryl oleate was used to prepare cholesteryl [9,10-3H]- stearate 29 with a molar radioactivity of 1.98 PBq mol71. Con- densation of [9,10-3H]oleic acid (0.07 PBq mol71) with lysophos- phatidylcholine in the presence of dicyclohexylcarbodiimide and L-dimethylaminopyridine (1 : 1 : 1 : 100) in ether (stirring for 20 h at room temperature) affords lecithin (yield 35%± 40%).4 The synthesis of phosphatidyl[3H]inositol can serve as an example of a synthesis of tritium-labelled fatty acid derivatives by biological methods.4 The starting compound, [2-3H]inositol, was synthesised by solid-phase reduction 35 of a commercially available scillo-inosose (2,4,6,3,5-pentahydroxycyclohexanone).35 In a search for optimum conditions for scillo-inosose reduction, 5%Pd/CaCO3, 5%Rh/Al2O3, 5%Pd/BaSO4, 5%Pt/C,5%Pd/C were tested as catalysts. The best results were obtained on a Ch O O O OP(OPh)2 Ac2O O HO O Ch Ch O O O O P OAg 1.NaI O OBzl 2. AgNO3 AcO O Ch 1. POCl3 OOl 2. NaHCO3 OPl Na2O3PO OOl OPl H2O3PO O[9,10-3H]St OPl H2O3PO 865 Scheme 5 Ch O O O OP(OPh)2 1. Pt/H2 O 2. PhCHN2 AcO O Ch Ch O O O O P O NH2NH2 O OOl OBzl OPl AcO PlO OlO I O Ch Ch O O O O P O O OOl OBzl OPl O Ch 1.NaI 2. H3O+ O HO OHO P O OH OOl OH OPl OH 3H2, 5%Pd/BaSO4 O HO OHO P O OH O[9,10-3H]St OH OPl OH rhodium catalyst (the catalyst : substrate ratio=10 : 1) with heat- ing to 150 8C for 15 min (the molar radioactivity of the labelled preparation was 0.56 PBq mol71, the yield was 25%). In addi- tion, scillo-inosose with a molar radioactivity of 0.11 PBq mol71 was isolated from the reaction mixture in*10% yield. The biosynthesis of the labelled phospholipid was performed by incubation of [2-3H]inositol with a culture of Kloeckera brevis (ATCC 9774) (7 days, 25 8C). Subsequent extraction, chromato- graphic purification and desalting resulted in phospha- tidyl[3H]inositol, which contained 54% of the original label and had a molar radioactivity*2 times as low as that of the original [2-3H]inositol. A combination of chemical and enzymic methods was used in the synthesis of 1,2-dioleoyl-sn-[2-3H]glycero-3-phosphoserine.55 The first step included the chemical synthesis of [2-3H]glycerol; its incubation with glycerol kinase in the presence of ATP gave sn- [2-3H]glycerol-3-phosphate (yield 98%± 99%).Boiling of the monotetraethylammonium sn-[2-3H]glycerol-3-phosphate tetrae- thylammonium oleate and oleic acid anhydride (1 : 10 : 10) in dry carbon tetrachloride (3 h, 80 8C) afforded 1,2-dioleoyl-sn- [2-3H]glycerol-3-phosphate in *56% yield. The pyridinium salt of the latter was condensed with the benzhydryl ester of N-Boc-866 serine in the presence of 2,4,6-triisopropylbenzenesulfonyl chlor- ide (1 : 3.2 : 6.4) in dry pyridine (24 h, 30 8C).Deprotection under acidic conditions and chromatographic purification yielded 1,2- dioleoyl-sn-[2-3H]glycero-3-phosphoserine (total yield 12%, molar radioactivity 0.93 PBq mol71). Labelled phospholipids play an important role in present-day biological studies. Thus in a study of the role of PAF in the regulation of cellular activity and cell-to-cell contacts, its labelled analogue was used in order to establish the effect of different antagonists blocking PAF activity at the level of PAF receptors on cell activation.51 A synthetic lipid antagonist CV-3988 inhibited the binding of the labelled phospholipid to platelets and sup- pressed the secretion and aggregation of platelets (0.1 ± 0.15 mmol litre71).57 The binding of this phospholipid to PAF receptors is known to trigger a cascade of biological reactions which result in the signal transfer into the cell and its subsequent activation.51 It has been shown 58 that 5 s after treatment of platelets with PAF, phosphatidylinositol-4,5-diphos- phate undergoes hydrolysis to inositol-1,4,5-triphosphate. This hydrolysis is catalysed by membrane-bound phospholipase C.It was found also that the antagonists inhibit both the binding of labelled PAF to platelets and the formation of inositoltriphos- phate in platelets.59 The role of inositoltriphosphates in the regulation of the Ca2+ level in cell cytoplasm is immense, which may account for the pronounced effect of phospholipids on vital activity of cells in general.50 b.Fatty acid amides Ethanolamides of fatty acids represent highly active endogenous bioregulators. Arachidonic acid ethanolamides, which has been named later anandamide, was isolated from porcine brain in the early 1990's.60 This compound proved to be an endogenous ligand of cannabinoid receptors, viz., it competitively inhibited the specific binding of cannabinoids to receptors and mimicked their biological effects.60 ± 67 Ethanolamides of other polyenoic fatty acids were isolated from mammalian brain more recently.68, 69 These compounds also bind specifically to cannabinoid receptors.It is established now that mammalian tissues manifest synthase and amidase activities responsible for the synthesis and inactiva- tion of anandamide.70 ± 77 Tritium-labelled ethanolamides of fatty acids were used in studies of their metabolism, the activity of enzymes of the arachidonic acid cascade in the oxidation of these fatty acid derivatives and the effects of these labelled compounds on functioning of the arachidonic acid cascade in different animal and plant cells. Ethanolamides can be labelled at both the acyl and ethanola- mide fragments. Highly labelled fatty acids can be synthesised by selective hydrogenation in an atmosphere of gaseous tritium, while labelling of ethanolamine can only be done by solid-phase methods.78, 79 Glycolic acid nitrile proved to be a promising starting compound, and 5% Pt/C, 5% Rh/C, 5% Pd/C, 5% Rh/ Al2O3 and 5% Rh/CaCO3 are effective catalysts.The highest yields were obtained with palladium catalysts; the molar radio- activity was higher on rhodium catalysts. The latter were used in the synthesis of labelled ethanolamine with a molar radioactivity of 1.04 ± 1.48 PBq mol71, but the yields of target products were so insignificant that preparative syntheses of the labelled compound could be carried out only in the presence of5%Pd/C. On platinum catalysts, dehydrogenation predominantly occurs. The data summarised in Table 9 suggest that T= 100 ± 120 8C is the optimum temperature. Under these condi- tions, the labelled product was obtained in 10%± 20% yield and had a molar radioactivity of 0.15 ± 0.22 PBq mol71.78 Isobutyl chloroformate 78 or N,N0-diisopropylcarbodiimide 79 were used for condensation of fatty acids with ethanolamine (Table 10).Studies on kinetics of oxidation of labelled anandamide and thymnodonic acid ethanolamide with soybean 15-lipoxygenase and on their possible oxidation in mouse splenocytes can serve as examples of application of these labelled compounds in biological studies.80 It was found that the values of Michaelis constants (Km) V P Shevcheu Nagaev, N F Myasoedov Table 9. The values of the relative degree of the label incorporation and yield of ethanolamine in the hydrogenation of glycolonitrile for different times and at different temperatures (the ratio 5% Pd/C : substrate=5 : 1).Yield (%) T /8C (20 min) Degree of label incor- Reaction time poration (arel) (%) /min (100 8C) 20 50 70 90 100 120 140 160 180 190 10 31 55 66 100 100 65 32 126 67 11 15 15 13 12863 5 10 15 20 30 40 50 60 75 90 Table 10. The values of molar radioactivities of ethanolamides. Compound Ref. a /PBq mol71 79 79 79 79 79 79 78 78 78 78 0.012 0.019 0.019 0.023 0.012 0.026 4.4 0.18 0.182 0.182 N-Arachidonoyl[3H]ethanolamine N-Dihomo-g-linolenoyl[3H]ethanolamine N-Eicosa-11,14-dienoyl[3H]ethanolamine N-Eicosa-11-enoyl[3H]ethanolamine N-Palmitoyl[3H]ethanolamine N-g-Linolenoyl[3H]ethanolamine N-[3H]Arachidonoylethanolamine N-Arachidonoyl[3H]ethanolamine N-Thymnodonoyl[3H]ethanolamine N-Docosahexaenoyl[3H]ethanolamine and Vmax for these compounds were equal to 3.2 and 4.9 mmol litre71 and to 0.11 and 0.15 mmol l71 s71, respec- tively.These values are of the same order of magnitude as those for arachidonic acid.79, 81 Experiments have shown that mouse splenocytes contain an anandamide-metabolising system. Spleno- cytes are related to cells continuously secreting 12-HETE and other eicosanoids. Anandamide-induced changes in the concen- tration of these products modulate functions of lymphocytes and other eicosanoid-controlled immune cells. Sphingolipids pertain to the class of lipid molecules with different chemical structures and functional activities.Sphinge- nin (sphingosine) is the most common sphingosine base of human and animal cell sphingolipids. The sphingenin molecule contains two chiral carbon atoms and can exist in four stereoisomeric forms. Natural sphingenin represents a D-erythro-isomer. N-Acylation of the amino group of sphingenin with fatty acids results in ceramides, the precursors of more complex sphingolipids (sphingomyelin and glycosphingolipids). Until the 1980's, the effect of the structure of the acyl residues of these lipids on their biological activity had been virtually neglected.82 However, studies carried out in the late 1980's provided evidence that sphingenin inhibited cell differentia- tion,83 while ceramides activated it.84 ± 87 More recent studies have shown that the biological properties of the fatty acid residues of ceramides depend on their lengths.Thus it was found that C18-ceramide does not inhibit lymphocyte proliferation (in contrast to the short-chain ceramides C2 and C6),88 whereas hydroxy acids present in natural ceramides strongly suppressed the induction of apoptosis.89 The activation of protein phospha- tase with ceramides also depends on the length of the acyl chain.90 This can be ascribed to the hydrophobic interactions between the hydrocarbon chains of sphingosine and the fatty acid which depend on the structure of both the sphingosine chain (length, the presence (sphingenin) or the absence (sphinganin) of a double bond) and the fatty acid residue (length, the presence or theMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids absence of double bonds, their number, position, the presence of hydroxy groups, etc.). Such interactions may attenuate the differ- ences in the effector activities of ceramides and dihydroceramides.Thus investigations into the mechanism of action of ceramides on target cells revealed that dihydroceramides, unlike the corre- sponding ceramides, inhibited (rather than stimulated) the activ- ity of protein phosphatases,90, 91 but had no effect on the activity of the protein kinase C isoform 92 and phospholipase D,93 which are inhibited by ceramides. However, in certain tests N-acetyl- sphingenin and N-acetylsphinganin manifested identical activ- ities.Radioactive labelling of sphingolipids can be performed in different ways, such as by condensation of the corresponding aminodiol with labelled fatty acids, by enzymic or chemical oxidation of hydroxy groups of sphingolipids with subsequent reduction with sodium or potassium borotritides, biosynthesis of labelled gangliosides starting from labelled N-acetyl-D-mannos- amine and hydrogenation of the corresponding unsaturated precursors with gaseous tritium (Table 11).The latter method was used for the synthesis of labelled gangliosides, sphingomyelin and ceramides. It was found 4, 5 that the 3H-label adds preferentially to the carbon atoms of multiple bonds. If liquid-phase isotope exchange is used for the labelling of unsaturated fatty acids, the label is predominantly incorporated into the allylic positions and into the double bonds.In phospho- lipids, the greater part of the label is localised in the fatty acid residues, whereas in the case of sphingolipids it is detected in both fatty acid and sphingosine moieties (Table 12) Labelling of cerebrosides on the Lindlar catalyst results in the label incorporation into the fatty acid moiety, while on 5% Pd/ BaSO4 as a catalyst, it incorporates into the sphingosine moiety. This can be explained by the fact that the molecules of these compounds contain several polar groups (carboxy, hydroxy, amino, etc.), which compete with one another for the active centres on the catalyst's surface.28 Presumably, the adsorption of the fatty acid residue on the Lindlar catalyst is more preferential than the adsorption of sphingosine bases.This fact determines the distribution of the tritium label. It should be noted that chemical methods can be used for the synthesis of compounds with a predetermined position of the label.95, 96 Complete chemical synthesis was used to obtain labelled D-erythro- and L-threo-sphingosines (Scheme 6). Starting from galactose, azidosphingosines were prepared; the allylic hydroxy group was oxidised with 2,3-dichloro-5,6-dicya- nobenzoquinone with subsequent reduction of the resulting ketone with sodium borotritide.95 An attractive approach was proposed by Toyokuni et al.96 Natural sphingenin was used as the starting compound; its primary hydroxy group was oxidised with CrO3 (the Collins reagent) after protection of the amino group and the allylic Table 12.The distribution of the tritium label in labelled glycosphingolipids after a catalytic reaction with gaseous tritium in solution.28 Catalyst Compound Tritium activity /PBq mol71 fatty acids 0.368 (19) 0.260 (29) Galactosyl ceramide Galactosyl ceramide 5% Pd/BaSO4 Lindlar catalyst Ganglioside GD1a G2 G1 GM2 0.009 (8) 0.086 (80) 0.065 (33) 0.298 (60) 0.140 (64) Sphingomyelin 5% Pd/BaSO4 (Ph3P)3RhCl 5% Pd/BaSO4 5% Pd/BaSO4 Lindlar catalyst Note: The percentage of the activity of a given moiety relative to the activity of the original molecule is given in parentheses.Table 11. The values of molar radioactivities of sphingolipids. Compound Chemical methods N-Oleoylsphinganin a (2S,3R,4E)-2-Amino[3-3H]dodec-4-ene- 1,3-diol b (2S,3S,4E)-2-Amino[3-3H]dodec-4-ene- 1,3-diol b (2S,3R,4E)-2-Amino[3-3H]octadec-4-ene- 1,3-diol b (2S,3S,4E)-2-Amino[3-3H]octadec-4-ene- 1,3-diol b (2S,3R,4E)-2-Amino[1-3H]octadec-4-ene- 1,3-diol b Hydrogenation with gaseous tritium Sphingomyelin Ganglioside GD1a GD2b GM1 GM2 GM3 G1 G2 Galactosyl ceramide c Galactosyl ceramide d Note: Brain and liver gangliosides are designated according to Svenner- holm (GD1a, GD2b, GM1, GM2, GM3); sea urchin (Strongylocentrotus intermedius) gangliosides are designated as G1 (N-glycolylneuraminosyl- (a,2?6)-glucosyl-(1?8)-N-glycolylneuraminosyl-(2?6)-glucosyl- (1?1)-ceramide) and G2 (8-sulfo-N-glycolylneuraminosyl-(a,2?6)- glucosyl-(1?8)-N-glycolylneuraminosyl-(2?6)-glucosyl-(1?1)-cer- amide).a The