|
1. |
Phase equilibria, phases and compounds in the Ti–C system |
|
Russian Chemical Reviews,
Volume 71,
Issue 6,
2002,
Page 439-463
Aleksandr I. Gusev,
Preview
|
|
摘要:
Russian Chemical Reviews 71 (6) 439 ± 463 (2002) Phase equilibria, phases and chemical compounds in the Ti ±C system A I Gusev Contents I. Introduction II. Phase diagrams of the Ti ±C system at temperatures above 1000 K and thermodynamic properties of disordered titanium monocarbide TiCy III. New chemical compounds in the Ti ±C system IV. Ordered phases of nonstoichiometric titanium carbide TiCy V. Phase diagrams of the Ti ±C system constructed with allowance for the ordering of TiCy VI. Conclusion Abstract. investiga- theoretical and experimental of results The The results of experimental and theoretical investiga- tions system carbon ± titanium the in equilibria phase the of tions of the phase equilibria in the titanium ± carbon system are are generalised. characteristics thermodynamic generalised The generalised.The generalised thermodynamic characteristics of of disordered of Peculiarities reported. are carbide, titanium disordered titanium carbide, TiC TiCy, are reported. Peculiarities of the hypothetical and known the all of structures crystal the crystal structures of all the known and hypothetical com- com- pounds detail. in considered are carbon with titanium of pounds of titanium with carbon are considered in detail. The The X-ray all of identification allow which patterns diffraction X-ray diffraction patterns which allow identification of all these these compounds ± Ti the of diagrams phase The presented. are compounds are presented. The phase diagrams of the Ti ±C system of ordering atomic for allowance with constructed system constructed with allowance for atomic ordering of non- non- stoichiometric molec- the of existence the for and carbide, stoichiometric carbide, TiC TiCy, and for the existence of the molec- ular cluster-like compounds Ti dis- are Ti ular cluster-like compounds Ti8C12 and and Ti13C22 (TiC2) are dis- cussed.The bibliography includes 142 references cussed. The bibliography includes 142 references. I. Introduction Carbides of Group IV and V transition d-metals belong to strongly nonstoichiometric interstitial compounds.1 ±3 A typical representative of strongly nonstoichiometric compounds is cubic titanium monocarbide TiCy (0.484y41.00) with the B1-type structure, which has found wide practical applications and has been the subject of intensive experimental and theoretical research.In 1823, Wollaston reported the discovery of pure crystalline titanium in metallurgic slags. However, after a lapse of 33 years WoÈ hler (Germany) showed that these crystals were titanium carbide containing nitrogen impurity. More recently in the 19th century contaminated titanium carbide was extracted from tita- nium-containing cast iron.4, 5 Some attempts at obtaining pure titanium carbide by carbothermal reduction of titanium dioxide, TiO2, were also made (see Ref. 6). The history of studies of titanium carbide and the Ti ±C system can arbitrarily be divided into three periods. In the first period (from the middle 1850s up to 1940), titanium carbide as well as the carbides and nitrides of other transition metals (zirconium, hafnium, vanadium, niobium, tantalum) were treated as ordinary chemical compounds of stoichiometric com- A I Gusev Institute of Solid State Chemistry, Urals Branch of the Russian Academy of Sciences, ul.Pervomaiskaya 91, 620219 Ekaterinburg, Russian Federation. Fax (7-343) 274 44 95. Tel. (7-343) 274 73 06, (7-343) 249 35 23. E-mail:gusev@ihim.uran.ru Received 4 April 2002 Uspekhi Khimii 71 (6) 507 ± 532 (2002); translated by A M Raevsky #2002 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2002v071n06ABEH000721 439 440 448 452 458 460 position. Most studies carried out in this period were related to the establishment of the types of crystal structure and to the determi- nation of the unit cell parameters of the titanium carbide samples obtained in one way or another.Titanium carbide was assumed to have a stoichiometric composition TiC. By the end of the first period, investigations of the thermodynamic and various physical properties of carbides and nitrides began and massive discrep- ancies between the results of different studies of the properties of titanium carbide were revealed. These discrepancies could not be rationalised by the measurement errors or by differences in the experimental methods. Detailed crystal-chemical studies showed that the discrepancies were due to the different content of carbon atoms in the carbide. It was found that, in contrast to other compounds, the composition of titanium carbide and, broadly, of the Group IV and V transition metal carbides and nitrides can vary over a very wide range while their crystal structure remains unchanged. In particular, the composition of titanium monocar- bide TiCy can vary from TiC0.48 to TiC1.00.In the second period (from 1940 to 1970s), emphasis was mainly placed on investigations of the composition ± property correlations for strongly nonstoichiometric compounds (in par- ticular, interstitial phases). Interest in the carbides peaked in the 1960s, when thousands of communications were reported, concerning the effect of composition on the crystal-chemical, thermodynamic, thermal-physical, electrical, magnetic, optical, mechanical and other properties of these substances.By this time, the main qualitative concepts of the nature of chemical bonding in nonstoichiometric carbides have been developed. In the 1970s, the number of publications concerning basic problems relating to nonstoichiometric carbides reduced considerably; however, this time was characterised by increasing interest in applied studies. Titanium carbide had found wide use in production of hard alloys for metal-cutting tools. Most of the researchers erroneously concluded that there was almost nothing left to study in non- stoichiometric carbides. The first (tentative) variants of the phase diagram of the Ti ±C system were reported,7, 8 which reflected the existence of the only compound, viz., cubic titanium monocarbide, TiCy, with a broad homogeneity region.These phase diagrams remained nearly unchanged up to the mid-1990s. For a long time after nonstoichiometric compounds were shown to have broad homogeneity regions, it was assumed that non-metal interstitial atoms and structural vacancies (unoccupied interstices of the metal sublattice) were randomly distributed over the lattice. However, crystallographic studies reported in 1967 (see Ref. 9) and more recently revealed that under particular condi-440 tions the interstitial atoms and vacancies can be redistributed over the sites of the crystal lattice, thus forming different types of ordered structures. The ordering of carbon atoms and structural vacancies was first observed in nonstoichiometric titanium car- bide.9 In the late 1960s, the third period began of studies of non- stoichiometric compounds, which is mainly related to investiga- tions into the ordering and its influence on the properties of these systems.Currently, one can say with certainty that at temper- atures below 1000 Knonstoichiometric carbides and nitrides exist in the thermodynamically equilibrium ordered state, whereas their disordered state is metastable in the same temperature range. In the 1980s, research into correlations between the properties of strongly nonstoichiometric compounds and the character of distribution of the interstitial atoms and vacancies began. It was shown that the effect of atomic and vacancy ordering on changes in the properties of a nonstoichiometric compound is comparable with the effect of varying the composition of the same system within the homogeneity region of the ordered phase.Therefore, varying the composition and redistribution of the atoms and vacancies in the crystal lattice of strongly nonstoichiometric compounds are two ways of controlling the properties of these compounds that possess equal rights. A large number of ordered phases of strongly nonstoichio- metric compounds was found during the decade from 1990 to 2000. The results of experimental studies and calculations of the phase equilibria allowed construction of the phase diagrams of the M±C and M±N (M=Ti, Zr, Hf, V, Nb, Ta) systems with allowance for both the presence of strongly nonstoichiometric carbides (MCy) and nitrides (MNy) and their ordering at temper- atures below 1300 K (see Refs 1 ± 3 and 10 ± 27).In 1992, a novel class of molecular metal ± carbon nano- clusters was discovered.28 ± 38 Three stable molecular clusters � Ti8C12, Ti13C14 and Ti13C22�were obtained in the Ti ±C system. Some experimental and theoretical studies on titanium dicarbide, TiC2, were also reported. Thus, the last two decades are characterised by avalanche-like growth of the body of information on the phase equilibria and compounds in the Ti ±C system. The phase diagram of this system was found to be much more intriguing and complicated than it was assumed previously. The aim of this review was to generalise the results of recent (reported between 1980 and 2000) experimen- tal and theoretical investigations of chemical compounds and phases in the Ti ±C system taking into account all available information.The emphasis is placed on the phase equilibria in this binary system and on the crystal structure of both the known and theoretically possible (model) phases and compounds of carbon and titanium. II. Phase diagrams of the Ti ±C system at temperatures above 1000 K and thermodynamic properties of disordered titanium monocarbide TiCy 1. Phase diagrams No phase diagrams of the Ti ±C system constructed with allow- ance for the ordering of nonstoichiometric carbide TiCy were reported until the mid-1990s. In the text below we will consider the principal results of investigations into the phase equilibria in the Ti ±C system at temperatures above the ordering temperature of nonstoichiometric titanium carbide (i.e., at T>1000 K).Detailed studies of phase equilibria in binary carbide-forming M±C (Mis a transition metal) systems were mostly carried out in 1961 ± 1965.39 Up to now, the experimental results [including those for the Ti ±C system at temperatures above 1800 K(see Ref. 7)] obtained at that time are the most reliable and accurate. The phase diagram of the Ti ±C system is presented in Fig. 1. In this system, cubic titanium carbide TiCy with the B1-type T /K 3200 L 2800 2400 TiCy+L 19237K 2000 19418K 1.50.5% 321% b-Ti+TiCy 1600 0 20 Figure 1. Experimental phase diagram of the Ti ±C system at T>1800 K.7 structure is in equilibrium with b-Ti (bcc structure) or a-Ti (hcp structure).The melting temperature of pure metallic b-Ti is 19418K (see Ref. 7). The addition of carbon to titanium reduces the melting temperature of Ti down to the eutectic temperature [19185,40 19237,7 19215,41 ± 43 1925K (see Ref. 44)]. The eutectic L.b-Ti+TiCy was reported to contain 4.4 at.% of carbon;40 however, metallographic tests7 showed that the content of carbon in the eutectic is 1.50.5 at.%. Difficulties in determi- nation of the eutectic composition by metallographic tests have been pointed out.7, 39 Allotropic transformation a-Ti.b-Ti occurs at 11555 K (see Ref. 8). The addition of carbon to titanium causes an increase in the temperature of this transformation up to the temperature of the b-Ti+TiCy.a-Ti peritectoid transformation, which equals 11933 K (see Refs 41 ± 43) or 1205 K (see Ref. 44).At temperatures above 1000 K, there is only one nonstoichio- metric compound in the binary system Ti ±C, viz., cubic titanium carbide TiCy with a very broad homogeneity region from TiC0.47 to TiC0.96 at T&1920 K.7, 39 Raising the temperature up to 3000 K causes a shift of both the lower and upper bounds of the homogeneity region of TiCy towards a higher content of carbon. At T&3000 K, the upper bound of the homogeneity region corresponds to a TiC0.998 composition (see Ref. 8). The highest melting temperature, 334015 K (see Refs 7, 39, 41 ± 43) or 3341K (see Ref.44), has carbide TiC0.80 which undergoes con- gruent melting. The temperature of the L.TiCy+C eutectic containing *63 at.% of carbon is 304912 K,7 30496 K (see Refs 39, 41, 42) or 3048 K (see Ref. 44). Figure 2 presents the phase diagram of the Ti ±C system constructed by Murray 41, 42 for T>800 K. The same phase diagram was also reported in a reference guide.43 Most of the phase equilibria in this diagram were shown tentatively. Murray supplemented the phase diagram shown in Fig. 1 with some essential details by indicating the regions of allotropic trans- formation a-Ti.b-Ti and peritectoid transformation b-Ti+TiCy.a-Ti. On the other hand, the phase diagram of this system (see Fig. 2) contained a tentatively shown existence region of ordered Ti2C phase that was thought to exist and be stable at temperatures below 2170 K.9, 45 However, the first report 9 on the stability of the ordered Ti2C phase at T42170 K was found to be erroneous: more recently, numerous studies of the ordering of nonstoichiometric carbide TiCy revealed that cubic ordered Ti2C phase (space group C5 4h) is formed at temperatures below 1000 K (see, e.g., Refs 12, 21, 46 ± 56).At T<1000 K, A I Gusev 334025K TiCy+L 441% 304912K *49.4%631% TiCy+C TiCy*48.8% 40 C (at.%)Phase equilibria, phases and chemical compounds in the Ti ±C system T /K L 3000 2500 1921K 2000 *32% *1.8% b-Ti 1500 1193K *38% a-Ti 1000 30 0 10 20 Figure 2. Phase equilibria in the Ti ±C system at T>800 K.41 ± 43 Ti2C is the ordered phase of nonstoichiometric titanium carbide TiCy.T /K L 3500 3341K 3000 2500 TiCy L+TiCy 1925K 2000 31.3% 2.6% b-Ti+TiCy 1500 1205K a-Ti+TiCy 34.4% 1000 40 0 20 Figure 3. Phase diagram of the Ti ±C system at T>1000 Kcalculated 44 taking into account the results of experimental studies.7, 8, 39 there are several ordered phases in the homogeneity region of carbide TiCy that are not shown in this phase diagram. Phase equilibria in the system under study at temperatures above 1000 Khave most reliably been shown in the phase diagram calculated by Jonsson 44 taking into account the experimental data reported by Rudy et al.7, 39 Jonsson's phase diagram is shown in Table 1.Crystal structure of phases that exist in the Ti ±C system at T>1000 K. Phase Existence region Space group [C (at.%)] h) h) 0 ± 0.6 0 ± 1.6 *32.0 ± 49.8 100 Im3m (O9 P63/mmc (D4 Fm3m (O5 P63/mmc (D4 b-Ti a-Ti TiCy C 3340K 3049K *63% TiCy TiCy+C Ti2C? 50 40 C (at.%) L+C 3046K65.1% TiCy+C C (at.%) 60 Lattice type Pearson notation 6h) A2 (W) A3 (Mg) B1 (NaCl) A9 (graphite) cI2 hP2 cF8 hP4 6h) 441 Table 2. Characteristics of special points in the phase diagram of the Ti ±C system at T>1000 K. T /K Phase transformation Phase trans- formation type Composition of phases involved in reaction [C (at.%)] 32 1.8 0.6 L.b-Ti+TiCy 19185 eutectic transformation 44.4 7 7 L.TiCy 334015 congruent melting L.TiCy+C 65.1 48.5 *100 30506 eutectic transformation b-Ti+TiCy.a-Ti 0.4 34.4 2.1 11933 peritectoid transformation 0 7 7 L.b-Ti 19418 melting 0 7 7 b-Ti.a-Ti 11555 allotropic transformation Fig.3. The results of investigations of the crystal structure of different phases and the characteristics of special points of the phase diagram of the Ti ±C system at T>1000 K are listed in Tables 1 and 2, respectively. 2. Structure and thermodynamic properties of disordered titanium carbide a. Structure Disordered cubic titanium monocarbide TiCy with the B1-type structure is the most important chemical compound that exists in the Ti ±C system at temperatures above 1000 K. The unit cell of TiCy comprises two face-centred (metal and non-metal) sublatti- ces displaced by half the lattice constant with respect to each other (Fig.4). All sites in the metal sublattice of TiCy are occupied by titanium atoms. The number of sites in the non-metal (carbon) sublattice is larger than that of interstitial atoms (i.e., carbon atoms), which is typical of nonstoichiometric compounds. In the disordered state, which is in thermodynamic equilibrium only at T>1000 Kand is treated as a quenched metastable state at lower temperes, carbon atoms are randomly distributed over the sites. The non-metal sublattice is characterised by the same site occupation probability equal to the value of y (the relative content of carbon in the carbide).This is responsible for the retention of cubic symmetry of the disordered non-metal sublattice and means that all sites in the carbon sublattice of disordered TiCy are crystallographically equivalent. A typical X-ray diffraction pat- tern of homogeneous cubic titanium carbide is shown in Fig. 5. The lattice constant of TiCy nonmonotonically increases as the content of carbon increases (Fig. 6).57 Nonstoichiometric titanium carbide can possess quite differ- ent thermodynamic properties within the homogeneity region. Therefore, of particular value are those studies in which the Ti C Figure 4. Unit cell of cubic disordered nonstoichiometric TiCy with the B1(NaCl)-type structure. All sites in the metal sublattice are occupied by Ti atoms, while the sites in the non-metal sublattice are randomly occupied by C atoms.442 40 60 Figure 5.X-Ray diffraction pattern of disordered titanium carbide TiC0.52 with cubic B1-type structure (aB1=0.4306 nm, CuKa1,2 radia- tion). aB1 /nm 0.432 0.431 II I 0.430 0.6 0.5 Figure 6. Lattice constant aB1 of the basic crystal lattice as a function of the composition of titanium carbide, TiCy, in the quenched disordered (full circles) and annealed ordered (open circles) states.57 Samples obtained by sintering Ti and TiC0.94 (1, 2 ), by sintering Ti and C (3, 4 ) and by hot pressing Ti and TiC0.98 (5 ). The existence regions of the ordered phases Ti2C, Ti3C2 and Ti6C5 are labelled I, II and III, respec- tively. measured thermodynamic characteristics were attributed to a certain composition of the carbide with the known (disordered or ordered) structural state.The presence of the temperature and concentration dependences of thermodynamic characteristics of disordered carbide allows quantitative description of high-tem- perature phase equilibria involving TiCy; however, of much greater importance is that using these data also provides the possibility of direct investigations into the disorder ± order phase transformations. Unfortunately, the available experimental data on thermodynamic characteristics of nonstoichiometric titanium carbide are scarce. b. Enthalpy (heat) of formation Burning of a substance in oxygen in a calorimetric bomb is the main method of experimental determination of the formation enthalpy.Most of the data on the formation enthalpy of disor- dered TiCy were obtained using this technique. The formation enthalpy can also be determined indirectly from the dependence of the partial pressure of the components of a compound on the evaporation temperature. Calorimetric measurements of the standard formation enthalpy (DHf298) of titanium carbide of a nearly stoichiometric Intensity (arb.u.) 80 100 2y /deg 12345 III 0.8 0.7 y 0.9 A I Gusev composition TiC0.996 by burning of carbide in oxygen revealed 58 that DHf298=7183.81.7 kJ mol71. Calculations using the refined value of the formation enthalpy of titanium dioxide gave a more correct value: DHf298=7184.6 1.7 kJ mol71.8 Morozova et al.59 determined the formation enthalpies of disordered cubic carbides of compositions TiC0.79, TiC0.91 and TiC1.00 by burning them in oxygen and calorimetrically measuring the heat effect.The DHf298 values for these carbides were found to be 7183.02.1, 7206.80.8 and 7192.6 2.1 kJ mol71, respectively. The dependence of the formation enthalpy (in kJ mol71) determined by burning technique 59 on the composi- tion of TiCy can be described by the following formula:1 ± 3, 27 DHf298(y)=791.077186.7 y+89.0 y2. f 298 (in The formation enthalpies of carbides TiC0.63, TiC0.72, TiC0.80 and TiC0.91 (7108.07.0, 7135.07.0, 7157.07.0 and 7184.07.0 kJ mol71, respectively) were determined from the data on vacuum evaporation of TiCy in the temperature range from 1873 to 2673 K.60, 61 The dependence of DH kJ mol71) on the composition of TiCy calculated using these values has the form 61 DHf298(y)=135.67467.8 y+128.3 y2.Comparison of the DHf298(y) dependences obtained from calorimetric and evaporation measurements showed that these results are in satisfactory agreement for carbides with y>0.9 but disagree at lower content of carbon. These dependences cannot be combined since the discrepancies between them seem to be due to the systematic errors of the methods employed. However, one can assume that the formation enthalpy of carbides TiCy (y<0.9) lies between the values found from calorimetric and evaporation measurements. Analysis of the experimental data on the formation enthalpy of titanium monocarbide taking into account the position of the lower bound of the homogeneity region of TiCy suggested 18 that the dependence of DHf298 (in kJ mol71) on the carbide composi- tion has the form DHf298(y)=16.757304.49 y+89.04 y2.f 298(y) (1) Mention may be made that dependence (1) is close to the average between the experimental dependences DH obtained from the results of calorimetric 59 and evaporation 61 measurements. c. Enthalpy (heat content) High-temperature enthalpy (HT7H298) measurements are usu- ally carried out by a mixing technique using an adiabatic-shell calorimeter. The high-temperature enthalpy of titanium carbide was measured 62 ± 65 and the following expressions for description of the temperature dependences of the enthalpy (in kJ mol71) were proposed.For carbide of composition TiC0.99 in the temperature range from 400 to 1800 K (see Ref. 62): (2) HT7H298=719.95+49.5361073T+ +1.6761076T2+1499T71. For carbide of composition TiC0.95 (the measurement error was 1.4%) in the temperature range from 1300 to 2700 K (see Ref. 63) HT7H298=713.68+41.1361073T+4.4861076T 2. (3) Storms 8 found the following formula for calculating the enthalpy (in kJ mol71) of stoichiometric compound TiC1.0 in the temperature range from 300 to 3000 K (4) HT7H298=722.19+55.6761073T7 74.0761076T 2+1763.6 T71+16.1610710T 3.Phase equilibria, phases and chemical compounds in the Ti ±C system and TiC0.99 The enthalpies of disordered carbides TiC0.64, TiC0.71, TiC0.82 were measured in the temperature range 1300 ± 2500 K by a mixing technique.64, 65 The authors suggested a polynomial description of the temperature dependence of the enthalpy; however, the polynomials did not fulfil the obvious normalisation condition: H T7H298 :0 at T=298 K.Because of this, we calculated the temperature dependences of the enthalpy (in kJ mol71) of these carbides taking into account the normalisation condition 1, 27 using the experimental data reported by Fesenko et al.64 and the following expression HT7H298=A+BT+CT 2+DT71. The corresponding formulae are as follows: for carbide of composition TiC0.64 HT7H298=72.42+24.5461073 T+ +7.4761076T271657.1T71, for carbide of composition TiC0.71 HT7H298=72.03+24.4661073 T+ +8.1261076T 271780.9T71, for carbide of composition TiC0.82 HT7H298=75.75+30.9061073 T+ +6.5461076T 271204.9T71 and for carbide of composition TiC0.99 HT7H298=712.97+36.6161073T+ +6.7961076T 2+431.1 T71.These equations can be combined to derive a common function (HT7H298)TiCy = =(15.86+12.0361073 T+9.6761076T276051.5T71)+ (722.80+6.1761073 T73.0961076T 2+6328.2T71) y+ +(75.680+19.0661073 T) y2 (kJ mol71), which allows estimation of the enthalpy of disordered monocar- bide TiCy with any content of carbon in the temperature range from 300 to 2000 K. d. Heat capacity and entropy The heat capacity, Cp , of titanium carbides of compositions TiC0.95 and TiC0.99 was measured using the adiabatic technique in the temperature range from 12 to 300 K.66 A decrease in the content of carbon in the TiCy carbide is accompanied by a decrease in Cp .Lorenzelli et al.67 measured the heat capacity of samples of disordered and ordered carbide of composition TiC0.625 between 0.5 and 5.0 K; however, they presented only the difference between the heat capacities of the samples. Since Cp=dÖHT ¡ H298Ü, dT the high-temperature heat capacity is usually determined using the results of experimental enthalpy measurements. This allows the functions (5) ± (9) to be used for estimating the heat capacity of disordered TiCy. The entropy, S 298, of stoichiometric titanium monocarbide TiC1.0 is 24.24 J mol71 K71 (see Ref.8). No experimental data on changes in entropy within the homogeneity region of cubic carbide TiCy are available so far. To a first approximation, the dependence of entropy on the composition of carbideMCy can be (5) (6) (7) (8) (9) 443 represented by the additive sum of the entropies of the compo- nents normalised to the entropy of stoichiometric carbide MC1.0 , i.e., (10) S298(y)=k(S298M+yS298C), 298 TiC where k=S298MC1:0 /(S298M+S298C) is the normalisation con- stant and S298M and S298C are the entropies of the metal and carbon, respectively. The entropies of a-Ti and carbon are 30.72 (see Ref. 68) and 5.39 J mol71 K71 (see Ref. 69), respectively. Taking into account these values and the value S 1:0=24.24 J mol71 K71 (see Ref.8), the normalisation constant, k, was found to be 0.671. To a first approximation, the dependence of entropy (in kJ mol71 K71) on the composition, TiCy, has the form (11) S298(y)=20.62+3.62 y. e. The free energy The free energy of disordered carbide can be calculated using the dependences (1), (9) and (11). Indeed, thermodynamic properties of solids are usually described assuming a constant pressure (usually, p is set to 0 or 1). Under isobaric conditions, the internal energy (configuration energy of a crystal), E, equals the enthalpy H. At 298 K, the enthalpy of a compound is equal to the formation enthalpy of this compound from corresponding elements, i.e., H298:DHf298. Taking into account the temperature dependence of the enthalpy of nonstoichiometric compound MXy, from here it follows that (12) E(y,T)=H(y,T)=H298(y)+ÖHT ¡ H298ÜMXy: :DHf298(y) +ÖHT ¡ H298ÜMXy .The free energy (F) of a disordered nonstoichiometric com- poundMXy is given by the following expression 1 ± 3, 14, 17, 20, 22, 70 (13) F(y,0,T)=E(y,T )7T[S(y,0,T)+Sc(y,0)]: :F0(T)+yF1(T)+y2F2(T )7TSc(y,0), where S(y,0,T ) is the non-configurational entropy, which is an explicit function of temperature; Sc(y,0) is the combinatorial (configurational) entropy, which is independent of temperature; and F0(T ), F1(T ) and F2(T ) are the free energy parameters. Using the standard entropy, the temperature dependence of entropy can be written as follows S(T)=S dÖHT ¡ H298Ü .T 298 á ÖT 298 In this case dÖHT ¡ H298ÜMXy . (14) T S(y,0,T)=S298ÖyÜ á ÖT 298 (15) The free energy of a disordered compound MXy minus the term responsible for the contribution of configuration entropy, Sc(y,0), is given by the expression F0(T)+yF1(T)+y2F2(T ):E(y,T)7TS(y,0,T) or, taking into account Eqns (12) and (14), (16) F0(T)+yF1(T)+y2F2(T)=DHf298(y)+ dÖHT ¡ H298ÜMXy : T +(HT¡H298)MXy¡T S298ÖyÜ á ÖT 298 The expression for a fraction of the free energy (in kJ mol71) of disordered carbide TiCy was derived 1 ± 3, 27 from Eqn (16) using the dependences (1), (9) and (11)444 (17) F0(T)+yF1(T)+y2F2(T )= =(32.61+31.6161073 T79.6761076T 273026T717 712.0361073T lnT)+(7327.29+71.4961073 T+ +3.0961076T 2+3164T7176.1761073T lnT)y+ +(83.36+127.6561073T719.0661073 TlnT )y2, where the parameters F0(T ), F1(T ) and F2(T) of the free energy F(y,0,T ) [see Eqn (13)] are given in explicit form.Knowledge of these parameters allows all types of thermodynamic calculations of the phase equilibria involving nonstoichiometric titanium carbide. 3. High-temperature evaporation of titanium carbide Thermodynamic properties of nonstoichiometric carbides are often determined using measurements of the vapour pressure above the solid compounds. Of particular interest is the case of congruent evaporation. According to the kinetic theory of gases, the rate of evapo- ration from the surface that is in equilibrium with vapour equals the condensation rate, which is proportional to the equilibrium vapour pressure p.Under nonequilibrium conditions (in the case of, e.g., vacuum evaporation from the open surface) the evapo- ration rate is independent of the condensation rate and equals the equilibrium evaporation rate at a given temperature;71 this holds if the residual pressure in the vacuum chamber does not exceed 10 Pa. Therefore, knowledge of the rate of steady-state vacuum evaporation allows determination of the equilibrium parameter, namely, the saturated vapour pressure, and calculation of the thermodynamic characteristics. If transition from the condensed phase to the vapour phase is accompanied by dissociation of the former, the observed evaporation rate and the partial pressure are lower than the equilibrium values; this can be taken into account by introducing the evaporation coefficient a41.0.The kinetic and thermodynamic conditions for congruent evaporation of disordered nonstoichiometric interstitial com- pounds have been studied in detail in monograph.14 Thermody- namics of congruent evaporation of nonstoichiometric compounds was also developed.72, 73. Evaporation of TiCy at temperatures above 1800 K was studied by many authors. Mass spectrometric studies of the composition and the pressure measurements of the vapour above titanium carbide evaporated from a graphite effusion chamber (evaporation in the presence of excess of free carbon) at 2200 ± 2500 K showed that the vapours contained elemental carbon and titanium and the C2 and C3 units; no molecular species TiC were found.74, 75 The vapour pressure of titanium above the titanium carbide ± graphite system at 2383 ± 2593 K measured 76 by the Knudsen effusion technique was found to be much higher than that determined by other authors.The value reported in this study seems to be overestimated, since evaporation of pure graphite through the effusion orifice was ignored. In the temper- ature range between 2518 and 2790 K, the vapours above the titanium carbide ± carbon system contained Ti, TiC2, TiC4 and C3 units.77 According to estimates,78 the gas phase above this system contains elemental titanium and carbon as the major components at T<2500K and Ti, C and C3 units at T>2500 K.Titanium carbide was shown to undergo dissociation during evaporation at 2200<T42500 Kin the presence of excess of free carbon; in this case, elemental titanium is released in the gaseous state while carbon is released in the solid state.74 ± 76 Studies of evaporation of carbide TiC0.97 at 2773 K using Langmuir's technique revealed that under these conditions the compound undergoes congruent evaporation and that the vapours contain mostly elemental titanium and carbon.79 The composition of carbide TiCy affects the vapour pressure, namely, the vapour pressure of titanium monotonically increases from 161074 Pa for TiC0.91 to 0.2 Pa for TiC0.55 as the content of carbon decreases (see Ref. 8). Evaporation and dissociation of A I Gusev carbide TiC0.91 in the temperature range between 1873 and 2673 K can be described by the reaction 80 (18) TiCy (s) Ti (g)+[y7z(T )]C (g)+z(T)C (s), where z(T)=y at T42100 K; z(T ) decreases from y down to 0 as the temperature increases from 2100 to 2273 K; z(T)=0 at T52273 K.The vapour pressure of titanium above carbide TiC0.91 at 1900K was found to be *661075 Pa,80 which is close to the value reported by Storms.8 Evaporation of titanium carbide was mainly studied near the upper bound of the homogeneity region of the cubic phase taking the TiCy samples with y>0.9 as examples (see Refs 74 ± 77, 79 and 80). The evaporation coefficient was assumed to be equal to unity, though if evaporation is accompanied by dissociation, this value is less than 1.The results of determination of the vapour pressure of titanium reported by different authors 8, 74 ± 77, 79, 80 are substantially different (from 2 to 40 times). Vacuum evaporation (at 0.0013 Pa) of samples of cubic monocarbide TiCy with different contents of carbon (0.6<y40.93) was studied in the temperature range between 1873 and 2673 K.60, 61 The samples represented disordered titanium carbide, since only disordered nonstoichiometric car- bides are in thermodynamic equilibrium at T>1300 ± 1500 K.1 ± 3, 14, 17, 70 All TiCy samples were found to be homoge- neous and contained one phase with the B1-type structure. Evaporation of titanium carbide was studied by Langmuir's technique. The evaporation rate V= Dm Sef t was determined from the mass loss (Dm) from the effective evaporation surface area, Sef, of a vacuum heated sample over a period t. Due to the presence of the open surface pores the Sef value was larger than the geometric evaporation surface area S , 1=3 9pP2 2 Sef=S 1 á where P is the relative porosity.The evaporation surface area was determined 60, 61 with allowance for porosity which was deter- mined as the average between the initial (before the experiment) and final (after heating) porosity. Evaporation was carried out until establishment of a constant total evaporation rate V (for a given temperature) and a constant crystal lattice parameter on the surface of the sample, i.e., until steady-state regime of the process. The overall time of vacuum evaporation of titanium carbide samples varied from 2 ± 8 h at 2673 K to 54 h at 1873 K.During evaporation the lattice constant on the surface of the samples remained virtually unchanged, which indicated a constant com- position of TiCy at T<2673 K. Evaporation of carbides TiCy at T<2273 K led to release of elemental carbon on the surface of the samples, whereas no elemental carbon was found on the surface of the TiCy samples at higher evaporation temperatures (T52273 K). These data and the results of evaporation rate measurements suggest that at T42273 K evaporation of titanium carbide TiCy is described by reaction (18). Indeed, at T<2100K the pressure and evapo- ration rate of metallic titanium are much higher than those of elemental carbon;68 because of this, carbon is mainly released in the solid phase and can thus be found on the surface of the samples.As the content of carbon in TiCy decreases (especially at y40.8), the amount of elemental carbon released on the surface rapidly decreases. Raising the temperature leads to an increase in the evaporation rate of carbon and no elemental carbon was found on the surface of TiCy samples at T>2300 K. At T>2273 K, which corresponds to complete evaporation of carbon released on the surface, the z(T ) value equals zero, and the equilibrium in system (18) loses ambiguity. Since at temper- atures below 2600K (in the absence of excess elemental carbon) the vapour above titanium carbide contains mostly elementalPhase equilibria, phases and chemical compounds in the Ti ±C system Table 3.Total evaporation rate V(kg m72 s71) and partial pressure pTi (Pa) in the temperature range between 2273 and 2673 K, evaporation coefficients aTi and aC, atomisation enthalpy (DatH298 /kJ mol71), formation enthalpy (DHf298 / kJ mol71) and evaporation enthalpy (DsH298 /kJ mol71) of TiCy samples.61 TiCy lnV=a7b/T aC aTi ln pTi=a7b/T (see a) a b a b see b 0.400.05 29.5380.084 73836209 10282 10252 71087 55370 0.450.05 29.7210.035 7633987 11192 11152 71357 64470 0.450.10 29.9870.170 78601423 11986 11934 71577 72370 TiC0.63 TiC0.72 TiC0.80 TiC0.91 0.800.05 0.900.05 0.900.10 0.600.05 72609225 7511185 77011322 7911567 0.300.05 29.9210.023 8034257 a 21.5450.091 21.8820.034 21.9980.129 21.7000.027 C =lnpTi +lny.b Found using the equation of the second law of thermodynamics (see below). c Found using the equation of the For carbide TiCy, lnp third law of thermodynamics (see below). titanium and carbon and does not contain C2 units and larger carbon molecular species and titanium carbide molecules,78 one can suggest congruent evaporation of TiCy in the temperature range between 2273 and 2673 K following the reaction (19) Ti (g)+yC (g). TiCy (s) This character of evaporation of TiCy