solid-phase method with subseqent condensation with oleic acid was used for introducing the label into sphinganin; b Sodium borotritide was used. c5% Pd/BaSO4 was used as a catalyst. d The Lindlar catalyst was used. hydroxy group. The reduction of the aldehyde formed with sodium borotritide and deprotection gave [1-3H]sphingenin with the natural configuration. Labelled sphingolipids were used to study the interaction of gangliosides with low density lipoproteins (using tritium-labelled ganglioside GM3) 97 and the role of gangliosides in the reception and penetration of the influenza virus into the cell.98 In the latter sphingosine bases hydroxy fatty acids 1.163 (60) 0.296 (33) 0.291 (15) 0.340 (38) 0.094 (82) 0.020 (19) 0.116 (59) 0.169 (34) 0.076 (35) 77777 867 Ref. a /PBq mol71 94 95 7.78 0.019 95 0.034 95 0.020 95 0.036 96 0.004 28 0.218 28 28 28 28 28 28 2844 0.115 0.137 0.196 0.496 0.014 0.159 0.107 1.94 0.895 polar fragments 0.116 (6) 70.012 (10) 0.001 (1) 0.016 (8) 0.030 (6) 0.002 (1)868 Ph PhCHO Galactose HO Ph O O Ph3P=CHCnH2n+1 (n=7 or 13) OH O H N3 HO O NH2 HO 3H OH D-erythro-Sphingenin case, Ehrlich ascites carcinoma cells, which are efficient receptors of the influenza virus, were employed.The viral adsorption decreased drastically after treatment of cells with Vibrio cholerae neuraminidase which points to the fact that sialo-containing compounds (gangliosides) are specific receptors of the virus on the cell surface. Indeed, gangliosides GM1, GD1a and GT1b restored the viral adsorption by cells. Based on the experimental data, it was concluded that: (1) the oligosaccharide chain of the ganglioside, Galb1?3 GalNAcb1?4(NeuAca2?3)Galb1?4 Glc, is the `minimum' structural element providing the adsorption O OH PGJ2 HOO OH PGD2 HO HO OH PGF2a O HO O HO HO 6-keto-PGF2a Scheme 6 O O NaIO4 O pH 4 OH OH Ph O O .. . OH CnH2n+1 H 1. NaB[3H]4 CnH2n+1 2. PPh3 NH2 CnH2n+1 CnH2n+1+ HO 3H HO L-threo-Sphingenin O COOH OH OO OOH PGG2 COOH OO PGH2 COOH COOH O O OH OH PGI2 OH COOH HO O OH of the influenza virus by the gangliosides; (2) the gangliosides probably take part in both binding and penetration of the virus into the cell; one receptor site binds *100 ganglioside molecules involved in the reception. III. Oxylipins Oxylipins comprise eicosanoids and eicosanoid-like compounds with shortened or lengthened carbon chains, which are synthes- ised, for the most part, by cyclooxygenase, lipoxygenase and cytochrome P450 pathways.5, 99 The main pathways of arachi- donic acid metabolism resulting in oxylipins, e.g., prostaglandins (PG), thromboxanes (TX), leukotrienes (LT), hydroxyeicosate- traenoic acids (HETE), hydroperoxyeicosatetraenoic acids (HPETE), lipoxins (LX), hepoxylins (HX), epoxytrienoic acids (EET), etc., are depicted in Scheme 7.It is of note that the scheme demonstrates exclusively transformations of arachidonic acid and its derivatives; however, other eicosapolyenoic acids can also be used as substrates in this reaction. Cyclooxygenase PGG PGH PG TX PGI COOH PGB2 COOH COOH OH COOH TXA2 COOH OH TXB2 V P Shevchenko, I Yu Nagaev, N F Myasoedov Arachidonic acid Lipoxygenase HPETE HETE DHETE THETE LT HX LX OOO HO 12-HHT OH Scheme 7 Cytochrome P450 o-Hydroxy derivatives and epoxides Scheme 8 COOH OH PGC2COOH PGA2 OH COOH OH PGE2 COOHMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids OOH COOH HO OOH 5-HETE COOH OH COOH 5-HPETE LTA4 Under the action of enzymes, arachidonic acid yields: (1) PG, TXand prostacyclines (PGI), (2) hydroxy acids (LT,HX, LX); (3) epoxy and o-hydroxy derivatives of fatty acids.Scheme 8 depicts preparation of PGandTXfromprostaglandinG2, Schemes 9 and 10, of hydroxy acids and LT from5-HPETE, Schemes 11 and 12, of hydroxy acids, LX and HX from 12- and 15-HPETE. The synthesis of o-hydroxy acids and epoxides is exemplified in oxidation of arachidonic acid in the presence of cytochromes P450 (Scheme 13). The biological and pharmacological effects of these com- pounds are extremely diverse.Thus lipoxins control the activity of killer cells and manifest spasmogenic activity;100 hepoxylins are the releasing factors in insulin secretion by pancreatic cells and regulate the transmembrane transfer of calcium;101 leukotrienes are mediators of hypersensitivity (allergy, anaphylaxis).102 Pros- taglandins have a wide application as medicinal drugs.103 How- ever, despite the diversity of oxylipins and the broad spectrumof their biological activities, it is the continuous interaction of these compounds with one another that provides for the normal course of vitally important reactions occurring in the organism. The validity of this concept was confirmed in the studies of the role of arachidonic acid metabolites in the maintenance of the normal functioning of hemostasis in mammals.104There exist at least two ways of maintaining the necessary level of arachidonic acid in a hemostatic system, viz., fromcell phospholipids and fromplasma lowdensity lipoproteins.The metabolites of polyunsaturatedfatty acids (prostacyclin, 13-hydroxyoctadecadienoic acid, 6-keto- PGE1, PGE1, HETE and HPETE, TXA2, PGE2, PGF2a, OH COOH Ê5H11 LTA4 OH OH COOH S C5H11 CH2CHCONHCH2COOH NHCOCH2CH2CHCOOH OH LTC4 NH2 OH OH COOH COOH OH S S C5H11 C5H11 CH2CHCHCONHCH2COOH CH2CHCOOH NHCOCH2CH2CHCOOH LTD4 NH2 NH2 LTF4 OH COOH S C5H11 CH2CHCHCOOH NH2 LTE4 869 Scheme 9 OH COOH 5,12-DHETE OH COOH OH 5,6-DHETE Scheme 10 COOH LTB4 OH COOH OH OH 20-hydroxy-LTB4 COOH COOH OH 20-carboxy-LTB4870 COOH COOH 15-HETE OH OOH 15-HPETE OH OH O OH HOOC (14S, 8E)-LXB4 OH HO OH OHOH COOH OH OH (6S)-LXA4 LXA4 PGD2b, LTA4, LTB4) participate in the regulation of practically all units of the blood coagulating system as well as in reactions of the plasma component of hemostasis by influencing the fibrino- lytic potential of blood and hemodynamic conditions.Thus, the urgency of investigations of these compounds is dictated by the fact that their effect on enzymic oxidation of polyenoic fatty acids is the central link in the mechanism of action of the majority of bioeffector lipids. The main investigations in the field of oxylipin with estab- lished structure are aimed at the study of their biosynthesis and metabolism. Labelled oxylipins have played a prominent role in these studies. For this reason, a part of the present review which is devoted to the synthesis of labelled oxylipins by enzymic methods can be regarded as an illustrative example of their application in COOH HO OOH 12-HPETE O 10-Hydroxy-11,12-epoxy-ETE HO OH HO COOH O H THETE OH H + HXA3 OH COOH HO Glu THETE HO OH SCysGly HO H H HXA3-C V P Shevchenko, I Yu Nagaev, N F Myasoedov OOH OOH OH COOH OH COOH LXB4 biological studies.In addition, studies of metabolic conversions of oxylipins also consider their properties, e.g., solubility, penetra- bility through biological membranes, etc.Labelled analogues turned out to be a helpful tool in pharmacokinetic studies of the first medicinal drugs (as a rule, natural prostaglandins and their synthetic analogues). Besides, they are indispensable components in the elaboration of highly sensitive methods for determination of oxylipins in biological samples and in reception studies (inter- action with protein molecules). Radioactive labels can be introduced into eicosanoids by several routes. Isotope exchange or chemical methods are used for labelling synthetic derivatives of oxylipins, where non-labelled prostaglandins are used as precursors or where the yields of the target products obtained by biological methods are negligibly OH COOH HOO H COOH COOH HO SCysGly HO H H HXA3-D Scheme 11 COOH OH OH HOOC (8E)-LXB4 COOH HO OH Scheme 12 COOH 12-HETE COOH H HXB3COOHMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids O OO O Epoxides small (e.g., the synthesis of epoxy compounds, prostacyclin, etc.), as well as for interconversion of labelled compounds.The isotope exchange and chemical methods are indispensable tools in the synthesis of labelled eicosanoids having a non-natural structure, (17S)-17-methyl-20-[11-3H]homoisocarba- e.g., cycline,105 fluoro prostaglandins,106 etc. These methods were used to obtain fluorinated compounds with molar radioactivities varying from 0.09 to 1.1 PBq mol71 (Table 13).Selective hydrogenation of prostaglandins of series 2 and 3 (e.g., of PGE3 to PGE2 and PGE1, Scheme 14) with gaseous tritium 107 afforded labelled preparations with molar radioactiv- ities of 1.5 ± 2.9 PBq mol71 (see Table 13). O HO OH PGE3 k3 PGE2 k2 In the latter method, the results of calculations in assumption that the rates of all reactions are described by first-order equations showed good agreement with experimental results. In this case, the ratio of hydrogenation rate constants (k1, k2, k3) were 2 : 1 : 4 for the reaction in acetone and 2 : 1 : 8 for the reaction in a ben- zene : acetone mixture. Obviously, for the synthesis of labelled 1 OH 2 34 5 OH OH OH OH o-Oxidation (o71)-Oxidation (o72)-Oxidation (o73)-Oxidation (o74)-Oxidation Arachidonic acid Cytochrome P450, epoxygenase OH HO COOH COOH C5H11 C5H11 HO OH COOH C5H11 C5H11 COOH C5H11 C5H11 OH HO COOH COOH C5H11 C5H11 OH HODiols Scheme 14 COOH k1 5,6-dihydro-PGE3 k3 k1 k2 PGE0 PGE1 k1 13,14-dihydro-PGE2 871 Scheme 13 COOH O COOH Cytochrome P450, allylic oxidation (5R/S)-HETE (8R/S)-HETE (9R/S)-HETE (11R/S)-HETE (12R/S)-HETE COOH (15R/S)-HETE COOH PGE2 from PGE3 the benzene : acetone mixture is more prefera- ble; under these conditions, the yields of [17,18-3H]PGE2 and [5,6,17,18-3H]PGE1 were 40%± 45% and 10%± 15%, respec- tively.The use of chemical methods for the synthesis of labelled compounds requires a detailed analysis of their chemical proper- ties.Thus it is known 5 that prostaglandins E are converted into PGA in acidic media, into PGB in alkaline media and into PGF after treatment with borohydrides. The latter in turn can serve as a precursor in the synthesis of prostacyclin or 6-keto-PGF1a (see Table 13). Chemical methods can also be used for the synthesis of labelled methyl esters, e.g., LTA4 and 8-epi-PGF2a (Scheme 15).108, 109, 110 The choice of catalysts for selective hydrogenation with gaseous tritium was carried out as described in Section II.1. The best results were obtained with palladium catalysts poisoned with lead diacetate. In the case of more labile compounds, triethyl- amine was added to a solvent in order to minimise the formation of side products.Selective hydrogenation with gaseous tritium can be used in the synthesis of other oxylipins, e.g., [5,6,8,9,14,15-3H]hepox- ylins B3 (Scheme 16).19 Enzymic methods with the use of labelled eicosapolyenoic acids as the starting substrates (see Scheme 7) are the most popular methods for the synthesis of labelled eicosanoids. It is impossible to mention all of the currently known data concerning this problem in a single review; therefore, we shall confine our description to the consideration of a few examples illustrating the versatility of this approach.111 Thus the use of only one enzyme, e.g., PG synthase, which catalyses the transformation of eicosa-872 Table 13.The values of molar radioactivities of oxylipins obtained by hydrogenation with gaseous tritium, isotope exchange and chemical methods. Compound Hydrogenation with gaseous tritium [13,14-3H]Leukotriene A4 ME (15S)-8-epi-[5,6-3H]PGF2a (15R)-8-epi-[5,6-3H]PGF2a [5,6,8,9,14,15-3H]HXB3 ME [17,18-3H]PGE2 [5,6,17,18-3H]PGE1 [5,6-3H]PGE1 Fluoro-15-deoxy-[5,6-3H]PGE1 11-Deoxy-[5,6,10,11-3H]PGE1 Liquid-phase isotope exchange 15-Fluoro-11,15-dideoxy-[G-3H]PGE1 ME 0.13 15-Fluoro-15-deoxy-[G-3H]PGF2a ME 0.09 5-Iodo-15-fluoro-15-deoxy-[G-3H]PGI1 ME 0.05 Chemical methods (17S)-17-methyl-20-[11-3H]homoiso- carbacycline a [8,11,12,14,15-3H]PGA1 b [5,6,8,11,12,14,15-3H]PGA2 b [5,6,8,11,12,14,15,17,18-3H]PGA3 b [11,14,15-3H]PGB1 b [5,6,11,14,15-3H]PGB2 b [5,6,11,14,15,17,18-3H]PGB3 b [8,11,12,14,15-3H]PGF1a b [5,6,8,11,12,14,15-3H]PGF2a b [5,6,8,11,12,14,15,17,18-3H]PGF3a b [8,11,12,14,15-3H]PGF1b b [5,6,8,11,12,14,15-3H]PGF2b b [5,6,8,11,12,14,15,17,18-3H]PGF3b b [5,8,11,12,14,15-3H]PGI2 MEc 6-Keto-[5,8,11,12,14,15-3H]PGF1aMEd Note: ME is methyl ester.a Sodium borotritide with a molar radioactivity of 3.04 PBq mol71 was used. b Prepared from [3H]PGE. c Prepared from [3H]PGF2aME; d Prepared from [3H]PGI2ME. polyenoic acids into PGH (its half-life in aqueous media is *5 min), made it possible to obtain a variety of labelled products, such as PDG, PGE, PGFa and 12-hydroxyheptadeca- trienoic acid (HHT) (see Scheme 8). The yield of labelled PGH and, consequently, of its degradation products depends on the concentrations of the eicosapolyenoic acid and PG synthase.Under optimum conditions, the yield of [5,6,8,9,11,12,- 14,15-3H]PGH2 was 70%; those of [5,6,8,9,11,12,14,15,- 17,18-3H]PGH3, [5,6,8,9,11,12-3H]HHT, [5,6,8,9,11,12,14,- 15-3H]PGF2a, [5,6,8,9,11,12,14,15,17,18-3H]PGF3a and [5,6,8,9,- 12,14,15,17,18-3H]PGD3 were 45%, 35%, 63%, 27% and 25%, respectively. The ratio of products of non-enzymic degradation of PGH2 in different buffer systems is variable (Table 14). The use of two-enzyme systems has made it possible to find conditions for the predominant formation of only one reaction product and to synthesise compounds which cannot be obtained by spontaneous degradation of PGH.In this case, PG synthase was supplemented with the enzymes specifically converting labile [3H]PGH into [3H]PGE, [3H]PGD, labelled thromboxane and [3H]HHT, viz., PGE synthase for the synthesis of [3H]PGE, PGD synthase for the synthesis of [3H]PGD and TX synthase for the synthesis of [3H]TXB and [3H]HHT. In a study with individual enzymes, conditions for eicosanoid biosynthesis were optimised by varying the concentrations of each enzyme and their ratios.112 All the preparations used in this study retained their activities Ref. a /PBq mol71 108 108 108 1.48 ± 1.67 1.85 ± 1.93 1.85 ± 1.93 2 ± 4 1.5 ± 1.6 2.8 ± 2.9 1.8 ± 2.0 0.9 ± 1.1 2.6 ± 3.0 19 29 29 29 29 29 29 29 29 0.68 4.1 5.2 7.3 2.9 3.7 5.7 4.6 5.8 7.3 4.6 5.8 7.3 5.0 4.9 105 29 29 29 29 29 29 29 29 29 29 29 29 29 29 V P Shevchenko, I Yu Nagaev, N F Myasoedov OHC(CH CH)2CH CH(CH2)3COOMe+ O +Ph3P(CH2)2C CC5H11 O Ê5H11 Ê5H11 3H 3H LTA4 O O 8 steps O OEt3SiO I Et3SiOHO HO R R0 3H 3H HO HO R R=OH, R0=H; R=H, R0=OH.upon storage (6 months at 760 8C). Even highly labile PGE synthase retained up to 80% of activity for 12 days at 760 8C. Inactivation of enzymes (PG synthase, 12-lipoxygenase, TX synthase) in the course of biosynthesis was taken into consider- ation using the following kinetic equation: de ds à V1 , where e is the concentration of the active enzyme species, s is the substrate concentration and V is the dimensionless parameter reflecting the number of the enzyme turnovers up to complete inactivation and depending on the chemical nature of the CHCHO+HC CCH2C C5H11C CCH2CHO 1.BuLi 2.CH2N2 C5H11 OH O 3 Pd H2 3H 3H OH C5H11 O 3H 3H 3H 3H Scheme 15 COOMe COOMe O COOMe 2 steps COOMe COOMe R0 8-epi-PGF2a Scheme 16 C(CH2)3CO2H CO2Me CO2MeMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids Table 14. The composition of PGH2 nonenzymic degradation products obtained in a PG synthase reaction in various buffers.111 Prostaglandin content (%) Buffer Buffer con- centration /mmol litre71 PGD2 PGE2 PGF2a 65 23 12 10 42 39 19 45 26 29 Tris buffer, pH 9.5 Potassium phosphate 50 buffer, pH 7.4 Potassium phosphate 50 buffer, pH 7.4 a a After 30-s incubation, hydrochloric acid was added to the reaction mixture to pH 3.0 and the incubation was continued (1 h, 25 8C).substrate. The experimental values of V for the PG synthase- catalysed conversions of arachidonic and thymnodonic acids 110 differed more than 10-fold. Quantitative calculations carried out with the use of a mathematical model and experiments with two-enzyme systems have shown 112 that the maximum yields of target products can only be obtained if the substrate is gradually added to the reaction mixture containing both enzymes. Under these conditions, the yields of target eicosanoids are restricted by enzyme inactivation in the course of the reaction. It was also found that the efficiency of a two-enzyme system can be significantly increased if some specified initial concentrations of the enzymes and the substrate are used.Thus, the strategy of enzymic synthesis of labelled eicosanoids is based on the use of artificial two-enzyme systems and can be formulated as follows.112 First, enzyme preparations must be purified to the maximum degree, since ballast proteins cause strong adsorption of the original labelled eicosapolyenoic acids, make them inaccessible for PG synthase and provoke undesirable conversions of PGH (Table 15). Second, the enzyme : substrate ratio should be an optimum, e.g., 747 and 63 nmoles of arachidonic and thymnodonic acids, respectively, should be taken per unit of PG synthase activity.Third, high yields of target products can be obtained at high concentrations of PGH-converting enzymes. Thus in the case of PGE2, a fourfold increase in the concentration of PGE synthase caused a 1.9-fold increase in the eicosanoid yield; in the case of PGD2, a threefold increase in the concentration of PGD synthase increased the eicosanoid yield twofold, whereas in the case of TXB2, a fivefold increase in the concentration of TX synthase Table 15. The effect of the degree of purification of enzyme preparations on the eicosanoid content in the reaction mixture.111 Yield (%) Enzyme preparation Synthesis of PGE2 50 80 Ram vesicular gland microsomes Partially purified PG and PGE synthases Synthesis of PGE3 30 45 Ram vesicular gland microsomes Partially purified PG and PGE synthases Synthesis of TXB2 15 80 Ram vesicular gland microsomes and human platelets Partially purified PG and thromboxane synthases Synthesis of 12-HETE 35 50 Human platelet microsomes Partially purified 12-lipoxygenase 873 Table 16.The yields of eicosanoids in two-enzyme systems with simulta- neous (I) and separate (II) addition of PG synthase and PGH-converting enzyme to the reaction mixture.111 Yield (%) Eicosanoid I II33 25 24 18 23 17 13 10 PGD2 PGD3 TXB2 TXB3 increased the yield of the target product 2.4-fold. Probably, this is associated with the necessity for creation of conditions where the rates of enzymic conversions of PGH will markedly exceed the rate of its spontaneous degradation.Fourth, the procedure of conducting the biosynthesis plays a very important role. The yield of the target product can be increased 1.5 ± 2-fold by separate addition of the enzymes to the reaction mixture (Table 16). The optimum time of incubation of an eicosapolyenoic acid with PG synthase prior to the addition of the second enzyme is 1.5 ± 2 min. Within this period, the accumu- lation of PGH is a maximum, i.e., the rate of its formation by the PG synthase reaction coincides with the rate of its nonenzymic degradation. Fifth, the temperature dependence of the yields of these eicosanoids in two-enzyme synthesis has a broad optimum (20 ± 25 8C).This implies that the optimal yields of labelled eicosanoids can be attained at room temperature. This strategy was successfully used in the synthesis of various highly labelled oxylipids (Table 17). Considerable interest recently arose in the studies of function- ing of various enzymic subsystems under heterophasic condi- tions.112 This interest is due to the fact that the catalytic activity and stability of many enzymes under reaction conditions can be significantly increased. Thus a search for the ways aimed at increasing the `productivity' of enzymes involved in prostaglan- din synthesis, which undergo irreversible inactivation in the course of the reaction, has demonstrated high efficiency of macroheterogenous systems (water-in-octane and octane-in- water).It was found that the activities of the enzymes of eicosanoid synthesis (PG synthase, PGE synthase, TX synthase, 12-lipoxygenase) were retained in all heterophasic systems under study. The highest values of activity, V, for the PG synthase Table 17. The values of molar radioactivities of various oxylipins obtained by enzymic methods. Ref. Compound a /PBq mol71 29 29 29 29 29 29 113 113 113 114 114 114 115 115 115 115 [8,11,12,14,15-3H]PGE1 [5,6,8,11,12,14,15-3H]PGE2 [5,6,8,9,11,12,14,15-3H]PGF2a [5,6,8,9,12,14,15-3H]PGD2 [5,6,8,9,11,12,14,15-3H]TXB2 [5,6,8,11,12,14,15,17,18-3H]PGE3 [5,6,8,9,12,14,15-3H]PGD2 13,14-Dihydro-15-keto-[3H]PGD2 a [9a,11b-3H]PGF2 a [15-3H]HETE [5,15-3H]DHETE [8,15-3H]DHETE 12-[3H]HPETE [12-3H]HETE [15-3H]HPETE [15-3H]HETE a Prepared from [5,6,8,9,12,14,15-3H]PGD2.4.65 5.90 6.10 5.98 5.27 7.35 4.44 776.20 5.10 6.00 5.26 5.26 5.37 5.37874 reaction were observed with octane-in-water emulsions stabilised with native microsomal membranes. It was also found that if water : octane : lyophilised membranes ratios were close to those at which the system is stratified, the catalytic activity of PG synthase increased nearly 10-fold in comparison with systems containing no organic solvents. Moreover, the main advantage of this approach to the synthesis of labelled preparations is that in stabilised emulsions the labelled eicosapolyenoic acid is predom- inantly accumulated in the organic solvent, whereas the oxidation products (e.g., PGE2) are detected only in the aqueous phase.As a result, the negative effect of radiolysis on the formation of target labelled eicosanoids is reduced and the reaction products, the enzyme and the nonconsumed substrate are readily separated by centrifugation. In addition to these methods for the synthesis of labelled oxylipins, special mention should be made of the possibility of synthesis of certain labelled prostaglandins by enzymic modifica- tion of other labelled prostaglandins. The use of chemical methods is coupled with the formation of stereoisomers and the necessity to perform additional steps for protection of functional groups with similar reactivities. Enzymic syntheses of prostaglandins starting from other prostaglandins as substrates are devoid of these drawbacks.Thus 15-keto-PGF2a, 13,14-dihydro-15-keto-PGF2a, PGE2, 15-keto-PGE2, 13,14-dihydro-15-keto-PGE2 and 9a,11b- PGF2 were synthesised from [3H]PGF2a and [3H]PGD2 using 9- and 15-hydroxyprostaglandin dehydrogenases, PGF synthase, etc., isolated from various sources (see Table 17).111 ± 113 Labelled 15-HETE, 5,15-DHETE, 8,15-DHETE and 12-HETE were synthesised in up to 60%± 65%, 20% ± 25%, 10%± 15% and 21%± 51% yields, respectively, using enzymes of the lipoxygenase pathway of eicosapolyenoic acid metabolism (e.g., 15- and 12-lipoxygenase) (see Schemes 11 and 12),. The optimum conditions for these reactions were selected in the studies of the dependence of the yields of target products on the presence of appropriate additives in the reaction mixture as well as on the substrate : enzyme ratio, substrate and enzyme concentrations, degree of enzyme purification, etc.111, 114, 115 An analysis of the data presented in Table 18 suggests that if the 12-lipoxygenase reaction is conducted in the presence of certain reducing agents, the yield of 12-HETE increases consid- erably.However, the use of enzymes is not always successful. Thus labelled leukotriene B4 was synthesised in but 0.1% yield using enzymes of the lipoxygenase pathway of eicosapolyenoic acid metabolism,29 whereas the yields of lipoxins 116 were negligibly small. These examples demonstrate that the use of enzymic methods demands a careful selection of reaction conditions and a thorough analysis of factors decreasing the yield of target products.Table 18. The yields of 12-HETE from [3H]arachidonic acid depending on the nature of the reducing agents.111 Yield (%) Reducing agent 34 51 49 44 36 34 32 32 31 29 7Dithiothreitol Glutathione L-Cysteine L-Tryptophan Aniline L-Adrenaline L-Methionine NADH Phenol IV. Steroids The fundamental problem of a relationship between the structure of biologically active compounds and their activities is still very V P Shevchenko, I Yu Nagaev, N F Myasoedov urgent. Its solution would make it possible to use both natural and modified steroids as medicinal drugs.It is known that steroid hormones fulfil several functions in the organism, each being mediated by a specific receptor. Studies of the mechanism of steroid ± receptor interaction shed light on what fragments of the steroid molecule are responsible for one or another of its biological function. This knowledge will allow the minimising of potential side effects of natural steroids used as medicinal drugs, on the one hand, and influencing their particular biological function by modifying natural steroid molecules, on the other hand.22 Tritium-labelled steroids possessing estrogenic, androgenic and anabolic activities are necessary for estimating the biological characteristics of pharmacological preparations used in studies on reproduction and increase of the productivity of animals. Labelled analogues of steroid hormones are used in investigation of various metabolic processes in normal and pathological states, in diag- nostics of diseases and quantitation of steroids in various bio- logical fluids.117 The topicality of synthesis of labelled steroids is connected with the application of their natural precursors in the treatment of various diseases, such as atherosclerosis, asthma, rheumatism, eczemas, allergies, polyarthritis, gout, radiation disease, diabetes mellitus, etc.The label is introduced into steroids by various routes. Let us consider the methods used in the synthesis of labelled cholesterol as an example. The data presented in Tables 19 ± 21 suggest that dehalogena- tion, solid-phase isotope exchange and isotope exchange with Table 19.The values of molar radioactivities of cholesterol obtained by isotope exchange and by chemical methods. Method of labelling Using recoil atoms 6Li(n,a)3H Isotope exchange: incubation in an atmosphere of gaseous tritium (the Wiltzbach method) activation of the reaction with an electric charge addition of a carrier microwave irradiation liquid-phase isotope exchange solid-phase isotope exchange isotope exchange with tritiated water Chemical methods: reduction with metal tritides dehalogenation of the 7-halogeno derivative dehalogenation of the 2,2-dihalogeno derivative hydrogenolysis of the 7,7-disulfur- containing derivative Table 20.The relative degree of isotope exchange (a) and yields of steroids depending on duration of the reaction.121 Estriol Progesterone t /h a a yield (%) 22 95 48 85 81 7560 18 10 2 48 95 4 75 90 6 97 85 100 70 98 20 100 10 100 95 96 10 24 50 Ref. a /TBq mol71 3, 4, 8, 118 0.0022 3, 4, 8, 118 74.1 3, 4, 8, 118 185.2 3, 4, 8, 118 3, 4, 8, 118 4119 121 0.0445 0.9207 1.7 997.3 218.0 118 120 785.2 370.0 118 1129.6 122 300.1 Cholesterol a yield (%) yield (%) 27 95 69 85 84 8065 15 10 100 95 94Methods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids Table 21. The values of molar radioactivities of steroids in isotope exchange with tritiated water.121 Compound a /PBq mol71 0.248 0.400 0.218 0.200 0.226 Progesterone Estradiol Cortisol Methylandrosta-1,4-dien-17a-ol Methylandrost-5-ene-3b,17a-diol tritiated water are the most promising methods.118 ± 122 The label can successfully be introduced into androstanes, pregnanes and some other steroids by a modified Wiltzbach procedure (see Ref.123). Labelled steroids can also be prepared by biological methods, 120 hydrogenation and selective hydrogenation of the corresponding precursors 124 ± 127 in combination with other chemical methods (Table 22).129 ± 133 Dehalogenation allows one to prepare stereospecifically labelled steroids. Molar radioactivities of the thus synthesised [7-3H]cholesterol, [7-3H]pregnenolone and [7-3H]dehydroepi- androsterone amount to 0.37 PBq mol71, and those of aromatic 2,4-dibromosteroids, to 1.1 PBq mol71.118 ± 120 Biological methods entail the conversion of labelled steroids into other steroids.Thus [16-3H]aldosterone (molar radioactivity 0.08 PBq mol71) was prepared from [16-3H]progesterone (0.5 PBq mol71) using bovine adrenal cortex enzymes, and [16-3H]cortisol (molar radioactivity 0.07 PBq mol71) was obtained from [16-3H]corticosterone with a molar radioactivity of 0.14 PBq mol71.120 It is known that the reactivity of double bonds of unsaturated steroid hormones decreases in the following order: 1,2>6,7>16,17>4,5. By changing the reaction conditions (solvents, catalysts, etc.), one can perform selective hydrogena- tion of a definite bond and obtain the target steroid with high molar radioactivity.118, 120 This method seems to be the most versatile for radioactive labelling of steroids; this results in steroids with high molar radioactivities.130, 133 The syntheses of 17a-ethynyl-19-nor[9,11-3H]testosterone and 18-methyl-17a- ethynyl-17b-hydroxy[14a,15a-3H]estr-4-en-3-one are shown in Schemes 17 and 18.These schemes provide illustrative examples of the potential- ities of chemical approaches to the synthesis of complex labelled compounds which are very labile under conditions of treatment with gaseous