has been confirmed experimentally.78 ± 80 An increase in the evaporation temperature causes a rapid increase in the evaporation rate: e.g., the evaporation rate of TiC0.92 increases by a factor of more than 20 000 as the temper- ature increases from 1873 to 2673 K.For TiCy samples with different contents of carbon the temperature dependence of the evaporation rate has the form lnV=a7 bT . In Table 3 we present the values of the coefficients a and b for some carbides. The evaporation rate of carbide of composition TiC0.92 calculated using this dependence for 2773 K is 10861075 kg m72 s71. This is rather close to the estimate (93 ± 113)61075 kg m72 s71 for the evaporation rate of TiC0.97 at 2773 K.79 If a compound MCy undergoes congruent evaporation, the observed partial pressures of its components, pM and pC, are 14, 71, 80 (20) pM à aMpM à VÖ2pRTAMÜ1=2 , M (21) pC à aCpC à yVÖ2pRTACÜ1=2 , M where aM, aC and pM, pC are the evaporation coefficients and the equilibrium partial pressures of the metal and non-metal compo- nents of condensed carbide MCy; V is the total evaporation rate; AM and AC are the atomic masses of the metal and carbon, respectively, andM=AM+yAC is the molecular mass of carbide MCy .Under equilibrium congruent evaporation one gets pC=ypM; hence from Eqns (20) and (21) it follows 14 that 1=2 C , aC=aM A AM pM. 1=2 AC pC=y AM In Fig. 7 we present the dependences of the partial pressure of titanium on the composition of carbide TiCy. At any temperature, the larger the deviation of the carbide composition from the stoichiometric composition, the higher the pTi value.Earlier, an analogous type of change in pTi was reported for T=1900 K.8 DatH298 13031 12991 71847 82870 pTi /Pa 100 1071 1072 1073 1074 123 1075 0.7 0.6 Figure 7. Partial pressure of titanium, pTi, as a function of the compo- sition of titanium carbide, TiCy,: pressure pTi above the TiCy carbide at 1900 K (1 );8 data 61 (1 ) and pressure pTi above the TiC0.97 carbide at 2773 K (3 ).79 Taking into account the relationship 1=2 C pM, A pC=y AM y=2 C K à pMpC à pÖ1áyÜ M à aÖ1áyÜÖp the equilibrium constant, K, of reaction (19) is (when calculating K values, the partial pressure is given in atmospheres): A yy AM Knowledge of the equilibrium constant K allowed estimation of the enthalpy of process (19);61 in the case of evaporation accompanied by dissociation this is the atomisation enthalpy DatH298.In accord with the second law of thermodynamics, the DatH298 value was found using the available temperature depend- ences of K and by solving Van't Hoff's isobar equation d lnK dT à DRT atH2T . Calculations of DatH298 were carried out with inclusion of correction D for the temperature dependence of the enthalpy of the components of reaction (19): 445 D DH sH298 f 298 see c 2673K 2573K 2473K 2273K 2173K 2073K 1873Ky 0.9 0.8 y=2 C . (22) M AAM MÜÖ1áyÜyy (23)446 D=(HT7H298)Ti(g)+y(HT7H298)C(g)7 7(HT7H298)TiCy(s). Ultradispersed transition-metal carbide powders are usually obtained by the interaction of chlorides of corresponding elements with dihydrogen and methane or other hydrocarbons in the low- temperature (4000 ¡¾ 8000 K) argon plasma.The particles of such powders are single crystals 10 to 100 ¡¾ 200 nm in size. Finely dispersed titanium, zirconium, niobium, tantalum, boron and The DatH298 values for carbides TiCy under study were found by the least-squares minimisation of the expression RlnK(T)=c ¢§ DatH298 a D, T where c is a constant, using the experimental dependences K(T ). The heat effect of reaction (19), i.e., the atomisation enthalpy of carbide TiCy, was also determined using the equation of the third law of thermodynamics. DatH298 a TDF0T ¢§ RTlnK T. with allowance for the calculated temperature and concentration dependence of the reduced potential DF0 The calculated standard atomisation enthalpies, DatH298, of titanium carbides of different composition and the evaporation coefficients aM and aC are listed in Table 3.The parameters of the dependences lnV¡¾T and ln pTi ¡¾ T (see Table 3) correspond to the temperature range of congruent evaporation of TiCy (from 2273 to 2673 K). Intensity (arb.u.) The dependence of the atomisation enthalpy on the composi- tion of carbide TiCy calculated by the least squares method using the DatH298 values calculated from the equations of the second and third laws of thermodynamics has the form: (24) silicon carbide powders were prepared by a plasmachemical technique.93, 94 Gas-phase synthesis involving laser heating of the mixture of reactant gases is a kind of plasmachemical synthesis.95 ¡¾ 98 Laser heating allows control of homogeneous nucleation and excludes contamination. As the intensity of laser radiation increases, the size of nanoparticles decreases due to an increase in the temper- ature and in the heating rate of the reactant gases.In the last decade, a mechanochemical technique has found wide use for the synthesis of nanocrystalline carbides. Mechanical milling is the most efficient method for production of large amounts of powders of different nanomaterials including car- bides. Mechanochemical synthesis of nanocrystalline titanium carbide from a mixture of the metal and carbon powders was reported.99 The mixture was ball milled.Formation of carbide occurred after pulverisation over a period of 4 to 12 h; the average particle size of the nanocrystalline powder after 48 h milling was 71 nm. Nanocrystalline powder of cubic titanium carbide was also prepared by pulverising metallic titanium and graphite in a sapphire ball mill. The components were taken in a ratio that provided a Ti44C56 composition of the product.100 The ball-to- powder mass ratio was 10 : 1. Milling was performed at room temperature in an argon atmosphere. DatH298(y)TiCy=335.4+1177.8 y7128.3 y27. For compoundsMXy, the atomisation enthalpy is given by the expression (25) After pulverisation for 2000 s the X-ray diffraction pattern of the reactant mixture exhibited only broad reflections of elemental carbon and titanium (Fig.8). Milling over a period of 1.16104 s led to nearly complete disappearance of the graphite reflections. New reflections appeared after pulverisation for 1.56104 s cor- responded to a new cubic phase with the B1-type structure, viz., titanium carbide. The lattice constant of titanium carbide pre- DatH298(y)=DsHM a yDsHX ¢§ DHf298OyU, where DsHM and DsHX are the evaporation enthalpies of the components M and X, respectively, and DHf298OyU is the forma- tion enthalpy of compound MXy. The expression for the depend- ence of the heat of formation of carbide TiCy on its composition derived using the relationships (24) and (25) and the reference data on the evaporation enthalpies of titanium and carbon, DsHTi=471 and DsHC=710 kJ mol71 (see Ref.68) has the form 60, 61 1 (26) DHf298OyUTiCy=135.67467.8 y+128.3 y27. 23456 The DHf298 values for carbides of compositions TiC0.79, TiC0.91 and TiC1.00 were found to be 7157, 7184 and 7204 kJ mol71, respectively;61 this is in fairly good agreement with the results of earlier studies [cf. the DHf298 values of 7183, 7207 (see Ref. 59) and7184 kJ mol71 (see Ref. 58) for carbides of compositions TiC0.79, TiC0.91 and TiC0.996, respectively]. The dissociation enthalpy of a gaseous titanium carbide molecule is 47360 kJ mol71 (see Refs 68, 81). Using this value, the evapo- ration enthalpy of titanium carbide TiCy was also roughly estimated 60, 61 as the difference between the atomisation and dissociation enthalpies (see Table.3). 7 4. Nanocrystalline titanium carbide 89 20 Figure 8. Changes in the X-ray diffraction patterns of a powder mixture Ti44C56 as a function of milling duration (t ) in the mechanochemical synthesis of titanium carbide (CuKa1,2 radiation).100 t=26103 (1 ), 66103 (2 ), 86103 (3 ), 1.156104 (4 ), 1.56104 (5 ), 2.26104 (6 ), 46104 (7 ), 86104 (8 ) and 7.26105 s ( 9 ). Establishment of the fact that reduction of the grain or crystallite size down to and below a certain threshold value can cause substantial changes in the properties of solids82 ¡¾ 86 gave a strong impetus to investigations of the nanocrystalline state of matter. Pioneering studies in this field were reported by Gleiter 87 ¡¾ 90 who developed a method for fabrication of compact nanomaterials from nanopowders (the so-called `compaction'). Ultradispersed and nanocrystalline transition metal carbides can be prepared by plasmachemical synthesis, mechanochemical synthesis, intense plastic deformation and structure ordering.91, 92 Investigations of the nanocrystalline state of carbides are directed at making them less brittle and at improving their plasticity with simultaneous retention of high hardness typical of these compounds.(002)C (010)Ti (011)Ti (002)Ti (012)Ti (004)C (111)TiC (200)TiC (220)TiC (311)TiC (222)TiC 60 50 40 30 A I Gusev (112)Ti (103)Ti (201)Ti (110)Ti 70 2y /degPhase equilibria, phases and chemical compounds in the Ti ±C system pared in this manner was 0.4326 nm.Prolonged milling (46104 s) led to complete disappearance of the reflections of metallic titanium and to an increase in the intensity of titanium carbide reflections. After 86104 s, mechanical deformation of the powder particles developed and the grain size reduced dramatically, which was manifested in substantial broadening of reflections in the X-ray diffraction pattern. Pulverisation for 726104 s resulted in the formation of titanium carbide nanocrystals. No observable changes in the state of the carbide sample were detected upon further extension of milling duration to 106 s. Figure 9 illustrates the changes in the powder grain size during pulverisation according to X-ray diffraction and electron micro- scopy studies.The formation of nanocrystalline titanium carbide involves four stages. The first stage (milling for no longer than 1.16104 s) is characterised by the formation of Ti/C composite particles with an average size of*1000 nm. Metallographic tests revealed an onion-like structure of these species comprised of titanium and carbon layers. In the second stage, which takes from 1.16104 to 26104 s, solid-phase interaction between titanium and carbon occurs, resulting in coarse-grained titanium carbide particles 800 to 1000 nm in size. The third stage takes from 26104 to 86104 s and involves intense milling of titanium carbide grains followed by the formation of finely dispersed powder with a rather broad grain size distribution (from 5 to 100 nm in diameter); in this stage, titanium carbide grains form agglomerates of size 5 to 10 mm.The last, fourth, stage takes 86104 to 16106 s and is characterised by `grain size homogenisation' of the nanocrystal- line powder. Carbide powder thus obtained had a particle size of *21 nm while its grains were agglomerated into spherulites of size no greater than 300 nm. d /nm Ti 1000 TiC C Ti/C composite particles 100 10 12 1 TiC nanoparticles t /s 103 105 104 102 101 Figure 9. Grain size (d ) of milled Ti44C56 powder as a function of milling duration t in the mechanochemical synthesis of titanium carbide.100 The grain size was determined from broadening of X-ray diffraction reflections (1) and by electron microscopy (2).Titanium carbide powders prepared 100 by milling over differ- ent periods of time (2.26104, 46104, 86104 and 7.26105 s) were sintered in activated plasma. This type of sintering is a kind of hot pressing. Salient features of the method are the use of pulsed voltage for plasma activation and combination of nonaxial torsion with pressure. Sintering for less than 8 min resulted in compact titanium carbide samples of high density (up to 5.2 g cm73) while the average grain size remained less than 70 nm. The microhardness of the sintered samples (323 GPa) was no greater than that of conventional coarse-grained titanium carbide TiCy (0.94y41.0). The results obtained by El-Eskandarany et al.100 pointed to high efficiency of mechanochemical synthesis for fabrication of nanocrystalline titanium carbide; however, no data on the chem- 2Dnano-TiC Intensity (arb.u.) 447 ical composition of the carbide were reported in this study.The assumption 100 that the composition of mechanochemically syn- thesised titanium carbide is identical to that of the initial mixture (Ti44C56) seems to be doubtful. The reported lattice constant of titanium carbide is somewhat smaller, while the measured density is higher than the corresponding parameters of stoichiometric carbide TiC1.0 (0.4328 nm and 4.91 g cm73, respectively). The reason for such a high density of titanium carbide obtained by El-Eskandarany et al.100 is particularly incomprehensible.Even assuming that the synthesised carbide with the B1-type structure has the same composition as the charge [i.e., Ti0.44C0.56 (or TiC1.27 !?)], the metal sublattice of this carbide should contain defects, since for the B1-type structure the number of sites in the non-metal sublattice is the same as that in the metal sublattice and there are no possibilities of arranging any sort of excess atoms. If a particular sort of atom is in relative excess, the sublattice formed by the other sort of atom contains structural vacancies. In our case titanium carbide with the defect metal sublattice can have a composition Ti0.79C1.00; however, its density should be 20% lower than the calculated (theoretical) value. Noteworthy is that no vacancies were found so far in the metal sublattice of carbides; therefore, the metal sublattice of transition metal carbides is assumed to be perfect.Taking into account the aforesaid, the results obtained by El-Eskandarany et al.100 should be thoroughly checked. We believe that titanium carbide synthesised in that study was highly contaminated with metal, which led to higher density of the material compared to the theoretical value. A promising way of producing compact nanomaterials is to employ intense plastic deformation.85, 92, 101 Preparation of com- pact samples of nanocrystalline titanium carbide using this technique was reported.102 Milling and compacting experiments were carried out with coarse-grained (d&2 ± 5 mm) powder of nonstoichiometric carbide TiC0.62.Intense plastic deformation was produced by applying torsion under quasi-hydrostatic pres- sure. X-Ray diffraction studies revealed retention of the B1-type cubic structure of all plastic deformed titanium carbide samples; however, the diffraction reflections were substantially broadened (Fig. 10). Evaluation of the grain size from the broadening magnitude showed that the compacted nanocrystalline samples of nonstoichiometric titanium carbide had an average grain size lying between *30 and 50 nm. Contrary to metals, titanium carbide is very hard and brittle; therefore, it is of considerable interest that the grain size of titanium carbide was smaller than that of the metal samples after nearly identical plastic deforma- tion.2DTiC0.62 12 0.62 36.5 36.0 35.5 35.0 2y /deg Figure 10. Broadening of the (111)B1 reflection of nanocrystalline tita- nium carbide of composition TiC0.62 prepared by intense plastic deforma- tion (1) and coarse-grained (grain size*5 mm) carbide TiC0.62 (2).102 2Dis the total width of the reflection at half-maximum (CuKa1,2 radiation).448 III. New chemical compounds in the Ti ±C system The discovery of molecular clusters of carbon and the first observations of the C60 fullerene molecule 103 ± 105 gave an impetus to investigations into the synthesis, structure and properties of fullerenes (see, e.g., Refs 106 ± 108) and to an intensive search for methods of obtaining molecular clusters of other substances. By analogy with fullerenes, molecular clusters were expected to have unique physical and chemical properties that differ from the properties of the known polymorphous modifications of the same substances.1. Molecular cluster Ti8C12 In 1992,28 search for new molecular clusters led to the discovery of unusual, stable charged cluster Ti8Cá12 corresponding to a penta- gondodecahedral molecule of stoichiometric composition Ti8C12 (Fig. 11 a). In an ideal dodecahedral molecule, all atoms are arranged on a sphere whose surface is obtained by linking adjacent atoms and comprises twelve regular pentagons. In this molecule, all carbon and titanium atoms are tricoordinated (similarly to fullerene C60), occupy identical sites and are distrib- uted over the vertices of the dodecahedron in such a manner that the Ti atom is linked to carbon atoms only while six C2 units alternate with eight Ti atoms.The dodecahedral structure of Ti8C12 can be represented as a cube with eight titanium atoms at the vertices and a C2 unit in each face. The point symmetry group, Th, of this structure includes 24 symmetry elements (rotations and reflections). High symmetry of the met-car (metallocarbohedrene) molecule implies its stability. b a Ti C Figure 11. Dodecahedral structure of a Ti8C12 molecular cluster of Th symmetry (a) and a fragment of structure of Td symmetry (b).29 Yet another possible structure of the Ti8C12 cluster is charac- terised by the point symmetry group Td (see Fig. 11 b).29 In this structure, titanium atoms occupy two types of lattice sites and each type of site forms a tetrahedron. The smaller tetrahedron is rotated by 90 8 about the larger tetrahedron. The sites in the titanium sublattice differ in arrangement with respect to the C2 units.Indeed, six C2 units are arranged parallel to the edges of the larger tetrahedron built of Ti(1) atoms and perpendicular to the edges of the smaller tetrahedron formed by four Ti(2) atoms. The Ti(1) atoms are linked to three adjacent carbon atoms while the Ti(2) atoms form bonds with six carbon atoms. The Ti(1)7C and Ti(2)7C distances are 0.193 and 0.219 nm, respectively.29 The issue of which structural type is practically realised is still to be clarified since no molecular crystals of met-cars have been synthesised so far.Clusters Ti8C12 were synthesised using the gas-phase plasma- chemical technique with helium as inert gas, hydrocarbons (methane, ethylene, acetylene, propylene and benzene) and tita- nium vapours as reactants; the pressure of the gas mixture in the reactor was 93 Pa. A rotating metallic titanium rod was evapo- rated by exposing it to focussed radiation of aNd laser operated at 532 nm, thus creating an ionised metal vapour beam. Neutral and ionised clusters were separated from other reaction products and A I Gusev analysed mass spectrometrically. The mass spectra of the reaction products exhibited a sharp peak corresponding to the Ti8C12 molecule. In addition to neutral molecules, the mixture of ionised gases also contained stable Ti8Cá12 ions.The cluster Ti8C12 was assumed28 to be a member of a novel class of molecular clusters and called metallocarbohedrene or met- car (see Ref. 106). In met-cars, both the transition metal and carbon atoms form a cage-like structure. Indeed, other M8C12 clusters of such transition metals as Zr, Hf, V,30, 31 Cr, Mo and Fe 32 were obtained after a short time. Met-cars have been reviewed.33, 34 The high stability of the cluster Ti8C12 is thought to be due to peculiarities of its molecular and electronic structure; chemical bonds in the Ti8C12 molecule are similar to those in the carbon fullerenes.28 However, in contrast to C60 fullerene, both ionised and neutral M8C12 molecules are comprised of five-membered rings only.The shape of the surface of stable cluster Ti8C12 is similar to that of hypothetical unstable (and, hence, unfeasible) C20 fullerene. This fact shows that full similarity of chemical bonds in the clusters M8C12 to those in carbon fullerene is hardly probable. Indeed, calculations of the equilibrium crystal and electronic structure of the cluster Ti8C12 with Th symmetry (see Ref. 35) showed that the bonds between the titanium atoms and three adjacent carbon atoms differ from the bonds between carbon atoms in graphite or C60 fullerene. In particular, the Ti7C and C7C bond lengths in the Ti8C12 molecule differ by nearly a factor of 1.5 and equal 3.76a0 and 2.63a0 (a0=0.052918 nm is the radius of the first Bohr orbit), respectively.According to Reddy et al.,36 the Ti7C bond is *30% longer than the C7C bond. On the other hand, the distances from carbon and titanium atoms to the centre of the cluster are nearly equal. This means that the real dodecahedral structure of Ti8C12 is strongly distorted. The bond- ing states of the cluster Ti8C12 are thought to be composed of the Ti d-AOs and the C2 MOs while the Fermi level lies between the bonding and antibonding states of titanium, thus providing stability for the cluster.35 Analogous conclusions that clusters M8C12 have a distorted rather than ideal pentagondodecahedron shape were obtained in calculations reported by other authors. Comparative analysis of the electronic structure of the Ti8C12 met-car of Th and Td { symmetry showed 37 that in both structural types the highest occupied energy level corresponds to a sharp peak of the density of states (DOS) composed mainly of the C2p- and Ti3d-AOs.High chemical stability of compound Ti8C12 is due to a combination of strong Ti3d±C2p interactions between the titanium atoms and C2 unit on the one hand and to the C±C interactions in the carbon units on the other hand. In both structural types, Ti8C12 has an open electron shell and can there- fore be both electron acceptor and donor. The atoms in met-car molecules are tightly bound. For instance, the bond energy in the Ti8C12 molecule is 6.1 ± 6.7 eV per atom.35, 36, 38 For comparison, this parameter is 7.4 ± 7.6 eV per atom for C60 fullerene molecule (see Refs 109, 110) and 7.2 eV per atom for titanium carbide TiC with the B1 (NaCl)-type fcc structure.35 Modern concepts of the geometry and electronic structure of molecular clusters Ti8C12 (see Refs 29, 35 ± 38) provide good explanation for their chemical stability and specific features of the reactivity towards both polar and nonpolar compounds. Combination of high reactivity and stability of Ti8C12 clusters was confirmed experimentally.111 Molecular cluster Ti8C12 has been the most detailed studied new compound in the Ti7C system; however, no macroscopic amounts of this substance were obtained so far.{ Calculations for Th and Td symmetry were carried out using the structure parameters and interatomic distances reported by Reddy et al.36 and by Lin and Hall,29 respectively.Phase equilibria, phases and chemical compounds in the Ti ±C system 2.Polymorphous modification of dicarbide TiC2 Comparative analysis of the electronic structure of hypothetical metastable polymorphous modifications of carbide TiC1.0 with the B2 (CsCl)-, B8 (NiAs)-, B3- and B4 (sphalerite and wuÈ rtzite ZnS)-type structures and that of the real carbide TiC1.0 with the B1-type structure showed 112 that search for new stable modifica- tions of titanium carbide should be performed among those substances that are structurally similar to or derived from the B1 type. Titanium dicarbide, which can have several types of crystal structure, was studied as a representative of this group of substances.112, 113 The ThC2-type monoclinic structure (space group C2/c) has a tetrahedral body-centred metal sublattice while the C2 units occupy the octahedral interstices.In the CaC2 (C11a)-type tetragonal body-centred structure [space group I4/m (C54h)] the C2 units also occupy octahedral interstices. The C11a structural type can be derived from the B1-type lattice by extend- ing it along the c axis; sites in the non-metal sublattice are occupied by the C2 units oriented along the c axis. The unit cell of titanium dicarbide with this structural type comprises two formula units TiC2 (Fig. 12). The atomic coordinates in the unit cell are listed in Table 4; the running coordinate of carbon atoms, z/c=0.427, was determined assuming the same length of all Ti7C distances in the unit cell because each type of atom occupies sites of the same type.Figure 13 presents the X-ray diffraction pattern of tetrago- nal TiC2 calculated using the atomic coordinates listed in Table 4. a Figure 12. Unit cells of body-centred tetragonal TiC2 with the CaC2 (C11a)-type structure, space group I4/m (C54h) (a), and with the CsCl (B2)-type structure, space group P4/mmm (D14h) (b). The non-metal sublattice of titanium dicarbide with the B2 (CsCl)-type tetragonal structure [space group P4/mmm (D14h)] is built of C2 units. The unit cell of this structure comprises one Table 4. Crystal-chemical characteristics of possible (model) modifications of titanium dicarbides. Dicarbide structure Tetragonal CaC2 (C11a) Tetragonal (derived from CsCl-type structure) Cubic of the CaF2 type (C1) Monoclinic of the ThC2 type (seea) Cubic a The angle b in the unit cell is 103.9 8.b Ti CUnit cell parameters /nm Space group a b c atom position 0.3668 4hÜ I4=m ÖC5 0.2594 P4=mmm ÖD14hÜ 0.487 Fm3m ÖO5h Ü 0.510 2hÜ C2=c ÖC6 0.588 F43m ÖT2d Ü 100 80 60 40 20 2y /deg Figure 13. X-Ray diffraction pattern of tetragonal TiC2 with the CaC2 (C11a)-type structure, space group I4/m (C54h). Calculated for CuKa1,2 radiation. formula unit TiC2 (see Fig. 12). The atomic coordinates in the unit cell are listed in Table 4. The running coordinate, z/c, of carbon atoms remains unknown so far and can vary between 0 and 0.5. The C7C separation in the C2 unit is 0.124 nm; in titanium dicarbide this distance should be somewhat longer due to tita- nium ± carbon interatomic interactions.This suggests that the most probable z/c values lie between 0.25 and 0.31. Figure 14 presents the X-ray diffraction patterns of tetragonal titanium dicarbide TiC2 (space group P4/mmm) calculated for several different positions of the carbon atom (z=0.10, 0.25, 0.30 and 0.40). Position of the carbon atom has a strong effect on the intensity ratio of diffraction reflections, namely, an increase in z from 0.1 to 0.4 causes a decrease in the intensities of the (001) and (111) reflections and an increase in the intensity of the (101) reflection. Similarly to cubic monocarbide TiCy, in titanium dicarbide with the CaF2 (C1)-type cubic structure [space group Fm3m (O5h )] (Fig.15) titanium atoms form a face-centred metal sublattice. However, carbon atoms occupy tetrahedral rather than octahe- dral (as in TiCy) interstices. Naturally, the lattice constant of titanium dicarbide must be larger than that of titanium mono- carbide, since only in this case is the size of the tetrahedral interstice sufficient to contain a carbon atom. Analysis of the crystal structure of cubic TiC2 showed that it is a typical representative of interstitial phases. In the nondistorted crystal Atomic coordinates in the unit cell z/c z/a y/a x/a 0.617 2 (a) 0 0 7 Intensity (arb.u.) 101 002 110 112 200 103 211 202 004 114 220 213 222 301204 310 110 449 7 0.337 7 TiC 4(e) 0 0 7 TiC 2(h) 1 (a) 0 0 7 1/2 1/2 C 8(c) 00.427 0 7 7 4 (a) 0 0 0 7 1/4 0.2 0.13 4 (a) 0 0 0 7 1/2 3/4312 231 224215 303 7 7 Ti 1/4 1/4 116 314 400 233 7 Ti 0.510 0.310 0.25 0.05 00.30 4 (e) C 8(f ) 77 7 7 Ti C1 C2 1/2 3/4 1/2 3/4 4 (b) 4 (d ) 77Figure 15.Unit cell of cubic TiC2 with the CaF2 (C1)-type structure, space group Fm3m (O5h ). Intensity (arb.u.) 1234 Figure 14. Effect of the position of carbon atom (z) in the unit cell of tetragonal TiC2 with the CsCl-type structure (space group P4/mmm, a=0.2594, c=0.337 nm) on the intensity ratio of diffraction reflections (calculated for CuKa1,2 radiation).z=0.40 (1); 0.30 (2); 0.25 (3) and 0.10 (4). Intensity (arb.u.) 001 100 111 200 110 002 220 450 111 20 40 lattice of this compound each carbon atom has a regular tetrahe- dral environment (four titanium atoms) while each titanium atom is surrounded by eight carbon atoms arranged at the cube vertices. The unit cell of cubic titanium dicarbide comprises four TiC2 formula units. The calculated X-ray diffraction pattern of this dicarbide is shown in Fig. 16. The electronic energy spectrum of hypothetical titanium dicarbide TiC2 with the ThC2-type monoclinic structure [space group C2/c (C62h)] was calculated.112, 113 The lattice constants of this compound are a=c= 0.510 and b=0.310 nm.112 The unit 40 Figure 16.X-Ray diffraction pattern of cubic TiC2 with the CaF2 (C1)- type structure, space group Fm3m (O5h ). Calculated for CuKa1,2 radiation. 101 20 80 311 102 222 200 112 201 400 210 003 211 103 202 331 60 60 100 2y /deg 420 100 2y /deg Ti C 422 80 In polymorphous modifications of TiC2 with the CaC2-, CsCl-, CaF2- and ThC2-type structures the Fermi level lies in the region of delocalised d-states of titanium;112, 113 therefore, all the four TiC2 modifications behave as metals. The most stable is dicarbide with the ThC2-type monoclinic structure. In dicarbide with the CaC2-type structure the C7C bonds in C2 units are the strongest, whereas the Ti7C bonds are poorly populated and, hence, weaker. Calculations for cubic TiC2 with the CaF2-type structure showed the absence of C7C bonds.According to calculations of the electronic structure of titanium dicarbides with the CaC2- and ThC2-type structures, the density of electron states at the Fermi level depends on the orientation of C2 units in the crystal lattice. This implies the possibility of modifying the 333 Figure 18. X-Ray diffraction pattern of monoclinic TiC2 with the ThC2- type structure, space group C2/c (C62h). Calculated for CuKa1,2 radiation. 113 212 220 c Intensity (arb.u.) a Figure 17. Unit cell of monoclinic TiC2 with the ThC2-type structure, space group C2/c (C62h). The two Ti atoms and one C atom arranged beyond the unit cell are also shown. cell of monoclinic titanium dicarbide constructed using the atomic coordinates for monoclinic ThC2 (see Ref.114) is shown in Fig. 17 (the atomic coordinates in the unit cell are listed in Table 4). The unit cell comprises four formula units TiC2. Distorted octahedra built of six titanium atoms contain two asymmetrically arranged carbon atoms (connected by a dashed line) separated by 0.251 nm. Atoms of the C2 units are shared by two neighbouring octahedra while the C7C bonds pass through their common edges; the C7C distance in the C2 units is 0.097 nm (in Fig. 17, the atoms of these units are shown connected by an arrow). The X-ray diffraction pattern calculated for polycrystal- line dicarbide TiC2 with this unit cell is shown in Fig. 18. 110 002 (111) 20 40 C b 113 022 400 114 222 222 80 111 202 112112 202 020 311(021) 310 312 60 223 A I Gusev Ti130 (204) 313 (404) 224(420) 132 (421) 513(315) 100 2y /degPhase equilibria, phases and chemical compounds in the Ti ±C system a b c The results reported by Perekrestov and Pavlov 116 and d e f C Ti Figure 19.Arrangement of C2 dimers in hypothetical titanium dicarbide with body-centred tetragonal structure of the CaC2 (YC2) type (a) and possible orientation disorder types of the carbon dimers (b ± f ).112 electronic structure and properties of titanium dicarbide by means of orientation ordering of the C2 units. Some of the possible types of orientation ordering of C2 units are shown in Fig.19.112, 113 The simplest form has the electronic spectrum of dicarbide TiC2 with ordered CaC2-type structure (see Fig. 19 a). Changes in the orientation of a fraction of C2 units are accompanied by complication of the DOS distribution profile and by the appearance of new peaks. The most probable orientation type of C2 units is to some extent similar to the ThC2-type structure (Fig. 19 d).112 The least probable is the orientation of a fraction of C2 units along the edges of the unit cell (Fig. 19 c). Experimental studies concerning the search for titanium dicarbide are scarce. A carbon ± titanium compound with the concentration of bound carbon of greater than 50 at.% was first obtained in the high-temperature interaction of ethylene with titanium tetrachloride (gas-phase synthesis).115 The carbide thus obtained had a CsCl-type cubic lattice with a lattice constant of 0.313 nm and was thought to have a TiC2 composition.However, attempts at reproducing these results 115 failed. A recent microdiffraction study 116 on films prepared by ion sputtering of carbon and titanium followed by vapour deposition revealed the formation of a compound with a body-centred lattice and composition TiC2. The deposition temperature was varied from 330 to 770 K; thin films were deposited on fresh KCl cleavages, coatings 10 to 18 mm thick were obtained on glassy substrates. Films condensed at 330 K and annealed at 700 K for 3 h contained only titanium monocarbide. As the deposition (condensation) temperature increased from 330 to 620 K, the phase composition of the films remained unchanged; however, the fcc lattice constant of titanium carbide in these films reached *0.455 nm at a carbon content of greater than 65 at.%.A gas- phase electron diffraction study also revealed a decrease in the intensity of reflections accompanied by an increase in the diffuse background at a carbon content of greater than 60 at.%. The electron diffraction patterns of the films containing *65 at.% of carbon and deposited at 670 K exhibited three diffuse lines assigned to the body-centred lattice with a lattice constant of *0.3 nm. A homogeneous film was prepared by condensation at 770 K; the partial pressure of reactants (gaseous carbon and titanium) was 861078 Pa, the film growth rate was *0.1 nm s71 and the content of carbon in the vapours was *66 at.%.The results obtained allowed one to suggest the formation of TiC2 films, a body-centred lattice of this compound with a lattice constant of 0.294 nm and the structure of titanium dicarbide. especially their interpretation provoke some objections. First of all, no direct determination of film composition was made (at best, the content of carbon in the gas flow was monitored indirectly). The proposed crystal lattice structure of TiC2 is not body-centred. This is a cubic lattice [space group F43m (T 2d )] with a face-centred metal (titanium) sublattice and a diamond-like non-metal (car- bon) sublattice [A4 type, i.e., two face-centred lattices displaced by (1/4 1/4 1/4) with respect to each other].In the unit cell of this structure (Fig. 20) half of the carbon atoms occupy the octahedral interstices of the titanium sublattice, while the rest of the carbon atoms occupy half of the tetrahedral interstices of the titanium sublattice, so each of them is tetrahedrally surrounded by four titanium atoms and by four carbon atoms. This unit cell comprises four formula units TiC2 and the lattice constant is 0.588 nm. The atomic coordinates in the cubic unit cell of TiC2 (space group F43m) are listed in Table 4 and the calculated X-ray diffraction pattern is presented in Fig. 21. Figure 20. Unit cell of cubic TiC2, space group F43m (T2d ), a=0.588 nm. 60 40 20 Figure 21.X-Ray diffraction pattern of polycrystalline cubic TiC2, space group F43m (T 2d ), a=0.588 nm. Calculated for CuKa1,2 radiation. In addition to clustersM8C12 (see Section III.1), laser-assisted plasmachemical synthesis led to the formation of cluster-type nanoparticlesMmCn (M=Ti, Zr, Hf, V; m: n&1 : 2).28, 30, 31 For M=Ti, molecular clusters of composition Ti13C22 were usually formed (i.e., the titanium : carbon ratio was close to the corre- sponding parameter in ideal TiC2). Preferable formation of nano- clusters with a high content of carbon (M :C&1 : 2) occurred at a high concentration of hydrocarbon in the gas phase and at an increased laser power, which led to an increase in the hydro- genation degree of the hydrocarbon and, correspondingly, to an increase in the content of carbon in the plasma.Thus, analysis of the available data suggests that titanium dicarbide TiC2 can belong to five structural types. These are the ThC2-type monoclinic structure [space group C2/c (C62h)]; CaC2 (C11a)- and CsCl (B2)-type tetragonal structures with I4/m (C5 and P4/mmm (D14h) space groups, respectively; CaF2 (C1)-type cubic structure [space group Fm3m (O5 Intensity (arb.u.) 111 220 220 311 222 400 331 C 80 420 422 451 Ti 100 2y /deg 4h) h)] and a cubic structure 333 440 531 442 620 533452 with a diamond-like carbon sublattice and the F43m (T2 d ) space group. 3. Endohedral fullerene Ti@C28 and rhombohedral titanium carbide The discovery of Cn fullerenes (n=60 ± 90) (see Refs 70 ± 73, 117, 118) and further studies of these compounds showed that the Cn clusters containing less than 60 carbon atoms are of low stability.Fullerenes with a small number of atoms (e.g., C28 fullerene) can be stabilised by preparing endohedral complexesM@C28 in which the dopant atom is placed inside the carbon sphere. A Ti@C28 endohedral complex was also synthesised.119 The possibility of stabilising unstable C28 fullerene by inter- calating it with both non-metal atoms of 2p-elements (B, C,Nand O) and metals (3d-elements such as Sc, Ti, V, Cr, Fe and Cu) was studied theoretically.112, 120 When estimating the formation prob- ability of endohedral M@C28 complexes with 3d-metals, one can take into account the geometric, chemical and kinetic factors.In the case of C28 fullerene the maximum radius of a metal ion that can be placed inside the carbon sphere lies between 0.09 and 0.10 nm (see Ref. 121); all 3d-metals fulfil this geometric criterion. The chemical factor is favourable if electron density transfer as a result of intercalation favours an increase in the bonding character of the MOs. The kinetic factor includes the formation mechanism of endohedral complexes. They are prepared by rolling a graphene sheet around the metal atom adsorbed on the graphite surface. This can occur if the interaction between the metal and the graphite surface causes strengthening of the M7C bonds and weakening of the C7C bonds (especially of the interlayer bonds). Recent studies 112, 120 showed that intercalation of metal atoms is accompanied by electron density transfer from these atoms to the atoms of the fullerene cage, by changes in the populations of the overlapping C-AOs, by the formation of theM7C bond and by a change in the pattern of the electronic energy spectrum.The endohedral complex Ti@C28 with the Ti atom at the centre of the C28 polyhedron was shown to be the most stable (with inclusion of the chemical and kinetic factors).120 Probably, this complex can be considered as a possible molecular cluster-like compound that can be formed in the Ti7C system. The results of a recent study 122 on the electronic structure of hypothetical metastable polymorphous modifications of TiC1.0 prompted us to mention a study by Dubrovinskaia et al.122 who experimentally found a rhombohedral modification of titanium carbide.The sample of cubic titanium carbide (no composition was reported) with a lattice constant of 0.4327 nm was subjected to quasi-hydrostatic pressure (*10 GPa) at 300 K. This resulted in broadening of the diffraction reflections of the B1-type cubic structure and in splitting of the (111)B1 reflection into two lines corresponding to the rhombohedral structure at a pressure of 18 and 38 GPa. The authors believed that rhombohedral distortion of the cubic unit cell is due to extension of its [111] diagonal, which led to splitting of the (111)B1 reflection into the (003) and (101) rhombohedral lines and to splitting of the (200)B1 reflection into the (104) and (110) lines.Rhombohedral titanium carbide pre- pared at a pressure of 38 GPa has a unit cell with the lattice constants a=0.29442 and c=0.73354 nm (in a hexagonal set- ting). IV. Ordered phases of nonstoichiometric titanium carbide TiCy The ordering of interstitial atoms was experimentally found for most of the nonstoichiometric transition metal carbides, nitrides and oxides. In disordered nonstoichiometric compounds the interstitial atoms and structural vacancies form a substitution solution in the non-metal sublattice, all sites of the non-metal sublattice being crystallographically equivalent. Ordering causes redistribution of the interstitial atoms and vacancies over sites of the non-metal sublattice, thus making the lattice sites inequiva- lent.The crystal lattice is divided into at least two sublattices A I Gusev differing in site occupation probabilities by the atoms of the same sort. Ordering is accompanied by symmetry reduction of the crystal, since some of the symmetry elements of a disordered non-metal sublattice that match the occupied and vacant sites do not belong to the group of symmetry elements of the ordered crystal in which these sites become crystallographically inequiva- lent.From the aforesaid it follows that the nonstoichiometry responsible for the fact that the number of interstitial sites is larger than that of atoms occupying these sites is a prerequisite to the ordering in defect carbides.Description of different ordered phases of nonstoichiometric carbides, nitrides, oxides and their solid solutions can be found in reviews 1, 22, 123 ± 127 and monographs.2, 3, 14, 17, 27, 70 Compared to other carbides (except for vanadium carbide VCy), the ordering in nonstoichiometric titanium carbide TiCy with the B1-type cubic structure has been studied in most detail; however, the mechanism of this process is still to be clarified. A great body of experimental information on the chemical compo- sition and structure of ordered phases of titanium carbide TiCy published earlier has not been confirmed later. The main charac- teristics of both experimentally observed and theoretically calcu- lated ordered phases of nonstoichiometric titanium carbide are listed in Table 5.As mentioned above, disordered titanium carbide TiCy (TiCy&17y) is characterised by an extremely broad homogeneity region from TiC0.48 to TiC1.00 (see Refs 2, 3, 14). Within this region, carbon atoms and structural vacancies & form a substi- tution solution in the non-metal sublattice. Depending on the composition and the synthesis and heat treatment conditions, TiCy can exist in a disordered or ordered state. The disordered state is in thermodynamic equilibrium at T>1100 K, while the ordered state is in thermodynamic equilibrium at T<1000 K. Due to the low diffusive mobility of the atoms, the disordered state of nonstoichiometric titanium carbide is retained upon quenching from T>1100K to rather low temperatures (*300 K) and exists as a metastable state at T<1000 K.Order-parameter functional (OPF) calculations1 ± 3, 18, 20, 22 showed that monocarbide TiCy can form three ordered phases of compositions Ti2C, Ti3C2 and Ti6C5. According to the inverse Monte Carlo calculations performed for a narrower composition range from TiC0.55 to TiC0.70 (see Ref. 12), the ordered phases Ti2C and Ti3C2 are in thermodynamic equilibrium at T<950 K , which is in agreement with the results of the above-mentioned calculations.1 ± 3, 18, 20, 22 Experimental studies of carbide TiCy in the composition range 0.54y<0.7 revealed the existence of ordered phases of the Ti2C type with cubic (space group Fd 3m) and trigonal (space group R3m or P3121, the former being more probable) symmetry (see Table 5), orthorhombic ordered phase Ti3C2 (space group C2221) and hexagonal ordered phase Ti6C5 (see Ref.130). The trigonal ordered phase Ti8C5 (space group R3m) was reported erroneously. Virtually no studies of the order- ing of TiCy (y>0.7) were performed until the mid-1990s. Let us consider in more detail which ordered phases of the compositions Ti2C and Ti3C2 can be formed in nonstoichiometric carbide TiCy. According to different studies of TiCy, an ordered phase of composition Ti2C with cubic (space group Fd 3m) (see Refs 9, 46 ± 48, 50) or trigonal (space group R3m) (see Refs 12, 49) symmetry is formed in the range 0.54y40.65 at T<1100 K.{ Cubic Ti2C phase is thought to be metastable.Initially,49, 129 the cubic phase was assumed to have a higher Ttrans { Information on the existence of the trigonal (space group P3121) Ti2C (Ti6C3+x) phase is erroneous: formation of a trigonal (space group P3121) pÅÅÅ M2C-type phase with the unit cell parameters a=b=aB1/ 2 (a={1/2 pÅÅÅ 1/2 0}B1) and c=3 26aB1 (c={222}B1) based on the basic B1-type structure is impossible, since in this case the sites of the metal and non- metal sublattices partially match one another.Phase equilibria, phases and chemical compounds in the Ti ±C system Table 5. Characteristics of ordered phases of nonstoichiometric titanium carbide TiCy with cubic B1-type structure. Com- Existence position region Ti2C Ti2C Ti3C2 Ti2C Ti2C Ti3C2 Ti6C5 Ti2C Ti3C2 Ti6C5 aTtrans is the disorder ± order transformation temperature.value than the trigonal phase, thus being an intermediate ordered high-temperature phase with respect to the disordered phase and low-temperature trigonal phase of composition Ti2C. Tashmetov et al.131 reported a consecutive phase transformation disordered phase (space group Fm3m) However, de Novion et al.12 suggested that in TiCy (y50.58) the cubic phase Ti2C can exist as a metastable phase for which Ttrans is *10 K lower than for the trigonal ordered phase. Analysis of the results of experimental structural stud- ies 9, 12, 46 ± 50, 128, 131 showed that cubic superstructure Ti2C is usually found in annealed TiCy samples with y<0.55 ± 0.56, whereas the samples with 0.584y40.65 are characterised by trigonal ordering.It should be noted that the X-ray diffraction patterns of both cubic and trigonal ordered phases of composition Ti2C exhibit identical sets of superstructure reflections and can be distinguished only given trigonal distortions in the phase with the space group R3m and taking into account the directions of static atomic displacements.126 Probably, this is the reason for attribut- ing the observed superstructure reflections in annealed carbide TiCy (y50.59) to the cubic phase Ti2C in the studies 46 ± 48, 50 concerned with the cubic model of ordering.9 Other authors 12, 49, 128, 129, 131 showed that trigonal phase Ti2C is the major ordered phase of TiCy (y50.6).The assumption of the existence of ordered phase Ti3C2 in the composition range TiC0.60 ± TiC0.70 was confirmed by some experimental studies. An elastic neutron scattering study 129 of annealed single crystal TiC0.61 revealed the presence of the (2/3 2/ 3 0) superstructure reflections. The X-ray diffraction pattern of annealed carbide TiC0.70 exhibited weak superstructure reflec- tions with the reduced diffraction vector j q j&2.03 characteristic of the orthorhombic phase Ti3C2 (space group C2221).132 Diffuse neutron scattering maxima corresponding to the (2/3 2/3 0) reflections and due to short-range order in TiC0.76 were detected.133 Estimation of the short-range order parameters in a Unit cell parameters /nm Space group Symmetry (lattice type) cubic TiC0.52 ± TiC0.71 Fd 3m a=2a0 trigonal R3m a=b=c=(a0 TiC0.58 ± TiC0.63 orthorhombic C2221 TiC0.64 ± TiC0.68 pÅÅÅ c=2a0 pÅÅÅ a=a0 2, b=3a0 2, cubic TiC0.49 ± TiC0.54 Fd 3m a=2a0 trigonal R3m a=b=c=(a0 TiC0.55 ± TiC0.59 orthorhombic a=a0 C2221 TiC0.63 ± TiC0.67 pÅÅÅ c=2a0 pÅÅÅ, 2 2, b=3a0 2 hexagonal a=b=(a0 7 7TiC0.47 ± TiC0.54 pÅÅÅ)/2=0.529, 130 pÅÅÅ)/3=0.996 6 c=(4a0 3 7 7 7 TiC0.62 ± TiC0.707 7 7 TiC0.80 ± TiC0.907 7 7 1070K 1050K cubic ordered phase (space group Fd 3m) trigonal ordered phase (space group R3m).453 Ref. Note a 9, 48 12, 128 pÅÅÅ)/2 6 12, 129 21 21 pÅÅÅ)/2 6 carbon sublattice is partially ordered; Ttrans&950 ± 1030 K calculated; Ttrans<900 K; the homogeneity region corresponds to 750 K calculated; Ttrans<720 K; the homogeneity region corresponds to 650 K Ttrans<1000 K; the homogeneity region corresponds to*600 K Ttrans<1010 K; the homogeneity region corresponds to*600 K 1 Ttrans<990 K; the homogeneity region corresponds to*600 K phase was found in thin films 18, 20, 22 calculated; Ttrans<950 K; the homogeneity region corresponds to 600 K 18, 20, 22 calculated; Ttrans<930 K; the homogeneity region corresponds to 600 K 18, 20, 22 calculated; Ttrans<920 K; the homogeneity region corresponds to 600 K single crystal of composition TiC0.64 12 using the diffuse neutron scattering data showed that agreement between the calculated and experimental data is the best when annealed carbide TiC0.64 contained a mixture of two ordered phases Ti2C and Ti3C2.The existence of a trigonal phase Ti3C2 was also confirmed by the OPF18, 20, 22 and Monte Carlo 10, 12 calculations. A theoretical study18 also suggested the formation of an ordered phase Ti6C5 in carbide TiCy (0.78<y<0.88). The calculated disorder ± order transformation temperatures for the Ti3C2 and Ti6C5 phases did not exceed 950 K.18, 20, 22 The structure of the ordered phases of nonstoichiometric carbide TiCy has been studied in detail.21 Samples of nonstoichiometric titanium carbide TiCy with different carbon content (0.524y40.98) were synthesised by hot pressing the mixtures of TiC0.98 and metallic titanium powders (T=1773 ± 2173 K, p=20 ± 30 MPa) in highly pure argon atmosphere for 30 min or by solid-phase vacuum sintering the TiC0.94, carbon black and metallic titanium powders (T= 2000 K, p=0.001 Pa) for 6 h.The ordered state of the carbides obtained was achieved by annealing the samples with gradual lowering of the temperature from 1070 down to 770 K for 340 h. Annealing resulted in the appearance of superstructure reflections in the X-ray diffraction patterns of carbides TiC0.50 , TiC0.52 , TiC0.54 , TiC0.55 , TiC0.58 and TiC0.62 . The X-ray diffraction patterns of the annealed carbides TiC0.50 , TiC0.52 and TiC0.54 exhibited identical sets of super- structure reflections. The first superstructure reflection with the reduced wave vector j q j=(2aB1siny)/l& 0.870 in the region 2y&17.8 ± 17.9 8 corresponds to the vector {1/2 1/2 1/2} of length j q j&0.866.The next three superstructure reflections at 2y&34.5, 45.9 and 55.4 8 correspond to the vectors {3/2 1/2 1/2}, {3/2 3/2 1/2} and {3/2 3/2 3/2}. The {3/2 3/2 1/2} super- structure reflection is very weak. Taking into account the posi- tions of the observed superstructure reflections and the absence of trigonal splitting of the (331)B1 , (420)B1 and (422)B1 lines, one can suggest that annealing led to the formation of the ordered cubic phase Ti2C (space group Fd 3m) in carbides TiC0.50 , TiC0.52 and TiC0.54. The channel of disorder ± order structural phase trans- formation TiCy (space group Fm3m) Ti2 C (space group454 a3 [100]B1 a1 Figure 22.Cubic unit cell (space group Fd3m) of the Ti2C superstructure in the lattice with the B1-type structure (Ti atoms are not shown).17, 27, 126 9 Fd 3m) includes all four arms of the {k9} Lifshitz star: kÖ1Ü =(b1+b2+b3)/2, kÖ2Ü=b1/2, kÖ3Ü=b2/2 and kÖ4Ü=b3/2.} The unit cell of cubic superstructureM2C (space group Fd 3m) is shown in Fig. 22. The translation vectors of and the atomic and vacancy coordinates in the unit cell are listed in Table 6. In completely ordered, ideal cubic superstructure M2C all metal atoms occupy sites of the M43 type (the subscript and superscript denote the number of vacancies in the first and second coordina- tion sphere of the metal atom, respectively).The non-metal sublattice of this superstructure is characterised by alternation of the (111)B1 atomic planes in which 1/3 and 2/3 of the total number of sites are vacant. Table 6. Characteristics of cubic [space group Fd 3m ÖO7hÜä M2C super- structure: a1=h200iB1 , a2=h020iB1 , a3=h002iB1.16, 27, 126 Position Atom 16 (c) 16 (d ) 32 (e) C1 (vacancy) C2 M1 The X-ray diffraction pattern of annealed carbide TiC0.58 exhibited the {1/2 1/2 1/2}, {3/2 1/2 1/2} and {3/2 3/2 3/2} super- structure reflections at 2y&17.9 8, 34.4 8 and 55.4 8 (see Ref. 21). No {3/2 3/2 1/2} reflection was observed. An important distinc- tion of the X-ray diffraction pattern of annealed carbide TiC0.58 from those of annealed carbides of compositions TiC0.50, TiC0.52 and TiC0.54 is the presence of trigonal splitting of the (220)B1, (311)B1, (222)B1, (331)B1, (420)B1 and (422)B1 lines.This is an indication of the trigonal ordered phase Ti2C (space group R3m) formed in carbide TiC0.58 upon annealing. The unit cell of the trigonal phase is shown in Fig. 23. Its translation vectors and the atomic and vacancy coordinates in ideal trigonal structureM2C are listed in Table 7. The TiCy (space groupFm3m) Ti2 C(space groupR3m) disorder ± order phase transition channel involving the formation of this superstructure includes one arm, kÖ3Ü 9 =b2/2, of the {k9} Lifshitz star. } For a detailed description of all stars {ks} of the wave vectors of the first Brillouin zone and their arms for a crystal with a fcc lattice, see Refs 2, 3 and 70.[001]B1 Cvacancy [010]B1 a2 9 9 9 Atomic coordinates in ideal ordered structure z/a3 y/a2 x/a1 1/8 5/8 3/8 1/8 5/8 3/8 1/8 5/8 3/8 A I Gusev [001]B1 CTi vacancy c a b [010]B1 [100]B1 Figure 23. Trigonal unit cell (space groupR3m) of the Ti2Csuperstructure in the lattice with the B1-type structure.17, 27, 126 Table 7. Characteristics of trigonal space group [R3m ÖD53dÜ] M2C super- structure: atr=21 h121iB1, btr=12 h211iB1, ctr=12 h112iB1.16, 27, 126 Position Atom Atomic coordinates in ideal ordered structure z/ctr y/btr x/atr 1 (a) 0 0 0 1/2 1/4 C1 (vacancy) C2 M1 1/2 1/4 1/2 1/4 1 (b) 2 (c) 4 4 4 3 3 3 3 The non-metal sublattice of completely ordered trigonal structure M2C (space group R3m) is characterised by alternation of the (111)B1 atomic planes with completely occupied (by interstitial atoms) and completely vacant sites along the [111]B1 direction. This superstructure must have a highly distorted metal sublattice. In completely ordered, ideal trigonal superstructure M2C all metal atoms occupy identical sites of the M43 type with three vacancies in the first and four vacancies in the second coordination sphere.The X-ray diffraction pattern of annealed carbide TiC0.62 exhibited both the {1/2 1/2 1/2} (2y&18.08) and {3/2 3/2 3/2} (2y&55.28) superstructure reflections from the trigonal ordered phase Ti2C (space group R3m) and some other superstructure reflections that were absent in the X-ray diffraction patterns of the carbides of compositions TiC0.50 , TiC0.52 , TiC0.54 , TiC0.55 and TiC0.58 (see Ref.21). These are the reflections in the ranges 2y&30.6 ± 30.7 8, 41.2 8, 42.6 8 and 55.4 ± 55.5 8 with reduced wave vectors of lengths j q j&1.488, 1.970, 2.038 and 2.607, respectively. The first two reflections are associated with the kÖ1Ü={2/3 2/3 0} and kÖ2Ü=7kÖ1Ü arms of the {k4} star, while the other two reflections are associated with the kÖ3Ü={1/3 72/3 71/2}, kÖ4Ü=7kÖ3Ü 3 , k3 Ö5Ü ={71/3 2/3 71/2} and kÖ6Ü=7kÖ5Ü arms of the {k3} star. This set of superstructure reflections can correspond only to the orthorhombic ordered phase Ti3C2 (space group C2221).The transition channel involv- ing the formation of this phase includes two arms of the {k4} star and four arms of the {k3} star (see Ref. 126). Characteristic of this superstructure are the {1/3 72/3 71/2} and {2/3 2/3 0} reflec- tions which should be observed in the regions 2y&18.4 8 and 19.4 8, respectively. The experimental X-ray diffraction pattern in this interval of angles exhibited a slightly `swollen' slope of the {1/2 1/2 1/2} superstructure reflection of the trigonal phase Ti2C.21 The orthorhombic unit cell of the M3C2 superstructure is shown in Fig. 24. Its translation vectors and the atomic andPhase equilibria, phases and chemical compounds in the Ti ±C system c [001]B1 CTi vacancy [010]B1 a [100]B1 b Figure 24.Orthorhombic unit cell (space group C2221) of the Ti3C2 superstructure in the lattice with the B1-type structure.2, 3, 17, 27, 126 2. vacancy coordinates in this superstructure are listed in Table 8. The existence of orthorhombic ordered phaseM3C2 (space group C2221) of titanium carbide TiC0.64 was suggested12 based on the results of Monte Carlo simulations. In completely ordered ortho- rhombic structureM3C2 two thirds of all metal atoms occupy the sites of the typeM32 while the rest metal atoms occupy the sites of the typeM2 Metallographic tests of annealed carbides TiC0.50 and TiC0.55 in reflected polarised white light showed that the former exhibited neither grain interference from the disordered phase nor surface precipitate interference from the ordered phase.21 This can occur if both phases belong to the cubic crystal system.Particular grains of annealed carbide TiC0.55 were coated with plate-like and prismatic precipitates of the ordered phase. The symmetry of these precip- itates was different from that of the disordered phase while their shape was typical of the crystals that belong to the trigonal crystal system and, unlike the crystals of the disordered cubic phase, exhibit an interference pattern (i.e., are anisotropic). Therefore, the results of diffraction studies of the ordered phases of non- Table 8. Characteristics of orthorhombic [space group C2221 (D52 )] M3C2 superstructure: arh=h110iB1 , brh=h330iB1 , crh=h002iB1.17, 27, 126 Position Atom Atomic coordinates in ideal ordered structure z/crh y/brh x/arh 001/4 000 1/4 1/4 01/4 1/4 1/4 1/4 00 4 (b) 4 (b) 8 (c) 4 (b) 4 (b) 4 (b) 4 (b) 8 (c) 8 (c) 01/4 1/4 1/6 1/3 5/12 01/2 2/3 5/6 1/12 1/4 4 (a) 0 0 0 01/12 1/4 5/12 1/6 1/3 C1 (vacancy) C2 (vacancy) C3 (vacancy) C4 C5 C6 C7 C8 C9 M1 M2 M3 M4 M5 M6 M7 4 (a) 8 (c) 8 (c) 8 (c) 8 (c) 8 (c) 01/4 1/4 1/4 1/2 1/2 1/2 1/4 1/4 1/4 00 311 222 (111)B1 400 (200)B1 104 331 015 a Intensity (arb.u.) 101 012 110 440 (220)B1 113 531 533 1010 116 622 (311)B1 444 (222)B1 2y /deg 0012 7222 024a 711 0111 1 024a2 731 119 455 stoichiometric carbide TiCy were confirmed by the results of metallographic tests.Thus, low-temperature annealing of carbide TiCy leads to the formation of some ordered phases. These are the cubic phase Ti2C (space group Fd 3m) in the composition range TiC0.50 ± TiC0.54, the trigonal phase Ti2C (space group R3m) for TiC0.55 ± TiC0.58 and the orthorhombic phase Ti3C2 (space group C2221) in the range TiC0.63 ± TiC0.68. Figure 25 presents the calculated X-ray diffraction patterns of the cubic and trigonal ordered phases Ti2C formed in carbides of composition TiC0.50 (aB1=0.4305 nm) and TiC0.55 (aB1=0.4310 nm), respectively. The X-ray diffraction patterns were calculated assuming the maximum degree of long-range order characterised by the parameter Z equal to 1.00 and 0.90 for the ordered carbides TiC0.50 and TiC0.55, respectively.The X-ray diffraction patterns of the cubic and trigonal ordered phases Ti2C exhibit identical sets of superstructure reflections; however, dis- tortion of the ideal lattice of the trigonal phase [slight extension along the (1 71 1)B1 axis] leads to splitting of the (311)B1, (222)B1, (420)B1 and (422)B1 reflections. Trigonal splitting of the (222)B1 reflection into two lines, viz., the (0 0 12) and (0 2 4) lines, is shown in the inset in Fig. 25 b as an example; the doublet a1,2 is due to the use of CuKa1,2 radiation. The superstructure reflections are of rather high intensity: for instance, the intensity of the first superstructure reflection is 30 times lower than that of the (111)B1 reflection and only 3 times lower than the intensity of the (222)B1 reflection.This makes the ordered phase Ti2C readily detectable. 60 40 20 Figure 25. X-Ray diffraction patterns of cubic (space group Fd3m, a=0.861 nm) (a) and trigonal (space group R3m, the lattice constants for hexagonal setting are a=b=0.3048 and c=1.4982 nm) (b) ordered Ti2C phases formed in titanium carbide of the compositions TiC0.50 (aB1=0.4305 nm) and TiC0.55 (aB1=0.4310 nm), respectively. The long-range order parameter, Z=Zmax, equals 1 (a) and 0.90 (b). The inset demonstrates trigonal splitting of the (222)B1 reflection into the (0 0 12) and (0 2 4) lines due to slight extension of the basic cubic lattice along the (1 71 1)B1 axis (lattice constant, c=1.4982 nm, is increased compared to the value c=1.4930 nm for the ideal trigonal cell).Calcu- lated for CuKa1,2 radiation. Intensity (arb.u.) 111 003 b 009 333 76.0 76.4 76.8111 040 022 151 220 113 202 (111)B1 171 260 (200)B1 311 044 173 351 282 313 371 260 (220)B1 175 260 (311)B1 260 (222)B1 375 456 60 40 20 2y /deg Figure 26. X-Ray diffraction pattern of the completely ordered (Z=1.00) orthorhombic Ti3C2 phase (space group C2221, a=0.6109, b=1.2328 and c=0.864 nm) formed in titanium carbide of composition TiC0.667 (aB1=0.432 nm). Calculated for CuKa1,2 radiation. The X-ray diffraction pattern of the ordered orthorhombic phase Ti3C2 (space group C2221) formed in the carbide of composition TiC0.667 is shown in Fig.26. A characteristic feature of the orthorhombic superstructure Ti3C2 is the presence of three superstructure reflections in the range 19 8<2y <318. A hexagonal ordered phase Ti6C5 with the packing defects characteristic of the 4H polytype was found in the Ti ±C thin films.130 The translation vectors of the unit cell are as follows: a={211}B1, b={121}B1 and c={111}B1. The films were pre- pared by electron beam sputtering of sintered carbide TiC0.9, by magnetron sputtering of titanium and carbon and by laser beam sputtering of titanium and graphite or carbide TiC0.9. The ordered phase Ti6C5 corresponds to disordered carbide TiC0.83 with a lattice constant, aB1, of 0.4326 nm; however, the lattice parame- ters of the hexagonal ordered phase 130 correspond to disordered carbide TiCy with the basis cubic lattice constant aB1=0.4319 nm characteristic of carbide TiC0.68 ± 0.70 (see Fig.6). The underesti- mated aB1 value can be due to oxygen impurity. Ordering of nonstoichiometric carbides MCy can cause the formation of monoclinic and trigonal superstructures M6C5 with C2/m (or C2) and P31 symmetry, respectively. The unit cell of monoclinic superstructure M6C5 (space group C2/m) is shown in Fig. 27 while its translation vectors and the atomic and vacancy coordinates are listed in Table 9. Translation of all superstructure [001]B1 c 123 b [010]B1 a [100]B1 Figure 27. Monoclinic (space group C2/m) unit cell of the M6C5 super- structure in the crystal lattice with the B1-type structure.Interstitial atom (1), metal atom (2) and vacancy (3).70, 126, 134 Intensity (arb.u.) A I Gusev Table 9. Characteristics of monoclinic [space group C2/m (C32h)] M6C5 superstructure: am=21 h112iB1 , bm=32 h110iB1 , cm=12 h112iB1.70, 126, 134 Position Atom Atomic coordinates in ideal ordered structure z/cm y/bm x/am 2 (a) 0 0 0 1/2 1/3 1/6 02/3 C1 (vacancy) C2 C3 C4 M1 M2 1/2 01/2 3/4 3/4 2 (d ) 4 (g) 4 (h) 4 (i ) 8 (j ) 0001/4 1/4 sites of the reciprocal lattice of the monoclinic ordered phase M6C5 (space group C2/m) showed that the first Brillouin zone of the disordered fcc lattice contains five inequivalent superstructure vectors belonging to three stars, {k9}, {k4} and {k3}.These superstructure vectors are included in the MCy (space group Fm3m) M6C5 (space group C2/m) phase transition channel involving the formation of the monoclinic ordered phaseM6C5. The non-metal sublattice of the monoclinic superstructure M6C5 (space group C2/m) is divided into four inequivalent sublattices differing in site occupation probabilities by carbon atoms. Because of this, the non-metal sublattice is characterised by alternation of two types of non-metal atomic planes along the [111]B1 direction. In the first type of atomic planes one third of all sites belongs to the first sublattice while the other sites belong to the third sublattice, each site of the first sublattice being sur- rounded by six sites of the third sublattice.In the second type of planes one third of all sites belongs to the second sublattice while the other sites belong to the fourth sublattice; each site of the second sublattice is surrounded by six sites of the fourth sublattice. The trigonal unit cell (space group P31) of superstructure M6C5 is shown in Fig. 28. Its translation vectors and the atomic and vacancy coordinates in ideal trigonal superstructureM6C5 are listed in Table 10. The disorder ± order phase transition channel involving the formation of the trigonal superstructure M6C5 includes thirteen arms of three stars, {k9}, {k4} and {k3}. The unit cell of monoclinic superstructureM6C5 (space group C2) is presented in Fig.29. The translation vectors of this unit cell and the atomic and vacancy coordinates in ideal monoclinic [001]B1 123 c b [010]B1 a [100]B1 Figure 28. Trigonal (space group P31) unit cell of theM6C5 superstructure in the crystal lattice with the B1-type structure. Interstitial atom (1), metal atom (2) and vacancy (3).70, 126, 134Phase equilibria, phases and chemical compounds in the Ti ±C system Table 10. Characteristics of trigonal [space group P31 (C23 )] M6C5 super- structure: atr=21 h211iB1, btr=12 h112iB1, ctr=2h111iB1.70, 126, 134 Position Atom 3 (a) 3 (a) 3 (a) 3 (a) 3 (a) 3 (a) 3 (a) 3 (a) 3 (a) C1 (vacancy) C2 C3 C4 C5 C6 M1 M2 M3 M4 3 (a) 3 (a) 3 (a) M5 M6 [001]B1 c a [100]B1 Figure 29.Monoclinic (space group C2) unit cell of the M6C5 super- structure in the crystal lattice with the B1-type structure. Interstitial atom (1), metal atom (2) and vacancy (3).70, 126, 134 superstructureM6C5 are given in Table 11. The transition channel involving the formation of this ordered phaseM6C5 includes nine superstructure vectors of the reciprocal lattice belonging to four stars, {k9}, {k4}, {k3} and {k0}. The non-metal sublattice of this superstructure is divided into five inequivalent sublattices and is characterised by alternation of two types of non-metal atomic planes along the [111]B1 direction. The first type of atomic plane contains an equal number of sites of the first, third and fifth sublattices, each site of a given sublattice being surrounded by the sites belonging to the other two sublattices. The second type of atomic plane is formed by the sites belonging to the second and fourth sublattices; here, each site of the fourth sublattice is surrounded by six sites of the second sublattice.In Fig. 30 we present the X-ray diffraction patterns of the ideal, completely ordered monoclinic phases Ti6C5 (space groups C2/m and C2) formed in titanium carbide TiC0.83 (aB1=0.4326 nm). Positions and intensities of superstructure reflections in the X-ray diffraction pattern of the ideal, completely ordered trigonal phase Ti6C5 (space group P31, a=b=0.5299, c=1.4987 nm) match those of the monoclinic phase Ti6C5 (space group C2/m), since the transition channels involving the forma- tion of both phases include the same superstructure vectors.For the real monoclinic and trigonal ordered phases Ti6C5, the intensities of superstructure reflections should be different due to Atomic coordinates in ideal ordered structure z/ctr y/btr x/atr 1/6 1/6 1/6 1/3 1/3 1/3 1/12 1/12 1/12 8/9 5/9 2/9 5/9 2/9 8/9 5/9 2/9 8/9 1/9 4/9 7/9 1/9 4/9 7/9 1/9 4/9 7/9 1/4 1/4 1/4 2/9 8/9 5/9 1/9 4/9 7/9 123 b [010]B1 Table 11. Characteristics of monoclinic [space group C2 (C32 )] M6C5 superstructure: am=21 h112iB1, bm=32 h110iB1, cm=h112iB1.70, 126, 134 Intensity (arb.u.) Intensity (arb.u.) 001 020 110 002 Position Atom 2 (a) 0 0 0 1/3 1/3 2/3 02/3 1/6 1/2 5/6 C1 (vacancy) C2 (vacancy) C3 C4 C5 C6 C7 C8 C9 2 (b) 2 (a) 2 (a) 2 (b) 2 (b) 4 (c) 4 (c) 4 (c) 4 (c) 4 (c) 4 (c) 4 (c) 4 (c) 4 (c) M1 M2 M3 M4 M5 M6 the difference between the directions and magnitudes of static atomic displacements.Analysis of the X-ray diffraction pattern taking into account these displacements allows unambiguous symmetry determination (monoclinic or trigonal) of the ordered phase Ti6C5. The intensities of superstructure reflections of the Ti6C5 phases are very low: for instance, the intensity of the first super- structure reflection is 230 ± 240 and 30 ± 32 times lower than those of the (111)B1 and (222)B1 reflections, respectively.Comparison of the X-ray diffraction patterns of completely ordered phases Ti2C 20 Figure 30. X-Ray diffraction patterns of completely ordered (Z=1.00), ideal monoclinic Ti6C5 phases (space group C2/m and C2) formed in titanium carbide of composition TiC0.83 (aB1=0.4326 nm). Calculated for CuKa1,2 radiation. Characteristics of the monoclinic Ti6C5 phase: space group C2/m, a=c=0.5299 nm, b=0.9177 nm, b=109.47 8 (a); space group C2, a=0.5299 nm, b=0.9177 nm, c=1.0597 nm, b=109.47 8 (b). 020 11-1 110 021 11-1 11-2 ab 111 021 022 11-3 111 112 023 Atomic coordinates in ideal ordered structure x/am 130 130 11-2 13-1 (111)B1 13-2 (111)B1 040 22-1 113 y/bm 00000000 1/6 1/2 5/6 01/3 2/3 1/4 1/4 1/4 1/4 1/4 1/4 30 40 22-1 220 131 (200)B1 220 132 (200)B1 1/2 001/2 1/2 1/4 1/4 1/42y /deg 22-3 041 115 201 221 22-2 22-4 457 z/cm 1/8 1/8 1/8 3/8 3/8 3/8 043 112 221 025458 and Ti6C5 shows that the intensity of the first superstructure reflection of the Ti2C phase is nearly an order of magnitude higher than that of the Ti6C5 phase.For all completely ordered phasesM6C5 the relative stoichio- metric concentration of interstitial atoms (yst) equals 5/6. Consid- eration of the non-metal sublattices of these superstructures reveals that the [111]B1 direction is characterised by alternation of the (111)B1 atomic planes with completely occupied (by interstitial atoms) sites and the defect (111)B1 planes in which one third of sites is vacant (see Figs 27 ± 29). Each vacancy is surrounded by six interstitial atoms that form a regular hexagon. Peculiarities of completely ordered carbide structures M6C5 have been studied in detail.134 Completely ordered monoclinic and trigonal carbide struc- turesM6C5 with C2/m(or C2) and P31 symmetry, respectively, are characterised by the identical environment of the metal atoms by the sites of the non-metal sublattice that form the first and second coordination spheres. Indeed, there are only two types of site for the metal atoms in these structures, viz.,M11 (metal atom with one vacancy in the first and one vacancy in the second coordination sphere) andM21 (metal atom with one vacancy in the first and two vacancies in the second coordination sphere).In ideal, completely ordered structures M6C5 (space groups C2/m, P31 and C2) two thirds of all metal atoms occupy the sites of the typeM11 while the other metal atoms occupy the sites of the typeM21 . V. Phase diagrams of the Ti ±Csystem constructed with allowance for the ordering of TiCy The experimental,14, 17, 70, 126, 127 results of theoreti- cal 1 ± 3,12, 22, 135 ± 137 and numerous structural studies of nonstoi- chiometric carbides MCy with the B1-type basis structure including titanium carbide TiCy revealed the possibility of for- mation of superstructures of the types M2C, M3C2 and M6C5.Particular types of superstructures formed during the ordering of a given carbide MCy and the corresponding concentration ranges are determined by the positions of the upper and lower boundaries of the homogeneity region of the disordered carbide and by the equilibrium conditions at these boundaries. The most efficient method for calculating the low-temperature phase equilibria associated with the ordering of nonstoichiometric compounds is the above-mentioned OPF approach. Calculations of the phase diagram of an ordering system by this method require only knowledge of the F0(T ), F1(T ) and F2(T ) parameters of the free energy F(y,0,T ) of disordered nonstoichiometric carbide MCy . These parameters for TiCy are listed in Section II.2.According to the results of thermodynamic calculations 8, 60, 65 and crystal structure analysis,4 theM2X3,M4X3,M6X5 andM8X7 superstructures can be formed by the first-order phase transition mechanism, whereas theM2Xsuperstructure can be formed by the second-order phase transition mechanism. Transition between two ordered phases occurs as a first-order phase transition. 