tritium in the presence of catalysts. If biological studies require labelled saturated steroids, solid-phase hydroge- nation is the method of choice.In this case, the molar activity of steroids increases owing to the isotope exchange (Table 23).120 Selective hydrogenation is indispensable for the synthesis of labelled derivatives of natural steroids.22, 129, 134 As can be seen from Scheme 19, several tritium-labelled, biologically active progesterone analogues can be prepared from the same precursor (see Table 22).129 OH 3H 3H2 3H Pd/C MeO MeO OH 3 3 H H 3H 3H CrO3 H2SO4, Me2CO O O Table 22. The values of molar radioactivities of labelled steroids. Compound Liquid-phase isotope exchange [G-3H]Cholestan-3b-ol b-[G-3H]Sitosterine Solid-phase isotope exchange 5a-[G-3H]Dihydrotestosterone 5a-[G-3H]Dihydrotestosterone enantate Hydrogenation with gaseous tritium 17a-Ethynyl-17b-acetoxy-19-nor- [6,7-3H]androst-4-en-3-one oxime 16a,17a[1,2-3H]Cyclopropanopregn- 4-ene-3,20-dione 6a-Methyl-16a,17-cyclohexano- [1,2-3H]pregna-1,4-diene-3,20-dione 16a,17a-[30,40-3H]Cyclohexanopregn- 4-ene-3,20-dione 16a,17a-[30,40-3H]Cyclophexanopregna- 1,4-diene-3,20-dione 16a,17a-Cyclohex-30-eno[1,2-3H]pregn- 4-ene-3,20-dione 16a,17a-[1,2,30,40-3H]Cyclohexanopregn- 4-ene-3,20-dione 5a,16a,17a-[1,2,30,40-3H]Cyclohexano- pregnane-3,20-dione 5b,16a,17a-[1,2,30,40-3H]Cyclohexano- pregnane-3,20-dione 3-O-Methyl-[9,11-3H]estradiol 5a-[16,17-3H]Androstan-3-one [4,5b-3H]Cholestan-3-one 17a-Hydroxy-[1,2-3H]androst-4-en-3-one 2.1 1.5 ± 1.7 1.8 ± 1.9 1.5 17a-Hydroxy-5a-[1,2,4,5-3H]androstan-3-one 1.9 2.1 18-Methyl-3-methoxy-[14a,15a-3H]-estra- 1,3,5(10),8-tetraen-17b-ol acetate The selectivity of the process is achieved through the use of homogenous catalysts.Kinetic studies have shown that hydro- genation of the C(1)=C(2) bond occurs 1.5 ± 2 times more easily than that of the C(30)=(C40) bond. However, reduction of the C(30)=(C40) bond is still the fast reaction, and after 1 ± 1.5 h the reaction mixture did not contain even more stable 16a,17a[1,2-3H]cyclohex-30-enopregn-4-ene-3,20-dione. Under optimum conditions (40 min), the reaction mixture contained 10%± 15% of 16a,17a-[30,40-3H]cyclohexanopregna-1,4-diene- 3,20-dione, 20%± 25% of 16a,17a-[1,2-3H]cyclohex-30-eno- pregn-4-ene-3,20-dione {[3H]CHE} and 60%± 70% of 16a,17a- [1,2,3040-3H]cyclohexanopregn-4-ene-3,20-dione. Hydrogenation of the latter in the presence of heterogenous catalysts yielded two OH OH 3H Na, NH3 3H THF MeO O 3H 3H LiC CH EDA, DMSO O 875 Ref.a/PBq mol71 004 004 0.0008 0.0021 120 120 0.56 1.50 128 1.6 022 1.5 022 1.6 022 2.0 129 1.5 129 1.6 129 3.2 129 3.2 129 3.2 130 014 131 132 132 133 Scheme 17 MeOH HCl OH C CH876 MeO MeO O 3HO Table 23. The values of molar radioactivities and the yields of labelled steroids obtained by a solid-phase method.120 T /8C t/h Starting compound�17-hydroxyandrosta-1,4,6-trien-3-one 5.0 501.5 1.5 120 120 150 150 Starting compound�androsta-1,4,6-triene-3,17-dione 1.5 1.5 26 26 26 20 20 50 50 50 additional labelled steroids, viz., 5a,16a,17a-[1,2,3,40-3H]cyclo- hexanopregnane-3,20-dione and 5b,16a,17a-[1,2,30,40-3H]cyclo- hexanopregnane-3,20-dione.The use of labelled reagents, e.g., metal borotritides, has been mentioned above (see Table 19). Reactions with these reagents OAc 3H2 Pd/BaSO4 MeO O HC CMgBr 3 3 H H O 3H2 3H 3H O 3 3H H2 3HO O 3H 3H 3H H Target product a /PBq mol71 6.65 5.10 5.55 4.5 androstanediol androstanol androstanediol androstanol 3.10 3.14 4.85 6.5 6.5 androstenediol testosterone androstanedione androstanediol androstanolone V P Shevchenko, I Yu Nagaev, N F Myasoedov OAc Na NH3 3 3H H 3H MeO OHC CH 3 3H H O MeO O 3H 3H2 3HO O 3 3H2 3HH H2 O 3 3 H 3 3 H HH O H and tritium-labelled methyl iodide and its derivatives afforded labelled steroids, such 4-3H]- and [16-3H]cholesterols, 1a,3b- dihydroxy-(20S)-[4-3H]-(3-hydroxy-3-methylbutyloxy)pregna- 5,7-diene, 1a,25-dihydroxy-22-oxavitamin D3 and some other compounds.3, 4, 118, 120, 135 The main disadvantage of this method is that the molar radioactivities of the resulting products depend on those of the labelled reagents, which are sometimes too low for conducting biological experiments.Presumably, syntheses of labelled preparations can be based on the use of non-labelled reagents and labelled steroidal precursors and reactions of condensation, oxidation, hydrolysis, acetylation or reduc- tion.3, 136 Several labelled 16a,17a-cycloalkano- and 16a,17a-cycloal- keno-substituted progesterones with different sizes of the addi- tional ring D0 were synthesised,22, 129 and their interactions with the progesterone receptor were studied for the first time in direct experiments.The steroid ± receptor binding was studied using labelled 16a,17a-cyclohexanopregn-4-ene-3,20-dione {[3H]CHA}, 16a,17a-cyclopropanopregn-4-ene-3,20-dione {[3H]CPA} and proteins from the soluble fraction of rat uterus homogenates responsible for progesterone reception. The interaction of CPA and CHA with the progesterone receptor and other rat uterus proteins was assayed by cross-inhibitory analysis, which is based on the displacement of a labelled ligand from complexes of proteins with non-labelled steroids. The values of relative com- Scheme 18 OH3H OHC CH 3H 3H Scheme 19 OMethods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids Table 24.The values of relative competitive activity (RCA) of natural progesterone calculated from the displacement of labelled ligands from complexes with the progesterone receptor from the soluble fraction of pregnant rat uterus. RCAa Unlabelled competitor Labelled ligand [3H]Progesterone [3H]CPA [3H]CHA 1.02 0.79 0.99 0.57 1.57 1.00 CPA CHA CPA CHA CPA CHA a The progesterone activity was taken as unity.petitive activity (RCA) were calculated from the concentration ratios of non-labelled compounds inducing 50% inhibition of specific binding of the labelled ligand (Table 24). It was shown that a progesterone receptor in the soluble fraction of a rat uterus homogenate is a protein that specifically binds CHA and CPA and manifests almost identical parameters of competitive binding. The same regularity was observed with [3H]CHE. This compound binds by the protein which is identical with the progesterone receptor, no other specific binding species in the soluble fraction of rat uterus were detected. This conclusion was confirmed by analysis of CHE inhibitory effect on [3H]progesterone binding. During biological experiments, no metabolic conversions of labelled steroids occur.No significant differences were found in the mode of interaction of [3H]CHE and [3H]progesterone with a receptor isolated from the cells of differ- ent animals either. It is interesting to note that CHE as a representative of this class of steroids manifested higher activity in a pregnancy preservation test and a relatively low activity in an endometrium proliferation test upon systemic administration than progesterone.137 These examples illustrate the possibility of the synthesis of steroid hormones with separated biological functions as regards progestagenic activity. V. Conclusion This review describes various methods for introducing tritium labels into biologically active compounds.Recent advances in radiochemistry have made it possible to solve numerous problems concerning the study of processes determining the vital activity of cells and the whole organism. There is no doubt, however, that further progress in medicine and biology will demand elaboration of novel and the development of existing methods of synthesis of labelled compounds. Studies on effects of micromodifications of molecules of various steroids are currently under way. These studies are necessary for the design of a new generation of medicinal drugs possessing strictly specific activity and having no adverse side effects.138 New approaches to modification of oxylipins and polyenoic fatty acids are being developed in order to elucidate the relationship between the structure and function of these compounds, since the use of natural oxylipins as medicinal drugs is strongly restricted by the wide range of their pharmaco- logical action and rapid metabolism.106 The synthesis of some derivatives of polyunsaturated acids (e.g., dicarboxylic acids) required for the analysis of enzyme ± substrate binding of mam- malian 15-lipoxygenases has been carried out.This is important for continuing investigations in the field of gastroenterology, cardiology, angiology, oncology, etc.139, 140 A vast variety of compounds (fungicides, immunodepressants, medicinal drugs influencing the central nervous system, pheromones, etc.), some- times having an exotic structure, are being synthesised.141 ± 143 Considerable effort is devoted to a search for novel approaches to the synthesis of tritium-containing preparations.These investigations include an elaboration of original procedures for the synthesis of new catalysts and labelled reagents which 877 represent fragments of biologically active compounds or sources of radioactive labels, methods of enzymic synthesis of labelled oxylipins using organic solvents are being developed, etc.144 ± 147 Thus data were obtained showing that enzymes functioning within micelles possess higher stability which increases the efficiency of synthesis of these labelled preparations.148 References 1. E V Dyatlovitskaya, V V Bezuglov Biokhimiya 63 3 (1998) a 2. G V Sidorov, N F Myasoedov Usp. Khim. 68 254 (1999) [Russ.Chem. Rev. 68 229 (1999)] 3. E A Evans Tritium and its Compounds (London: Butterworth Publishers, 1974) 4. V P Shevchenko, N F Myasoedov, in Isotopes in the Physical and Biochemical Sciences (Labelled Compounds) (Eds E Buncel, J R Jones) (New York: Elsevier, 1987) Vol. 1, Pt. A, p. 237 5. V P Shevchenko, N F Myasoedov, in Isotopes in the Physical and Biochemical Sciences (Labelled Compounds) (Eds E Buncel, J R Jones) (New York: Elsevier, 1991) Vol. 1, Pt. B, p. 179 6. V P Shevchenko, I Yu Nagaev, N F Myasoedov J. Label. Comp. Radiopharm. 27 1195 (1989) 7. A V Filikov, N F Myasoedov J. Catal. 130 569 (1991) 8. N F Myasoedov J. Label. Comp. Radiopharm. 33 391 (1993) 9. E V Laskatelev, Candidate Thesis in Chemical Sciences, Institute of Organoelement Compounds, Russian Academy of Sciences, Moscow, 1997 10.V P Shevchenko, in II Vsesoyuz. Soveshch. po Probleme `Fiziologi- cheski Aktivnye Soedineniya, Mechennye Radioaktivnymi Izotopami' (Tez. Dokl.), Zvenigorod, 1988 [The Second All-Union Meeting on the Problem `Physiologically Active Compounds Labelled by Radioactive Isotopes (Abstracts of Reports), Zvenigorod, 1988] p. 35 11. V P Shevchenko, I Yu Nagayev, N F Myasoedov, A Susan, in The Sixth International Symposium on the Synthesis and Applications of Isotopes and Isotopically Labelled Compounds (ISSAIILC6) (Abstracts of Reports), Philadelphia, PA, 1997 p. 73 12. Yu A Borisov, Yu A Zolotarev, E V Laskatelev, N F Myasoedov Izv. Akad. Nauk, Ser. Khim. 1852 (1996) b 13.Yu A Borisov, Yu A Zolotarev, E V Laskatelev, N F Myasoedov Izv. Akad. Nauk, Ser. Khim. 428 (1997) b 14. V P Shevchenko, A V Potapova, N F Myasoedov Radiokhimiya 31 (5) 75 (1989) c 15. V P Shevchenko, N F Myasoedov Radiokhimiya 31 (3) 95 (1989) c 16. V P Shevchenko, I Yu Nagaev, N F Myasoedov, D V Kuklev, A B Imbs, N A Latyshev, V V Bezuglov Radiokhimiya 31 (6) 119 (1989) c 17. D V Kuklev, V P Shevchenko, I Yu Nagaev, N A Latyshev, V V Bezuglov Bioorg. Khim. 17 1574 (1991) d 18. V P Shevchenko, G I Myagkova, T Yu Lazurkina, P M Dyomin, S I Shram, D A Zabolotsky, I Yu Nagaev, Yu Yu Belosludtsev, R P Evstigneeva, N F Myasoedov J. Label. Comp. Radiopharm. 27 1177 (1989) 19. I Yu Nagaev, V P Shevchenko, I V Ivleva, G I Myagkova, N F Myasoedov Radiokhimiya 37 263 (1995) c 20.J W Coe, C R Hawes, P Towers, in The 5th International Symposium on the Synthesis and Applications of Isotopes and Isotopically Labelled Compounds (Abstracts of Reports), Strasbourg, 1994 p. 127 21. U Pleiss, U Rosentreter, J Stoltefuss, O Behner, in The Sixth International Symposium on the Synthesis and Applications of Isotopes and Isotopically Labelled Compounds (ISSAIILC6) (Abstracts of Reports), Philadelphia, PA, 1997 p. 71 22. A V Kamernitskii, I S Levina, L E Kulikova, T N Galakhova, V P Shevchenko, I Yu Nagaev, N F Myasoedov, A N Smirnov, E V Pokrovskaya, T A Shchelkunova Izv. Akad. Nauk, Ser. Khim. 1532 (1997) b 23. G S Kogteva, V V Bezuglov Biokhimiya 63 6 (1998) a 24. R Grabe, C Sumida, E A Nunez J.Lipid Med. Cell Sign. 9 91 (1994) 25. S D Clarke, D B Jump Prog. Lipid Res. 32 139 (1993) 26. M Haourigui, G Vallett,M E Martin, C Sumida, C Benassayag, E A Nunez Steroids 59 46 (1994) 27. J F Mead J. Lipid Res. 25 1517 (1984)878 28. V P Shevchenko, Candidate Thesis in Chemical Sciences, Institute of Molecular Genetics, Academy of Sciences of the USSR, Moscow, 1981 29. V P Shevchenko, Doctoral Thesis in Chemical Sciences, Institute of Molecular Genetics, Russian Academy of Sciences, Moscow, 1992 30. D V Kuklev, V G Rybin, A B Imbs, V V Bezuglov Bioorg. Khim. 19 1122 (1993) d 31. E D Matveeva, A L Kurts, Yu G Bundel' Usp. Khim. 55 1198 (1986) [Russ. Chem. Rev. 55 672 (1986)] 32. G D Prestwich, in Isotopes in the Physical and Biochemical Sciences (Labelled Compounds) (Eds E Buncel, J R Jones) (New York: Elsevier, 1991) Vol.1, Pt. B, p. 137 33. V P Shevchenko, S V Faradzheva, I Yu Nagaev, N F Myasoedov Radiokhimiya 40 84 (1998) c 34. S V Faradzheva, Candidate Thesis in Chemical Sciences, Moscow State University, Moscow, 1990 35. V P Shevchenko, I Yu Nagaev, N F Myasoedov, in Vseros. Konf. `50 Let Proizvodstva i Primeneniya Izotopov v Rossii' (Tez. Dokl.), Obninsk, 1998 [The All-Union Conference `50 Years of Production and Application of Isotopes in Russia' (Abstracts of Reports), Obninsk, 1998] p. 33 36. B Devadas, S P Adams, J I Gordon J. Label. Comp. Radiopharm. 29 157 (1991) 37. K N Semenenko, V V Burnasheva, N A Yakovleva, E A Ganich Izv. Akad.Nauk, Ser. Khim. 214 (1998) b 38. A R Ubbelhode J. Chem. Soc. 1143 (1950) 39. T R Gibb, in Progress in Inorganic Chemistry (New York: Interscience, 1962) p. 315 40. S Tanaka, A Yashiro, H Tasaki, Y Nakashima Lipids 33 385 (1998) 41. V Di Martso Biokhimiya 63 16 (1998) a 42. T Elbert J. Label. Comp. Radiopharm. 27 107 (1989) 43. R D Schmid, R Verger Angew. Chem., Int. Ed. Engl. 37 1608 (1998) 44. T Sugiura, S Kondo, A Sukagawa, S Nakane, A Shinoda, K Iton, A Yamashita, K Waku Biochem. Biophys. Res. Commun. 