1. Calculated phase diagrams of the Ti ±C system Let us consider the phase diagram of the Ti ±C system obtained from the OPF calculations with allowance for the ordering of nonstoichiometric titanium carbide TiCy.1, 20 There is only one compound capable of ordering in the Ti ±C system, cubic carbide TiCy. At the lower boundary of the homogeneity region, titanium carbide is in equilibrium with metallic a-Ti (at T<1150 K) or b-Ti (at T>1150 K).According to calculations,18 at 1000<T<1900 K the carbide that exists near the lower boundary of the homogeneity region of the disordered phase TiCy has composition TiC0.32 ± TiC0.37. The exact position of the lower boundary of the homogeneity region has not experimentally been determined so far; different estimates gave TiC0.48 (see Ref. 8) or TiC0.47 (see Ref. 41) at 1900 K. According to calculations,44 at 1900 K near the lower boundary of the homogeneity region the carbide has a TiC0.52 composition. Calculations 8, 41, 44 revealed that as the temperature is lowered, A I Gusev the lower boundary of the homogeneity region of disordered TiCy is shifted towards the region with higher y at a rate that is somewhat higher than that predicted by calculations.18 The low-temperature portion of the phase diagram of the Ti ±C system, corresponding to the ordering of carbide TiCy, was calculated by the OPF method.18, 20 The broad homogeneity region of TiCy allows the formation of phases of the types Ti2C, Ti3C2 and Ti6C5 (Fig.31) during the ordering under thermodynamic equilibrium conditions. Accord- ing to calculations, an ordered phase Ti2C is formed by the second-order phase transition mechanism with the disorder ± order transition temperature Ttrans=920 ± 950 K. The homoge- neity region of the Ti2C phase is fairly broad (TiC0.42 ± TiC0.56 at 700 Kand TiC0.40 ± TiC0.54 at 800 K) and is confined from the left and from the right by two-phase regions, (a-Ti+Ti2C) and (Ti2C+Ti3C2), respectively.The ordered Ti3C2 and Ti6C5 phases have homogeneity regions of comparable width (TiC0.59 ± TiC0.71 for Ti3C2 and TiC0.74 ± TiC0.87 for Ti6C5 at 700 K) and are formed by the first-order phase transition mechanism. All order ± order phase transformations in titanium carbide also occur as first-order phase transitions. The calculated values of Ttrans , DStrans and the heat of phase transition, DHtrans, for all possible order ± order and disorder ± order first-order phase transformations of titanium carbide are listed in Table 12. By and large, the ordering in carbide TiCy occurs at T<1000 K. T /K L TiCy+L TiCy 3000 3050K 2500 TiCy 2 964 3 L+TiCy 2000 Ti2C 960 1Ti3C2Ti6C5 1918K 0.50 0.52 0.54 0.48 1500 b-Ti b-Ti+TiCy Ti2C+Ti3C2Ti3C2+Ti6C5 TiCy+C 1000 1193K a-Ti+TiCy Ti6C5+TiCy Ti2C Ti3C2 919K a-Ti a-Ti+Ti2C Ti6C5 500 y 0.6 0.4 0.2 0 1.0 0.8 Figure 31.Equilibrium phase diagram of the Ti ±C system constructed with allowance for atomic ordering of nonstoichiometric cubic titanium carbide TiCy (see Ref. 1). A portion of the ordering region corresponding to peritectoid transformation Ti3C2+TiCy ? Ti6C5 (T=963 K) is shown in the inset (magnified); the coordinates, y=C/Ti, of points 1, 2 and 3 are 0.5236, 0.524 and 0.5255, respectively. The calculated phase diagram of the Ti ±C system (see Fig. 31) is in agreement with the results of experimental studies (see Section IV).Different authors reported the formation of an ordered phase Ti2C with cubic or trigonal symmetry (space group Fd 3m or R3m, respectively) in carbide TiCy in the region 0.54y40.65 at T<1100 K. The X-ray diffraction patterns of annealed titanium carbide of composition TiC0.60 ± TiC0.70 exhib- ited superstructure reflections characteristic of the ordered phase Ti3C2. Recently,130 an ordered phase Ti6C5 with hexagonal symmetry was found in thin Ti ±C films. The existence of a trigonal ordered phase Ti2C and an orthorhombic ordered phase Ti3C2 was also confirmed by the Monte Carlo calculations of the phase diagram of the Ti ±C system for the composition range TiC0.57 ± TiC0.70 (Fig. 32).10, 12 Landesman et at.138 showed that short-range order and disorder ±Phase equilibria, phases and chemical compounds in the Ti ±C system Table 12.Thermodynamic characteristics of the disorder ± order and order ± order phase transformations in nonstoichiometric titanium carbide TiCy (see Ref. 1). y Ti2C±Ti3C2 Ttrans DStrans /K /J mol71 K71 948 926 893 853 795 460 0.09 0.15 0.25 0.28 0.32 0.29 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 0.52 0.53 0.54 0.55 0.56 0.58 0.60 0.62 0.63 0.64 0.65 0.66 0.68 0.70 0.72 0.73 0.74 0.75 0.76 0.78 0.80 0.82 0.83 0.84 0.85 0.86 0.88 0.90 0.92 0.94 0.95 T /K 800 Ti2C(R3m) 700 Ti2C+Ti3C2 600 0.65 0.60 Figure 32.Portion of the phase diagram of the Ti ±C system obtained from Monte Carlo calculations with allowance for the ordering of non- stoichiometric titanium carbide TiCy in the composition range TiC0.57 ± TiC0.70 (see Refs 10, 12). SRO denotes a phase with the B1-type structure and short-range order. order transitions in nonstoichiometric carbides and nitrides can be described by using the Ising Hamiltonian for the non-metal fcc sublattice. In this case the Hamiltonian is defined as the sum of the effective energies, Vn, of the pairwise interactions between the sites Ti3C2±Ti6C5 Ttrans DStrans DHtrans /K /J mol71 K71 /kJ mol71 961 959 956 951 940 924 902 891 877 860 841 794 733 639 566 472 320 0.08 0.14 0.22 0.24 0.25 0.13 7777777777777777777777777 7 70.003 0.013 0.03 0.05 0.12 0.19 0.29 0.35 0.41 0.48 0.56 0.74 0.91 0.99 1.00 0.90 0.69 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 SRO Ti3C2 SRO+Ti3C2 0.70 y DHtrans /kJ mol71 70.003 0.013 0.03 0.05 0.12 0.18 0.26 0.31 0.36 0.42 0.47 0.59 0.66 0.63 0.57 0.42 0.22 7777777777777 in the non-metal sublattice that are the nth meighbours.The energies Vn of the pairwise interactions in the non-metal sublattice of the carbides TiC0.76 and TiC0.65 were calculated 10, 12 not only for the nearest (V1) and next-nearest (V2) neighbours, but also for more distant sites in the third (V3) and fourth (V4) coordination spheres.The energies of the pairwise interactions were calculated from the experimental data on high-temperature diffuse neutron scattering 139 using three methods, namely, the mean field approx- imation, the inverse Monte Carlo method and the cluster variation method. The best agreement with the experiment was achieved when the phase diagram was calculated by the Monte Carlo method. The formation of an ordered orthorhombic phase Ti3C2, was reported to occur as a first-order phase transition.12, 139 This conclusion is consistent with the results of OPF calculations of the ordering in TiCy.18, 20 Some distinctions in the mutual positions of the ordered phases Ti2C and Ti3C2 in the phase diagrams of the Ti ±C system obtained from the OPF (see Fig.31) and Monte Carlo calcula- tions (see Fig. 32) are due to the fact that the OPF method treats the free energy of a compound as a function of its composition and long-range order parameters, whereas in the second method the energies of the interatomic interactions are determined by the short-range order parameters. 2. Experimental phase diagram of the Ti ±C system constructed with allowance for low-temperature (at T<1000 K) equilibria The phase diagram of the Ti ±C system at T<1000 K was constructed based on the experimental results on the ordering of carbide TiCy.21 Using the data on the structure and electrical Ti6C5 ± TiCy Ttrans DStrans /K /J mol71 K71 7 70.02 0.03 0.04 0.06 0.10 0.15 0.22 0.26 0.30 0.35 0.40 0.50 0.62 0.75 0.82 0.90 0.97 1.05 1.20 1.34 1.46 1.49 1.49 1.48 1.43 1.27 1.05 0.82 0.58 0.46 962 961 960 959 955 949 942 938 934 929 924 912 898 882 874 864 854 843 818 789 755 728 713 689 662 601 528 444 348 294 459 DHtrans /kJ mol71 70.02 0.03 0.04 0.05 0.09 0.14 0.21 0.24 0.28 0.32 0.37 0.46 0.56 0.67 0.72 0.78 0.83 0.88 0.98 1.06 1.10 1.08 1.07 1.02 0.95 0.76 0.55 0.36 0.20 0.14460 resistance of annealed and quenched samples of TiCy, it was (space group Fd 3m) and trigonal (space group R3m) ordered shown that cubic ordered phase Ti2C (space group Fd 3m) has a phase Ti2C and orthorhombic ordered phase Ti3C2 (space group homogeneity region extending from TiC0.49 ± 0.51 to TiC0.54 ± 0.55, C2221).This is consistent with the phase diagram of the Ti ±C while the trigonal superstructure Ti2C (space group R3m) is system (see Fig. 31) obtained from OPF calculations 18 (note that formed in the composition range TiC0.55 ± TiC0.59 . The for Ti2C-type superstructures the OPF method leads to a common TiC0.59 ± TiC0.63 range corresponds to a two-phase region [Ti2C region of existence, from TiC0.46 to TiC0.58, without separation (space group R3m)+Ti3C2 (space group C2221)].The existence into the cubic and trigonal phases). region of the orthorhombic ordered phase Ti3C2 (space group As mentioned above (see Section III), recent studies of the C2221) is fairly narrow and does not exceed TiC0.63 ± TiC0.67. The Ti ±C system revealed the existence of molecular cluster-like lowest annealing temperature was 770 K. Since no ordered phase compounds Ti8C12 and Ti13C22 in addition to nonstoichiometric Ti6C5 was detected after annealing the TiC0.83 and TiC0.85 samples cubic titanium carbide TiCy. The phase diagram of the Ti ±C at such a low temperature, the transition temperature of this phase system constructed with allowance for the ordering of nonstoi- was assumed to be lower than 770 K.The portion of the chiometric carbide TiCy and for the existence of molecular cluster- equilibrium phase diagram of the Ti ±C system corresponding to like compounds Ti8C12 and Ti13C22 (TiC2) is shown in Fig. 34. the ordering of nonstoichiometric titanium carbide TiCy is pre- sented in Fig. 33. A consecutive transition was suggested 21 in the range 0.544y40.57 99020K disordered (space group Fm3m) carbide TiCy 96020K cubic (space group Fd 3m) ordered phase Ti2C trigonal (space group R3m) ordered phase Ti2C. The ordered orthorhombic phase Ti3C2 is formed following a peritectoid reaction Ti3C2 Ti2C+TiCy at 99010 K in the composition range 0.614y<0.63. The ordered phases Ti2C with cubic and trigonal symmetry can be observed over a broad composition range of nonstoichiometric titanium carbide, from TiC0.40 to TiC0.63.However, only in the TiC0.49 ± 0.50 ± TiC0.58 ± 0.59 range is the homogeneity region of the ordered phase Ti2C (see Fig. 33). The composition range in which the formation of an ordered phase Ti6C5 is possible is shown tentatively since this phase was not found experimentally. The hysteresis in the temperature dependences of the electrical resist- ance 21 suggested that the TiCy>Ti2C and TiCy>Ti3C2 trans- formations are first-order phase transitions with Ttrans=980 ± 1000 K. This value is in good agreement with the value Ttrans&1000 K found for titanium carbide TiC0.55 from electrical resistance measurements.140 Thus, experiments revealed that ordering of TiCy in the ranges 0.524y40.55, 0.564y40.58 and 0.624y<0.68 at T<1000 K leads respectively to the formation of the cubic T /K TiCy a-Ti+TiCy TiCy+Ti2C 1000 919K Ti2C (R3m) TiCy+Ti3C2 Ti2C (R3m)+Ti3C2 Ti3C2 (C2221) 800 Ti6C5+TiCy 600 Ti6C5 a-Ti+ +Ti2C (Fd3m) Ti3C2+ +Ti6C5 Ti2C (Fd3m) 400 0.8 0.6 0.9 0.7 y 0.5 0.4 Figure 33.Low-temperature portion of the equilibrium phase diagram of the Ti ±C system.21 Formation of the ordered orthorhombic Ti3C2 phase occurs at 99010 K as the peritectoid transformation Ti2C+TiCy ? Ti3C2; the region of phase equilibria involving ordered Ti6C5 phase is tentatively shown by the dashed line. A I Gusev T /K 3500 L TiCy+L 3000 3050K 2500 TiCy+C TiCy L+TiCy 2000 1918K b-Ti Ti8C12 TiC2 2+3 1500 b-Ti+TiCy 1 2 3 3+4 1000 1193K a-Ti+TiCy 919K 4 a-Ti 500 a-Ti+Ti2C (Fd3m) 40 80 C (at.%) 60 20 0 Figure 34.Phase diagram of the Ti ±C system constructed with allowance for the existence of molecular cluster-like compounds Ti8C12 and Ti13C22 (TiC2) and for the atomic ordering of nonstoichiometric cubic titanium carbide TiCy . Ti2C (space group Fd3m) (1), Ti2C (space group R3m) (2), Ti3C2 (space group C2221) (3) and Ti6C5 (Ti2C, Ti3C2 and Ti6C5 are the ordered phases of titanium carbide TiCy) (4). VI. Conclusion For more than a century it was thought that there is only one compound in the Ti ±C system, namely, cubic titanium carbide TiCy with a broad homogeneity region. However, the studies of this system carried out in 1990s revealed the formation of (i) ordered phases of nonstoichiometric titanium carbide with differ- ent composition and symmetry and (ii) molecular cluster-like compounds. Among the new phases of titanium carbide, the ordered phase Ti2C, which can have a cubic or trigonal symmetry, has been studied in most detail.This phase is readily formed as a result of annealing the TiCy samples at 1300 ± 1500 K followed by gradual lowering of the temperature down to 600 ± 700 K over a period of several hours. It can also be found even in the TiCy samples (y=0.5 ± 0.58) synthesised by solid-phase vacuum sintering of titanium and carbon at 2000 ± 2200 K and cooled down to 300 K during further self-cooling of the furnace.Using a special regime of long-term annealing, one can obtain the Ti2C phase with the maximum extent of long-range order. Ordered phases of the Ti3C2 and Ti6C5 types are much more hard to obtain. Similar to the Ti2C phase, they can be of different symmetry while their chemical composition remains unchanged.Phase equilibria, phases and chemical compounds in the Ti ±C system Bulk samples of these phases can be obtained only after long-term (several hundreds of hours) annealing at T<1000 K. The results of experimental and theoretical studies suggest that these phases are in thermodynamic equilibrium at temperatures below 900 ± 1000 K. A hexagonal Ti6C5 phase was found in thin films synthesised by electron beam and laser sputtering of titanium carbide.Molecular cluster-like compounds Ti8C12 (Ti2C3) and Ti13C22 (*TiC2) were obtained by laser-assisted plasmachemical syn- thesis. The most convenient procedure for preparation of these compounds in the crystalline form is to stabilise them in thin films deposited on a substrate (molecular clusters with a characteristic atomic ratio, C: Ti=1.5 or 2).141 This allows one to obtain thin films of a compound Ti2C3 or TiC2 with the crystal structure similar to that of the substrate. Indeed, the synthesis of titanium oxide, fullerene and amorphous carbon nanocrystals with a different structure on the solid surfaces of different chemical nature (graphite, polymers, oxides) was reported.142 Chemical modification of the substrate surface or heat treatment of the film cast provide additional possibilities of forming a particular type of the crystal structure of the film.The main methods of experimental detection of new com- pounds and phases include X-ray, electron or neutron diffraction. We believe that the model X-ray diffraction patterns of most of the phases of the Ti ±C system, first reported in this review will essentially facilitate identification of new phases and interpreta- tion of the results obtained. Thus, analysis of studies on the Ti ±C system carried out in 1985 ± 2000 showed that not only titanium carbide TiCy, but also other chemical compounds and phases are formed in this system. They were discovered using novel, advanced methods of synthesis, thorough studies of phase equilibria at temperatures below 1300 K and highly sensitive mass spectrometric and diffraction methods of analysis of products of the chemical interaction.One can expect (and some experimental proofs have already been reported) that binary systems comprising carbon and other transition metals can appear to be no less interesting than the Ti ±C system. This work has been written with the financial support of the Russian Foundation for Basic Research (Project No. 01-03- 96510). References 1. A I Gusev Usp. Fiz. Nauk 170 3 (2000) a 2. A I Gusev, A A Rempel, A J Magerl Disorder and Order in Strongly Nonstoichiometric Compounds: Transition Metal Carbides, Nitrides and Oxides (Berlin; Heidelberg; New York; London: Springer, 2001) 3.A I Gusev, A A Rempel Nestekhiometriya, Besporyadok i Poryadok v Tverdom Tele (Nonstoichiometry, Disorder and Order in Solid) (Ekaterinburg: Urals Branch of the Russian Academy of Sciences, 2001) 4. P W Shimer Proc. R. Soc. 42 89 (1887) 5. P W Shimer Chem. News 55 156 (1887) 6. H Moissan Compt. Rend. 120 290 (1895) 7. E Rudy, D P Harmon, C E Brukl, in Ternary Phase Equilibria in Transition Metal-Boron-Carbon-Silicon Systems. Related Binary Systems (Tech. Report AFML-TR-65-2) Vol. 2, Pt. I (Ohio, USA: Metals and Ceramics Division, Air Force Materials Laboratory Wright-Patterson Air Force Base, 1965) p. 1 8. E K Storms The Refractory Carbides (New York; London: Academic Press, 1967) 9.H Goretzki Phys. Status Solidi 20 K141 (1967) 10. T Priem Etude de L'ordre a Courte Distance Dans les Carbures et Nitrures Non-Stoechiometriques de Metaux de Transition par Diffusion Diffuse de Neutrons (Rapport CEA-R-5499) (Gif-sur-Yvette: Commissariat a l'Energie Atomique Centre d'Etudes Nucleaires de Saclay, 1989) 11. A I Gusev Dokl. Akad. Nauk SSSR 313 887 (1990) b 461 12. C H de Novion, B Beuneu, T Priem, N Lorenzelli, A Finel, in The Physics and Chemistry of Carbides, Nitrides and Borides (Ed. R Freer) (Netherlands: Kluwer Academic, 1990) p. 329 13. A I Gusev Fiz. Tv. Tela 32 2752 (1990) c 14. A I Gusev Fizicheskaya Khimiya Nestekhiometricheskikh Tugoplav- kikh Soedinenii (Physical Chemistry of Nonstoichiometric Refractory Compounds) (Moscow: Nauka, 1991) 15.A I Gusev, A A Rempel' Zh. Fiz. Khim. 65 625 (1991) d 16. A I Gusev Dokl. Akad. Nauk 322 918 (1992) b 17. A A Rempel' Effekty Uporyadocheniya v Nestekhiometricheskikh Soedineniyakh Vnedreniya (The Ordering Effects in Nonstoichio- metric Intercalation Compounds) (Ekaterinburg: Nauka, 1992) 18. A I Gusev, A A Rempel Dokl. Akad. Nauk 332 717 (1993) b 19. A I Gusev, A A Rempel J. Phys. Chem. Solids 55 299 (1994) 20. A I Gusev, A A Rempel Phys. Status Solidi A 163 273 (1997) 21. V N Lipatnikov, A Kottar, L V Zueva, A I Gusev Fiz. Tv. Tela 40 23. V N Lipatnikov, A I Gusev, P Ettmayer,W Lengauer J. Phys., 1332 (1998) c 22. A I Gusev, A A Rempel, in Materials Science of Carbides, Nitrides and Borides (Eds Y G Gogotsi, R A Andrievski) (Dordrecht: Kluwer Academic, 1999) p.47 Condens. Matter 11 163 (1999) 24. V N Lipatnikov,A I Gusev, P Ettmaier, V Lengauer Fiz. Tv. Tela 41 529 (1999) c 25. A I Gusev Dokl. Akad. Nauk 365 644 (1999) b 26. A I Gusev Zh. Fiz. Khim. 74 600 (2000) d 27. V N Lipatnikov, A I Gusev Uporyadochenie v Karbidakh Titana i Vanadiya (Ordering in Titanium and Vanadium Carbides) (Ekaterinburg: Urals Branch of the Russian Academy of Sciences, 2000) 28. B C Guo, K P Kerns, A W Castleman Science 255 1411 (1992) 29. Z Lin,M B Hall J. Am. Chem. Soc. 115 11165 (1993) 30. B C Guo, S Wei, J Purnell, S Buzza, A W Castleman Jr Science 256 515 (1992) 31. S Wei, B C Guo, J Purnell, S Buzza, A W Castleman J. Phys. Chem. 96 4166 (1992) 32.J S Pilgrim, M A Duncan J. Am. Chem. Soc. 115 4395 (1993) 33. B C Guo, A W Castleman Adv. Met. Semicond. Clusters 2 137 (1994) 34. A W Castleman Proc. Electrochem. Soc. 92 ± 94 346 (1994) 35. M Methfessel,M van Schilfgaarde, S Scheffler Phys. Rev. Lett. 70 29 (1993) 36. B V Reddy, S N Khanna, P Jena Science 258 1640 (1992) 37. M V Ryzhkov, A L Ivanovskii Zh. Strukt. Khim. 40 630 (1999) e 38. S N Khanna, B V Reddy Comput. Mater. Sci. 2 638 (1994) 39. E Rudy, in Compendium of Phase Diagram Data. Metals and Ceramics Division (Final Tech. Report AFML TR-65-2) (Ohio, USA: Air Force Materials Laboratory, Wright-Patterson Air Force Base, 1969) Pt. V 40. R L Bickerdike, G L Hughes J. Less-Common Met. 1 42 (1959) 41. J L Murray, in Phase Diagrams of Binary Titanium Alloys (Ed.J L Murray) (Metals Park, OH: ASM Int. Publ., 1987) p. 47 42. J L Murray, in Binary Alloy Phase Diagrams Vol. 1 (Ed. T B Massalski) (Metals Park, OH: ASM Int. Publ., 1987) p. 593 43. T B Massalski, P R Subramanian, H Okamoto, L Kasprzak (Eds) Binary Alloy Phase Diagrams Vol. 1 (Metals Park, OH: ASM Int. Publ., 1990) 44. S Jonsson, PhD Thesis, Royal Institute of Technology, Stockholm, 1993 45. H Bittner, H Goretzki Monatsh. Chem. 93 1000 (1962) 46. M P Arbuzov, B V Khaenko, E T Kachkovskaya, S Ya Golub Ukr. Fiz. Zh. 19 497 (1984) 47. V T Em, I A Karimov, V F Petrunin, I Khidirov, I S Latergaus, A G Merzhanov, I P Borovinskaya, V K Prokudina 48. I Karimov, V T Em, I Khidirov, I S Latergaus Izv.Akad. Nauk 49. V Moisy-Maurice, N Lorenzelli, C H de Novion, P Convert Acta 50. A Sh Remeev, I Karimov Izv. Akad. Nauk Uzb. SSR, Ser. Fiz.-Mat. Kristallografiya 20 320 (1975) f Uzb. SSR, Ser. Fiz.-Mat. Nauk (4) 81 (1979) Metall. 30 1769 (1982) Nauk (2) 87 (1986) 51. V A Vlasov, I A Karimov, L V Kustova Izv. Akad. Nauk SSSR, Neorg. Mater. 22 231 (1986) g 52. A N Emel'yanov Teplofiz. Vys. Temp. 28 269 (1990) h462 53. A V Karpov, V P Kobyakov, E A Chernomorskaya Neorg. Mater. 31 655 (1995) g 54. A V Karpov, V P Kobyakov Fiz. Tv. Tela 38 1638 (1996) c 55. V N Lipatnikov, A I Gusev Phys. Status Solidi B 212 R11 (1999) 56. V N Lipatnikov, A I Gusev Pis'ma Zh. Eksp. Teor. Fiz. 69 631 (1999) i 57. L V Zueva, A I Gusev Fiz. Tv. Tela 41 1134 (1999) c 58.G L Humphrey J. Am. Chem. Soc. 73 2261 (1951) 59. M P Morozova,M K Khripun, S M Ariya Zh. Obshch. Khim. 32 2072 (1962) j 60. A N Zyryanova, L V Zueva, A I Gusev, in Khimiya Tverdogo Tela. Struktura, Svoistva i Primenenie Novykh Neorganicheskikh Materialov (The Solid State Chemistry. Structure, Properties and Application of New Inorganic Materials) (Ekaterinburg: Institute of Solid State Chemistry, Urals Branch of the Russian Academy of Sciences, 1998) p. 36 61. L V Zueva, A I Gusev Zh. Fiz. Khim. 73 792 (1999) d 62. B Naylor Am. Chem. Soc. 68 370 (1946) 63. L S Levinson J. Chem. Phys. 42 2891 (1965) 64. V V Fesenko, A G Turchanin, E A Guseva Zh. Fiz. Khim. 42 1026 (1968) d 65. V V Fesenko, A G Turchanin, in Tugoplavkie Karbidy (The Refractory Carbides) (Kiev: Naukova Dumka, 1970) p.200 66. V E Ganenko, G A Berezovskii, V S Neshpor, G M Klimashin, in Raboty po Fizike Tverdogo Tela (Studies on Physics of Solid State) (Novosibirsk: Nauka, 1968) Vol. 3, p. 127 67. N Lorenzelli, R Caudron, J P Landesman, C H de Novion Solid State Commun. 59 765 (1986) 68. V P Glushko (Ed.) Termodinamicheskie Svoistva Individual'nykh Veshchestv. Spravochnik (Thermodynamic Properties of Individual Compounds. Handbook) (Moscow: Nauka, 1979) Vol. 2, Pts 1, 2 69. R B Kotel'nikov, S N Bashlykov, Z G Galiakbarov, A I Kashtanov Osobo Tugoplavkie Elementy i Soedineniya (Especially Refractory Elements and Compounds) (Moscow: Metallurgiya, 1969) 70. A I Gusev, A A Rempel Strukturnye Fazovye Perekhody v Neste- khiometricheskikh Soedineniyakh (Structural Phase Transitions in Nonstoichiometric Compounds) (Moscow: Nauka, 1988) 71.J P Hirth,G M Pound Condensation and Evaporation-Nucleation and Growth Kinetics (Oxford: Pergamon Press, 1963) 72. A I Gusev Dokl. Akad. Nauk SSSR 311 631 (1990) b 73. A I Gusev Teplofiz. Vys. Temp. 29 695 (1991) h 74. W A Chupka, J Berkowitz, C F Giese,M G Inghram J. Phys. Chem. 62 611 (1958) 75. T S Starostina, L N Sidorov, P A Akishin, N M Karasev Izv. Akad. Nauk SSSR, Neorg. Mater. 3 727 (1967) g 76. S Fujishiro, N A Gokcen J. Phys. Chem. 65 161 (1961) 77. C A Stearns, F J Kohl High Temp. Sci. 2 274 (1970) 78. I S Kulikov Termodinamika Karbidov i Nitridov (Thermodynamics of Carbides and Nitrides) (Chelyabinsk: Metallurgiya, 1988) 79.V V Fesenko, A S Bolgar Teplofiz. Vys. Temp. 7 244 (1969) h 80. A I Gusev, A A Rempel Izv. Akad. Nauk SSSR, Neorg. Mater. 27 963 (1991) g 81. T Ya Kosolapova (Ed.) Svoistva, Poluchenie i Primenenie Tugoplav- kikh Soedinenii. Spravochnik (Properties, Preparation and Applications of Refractory Compounds. Handbook) (Moscow: Metallurgiya, 1986) 82. H Gleiter Nanostruct. Mater. 1 1 (1992) 83. R Birringer, H Gleiter, in Encyclopedia of Material Science and Engineering (Suppl.) Vol. 1 (Ed. R W Cahn) (Oxford: Pergamon Press, 1988) p. 339 84. R W Siegel Ann. Rev. Mater. Sci. 21 559 (1991) 85. A I Gusev Usp. Fiz. Nauk 168 55 (1998) a 86. Yu I Petrov Fizika Malykh Chastits (The Physics of Small Particles) (Moscow: Nauka, 1982) 87.H Gleiter, in Deformation of Polycrystals: Mechanisms and Microstructures (Roskilde: Risù National Laboratory, 1981) p. 15 88. H Gleiter, P Marquardt Z. Met.kd. 75 263 (1984) 89. R Birringer, U Herr, H Gleiter Trans. Jpn. Inst. Met. (Suppl.) 27 43 (1986) 90. H Gleiter Prog. Mater. Sci. 33 233 (1989) 91. A I Gusev Nanokristallicheskie Materialy: Metody Polucheniya i Svoistva (Nanocrystalline Materials: Preparation Methods and Properties) (Ekaterinburg: Urals Branch of the Russian Academy of Sciences, 1998) A I Gusev 92. A I Gusev, A A Rempel Nanokristallicheskie Materialy (Nanocrystalline Materials) (Moscow: Fizmatlit-Nauka, 2000) 93. E Neuenschwander J. Less-Common Met. 11 365 (1966) 94. T Ya Kosolapova, G N Makarenko, D P Zyat'kevich Zh. Vsesoyuz. Khim. O-va im. D I Mendeleeva 24 228 (1979) k 95. J I Steinfeld (Ed.) Laser-Induced Chemical Processes (New York: Plenum, 1981) 96. R A Bauer, J G M Becht, F E Kruis, B Scarlett, J Schoonman J. Am. Ceram. Soc. 74 2759 (1991) 97. M Suzuki, Y Maniette, Y Nakata, T Okutani J. Am. Ceram. Soc. 76 1195 (1993) 98. N V Karlov, M A Kirichenko, B S Luk'yanchuk Usp. Khim. 62 223 (1993) [Russ. Chem. Rev. 62 203 (1993)] 99. A Teresiak, H Kubsch Nanostruct. Mater. 6 671 (1995) 100. M S El-Eskandarany,M Omori, T Kamiyama, T J Konno, K Sumiyama, T Hirai, K Suzuki Sci. Rep. Res. Inst. Tohoku Univ. A, Phys. Chem. Metall. 43 181 (1997) 101. R Z Valiev, A V Korznikov, R R Mulyukov Mater. Sci. Eng. A 168 141 (1993) 102. A A Rempel, A I Gusev, R R Mulyukov, in Ul'tradispersnye Poroshki, Materialy i Nanostruktury. Materialy Mezhregional'noi Konferentsii (Ultradispersed Powders Materials and Nanostruc- tures. Materials of International Conference) (Ed. V E Red'kin) (Krasnoyarsk: Krasnoyarsk State Technological University, 1996) p. 131 103. E A Rohlfing, D M Cox, A Kaldor J. Chem. Phys. 81 3322 (1984) 104. H W Kroto, J R Heath, S C O'Brien, R F Curl, R E Smalley Nature (London) 318 162 (1985) 105. R F Curl, R E Smalley Science 242 1017 (1988) 106. H W Kroto, A W Allaf, S P Balm Chem. Rev. 91 1213 (1991) 107. A V Eletskii, B M Smirnov Usp. Fiz. Nauk 165 977 (1995) a 108. G S Hammond, V J Kuck (Eds) Fullerenes. Synthesis, Properties, and Chemistry of Large Carbon Clusters. ACS Symposium Series Vol. 481 (Washington, DC: American Chemical Society, 1992) 109. J H Weaver, J L Martins, T Komeda, Y Chen, T R Ono, G H Kroll, N Troullier, R E Haufler, R E Smalley Phys. Rev. Lett. 66 1741 (1991) 110. S Saito, A Oshiyama Phys. Rev. Lett. 66 2637 (1991) 111. B C Guo, K P Kerns, A W Castleman J. Am. Chem. Soc. 115 7415 (1993) 112. A A Sofronov, Candidate Thesis in Chemical Sciences, Institute of Solid State Chemistry, Urals Branch of the Russian Academy of Sciences, Ekaterinburg, 2000 113. A L Ivanovskii, A A Sofronov, Yu N Makurin Teor. Eksp. Khim. 35 289 (1999) 114. D W Jones, I J McColm, R Steadman, J Yerkess J. Solid State Chem. 68 219 (1987) 115. S Raman, G N Ramanchandran Curr. Sci., A 31 321 (1962) 116. V I Perekrestov, A V Pavlov Pis'ma Zh. Eksp. Teor. Fiz. 73 17 (2001) i 117. M S Dresselhaus, G Dresselhaus, P C Eklund Science of Fullerenes and Carbon Nanotubes (San-Diego: Academic Press, 1996) 118. A L Buchachenko Usp. Khim. 68 99 (1999) [Russ. Chem. Rev. 68 85 (1999)] 119. T Guo,M D Diener, Y Chai,M J Alford, R E Hauter, S M McClure, T Ohto, J H Weaver, G E Scuseria, R E Smalley Science 257 1661 (1992) 120. Yu N Makurin, A A Sofronov, A I Gusev, A L Ivanovskii Chem. Phys. 270 293 (2001) 121. T Guo, R E Smalley, G E Scuseria J. Chem. Phys. 99 352 (1993) 122. N A Dubrovinskaia, L S Dubrovinsky, S K Saxena, R Ahuja, B Johansson J. Alloys Comp. 289 24 (1999) 123. B V Khaenko Izv. Akad. Nauk SSSR, Neorg. Mmater. 15 1952 (1979) g 124. C H de Novion, J P Landesman Pure Appl. Chem. 57 1391 (1985) 125. A I Gusev Phys. Status Solidi B 163 17 (1991) 126. A I Gusev, A A Rempel Phys. Status Solidi A 135 15 (1993) 127. A A Rempel Usp. Fiz. Nauk 166 33 (1996) a 128. B V Khaenko, S Ya Golub, M P Arbuzov Kristallografiya 25 112 (1980) f 129. V Moisy-Maurice, in Structure Atomique des Carbures Non-stoechiometriques de Metaux de Transition (Rapport CEA-R- 5127) (Gif-sur-Yvette: Commissariat a l'Energie Atomique, 1981)463 Phase equilibria, phases and chemical compounds in the Ti ±C system 130. N V Dzhalabadze, B G Eristavi, N I Maisuradze, E R Kuteliya Fiz. Met. Metalloved. 86 85 (1998) m 131. M Yu Tashmetov, V T Em,M U Kalanov, V M Shkiro Metallofizika 13 100 (1991) 132. V N Lipatnikov, A A Rempel, A I Gusev Int. J. Refract. Met. Hard Mater. 15 61 (1997) 133. V Moisy-Maurice, C H de Novion, A N Christensen, W Just Solid State Commun. 39 661 (1981) 134. A I Gusev, A A Rempel J. Phys. C, Solid State Phys. 20 5011 (1987) 135. A I Gusev, A A Rempel Phys. Status Solidi B 131 43 (1985) 136. A I Gusev, A A Rempel Phys. Status Solidi B 140 335 (1987) 137. A I Gusev Philos. Mag. B 60 307 (1989) 138. J P Landesman, G Treglia, P Turchi, F Ducastelle J. Phys. (France) 46 1001 (1985) 139. T Priem, B Beuneu, C H de Novion, J Chevrier, F Livet, A Finel, S Lefevbre Physica B 156 ± 157 47 (1989) 140. V N Lipatnikov, A I Gusev Pis'ma Zh. Eksp. Teor. Fiz. 70 284 (1999) i 141. R B King Izv. Akad. Nauk, Ser. Khim. 5 862 (1998) n 142. E P Smirnov, A M Abyzov, V Vege, in Khimiya Vysokoorganizo- vannykh Veshchestv i Nauchnye Osnovy Nanotekhnologii (3-ya Mezhdunarodnaya Konferentsiya) [The Chemistry of Highly Organised Substances and Scientific Foundations of Nano- technology (The 3th International Conference)] (St Petersburg: St Petersburg State University, 2001) p. 58 a�Physics-Uspekhi (Engl. Transl.) b�Dokl. Chem. (Engl. Transl.) c�Phys. Solid State (Engl. Transl.) d�Russ. J. Phys. Chem. (Engl. Transl.) e�J. Struct. Chem. (Engl. Transl.) f�Crystallogr. Rep. (Engl. Transl.) g�Inorg. Mater. (Engl. Transl.) h�High Temp. (Engl. Transl.) i�J. Exp. Theor. Phys. Lett. (Engl. Transl.) j�Russ. J. Gen. Chem. (Engl. Transl.) k�Mendeleev Chem. J. (Engl. Transl.) m�Phys. Met. Metall. (Engl. Transl.) n�Russ. Chem. Bull.
ISSN:0036-021X
出版商:RSC
年代:2002
数据来源: RSC
|
2. |
Redox reactions of actinides in carbonate and alkaline solutions |
|
Russian Chemical Reviews,
Volume 71,
Issue 6,
2002,
Page 465-488
Vladimir P. Shilov,
Preview
|
|
摘要:
Russian Chemical Reviews 71 (6) 465 ± 488 (2002) Redox reactions of actinides in carbonate and alkaline solutions V P Shilov, A B Yusov Contents I. Introduction II. Redox potentials III. Reactions in carbonate solutions IV. Reactions in alkaline media V. Conclusion Abstract. neptunium, uranium, involving reactions redox on Data Data on redox reactions involving uranium, neptunium, plutonium alkaline and carbonate in ions americium and plutonium and americium ions in carbonate and alkaline solu- solu- tions of studies kinetic of results The generalised. are tions are generalised. The results of kinetic studies of these these reactions are analysed and their mechanisms are discussed. The reactions are analysed and their mechanisms are discussed. The bibliography references 169 includes bibliography includes 169 references.I. Introduction Progress in the chemistry of actinides has long been associated predominantly with the necessity of solving technological prob- lems. Early chemical investigations were concerned primarily with the reactions of actinides in acidic media, whereas the reactions in alkaline media were little studied. However, these scarce data are of most interest. Thus, the use of carbonate and alkaline solutions made it possible to stabilise valence states of actinides, which are either unstable or difficult to prepare in acidic media. For instance, Am(IV) was stabilised and Cf(V) and Cm(VI) were first obtained in carbonate media. The most prominent example is the discovery of heptavalent neptunium, plutonium and americium, which gave impetus to the outburst of studies of alkaline solutions of actinides in the late 1960s and in the early 1970s.In recent years, interest in redox reactions of actinides in alkaline media was rekindled. The reason is that there is an acute need to solve a number of problems associated with storage and processing of alkaline waste of radiochemical industry, which were accumulated in large amounts at plants both in Russia and abroad. Extensive studies in this field gave new interesting results, which are not only of applied but also of scientific importance. In the last decade, the properties of actinides in carbonate and alkaline media were surveyed in reviews. For example, the compositions and structures of carbonate complexes in solutions and in the solid state were covered in detail in the review;1 however, no consideration was given to redox reactions.The report 2 was devoted to the chemistry of actinides in alkaline media, including redox reactions. Unfortunately, this report is not easily accessible to Russian researchers. Besides, the results of more recent investigations were published in the literature. V P Shilov, A B Yusov Institute of Physical Chemistry, Russian Academy of Sciences, Leninsky prosp. 31, 119991 Moscow, Russian Federation. Fax (7-095) 952 53 08. Tel. (7-095) 330 41 70 E-mail: Shilov@ipc.rssi.ru (V P Shilov), Ioussov@ipc.rssi.ru (A B Yusov) Received 20 March 2002 Uspekhi Khimii 71 (6) 533 ± 558 (2002); translated by T N Safonova #2002 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2002v071n06ABEH000719 465 465 465 473 486 The aim of the present review is to systematise and analyse the available data on redox reactions of uranium, neptunium, pluto- nium, americium and curium in bicarbonate, carbonate and alkaline solutions.{Wedo not consider the data on photochemical and radiochemical reactions, which were generalised in the publications 3, 4 in sufficient detail.Studies of electrochemical reactions involving actinide ions are also beyond the scope of our review. II. Redox potentials To predict the possibility of a particular redox reaction occurring, the data on redox potentials (hereinafter referred to as potentials) of actinide pairs in carbonate and alkaline media are required.The measured and estimated (calculated) formal potentials (i.e., potentials at equal concentrations of the oxidised and reduced forms) are given in Table 1. III. Reactions in carbonate solutions It can be seen from Table 1 that tri- and tetravalent uranium, neptunium and plutonium can be oxidised by molecular oxygen, whereas trivalent neptunium and plutonium can also be oxidised by water (in the pH range of 9 ± 12, the potentials of the O2/H2O and H2O/H2 pairs vary from 0.70 to 0.52 V and from 70.53 to 70.71 V, respectively).38 Besides, U(V), Pu(V) and Am(IV) are unstable in disproportionation reactions, Np(VI) must oxidise Np(IV), whereas Am(VI) can oxidise Am(III) and Am(IV). 1.Disproportionation and reproportionation reactions a. Disproportionation of actinides(V) The kinetics of disproportionation of U(V) in carbonate solutions was examined by cyclic chronopotentiometry.39 It was found that U(V), which was generated on the electrode surface through reduction of U(VI), rapidly disappeared in solution. This process is best described by a model on the basis of the second-order reaction { It should be noted that the reaction mechanisms were not considered in most of original studies and it is impossible to make justified conclusions about these mechanisms based on the available data. In this connection, we write the reaction equations in the formal form, i.e., denote actinide ions as Np(VI), Pu(IV), etc.In addition, many reactions proceed in rather broad ranges of pH or alkali concentrations due to which the valence state of the actinide ion may vary.466 Table 1. Formal potentials of actinide pairs. Solution Potential of the pair /V (relative to the normal hydrogen electrode) VII/VI Uranium 7777 0.1 M NaHCO3 1 M NaHCO3 0.05 M Na2CO3 0.75 M Na2CO3+ +0.6 M NaClO4 0.1 ¡¾ 3 M Na2CO3 1 M K2CO3 1 M NaOH Neptunium 0.05 M Na2CO3 1 M Na2CO3 2 M Na2CO3 (pH 13) 2 M Na2CO3 (pH 13 ¡¾ 14) 0.2 M K2CO3 0.8 M K2CO3 1 M K2CO3 777 3 M K2CO3 1 M NaOH 70.582 0.587 0.462 0.225 0.604 7777777 4 M NaOH 10.4 M NaOH 1 M LiOH 2.5 M LiOH 4 M LiOH Plutonium 0.05 M Na2CO3 1 M Na2CO3 2 M Na2CO3 2 M Na2CO3+ +1 M NaOH 2 M Na2CO3+ +2 M NaOH a Determined by polarography. b Determined by cyclic voltammetry scan.c The diffusion potential between a 2 M Na2CO3 solution and a saturated calomel electrode was ignored. d Estimated by us based on the data of the study;24 under these conditions, an equilibrium mixture of Pu(VI), Pu(V) and Pu(IV) was formed. ¢§daUOVUa a 2kdaUOVUa2, dt where kd is the rate constant of disproportionation. The rate constants for disproportionation of U(V) in a 1 M NaHCO3 solution (pH 8.57, T=25 8C) remains unchanged as the concentration of U(VI) is varied in the range from 1 to 10 mmol litre71. An increase in pH{ leads to a decrease in the constant kd both at a constant CO2¢§ content (17 mmol litre71) and at a constant HCO¢§3 content (0.68 mol litre71).In the pH range of 8.57 ¡¾ 9.23, the constant kd varies from *15 ¡¾ 20 to 2 ¡¾ 3 litre mol71 s71. The following mechanism of the process was proposed:39 U(VI)O2(CO3)4¢§ 3 +e U(V)O2(CO3)53 ¢§, { In bicarbonate ¡¾ carbonate solutions, pH is generally controlled by changing the ratio between [CO2¢§ 3 ] and [HCO¢§3 ]. V P Shilov, A B Yusov Ref. Solution Ref. Potential of the pair /V (relative to the normal hydrogen electrode) IV/III V/IV VI/V VII/VI IV/III V/IV VI/V Plutonium 7 7 0.3 ¡¾ 0.35 d 7 70.98 a 70.63 b7 7 5 7 7 5 70.671 7 7 6 70.527 7 7 7 70.495 8, 13 1 M (NaHCO3+ +Na2CO3) (pH 11.0) 0.1 M K2CO3 1 M K2CO3 1 M NaOH 72.08 8 770.849 7 7 7 5 07 72.6 9, 10 70.71 70.6 70.65 4 M NaOH 7 7 13 0.67 7 20, 25 7 70.95 26 25 27 7 0.26 ¡¾ 0.29 7 0.445 0.44 0.47 b, c 7 7 6 7 7 11 7 71.4 d 12 0.30 0.32 0.23 0.21 7 7 0.44 77 7 7 20 0.21 0.20 0.2518 0.2556 0.715 70.540 77 7 7 28 7 7 20, 28 7 7 22 7 7 22 0.27 b 7 7 12 0.924 29, 30 0.9 7 7 0.43 0.44 0.44 7 7 13 7 7 13 0.1 0.887 30 0.92 31 7 7 7 7 7 0.89 30, 32, 0.975 71.32 8 71.3 b 12 7 7 13 7 7 14, 15 0.13 0.44 0.18 0.12 7 7 0.5 7 0.93 0.90 7 7 0.997 7 32 0.86 8, 13 34 71.8 16, 17, 18, 19 7 7 7 20 7 7 7 20 7 7 7 21 0.1527 0.1498 7 7 22 7 7 22 34 7 7 0.81 0.334 7 7 6 4.6 M NaOH 7.9 M NaOH 10.1 M NaOH 2.5 M LiOH 4 M LiOH Americium 1 M Na2CO3 2 M Na2CO3 2 M (NaHCO3+ +Na2CO3) (pH 9.7) 2 M (NaHCO3+ +Na2CO3) 33 (pH 10) 1 M K2CO3 2.82 M (KHCO3+ +K2CO3) (pH 8.87) 2.82 M (KHCO3+ +K2CO3) (pH 11.2) 1 M NaOH 1.05 0.65 0.68 7 7 35, 36 0.25 ¡¾ 0.50 <0.25 37 0.33 0.35 c 0.33 c 7 7 23 7 7 24 7 7 24 Curium 1.6 0.31 c 7 7 24 8 0.7 10 1 M K2CO3 1 M NaOH 7 7 0.5 7 7 7 K (3) U(V)O2(CO3)5¢§ 3 +HCO¢§3(1) U(V)O2(HCO3)(CO3)27+2CO2¢§ 3 ,kd (4) U(V)O2(CO3)5¢§ 3 +U(V)O2(HCO3)(CO3)27 U(VI)O2(CO3)4¢§ 3 +U(IV)O2(HCO3)(CO3)37, stable form of U(IV).U(IV)O2(HCO3)(CO3)37 3 The experimental logkd7log[HCO¢§3 ] plot yields a straight line (2) with a slope of 0.840.06. Consequently, the reaction order with respect to the bicarbonate ion is close to unity.The slope of the plot of kd vs. the concentration of free carbonate ions (on the logarithmic coordinates) is equal to 71.870.1, i.e., is close to 72.It should be emphasised that the reaction (2) proceeds on the electrode surface, whereas the reactions (3) and (4) take place in solution. In 0.25 ¡¾ 1 M NaHCO3 solutions, the solvated electron can lose the `reducing' properties. Thus, it was found that Np(V) was oxidised to Np(VI) upon g irradiation of its aerated or argon-Redox reactions of actinides in carbonate and alkaline solutions saturated bicarbonate solutions. This oxidation did not take place in a solution of Na2CO3, whereas irradiation of Np(VI) led to its reduction.40 Disproportionation of Pu(V) was not examined.It was only noted that Pu(V), which was generated upon electrochemical reduction of Pu(VI) in a bicarbonate solution, underwent dispro- portionation with a high rate and exclusively Pu(IV) was accumu- lated in the solution.23 An increase in pH slows down disproportionation of Pu(V) [like disproportionation of U(V)]. At pH 11.0, the process is completed in 6 ¡¾ 8 h. The hexa-, penta- and tetravalent plutonium ions were found in the equilibrium mixture, i.e., the following reaction took place (5) Pu(VI)+Pu(IV). 2 Pu(V) b. Disproportionation of Am(IV) Americium(IV), which was prepared in bicarbonate-carbonate solutions by electrolytic oxidation of Am(III), was reduced with water (the first-order reaction with respect to americium) and underwent disproportionation (second-order reaction).32, 33 (6) 4 Am(III)+O2 , 4 Am(IV)+4H2O (7) Am(III)+Am(V).Am(IV)+Am(IV) In a 1.3 M NaHCO3+Na2CO3 solution (pH 10), the activation energy of the second-order reaction in the temperature range of 30 ¡¾ 65 8C is 100.84.0 kJ mol71. In a 1.5 MNaHCO3+Na2CO3 solution, the rate constant for disproportionation of Am(IV) was measured at different pH by spectrophotometry.41 Americium(IV) was prepared by irradiating a solution of Am(III) containing the BrO7 ions with UV light. After termination of irradiation, Am(IV) was transformed sponta- neously into Am(III) and Am(V) according to the reactions (6) and (7) in a time varying from a few minutes to 3 ¡¾ 4 days (depending on pH).The conversion of Am(IV) into Am(III) was 50%¡¾ 65%. Taking into account the final proportion of Am(III), it can be concluded that the second-order reaction (7) prevailed. The rate constant of this reaction increased from 6.661072 to 262 litre mol71 s71 as pH was increased from 10.39 to 12.13. In a 1.5 M KHCO3+K2CO3 solution, 2kd increased from 6.661072 to 450 litre mol71 s71 as pH was increased from 8.40 to 11.70. Judging from the slopes of the plots of 2kd vs. pH (on the logarithmic coordinates), the disproportionation reaction of Am(IV) in KHCO3+K2CO3 and NaHCO3+Na2CO3 solutions is second order with respect to the OH7 ions. This is attributable to the fact that the first step involves hydrolysis of americium carbonate complexes: (8) Am(CO3)n(H2O)O2n¢§4U¢§+OH7 m Am(OH)(CO3)n(H2O)O2n¢§3U¢§+H2O, m¢§1 (9) 2 Am(OH)(CO3)n(H2O)O2n¢§3U¢§ m¢§1 Am(H2O)x(CO3)3¢§ 3 +AmO2(H2O)y(CO3)32 ¢§+zCO23 ¢§.If the concentration of K2CO3 in a KHCO3+K2CO3 solution is higher than 3 mol litre71, disproportionation of Am(IV) proceeds much more slowly and this reaction is very slow at a concentration of 4.0 ¡¾ 5.88 mol litre71. Apparently, this is associated with an increase in the number of the carbonate groups in the coordination sphere of Am(IV) resulting in a decrease in the potential of the Am(IV)/Am(III) pair and an increase in the potential of the Am(V)/Am(IV) pair. As a con- sequence, the equilibrium of the reaction (9) is sharply shifted to the left and Am(IV) becomes stable.c. Reproportionation of Np(IV)+Np(VI) The reaction (10) 2 Np(V) Np(IV)+Np(VI) 467 was examined in 1.7 ¡¾ 4.0 M K2CO3 solutions by spectrophotom- etry (l=490 nm).42 The concentrations of Np(VI) and Np(IV) are lowered according to the equation (11) ¢§daNpOVIUa a ¢§daNpOVIUa a k0[Np(VI)][Np(IV)]. dt dt The apparent rate constant of the reaction k0 remains virtually unchanged upon a twofold change in the concentrations of Np(VI) and Np(IV) and decreases by a factor of 4 as the K2CO3 content is increased from 1.71 to 4 mol litre71. The authors of the cited study 42 attributed this fact to an increase in the number of the carbonate groups in the complex Np(IV) ion [like in the Am(IV) ion, see above], which hinders charge transfer.In the temperature range of 52 ¡¾ 76.5 8C, the activation energy of the reaction (10) in K2CO3 solutions is 1032 kJ mol71, which is virtually equal to this value (104 kJ mol71) for the reproportionation reaction Np(IV)+Np(VI) in solutions of HClO4 and HNO3 . 4 In the presence of sulfate ions, the reaction (10) slows down. The addition of SO2¢§ (0.21 mol litre71) to a 2.08 M K2CO3 solution led to a decrease in k 0 from 1470 to 390 litre mol71 min71. Hence, the sulfate ion has a more sub- stantial effect on the reaction rate than the carbonate ion. This fact remains to be explained. 2. Reactions analogous to disproportionation and reproportionation a. Reactions of Am(III) with Am(VI) Judging from the potentials of the Am(VI)/Am(V) and Am(IV)/ Am(III) pairs in carbonate-bicarbonate solutions (see Table 1), Am(VI) can oxidise Am(III) K (12) Am(IV)+Am(V).Am(III)+Am(VI) Actually, Am(IV) and Am(V) are accumulated in solution when equimolar carbonate solutions containing tri- and hexava- lent americium (6.261074 M) are mixed in different propor- tions.30 If the difference between the formal potentials is 60 mV ([CO2¢§ 3 a=1 mol litre71), the rate constant K a aAmOIVUaaAmOVUa aAmOIIIUaaAmOVIUa Hence, four valence forms, viz., Am(III), Am(IV), Am(V) and must be equal to 10 (the experimental value is 11.21.5). Am(VI), can coexist in solutions with pH 9.5 ¡¾ 10.5. b. Reactions of Np(V) with Am(V) The reaction Np(VI)+Am(III) Np(V)+Am(V) was studied by spectrophotometry from the intensity of light absorption of Am(III) (l=508 nm) and Np(VI) (470 nm) in 0.45 ¡¾ 1.71 M Na2CO3 solutions.43 The Am(III) concentration was increased half as rapidly as the Np(VI) concentration.Hence, the following equation was proposed for this reaction:(13) 2 Np(V)+Am(V)=2 Np(VI)+Am(III), and a change in the concentrations of the reacting compounds is described by the equation (14) ¢§daAmOVUa a ¢§1 daNpOVUa a k0aAmOVUaaNpOVUa. dt 2 dt 3 The validity of the Eqn (14) is confirmed by the fact that the constant k0 (T=64 8C) is independent of the initial concentra- tions of Np(V) (0.65 ¡¾ 1.9 mmol litre71) and Am(V) (0.65 ¡¾ 1.3 mmol litre71). The reaction rate (the order with respect to the CO2¢§ ions n=71.53) decreases monotonically as the Na2CO3 content is increased.In the temperature range of468 3 50.5 ± 69.7 8C, the activation energy is 604 kJ mol71 regardless of the Na2CO3 concentration. Most likely, the process starts with dissociation of the Am(V) and Np(V) carbonate complexes [like in the case of disproportio- nation of U(V)]. Hence, we believe that at a constant ionic strength of solution (m), the reaction order with respect to CO2¡ would be more close to72. In this medium, pentavalent actinides can exist as tricarbonate complexes.1 In this case, it can be assumed that the following reactions proceed: (15) Am(V)O2(CO3)5¡ 3 Am(V)O2(CO3)3¡ 2 +CO23 ¡, (16) Np(V)O2(CO3)5¡ 3 Np(V)O2(CO3)3¡ 2 +CO23 ¡.Next, the following charge transfer processes take place (17) Z, Np(V)O2(CO3)3¡ 2 +Am(V)O2(CO3)32 ¡ slow Z (18) Np(VI)O2(CO3)x¡ n +Am(IV)(CO3)ym¡ , fast (19) Np(VI)+Am(III), Np(V)+Am(IV) where Z is an activated complex. c. Reactions of Am(III) with Np(VII) In a 2.82 M KHCO3+K2CO3 solution, the potential of the Am(IV)/Am(III) pair varies from 0.99 to 0.81 V as pH is increased from 8.9 to 11.2.34 In carbonate and bicarbonate media, the potentials of the Np(VII)/Np(VI) pair vary in the ranges which are apparently comparable with the range mentioned above or are even broader. Actually, the potential of this pair in a 1 M NaOH solution is 0.582 or 0.587 V (see Table 1) and increases by 0.118 V as pH is decreased by unity (m=1),16 i.e., this potential at pH 11 and pH 10 must be no lower than 0.9 V and no higher than 1 V, respectively.In carbonate solutions, these values may vary due to the formation of the Np(VI) and Np(VII) complexes with the carbonate ions, but these changes are insignificant. Hence, Np(VII) can serve as an oxidiser of Am(III) in bicarbonate and carbonate media. However, Np(VII) is reduced with water under the same conditions. The yield of Am(IV) depends on the ratio between the reaction rates (20) Np(VI)+Am(IV), Np(VII)+Am(III) (21) 4 Np(VI)+O2 . 4 Np(VII)+2H2O After the addition of an aliquot of an alkaline solution of neptunium(VII) to a 1.5 MKHCO3 solution of americium(III), the latter is oxidised already in the course of stirring (10 ± 30 s).44 The intensity of the absorption band of Am(III) carbonate complexes with a maximum at 507 nm decreases (in some cases, the band even disappears).The light absorption resulting from superposi- tion of broad bands belonging to Am(IV) and Np(VI) is observed at wavelengths shorter than 450 nm. When the dry salt K4Fe(CN)6 .3H2O was placed in a cell, light absorption at 350 ± 400 nm was decreased and the band with a maximum at *507 nm simultaneously regained its initial intensity. This unam- biguously indicates that Np(VII) oxidises Am(III) to Am(IV). The rate constant for the bimolecular reaction (20) is larger than 56103 litre mol71 s71.44 This value is many orders of magnitude larger than the rate constant of the reaction (21) for which the conversion of Np(VII) at pH 9.08 reaches 50% in 1.2 min.45 In the presence of a two- or threefold excess of Np(VII), Am(III) is also oxidised only to the tetravalent state.The same phenomena are observed in KHCO3+K2CO3 or NaHCO3+Na2CO3 mixtures at pH410.9. At higher pH (including 1.5 M K2CO3 or Na2CO3 solutions without additives of bicarbonate), the addition of a stoichiometric amount of Np(VII) leads to complete oxidation of Am(III) to Am(IV), which then disappears due to disproportionation [see the reaction (7)]. This is evidenced by the appearance of Am(III) in an amount half as large as the initial value. The reaction of Am(III) with Np(VII) V P Shilov, A B Yusov taken in a threefold excess afforded Am(VI), which is character- ised by light absorption at 450 ± 600 nm.Hence, Np(VII) oxidises not only Am(III) but also Am(V) (22) Am(VI)+Np(VI). Am(V)+Np(VII) The addition ofK4Fe(CN)6 .3H2Ocrystals to this solution led to a decrease in absorption at 450 ± 600 nm but the absorption of Am(III) was not observed, i.e., Am(VI) was reduced to Am(V). The addition of a solution of Np(VII) in 0.01 M NaOH to a solution of Am(III) in 6 M K2CO3 resulted in oxidation of approximately one-half of the initial amount of Am(III). How- ever, the absorption bands of Am(IV) and Am(VI) were not observed. The concentration of Am(III) was not increased upon the addition of K4Fe(CN)6 .3H2O. Hence, Am(III) is partially oxidised to Am(V) in this medium (23) Am(V)+2 Np(VI).Am(III)+2 Np(VII) d. Reactions of Am(IV) with Np(IV) or Pu(IV) In a 0.5 M NaHCO3+1 M Na2CO3 solution, americium(IV) oxidises neptunium(IV), which is taken in an equimolar amount, to neptunium(VI) in 1 ± 2 min.46 The processes starts with the reaction (24) Am(III)+Np(V) Am(IV)+Np(IV) followed by the more rapid reaction (19), because only Am(III), Np(VI) and Np(IV) are present in the solution after the disappear- ance of Am(IV), i.e., the process is described by the equation Np(IV)+2 Am(IV)=Np(VI)+2 Am(III). Compared to the reaction (24), oxidation of Pu(IV) with americium(IV) (25) Am(III)+Pu(V) Am(IV)+Pu(IV) proceeds more slowly.47 The subsequent reaction (26) Am(III)+Pu(VI) Am(IV)+Pu(V) must proceed faster. 3. Oxidation reactions a.Oxidation with molecular oxygen Oxidation of uranium(V) in a 1 MNa2CO3 solution and oxidation of uranium(IV) in a 1 M NaHCO3 solution with atmospheric oxygen were examined in the study. 5 In both cases, U(VI) was accumulated in the reaction products. The kinetic parameters of the reactions were not reported. The kinetics of U(IV) oxidation in carbonate-bicarbonate solutions was investigated. 48 When air was passed with a rate of 130 ml min71 through a 1 M NaHCO3+0.25 M Na2CO3 solution (40 ml) containing uranium (1.361072 mol litre71), U(VI) was formed with a rate of 1.361075 mol litre71 min71. A threefold decrease in the concentration of HCO¡3 +CO23 ¡ led to an increase in the rate of oxidation by *30%; the addition of copper ammoniate caused acceleration of oxidation.Plutonium(III), which was prepared by electrolysis of Pu(IV) in a 1 M K2CO3 solution, underwent oxidation with atmospheric oxygen within a few minutes after termination of electrolysis. The supply of atmospheric oxygen was increased on stirring of the solution and oxidation was accelerated.24 Dissolution of NpO2 crystals in an aerated 1 M NaHCO3 solution was examined.49 The rate of dissolution of NpO2 in this solution (210 ppm per day) proved to be much higher than that in water (2.5 ppm per day) or in a saturated NaCl solution (25 ppm per day). Neptunium(V) was detected in bicarbonate solutions, which indicates that NpO2 is oxidised under the action of O2 and HCO¡3 . b. Oxidation with hydrogen peroxide Hydrogen peroxide was also used for uranium oxidation (a suspension of a UO2 powder in 0.5 M NaHCO3 or NH4HCO3Redox reactions of actinides in carbonate and alkaline solutions 6 6 solutions, T=20 8C).50 To accelerate dissolution of UO2, Fe(CN)4¢§ and Fe(CN)3¢§ were added to the system.The propor- tion of uranium oxidised during the same period was approx- imately halved when pH was increased from 7.5 to 9, whereas a subsequent increase in pH has no effect on the oxidation rate. c. Oxidation with water Neptunium(III), which was prepared by an electrochemical method in a carbonate solution,12 was spontaneously oxidised with water, which hindered the measurement of its absorption spectrum. d. Oxidation with ozone In the study,51 solutions of Am(VI) and Am(V) were prepared with the use of ozone.When oxygen containing 5% O3 was passed through a 0.1 M NaHCO3 solution containing a suspension of Am(OH)3 during 1 h, Am(VI) was formed quantitatively at any temperature in the range from 0 to 90 8C. In a 2 M Na2CO3 solution, Am(III) is oxidised with ozone to Am(VI) at room and lower temperatures. In heated solutions, accumulation of Am(V) takes place. Americium(V) involved in double carbonate with sodium was readily oxidised to Am(VI) with ozone in NaHCO3 solutions at room temperature. However, ozone oxidised Am(OH)3 in 0.03 ¡¾ 0.1 MKHCO3 solutions to form Am(V), which precipitated as the salt KAmO2CO3 . 2 The kinetics of ozonation was not studied in detail.Never- theless, some conclusions can be made about the mechanisms of the reactions that take place. Experiments on oxidation of Pu(III) with ozone in aqueous 0.25 ¡¾ 1 M HClO4 solutions enriched with the 18O isotope demonstrated that water and ozone each supply one oxygen atom to form the PuO2a ion.52 Oxidation of U(IV) with ozone is also accompanied by the transfer of the oxygen atom from ozone (apparently, as an inner-sphere process).53 Oxidation of Am(III) with ozone may be schematically described by the following equations (without considering the complex formation with the carbonate ions): AmOa2 +O2+H2O, AmO2a 2 +O¢§3 , Am(IV)+O2 , 3 (27) (28) (29) (30) Am(III)+O3+2OH7 AmOa2 +O3 Am(III)+O¢§ Am(IV)+O3+2OH7 AmO2a 2 +O2+H2O.In carbonate solutions, Am(III) exists as the complex AmCOa3 , Am(CO3)¢§2 or Am(CO3)33 ¢§ ions depending on the composition of these solutions.54 Based on the two last-mentioned complex ions, the crystalline compounds NaAm(CO3)2 .4H2O and Na3Am(CO3)3 .3H2O containing water molecules in the coordination sphere of Am(III) were obtained.55 Oxidation with ozone is accompanied by the replacement of the water molecule in the coordination sphere of Am(III) followed by elimination of the oxygen molecule [see the reaction (27)]. This mechanism was confirmed by our experiments on oxidation of Am(III) with ozone in a 5 M K2CO3 solution. Apparently, an increase in the concentration of the carbonate ions leads to an increase in the number of the CO2¢§ 3 groups in the coordination sphere of Am(III) due to the displacement of the water molecules as a result of which the insertion of the ozone molecule is hindered and oxidation occurs as an outer-sphere process to form Am(IV).e. Oxidation with S2O2¢§ 8 ions Neptunium(V). The kinetics of Np(V) oxidation with the persulfate ions was studied by spectrophotometry from the change in absorption of Np(VI) at 470 nm.56Adecrease in the concentration of Np(V) is described by the following equation (31) ¢§daNpOVUa a2{k1+k2[Np(V)]}[S2O28¢§]. dt 469 It was found that two competitive reactions take place, viz., the first-order (rate constant k1) and second-order (rate constant k2) processes (the zero and first orders with respect to neptunium, respectively).Variation in the S2O2¢§ 8 concentration from 50 to 500 mmol litre71 ([Np(V)]=16 mmol litre71 at 50 8C) has no effect on the rate constants k1 and k2 for Np(V) oxidation in a 0.94 M Na2CO3 solution. As the Na2CO3 concentration is increased from 0.84 to 1.92 mol litre71 (T=50.4 8C), k1 decreases from 4.761075 to 2.361075 min71, whereas the constant k2 (k2=0.15 litre mol71 min71) remains unchanged. In a 0.93 M K2CO3 solution, the constants k1 and k2 remain unchanged as [Np(V)]0 is increased from 1.07 to 2.13 mmol litre71 and [S2O2¢§ 8 ]0 is increased from 30 to 150 mmol litre71, but an increase in theK2CO3 concentration from 0.93 to 4.50 mol litre71 (T=51.0 8C) is accompanied by a decrease in k1 and an increase in k2 .8 8 The activation energies of the first- and second-order reac- tions (E1 and E2, respectively) are 138 and 71 kJ mol71, respec- tively. The energy E1 is virtually equal to the activation energy of thermal decomposition of S2O2¢§ (140 kJ mol71).57 Hence, it can be assumed that the direct reaction of S2O2¢§ with Np(V) in carbonate solutions occurs simultaneously with thermal decom- position of S2O2¢§ 8 followed by the reaction of the SO¢§4 radical ions with Np(V). k1 , slow 4 , 8 fast SO2¢§ 4 +CO¢§3 , fast Np(VI)+CO2¢§ 3 , 3 (32) (33) (34) (35) S2O2¢§ 2SO¢§ SO¢§4 +CO23 ¢§ Np(V)+CO¢§ Np(V)+S2O2¢§ k2 , slow 8 Np(VI)+SO¢§4 +SO24 ¢§. Neptunium(IV). Oxidation of Np(IV) with persulfate ions was studied in 0.86 ¡¾ 3.30 M K2CO3 solutions by spectrophotometry from a change in absorption of Np(VI) (l=490 nm) and Np(IV) (l=800 nm) .58 Typical kinetic curves for a decrease in the amount of Np(IV) and accumulation of Np(VI) assume an S shape.The concentration of Np(V) (calculated) passes through a maximum, which is reached within 10 ¡¾ 15 min after the beginning of the reaction. 8 At constant temperature and concentration of K2CO3, the initial oxidation rate (V0) of neptunium(IV) is independent of its initial concentration. The rate V0 is comparable with the rate of thermal decomposition of the S2O2¢§ ions and is directly propor- tional to their concentration. Consequently, the process starts with the reactions (32) and (33) followed by the reaction (36) Np(V)+CO2¢§ Np(IV)+CO¢§ 3 .3 The rate constant for this reaction was estimated at 6.96103 litre mol71 s71. 59 As Np(V) is accumulated, the contri- bution of the reactions (34) and (35) becomes more and more noticeable. Neptunium(VI) reacts with neptunium(IV) according to the reaction (10). Neptunium(VI). In bicarbonate solutions containing Np(VI) (1.5 mmol litre71) and S2O2¢§ 8 (50 mmol litre71), Np(VI) was only partially oxidised at 60 8C in the presence of alkali metal (potas- sium, sodium or lithium) or ammonium cations.60 Accumulation of Np(VII) as MNpO4 took place within first 30 min after mixing of the reagents, whereas subsequent storage of these reagents for 120 min led to only a slight increase in the Np(VII) content. As the concentration of MHCO3 was increased from 16 to 46 mmol litre71, the yield of Np(VII) was increased from 17% to 44% in a KHCO3 medium (pH 7.30 ¡¾ 7.70) and from 18% to 29% in a NaHCO3 medium (pH 7.30).A further increase in the MHCO3 concentration led to a decrease in the yield of Np(VII). In aqueous solutions containing KHCO3 (93 mmol litre71) or NaHCO3 (46 mmol litre71), the formation of Np(VII) was not observed. In solutions of LiHCO3 and NH4HCO3, the maximum yields of Np(VII) were lower.470 The following scheme of the process can be proposed. Ther- 8 affords the SO¡4 radical ions [see the mal decomposition of S2O2¡ reaction (32)]. Then, the following reactions proceed: (37) SO4¡ +H2O SO4 2¡+OH.+H+, CO¡3 +SO24 ¡+H+ HCO¡3 +SO¡4 (38) with the rate constants of 60 and 9.16106 litre mol71 s71, respectively.61 3 4 At a bicarbonate concentration of 10 ± 100 mmol litre71, the SO¡4 radical anions are predominantly consumed in the reaction (38).However, it is known that the CO¡ radical anion cannot oxidise Np(VI) in a carbonate medium.62 Hence, either the SO¡ radical anion or the S2O2¡ 8 ion must serve as an oxidiser of Np(VI) (39) Np(VII)+SO2¡ Np(VI)+SO¡ 4 , 4 (40) Np(VI)+S2O2¡ 8 Np(VII)+SO¡4 +SO24 ¡. Neptunium(VII) is involved in two competitive processes. In one process, neptunium(VII) partially precipitates as the solid compound MNpO4 (see Ref. 63) and is stabilised. In another process, the remaining Np(VII) is involved in a carbonate complex and reduced with water.45 8V0=a+b[Pu(IV)].Plutonium(IV). The kinetics of Pu(IV) oxidation with the S2O2¡ ions in 1.07 ± 1.57 M Na2CO3 solutions was studied by spectrophotometry from a change in light absorption of Pu(IV) and Pu(VI).64 The initial oxidation rates of Pu(IV) to Pu(VI) linearly increased as the concentration of Pu(IV) was increased from 2 to 8 mmol litre71 according to the empirical dependence (41) It was assumed that two competitive processes of Pu(IV) 8 oxidation occurred. The rate V0 nonlinearly increased as the concentration of the S2O2¡ 8 ions was increased. Based on analysis of the experimental data, the following sequence of Pu(IV) oxidation with the S2O2¡ ions was proposed. First, the reactions (32) and (33) proceed followed by the reactions (42) Pu(V)+CO2¡ Pu(IV)+CO¡ 3 , 3 (43) Pu(VI)+CO2¡ Pu(V)+CO¡ 3 , 3 (44) products.CO¡3 +CO¡3 Simultaneously, Pu(IV) reacts with the S2O2¡ 8 ions that remain undecomposed (45) complex, Pu(IV)+S2O2¡ 8 (46) Complex Pu(V)+SO2¡ 4 +SO¡4 . 4 It was assumed 64 that Pu(IV) is oxidised directly with the SO¡ radicals. However, this reaction plays virtually no part in this medium. A change in the concentration of CO2¡ 3 has a substantial effect on the mechanism of the radical pathway of Pu(IV) oxidation and is only slightly reflected in intramolecular oxida- tion.In the temperature range of 70 ± 81.5 8C, the activation energies of radical and intramolecular oxidation are 1348 and 968 kJ mol71, respectively.Americium(III) and americium(V). Upon heating in a 0.1 M 8 NaHCO3 solution, americium(III) is oxidised under the action of Na2S2O8 to Am(VI).51 In a 2 M Na2CO3 solution in the presence of Na2S2O8 (0.01 mol litre71), americium(VI) is formed only at 90 8C.51 In a 0.1 M KHCO3+K2CO3 solution, americium(III) is oxidised under the action of the S2O2¡ ions to americium(V), which precipitates as the KAmO2CO3 complex. The kinetics of Am(III) oxidation with the persulfate ions was examined in solutions of K2CO3 (see Refs 65 and 66), (NH4)2CO3 and Na2CO3 (see Ref. 66).} The course of the reaction was followed from a change in light absorption of Am(III) } Semiquantitative data on the kinetics of this process published in earlier studies were surveyed in the review.65 V P Shilov, A B Yusov 8 3 (l=508 nm) and Am(VI) (l=688 nm65 or 550 nm66). The concentration of Am(III) was decreased with a constant rate and this change slowed down only toward the end of the process.Accumulation of Am(VI) started when only 10%± 20% of Am(III) remained in the solution. Until then Am(VI), apparently, disap- peared due to the involvement in the reaction (12) although the data on this process at high temperature are lacking. The higher was the CO2¡ concentration and the lower was the S2O2¡ concentration, the longer was the induction period. After com- pletion of the induction period, Am(VI) was accumulated in solution almost linearly. The rate of Am(III) oxidation is independent of the concen- trations of americium (in the range of 0.25 ± 1.58 mmol litre71) and K2CO3 and is directly proportional to the concentration of S2O2¡ 8 (12.5 ± 200 mmol litre71).65, 66 At the same time, the rate of Am(V) oxidation, i.e., the rate of Am(VI) accumulation, increases as the concentration of Am(V) is increased only to 0.6 mmol - litre71 and then remains constant.The formation of Am(VI) is accelerated as the concentration of S2O2¡ 8 is increased, whereas an increase in theK2CO3 content slows down the Am(VI) formation. The activation energy of Am(III) oxidation is equal to that of Am(V) oxidation. These values are 1438 kJ mol71 in a 1.9 M K2CO3 solution (T=58 ± 83 8C),65 1438 kJ mol71 in a 1.5 M Na2CO3 solution and 1598 kJ mol71 in a 1.9 M (NH4)2CO3 solution.66 8 The following sequence of Am(III) and Am(V) oxidation can be proposed.Thermal decomposition of S2O2¡ [see the reactions (32) and (33)] affords the SO¡4 and CO¡3 radical anions followed by the reactions (47) Am(IV)+CO2¡ Am(III)+CO¡ 3 3 (48) Am(VI)+CO2¡ Am(V)+CO¡ 3 , 3 3 . and the reaction (12). The rate constant of the reaction (47) is 2.56107 litre mol71 s71.67 To our knowledge, americium(IV) does not react with CO¡ and, hence, Am(V) may be formed only upon disproportionation of Am(IV). 3 3 The plot of the rate constant for An(V) oxidation (An=U, Np or Pu) with the CO¡ radical anions vs. the potential of the An(VI)/An(V) pair yields a straight line on the logarithmic coordinates, which is characteristic of the Marcus outer-sphere electron transfer.6 It is reasonable to suggest that oxidation of Am(V) will be described by an analogous plot.Extrapolation of the straight line to the potential of 0.92 V [i.e., to Am(V)] gave log k&6 (litre mol71 s71). This value is smaller than log k for the reaction (47) although the Am(VI)/Am(V) and Am(VI)/ Am(III) pairs have similar potentials (see Table 1) and oxidation of both ions occurs without the structural rearrangements. Appa- rently, the lower rate of Am(V) oxidation as compared to the rate of Am(III) oxidation is responsible for the occurrence of the induction period observed upon accumulation of Am(VI). Taking into account that kAm(III)[Am(III)]&kAm(V)[Am(V)], for oxidation of Am(III) and Am(V) with persulfate ions we have kAm(V)&0.1 kAm(III) .The results obtained with the use of this ratio agree with the results of extrapolation. Our data on radiolysis and photolysis of Am(III) and Am(V) also indicate that these ions differ in their ability to be oxidised in carbonate solutions. Apparently, the CO¡ radical anions interact with Am(III) and Am(V) through different mech- anisms. Oxidation of Am(V), like oxidation of U(V), Np(V) and Pu(V), proceeds through the outer-sphere charge transfer. By contrast, oxidation of Am(III) can be accompanied by hydrogen abstraction from the water molecule. It should be noted that oxidation of Am(III) with the OH. radical in HClO4 solutions proceeds through hydrogen abstraction from the water molecule involved in the hydration shell about the Am(III) ion.68Redox reactions of actinides in carbonate and alkaline solutions 4 A comparison of the calculated rates of formation of the SO¡ radical ions and oxidation of Am(III) shows that *40% of the radical oxidisers goes into the latter reaction.The fact that Am(III) oxidation is somewhat accelerated in solutions containing more than 2 mol litre71 of K2CO3 is indicative of the less significant role of side reactions rather than of the direct interaction of Am(III) with S2O2¡ 8 as has been suggested in the study. 66 f. Oxidation with silver compounds Americium(IV), which is unstable in acidic media, was stabilised in carbonate solutions.Americium(III) was subjected to the action of a mixture of AgNO3 and K2S2O8 . In this case, several oxidisers appeared in the solution 69 (49) Ag(I)+S2O2¡ 8 Ag(II)+SO¡4 +SO24 ¡, (50) Ag(III)+2SO2¡ Ag(I)+S2O2¡ 4 , 8 (51) Am(IV)+Ag(I). Am(III)+Ag(II) In addition, the reactions (33) and (47) took place. Solid AgO also oxidises Am(III) to Am(IV).69 Silver oxide was used also for oxidation of Pu(IV) in a 45%K2CO3 solution.10, 70 In the temperature range of 75 ± 90 8C, complete oxidation of plutonium occurred during 45 min. Apparently, the following reactions proceeded: (52) Pu(V)+Ag, Pu(IV)+Ag(I) (53) Pu(VI)+Ag, Pu(V)+Ag(I) along with the reaction (5). g. Oxidation with ClO7 ions It is known that macroamounts of Pu(IV) in a 45% K2CO3 solution are oxidised to Pu(VI) with the hypochlorite ions.70 When heated to an above-room temperature, the reaction was completed in 5 ± 10 min.The data on this process are scarce, which does not allow one to make conclusions about the reaction mechanism. h. Oxidation with xenon compounds The reactions of the americium ions with xenon compounds in bicarbonate and carbonate media were examined in the stud- ies.71 ± 73 Xenon compounds were chosen as oxidisers in view of the fact that the potentials of the Xe(VIII)/Xe(VI) and Xe(II)/Xe(0) pairs are higher than the potentials of the Am(IV)/Am(III) and Am(VI)/Am(V) pairs. The calculated potential of the XeF2/Xe pair in an acidic medium is 2.3 V.74 In an alkaline medium, the potential of this pair was not determined but it was found that xenon difluoride partially oxidises Pu(VI) to Pu(VII) in 0.2 ± 1.2 M NaOH solutions.75 In a 1 M NaOH solution, the potential of the latter pair is 0.85 V.Hence, XeF2 must oxidise Am(III). The addition of a XeF2 solution, which was cooled to 0 8C, to a 1.5 M KHCO3 solution of Am(III) led to oxidation of the latter to Am(IV) in the course of stirring, which was established by spectrophotometry. The stoichiometric coefficient n= D[Am(IV)]/D[XeF2] is 0.23.71, 72 It should be noted that this is yet another known procedure for the chemical preparation of Am(IV) in carbonate solutions (photo- and electrochemical methods are not considered). Upon the addition of dry K4Fe(CN)6 .3H2O to the resulting solution, americium reverted to Am(III).In the presence of a tenfold excess of XeF2, americium(III) was com- pletely oxidised to Am(IV). Other valence states were not found. In a KHCO3+K2CO3 solution, the coefficient n decreased as the K2CO3 content was increased. In a 1.5 M K2CO3 solution, n<0.01. In this medium, Am(III) was oxidised by at most 5%. In more concentrated K2CO3 solutions, americium(III) was not oxidised. In aqueous solutions, not only does XeF2 react with reducing agents but it also undergoes hydrolysis. The scheme of hydrolysis was proposed in the studies 76, 77 and was thereafter substanti- ated.78 Taking into account this scheme and the reactions of 471 Am(III) and Am(IV), the following sequence of processes occur- ring in the course of Am(III) oxidation can be suggested.The reactions (54) XeO+2 HF, XeF2+H2O (55) XeO+H2O Xe+H2 O2 , (56) XeO+H2O2 Xe+OH.+HO2 , HO (57) 2+OH7 O¡2 +H2O, (58) XeF.+O2+F7, XeF2+O¡2 (59) Xe+OH.+F7, XeF.+OH7 (60) OH.+HCO¡3 CO¡3 +H2O, (61) XeF.+HCO¡ Xe+F7+HCO3 , 3 are followed by the reaction (47) and then by the reactions (62) Am(III)+HO Am(IV)+HO¡ 2 , 2 (63) Am(IV)+O¡ Am(III)+O2 . 2 The direct reactions of Am(III) with xenon compounds are also not ruled out. (64) Am(IV)+XeF.