215 89 (1995) 45. S Munro, K L Thomas, M Abu-Shaar Nature (London) 365 61 (1993) 46. R Mechoulam, S Ben-Shabat, L Hanus,M Ligumsky, N E Kaminski, A R Schatz, A Gopher, S Almog, B R Martin, D R Compton, R G Pertwee, G Griffin, M Bayewitch, J Barg, Z Vogel Biochem.Pharmacol. 50 83 (1995) 47. S Maurelli, T Bisogno, L De Petrocellis, A Di Luccia, G Marino, V Di Marzo FEBS Lett. 377 82 (1995) 48. E Fride, T Bisogno, V Di Marzo,M Bayewitch, Z Vogel, R Mechoulam, in Abstract Book of the Society for Neuroscience Annual Meeting, New Orleans, 1997 p. 25 49. E Di Tomaso, M Beltramo, D Piomelli Nature (London) 382 677 (1996) 50. V A Tkachuk Biokhimiya 63 47 (1998) a 51. V I Kulikov, G I Muzya Biokhimiya 63 57 (1998) a 52. N V Prokazova, N D Zvezdina, A A Korotaeva Biokhimiya 63 38 (1998) a 53. A Tokumura, T Tsutsumi, Y Nishioka, H Tsukatani Chem. Phys. Lipids 86 75 (1997) 54. U H Do, Y Hong, P Tam, P Srinivasan J. Label. Comp. Radiopharm. 38 117 (1996) 55. L Binaglia, G Alunni-Bistocchi, M Oriando, P Orlando, G Rosa, R Trenta J.Label. Comp. Radiopharm. 29 95 (1991) 56. V P Shevchenko,M L Tsirenina, J G Molotkovsky, L D Bergelson Chem. Phys. Lipids 15 95 (1975) 57. Z I Terashita, S Tsushima, Y Yoshioka, H Nomura, Y Inada, K Nishikawa Life Sci. 32 1975 (1983) 58. S D Shukla, D J Hanahan Arch. Biochem. Biophys. 227 626 (1983) 59. S D Shukla Lipids 26 1028 (1991) 60. W A Devane, L Hanus, A Breuer, R G Pertwee, L A Stevenson, G Griffin, D Gibson, A Mandelbaum, A Etinger, R Mechoulam Science 258 1946 (1992) 61. E Fride, R Mechoulam Eur. J. Pharmacol. 231 313 (1993) 62. J N Crawley, R L Corwin, J K Robinson, C C Felder, W A Devane Pharmacol. Biochem. Behav. 46 967 (1993) 63. P B Smith, D R Compton, S P Welch, R K Razdan, R Mechoulam, B R Martin J.Pharmacol. Exp. Ther. 270 219 (1994) 64. Z Vogel, J Barg, R Levy, D Saya, E Heldman, R Mechoulam J. Neurochem. 61 352 (1993) V P Shevchenko, I Yu Nagaev, N F Myasoedov 65. C C Felder, E M Briley, J Axelrod, J T Simpson, K Mackie, W A Devane Proc. Natl. Acad. Sci. USA 90 7656 (1993) 66. SRChilders, T Sexton,MB Roy Biochem. Pharmacol. 47 711 (1994) 67. K Makie, W A Devane, B Hille Mol. Pharmacol. 44 498 (1993) 68. L Hanus, A Gopher, S Almog, R Mechoulam J. Med. Chem. 36 3032 (1993) 69. J Priller, E M Briley, J Mansouri, W A Devane, K Mackie, C C Felder Mol. Pharmacol. 48 288 (1995) 70. D G Deutsch, S A Chin Biochem. Pharmacol. 46 791 (1993) 71. K K Kruszhka, R W Gross J. Biol. Chem. 269 14 345 (1994) 72.W A Devane, J Axelrod Proc. Natl. Acad. Sci. USA 91 6698 (1994) 73. V Di Marzo, A Fontana, H Cadas, S Schinelli, G Cimino, J-C Schwartz, D Piomelli Nature (London) 372 686 (1994) 74. N Ueda, K Yamamoto, S Yamamoto, T Tokunaga, E Shirakawa, H Shinkai,M Ogawa, T Sato, I Kudo, K Inoue, H Takizawa, T Nagano,M Hirobe, N Matsuki, H Saito Biochim. Biophys. Acta 1254 127 (1995) 75. M Yu Bobrov, G S Kogteva, E V Fomina-Ageeva, G N Zinchenko, N M Gretskaya, D V Kuklev, V V Bezuglov, in Modifitsirovannye Zhirnye Kisloty N-6- i N-3-Semeistv. Mediko- biologicheskie, Biokhimicheskie i Biotekhnologicheskie Aspekty (Tez. Dokl.), Vladivostok, 1995 [Modified Fatly Acids of the N-6- and N-3-Families. Medicobiological, Biochemical and Biotechno- logical Aspects (Abstracts of Reports), Vladivostok, 1995] p.19 76. A J Hampson,W A Hill, M Zan-Phillips, A Makriyannis, E Leung, R M Eglen, L M Bornheim Biochim. Biophys. Acta 1259 173 (1995) 77. F Desarnaud, H Cadas, D Piomelli J. Biol. Chem. 270 6030 (1995) 78. S I Rogov, V P Shevchenko, I Yu Nagaev, N F Myasoedov, V V Bezuglov,M Yu Bobrov, V M Fedoseev Radiokhimiya 39 458 (1997) c 79. H W Cook,W E M Lands Can. J. Biochem. 53 1220 (1975) 80. C P A Van Os, G P MRijke-Schilder, H van Halbeek, J Verhagen, J F G Vliegenthart Biochim. Biophys. Acta 663 177 (1981) 81. R T Holman, P O Egwim,W W Christie J. Biol. Chem. 244 1149 (1969) 82. E V Dyatlovitskaya Biokhimiya 63 67 (1998) a 83. A H Merrill, S Nimkar Jr, D Menaldino, Y A Hannun, C Loomis, R M Bell, S R Tyagi, J D Lambeth, V L Stevens, R Hunter, 84.R T Dobrowsky, M H Werner, A M Castellino, M V Chao, 85. T Okazaki, A Bielawska, R M Bell, Y A Hannun J. Biol. Chem. 86. A Bielawska, C M Linardic, Y A Hannun FEBS Lett. 307 211 87. L Riboni, A Prinetti, R Bassi, A Caminiti, G Tettamanti J. Biol. 88. R Olshefski, B Taylor, A Heitger, A Hasegawa, S Ladisch Eur. J. 89. L Ji, G Zhang, S Uematsu, Y Akahori, Y Hirabayashi FEBS Lett. 90. R T Dobrowsky, C Kamibayashi, M C Mumby, Y A Hannun D C Liotta Biochemistry 28 3138 (1989) Y A Hannun Science 265 1596 (1994) 265 15 823 (1990) (1992) Chem. 270 26 868 (1995) Biochem. 241 47 (1996) 358 211 (1995) J. Biol. Chem. 268 15 523 (1993) 91. R A Wolff,R T Dobrowsky, A Bielawska, LMObeid, YA Hannun J.Biol. Chem. 269 19 605 (1994) 92. J Y Lee,Y A Hannun, L M Obeid J. Biol. Chem. 271 13 169 (1996) 93. M E Venable, A Bielawska, L M Obeid J. Biol. Chem. 271 24 800 94. V P Shevchenko, I Yu Nagaev, A V Potapova, N F Myasoedov 95. R Birk, G Brenner-Weib, A Giannis, K Sandhoff, R R Schmidt 96. T Toyokuni,M Nisar, B Dean, S Hakomori J. Label. Comp. (1996) Radiokhimiya 37 265 (1995) a J. Label. Comp. Radiopharm. 29 289 (1991) Radiopharm. 29 567 (1991) 97. I A Mikhailenko, O V Korepanova, N V Prokazova Biol. Membrany 12 82 (1995) 98. L D Bergelson, A G Bukrinskaya, N V Prokazova, G I Shaposhnikova, S L Kocharov, V P Shevchenko, G V Kornilaeva, E V Fomina-Ageeva Eur. J. Biochem. 128 467 (1982) 99. I V Ivanov, N V Groza, D M Kochev, G I Myagkova Bioorg.Khim. 24 83 (1998) d 100. C N Serhan Biochim. Biophys. Acta 1212 1 (1994)Methods for the synthesis of tritium-labelled fatty acids and their derivatives, oxylipins and steroids 101. C R Pace-Asciak Gen. Pharmacol. 24 805 (1993) 102. P M Demin, G I Myagkova, R P Evstigneeva Bioorg. Khim. 17 900 (1991) d 103. S D Varfolomeev, A T Mevkh Prostaglandiny�Nizkomoleku- lyarnye Bioregulyatory (Prostaglandins as Low-Molecular Bioregulators) (Moscow: Moscow State University, 1985) 104. G N Petrukhina, V A Makarov Biokhimiya 63 111 (1998) a 105. S Sugiura, S Kurozumi J. Label. Comp. Radiopharm. 38 129 (1996) 106. V V Bezuglov, Doctoral Thesis in Chemical Sciences, M M Shemyakin and Yu A Ovchinnikov Institute of Bioorganic Chemistry, Russia Academy of Sciences, Moscow, 1997 107.V P Shevchenko, I Yu Nagaev, N F Myasoedov Radiokhimiya 33 172 (1992) c 108. V P Shevchenko, I Yu Nagaev, N F Myasoedov, A Guy, T Durand, A Vidal, J-P Vidal, J-C Rossi, V V Bezuglov J. Label. Comp. Radiopharm. (1999) (in the press) 109. M Garcia, T Durand, J-C Rossi, S Zarini, M Bolla, G Folco, P Wheelan, A Sala Bioorgan. Med. Chem. Lett. 7 105 (1997) 110. A Guy, T Durand, J-P Vidal, J-C Rossi Tetrahedron Lett. 38 1543 (1997) 111. S I Shram, T Yu Lazurkina, V P Shevchenko, I Yu Nagaev, N F Myasoedov J. Label. Comp. Radiopharm. 41 741 (1998) 112. S I Shram, Candidate Thesis in Chemical Sciences, Institute of Molecular Genetics, Russian Academy of Sciences, Moscow, 1995 113. S I Shram, T Yu Lazurkina, V P Shevchenko, I Yu Nagaev, N F Myasoedov J. Label. Comp. Radiopharm. 34 359 (1994) 114. I V Ivleva, V P Shevchenko, I Yu Nagaev, N F Myasoedov Radiokhimiya 35 99 (1993) c 115. I V Ivleva, V P Shevchenko, S I Shram, I Yu Nagaev, T Yu Lazurkina, G I Myagkova, N F Myasoedov Radiokhimiya 37 258 (1995) c 116. G F Sud'ina, O K Mirzoeva, I A Shchukin, V E Shevchenko, R S Vorisova, G I Myagkova, S D Varfolomeev Biokhimiya 56 1113 (1991) a 117. A G Reznikov Metody Opredeleniya Gormonov (Methods of Determination of Hormones) (Kiev: Naukova Dumka, 1982) 118. M Campbell, in Isotopes in the Physical and Biochemical Sciences (Labelled Compounds) (Eds E Buncel, J R Jones) (New York: Elsevier, 1991) Vol. 1, Pt. B, p. 71 119. V P Shevchenko, I Yu Nagaev, N F Myasoedov Radiokhimiya 35 106 (1993) c 120. V N Kramerov, Candidate Thesis in Chemical Sciences, Scientific Industrial Joining State Institute of Applied Chemistry, Leningrad, 1987 121. N F Myasoedov, G V Sidorov, V N Kramerov, V I Mishin J. Label. Comp. Radiopharm. (1999) (in the press) 122. U Pleiss, in Proceedings of the Third International Symposium `Synthesis and Applications of Isotopically Labelled Compounds' (Abstracts of Reports), Innsbruck, 1988 (Eds T A Baillie, J R Jones) p. 521 123. G Z Tang, C T Peng J. Label. Comp. Radiopharm. 25 585 (1988) 124. C S Markos, C R Dorn, D J Zitzewitz J. Label. Comp. Radiopharm. 25 515 (1988) 125. M D Meyer, G L Carson, D O Toft J. Label. Comp. Radiopharm. 24 171 (1987) 126. P A McCarthy J. Label. Comp. Radiopharm. 28 1149 (1990) 127. C K Acosta, J W Cessac, P N Rao, H K Kim J. Label. Comp. Radiopharm. 28 1201 (1990) 128. X Zhao,M Zhou, G Wang, Q Peng, in Proceedings of the Third International Symposium `Synthesis and Applications of Isotopically Labelled Compounds' (Abstracts of Reports), Innsbruck, 1988 (Eds T A Baillie, J R Jones) p. 553 129. V P Shevchenko, I Yu Nagaev, A V Potapova, N F Myasoedov, A V Kamernitsky, I S Levina, L E Kulikova, A N Smirnov J. Label. Comp. Radiopharm. 41 919 (1998) 130. H Parnes, E J Shelton J. Label. Comp. Radiopharm. 29 1157 (1991) 131. J P DiBattista, F X Webster J. Label. Comp. Radiopharm. 38 1083 (1996) 132. A Kasal, K Fuksova, V Pouzar Collect. Czech. Chem. Commun. 57 2166 (1992) 133. H Wagner, J Romer J. Label. Comp. Radiopharm. 25 645 (1988) 134. E J Corey, H Niwa, J Knolle J. Am. Chem. Soc. 100 1942 (1978) 135. H Watanabe,M Akiyama, T Kawanishi, N Kubodera J. Label. Comp. Radiopharm. 36 645 (1995) 879 136. L N Kaklyushkina, in Khimiya i Tekhnologiya Izotopov i Meche- nykh Soedinenii (The Chemistry and Technology of Isotopes and Labelled Compounds) (Leningrad: State Institute of Applied Chemistry, 1984) p. 3 137. A V Kamernitskii, I S Levina Khim.-Farm. Zh. 25 (10) 4 (1991) e 138. V P Shevchenko, I Yu Nagaev, N F Myasoedov, A V Kamernitsky, I S Levina, L E Kulikova, E V Pokrovskaya, T A Shchelkunova, A N Smirnov Bioorg. Chem. 27 207 (1999) 139. I V Ivanov, N V Groza, Kh Kyun, G I Myagkova Bioorg. Khim. 24 454 (1998) d 140. N V Groza, I V Ivanov, G I Myagkova Bioorg. Khim. 24 458 (1998) d 141. V P Shevchenko, I Yu Nagaev, N F Myasoedov Radiokhimiya 41 80 (1999) c 142. V P Shevchenko, I Yu Nagaev, N F Myasoedov Radiokhimiya 41 82 (1999) c 143. V P Shevchenko, I Yu Nagaev, N F Myasoedov Radiokhimiya (1999) (in the press) c 144. J R Jones, S-Y Lu J. Label. Comp. Radiopharm. 41 1024 (1998) 145. J-C Cintrat, B Rousseau J. Label. Comp. Radiopharm. 41 1029 (1998) 146. A Y L Shu,W Chen, J R Heys J. Organomet. Chem. 524 87 (1996) 147. W Chen, K T Garnes, S H Levinson, D Saunders, S G Senderoff, A Y L Shu, A J Villani, J R Heys J. Label. Comp. Radiopharm. 39 291 (1998) 148. S I Shram, T Yu Lazurkina, N F Myasoedov Dokl. Akad. Nauk 359 758 (1998) f a�Biochemistry (Engl. Transl.) b�Russ. Chem. Bull. (Engl. Transl.) c�Radiochemistry (Engl. Transl.) d�Russ. J. Bioorg. Chem. (Engl. Transl.) e�Pharm. Chem. J. (Engl. Transl.) e�Dokl. Chem. Technol., Dokl. Chem. (

ISSN:0036-021X    

出版商:RSC    

年代:1999

数据来源: RSC

摘要:

Russian Chemical Reviews 68 (10) 881 ± 887 (1999) Regulation of cytochrome P450 activity by physicochemical methods V V Shumyantseva, T V Bulko, A I Archakov Contents I. Introduction II. The use of electrochemical methods in the study of hemoproteins III. Photoinduced electron transport IV. Changes in the microenvironment of the protein and its active centre V. Chemical modification of proteins for triggering new catalytic functions VI. Effect of temperature on hemoprotein-catalysed reactions VII. Effect of high pressure on the catalytic activity of hemoproteins VIII. Conclusion Abstract. Physicochemical factors influencing the catalytic activ- ities of hemoproteins (of cytochrome P450, in particular) that find use in biosensors and bioreactors are considered.The authors' data on the preparation of semisynthetic hemoproteins based on cytochrome P450 2B4 and NADPH-dependent cytochrome P450 reductase are presented. The use of alternative electron sources (electrochemical reduction and photoreduction) for the redox cycle of hemoproteins is discussed. The effects of temperature, pressure, chemical modification of proteins and organic solvents on the efficiency of hemoprotein-catalysed enzymic reactions are analysed. The bibliography includes 86 references. I. Introduction Cytochromes P450 are unique enzymes in that they are able to hydroxylate substrates with nonactivated C7H bonds. Cyto- chromes metabolise up to 200 000 compounds and catalyse about 60 types of chemical reactions, such as hydroxylation, N-, O- or S-demethylation, dealkylation, epoxidation, etc.1 This feature of cytochromes P450 can be employed for quantification of medicinals and/or their metabolites and xenobiotics in different media as well as for stereodirected synthesis of steroids and other classes of biologically active compounds.The increase in the efficiency of catalysis by cytochromes P450 is a problem of considerable practical importance. The principles of bioelectroca- talysis have been discussed in detail.2, 3 The present review is devoted to the analysis of effects of various physicochemical factors on different stages of enzymic reactions with special emphasis on the advantages of alternative electron sources (electrochemical and photoreduction) over tradi- tional reducing agents, such as NADH or NADPH.Reduction with the use of pyridine nucleotides is not a continuous process due to rapid exhaustion of NADH or NADPH. NADPH- V V Shumyantseva, T V Bulko, A I Archakov Research Institute of Biomedical Chemistry, Russian Academy of Medical Sciences, Pogodinskaya ul. 10, 119832 Moscow, Russian Federation. Fax (7-095) 245 08 57. Tel. (7-095) 246 50 72. E-mail: victoria@ibmh.msk.su (V V Shumyantseva), Tel. (7-095) 246 69 80 (A I Archakov) Received 20 April 1998 Uspekhi Khimii 68 (10) 967 ± 975 (1999); translated by R L Birnova #1999 Russian Academy of Sciences and Turpion Ltd UDC 612.015.1 : 577.152.162.08 881 881 883 884 884 885 886 886 regenerating enzymes are normally introduced into a biochemical system to maintain the required concentration of reducing agents, which makes the system more complicated.The use of cathodic current instead of NADH or NADPH allows the reduction process to become more long-lasting and controllable by means of physicochemical factors, such as voltage and electric current. In the case of photoreduction, this process is monitored by recording the generated photocurrent and can be controlled by changing the excitation wavelength. The published data obtained in the past decade on cytochromes P450 are surveyed. II. The use of electrochemical methods