+F7, Am(III)+XeF2 (65) Am(IV)+Xe+F7. Am(III)+XeF. Oxidation of Am(III) under the action of Na4XeO6 in carbo- nate solutions was examined in the studies.71, 73 In an alkaline medium, the potential of the Xe(VIII)/Xe(VI) pair is close to 0.9 V.79 This potential increases as pH is decreased and its values in a carbonate medium is higher than 1 V.Hence, the perxenate ion can oxidise americium(III) in bicarbonate and carbonate solutions. When Na4XeO6 . nH2O was added to a 0.75 M KHCO3+K2CO3 solution (pH 10) containing Am(III) (261074 mol litre71) until the concentration of the former reached 161074 mol litre71, the mixture turned yellow already in 1 min. Analysis of the solution for different valence states of americium [by spectrophotometry and by reduction with K4Fe(CN)6 .3H2O] revealed the presence of Am(IV), Am(V) and Am(VI). [Recall that XeF2 oxidises Am(III) only to Am(IV)]. The 7D[Am(III)]/D[Xe(VIII)] ratio varied from 0.3 to 0.7.In the presence of a 30-fold excess of perxenate ions, americium was quantitatively oxidised to the hexavalent state. The same situation was observed in a 1.5 M KHCO3 or K2CO3 solution. In a 5.9 M K2CO3 solution, perxenate ions decomposed and did not oxidise Am(III). As regards the mechanism of the reaction of Na4XeO6 with Am(III), the following conclusions can be made. If the process (66) CO2¡ 3 +Xe(VIII) CO¡3 +Xe(VII) 3 took place, the subsequent reaction (47) would afford only Am(IV), which does not react with the CO¡ radical anions. For instance, this situation was observed upon photochemical oxida- tion of Am(III) in the presence of S2O2¡ 8 , N2O, BrO7 or BrO¡3 at pH 8.5 ± 10.5.41, 80 ± 82 The appearance of Am(V) and Am(VI) in solutions, where disproportionation of Am(IV) proceeds very slowly, indicates that Xe(VIII) serves as an oxidiser of americium to the pentavalent state (67) Am(V)+Xe(VI), Am(III)+Xe(VIII) (68) Am(VI)+Xe(VII), Am(V)+Xe(VIII) (69) Am(VI)+Xe(VI).Am(IV)+Xe(VIII) Two-electron reactions can proceed, for example, through the transfer of the oxygen atom. Americium(IV) involved in the reaction (69) is generated in the reaction (12). Like Xe(VIII), xenon(VII) can apparently oxidise Am(III), Am(IV) and Am(V). In addition, it was suggested that Xe(VII) is involved in the following processes:83472 Xe(VII) . Xe(VIII), (70) Xe(VII)+Xe(VIII) Xe(VI)+Xe(IX), (71) Xe(VII) . Xe(VIII) (72) Xe(VIII)+Xe(VI)+O 2 Xe(IX) 2 The rate constant of the reaction (72) is 26107 litre mol71 s71.The reactions of Am(III) and Am(IV) with the perxenate ions can also proceed by a one-electron mechanism: (73) Am(IV)+Xe(VII), Am(III)+Xe(VIII) (74) Am(V)+Xe(VII). Am(IV)+Xe(VIII) The ability of Na4XeO6 to oxidise Am(IV) was proved as follows. Trivalent americium (0.13 mmol litre71) in a 1.5 M KHCO3+K2CO3 solution was quantitatively oxidised to Am(IV) under the action of XeO3 with the use of the photo- chemical method.71 The addition of perxenate ions to the solution led to the rapid formation of Am(V) and Am(VI). The stoichio- metric coefficient 7D[Am(IV)]/D[Xe(VIII)] was 0.2. The potential of the XeO3/Xe pair was estimated 84 at 2.10 and 1.24 V in acidic and alkaline solutions, respectively. Hence, XeO3 could oxidise Am(III) in carbonate solutions.However, we exper- imentally established71 that this reaction proceeded only under UV or g irradiation. i. b Decay as an oxidative process Attempts to prepare curium(VI) by chemical methods failed due to the high oxidation potential of the Cm(IV)/Cm(III) pair and, presumably, low stability of curium(V). At the same time, there is reason to believe that Cm(VI) in carbonate solutions must be rather stable. Hence, another procedure for the preparation of Cm(VI) was chosen. This procedure makes use of b decay of the 242Am isotope (half-life is 16.07 h). The compound K3AmO2(CO3)2, which was synthesised in the study,85 contains americium with the following isotopic composition: 92% 241Am, 1.4% 242Am and 6.6% 243Am.Before the synthesis, the starting americium specimen was purified from 242Cm that was accumu- lated. The resulting compound was kept over a period from 18 to 40 h and then dissolved in a 0.1 M NaHCO3 solution upon ozone bubbling to oxidise Am(V) to Am(VI). Under these conditions, curium(III) was not oxidised. It was expected that the daughter curium-242 isotope, which was generated via b decay of the 242Am isotope in the course of storage of K3AmO2(CO3)2, would exist as the `hot' ion, which has the parent composition but whose charge is larger by unity, i.e., b decay AmOa2 CmO2a 2 +e . The estimation, which was carried out taking into consider- ation all nuclear-physics processes attendant on b decay of the 2 2 2 242Am isotope, demonstrated that up to 60% ¡¾70% of the daughter 242Cm isotope must be stabilised as the CmO2a ions.The latter can be found after dissolution ofK3AmO2(CO3)2 . With the aim of establishing the valence state of 242Cm, K4UO2(CO3)3 was precipitated from the resulting solution. The precipitate captured 96%¡¾99% of AmOa and AmO2a ions and only 3%¡¾ 5% of Cm(III) ions. When the solution was in contact with freshly precipitated LaF3, 97%¡¾ 99% of the Cm(III) ions and no more than 3% of the Am(VI) ions were sorbed. Under the same conditions, 92%¡¾ 95% of the NpOa2 ions (analogues of CmOa2 ) were sorbed. Experiments demonstrated that curium-242 was concentrated either with penta- and hexavalent actinides [in the K4UO2(CO3)3 precipitate] or only with hexavalent actinides (in a solution over the LaF3 precipitate). Hence it follows that 242Cm differs in chemical properties from curium(III), i.e., 242Cm was obtained in the hexavalent state.This procedure was used also for the preparation of califor- nium(V).86 In carbonate solutions, the 249Cf isotope, which is generated upon b decay of tetravalent berkelium-249, exists as Cf(V) sorbed on the Na4UO2(CO3)3 precipitate. V P Shilov, A B Yusov 4. Reduction reactions a. Reduction with water Neptunium(VII). The gradual transformation of Np(VII) into Np(VI) in carbonate solutions was studied by spectrophotometry from a change in the intensity of absorption at l=460 nm.45 For a 0.5 M KHCO3+K2CO3 solution with m=1, the change in the Np(VII) concentration is described by the equation (75) ¢§daNpOVIIUa a kaNpOVIIUa.dt At 25 8C, the constant k decreases from 56.461072 to 0.7961072 min71 as pH is increased from 9.08 to 11.03. The plot of logk vs. pHyields a straight line with a slope of70.98, i.e., this reaction is first order with respect to the hydrogen ions (or reciprocal first order with respect to the OH7 ions). In the temperature range of 25 ¡¾ 35 8C, the activation energy of reduc- tion of Np(VII) to Np(VI) was estimated 45 at 33 kJ mol71 (pH 10.0). Instability of Np(VII) in carbonate solutions results from the high oxidation potential of the Np(VII)/Np(VI) pair (51 V; see Section III.2.c). Under these conditions, the potential of oxygen liberation from water is 0.64 V.It is commonly assumed that the first step of H2O decom- position under the action of strong oxidisers involves hydrogen abstraction to form the OH. radical. However, the oxidation potential of the OH./OH7 pair at pH 10 is 2.1 V. Due to the large difference in the potentials of the above pairs, reduction of Np(VII) with water cannot proceed through the formation of the OH. radical. However, another reaction, viz., oxidation of water to H2O2 is also thermodynamically possible. The standard poten- tial of the H2O2/H2O pair is 1.77 V. In carbonate solutions, this potential is lowered to 1.1 ¡¾ 1.2 V. For the reaction to follow this pathway, two Np(VII) ions must be simultaneously involved in this reaction.However, this reaction is first order with respect to neptunium. To resolve this contradiction, it was suggested 87, 88 that the thermally excited Np(VII) ion and the nonexcited Np(VII) ion form a dimer through two OH groups. This dimer subse- quently decomposes into two Np(VI) ions and the H2O2 water molecule. Hydrogen peroxide reduces two more Np(VII) ions. Americium(VI) and americium(IV). The overall equation for reduction of Am(VI) with water in a 1.3 M Na2CO3 solution (pH 11.25) has the form (76) 4 Am(VI)+4OH7=4 Am(V)+O2+2H2O (taking into account the high pH of the medium, the hydroxy ions rather than water molecules formally serve as a reducing agent). At 25 8C, the rate constant for the first-order reaction was estimated at (0.140.01)61074 s71.In the temperature range of 25 ¡¾ 75 8C, the activation energy is 63.84 kJ mol71 (see Refs 32 and 33). In a 1.3 M NaHCO3+ Na2 CO3 solution (pH 10), the rate constant (at 30 8C) for reduction of Am(IV) with water [see the reaction (6)] is (0.60.3)61074 s71. In the temperature range of 30 ¡¾ 65 8C, the activation energy is 96.612.0 kJ mol71 (see Ref. 33). Probably, the mechanisms of Am(VI) and Am(IV) reduction with water are analogous to the mechanism of Np(VII) reduction. b. Reduction with hydrogen peroxide Plutonium(VI). The reaction of Pu(VI) with H2O2 was studied in 0.005 ¡¾ 0.1 M NaHCO3 solutions by stopped-flow spectropho- tometry.89 The results of this study indicate that the following reaction takes place (77) 2 Pu(V)+O2+2H2O.2 Pu(VI)+H2O2+2OH7 It was established experimentally that Pu(VI) forms a complex with H2O2 . Thus, a 0.05 M NaHCO3 solution containing Pu(VI) (0.6 mmol litre71) was injected into a 0.05 M NaHCO3 solution containing H2O2 (3 mmol litre71). The rate constant for the formation of the Pu(VI) complex with H2O2 (k) was estimated atRedox reactions of actinides in carbonate and alkaline solutions (6.90.8)6103 litre mol71 s71. Doubling of the initial H2O2 concentration led to an increase in the constant k to (7.20.7)6103 litre mol71 s71. The rate of disappearance of the complex cannot be described by the first-order equation. Reduction began with the reactions 89 (78) complex, Pu(VI)+H2O2 (79) Pu(V)+HO Complex 2 .According to the scheme proposed in the study,89 the subse- quent step involves the reaction of Pu(VI) with HO2 followed by other reactions. However, rather high pH was not taken into account. Since the reaction (57) is efficient at pH 8.5, it is apparently followed by the processes (80) Pu(VI)+O¢§ Pu(V)+O2 , 2 (81) OH.+OH7+O2 , O¢§2 +H2O2 3 and then by the reactions (60) and (43). When Pu(VI) and H2O2 were taken at the initial concentra- tions of 0.11 and 11.2 mmol litre71, respectively, the apparent rate constant of the first-order reaction [see Eqn (79)] increased from 1.12 to 24.2 s71 as the concentration of HCO¢§ was decreased from 0.05 to 0.005 mol litre71. An increase in the concentration of H2O2 in a 0.005 M NaHCO3 solution from 11 to 22 mmol litre71 led to an increase in the first-order rate constant from 24 to 34 s71.c. Reduction with Fe(CN)4¢§ 6 ions Neptunium(VI). Reduction of Np(VI) with the Fe(CN)4¢§ 6 ions was studied in aqueous solutions of Na2CO3 ([Na2CO3]40.05 mmol litre71) by the stopped-flow method.90 For the process in a 0.05 M Na2CO3 solution, DaFeOCNU6 4¢§a a 1:01 0:01, DaNpOVIUa which corresponds to the reaction (82) Np(V)+Fe(CN)3¢§ Np(VI)+Fe(CN)4¢§ 6 . 6 The kinetics of the reaction is described by the empirical equation (83) ¢§daNpOVIUa a k0aNpOVIUaaFeOCNU4¢§ dt 6 a , where k0 =(a+b[CO2¢§ 3 ])71. In a 0.004 M Na2CO3 solution (m was maintained at 0.2 by adding NaCl) k 0=2.86103 litre mol71 s71 (at 25 8C). At a constant concentration of Na2CO3 (0.004 mol litre71), the change in the ionic strength from 0.1 to 1 had no effect on k 0 but the same increase in m in a 0.05 MNa2CO3 solution led to doubling of k 0.The mechanism of reduction involves the following reactions: k1 (84) A+CO2¢§ 3 , NpO2(CO3)4¢§ 3 +Fe(CN)46 ¢§ k71 k2 (85) products, A where A is an intermediate formed upon the replacement of one CO2¢§ 3 group in NpO2(CO3)43 ¢§ by the Fe(CN)46¢§ ion. Plutonium(VI). The reaction of Pu(VI) with the Fe(CN)4¢§ 6 ions proceeds 91 analogously to the reaction of Np(VI) with these ions considered above kf (86) Pu(V)+Fe(CN)3¢§ Pu(VI)+Fe(CN)4¢§ 6 . 6 kr In a 0.05 M Na2CO3 solution (25 8C), kf=121420 and kr=(3.640.20)6104 litre mol71 s71 for the forward (f) and reverse (r) reactions, respectively.The empirical dependence of the 473 rate constant for the forward reaction on the concentration of the carbonate ions takes the form lnkf=lna+n ln[CO2¢§ 3 ]. 3 (87) At 25 8C, ln a=4.740.11 and n=0.800.03 for m=0.2 (NaCl), [Pu(VI)]=0.2 mmol litre71, [Fe(CN)6]47=0.4 mmol litre71. The rate constant kr is virtually independent of the CO2¢§ concentration. The reactions (82) and (86) have much in common. However, it is not improbable 91 that these reactions proceed through different mechanisms. In particular, the characteristics of the reaction (82) are inconsistent with the mechanism of the outer- sphere electron transfer, which seemed to be most probable. 6 6 Americium(IV) and americium(VI).The reactions of Am(IV) and Am(VI) with the Fe(CN)4¢§ ions were not studied in detail. Nevertheless, it is known that these reactions in bicarbonate- carbonate media afford Am(III) and Am(V), respectively. Amer- icium(V) is resistant to the hexacyanoferrate(II) ions. As men- tioned above, the ratio between Am(IV) and Am(VI) in the initial mixture can be precisely established with the use of reduction with the Fe(CN)4¢§ ions followed by the determination of the reaction products by spectrophotometry. d. Reduction with the [Tc(DMPE)3]SO3CF3 and [Re(DMPE)3]SO3CF3 compounds The [M(DMPE)3]SO3CF3 compounds [M=Tc, Re, DMPE is 1,2-bis(dimethylphosphino)ethane] serve as outer-sphere reduc- ing agents. These compounds were used in studies of the reactions with Np(VI) and Pu(VI) in carbonate solutions.92 The equilibrium constants of the reactions [An(VI)O2(CO3)3]47+[M(I)(DMPE)3]+ [An(V)O2(CO3)3]57+[M(II)(DMPE)3]2+ (T=25 8C, m=0.3, [CO2¢§ 3 ]=0.05 mol litre71) were calculated.The rate constants and the activation energies of these reactions were determined. The rate constants of the self-exchange reactions for the Np(VI)/Np(V) and Pu(VI)/Pu(V) pairs, which were calcu- lated according to the Marcus equation, are 0.69 and 3.0 litre mol71 s71, respectively, when Tc(DMPE)a3 was used as a reducing agent and are 0.64 and 8.9 litre mol71 s71, respec- tively, when Re(DMPE)a3 was used as a reducing agent. IV. Reactions in alkaline media 1. Disproportionation and reproportionation reactions f f a.Disproportionation of Np(VI) and Am(VI) The necessary condition for self-oxidation ¡¾ self-reduction reac- tions of hexavalent actinides is the similarity of the formal potentials Ef for the VII/VI and VI/V pairs. In the case of neptunium and americium, the potentials of these pairs in 1 M NaOH solutions differ by *0.4 V. An increase in the alkali concentration leads to a sharp decrease in the potentials EVII=VI, whereas EVI=V are changed only slightly. Hence, penta-, hexa- and heptavalent neptunium or americium can coexist in 10 M and more concentrated NaOH solutions. Upon rapid 10 ¡¾ 20-fold dilution with a 17 M NaOH solution, a colourless 561073 M neptunium(VI) solution turned green, which is characteristic of Np(VII).93 The visible regions of the absorption spectra of these solutions, which were measured after the removal of a suspension of air bubbles, were similar to the absorption spectrum of Np(VII).The difference was observed only in the short-wavelength region [due to the presence of Np(VI) in the solution]. The equilibrium (88) Np(VII)+Np(V) 2 Np(VI) was established at room temperature in less than 1 min. The position of the equilibrium depends substantially on the NaOH concentration. The determined equilibrium constants474 K a aNpOVIIUaaNpOVUa aNpOVIUa2 in solutions with different compositions are given in Table 2. Disproportionation of Np(VI) was studied also in hot NaOH solutions.94 Under these conditions, the solubilities of neptunates are much higher as compared to those in the cold.In addition, the viscocity of alkaline solutions decreases as the temperature is increased due to which the formation of a suspension of air bubbles ceases. The determined equilibrium constants are well reproducible. Table 2. Equilibrium concentrations of Np(VII) and equilibrium constants K for disproportionation reactions of Np(VI) in alkaline solutions (T=21 ¡¾ 23 8C).93 102 K 104 [Np(V)] /mol litre71 105 [Np(VII)]p /mol litre71 104 [Np(VI)]0 /mol litre71 [NaOH] /mol litre71 0.017 0.16 0.44 0.8 2.4 6.8 8.0 13.2 2.2 3.4 2.5 1.2 3.5 1.4 3.8 2.8 7.8 4.3 5.1 1.8 1.5 0.45 9.52 9.52 2.38 5.00 2.38 4.55 2.38 2.38 2.38 2.38 2.38 8.7 10.7 11.8 12.4 13.6 14.8 15.3 17.2 13.6 13.6 13.6 777777770.33 0.84 2.50 Interestingly, the equilibrium constants are virtually inde- pendent of the temperature under the conditions studied. Con- sequently, the activation energies of Np(VI) disproportionation and Np(VII) reduction with pentavalent neptunium must be equal within the experimental error.When studying disproportionation of actinides in aqueous solutions of KOH and CsOH and in aqueous-methanolic solu- tions of KOH, Tananaev 95 found that the nature of alkali is of little importance, whereas a decrease in the water content is favourable for disproportionation. For example, disproportiona- tion in a solution containing 60 vol.% of methanol proceeds to a noticeable extend even at [KOH]=4 mol litre71.Disproportionation of Am(VI) was examined in the study.96A 18 M NaOH solution (2.5 ¡¾ 3.0 ml) cooled to 0 8C and a cooled weakly acidic 0.03 M americium(VI) solution (0.2 ml) were mixed in a quartz cell and the absorption spectrum of the mixture was recorded. This spectrum had a band with a maximum at 740 ¡¾ 750 nm, which is characteristic of Am(VII). Since there were no other oxidisers in the solution, the formation of Am(VII) was attributed to self-oxidation ¡¾ self-reduction of Am(VI) (89) Am(VII)+Am(V). 2 Am(VI) Table 3. Equilibrium constants for disproportionation reaction of Am(VI) in NaOH solutions at 0 ¡¾ 3 8C. Equilibrium concentration K /mol litre71 [NaOH] 103 [Am(VI)]0 /mol litre71 /mol litre71 Am(VII) Am(V) Am(VI) 1.74 8.5 7 7 7 not found 0.20 0.34 0.80 0.45 0.61 1.21 1.36 0.70 0.47 0.17 0.20 0.34 0.80 0.45 0.61 1.61 2.04 2.30 1.37 1.39 12.4 14.2 15.4 16.6 17.2 2.761072 6.261072 1.3 0.9 13 V P Shilov, A B Yusov The estimated equilibrium constants of this reaction are given in Table 3.Americium(VI) is somewhat more prone to disproportiona- tion than neptunium(VI). In both cases, the NaOH concentration has a similar effect on the equilibrium of the reaction. b. Disproportionation of Pu(V) Comparison of the potentials of the Pu(V)/Pu(IV) and Pu(VI)/ Pu(V) pairs shows that Pu(V) must undergo disproportionation in a wide range of alkali concentrations.The kinetics of this process was studied by spectrophotometry.97 In experiments, Pu(V) was prepared predominantly by reducing Pu(VI) with H2O2 in an alkaline medium H2O2+OH7 HO¢§2 +H2O, (90) 2 2 PuO2(OH)2¢§ 4 +HO¢§ 2 PuO2(OH)2¢§ 3 +O2+OH7+H2O. Then the accumulation of Pu(VI) in the solution was tracked. When precipitating, plutonium(IV) captured a portion of Pu(V). Taking into account the balance of all valence states of plutonium in the solution and in the precipitate [without considering reduc- tion of Pu(VI) with a-radiolysis products of water], it was concluded that the following reaction proceeds in the solution:(5) Pu(VI)+Pu(IV). 2 Pu(V) 4 In alkaline media, Pu(V) exists as the PuO2(OH)3(H2O)27 and PuO2(OH)3¢§ ions, freshly precipitated Pu(IV) exists as Pu(OH)4 (more precisely, as PuO2 .nH2O) and Pu(VI) occurs as PuO2(OH)2¢§ 4 (see Ref. 2). Hence, the reaction (5) can be described as follows: (91) 2 PuO2(OH)3(H2O)27 PuO2(OH)2¢§ 4 +PuO2 . nH2O+mOH7. The constant 2k was estimated from the initial rate of disproportionation of Pu(V) V (92) 0 a ¢§daPuOVUa a 2kaPuOVUa2. dt It was found that the rate constant sharply increased as the NaOH concentration was decreased from 7.6 to 0.5 mol litre71. On the logarithmic coordinates, the plot of 2k vs. the activity of the OH7 ions yields a straight line with a slope of 72.8. However, taking into account that the activity coefficient of Pu(V) decreases on going from a 0.5 M NaOH solution to a 7.6 M solution, it can be assumed that the reaction order with respect to the OH7 ions is close to 72.Consequently, the fast step preceding the formation of an activated complex, proceeds with the replacement of the OH group in the Pu(V)-containing PuO2(OH)3(H2O)27 ion by the H2O molecule PuO2(OH)3(H2O)27+H2O PuO2(OH)2(H2O)¢§2 +OH7, (93) i.e., as expected, lower-charged species are involved in the reac- tion.The constant 2k increases as the temperature is increased. Thus, an increase in the temperature of a 3.3 M NaOH solution from 10 to 30 8C leads to an increase in 2k from 0.5161072 to 6.1561072 litre mol71 s71. The activation energy is approxi- mately 88 kJ mol71. The results of experiments with Np(V) in 0.5 ¡¾ 1 M LiOH, NaOH, KOH and CsOH solutions at 25 ¡¾ 80 8C demonstrated that Np(V) remained unchanged in the conditions under study.c. Reproportionation Np(VII)+Np(V) and Pu(VII)+Pu(V) As mentioned above, the potential of the Np(VII)/Np(VI) pair sharply increases at the alkali concentration of 410 mol litre71 and the reaction (88) goes in the reverse directionRedox reactions of actinides in carbonate and alkaline solutions (94) 2 Np(VI). Np(VII)+Np(V) An analogous reaction proceeded also with plutonium (95) 2 Pu(VI), Pu(VII)+Pu(V) because the potential of the Pu(VII)/Pu(VI) pair is substantially higher than the potential of the Pu(VI)/Pu(V) pair in a wide range of alkali concentrations. The kinetics of the reaction (94) in a LiOH solution was studied by pulse radiolysis.98 It was estab- lished that the rate constant of the reaction is independent of the alkali concentration in the range of 0.033 ¡¾ 2.0 mol litre71 and its average value in a LiOH solution is (2.30.9)6107 litre mol71 s71.In NaOH solutions, the rate constants of the reaction are smaller. Apparently, the reaction (95) also proceeds with a high rate. d. Reproportionation Np(VI)+Np(IV) and Pu(VI)+Pu(IV) The reaction (10) 2 Np(V) Np(IV)+Np(VI) in deaerated 0.025 ¡¾ 4.0 M LiOH solutions was examined in the study.18 The reaction was completed in the course of stirring of a suspension of Np(IV) hydroxide in a LiOH solution with a Np(VI) solution in the same alkali. The formal potential of the Np(V)/Np(IV) pair was estimated from the equilibrium constants.The addition of an equimolar amount of Pu(VI) to a 4 M NaOH solution containing Pu(IV) hydroxide led to the rapid formation of Pu(V),25 i.e., the following reaction proceeded: (96) 2 Pu(V). Pu(VI)+Pu(IV) 2. Reactions analogous to disproportionation and reproportionation a. Reactions of Np(VI) with Pu(VII) and Pu(VI) The reaction (97) Np(VII)+Pu(VI) Np(VI)+Pu(VII) proceeds rapidly at room temperature.99 This reaction is irrever- sible at any concentrations of alkali because Pu(VII) is a stronger oxidiser than Np(VII) and the potentials of the Pu(VII)/Pu(VI) and Np(VII)/Np(VI) pairs change in parallel with the OH7 concen- tration. In alkaline solutions, Np(VI) can be oxidised not only with heptavalent but also with hexavalent plutonium.28 The reaction (98) Np(VII)+Pu(V) Np(VI)+Pu(VI)2 is reversible and proceeds with a high rate, which indicates that the ions of penta-, hexa- and heptavalent actinides in concentrated alkaline solutions have similar structures.According to the recent EXAFS study,100 neptunium(VII) in alkaline solutions is involved in the NpO4(OH)3¢§ ion. Hence, neptunium(VII), like Np(V) and Np(VI), is in a tetragonal ¡¾ bipyramidal environment formed by oxygen atoms due to which the electron transfer is not accom- panied by a substantial structural rearrangement of the ions involved in the reaction. The position of the equilibrium depends on the alkali concentration. In a 7.8 M NaOH solution (at 24 8C), the equilibrium constant K a aNpOVIIUaaPuOVUa aNpOVIUaaPuOVIUa is 1.861072; its value increases to 1461072 as the NaOH concentration is increased to 10.1 mol litre71.b. Reactions of Np(VI) with Am(VI) and Am(VII) Neptunium(VII) is rapidly formed upon the addition of a bicar- bonate or weakly acidic Am(VI) solution to a 1 MNaOH solution containing Np(VI) (see Ref. 36) (99) Np(VII)+Am(V). Np(VI)+Am(VI) 475 The position of the reaction equilibrium depends on the concentration of the OH7 ions. Oxidation of Np(VI) starts at [OH7]&0.016 mol litre71. The yield of Np(VII) increases as the alkali concentration is increased and it reaches 100% in 3 ¡¾ 4 M NaOH solutions. Americium(V) does not react with Np(VI) in alkaline solutions.The addition of a cooled freshly prepared (by ozonation) Am(VII) solution in 3 ¡¾ 4 M NaOH to an alkaline solution of Np(VI) rapidly affords Np(VII) 101, 102 (100) 2 Np(VII)+Am(V). 2 Np(VI)+Am(VII) c. Reactions of Pu(VI) with Am(VI) and Am(VII) The equilibrium of the reaction (101) Pu(VII)+Am(V) Pu(VI)+Am(VI) is established immediately after mixing the reagents,36 but is is noticeably shifted to the right only at an alkali concentration higher than 7.5 mol litre71. Plutonium(VII) is virtually immedi- ately formed upon the addition of freshly prepared (by ozonation) Am(VII) to a 1 M NaOH solution containing Pu(VI)101, 102 (102) Pu(VII)+Am(VI). Pu(VI)+Am(VII) 3. Oxidation reactions a. Oxidation with molecular oxygen Neptunium(V) and plutonium(V).Oxygen bubbling through 0.5 ¡¾ 2 M LiOH or NaOH solutions containing neptunium(V) for 24 h did not yield Np(VI). As the alkali concentration and the reaction time were increased, Np(V) was slowly transformed into Np(VI) under the action of both pure O2 and atmospheric oxy- gen.103 In an aerated 8 M NaOH solution containing Np(V) (1 mmol litre71),*10% of Np(V) was oxidised during 3 months. In the same solution containing F7 (0.1 mol litre71), *50% of Np(V) was oxidised during one month. A change in the temper- ature of the solution from 0 to 50 8C exerts only a slight effect on the kinetics of Np(V) oxidation. Plutonium(V) is also oxidised with atmospheric oxygen in 4 M and 8 M NaOH solutions. In the latter case, the rate constant of the reaction (1.461073 h71) is smaller than the rate constant of Np(V) oxidation (4.761073 h71), whereas plutonium(V) in a 4 M NaOH solution is oxidised faster than Np(V).This is, apparently, associated with a change in the reaction mechanism. Most likely, oxygen reacts with Pu(IV) that is generated through disproportio- nation of Pu(V). Neptunium(IV) and plutonium(IV). A freshly precipitated sus- pension of neptunium(IV) hydroxide [we will denote this com- pound and plutonium(IV) hydroxide as NpO2 . nH2O and PuO2 . nH2O, respectively] is slowly oxidised with oxygen.104 At 23 8C, a suspension of NpO2 . nH2O (7 mmol litre71) in 2¡¾14 M NaOH solutions is completely oxidised during *1 ¡¾ 2 days. In addition to neptunium(V), Np(VI) is formed in substantial amounts. As the alkali concentration is increased, the reaction slightly slows down and the proportion of Np(VI) is decreased.It is believed that oxygen is adsorbed on the surface of Np(IV) hydroxide species followed by successive or synchronous electron transfer. The intermediate that formed in the course of O2 reduction (hydrogen peroxide) is not accumulated in solution. A suspension of PuO2 . nH2O in 0.5 ¡¾ 4 M NaOH solutions does not react with oxygen. Prolonged air bubbling through a 5 M NaOH solution containing this suspension affords different ionic forms of plutonium whose concentration is 10% higher than that obtained upon argon bubbling.104 b. Oxidation with hydrogen peroxide Neptunium(IV).The addition of H2O2 (100 mmol litre71) to a 0.5 M NaOH solution containing a NpO2 . nH2O suspension (10 mmol litre71) at 20 8C gave rise to a dissolved Np(V) peroxide complex. Oxidation of Np(IV) was completed in less than 10 min. Then Np(V) gradually precipitated.105476 In NaOH 1 ¡¾ 16 M solutions, the formation of Np(V) starts once the reagents have been mixed but the reaction rate decreases as the NaOH concentration is increased. The Np(V) peroxide complex is stable over a long period and the solid phase does not appear for at least 3 h.105 The formation of Np(V) slows down as the H2O2 concen- tration is decreased. At [H2O2]=5 mmol litre71, 9% of Np(IV) remains in the precipitate after completion of the reaction. Taking into account that hydrogen peroxide in an alkaline medium exists as the HO¢§2 ion, it may be written 2NpO2OH+OH7+mH2O.(103) 2NpO2 . nH2O+HO¢§2 Decomposition of H2O2 proceeds as a competitive reaction. The formation of Np(V) (and precipitation) is accelerated as the temperature is increased from 20 to 60 8C. Plutonium(IV). When a 1 M NaOH solution containing a freshly precipitated PuO2 . nH2O suspension (1 mmol litre71) and H2O2 (10 mmol litre71) was kept at 20 8C for 7 h, the concentration of plutonium in the ionic form remained the same as that in a check experiment performed in the absence ofH2O2. In more concentrated alkaline solutions, the higher is the NaOH concentration, the higher is the rate of Pu(IV) oxidation.Thus in a 12 M NaOH solution containing PuO2 . nH2O (10 mmol litre71) and H2O2 (50 mmol litre71), the Pu(V) peroxide complex was detected in an amount of 0.4 mmol litre71 already after 8 min.105 Slow oxidation of Pu(IV) with hydrogen peroxide in a 1 M NaOH solution takes place upon the addition of a trivalent iron salt (to the concentration of 0.1 mol litre71). The addition of Co(II) and Cu(II) salts has a weaker effect on oxidation of Pu(IV) with hydrogen peroxide. No effect of Ni(II) was observed.105 Americium(III). Oxidation of Am(OH)3 to Am(IV) under the action of hydrogen peroxide was first presented in the paper.106 These data are confirmed by the results of a more detailed study. 107 An acidic solution of Am(III) containing H2O2 was neutralised with ammonia.The precipitate that formed [presum- ably, Am(III) hydroperoxide] was treated with a concentrated ammonia or KOH solution to obtain Am(IV) hydroxide. The process was accelerated upon heating to 90 8C. At this temper- ature, Am(IV) hydroxide was obtained in quantitative yield according to several procedures: �¢ by treatment of Am(OH)3 with aH2O2 andKOH solution; �¢ by treatment of Am(III) hydroperoxide with a H2O2 and KOH solution; �¢ by treatment of Am(III) hydroperoxide with a KOH solution. c. Oxidation with ozone Neptunium(VI) and neptunium(V). Neptunium(VII) was first pre- pared by oxidation of a suspension of sodium or potassium neptunates with ozone in 0.5 ¡¾ 3.5 M alkaline solutions.108, 109 The kinetics of the reaction of ozone with Np(VI) depends on the state of the latter.In the case of ozonation of neptunate suspen- sions in NaOH (or KOH) solutions, the rate of Np(VII) formation depends on the mode of ozone supply and the properties of the solid phase. Thus, the rate of Np(VI) oxidation is decreased upon prolonged storage of the precipitates. In the study,14 it was suggested that a suspension of Np(V) hydroxide in NaOH solutions be used in these reactions. In the course of ozonation, Np(V) is oxidised to soluble Np(VI), which is transformed into the heptavalent state. The maximum concentration of Np(VII) obtained in NaOH (or KOH) solutions is 6 ¡¾ 8 g litre71. The Np(VII) content of 30 g litrs achieved upon ozonation of neptunates in 2 ¡¾ 3 M LiOH solutions 110 due to high solubility of lithium neptunates.111 The kinetics of Np(VII) formation in alkaline solutions con- taining Np(VI) ions is determined by the rate of ozone supply and intensity of stirring, i.e., by diffusion factors.109, 112, 113 At low concentrations of the reagents, the reaction of Np(VI) with O3 is kinetically controlled.The accumulation of Np(VII) obeys the equation 112 V P Shilov, A B Yusov (104) daNpOVIIUa a kaNpOVIUaaOH¢§a0:5aO3a0:5. dt In the temperature range of 0 ¡¾ 20 8C, the activation energy is close to 20 kJ mol71.112 An analogous dependence of the rate of Np(VII) accumulation on the ozone concentration was found in the study. 113 In a solution with a constant LiOH concentration, the rate of oxida- tion (V) of Np(VI) (10 mmol litre71) at a constant rate of ozone supply remains unchanged, i.e., (105) V a daNpOVIIUa a const.dt The plot of V vs. the LiOH concentration yields a curve with a maximum at [OH7]& 1.5 ¡¾ 2 mol litre71. It was proposed that the OH. radical 112 or the O7. radical ion113 serve as a direct oxidiser of Np(VI). However, in our opinion, the O3¢§. radical ion 114 serves as an oxidiser. Thus as the LiOH concentration is lowered beyond 1 mol litre71, the rate of Np(VI) oxidation decreases due to acceleration of O¢§3 . decomposition giving rise to O¢§2 .Neptunium(VI) is oxidised with ozone even at pH 9 ¡¾ 10 although the reaction rate is noticeably lower.115 Moreover, heptavalent neptunium is partially generated upon ozonation of a NpO2OH suspension in water.116 The completeness of Np(VI) oxidation with ozone in alkaline solutions depends on its concentration. For example, Np(VI) was oxidised by 80% at its concentration of (2 ¡¾ 4)61075 mol litre71,112 whereas 98%¡¾ 100% of Np(VI) was oxidised at the concentration of 361073 mol litre71.113 Neptunium(IV).When a mixture of O2 and O3 (0.7 vol.%) is passed through 0.5 ¡¾ 1 M NaOH solutions containing a suspen- sion of NpO2 . nH2O (1 mmol litre71), neptunium(IV) is oxidised to Np(V) and Np(VI) already within minutes after the onset of the process and the suspension is dissolved. Further ozonation affords Np(VII). Accumulation of Np(VII) is accelerated as the alkali concentration is increased.In a 14 M NaOH solution, neptunium is quantitatively transformed into the heptavalent state already within 8 min after the onset of oxidation.104 Ozone can directly oxidise tetra-, penta- and hexavalent neptunium. However, in a strongly alkaline medium, ozone initially reacts with the OH7 ions to form the ozonide ions. In this case, the reactions may be described by the scheme 117 (106) O3+OH7 O¢§3 +OH., (107) OH.+OH7 O7.+H2O, (108) O¢§ O7.+O2 3 , or the scheme 118 (109) O2+HO¢§2 , (110) 2 O3¢§ +HO2. . (57) O¢§2 +H2O, (111) O3+OH7 O3+HO¢§ HO2.+OH7 O3+O¢§2 O¢§3 +O2 . The ozonide ions oxidise all valence forms of neptunium to the heptavalent state. The process starts with the reaction (112) NpO2(OH)+O2+OH7+mH2O. NpO2 .nH2O+O¢§3 n Neptunium(V) is oxidised both in the solid phase and after the formation of the soluble NpO2(OH)1¢§n complex (n=2 ¡¾ 3). In this case, the following reactions proceed (113) n NpO2(OH)1¢§n+O¢§3 +H2O NpO2(OH)2¢§ 4 +O2+(n72)OH7, (114), NpO4(OH)3¢§ 2 +O2+H2O NpO2(OH)2¢§ 4 +O¢§3 the rate constants are (2.10.3)6106 and (2.10.2)6105 litre mol71 s71, respectively.114Redox reactions of actinides in carbonate and alkaline solutions Accumulation of Np(VII) occurs after the disappearance of Np(IV) and Np(V). Plutonium(VI). Like Np(VII), heptavalent plutonium was first prepared by ozonation.108, 119 Presently, this procedure is com- monly used for the preparation of pure Pu(VII) solutions with different concentrations. The reaction of Pu(VI) with O3 in alkaline media proceeds analogously to the reaction of Np(VI).113 In homogeneous sol- utions, the reaction rate is determined by the diffusion factors.In the mode of steady ozone supply, the rate of this reaction is described by the zero-order equation (115) V a daPuOVIIUa a const. dt The completeness of oxidation of Pu(VI) (7.5 mmol litre71) in alkaline solutions ([NaOH]<1 mol litre71) depends on insta- bility of Pu(VII) whose reduction with water is accelerated as the concentration of the OH7 ions is decreased.120 Plutonium(IV). Oxidation of Pu(IV) with ozone was examined in the study.104 A suspension of PuO2 . nH2O (0.3 ¡¾ 0.5 mmol litre71) in a 0.5 ¡¾ 8 M NaOH solution was oxidised with a mixture of O2 and O3 (0.7 vol.%).The resulting Pu(VII) and Pu(VI) rapidly reacted with Pu(IV). Hence, accumulation of Pu(VII) becomes possible after the complete disappearance of Pu(IV). In a 1 M NaOH solution, Pu(VII) is formed most rapidly (84 min), whereas the time of this reaction increases to 300 min (and more) in an 8 M NaOH solution. The oxidation process involves the reactions (106) ¡¾ (108) and reactions analogous to (112) ¡¾ (114). Americium(VI). When a gas mixture containing O3 (20 ¡¾ 50 mg litre71) was passed through a cooled (0 8C) solution of americium(VI) in 3 ¡¾ 4 M NaOH during 30 ¡¾ 60 min, the colour and absorption spectrum of the solution were changed.101, 102 After the addition of the resulting freshly ozonated alkaline solution of americium to an excess of Pu(VI) in 1 M NaOH, plutonium immediately took on the colour characteristic of Pu(VII).The absorption spectrum of the resulting solution agrees satisfactorily with the absorption spectrum of pure Pu(VII). It is known 37 that Am(VI) does not oxidise Pu(VI) in solutions containing NaOH at a concentration lower than 3 mol litre71. Hence, the formation of Pu(VII) in the experiments described above indicates that americium(VI) is transformed into Am(VII) upon ozonation of its alkaline solutions. Heptavalent americium reacts with Pu(VI) according to Eqn (102). The yield of Am(VII) upon ozonation can be estimated from the amount of Pu(VII) formed.In different experiments, its value varied from 40% to 60%. The same results were obtained in the study of the stoichi- ometry of the reaction (100). Americium(III). Oxidation of Am(III) was examined in a wide pH range.121 Ozone oxidises Am(OH)3 to Am(VI) both in neutral and alkaline (1 M NaOH) media. In a 1 M NaOH solution, oxidation proceeds slowly. Therefore, it is necessary to lower the temperature to 0 8C for the enhancement of solubility of O3. In a 0.1 M NaOH solution (at 25 8C), Am(OH)3 is oxidised with a moderate rate to form Am(VI), which gives a soluble yellow complex. The absorption spectrum of this complex is identical to the spectrum of the solution, which was prepared by adding alkali to an acidic solution of AmO2a 2 .In water, Am(III) is oxidised by passing ozone for 1 ¡¾ 2 h (T=90 8C). In the course of ozonation, pH of the solution must be no lower than 5. Apparently, the mechanism of Am(III) oxidation in solutions with pH lower than 12 is analogous to the mechanism of oxidation in carbonate solutions. This mechanism involves the inner-sphere transfer of the oxygen atom from O3 as the first step. The general scheme of oxidation is as follows: (116) AmO2OH+O2+H2O, Am(OH)3+O3 (117) AmO2(OH)2+O¢§ AmO2OH+O3+OH7 3 , AmO2 . nH2O+O2+OH7, (118) Am(OH)3+O¢§3 +H2O 477 AmO2 . nH2O+O3 AmO2(OH)2+O2+(n71)H2O. (119) In alkaline solutions (pH>12), the reactions (106) ¡¾ (108), (118) and (119) and also the reaction (120) AmO2 . nH2O+O¢§3 AmO2OH+O2+(n71)H2O+OH7 are the major processes.Americium(V) is oxidised in both the solid phase and in the dissolved form existing as the AmO2(OH)1¢§n complex (n=2¡¾3) n (121) n AmO2(OH)O1¢§nU+O¢§3 +H2O AmO2(OH)2¢§ 4 +O2+(n72)OH7. At 0 8C, Am(VII) is formed in 3 ¡¾ 4 M NaOH solutions (122) Am(VII)+O2 . AmO2(OH)2¢§ 4 +O¢§3 8 d. Oxidation with S2O2¢§ 8 ions Neptunium(VI). The kinetics of Np(VI) oxidation was examined in 0.1 ¡¾ 0.7 M NaOH solutions at T>50 8C.122 The initial reaction rate did not depend on the Np(VI) concentration (up to 0.3 mmol litre71) and increased in directly proportion to the S2O2¢§ concentration (the latter was increased from 25 to 300 mmol litre71) (123) 2O28 ¢§a. daNpOVIIUa a ¢§ daNpOVIUa a kaS dt dt In a 0.2 M NaOH solution heated to 60 8C, the rate constant of the reaction is 261074 min71.At [OH7]=0.7 mol litre71 and m=1, the activation energy is 140 kJ mol71, which is consistent with the activation energy of thermal decomposition of the persulfate ions.57 8 The first step of Np(VI) oxidation with the persulfate ions in 0.2 ¡¾ 0.7 M NaOH solutions involves decomposition of the S2O2¢§ ions [reaction (32)] followed by the reaction (124) SO4 2¢§+OH.. SO¢§4 +OH7 The fast reactions (124), (107) and (108) afford the O¢§3 radical ions, which oxidise Np(VI) to Np(VII) according to the reaction (114). Calculations demonstrated that the rate of Np(VII) accumu- lation in a NaOH solution is three times lower than the rate of formation of the SO¢§4 radical ions. Consequently, decomposition of O¢§3 is of considerable importance in this process, as in the case of Np(VI) oxidation with ozone at [OH7]<1 mol litre71. The rate of Np(VII) accumulation decreases as the alkali concentration is increased to 0.9 mol litre71.The plot of Np(VII) accumulation in a 1 M NaOH solution (m=2) goes through a maximum. In 3 ¡¾ 4 M NaOH solutions, neptunium(VII) is not formed at all. 8 Apparently, decomposition of S2O2¢§ in alkaline solutions proceeds not only through thermal decomposition [reaction (32)] but also under the action of the OH7 ions, reducers being formed in the latter case. This pathway becomes dominant in 3 ¡¾ 4 M NaOH solutions. On a further increase in the alkali concentration to 5 ¡¾ 6 mol litre71, the formation of Np(VII) resumes.Under these conditions, the mechanism of decomposition of persulfate ions is, evidently, changed once again. An increase in the ionic strength of solution has virtually no effect on the rate of Np(VI) oxidation in a 0.2 M NaOH solution but the reaction rate is decreased in a 0.7 M NaOH solution. 8 Hydroxides of some d elements serve as catalysts of the Np(VII) formation. In a 4 M NaOH solution, the S2O2¢§ ions oxidise N(VI) even at 20 8C in the presence of Fe(III), Ag(I), Cu(II), Ni(II) or Co(II) salts in concentrations of 561075, 161075, 561076, 161077 or 261078 mol litre71, respec- tively. In alkaline solutions containing Ni(II) or Co(II) com-478 pounds, Np(VII) is completely reduced during 15 ¡¾ 20 h.After storage for one day, silver(I) salts retain catalytic activity. In a 0.5 MNaOH solution, the catalytic activity of the d-metal ions is less pronounced. In the temperature range of 30 ¡¾ 50 8C, the activation energies of Np(VI) oxidation in the presence of the Ag(I) and Co(II) ions are 67 and 42 kJ mol71, respectively. Neptunium(IV). In deaerated 0.5 ¡¾ 14 M NaOH solutions con- 8 taining a suspension of hydrated Np(IV) oxide (7 mmol litre71), Np(V), Np(VI) and Np(VII) are successively accumulated under the action of S2O2¢§ 8 .123 All steps of oxidation are accelerated as the concentrations of NaOH and S2O2¢§ 8 are increased and the temper- ature is raised. In a 4 M NaOH solution containing S2O2¢§ (20 mmol litre71), the transformation of neptunium(IV) into the dissolved Np(VI) form proceeds by 95% at 21 8C during 11 ¡¾ 12 min.After the next 5 ¡¾ 7 min, Np(VII) is generated. In a 4 M NaOH solution, the activation energy of Np(VI) oxidation with persulfate ions is 68 kJ mol71. In the presence of insufficient persulfate, Np(V) and Np(VI) were found in the solution (T=36 8C) after completion of the reaction. The oxidation process involves the following reactions [Np(V) and Np(VI) react both in the solid and liquid phases after the formation of soluble complexes]: (125) Np(IV)+S2O2¢§ 8 Np(V)+SO¢§4 +SO24 ¢§, (35) Np(V)+S2O2¢§ 8 Np(VI)+SO¢§4 +SO24 ¢§, (40) Np(VI)+S2O2¢§ 8 Np(VII)+SO¢§4 +SO24 ¢§. Then, the reactions (124), (107), (108) and (112) ¡¾ (114) proceed.8 Plutonium(VI). It was found 124 that the reaction of Pu(VI) (0.75 mmol litre71) with the S2O2¢§ ions in KOH solutions afforded Pu(VII), which was accumulated according to the kinetic equation (reaction is zero order with respect to plutonium and first order with respect to S2O2¢§ 8 ) (126) 2O28 ¢§a. daPuOVIIUa a ¢§ daPuOVIUa a kaS dt dt 3 In a 1 M KOH solution, the activation energy of Pu(VI) oxidation is 146 kJ mol71, which is similar to the activation energy of thermal decomposition of the persulfate ions. Oxidation proceeds with the participation of the O¢§ radical ions [see the reactions (32), (124), (107) and (108)] (127) PuO2(OH)2¢§ 4 +O¢§3 PuO4(OH)3¢§ 2 +O2+H2O. In hot KOH solutions containing the S2O2¢§ 8 ions, oxidation of 8 Pu(VI) is accompanied by reduction of Pu(VII).As the KOH concentration is increased from 0.2 to 1.2 mol litre71 ([S2O2¢§ 8 ]=0.1 mol litre71, T=80 8C), the rate of Pu(VII) accu- mulation initially somewhat increases, peaks at the KOH concen- tration of 0.5 mol litre71 and then decreases. The yield of Pu(VII) {the [Pu(VII)]/[Pu(VI)]0 ratio} is linearly increased as the KOH concentration is increased to *1 mol litre71 and then is sharply decreased. At a low alkali concentration, the rate of Pu(VII) reduction with water increases and, hence, its yield is decreased under these conditions. An increase in the KOH concentration leads to the enhancement of stability of Pu(VII) and an increase in its yield.In solutions with [KOH]>1 mol litre71, oxidation of Pu(VI) ceases. This is attributed to catalytic decomposition of S2O2¢§ to form reducers. One would expect a change in these dependences at other temperatures and concentrations of persul- fate ions. 8 Plutonium(IV). Hexavalent plutonium is accumulated in 0.5 ¡¾ 8 M NaOH solutions containing a suspension of freshly precipitated hydrated Pu(IV) oxide and S2O2¢§ (10 ¡¾ 50 mmol litre71). At [NaOH]54 mol litre71, Pu(VII) is also gen- erated.125 The rate of dissolution of hydrated Pu(IV) oxide V P Shilov, A B Yusov increases as the concentrations of NaOH and S2O2¢§ 8 are increased and the temperature is raised from 23 to 62 8C. The following solid-phase reactions are of importance in oxidation and dissolu- tion: (128) Pu(IV)+S2O2¢§ 8 Pu(V)+SO¢§4 +SO24 ¢§, (129) Pu(V)+S2O2¢§ 8 Pu(VI)+SO¢§4 +SO24 ¢§.Simultaneously with these reactions, Pu(V) goes into the solution where it is also oxidised to Pu(VI) and undergoes disproportionation. Plutonium(VI) reacts with PuO2 . nH2O. In 4 ¡¾ 8 M NaOH solutions, Pu(VI) is oxidised (130) Pu(VI)+S2O2¢§ 8 Pu(VII)+SO¢§4 +SO24 ¢§. 3 In addition, O¢§ is formed under these conditions and reactions involving this radical proceed. Americium(III). Oxidation of Am(OH)3 (a pink suspension) with the S2O2¢§ 8 ions in an alkaline medium affords a black or dark- brown precipitate of hydrated Am(IV) oxide.106 The reaction of Am(OH)3 (5 mg) with excess K2S2O8 in a 0.1 M NaOH solution on a boiling water bath (2 h) gave rise to an olive-brown precipitate. The examination of products of its dissolution in acid demonstrated121 that the average oxidation number of americium is 4.53, i.e., the precipitate contains not only Am(IV) but also americium in higher oxidation states. Pure Am(IV) was prepared by adding Am(III) to a hot 7 M KOH solution saturated with K2S2O8.Then the reaction mixture was kept at 90 8C for 1 h. The precipitate was washed with water to remove an excess of the oxidiser. The average oxidation state of americium in hydrated oxide is 4.00.1.121 e. Oxidation with ClO7 ions Neptunium(IV). In deaerated 0.5 ¡¾ 16 M NaOH solutions contain- ing NpO2 . nH2O, the successive formation of Np(V), Np(VI) and Np(VII) takes place under the action of the ClO7ions.126 The time dependence of the Np(VI) concentration in solution passes through a maximum.Initially, a supersaturated solution of Np(VI) is formed from which Np(VI) hydroxide gradually precip- itates. Neptunium(VII) is formed in a 2 M NaOH solution (T=25 8C) at [ClO7]=48 mmol litre71. In more concentrated alkaline solutions, a lower concentration of the hypochlorite ions is sufficient for the formation of Np(VII). The activation energy of the Np(VII) formation is 34 kJ mol71 (T=25 ¡¾ 60 8C). The oxidation reactions of Np(IV), Np(V) and Np(VI) are accelerated as the temperature is raised and the concentrations of alkali and hypochlorite ions are increased. The oxidation process involves the following steps: (131) NpO2 .nH2O+OH7+ClO7 NpO2(OH)+ClO27.+mH2O, (132) NpO2 . nH2O+ClO27. NpO2(OH)+Cl7+OH7+mH2O. It is not improbable that Np(IV) is oxidised due to the transfer of the oxygen atom taking into account that oxygen is involved in the oxygenated Np(V) and Np(VI) ions (133) NpO2 . nH2O+ClO7 NpO2(OH)2+Cl7+(n71)H2O. Then the reaction (10) proceeds in the precipitate. The mechanism of neptunium oxidation can be conclusively established with the use of the 18O-labelled hypochlorite ions. In a precipitate, neptunium(V) reacts with ClO7 (134) NpO2(OH)2+ClO27., NpO2OH+ClO7+OH7 NpO2(OH)2+Cl7+OH7. (135) NpO2OH+ClO27.+H2O In solution, the latter reaction proceeds differentlyRedox reactions of actinides in carbonate and alkaline solutions NpO2(OH)27+ClO27.+OH7 NpO2(OH)2¢§ 4 +Cl7.(136) Neptunium(VI) is, in turn, oxidised to Np(VII) (137) NpO2(OH)2¢§ 4 +ClO7+2OH7 NpO4(OH)3¢§ 2 +ClO27.+2H2O, NpO4(OH)3¢§ 2 +Cl7+H2O. (138) NpO2(OH)2¢§ 4 +ClO27. Reproportionation of Np(VII) and Np(V) [reaction (94)] occurs in a dilute alkaline solution, and disproportionation of Np(VI) takes place in 9 ¡¾ 16 M NaOH solutions. Neptunium(VI). In 1¡¾4 M LiOH solutions or 1 ¡¾ 8.5 M NaOH solutions, neptunium(VI) is oxidised with K2IrCl6 .127 (139) Np(VII)+IrCl3¢§ Np(VI)+IrCl2¢§ 6 . 6 6 Neptunium(VII) thus prepared slowly disappears because IrCl3¢§ is hydrolysed and the resulting Ir(III) hydroxide reduces Np(VII). The addition of a NaClO solution to an alkaline solution of neptunium(VI) containing iridium hydroxides leads to the slow formation of Np(VII) because iridium is transformed into higher oxidation states under the action of the ClO7 ions.The ClO27. radical ions that are formed react with Np(VI) [see the reac- tion(138)]. Plutonium(IV). The hypochlorite ions oxidise a suspension of freshly prepared PuO2 . nH2O even at room temperature.126 In a precipitate, plutonium is transformed into a hexavalent state. Plutonium(VI) remains partially in solution. Plutonium(VII) appears at the NaOH concentration of about 10 mol litre71. The rate and completeness of dissolution of Pu(IV) (6.8 mmol litre71; 60 8C, [ClO7]=12 mmol litre71) are increased as the NaOH concentration is increased.In a 8 MNaOH solution, 92%¡¾ 94% of plutonium is dissolved. The initial rate of accumu- lation of plutonium in solution increases as the ClO7 concen- tration and temperature are increased. Oxidation of Pu(IV) proceeds analogously to oxidation of Np(IV) and involves the reactions (140) PuO2 . nH2O+OH7+ClO7 PuO2(OH)+ClO27.+mH2O, (141) PuO2 . nH2O+ClO27. PuO2(OH)+Cl7+OH7+ (n71)H2O, which take place in the solid phase. In 0.4 ¡¾ 1.5 M NaOH solu- tions, plutonium(V) either undergoes disproportionation or is oxidised (142) PuO2(OH)2+ClO27., PuO2(OH)+ClO7+OH7 (143) PuO2(OH)+ClO27.+H2O PuO2(OH)2+Cl7+OH7. Plutonium(V) and Pu(VI) partially go into solution where they are further oxidised (144) PuO2(OH)2¢§ 4 +ClO7+2OH7 PuO4(OH)3¢§ 2 +ClO27+2H2O, PuO4(OH)3¢§ 2 +Cl7+H2O.(145) PuO2(OH)4 2¢§+ClO27. Oxidation of Pu(IV) can also occur through the transfer of the oxygen atom from ClO7 PuO2 . nH2O+ClO7 PuO2(OH)2+Cl7+(n71)H2O. (146) Americium(III). Precipitation of a pink suspension of Am(III) in an alkaline solution containing ClO7 afforded a dark precip- itate.128 This process was studied in detail.121 In a 0.2 M NaOH solution containing ClO7 (0.2 mol litre71), Am(OH)3 was trans- formed into hydrated Am(IV) oxide at 100 8C during 30 min. 479 f. Oxidation with BrO7 ions Neptunium(VI). In alkaline solutions containing Np(VI) (0.2 mmol litre71), neptunium(VI) is oxidised with the hypobro- mite ions upon heating to 70 8C.129 In 1.74 ¡¾ 6.5 M NaOH solutions, the kinetics of the reaction is described by the empirical equation 130 daNpOVIIUa a ¢§ daNpOVIUa a kaNpOVIUaaBrO¢§a0:6aOH¢§a1:6, (147) dt dt where k=(6.650.35)61072 litre2.2 mol72.2 min71 at 70 8C.In a 6.5 M NaOH solution, the activation energy of the reaction is 87.5 kJ mol71 (T=50.7 ¡¾ 70 8C). The rate of Np(VI) oxidation with the BrO7 ions depends on the OH7 concentration in a complicated manner, which reflects both the participation of hydroxyl ions in the formation of an activated complex and the effect of the ionic strength of solution. The fractional order of the reaction with respect to the BrO7 ions results from the fact that the processes (148) NpO2(OH)2¢§ 4 +BrO7+2OH7 NpO4(OH)3¢§ 2 +BrO27.+2H2O, NpO4(OH)3¢§ 2 +Br7+H2O (149) NpO2(OH)2¢§ 4 +BrO27.are accompanied by the reactions of Np(VI) with radical products of BrO7 decomposition. Oxidation of Np(VI) with the hypobromite ions is accelerated upon the addition of d-metal salts. In dilute alkali (0.5 mol litre71), the formation of Np(VII) was observed only in the presence of 2.561075 mol litre71 of Cu(II) or (1¡¾5)61076 mol litre71 of Ni(II) or Co(II). Neptunium(IV). In deaerated 0.8 ¡¾ 16 M NaOH solutions con- taining a suspension of NpO2 . nH2O (5 mmol litre71), Np(V), Np(VI) and Np(VII) are formed in the presence of BrO7.131 At 25 8C, the rate of Np(VII) accumulation is approximately proportional to the NaOH concentration and the process is accelerated as the BrO7 concentration is increased. In the temper- ature range of 25 ¡¾ 40 8C, the activation energy of this reaction is 32 kJ mol71.Oxidation is most likely to occur according to the reactions analogous to the reactions (131) ¡¾ (136) followed by the reactions (10), (148) and (149). Plutonium(VI). Oxidation of Pu(VI) (0.85 mmol litre71) with the hypobromite ions was studied in 10.4 ¡¾ 12.4 MKOH solutions at 50 ¡¾ 90 8C.124 The completeness and rate of oxidation are increased as the concentrations of alkali and KBrO are increased (7 ¡¾ 100 mmol litre71). The initial regions of the kinetic plots yield straight lines, which are indicative of oxidation of Pu(VI) with radical products. The activation energy, which we estimated from the results of the study,124 is 45 ¡¾ 50 kJ mol71. The addition of Co(II) and Ni(II) salts to 1075 mol litre71 leads to substantial acceleration of BrO7 decomposition and Pu(VII) accumulation, which suggests that these processes corre- late with each other.Plutonium(IV). In 0.8 ¡¾ 16 M NaOH solutions containing a sus- pension of PuO2 . nH2O, Pu(VI) is accumulated both in the precipitate and solution under the action of the hypobromite ions.131 Plutonium(VII) is formed in 12 ¡¾ 16 M NaOH solutions. The rates of all steps increase as the concentrations of alkali and the oxidiser are increased. The activation energy of the Pu(VI) formation in solution is 23 kJ mol71 and the activation energy of the Pu(VII) formation in a 16 MNaOH solution is*35 kJ mol71. Oxidation of plutonium(IV) proceeds analogously to oxidation of neptunium(IV).g. Oxidation with Fe(CN)3¢§ 6 ions Neptunium(VI). The reaction (150) Np(VII)+Fe(CN)4¢§ Np(VI)+Fe(CN)3¢§ 6 6480 is reversible and its equilibrium depends substantially on the concentration of the hydroxyl ions. In 1 ¡¾ 4 M KOH solutions, the equilibrium constant is proportional to [OH7]4, whereas this constant in solutions with m=4 is proportional to [OH7] 3.93 In a 3.2 M KOH solution, the equilibrium constant (at 23 8C) is 2.00.2. At the concentrations of the reagents of *0.1 mmol litre71, the rate of the reaction (150) is described by the equa- tion 132 daNpOVIIUa a ¢§ daNpOVIUa a k1aNpOVIUaaFeOCNU3¢§ dt 6 a. dt In a 1.5 M KOH solution (at 10 8C), the ionic strength exerts only a slight effect on the rate constant.Thus the constant k1 changes from 170 to 215 litre mol71 min71 as m is increased from 1.5 to 3 (KNO3). The constant k1 changes almost directly propor- tional to the concentration of the hydroxyl ions, viz., it increases from 215 to 400 litre mol71 min71 (m=3) as the concentration of OH7 is increased from 1.5 to 3 mol litre71. In a 3 M KOH solution, the activation energy of the reaction (150) is 39 kJ mol71 (10 ¡¾ 25 8C). Neptunium(IV). In a 0.5 M NaOH solution containing a NpO2 . nH2O suspension (5.1 mmol litre71) and K3Fe(CN)6 (50 mmol litre71), Np(VI) is formed in less than 15 min.104 In 6 ¡¾ 14 M NaOH solutions, neptunium(VI) is further oxidised to Np(VII). 6 6 6 Plutonium(VI). The reaction of Pu(VI) with Fe(CN)3¢§ pro- ceeds at a noticeable rate at [KOH]>8 mol litre71.133 In a 12 M KOH solution containing the ferricyanide ions (2 mmol litre71), Pu(VI) (0.5 mmol litre71) is completely oxidised.The reaction rate is high even in the cold. However, the resulting Pu(VII) solutions appeared to be unstable. Apparently, the cyanide ions that are generated upon hydrolysis of Fe(CN)3¢§ and Fe(CN)4¢§ serve as a reducing agent. Plutonium(IV). In a 0.2 M NaOH solution containing a suspension of freshly precipitated PuO2 . nH2O (1 mmol litre71), Pu(VI) (0.08 mmol litre71) was accumulated within 10 min after the addition of K3Fe(CN)6 (3.3 mmol litre71). During the same period of time, the concentration of Pu(VI) reached 0.6 and 0.75 mmol litre71 in 1 M and 14 M NaOH solutions, respec- tively.104 Americium(III).In alkaline solutions, Am(III) is oxidised with an excess of the ferricyanide ions to Am(V). This reaction affords the soluble (AmO2)3Fe(CN)6 complex, which gradually decom- poses to give a precipitate of Na2AmO2(OH)3 .H2O. Hydrated Am(IV) oxide is formed at equal concentrations of Am(III) and K3Fe(CN)6.134 h. Oxidation with CrO2¢§ 4 ions In a 1 MNaOH solution, the potential of the CrO2¢§ 4 /CrO34 ¢§ pair is estimated at 0.1 V.135 Hence, the chromate ions can partially or completely oxidise Np(IV). At room temperature, a suspension of NpO2 . nH2O in 0.5¡¾2 MNaOH solutions is slowly oxidised with potassium chromate (the reaction does not proceed to completion during several hours).104 The resulting Np(V) almost completely remains in the precipitate.The completeness of oxidation is increased as the alkali concentration is increased. In a 14 M NaOH solution, the quantitative transformation of Np(IV) into Np(V) can be achieved already during 1 h. Plutonium(IV) does not react with chromate ions in the range of NaOH concentrations from 0.5 to 14 mol litre71.104 i. Oxidation with MnO¢§4 ions The standard potential of the MnO¢§4 /MnO24 ¢§ pair is 0.56 V.135 The formal potential increases as the ionic strength of the solution is increased. Hence, the MnO¢§4 ion must oxidise Np(IV), Np(V), Np(VI) and Pu(IV), Pu(V), Pu(VI). Neptunium(VI) and plutonium(VI). The reactions of Np(VI) and Pu(VI) with MnO¢§4 are reversible 99, 136 (151) Np(VII)+MnO2¢§ Np(VI)+MnO¢§ 4 , 4 V P Shilov, A B Yusov (152) Pu(VII)+MnO2¢§ Pu(VI)+MnO¢§ 4 , 4 4 their equilibria are rapidly established at 20 ¡¾ 25 8Cand depend on the alkali concentration.Neptunium(VI) is oxidised in 3 ¡¾ 4 M NaOH solutions but the formation of Np(VII) is generally accompanied by the side reaction, viz., gradual reduction of excess MnO¢§4 ions with water to MnO24 ¢§. The potential of the Np(VII)/ Np(VI) pair sharply increases as the alkali concentration is decreased, whereas the potential of the MnO47/MnO2¢§ pair remains virtually unchanged. Because of this, neptunium(VII) rapidly oxidises MnO2¢§ 4 to MnO¢§4 in solutions with pH 11. At high concentrations of alkali, the equilibrium of the reaction involving plutonium is shifted to the right.Plutonium(VI) is oxidised with the MnO¢§4 ions only in 10 ¡¾ 14 M NaOH solu- tions.136 The process is accompanied by the side reaction, viz., reduction of excess MnO¢§4 ions with water to MnO24 ¢§. Dilution of alkali to 2 ¡¾ 4 mol litre71 leads to rapid oxidation of the man- ganate ions with plutonium(VII) to MnO¢§4 . Neptunium(IV). In 0.5 ¡¾ 1 M NaOH solutions containing a NpO2 . nH2O suspension (5 mmol litre71), neptunium (IV) is oxidised with a fourfold excess of potassium permanganate to Np(VI) during 10 min. Approximately a half amount of Np(VI) remains in the precipitate, whereas the remainder of Np(VI) goes into solution. In 2 ¡¾ 14 M NaOH solutions, neptunium passes entirely into solution during the same period of time.At [NaOH]>6 mol litre71, Np(VII) was also found in the solu- tion.104 Plutonium(IV). In a 0.5 MNaOH solution containing a freshly precipitated suspension of hydrated Pu(IV) oxide (1 mmol lit- re71), 10% of dissolved Pu(VI) appears within minutes after the addition of a threefold excess of permanganate ions. The Pu(VI) concentration is not increased with time. An increase in theNaOH concentration is accompanied by an increase in the proportion of dissolved plutonium. In a 14 M NaOH solution, 60% of pluto- nium is already dissolved.104 Americium(III). In alkaline solutions, a suspension of Am(OH)3 is oxidised with the MnO¢§4 ions to give hydrated Am(IV) oxide.123 4 j.Oxidation with FeO2¢§ 4 ions In a 1 MNaOH solution, the potential of the Fe(VI)/Fe(III) pair is 0.8 V.135 This value is higher than the potentials of the Np(V)/ Np(IV), Np(VI)/Np(V), Np(VII)/Np(VI) and Pu(V)/Pu(IV), Pu(VI)/ Pu(V) pairs in alkaline solutions with the same concentrations. The FeO2¢§ ion can oxidise Np(IV) and Pu(IV), but the multi- electron transition from Fe(VI) to Fe(III) is accompanied by the structural rearrangement, which is reflected in the rates of the processes. Neptunium(IV). A suspension (5.1 mmol litre71) of hydrated Np(IV) oxide in a 0.5 M NaOH solution is slowly oxidised to Np(V) and Np(VI) under the action of a 10-fold excess ofK2FeO4 . An increase in the NaOH concentration is accompanied by acceleration of the process.In a 16 M NaOH solution, neptuni- um(IV) disappears during several minutes and Np(VII) is formed along with Np(V) and Np(VI).104 Plutonium(IV). In the presence of potassium ferrate (20 mmol litre71), a freshly precipitated suspension of hydrated Pu(IV) oxide in a 0.5 M NaOH solution is slowly dissolved. At [PuO2 . nH2O]=1 mmol litre71, 5.7% of plutonium goes into solution after its storage in the presence of an oxidiser for two days, whereas only 1.5% of plutonium passes into solution in the absence of an oxidiser. In a 14 M NaOH solution, 21% of plutonium goes into solution during the same period of time.104 6 k. Oxidation with XeO4¢§ 6 ions Neptunium(VI). In alkaline solutions, neptunium(VI) is oxidised with the XeO4¢§ ions only upon heating.In a 1 M KOH solution (at 70 8C), 50% of Np(VI) is oxidised with the XeO4¢§ 6 ions already in 15 min ([Np(VI)]=0.2 mmol litre71, [XeO4¢§ 6 a=1 mmol litre71).129 The reaction rate somewhat increases as the alkaliRedox reactions of actinides in carbonate and alkaline solutions concentration is decreased. The addition of the perxenate ions to weakly acidic or neutral solutions of Np(VI) does not afford heptavalent neptunium. Plutonium(VI). In 0 ¡¾ 12 M NaOH solutions, plutonium(VI) does not react with the perxenate ions.119 This is to some extent associated with a decrease in solubility of Na4XeO6 in concen- trated NaOH solutions. In hot concentrated KOH solutions, plutonium(VI) is partially oxidised with xenon(VIII), but Pu(VII) is formed in low yield.l. Oxidation with xenon trioxide In alkaline solutions, xenon trioxide reacts with Np(VI) more actively than the perxenate ions.129A1 MKOH solution contain- ing Np(VI) gradually turns green upon the addition of XeO3 even in the cold. The reaction proceeds during 3 h. Oxidation of a half amount of neptunium(VI) occurs in a matter of minutes and during *1 min on heating to 50 and 70 8C, respectively. Oxida- tion of Np(VI) slows down when the alkali concentration is decreased to 0.25 mol litre71, whereas an increase in this concen- tration leads to acceleration of oxidation. In 10 ¡¾ 11 M NaOH solutions, Np(VI) (0.2 mmol litre71) is completely oxidised with a tenfold excess of XeO3 at 20 8C during 35 min.74 When Np(VI) and XeO3 are present in equal concentrations, approximately a half amount of hexavalent neptunium is oxidised during the same period of time.The reaction of Np(VI) with XeO3 in alkaline solutions is accompanied by a side process, viz., decomposition of the oxidiser to form perxenate, which can also be involved in oxidation of Np(VI). m. Oxidation with xenon difluoride Xenon difluoride oxidises Np(VI) only in solutions containing alkali in a moderate concentration.74, 75 When the reactions are carried out with the use of solid XeF2, Np(VII) is formed in solutions containing NaOH in a concentration from 0.01 to 1.7 mol litre71. For achieving quantitative oxidation of Np(VI) in 0.2 ¡¾ 1.0 M NaOH or KOH solutions, it is necessary to use a 200-fold excess of solid XeF2 .75 Aqueous solutions of XeF2 are more efficient and can oxidise Np(VI) in 5 MKOHsolutions.(The reaction proceeds rapidly.74) Xenon difluoride can reduce Np(VII). Thus, the addition of XeF2 (5 ¡¾ 200 mg) to a 1 M NaOH solution (20 ml) containing Np(VII) (0.2 mmol litre71), which was prepared by ozonation, afforded Np(VI) in 3%¡¾ 12% yields. The addition of microconcentrations of Co(II) and Ni(II) salts has no noticeable effect on oxidation of Np(VI) with xenon difluoride in a 0.5 M NaOH solution. In the presence of Cu(II) salts (1075 mol litre71), neptunium(VI) is not oxidised. When 0.2 ¡¾ 1.2 MNaOH solutions of Pu(VI) were treated with a 5 ¡¾ 200-fold excess of XeF2, a colour characteristic of Pu(VII) solutions appeared for only a short period (10 min).Within 3 min after the addition of XeF2, the yield of Pu(VII) was no higher than 10%.75 Oxidation of Pu(VI) was not observed in alkaline solutions at a concentration higher than 2 mol litre71. Upon the addition of XeF2 to 0.5 ¡¾ 3 MKOHor NaOH solutions containing Pu(VII), which was prepared by ozonation, the mixture rapidly turned colourless and the absorption spectra of the solutions were identical with the absorption spectrum of Pu(VI). n. Oxidation with silver oxides Neptunium(VI). Neptunium(VII) was prepared in alkaline solu- tions with the use of Ag2O and AgO.93 Oxidation of Np(VI) with silver(I) oxide proceeds at a noticeable rate only in solutions containing alkali in a concentration higher than 8 mol litre71.At room temperature, the reaction rate was low. It is recommended that the solutions be heated to at least 50 8C. At 60 8C, the process was completed in 15 ¡¾ 20 min. Heating of 9 M and 12 M NaOH solutions containing Np(VI) (0.5 mmol litre71) and freshly prepared Ag2O (10 mg ml71) to 60 8C afforded Np(VII) in 10%¡¾ 15% and 70%¡¾ 80% yields, 481 respectively. The yield of Np(VII) was approximately halved as the amount of Ag2O was decreased from 10 to 2 mg ml71. The completeness of Np(VI) oxidation was somewhat improved as the temperature was raised to 90 8C. Oxide AgO oxidised Np(VI) even in a 0.1 M KOH solution. The reaction rate was high even at room temperature. In solutions heated to 50 8C, the reaction equilibrium was established in 3 ¡¾ 5 min.In 2 ¡¾ 3 M solutions of alkali, an excess of silver(II) oxide oxidised 92%¡¾ 93% of the total amount of Np(VI).93 Plutonium(VI). Silver(II) oxide oxidises Pu(VI) only at an alkali concentration higher than 8 mol litre71.136 In an 11 M NaOH solution, the yield of Pu(VII) is close to 100%. In the cold, the reaction rate is low. At 80 ¡¾ 90 8C, oxidation of Pu(VI) is com- pleted in less than 15 min. o. Oxidation with IO¢§4 ions, copper(III) periodate and sodium bismuthate Neptunium(VI) is oxidised with the periodate ions only in hot alkaline solutions.129 At 85 8C, a half amount of Np(VI) in a 1 M KOH solution containing KIO4 (0.1 mol litre71) and Np(VI) (0.2 mmol litre71) was oxidised during 25 min, whereas the reaction in a 5 M KOH solution was completed in*10 min.Plutonium(VI) is not oxidised with the periodate ions both in dilute and concentrated alkaline solutions.119 An excess of copper(III) involved in the complex with the periodate ions rapidly oxidises Np(VI) and Pu(VI) in solutions containing NaOH at concentrations higher than 1 and 11 mol litre71, respectively.137 4. Reduction reactions hydroxyl ions is increased. The activation energy of reduction is 60.5 kJ mol71 (m=1.0, [KOH]=0.1 mol litre71, T=25 ¡¾ 45 8C). 2 (154) O7.+O7.+H2O HO¢§2 +OH7. Heterogeneous oxidation of Np(VI) with sodium bismuthate in dilute alkaline solutions proceeds rather rapidly only at the boiling temperature of the solution.129 The absorption spectra of the green solutions, which are obtained after completion of the reaction and separation of the excess oxidiser by centrifugation, are characteristic of Np(VII). a.Reduction with water Plutonium(VII). The kinetics of reduction of Pu(VII) (0.16 ¡¾ 0.25 mmol litre71) in 0.044 ¡¾ 0.1 M KOH solutions is governed by the equation 120 ¢§ daPuOVIIUa a kaPuOVIIUa1:2. (153) dt The rate constant k decreases as the concentration of the It was suggested 120 that theOH. radical is abstracted from the hydrolysed Pu(VII) anion followed by its transformation into the O7. radical ion [see the reaction(107)] and then into HO¢§ The peroxide ion reduces two Pu(VII) ions. The rate of reduction of Pu(VII) (1.7 mmol litre71) in 0.3 ¡¾ 1.96 M NaOH solutions obeys the equation 138 ¢§ daPuOVIIUa a kaPuOVIIUa.(155) dt The apparent rate constant k is proportional to [OH7] n. Here, 2 n=1 and 2 if [OH7]>1 mol litre71 and [OH7]= 0.3 ¡¾ 0.62 mol litre71, respectively. To detect the OH. radicals, which can be generated in this reaction, Pu(VII) was added to 0.5 M and 1.0 M NaOH solutions saturated with benzene. In both cases, the final product of the reaction of OH. with benzene, viz., phenol, was not detected. Hence, it was concluded 138 that Pu(VII) involved in the complex PuO4(OH)3¢§ anion eliminates 1/2O2 and is transformed into Pu(V), which rapidly reacts with Pu(VII). However, no explan- ation was given as to how one-half of the oxygen molecule was eliminated.482 2 Based on comparison of the standard potentials of the OH./OH7 (1.9 V) and Pu(VII)/Pu(VI) pairs in a 1 M NaOH solution, it was concluded that one-electron oxidation of OH7 orH2O with the Pu(VII) ions is thermodynamically impossible.At the same time, the potential of the HO¢§2 /OH7 pair in an alkaline medium is 0.867 V.135 As in the case of Np(VII) reduction in carbonate solutions, the reaction of Pu(VII) with water proceeds through the formation of a dimer.87, 88 It was established that in an alkaline medium Pu(VII) exists as the PuO4(OH)3¢§ anion.133 A decrease in the alkali concentration leads to the appearance of the PuO4(OH)(H2O)27 species, which are dimerised and then decom- pose into two Pu(VI) ions and H2O2 (or HO¢§2 ).In a 0.15 M KOH solution, reduction of Pu(VII) is accelerated in the presence of Fe(III), Cu(II), Ni(II) or Co(II) salts.120 The effect of trivalent iron becomes noticeable starting with the concentration of 7.761077 mol litre71. The rate of Pu(VII) reduction increases proportionally to an increase in the Fe(III) concentration to at least 3.161076 mol litre71. Analogous results were obtained also for Cu(II) and Ni(II). The effect of Co(II) in a 0.15 MKOHsolution is manifested at its concentration of *1079 mol litre71. The kinetic curves for Pu(VII) reduction take the shape characteristic of the kinetic curves of autocatalytic reactions. The sensitivity of Pu(VII) to the presence of Co(II) decreases sharply as the OH7 concentration is increased and the curves of Pu(VII) reduction assume the form typical of the kinetic curves of the first-order reactions [as in the case of solutions containing additives of Fe(III), Cu(II) or Ni(II)].The catalytic action of d elements is attributed to their capability for being transformed into higher valence forms, which rapidly react with water. b. Reduction with hydrogen peroxide Neptunium(VI) and neptunium (VII).The kinetics of the reactions Np(V)+HO2. , 2 Np(V)+O2 , 2 Np(VI)+HO2. , 2 (156) (157) (158) (159) Np(VI)+HO¢§ Np(VI)+O¢§ Np(VII)+HO¢§ Np(VII)+O¢§ Np(VI)+O2 2 in alkaline solutions containing Np(VI) or Np(VII) (0.1 mmol litre71) at pH 9.2 ¡¾ 13.7 was studied by pulse radiolysis.139 The rate constants for all reactions decrease from 104¡¾ 105 to 102 ¡¾ 103 litre mol71 s71 as pH is increased in the above-men- tioned range. This is associated with a decrease in the oxidation potentials of the Np(VI)/Np(V) and Np(VII)/Np(VI) pairs.The kinetics and the mechanism of the reactions of Np(VI) and Np(VII) withH2O2 in 1 ¡¾ 8.5 MNaOH solutions were examined in the study.140 It was found that Np(VI) interacts withH2O2 to form complexes with 1 : 1 and 1 : 2 compositions. In the latter complex, intramolecular reduction takes place to give the nonequilibrium Np(V)(O2¢§ 2 ) complex. In the presence of an excess of H2O2, the concentration of this complex tends to an equilibrium value. In the presence of an excess of Np(VI), the Np(V)(O2¢§ 2 ) complex reacts directly with Np(VI).In addition, the complex is decomposed to liberate H2O2 , which reduces Np(VI). Hydrogen peroxide reduces Np(VII) to Np(VI) (outer-sphere reduction). Plutonium(VI) and plutonium(VII). In 1 ¡¾ 7 M NaOH solutions (at 23 8C), the stoichiometric coefficient for the reaction of Pu(VI) withH2O2 varies from 1.6 to 2.0 and approaches 2 as the [Pu(VI)]0/ [H2O2] ratio is increased. The stoichiometry changes only slightly as the temperature is lowered to 10 8C.141 It can be assumed that the reaction of Pu(VI) with H2O2 proceeds according to Eqn (90). At room temperature, the rate of the process in 2 ¡¾ 4 M NaOH solutions obeys the equation (160) ¢§ daPuOVIUa a kaPuOVIUaaH2O2a, dt where k=56102 litre mol71 s71.V P Shilov, A B Yusov (161) complex Pu(V)+O¢§ Pu(VI)+HO¢§ 2 . 2 The mechanism of Pu(VI) reduction involves reactions analo- gous to the reactions (78) and (80), i.e., Pu(VI) forms complexes with hydrogen peroxide. Intramolecular transformations afford Pu(V) The rate of the reaction of Pu(VI) with H2O2 decreases as alkaline solutions of Pu(VI) are aged. For example, the rate constant for the process in a 2 M NaOH solution ([Pu(VI)]=1 mmol litre71), which was previously kept for 4 days, decreased to 5 litre mol71 s71. Probably, this effect is caused by slow polymerisation of Pu(VI). The reactions of Pu(VII) and Am(VII) with H2O2, like those of Np(VII), proceed through an outer-sphere mechanism. In the study,142 the rate constants for the reactions Pu(VII)+HO¢§ Pu(VI)+HO2., 2 Am(VII)+HO¢§ Am(VI)+HO2. , 2 2 / 2 were estimated at (4 ¡¾ 5)6104 and (1 ¡¾ 4)6106 litre mol71 s71, respectively, from the dependence of log k on the difference between the redox potentials of the An(VII)/An(VI) and HO¢§ OH7 pairs. Americium(VI). It is impossible to perform complete oxidation of Am(VI) with ozone because of instability of Am(VII) to reduction with water and also due to side reactions involving ozone.102 It was expected that an increase in the O3 concentration in the gaseous phase and solution would lead to acceleration of Am(VI) oxidation, whereas reduction of the resulting Am(VII) with water would be of less importance. However, bubbling of a gaseous mixture with a high ozone content through an alkaline solution of Am(VI) resulted in reduction rather than in oxidation of Am(VI).143 This is associated with the fact that a large amount of O¢§3 radicals was formed due to a high concentration of O3 and these radicals rapidly decomposed.At a concentration lower than 1075 mol litre71, the rate of their decomposition is described by the first-order rate equation, whereas this decomposition obeys the rate law for a second-order reaction at a concentration higher than 1075.144 In the first case, the reaction affords O¢§ ., which rapidly reacts with O3 and, to a small extent, with Am(VI). In the second case, the reaction yields HO¢§2 , which reacts rather slowly with O3 but efficiently reduces Am(VI).c. Reduction with hydrazine N2H4 Neptunium(VII). Neptunium(VII) is rapidly reduced with hydra- zine. The examination of the reaction stoichiometry demon- strated 145 that 0.25 g-mol of N2H4 were consumed per g-mole of Np(VII) (162) 4NpO4(OH)3¢§ 2 +N2H4 +4H2O 4NpO2(OH)2¢§ 4 +N2+4OH7. Neptunium(V). In a 2 M NaOH solution, neptunium(V) (0.2 mmol litre71) does not react with N2H4 . 