in the study of hemoproteins 1. The use of electrolysis for reduction of hemoproteins Electron transport in biological systems normally occurs in the presence of electron donors (e.g., NADH and NADPH) and an electron mediator, such as flavin nucleotides.In recent years, electrical systems are finding increasing use in the study of redox conversions of enzymes (of metalloenzymes, in particular).4 Electrons can be transferred to the electrode either directly, by electrochemical transport from the protein redox centre, or in the presence of mediator molecules.5 Different types of electrodes, such as graphite (glassy carbon, pyrolytic graphite), gold, plati- num, semiconducting (indium oxide, tin oxide) and chemically modified graphite electrodes combined with redox catalysts (e.g., rhodium catalysing oxygen reduction to hydrogen peroxide or water, metal complexes which efficiently mediate electron trans- port, etc.).6±8 The electron transfer in enzyme systems containing cyto- chrome P450 occurs according to the scheme: NADPH-dependent cytochrome P450 reductase NADPH cytochrome P450.Here, NADPH is the electron donor and the microsomal flavo- protein (NADPH-dependent cytochrome P450 reductase) or a mitochondrial system consisting of a flavoprotein and a protein containing a Fe7S cluster is the electron carrier from NADPH to the hemoprotein (cytochrome P450). The use of cathodic current instead of NADPH for electro- reduction of flavins as mediators in oxygenase reactions has made it possible to obtain highly efficient electrochemical systems (Table 1).9 The hydroxylation of lauric acid in the presence of an882 Table 1. Comparison of rates of electrochemical and NADPH-dependent substrate metabolism reactions catalysed by recombinant proteins.9 Reaction Substrate Protein (reductase+cytochrome P450) rF450(mBov17a/mRatOR)L1 (reductase+P450 17a) rF450(mHum3A4/ mRatOR)L1 (reductase+P450 3A4) Progesterone Pregnenolone Testosterone Erythromycin Benzphetamine Caffeine Imipramine Lauric acid rF450(mHum1A2/mRatOR)L1 (reductase+P450 1A2) CYP102-BM3 (P450 BM3) 17a-Hydroxylation "6b-Hydroxylation N-Demethylation """o-Hydroxylation electrochemical system consisting of a gold or platinum electrode and a purified fused protein containing P450 4A1 and NADPH- dependent cytochrome P450 reductase has been studied.6, 9, 10 In addition to the fused protein, the electrochemical system contained a coordination compound of Co(III), viz., [(S)- 1,3,6,8,10,13,16,19-octaazabicyclo[6.6.6]eicosanocobalt(III)] as a mediator.The standard redox potential, U, of this compound is 7350 mV (vs. standard hydrogen electrode). Presumably, the reduction of the Co(III) ion on the cathode occurs within the coordination compound according to the scheme: Co3++e?Co2+. If the electrolysis is carried out in a system containing only the flavoprotein (reductase) or a hemoprotein (cytochrome P450) no hydroxylation of lauric acid occurs, which suggests that the Co2+ ion plays the role of an activator. This reduces flavin nucleotides of the reductase, which, in turn, reduce the heme iron of cytochrome P450.Oxygen activation occurs on the reduced heme iron within the triple complex `sub- strate7Fe2+7O2'; in this case, the oxidation of various sub- strates under physiological conditions occurs at high rates.1 Electrochemical reduction of the fusion protein generates hydro- gen peroxide, which destroys the enzyme system. The addition of catalase increases the efficiency of the system in the hydroxylation reaction. It was shown that the highest rate of lauric acid hydroxylation is observed at 7450 mV (vs. standard hydrogen electrode).6, 9, 10 semiartificial A (flavocytochrome flavohemoprotein P450 2B4) possessing reductase and oxygenase activities was obtained earlier in our laboratory.11, 12 Using an electrochemical system consisting of flavocytochrome 2B4 or 1A2 and rho- dium ± graphite microelectrodes, we performed successful aniline p-hydroxylation, aminopyrine N-demethylation and 7-ethoxy- and 7-pentoxyresorufin O-dealkylation.The reduction was car- ried out at U=7450 mV (vs. AgCl electrode).13 The catalytic constants of NADPH-dependent and electrochemical reactions catalysed by flavocytochromes 1A2 and 2B4 were comparable. Several factors have plagued direct electron transfer between electrodes and proteins. These include: (i) electroactive prosthetic groups deep within the protein structure, (ii) adsorptive denatura- tion of proteins onto electrodes, and (iii) unfavourable orienta- tions at electrodes. Remarkable recent progress provides several strategies for achieving direct electron exchange between electro- des and proteins.Immobilisation of enzymes on the electrode surface is used to increase the efficiency of the electron transfer from the electrode to the electroactive prosthetic group of the protein 14 ± 16 by means of adsorption, covalent binding and incorporation of enzymes into polymeric films coating the electrodes. For example, positively charged proteins are immobilised by adsorption on electrodes of pyrolytic graphite containing negatively charged carboxy groups, whereas immobilisation of hemoproteins on graphite electrodes involves covalent binding in the presence of a cross-linking bifunctional reagent (glutaraldehyde) and albumin.17 In the latter case, albumin is strongly adsorbed on the surface of graphite electrodes, while glutaraldehyde cross-links the redox protein with V V Shumyantseva, T V Bulko, A I Archakov Reaction rate /nmol min71 (nmol of P450)71 NADPH-dependent electrochemical 285 10.0 2.2 4.6 1.5 2.0 900 2052.6 1.2 2.0 0.75 1.2 110 albumin through the amino groups thereby binding the former to the electrode.Yet another approach consists of the use of promoter molecules which favour the electron transport through proper orientation of the proteins on the electrode, which thus blocks their adsorptive denaturation. The role of promoters can be played by alkanethiols containing terminal functional groups (e.g., mercaptoundecanoic acid).18 The chemosorption of mer- captoundecanoic acid on the surface of gold electrodes is accom- panied by the formation of a stable monolayer, which facilitates subsequent binding (both covalent and ionic) of proteins to the electrode surface.Au(electrode)7S(CH2)10COO7 (NH3)+-hemoprotein, Au(electrode)7S(CH2)10COO7 (NH3)+(CH2)n(NH3)+ 7OOC-hemo- protein. Double-stranded DNA, which is strongly adsorbed on elec- trodes of pyrolytic graphite, can also be used as a promoter molecule.19 Such DNA-containing electrodes can readily extract hemoproteins from the solution. The rate of the electron transfer from the electrode to the redox protein, can be increased if the electrodes containing covalently bound or adsorbed electron transport mediators, such as riboflavin or methylviologen, are chemically modified;20, 21 the stability of hemoproteins in electrochemical systems is increased if they are incorporated into phospholipid films which coat the electrodes.22 ± 24 Surfactants containing charged or polar head groups and a hydrophobic tail (didodecyldimethylammonium bromide, polystyrenesulfonate, dihexadecyl phosphate, dimyris- toylphosphatidylcholine) are used as films mimicking natural biomembranes. It is of note that electrochemical processes involving redox proteins are reversible only in liquid-crystalline films.22 If cyto- chrome P450cam is incorporated into liquid-crystalline films formed by dimyristoyl-L-a-phosphatidylcholine or didodecyl- dimethylammonium bromide, direct reversible electron transfer occurs from the electrode to the heme iron of the enzyme as is judged by direct current cyclic voltammograms.22 ± 24 Thus myo- globin reduction in liquid-crystalline didodecyldimethyldi- ammonium bromide films coating the pyrolytic graphite electrodes is reversible in contrast with its reduction on electrodes without biomembrane-like surfactant films.24 On addition of a substrate (e.g., trichloroacetic acid), this is reduced, as follows from the increase in the cathodic current.The incorporation of cytochrome P450cam into DNA films adsorbed onto gold electro- des has made it possible to perform direct reduction of the heme iron and catalytic epoxidation of styrene.25 Electrochemical reduction of metmyoglobin (X7MbFeIII), which is a perfect functional model of cytochrome P450 (the catalytic activity of myoglobin is close to that of cytochrome P450), was studied by Onuoha et al.24 The electroreduction was carried out on electrodes of pyrolytic graphite at the potential, U=7400 mV(vs.SCE) in aqueous buffers or microemulsions ofRegulation of cytochrome P450 activity by physicochemical methods hydrocarbons, water and cationic detergents. Under these con- ditions metmyoglobin is reducted on the electrode [Scheme 1, reaction (1)]. The reduced metmyoglobin rapidly reacts with oxygen to yield an X7MbFeII7O2 complex, which is reduced with the liberation of hydrogen peroxide on the electrode. The hydrogen peroxide evolved effects stepwise oxidation of metmyo- globin (X7MbFeIII) to ferrylmyoglobin (HX7MbFeIV=O2), an active oxidised form of myoglobin [Scheme 1, reactions (4) and (5)].Scheme 1 (1) X7MbFeII (electrode), (2) X7MbFeII7O2 , X7MbFeIII+e X7MbFeII+O2 X7MbFeII7O2+2e+2H+ X7MbFeIII+H2O2 (electrode), .X7MbFeIV=O+H2O, X7MbFeIII+H2O2 (3) (4) (5) 2HX7MbFeIV=O+O2 , 2 .X7MbFeIV=O+H2O2 whereXis an amino acid residue, e.g., tyrosine-130. Styrene added to the reaction system is oxidised by hydrogen peroxide to give styrene oxide and benzaldehyde. The efficiency of electrolysis used for the reduction of hemoproteins is comparable with the traditional reduction with pyridine nucleotides. The results obtained in the study of electro- chemical and NADPH-dependent reactions of several isoforms of cytochrome P450 are presented in Table 1.At present, attempts are being undertaken aimed at designing cytochrome P450-containing bioreactors and biosensors based on electrochemical systems.9, 22, 26 2. The use of inorganic semiconductors in redox systems The effectiveness of inorganic semiconductors (TiO2, ZnO, CdS) in combination with enzymes (hydrogenases, dehydrogenases) in photoproduction of hydrogen, photoreduction of NADP+ and photosynthesis of organic acids and amino acids has been demonstrated in a number of studies.27 ± 31 The system `inorganic semiconductor ± enzyme' can transform light energy by semicon- ductor-photosensitised oxidation of organic electron donors and by enzyme-catalysed reduction of substrates (H+, NADP+, organic acids and amino acids) with electrons photogenerated in the semiconductor.Coupling of an inorganic semiconductor with an enzyme is usually attained with the help of electron carriers (mediators), e.g., methylviologen, a rhodium bipyridyl complex and other compounds subject to reversible oxidation ± reduction. Direct mediator-free transfer of electrons photogenerated in the conduction band of the semiconductor to the enzyme reaction centre has been described by Nikandrov et al.28 A prerequisite for the conjugate effect of a semiconductor and an enzyme in the absence of an electron carrier is the sorption of the enzyme on the semiconductor surface. 3. The construction of an enzymic `galvanic couple' for redox reactions The operating principle of a galvanic element is based on the difference in the electrode potentials of the oxidant and the reducing agent.Active metals (e.g., zinc) are most commonly used as reducing agents, while hemoproteins are used as oxidants. Normally, hemoproteins have a more positive standard electrode potential than zinc [E0(Zn0/Zn2+)=7763 mV]; therefore zinc serves as a reducing agent in a zinc ± hemoprotein couple. We used metallic zinc as an electron donor and a hemin ± human serum albumin complex as an artificial hemoprotein and succeeded in effecting N-demethylation of aromatic amines.32 The rates of these reactions were comparable with those of NADH-dependent reactions carried out in the presence of flavin nucleotides or riboflavin as electron transfer mediators.Metallic cadmium [E0(Cd0/Cd2+)=7403 mV, E0(NAD+/ NADH)=7340 mV] was used as a reducing agent in a system 883 containing NAD+-dehydrogenase.33 In this case, too, the `gal- vanic couple' principle is realised. III. Photoinduced electron transport Light energy is utilised not only in traditional photosynthetic reactions or in photoinduction of semiconductors, it also triggers electron transport in redox systems. Direct photoreduction of hemoproteins in the absence of electron donors has been described by Gu et al.34 Metmyoglobin, cytochromes c and b5 and heme complexes with octapeptides and cytochrome P450 were photo- reduced upon laser irradiation at wavelengths of 430 ± 254 nm. UV irradiation resulted in photooxidation.This study has shown that the heme ligands play a crucial role in photoreduction. The photoreduction products were characterised by UV and visible absorption spectra and Raman spectra and compared with the products formed upon chemical reduction. Heme photoreduction in hemoproteins can be utilised for `triggering' redox reactions, such as oxygen activation by cyto- chrome P450 or binding of ligands (e.g., O2) with myoglobin or hemoglobin. Studies of model complexes (hemin ± octapeptides) corroborated the viewpoint that aromatic amino acids are not the primary electron donors, since the model peptides were devoid of aromatic amino acid residues. In the presence of organic solvents (glycerol and other alcohols), the rate of photoreduction increased.34 Irradiation with a nonfocused laser (l=457.9 nm, power 40 mV) of a `riboflavin ± hemin ± albumin' complex resulted in riboflavin photoreduction as could be evidenced from changes in the absorption spectrum of the complex.Amine demethylation and aniline hydroxylation products were formed in the presence of N-substituted amines or aniline; the rate constants (kcat) for dimethylaniline N-demethylation and aniline hydroxylation are equal to 0.5 min71 and 0.13 min71, respec- tively, i.e., they are comparable with those of NADPH-dependent reactions.32 The photooxidation of the reduced flavocytochrome b2 { from Saccharomyces cerevisiae (lactate dehydrogenase) upon irradia- tion with a laser (l=400 nm) has been described by Hazzard et al.35 At this wavelength, the oxidised form of 5-deazariboflavin passes into an excited triplet state; it is this state that is involved in the oxidation of the reduced heme.The rate of intermolecular electron transport was 2200 s71. A series of investigations 36 ± 40 have been devoted to the study of electron transport in hemoproteins (in cytochromes c and b5, in particular) modified with ruthenium or cobalt complexes. A characteristic feature of ruthenium complexes is their ability to undergo reversible photoreduction and to transfer electrons to other acceptors. Thus a ruthenium complex covalently bound to cysteine-65 of cytochrome b5 effected rapid reduction of cyto- chrome c according to the scheme.36 7e Ru2+ ± cyt b5(Fe3+) Ru3+ ± cyt b5(Fe2+), 7e Ru3+ ± cyt b5(Fe2+)+cyt c (Fe3+) Ru3+ ± cyt b5(Fe3+)+cyt c (Fe2+) .The electron transfer in these systems was studied in direct photoreduction of ruthenium complexes and using a flash- photolysis technique. The photoreduction of the flavoprotein cytochrome P450 reductase in the presence of electron acceptors (ferricyanide, Tetrazolium Nitro Blue) could be carried out in the absence of NADPH.40 { Reduced flavocytochrome was obtained by titration of the oxidised enzyme with a 1.5-fold excess of L-lactate.884 Thus, photoinduced electron transport can serve as an alternative source of electrons in redox reactions provided the proteins contain photosensitive prosthetic groups or their ana- logues. IV.Changes in the microenvironment of the protein and its active centre 1. Modelling of the membrane environment The environment, especially the cell membrane, strongly influ- ences the enzyme activity in living organisms. The construction of an artificial membrane-like environment may significantly enhance the enzyme activity. Among various synthetic mem- branes studied, synthetic membrane-like phospholipid bilayers were shown to possess an activating effect.41 Cytochrome c incorporated into such membranes increases the rate of H2O2- induced N,N-dimethylaniline N-demethylation 10.6-fold. A 20- fold decrease in the Michaelis constant for H2O2 is observed, which suggests increased affinity of the enzyme towards hydrogen peroxide.Changes in the spectral characteristics of cytochrome c incorporated into the phospholipid bilayer suggest significant structural and functional changes. The oxidised form of the protein is characterised by a short-wave shift of the optical density maximum (from 408 to 406 nm) upon incorporation, the absorption maximum of the reduced form is shifted from 416 to 426 nm and one maximum at 550 nm is observed instead of two maxima at 520 and 550 nm. The incorporation of myoglobin into membrane-like films results in a decrease in the activation energy of electrochemical reduction of dibromoethylene and trichloroacetic acid, the rate of electron transport increases 1000-fold.16 Acetylcholinesterase, cholinesterase and horseradish peroxidase incorporated into polymeric films prepared from polyvinylpyridine and polyethy- lene glycol can be used as amperometric sensors for detecting acetylcholine, choline and hydrogen peroxide.8 2.Enzymatic catalysis in reversed micelles and organic solvents Many properties of biological membranes are efficiently mim- icked by reversed micelles of surfactants in organic solvents, which can be regarded as biomembrane models. In 1977, Martinek et al.42 showed that enzymes solubilised in organic solvents with the aid of reversed micelles of synthetic surfactants retain their catalytic activities. Some hemoproteins, e.g., cytochrome c and methemoglobin, induce the formation of reversed micelles by incorporating into bilayer membranes with simultaneous preservation of their enzymic activities.43 ± 45 Horse- radish peroxidase solubilised in a sodium diisooctyl sulfosucci- nate ± water ± octane system catalyses pyrogallol oxidation by hydrogen peroxide much more efficiently than the water-soluble enzyme.46 Efficient catalysis of redox reactions by hemoproteins incor- porated into reversed micelles of surfactants was used for increas- ing the operational stability of catalase, superoxide dismutase and glutathione peroxidase.47 Systems mimicking hemoproteins (albumin ± hemin complexes and ferritin) were also incorporated into reversed micelles.48 Cytochrome P450 1B1 and cytochrome P450 reductase as a fusion protein was incorporated into oil-in-water macroemulsion, termed polyaphron, (the internal organic phase content was more than 0.74).The efficiency of such immobilisation exceeded 85%; the enzyme preparation remained active towards substrates (erythromycin, chlorotolurone) for 24 h at 15 8C.49 In order to decrease the water content in colloidal media, the substitution of a water-organic mixture for the water surrounding the enzyme in reversed micelles of surfactants was suggested.50 The hemoglobin-catalysed oxidation of dibenzothiophene in various water-organic mixtures was studied by Klyachko and Klibanov.51 The preparation of suspensions of lyophilised enzymes in organic solvents is an alternative procedure for V V Shumyantseva, T V Bulko, A I Archakov excluding water. The enzymes suspended in organic solvents acquire new properties, which affects virtually all the aspects of their functioning, viz., catalytic activity, substrate specificity, stereoselectivity, thermal stability, etc.52, 53 In addition to ordinary enzymic reactions, the reactions nonspecific for enzyme behaviour in aqueous media may occur either in nonaqueous organic solvents or in organic solvents with the H2O content of 0.01 vol.%.The reasons for the alteration of the substrate specificities are different for individual enzymes. Nevertheless, enzymatic catalysis in organic solvents obeys the Michaelis ± Menten kinetics, which suggests the formation of a classical enzyme ± substrate complex in nonaqueous solutions. The phenomenon of `enzyme memory' has been discovered in experiments with enzyme suspensions in organic solvents.An enzyme lyophilised from an aqueous solution containing a specific ligand and then suspended in an organic solvent binds the ligand several tens of times more efficiently. This property disappears after repeated dissolution in water. A possible explanation of this phenomenon is that the enzyme retains an `imprint' of the ligand even after its removal upon lyophilisation. In nonaqueous media where the conformation of the protein molecule becomes more rigid, the enzyme manifests higher affinity for the ligand.54, 55 The ability of enzymes to retain the `imprints' of ligands has a certain practical significance for the synthesis of chiral pharmacologically useful intermediates.The effects of organic solvents on the properties of cyto- chrome P450 have been studied to a lesser degree. It was found that organic solvents favour the transition of cytochrome P450 into the inactive form, P420.56 This transition is due to the changes in the nearest coordination sphere (sulfur atoms) around the iron atom. A similar transition is observed in the case of chloroperox- idase.56 However, cytochrome c suspended in tetrahydrofuran containing 1% D2O fully preserves its tertiary structure.57 These conclusions were made on the basis of analysis of the tertiary structure of cytochrome c by two-dimensional NMR spectro- scopy. V. Chemical modification of proteins for triggering new catalytic functions This section considers the methods used for enhancement of the enzymatic activity by introducing new catalytically active groups into the enzyme molecule.Thus cytochrome P450 manifests monooxygenase activity only within a complex with the flavoprotein reductase. Here flavin nucleotides represent electron carriers which reduce the heme iron of cytochrome P450. The introduction of flavin nucleotides or their analogues into cytochrome P450 converts the latter into a self-competent enzyme which does not require the involvement of partner proteins in the catalytic cycle. Semisynthetic flavohemoglobin was prepared by attachment of activated flavin derivatives 58 to the b-subunit of hemoglobin used in a 2 : 1 ratio.59, 60 Hemoglobin catalyses monooxygenase reactions occurring in a system containing a hemoprotein, oxygen, NADPH and a flavoprotein (NADPH-dependent cytochrome P450 reductase).Covalent binding of flavin to hemoglobin makes it possible to conduct similar reactions in the absence of reductase. The rates of the catalytic reaction of aniline hydrox- ylation in the presence of flavohemoglobin and the microsomal monooxygenase system are comparable. The rate of aniline hydroxylation in the presence of semisynthetic flavohemoglobin increased in comparison with hydroxylation rates in the presence of hemoglobin and the hemoglobin ± reductase system (7.7- and 1.3-fold, respectively). The catalytic activity of the covalent complexes `cytochrome P450 ± flavin adenine dinucleotide' (FAD) and `cytochrome P450 ± flavin mononucleotide' (FMN) has been studied by Uvarov et al.61 The complex in which cytochrome P450 is covalently bound to FMN (1 : 3) turned out to be more active.The catalytic rate constants for dimethylaniline and aminopyrineRegulation of cytochrome P450 activity by physicochemical methods N-demethylation and aniline hydroxylation were about 2 times lower than those of the analogous reactions catalysed by micro- somes, which represent natural complexes of cytochrome P450 and reductase. The rates of reactions catalysed by covalent and noncovalent P450 ±FAD complexes are 10 times lower than those of microsome-catalysed reactions. Amonomolecular monooxygenase system formed by covalent binding of the electron-accepting antibiotic bleomycin to a flavoprotein (NADPH-dependent cytochrome P450 reductase) has been described by the same authors.61 The rates of reactions involving the reductase ± bleomycin complex are 10 times lower than those of microsome-catalysed reactions. Reductase does not possess its own catalytic activity in monooxygenase reactions; however, after incorporation of 1 to 10 hemin residues into the reductase molecule the semi-artificial hemoreductase manifested activity in NADPH-dependent N-demethylation of amidopyrine (kcat=0.7 min71).The kcat values for the analogous reactions catalysed by the microsomal monooxygenase system and the semi-artificial flavocytochrome 2B4 are 2.4 and 0.98 min71, respectively.11, 12 A semi-artificial hemoreductase 62 which represents a covalent complex of hemin and NADPH-dependent cytochrome P450 reductase manifested reductase activity with respect to electron acceptors, such as potassium ferricyanide, cytochrome c and hemin.The covalent binding of flavodoxin to ferredoxin-NADP+ reductase 63 (1 : 1) gave an artificial enzyme which catalyses the reduction of cytochrome c with high efficiency. The transfer of electrons in this system occurs according to the same scheme as in NADPH-dependent cytochrome P450 reductase. FAD (from ferredoxin-NADP+ reductase) NADPH cytochrome c. FMN (flavodoxin) Ferredoxin-NADP+ reductase modified with viologen in the presence of carbodiimide manifests oxidase activity, which is absent in the native enzyme.64 In the absence of ferredoxin, the reductase does not reduce cytochrome c, whereas the modified enzyme displays high activity.In the presence of electron accept- ors such as 2,6-dichloroindophenol and ferricyanide, the activity of the modified enzyme decreases twofold, whereas the rate of the cytochrome c reductase reactions increases 30-fold. The scheme of electron transport in a system with the nature enzyme appears as: cytochrome c. ferredoxin reductase NADH When the reactions are carried out in the presence of super- oxide dismutase, the viologen ± reductase systems utilise the oxy- gen molecule reduced to the superoxide radical anion as the find electron acceptor. It is the presence of active oxygen species that accounts for the appearance of oxidase activity in the modified enzyme.Titration of the reduced viologen ± reductase system revealed that the electron transfer to viologen occurs through the flavin group of FAD present in the reductase. The standard electrode potential of viologen covalently bound to ferredoxin- NADP+ reductase is 7365 mV, while that of soluble viologen is 7420 mV. The electrode potentials of other reaction participants are 7320 (NADPH), 7344 (reductase) and 7420 mV (ferre- doxin). The electrochemical preparation of NADPH using the native enzyme requires an electron donor capable of reduction of ferredoxin, since in this case the scheme of electron transport appears as: NADP+. reductase ferredoxin The electric current generated by the electrochemical system `viologen ± reductase' in the presence of NADP+ points to the electron transfer between the electrodes and the system.The latter can therefore be used for regeneration of NADPH. Ferrocene and its derivatives are known to be efficient electron carriers. Modification of glucose oxidase (GOx) with ferrocene- carboxylic acids (FCA) has been described by Badia et al.65 Covalent derivatisation of the GOx molecule containing two 885 FAD molecules with additional electron carrier groups imparts the ability to be involved in the electrochemical reaction. The optimum number ofFCA groups is 13 residues per GOx molecule. In native glucose oxidase, reduced FADH2 is oxidised with atmospheric oxygen, whereas the role of oxygen in ferroceneglu- cose oxidase is played by FCA responsible for `internal' electron transport.Ferroceneglucose oxidase undergoes electrochemical conversions on the electrodes according to the scheme: FADH27GOx7(FCA+)n+ne FADH27GOx7(FCA)n FAD7GOx7(FCA)2(FCA+)n72+2H++ne. The arrangement of bound FCA residues relative to the FAD molecule is more important for electron transport than their number. The electrode potential of the FADH2/FADH. couple is7219 mV, that of the FCA+/FCA couple is +773 mV. Hence, the motive force of the electron transport in the given system is about 1 V. Modification of cytochrome P450cam and b-lactamase with a new electron-active reagent, N-(2-ferrocenylethyl)maleimide, was described by Gleria et al.66 ± 68 who showed that modified enzymes are electrochemically active.These authors suggest using N-(2- ferrocenylethyl)maleimide as a promising reagent for the conver- sion of electropassive enzymes into electroactive ones as well as for the design of electrochemical sensors and for studies of mecha- nisms of intraprotein electron transport. The covalent binding of glutathione reductase to the photo- sensitive dye eosin (1 : 2) was used to obtain a photoactivated system which can be used for glutathione reduction.69 Eosin- modified glutathione reductase retains 50% of activity of the native enzyme in a NADPH-dependent reaction. However, the photosystem thus obtained was 1000 times less active in gluta- thione reduction than the glutathione reductase with NADPH as the electron source.VI. Effect of temperature on hemoprotein- catalysed reactions Studies of reactions catalysed by cytochrome P450 2B4 have shown that an increase in temperature causes an increase in the rate of substrate hydroxylation. For example, with a rise in temperature from 20 to 37 8C, the catalytic constant, kcat, for peroxide-dependent O-dealkylation of p-nitroanisole increases twofold.70 The catalytic cycle of cytochrome P450 is rather complex and comprises at least seven steps.1, 71 The effect of temperature on the binding of various substrates (e.g., benzphetamine) to cytochrome P450 and on the spin equilibrium between low- and high-spin forms of the hemoprotein was studied in most detail. It was shown 72 that a change in temperature from 12 to 27 8C has no effect on the absorption spectra of cytochrome P450 2B4.The spectra are not changed after addition of benzphetamine (0.8 mol litre71) either. The dissociation constants (Kd) and concentration [ES]max of enzyme ± substrate complexes show the following temperature dependence: [ES]max /mmol litre71 Kd /mmol litre71 Temperature /8C 13.8 23.8 0.0850.002 0.0870.004 1895 865 As can be seen, the concentration of enzyme ± substrate complexes is practically constant at different temperatures. How- ever, the use of the `temperature jump' technique, which consists of instantaneous increase in the temperature in the reaction cell upon high-power electric discharge revealed enhanced binding of the substrate to the enzyme.72 The substrate-induced spin equilibrium obeys a bimolecular mechanism: P450(l.s.)+S P450(h.s.)7S ,886 where l.s.and h.s. are the concentrations of the low-spin and high- spin forms of the enzyme, respectively. The catalytic constant of this reaction can be written as: kcat = 1tf=k1([E]eq+[S]eq)+k71, where tf is the fast relaxation time, i.e., the time during which the system passes into a new state and [E]eq and [S]eq are the equilibrium concentrations of the enzyme and the substrate, respectively. In the absence of the substrate, the relaxation does not occur, which is consistent with the experimental data con- cerning the effect of temperature on the absorption spectra of cytochrome P450 not bound to the substrate.A study of temperature-dependent electrochemical reduction of the fusion protein { (within the temperature range of 4 ± 36 8C) revealed that the rate of reduction is a maximum at 22 8C.6 The reason should be sought in the simultaneous occurrence of several parallel reactions in electrochemical systems that are differently affected by temperature. VII. Effect of high pressure on the catalytic activity of hemoproteins The effect of high pressure on the kinetics and dynamics of conformational conversions of proteins has been studied by Morild.73 High pressures are used for shifting chemical and biochemical equilibria as well as for changing the rates of individual steps of enzymatic reactions. The changes in the reaction and activation (DV* and DV**) volumes calculated from the experimental data on the effect of high pressure can be utilised in the study of molecular mechanisms of different enzymes and their transition states during the formation of new bonds.If DV* and DV** values are negative, an increase in pressure will shift the equilibrium towards the formation of enzymatic reaction products in accordance with the Le Chaà telier principle.74, 75 The effect of high pressure on bacterial camphor-metabolising cytochrome P450cam was extensively studied by several authors.76 ± 79 It was found that the pressure affects the equili- brium between low-spin and high-spin states. An increase in pressure to 2.2 kbar induces the transition of cytochrome P450cam into the inactive form, P420.Chloroperoxidase is also converted into C420 at high pressure.75 The effect of high pressure on the redox potential of cytochrome c was studied by Tschirret- Guth et al.80 By measuring changes in the standard potential (E 0) as the function of pressure (P), one can calculate the reaction volume, DV0: à DV0 nF , qP t qE0 where F is the Faraday constant, DV0=~V0p7~V0r (~V0p and ~V0r are the sums of standard molar volumes of the products and reactants, respectively). As the pressure is increased, the reaction accompanied by a decrease in the molar volume will be preferred. In the case of a redox process, this will induce a positive shift of the redox potential. According to Tschirret-Guth et al.,80 the increase in pressure to 5 kbar results in the changes of the potential from 75 mV to +76 mV; the molar volume also changes upon transition from the oxidised to the reduced form of cytochrome c: V[cyt c(Fe2+)]7V[cyt c(Fe3+)]=724 ml mol71.These findings are consistent with the results of structural and physicochemical studies, viz., ferrocytochrome c has a more compact structure than ferricytochrome c. By changing the pressure on the enzymic system, one can shift the chemical { The fusion protein is a gene-engineered protein containing both reductase and cytochrome P450 4A1 in the same polypeptide chain. V V Shumyantseva, T V Bulko, A I Archakov equilibrium in redox hemoproteins towards the formation of the reduced form. VIII. Conclusion Most enzymes, including hemoproteins, are rather sensitive to physical and chemical influences which can induce complete or partial loss of their activities.This necessitates the design of artificial enzymes based on proteins with higher resistance to such influences. In some studies, artificial hemoproteins were prepared as a result of the formation of noncovalent specific complexes of hemin 32, 81 or its analogues (cobalt octaethylbis- porphyrinate, cobalt octapyridinemethylphthalocyanine and lute- cium diphthalocyanine) 82 with human or bovine serum albumin and synthetic peptides.83, 84 These complexes manifest oxygenase- like catalytic activity in aminopyrine N-demethylation and aniline hydroxylation. Yet another trend, viz., the design and de novo synthesis of artificial proteins with predetermined structure and functional properties by gene engineering methods is also in progress.85, 86 This review summarises the data on targeted action of various physicochemical factors on individual steps of enzymic reactions catalysed by hemoproteins.A rational combination of physico- chemical approaches to the study of natural, semi-artificial and artificial enzymic systems can be useful in the design of new effective biosensors, test systems and bioreactors. References 1. A I Archakov, G I Bachmanova Cytochrome P450 and Active Oxygen (London, New York, Philadelphia: Taylor and Francies, 1990) p. 105 2. I V Berezin, S D Varfolomeev, in Frontiers in Bioorganic Chemistry and Molecular Biology (Oxford: IUPAC, Pergamon Press, 1980) p.467 3. S D Varfolomeev, I N Kurochkin, A I Yaropolov Biosens. Bioelectron 11 863 (1990) 4. H A Hill, N I Hunt Methods Enzymol. 227 501 (1993) 5. P Myur, F W Scheller, M Kyun Prikl. Biokhim. Mikrobiol. 18 481 (1982) 6. K M Faulkner, M S Shet, C W Fisher, R W Estabrook Proc. Natl. Acad. Sci. USA 92 7705 (1995) 7. S A Wring Analyst 117 1281 (1992) 8. E F Bowden, F M Hawkridge J. Electroanal. Chem. 161 355 (1984) 9. R W Estabrook, K M Faulkner, M S Shet, C W Fisher Methods Enzymol. (Cytochrome P450, Part B) 272 44 (1996) 10. R W Estabrook, M S Shet, C W Fisher, C M Jenkins, M R Waterman Arch. Biochem. Biophys. 333 308 (1996) 11. V V Shumyantseva, V Yu Uvarov, O E Byakova, A I Archakov Biochem.Mol. Biol. Int. 38 829 (1996) 12. V V Shumyantseva, V Yu Uvarov, O E Byakova, A I Archakov Arch. Biochem. Biophys. 354 133 (1998) 13. V V Shumyantseva, A I Archakov, R D Shmidt, in II S'ezd Biokhimicheskogo Obshchestva RAN (Tez. Dokl.), Moskva, 1997 [The Second Congress of Biochemical Society of Russian Academy of Sciences (Abstracts of Reports), Moscow, 1997] p. 544 14. I Cruz Vieira, O Fatibello-Filho Anal. Lett. 30 895 (1997) 15. W Jin, F Bier Biosens. Bioelectron. 10 823 (1995) 16. J F Rusling, A-E F Nassar J. Am. Chem. Soc. 115 11 891 (1993) 17. E I Iwuoha, S Joseph, Z Lung,M R Smyth, U Fuhr, P R O Demontellano J. Pharm. Biomed. Anal. 17 1101 (1998) 18. W Jin, U Wollenberger, R Bernhardt, W F M Stocklein, F W Scheller Bioelectrochem.Bioenerg. 47 75 (1998) 19. A-E F Nassar, J F Rusling J. Am. Chem. Soc. 118 3043 (1996) 20. W Sun, J Kong, J Deng Anal. Lett. 29 2425 (1996) 21. M Somasundrum, M Tanticharoen, J Kirtikara J. Electroanal. Chem. 407 247 (1996) 22. J F Rusling Acc. Chem. Res. 31 363 (1998) 23. Z Zhang, A-E F Nassar, Z Lu, J B Schenkman, J F Rusling J. Chem. Soc., Faraday Trans. 93 1769 (1997) 24. A C Onuoha, X Zu, J F Rusling J. Am. Chem. Soc. 119 3979 (1997)Regulation of cytochrome P450 activity by physicochemical methods 25. R W Estabrook, M S Shet, K M Faulker, C W Fisher Endocryn. Res. 22 665 (1996) 26. Y M Lvov, Z Lu, J B Schenkman, X Zu, J F Rusling J. Am. Chem. Soc. 120 4073 (1998) 27. M A Shlyk, V V Nikandrov, N A Zorin, A A Krasnovskii Biokhimiya 54 1598 (1989) a 28.V V Nikandrov, A I Aristarkhov, M A Shlyk, A A Krasnovskii Dokl. Akad. Nauk SSSR 319 242 (1991) b 29. P Cuendet, K K Rao,M Gratrel, D O Hall Biochemie 68 217 (1986) 30. I A Shumilin, V V Nikandrov, V O Popov, A A Krasnovsky FEBS Lett. 306 125 (1992) 31. V V Nikandrov Biol. Membrany 15 598 (1998) c 32. V V Shumyantseva, T V Bulko, A G Zimin, V Yu Uvarov, A I Archakov Biochem. Mol. Biol. Int. 39 503 (1996) 33. I A Shumilin, V V Nikandrov, A A Krasnovsky, V O Popov FEBS Lett. 328 189 (1993) 34. Y Gu, P Li, J T Sage, P M Champion J. Am. Chem. Soc. 115 4993 (1993) 35. J T Hazzard, C A McDonough, G Tollin Biochemistry 33 13 445 (1994) 36. A Willie, P S Stayton, S G Sligar, B Durham, F Millett Biochemistry 31 7237 (1992) 37.E H Yonemoto, R L Riley, Y Il Kim, S J Atherton, R H Schmehl, T E Mallouk J. Am. Chem. Soc. 114 8081 (1992) 38. D S Wuttke, H B Gray J. Am. Chem. Soc. 115 8455 (1993) 39. D W Conrad, H Zhang, D E Stewart, R A Scott J. Am. Chem. Soc. 114 9909 (1992) 40. A V Hodgson, H W Strobel Anal. Biochem. 243 154 (1996) 41. I Hamachi, A Fujita, T Kunitake J. Am. Chem. Soc. 116 8811 (1994) 42. K Martinek, A V Levashov, N L Klyachko, I V Berezin Dokl. Akad. Nauk SSSR 236 920 (1977) b 43. N L Klyachko, A V Levashov, K Martinek Mol. Biol. 18 1019 (1984) d 44. A V Kabanov, A V Levashov, N L Klyachko, K Martinek J. Theor. Biol. 133 327 (1988) 45. A V Levashov Biotekhnologiya. (Itogi Nauki i Tekhniki) [Biotechnology (Advances in Sciences and Engineering Series)] (Moscow: Izd.VINITI, 1987) p. 112 46. D I Metelitsa Usp. Sovrem. Biol. 28 145 (1988) 47. A N Eremin, A V Litvinchuk, D I Metelitsa Biokhimiya 61 664 (1996) a 48. G S Arapova, A N Eremin, D I Metelitsa Biokhimiya 63 1400 (1998) a 49. S B Lamb, D C Lamb, S L Kelly, D S Stuckley FEBS Lett. 431 343 (1998) 50. N L Klyachko, N G Bogdanova, A V Levashov, A V Kabanov, A V Pshezhetskii, Yu L Khmel'nitskii, K Martinek, I V Berezin Dokl. Akad. Nauk SSSR 297 483 (1987) b 51. N L Klyachko, A M Klibanov Appl. Biochem. Biotechnol. 37 53 (1992) 52. A M Klibanov Trends Biochem. Sci. 14 141 (1989) 53. K Xu, K Grebenow, A M Klibanov Biotechnol. Bioeng. 56 485 (1997) 54. A M Klibanov Nature (London) 374 596 (1995) 55.U R Desai, J J Osterhout, A M Klibanov J. Am. Chem. Soc. 116 9420 (1994) 56. S A Martins, S R Blanke, L P Hager, S G Sligar Biochemistry 35 14 530 (1996) 57. J Wu, D G Gorenstein J. Am. Chem. Soc. 115 6843 (1993) 58. H L Levine, E T Kaiser J. Am. Chem. Soc. 100 7670 (1978) 59. T Kokubo, S Sassa, E T Kaiser J. Am. Chem. Soc. 109 606 (1987) 60. J Kuriyan, R J Simon, T Kokubo, E T Kaiser, A Pahler J. Am. Chem. Soc. 110 6261 (1988) 61. V Yu Uvarov, V V Shumyantseva, E A Bukovskaya, L N Kolyada, A I Archakov Biochem. Biophys. Res. Commun. 200 62. V V Shumyantseva, T L Moskvitina, T V Bulko, A I Archakov 63. M C Pirola, F Monti, A Aliverti, G Zanetti Arch. Biochem. 64. M T Bes, A L de Lacey,M L Peteato, V M Fernandez, 65. A Badia, R Carlini, A Fernandez, F Battaglini, S R Mikkelsen, 722 (1994) FASEB J.Abstr. 11 288 (1997) Biophys. 311 480 (1994) C Gomez-Moreno Eur. J. Biochem. 233 593 (1995) A M English J. Am. Chem. Soc. 115 7053 (1993) 887 66. K Di Gleria, H A O Hill, L L Wong FEBS Lett. 390 142 (1996) 67. K Di Gleria, C M Halliwell, C Jacob, H A O Hill FEBS Lett. 400 155 (1997) 68. K Di Gleria, D P Nickerson, H A O Hill, L L Wong, V FuÈ loÈ p J. Am. Chem. Soc. 120 46 (1998) 69. I Willner, E Zahavy Angew. Chem., Int. Ed. Engl. 33 581 (1994) 70. V Yu Uvarov, V E Tretiakov, A V Leschenko, C S Dzuzenova, L Z Tretiakova, I G Rukavishnikov, A I Archakov Eur. J. Biochem. 181 391 (1989) 71. X Fang, Y Kobayashi, J R Halpert FEBS Lett. 416 77 (1997) 72. Sh Narasimhulu Biochemistry 32 10344 (1993) 73. E Morild Adv. Protein Chem. 34 93 (1981) 74. K Herenamans Ann. Rev. Biophys. Bioeng. 11 1 (1982) 75. G Hui Bon Hoa, C Di Primo, I Dondaine, S G Sligar, I Gunsalus, P Douzon Biochemistry 28 651 (1989) 76. G Hui Bon Hoa, C Di Primo, E Deprez, in Proceedings of the 7th International Conference on Biochemistry and Biophysics of P450. Structure and Function, Biotechnological and Ecological Aspects, Moscow, 1992 p. 21 77. M T Cruanes, K K Rodgers, S G Sligar J. Am. Chem. Soc. 114 9660 (1992) 78. D R Davydov, E Deprez, G Hui Bon Hoa, T V Khyushko, G P Kuznetsova, Ya M Koen, A I Archakov Arch. Biochem. Biophys. 320 330 (1995) 79. D R Davydov, G Hui Bon Hoa, J Peterson Biochemistry 38 751 (1999) 80. R A Tschirret-Guth, G Hui Bon Hoa, P R Ortiz de Montellano J. Am. Chem. Soc. 120 3590 (1998) 81. O B Rus', A V Puchkaev,D I Metelitsa Biokhimiya 61 1813 (1996) a 82. V V Shumyantseva, T L Moskvitina, L G Tomilova, G N Novodarova, V Yu Uvarov, G V Ponomarev, A I Archakov, in XIth International Symposium on Microsomes and Drug Oxidations (Abstracts of Reports), Los Angeles, 1996 p. 198 83. R E Sharp, C C Moser, F Rabanal, P L Dutton Proc. Natl. Acad. Sci. USA 95 10465 (1998) 84. M W Mutz, M A Case, J F Wishart, M R Ghadiri, G L McLendon J. Am. Chem. Soc. 121 858 (1999) 85. D A Dolgikh,M P Kirpichnikov, O B Ptitsyn, V V Chemeris Mol. Biol. 30 261 (1996) d 86. D A Dolgikh, V N Uversky, A E Gabrielian, V V Chemeris, A N Fedorov, E V Navolotskaya, V A Zav'yalov, M P Kirpichnikov Protein Eng. 2 195 (1996) a�Biochemistry (Engl. Transl.) b�Dokl. Chem. Technol., Dokl. Chem. (Engl. Transl.) c�Membr. Cell Biol. (Engl. Transl.) d�Mol. Biol. (Engl.

ISSN:0036-021X    

出版商:RSC    

年代:1999

数据来源: RSC