2 HCl and other hydrazine derivatives, except forN2H4 .HNO3 . The latter reduces Np(V) to Np(IV) at a moderate rate.146, 147 At 20 ¡¾ 65 8C, the reaction with N2H4 .HNO3 (0.1 ¡¾ 0.2 mol litre71) may take from several hours to several days to give a precipitate consisting of Np(IV) and a portion of captured Np(V). Precipitation of neptu- nium slows down as the NaOH concentration is increased.In a 14 M NaOH solution, the precipitate does not form during the time of the experiment. Apparently, the starting N2H4 .HNO3 specimen contains an active compound, which appears after prolonged storage of the reagent at 18 ¡¾ 20 8C. For example, HN3 or H2N2O2 can be accumulated in hydrazine. However, these compounds do not reduce Np(V) in an alkaline medium. Hence, the nature of the active compound remains unclear.147 In solutions of LiOH containing hydrazine hydrate, Np(V) is not reduced in the temperature range of 40 ¡¾ 80 8C. Under theRedox reactions of actinides in carbonate and alkaline solutions same condition, the reaction performed in the presence of PdCl2 affords a finely dispersed palladium metal on which hydrated Np(IV) oxide precipitates.The processes are accelerated as the PdCl2 andN2H4 concentrations are increased and the temperature is raised, whereas an increase in the alkali concentration from 1 to 5 mol litre71 slows down these processes.148 Plutonium(VII). In a 2 M NaOH solution, plutonium(VII) is reduced with hydrazine 119 4 PuO4(OH)3¢§ 2 +N2H4 +4H2O 4 PuO2(OH)2¢§ 4 +N2+4OH7. Plutonium(VI). It was found 149 that in 5 ¡¾ 8 M NaOH solu- tions, N2H4 . HCl or N2H4 .HNO3 taken in a molar ratio of 1 : 3 with respect to Pu(VI) reduce the latter to Pu(V) in the course of stirring, i.e., the following reactions proceed (163) Pu(V)+N2H3. , Pu(VI)+N2H4 (164) Pu(V)+N2H2 Pu(VI)+N2H3.etc. Plutonium(V). In 0.5¡¾4M NaOH solutions, Pu(VI) (0.42 mmol litre71) is reduced with hydrazine (5 mmol litre71) to Pu(IV) at 25¡¾55 8C during several minutes.149 Under these conditions, the reactions (163) and (164) are followed by dispro- portionation of Pu(V). Accumulation of Pu(IV), while slows down, does not cease as the NaOH concentration is increased (higher than 4 mol litre71). Under these conditions, disproportionation of Pu(V) may be ignored, i.e., the whole of Pu(IV) is formed as the product of Pu(V) reduction (165) Pu(IV)+N2H3. . Pu(V)+N2H4 An increase in theN2H4 concentration or temperature leads to acceleration of the Pu(IV) formation. Selective spectrophotomet- ric analysis of dissolved precipitates showed that these precipitates always contain Pu(IV), whereas other valence forms were not detected.Americium(V). In a 3 M NaOH solution (at 22 ¡¾ 238C), N2H4 .HNO3 (10 mmol litre71) reduces 50% of pentavalent americium (0.2 mmol litre71) to Am(III) during *30 h.150 The process is accelerated as the concentration of hydrazine is increased or the temperature is raised to 50 8C. An increase in the alkali concentration to 6 mol litre71 leads to acceleration of the reaction, whereas a further increase in the alkali concentration slows down this process. d. Reduction with hydroxylamine Neptunium(VII). In a 1 M NaOH solution (T&20 8C), NH2OH rapidly reduces Np(VII).145 The consumption of the reducing agent is 0.75 ¡¾ 0.80 mol per mole of Np(VII). Apparently, after the reactions (166) Np(VI)+NHOH., Np(VII)+NH2OH (167) Np(VI)+NOH., Np(VII)+NHOH.the NHOH. and NOH. radicals are partially consumed in dimer- isation reactions. Neptunium(V). In 0.5 ¡¾ 6 M NaOH solutions, NH2OH (0.5 mol litre71) reduces Np(V) (0.4 mmol litre71) at a temper- ature higher than 60 8C.146 Efficient reduction occurs at 80 ¡¾ 95 8C. When precipitating, the resulting Np(IV) captures a portion of Np(V) from the solution. Plutonium(VII). The reaction of NH2OH (without an excess) with Pu(VII) rapidly affords Pu(VI). The reaction stoichiometry is expressed as a fraction,119 i.e., the process occurs analogously to the reaction of NH2OH with Np(VII). reduces rapidly Plutonium(VI). In 1 ¡¾ 14 M NaOH solutions (at 25 ¡¾ 60 8C), (5 mmol litre71) Pu(VI) NH2OH (0.5 mmol litre71) to Pu(V).149 483 Plutonium(V).In NaOH solutions (at an alkali concentration higher than 4 mol litre71), hydroxylamine reduces Pu(V) (0.5 mmol litre71) to Pu(IV) at 25 8C (see Ref. 149) (168) Pu(IV)+NHOH., Pu(V)+NH2OH (169) Pu(IV)+NOH.. Pu(V)+NHOH. The rate of Pu(IV) precipitation and, consequently, of Pu(V) reduction decreases as the NaOH concentration is increased. However, this dependence is not so sharp as that observed in the case of hydrazine. At room temperature, NH2OH reduces Pu(V) at a moderate rate even in a 14 M NaOH solution. The rate of this process depends substantially on the NH2OH concentration and temperature. Americium(V).In a 3 M NaOH solution (at 22 ¡¾ 23 8C), hydroxylamine (0.1 mol litre71) slowly reduces Am(V) (0.2 mmol litre71) to Am(III).150 The reaction is accelerated as theNH2OH concentration is increased, whereas an increase in the alkali concentration slows down this process. 3 e. Reduction with SO2¢§ 3 ions Neptunium(VII). The kinetics of the reaction of neptunium(VII) with the SO2¢§ ions was studied in 0.1 ¡¾ 0.5 M NaOH solutions with m=0.7 at T=20 ¡¾ 35 8C.145 Under these conditions, Np(VII) is reduced with an excess of sulfite at a moderate rate. The initial regions of the kinetic curves are satisfactorily described by the first-order reaction equation. However, the rate of reduc- tion increases toward the end of the process. This is associated with the fact that Np(VI), which is generated in the course of the process, is reduced with sulfite to Np(V), which reacts with Np(VII) much faster than the sulfite ions.Autocatalytic acceler- ation of the reaction becomes noticeable when a sufficient amount of Np(VI) is accumulated. 3The rate constant for the first-order reaction (k 0) can be determined from the initial regions of the kinetic curves and then the rate constant k for a bimolecular reaction can be calculated from the k 0/[SO2¢§ 3 ] ratio. The k value is constant in the range of SO2¢§ concentrations from 22 to 135 mmol litre71, i.e., the reaction order with respect to the sulfite ions is equal to unity. The dependence of k on the OH7 concentration k2 k a aOH¢§a a k1 aOH¢§a2 shows that the reaction involves low-charged complex neptu- nium(VII) anions formed in the processes (170) NpO4(OH)(H2O)27+OH7, NpO4(OH)3¢§ 2 +H2O (171) NpO4(OH)3¢§ 2 +2H2O NpO4(H2O)¢§2 +2OH7.In a 0.5 M NaOH solution, the activation energy of Np(VII) reduction is 94 kJ mol71. The reaction with sulfite proceeds as follows: (172) Np(VI)+SO¢§ Np(VII)+SO2¢§ 3 , 3 (173) Np(VII)+SO¢§3 Np(VI)+SO3(SO2¢§ 4 ), (174) Np(V)+SO¢§ Np(VI)+SO2¢§ 3 3 and is accompanied by reproportionation [see the reaction (94)]. The process of Np(VII) reduction with sulfite ions is affected by impurities of Cu(II), Co(II) and Mn(VII) compounds. The catalytic action of copper(II) becomes noticeable at its concen- tration of 261077 mol litre71.145 The rate constant for the reaction linearly increases as the CuCl2 concentration is increased from 561077 to*561076 mol litre71 (175) k=k0+k0[Cu(II)], where k0 is the rate constant for the reaction in the absence of a catalyst, k0=1.46106 litre2 mol72 min71 at m=0.7 and484 T=25 8C.Cobalt(II) and manganese(II) are similar in catalytic activity to copper(II). serves as a catalyst of this reaction but the concentration of the latter must be no lower than 1075 mol litre71. Silver(I) com- pounds have no effect on the rate of Np(VII) reduction with the SO2¢§ sulfite ions have no effect on Np(V).146 High concentrations of sulfite are required for reduction of Np(V) in 3¡¾5 M NaOH solutions at 20 8C. However, Np(V) (0.2 mmol litre71) is reduced only by 15% during 24 h even at [SO2¢§ In addition to the above-mentioned metal ions, iron(III) also 3 ions.Neptunium(V). At a concentration of 5 mmol litre71, the 3 ]=1 mol litre71.151 Plutonium(VII) and plutonium(V). In 1.2 ¡¾ 2.5 M KOH solu- 3 tions (T=25 8C), Pu(VII) is rapidly reduced with the SO2¢§ ions.152 In a 4 M NaOH solution, plutonium(V) does not react with sulfite ions at a concentration of 0.1 mol litre71.149 Americium(V). The SO2¢§ 3 ions slowly reduce Am(V). In a 1.5 M NaOH solution containing Am(V) (0.23 mmol litre71) and Na2SO3 (0.11 mol litre71), a half amount of Am(V) is reduced at 22 ¡¾ 23 8C during 9 h.150 The reaction is accelerated as the sulfite concentration is increased. f. Reduction with S2O2¢§ 3 ions Neptunium(VII) and neptunium(V).The thiosulfate ion acts as a weak reducing agent with respect to Np(VII) and Np(V). In a 2 M NaOH solution containing Na2S2O3 (0.05 mol litre71), a half amount of Np(VII) is reduced at 258C in a time longer than 100 min,145 whereas Np(V) is reduced only by 18% even under the action of Na2S2O3 (1 mol litre71) during 24 h.151 Heating or a decrease in the NaOH concentration to 0.1 mol litre71 leads to a slight acceleration of Np(VII) reduction.145 Plutonium(VII). The reaction of Pu(VII) with the thiosulfate ions proceeds at a moderate rate. The time of reduction of a half amount of Pu(VII) (0.5 mmol litre71) in a 2.5 M KOH solution containing Na2S2O3 (2.5 mmol litre71) is 2 min. The reaction rate increases as the concentration of the reducing agent is increased.152 g.Reduction with S2O2¢§ 4 ions and thiourea dioxide The dithionite ions and thiourea dioxide also reduce Np(V), Pu(V) and Am(V). In a 4 M NaOH solution, dithionite (0.1 mol litre71) reduces only 30% of neptunium(V) (0.4 mmol litre71) at 208C during one day. In the presence of (NH2)2CSO2 (0.1 mol litre71), 20% of Np(V) is reduced at 65 8C for 3 h.146 4 In a 4 M NaOH solution containing S2O2¢§ (5 mmol litre71), 90% of plutonium(V) (0.4 mmol litre71) precipitates at 50 8C for 20 min. In the presence of thiourea dioxide (5 mmol litre71), the same amount of plutonium precipitates during 2 min.149 The Pu(IV) and Pu(V) contents in the precipitate are unknown. In a 1.5 M NaOH solution containing dithionite (5 mmol litre71), 50% of Am(V) (0.23 mmol litre71) is reduced at 22 ¡¾ 23 8C for 5 h.In the presence of (NH2)2CSO2 (0.1 mol litre71), the same amount of Am(V) is reduced during 0.2 h. Reduction is accelerated as the concentrations of the reagents are increased.150 h. Reduction with iodide ions Neptunium(VII). The iodide ions (90 ¡¾ 160 mmol litre71) reduce Np(VII) at a moderate rate without heating at [OH7]< 0.03 mol litre71.145 The reaction of Np(VII) directly with I7 proceeds rather slowly. The reaction is accelerated as oxidation products of iodide ions are accumulated in solution. The addition of a mixture of I2 and KI, i.e., the I¢§3 ions, to the reaction mixture exerts an even stronger effect. The presence of Np(VI) has no effect on the reaction.Reduction of Np(VII) is accelerated as the I7 concentration is increased or the alkali concentration is decreased. The process involves the reactions very slow (176) Np(VI)+I, Np(VII)+I7 V P Shilov, A B Yusov fast 2 , I¢§ fast Np(VI)+I2, 2 fast I¢§3 , (177) (178) (179) (180) IO7+I7+H2O, I7+I Np(VII)+I¢§ I7+I2 I2+2OH7 slow (181) Np(VII)+I¢§ Np(VI)+I2+I. 3 The possibility of the reaction of Np(VII) with IO7 was not examined. The addition of microamounts of Co(II), Cu(II) or Mn(II) leads to acceleration of Np(VII) reduction with the I7 ions. The catalytic effect becomes noticeable when these metals are present in concentrations of (1 ¡¾ 2)6107 mol litre71. Plutonium(VII).In KOH solutions, plutonium(VII) is reduced with equimolar amounts of KI at a high rate even at room temperature.152 The reaction proceeds to completion even if the concentration of the reducing agent is 7 ¡¾ 8 times lower than the concentration of Pu(VII). Consequently, all intermediate products of oxidation of I7 to IO¢§4 (or H3IO26 ¢§) react with Pu(VII) due to the high potential of the Pu(VII)/Pu(VI) pair. The possibility of Pu(VII) reduction with the IO7 and IO¢§3 ions was confirmed experimentally. Hence, the reaction of Pu(VII) with insufficient I7 proceeds according to the equation (182) 8Pu(VI)+I(VII). 8 Pu(VII)+I7 In the presence of an excess of the reducing agent, the character of the reaction depends substantially on the alkali concentration.At [KOH]<2 mol litre71, the rate of Pu(VII) reduction is described by the equation (183) ¢§ daPuOVIIUa a kaPuOVIIUaaI¢§a. dt In a 1 M KOH solution (at 25 8C), the constant k is 6.56103 litre mol71 min71. The constant k decreases by a factor of*50 as the alkali concentration is increased to 3.3 mol litre71. For a 2.5 M KOH solution, the temperature dependence of k in the range of 25 ¡¾ 53 8C corresponds to the apparent activation energy of 25 kJ mol71. Reduction of Pu(VII) begins with the reaction slow (184) Pu(VI)+I, Pu(VII)+I7 followed by the reaction (177) fast (185) Pu(VII)+I¢§ Pu(VI)+I2 . 2 and then the reactions (179) and (180) take place. The kinetic curves for Pu(VII) reduction with an excess of KI in solutions with [KOH]>2 mol litre71 are distorted, the dis- tortion being stronger at higher alkali concentrations.For 4 M KOH solutions, the curves assume an S shape, which was attributed 152 to the fact that intermediate products of I7 oxida- tion to I(VII) are involved in Pu(VII) reduction to an increasing extent. In our opinion, under these conditions I¢§2 reduces Pu(VI) to Pu(V), which rapidly reacts with Pu(VII). i. Reduction with iodate ions Plutonium(VII). In 0.5 ¡¾ 3.5 M KOH solutions (at 15 ¡¾ 35 8C), reduction of Pu(VII) with iodate ions proceeds irreversibly accord- ing to the stoichiometric equation 152 (186) 2 Pu(VII)+IO¢§3 =2 Pu(VI)+I(VII). The rate of the process is governed by the equation (187) ¢§ daPuOVIIUa a kaPuOVIIUaaIO¢§3 a.dt In a 1 M KOH solution (at 25 8C), the rate constant for the reaction (186) is 1.16103 litre mol71 min71. As the alkali con- centration of increased from 0.17 to 1.7 mol litre71, the constant k decreases in an inverse proportion to [OH7] and then remains virtually unchanged up to [KOH]=3.5 mol litre71.Redox reactions of actinides in carbonate and alkaline solutions An increase in the temperature leads to a rather small increase in the rate of Pu(VII) reduction. In a 2.5 M KOH solution, the apparent activation energy is 24 kJ mol71. The mechanism of Pu(VII) reduction with the iodate ions can be described by the following scheme: Pu(VII)+I(V)O¢§ Pu(VII)+I(VI) 6 j. Reduction with Fe(CN)4¢§ 6 , tin(II) and vanadium(IV) ions The Fe(CN)4¢§ ions rapidly reduce Np(VII) in a 1 M NaOH solution145 [the equilibrium of the reaction (150) is shifted to the left] and Pu(VII) in a 5 M NaOH solution.119 According to our estimation, the equilibrium constant of the reaction (150) in a 1.5 M KOH solution is 0.1 (T=23 8C).We estimated the rate constant for oxidation [the equilibrium of the reaction (150) is shifted to the right] as 350 litre mol71 min71 in going from 10 to 23 8C. The rate constant for reduction was estimated at 3500 litre mol71 min71. Tin(II) reduces Np(VII) in a 1 M NaOH solution145 and Pu(VII) in 1.5 ¡¾ 2.5 MNaOH solutions in the course of stirring.152 At 20 ¡¾ 80 8C, vanadium(IV) (20 mmol litre71) reduces Np(V) (0.2 mmol litre71) in 1¡¾5 MNaOH solutions as a result of which Np(IV) precipitates and captures a portion of Np(V) from the solution.153 Reduction slows down as the alkali concentration is increased and is accelerated as the temperature is raised.k. Reduction with organic compounds Neptunium(VII). The ascorbate ions rapidly reduce Np(VII).145 The semiquantitative data on reduction of Np(VII) with other organic compounds are given in Table 4. At 25 8C, Np(VII) does not virtually react with ethanol, acetate ions and oxalate ions both in 2 M and 0.1 M alkaline solutions. The resistance of Np(VII) to ethanol and oxalate ions is decreased (particularly, at a low concentration of the OH7 ions) when the temperature is raised to 65 8C. Formaldehyde rapidly reduces Np(VII) under all conditions being studied, whereas the other compounds react with Np(VII) at a moderate rate.The kinetics of the reactions of Np(VII) with aliphatic alco- hols, formaldehyde and formate ions in alkaline solutions was examined in the studies.154 ¡¾ 159 For all systems, the reaction rate is governed by the equation ¢§ daNpOVIIUa a kaNpOVIIUa a k0aRedan[OH7]m[Np(VII)], dt Table 5. Effect of the nature of the reducing agent (Red) and the reaction conditions on stability of Np(VII) and Pu(VII). Red Data for Np(VII) CH3OH C2H5OH n-C3H7OH n-C4H9OH n-C5H11OH iso-C3H7OH iso-C4H9OH iso-C5H11OH CH2O (CH3)2CO HCOONa Data for Pu(VII) CH3OH CH2O C2H5OH aNaOH was used as the medium. (188) Pu(VI)+I(VI), 3 (189) Pu(VI)+I(VII).[Red] /mol litre71 [KOH] /mol litre71 T /8C 102k /min71 28.2 38.9 60.4 60.4 60.4 55.6 55.6 60.4 29.6 30.3 65.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 0.5 1.0 1.0 0.2 a 16.4 3.4 0.28 0.22 0.10 0.26 0.22 0.11 0.0075 0.27 0.01 25.0 25.2 25.0 25.0 25.2 2.5 0.95 a 2.5 2.5 0.94 a 0.55 0.034 0.004 0.55 0.024 Table 4. Time of reduction of a half amount of Np(VII) (t1/2) with organic compounds.145 [Red] a t1/2 /min Reducing agent (Red) [OH7]=0.1 mol litre71 [OH7]= 2 mol litre71 25 8C 658C 258C 658C 0.010 0.090 0.090 32 >1001.1 0.4 2.5 50.1 0.070 >100 0.200 >100 75 12 50 1.500 >100 1.5 Oxalate ion Acetate ion EDTA Citrate ion Formate ion Ethanol Formaldehyde 0.060 a In mol litre71.where k0 is the rate constant at [Red]=1 mol litre71, [OH7]=1 mol litre71, T=25 8C. The kinetic parameters are given in Table 5. Analysis of the data from Table 5 shows that the rate constant for oxidation decreases as the number of atoms in normal alcohols is increased. Oxidation products of alcohols are more reactive than alcohols as such. Neptunium(V). In 1 ¡¾ 5 M NaOH solutions containing up to 1 mol litre71 of HCOONa, Np(V) is not reduced at 60 ¡¾ 90 8C during 5 h. Under the same conditions, the reactions performed in the presence of palladium salts afford palladium metal on which hydrated Np(IV) oxide precipitates. Platinum salts also exert a catalytic effect on Np(V) reduction, although this process is less efficient than that occurring in the presence of palladium salts.Reduction of neptunium(V) with the formate ions proceeds on its contact with an anion-exchange resin covered with palladium or platinum metal.161 Plutonium(VII). The ascorbate ions rapidly reduce Pu(VII).152 The semiquantitative data on the reactions with some other compounds, which react with Pu(VII) more slowly, are presented in Table 5. It can be seen that the reaction of Pu(VII) with ethanol is less efficient than that with methanol and the rate constants for Pu(VII) reduction with these alcohols are much higher than the analogous constants for Np(VII). n m Ea /kJ mol71 50.73.4 49.72.6 71.1 70.96 46.33.0 71.0 58.43.6 81.44.5 52.02.0 54.02.0 3.84 0.69 0.80 0.28 0.14 0.37 0.23 0.40 1.23 0.42 1.45 0.74 71.5 1.5 0.9 0.6 0.8 0.8 0.8 1.0 1.0 0.9 0.6 71.0 71.0 71.0 70.95 71.03.0 71.0 59.04.0 49.02.0 71.0 70.33 56.02.0 7 7 7 1.5 70.3 60.3 0.26 55.5 39.6 0.23 50.04.0 7 7 7 7 7 7 0.9 70.8 44.04.0 485 >100 >100 45 22 20 300.3 >100 >100 >100 7 >100 >100 7 Ref. 104 k0 154 154 155 156 156 157 157 155 158 158 159 44.0 4.47 34.5 5.77 2.71 16.4 10.9 9.0 1.146104 96.0 93.0 152 160 152 152 160 746103 776.36102486 l.Reduction with metals Reduction of Np(VII), Pu(VI) and Am(VI) in alkaline solutions with aluminium metal was examined in the study.162 The reactions were carried out at room temperature.A 2 ± 6 cm2 aluminium plate was submerged in an alkaline solution of actinide, which was vigorously stirred. Reduction of neptunium(VII) (0.1 ± 2 mmol - litre71) in a 0.1 ± 2 MNaOH solution was completed in 8 ± 20 min depending on the alkali concentration and the size of the plate. Neptunium (25% ± 50%) partially precipitated on the aluminium plate as a thin layer, which was firmly attached to the support. The precipitate contained only Np(V). Reduction of Pu(VI) was completed in 5 ± 20 min to give a colloid of Pu(IV) and a precipitate of hydrated Pu(IV) oxide; 5 ± 7 mg cm72 of plutonium precipitated on the aluminium surface.The amount of the precipitate varied with the concentrations of plutonium and alkali. A precipitate of Am(III) hydroxide rapidly formed upon reduction of Am(VI). The amount of americium precipitated on the aluminium surface was no higher than 6 mg cm72. The reactions of Np(V) and Pu(VI) with zinc, chromium, tin and their alloys in 0.5 ± 4 MNaOHsolutions were examined under static and dynamic conditions (filtration of solutions through a column filled with metal granules).163 Reduction of actinides and precipitation of hydroxides took place on the surface of metallic granules. In the case of plutonium, the best result was achieved on contact of chromium granules with a solution of Pu(VI) in 0.5 M NaOH at 60 8C for 2 ± 5 h.The best sorption of neptunium from a 1 M NaOH solution was attained upon prolonged contact with chromium and zinc granules. An increase in the alkali concentration and the addition of chromate ions and complex-forming agents slow down sorption of plutonium. V. Conclusion Abundant data were accumulated on redox reactions of actinides in carbonate and alkaline solutions. Procedures for oxidation of low-valence forms of neptunium, plutonium and americium and stability of high-valence forms of these elements were studied in detail. In many cases, the transitions between the oxidation states V, VI and VII proceed easily because they do not require sub- stantial structural rearrangements of the ions. By contrast, the transition IV?V giving rise to an `yl' group proceeds slowly.The proposed mechanisms of redox processes are consistent with the concepts of general chemistry. However, many of these mecha- nisms call for experimental or quantum-chemical verification. Among the reactions to be studied, the following reactions, in our opinion, are of most interest: disproportionation of Pu(V) in a wide range of concentrations of bicarbonate and carbonate ions; the reactions of Np(IV), Pu(IV) and Am(IV) with ozone in carbonate solutions; oxidation of Np(V) and Pu(V) with ferricya- nide ions in concentrated carbonate solutions; reactions of Np(VI) and Pu(VI) with hydrogen peroxide in bicarbonate-carbonate solutions; and disproportionation of Pu(VI) in concentrated alkaline solutions (if this process does not occur, it is necessary to establish the causes).Problems of the preparation of macroamounts of curium(VI) (attempts to solve this problem were unsuccessful 164) and Pu(VIII) are of particular interest. It was suggested 165 that Pu(VIII) can be produced upon b decay of 239Np(VII) in the solid state 239PuO4¡ 239NpO657 6 . In this connection, the fundamental question arises as to the form of existence of Pu(VIII) and its stability. The estimated potential of the Pu(VIII)/Pu(VII) pair is higher than 3 V,166 which excludes the existence of Pu(VIII) in solution or the gaseous phase. However, it was reported 167, 168 that plutonium was oxidised with molecular oxygen to PuO4 under the conditions of gas thermo- chromatography.This compound is comparable in volatility to octavalent osmium oxide (OsO4). Undoubtedly, there is a need to V P Shilov, A B Yusov carry out further experiments on the determination of the compo- sition of the volatile plutonium compound, including studies with the use of modern methods of analysis, which allow one to determine the structures and compositions of specimens with a weight of several nanograms (in particular with the use of X-ray synchrotron radiation).169 References 1. D L Clark, D E Hobart,M P Neu Chem.Rev. 95 25 (1995) 2. V F Peretrukhin, V P Shilov, A K Pikaev Alkaline Chemistry of Transuranium Elements and Technetium and the Treatment of Alkaline Radioactive Wastes (Report WHC-EP-0817 UC-601) (Richland, Washington, DC: Westinghouse Hanford Co., 1995) 3.A K Pikaev, V P Shilov, A V Gogolev Usp. Khim. 66 845 (1997) [Russ. Chem. Rev. 66 763 (1997)] 4. A B Yusov, V P Shilov Izv. Akad. Nauk, Ser. Khim. 1957 (2000) a 5. D W Wester, J C Sullivan Inorg. Chem. 19 2838 (1980) 6. W A Mulac, S Gordon, K N Schmidt, D Wester, J C Sullivan Inorg. Chem. 23 1639 (1984) 7. E Haltier, B Fourest, F David Radiochim. Acta 51 107 (1990) 8. A M Fedoseev, V F Peretrukhin, N N Krot Dokl. Akad. Nauk SSSR 244 1187 (1979) b 9. C Musicas, G Rousset Note CEA 1419 431 (1971) 10. J J Katz, G T Seaborg, L Morss (Eds), in The Chemistry of Actinide Elements Vol. 2 (London; New York: Chapman and Hall, 1986) p. 1143 11. D W Wester, J C Sullivan J.Inorg. Nucl. Chem. 43 2919 (1981) 12. P G Varlashkin, D E Hobart, G M Begun, J R Peterson Radiochim. Acta 35 91 (1984) 13. G A Simakin, Yu F Volkov, G I Visyashcheva, I I Kapshukov, P F Baklanova, G N Yakovlev Radiokhimiya 16 859 (1974) c 14. A J Zielen, D Cohen J. Phys. Chem. 74 394 (1970) 15. G A Simakin, I V Matyashchuk,N A Vladimirova Radiokhimiya 15 98 (1973) c 16. V P Shilov, N N Krot, A D Gel'man Radiokhimiya, 12 697 (1970) c 17. V F Peretrukhin, D P Alekseeva Radiokhimiya 16 836 (1974) c 18. V P Shilov, A B Yusov Radiokhimiya 39 503 (1997) c 19. V F Peretrukhin, V I Spitsyn Izv. Akad. Nauk SSSR, Ser. Khim. 4 826 (1982) a 20. V F Peretrukhin, N N Krot, A D Gel'man Radiokhimiya 14 72 (1972) c 21. V S Ermakov, V F Peretrukhin, N N Krot Radiokhimiya 19 324 (1977) c 22.V S Ermakov, V F Peretrukhin, N N Krot Radiokhimiya 19 253 (1977) c 23. D W Wester, J C Sullivan Radiochem. Radioanal. Lett. 57 35 (1983) 24. P G Varlashkin, G M Begun, J R Peterson Radiochim. Acta 35 211 (1984) 25. J Bourges Radiochem. Radioanal. Lett. 12 111 (1972) 26. V F Peretrukhin, D P Alekseeva Radiokhimiya 16 843 (1974) c 27. N A Bubantseva, I G Tananaev, A M Fedoseev, C H Delegard J. Alloys. Comp. 271 ± 273 813 (1998) 28. V P Shilov Radiokhimiya 18 688 (1976) c 29. G A Simakin Radiokhimiya 19 518 (1977) c 30. J Y Bourges, B Guillaume, G Koehly, D E Hobart, J R Peterson Inorg. Chem. 22 1179 (1983) 31. D E Hobart, K Samhoun, J R Peterson Radiochim. Acta. 31 139 (1982) 32.V Ya Frenkel', I A Lebedev, P L Khizhnyak, B F Myasoedov Radiokhimiya 27 405 (1985) c 33. V Ya Frenkel', I A Lebedev, P L Khizhnyak, B F Myasoedov Radiokhimiya 27 576 (1985) c 34. P Berger, P Blank, J Bourges Radiochim. Acta 43 217 (1988) 35. V B Nikolaevskii, V P Shilov, N N Krot, V F Peretrukhin Radiokhimiya 17 426 (1975) c 36. V B Nikolaevskii, V P Shilov, N N Krot Radiokhimiya 16 122 (1974) c 37. V F Peretrukhin, V B Nikolaevskii, V P Shilov Radiokhimiya 16 833 (1974) c 38. L L Borin, A I Karelin Termodinamika Okislitel'no-Vosstanovi- tel'nykh Protsessov v Tekhnologii Aktinidov (Thermodynamics ofRedox reactions of actinides in carbonate and alkaline solutions Redox Processes in the Technology of Actinides) (Moscow: Atomizdat, 1977) 39.M Vokovic', D Cukman, V Pravdic' J. Electroanal. Chem. 54 209 (1974) 40. V P Shilov, A V Gogolev, A K Pikaev Dokl. Akad. Nauk 375 362 (2000) b 41. V P Shilov, A B Yusov Radiokhimiya 35 (5) 61 (1993) c 42. L M Frolova, V M Chistyakov, V M Ermakov, A G Rykov Radiokhimiya 18 281 (1976) c 43. V M Chistyakov, V M Ermakov, A G Rykov Radiokhimiya 16 553 (1974) c 44. V P Shilov, A B Yusov Radiokhimiya 37 201 (1995) c 45. V P Shilov, E S Stepanova, N N Krot Radiokhimiya 18 350 (1976) c 46. V P Shilov, A B Yusov Radiokhimiya 37 203 (1995) c 47. V P Shilov, L N Astafurova, A B Yusov Radiokhimiya 37 507 (1995) c 48. E A Kanevskii, I V Goncharov, V B Rengevich Radiokhimiya 7 579 (1965) c 49. V Hirsch, K H Lieser Radiochim. Acta 44/45 Pt 1 7 (1988) 50.A P Filippov, V N Fedoruk Radiokhimiya 32 (1) 15 (1990) c 51. J S Coleman, T K Keenan, L N Jones,W T Carnall, R A Penneman Inorg. Chem. 2 58 (1963) 52. S W Rabideau, B J Masters J. Phys. Chem. 67 318 (1963) 53. G Gordon, H Taube Inorg. Chem. 1 69 (1962) 54. A R Felmy, D Rai, R W Fulton Radiochim. Acta 50 193 (1992) 55. C Keller, D Fang Radiochim. Acta 11 123 (1969) 56. L M Frolova, V M Chistyakov, V M Ermakov, A G Rykov Radiokhimiya 17 410 (1975) c 57. I M Kolthoff, I K Miller J. Am. Chem. Soc. 73 3055 (1951) 58. L M Frolova, V M Chistyakov, V M Ermakov, A G Rykov Radiokhimiya 18 276 (1976) c 59. V P Shilov, L N Astafurova, A B Yusov Radiokhimiya 37 509 (1995) c 60. V P Shilov, N N Gavrilova Radiokhimiya 22 603 (1980) c 61.P Neta, R E Huie J. Phys. Chem. Ref. Data 17 1027 (1988) 62. A V Gogolev, V P Shilov, A K Pikaev Izv. Akad. Nauk SSSR, Ser. Khim. 20 (1989) a 63. M P Mefod'eva, N N Krot, A D Gel'man, T V Afonas'eva Radiokhimiya 18 93 (1976) c 64. V M Ermakov, V M Chistyakov, L M Frolova, A G Rykov Radiokhimiya 19 234 (1977) c 65. V M Ermakov, A G Rykov, G A Timofeev, Ashok V Dzhadov, G N Yakovlev Radiokhimiya 15 380 (1973) c 66. V M Chistyakov, V M Ermakov, A D Mokrousov, A G Rykov Radiokhimiya 16 810 (1974) c 67. A V Gogolev, V P Shilov, A M Fedoseev, A K Pikaev Izv. Akad. Nauk SSSR, Ser. Khim. 1 28 (1990) a 68. A K Pikaev, V P Shilov, V I Spitsyn Dokl. Akad. Nauk SSSR 232 387 (1977) b 69. P L Khizhnyak, Author's Abstracts of Candidate Thesis in Chemical Sciences, V I Vernadsky Institute of Geochemistry and Analytical Chemistry, Academy of Sciences of the USSR, Moscow, 1987 70.G T Seaborg, J J Katz (Eds) The Actinide Elements (New York; Toronto; London: McGraw-Hill, 1954) 71. A B Yusov, A V Anan'ev, V P Shilov Izv. Akad. Nauk, Ser. Khim. 953 (1994) a 72. V P Shilov, A B Yusov Radiokhimiya 36 396 (1994) c 73. A V Anan'ev, V P Shilov, A B Yusov Radiokhimiya 36 247 (1994) c 74. V D Klimov, V N Prussakov, V B Sokolov Radiokhimiya 13 725 (1971) c 75. Yu K Gusev, V F Peretrukhin, V P Shilov, N S Kirin Radiokhimiya 14 888 (1972) c 76. E H Appelman Inorg. Chem. 6 1305 (1967) 77. E H Appelman Inorg. Chem. 10 1881 (1971) 78. A A Goncharov, Yu N Kozlov, A P Purmal' Zh.Fiz. Khim. 55 1633 (1981) d 79. E H Appelman, J G Malm J. Am. Chem. Soc. 86 2141 (1964) 80. A B Yusov, V P Shilov Radiokhimiya 34 (6) 31 (1992) c 81. V P Shilov, A B Yusov Radiokhimiya 35 (1) 38 (1993) c 82. A B Yusov, V P Shilov Radiokhimiya 35 (2) 38 (1993) c 83. U K KlaÈ ning, K Sehested, T Wolff, E H Appelman J. Chem. Soc., Faraday Trans. 1 1539 (1982) 487 84. P A G O'Hare, G K Johnson, E H Appelman Inorg. Chem. 9 332 (1970) 85. V F Peretrukhin, E A Erin, V I Dzyubenko, V V Kopytov, V G Polyukhov, V Ya Vasil'ev, G A Timofeev, A G Rykov, N N Krot, V I Spitsyn Dokl. Akad. Nauk SSSR 242 1359 (1978) b 86. V N Kosyakov, E A Erin, V M Vityutnev, V V Kopytov, A G Rykov Radiokhimiya 24 551 (1982) c 87. V P Shilov Zh. Fiz. Khim. 70 1915 (1996) d 88. V P Shilov, A B Yusov Radiokhimiya 41 416 (1999) c 89.K Nash,M E Noon, S Fried, J C Sullivan Inorg. Nucl. Chem. Lett. 16 33 (1980) 90. J C Sullivan,W Brown, K Nash,M Thompson Radiochim. Acta 30 45 (1982) 91. J Beitz, C Johan, J C Sullivan, M Woods Radiochim. Acta 40 7 (1986) 92. C J Anderson,M Woods, E Deutsch, K Deutsch, J C Sullivan Radiochim. Acta 61 137 (1993) 93. N N Krot, M P Mefod'eva, V P Shilov, A D Gel'man Radiokhimiya 12 471 (1970) c 94. V S Ermakov, T N Bukhtiyarova, N N Krot Radiokhimiya 19 250 (1977) c 95. I G Tananaev Radiokhimiya 32 (1) 20 (1990) c 96. V B Nikolaevskii, V P Shilov, N N Krot, V F Peretrukhin Radiokhimiya 17 431 (1975) c 97. V P Shilov Radiokhimiya 39 330 (1997) c 98. A K Pikaev, V P Shilov Izv.Akad. Nauk SSSR, Ser. Khim. 9 2136 (1978) a 99. V P Shilov, N N Krot, A D Gel'man Radiokhimiya 12 911 (1970) c 100. C W Williams, J-P Blaudeau, J C Sullivan, M R Antonio, B Bursten, L Soderholm J. Am. Chem. Soc. 123 4346 (2001) 101. N N Krot, V P Shilov, V B Nikolaevskii, A K Pikaev, A D Gel'man, V I Spitsyn Dokl. Akad. Nauk SSSR 217 589 (1974) b 102. V P Shilov Radiokhimiya 18 659 (1976) c 103. N A Budantseva, V P Shilov, N N Krot Radiokhimiya 40 544 (1998) c 104. V P Shilov, L N Astafurova, A Yu Garnov, A B Yusov, N N Krot Radiokhimiya 39 437 (1997) c 105. V P Shilov, L N Astafurova, A Yu Garnov, N N Krot Radiokhimiya 38 231 (1996) c 106. R A Penneman, L B Espri, in Materialy 1-oi Mezhdunarodnoi Konferentsii po Mirnomu Ispol'zovaniyu Atomnoi Energii, Zheneva, 1955 (Proceedings of the First International Conference on Peaceful Use of Nuclear Energy, Geneva, 1955) (Moscow: GNTIKhL, 1959) Vol. 7, p. 435 107. K Buijs, K Louwrier J. Inorg. Nucl. Chem. 28 2463 (1966) 108. N N Krot, A D Gel'man Dokl. Akad. Nauk SSSR 177 124 (1967) b 109. N N Krot, M P Mefod'eva, T V Smirnova, A D Gel'man Radiokhimiya 10 412 (1968) c 110. V I Blokhin, N N Krot, A D Gel'man Radiokhimiya 12 792 (1970) c 111. D Cohen, S Fried Inorg. Nucl. Chem. Lett. 5 653 (1969) 112. E A Shashukov, A A Kozlov Radiokhimiya 12 237 (1970) c 113. M P Mefod'eva, N N Krot, T V Afonas'eva Radiokhimiya 19 245 (1977) c 114. A V Gogolev, V P Shilov, A K Pikaev Radiokhimiya 31 124 (1989) c 115. V P Shilov, E S Stepanova, N N Krot Radiokhimiya 19 80 (1977) c 116. A A Chaikhorskii,M Yu Matuzenko, Yu I Belyaev Radiokhimiya 16 850 (1974) c 117. D S Gorbenko-Germanov, I V Kozlova Zh. Fiz. Khim. 48 1998 (1974) d 118. L Forni, D Bathemann, E J Hart J. Phys. Chem. 86 255 (1982) 119. Yu A Komkov, N N Krot, A D Gel'man Radiokhimiya 10 625 (1968) c 120. Yu A Komkov, N N Krot Radiokhimiya 12 227 (1970) c 121. R A Penneman, J S Coleman, T K Keenan J. Inorg. Nucl. Chem. 17 138 (1961) 122. V P Shilov, N N Krot, A D Gel'man Radiokhimiya 13 9 (1971) c 123. A B Yusov, A Yu Garnov, V P Shilov, N N Krot Radiokhimiya 39 515 (1997) c 124. Yu A Komkov, V F Peretrukhin, N N Krot, A D Gel'man Radiokhimiya 11 407 (1969) cV P Shilov, A B Yusov 488 125. A B Yusov, A Yu Garnov, V P Shilov, N N Krot Radiokhimiya 39 519 (1997) c 126. A Yu Garnov, A B Yusov, V P Shilov, N N Krot Radiokhimiya 40 18 (1998) c 127. V P Shilov, A Yu Garnov, A V Gogolev Radiokhimiya 40 308 (1998) c 128. B B Cunningham, in The Transuranium Elements (Eds G T Sea- borg, J J Katz) (New York; Toronto; London: McGraw-Hill, 1949) 167. V P Domanov, G V Buklanov, Yu V Lobanov, in Tret'ya Rossiiskaya Konferentsiya po Radiokhimii (Tez. Dokl.) [Third Russian Conference on Radiochemistry (Abstracts of Reports)] (St Petersburg: State Unitary Enterprises Research and Production Union `Radievyi Institut im. V G Khlopina', 2000) p. 37 168. V P Domanov, G V Buklanov, Yu V Lobanov, in Actinides-2001 International Conference (Final Program and Abstracts), Hayama, Japan, 2001 p. 166 169. Ya V Zubavichus, Yu L Slovokhotov Usp. Khim. 70 429 (2001) [Russ. Chem. Rev. 70 373 (2001)] Vol. 14B, p. 1363 129. V I Spitsyn, N N Krot, M P Mefod'eva, A D Gel'man Dokl. Akad. Nauk SSSR 181 128 (1968) b 130. V P Shilov, N N Krot, A D Gel'man Radiokhimiya 12 896 (1970) c a�Russ. Chem. Bull., Int. Ed. (Engl. Transl.) b�Dokl. Chem. (Engl. Transl.) c�Radiochemistry (Engl. Transl.) d�Russ. J. Phys. Chem. (Engl. Transl.) e�High Energy Chem. (Engl. Transl.) 131. A Yu Garnov, A B Yusov, V P Shilov, N N Krot Radiokhimiya 40 213 (1998) c 132. N N Krot, M P Mefod'eva, T V Afonas'eva Radiokhimiya 19 230 (1977) c 133. N N Krot, A D Gel'man, M P Mefod'eva, V P Shilov, V F Peretrukhin, V I Spitsyn Semivalentnoe Sostoyanie Neptuniya, Plutoniya, Ameritsiya (Heptavalent States of Neptunium, Plutonium and Americium) (Moscow: Nauka, 1977) 134. Yu M Kulyako, T I Trofimov, D A Malikov, I A Lebedev, B F Myasoedov Radiokhimiya 35 (5) 68 (1993) c 135. S G Bratsch J. Phys. Chem. Ref. Data 18 1 (1989) 136. V F Peretrukhin, N N Krot, A D Gel'man Radiokhimiya 13 772 (1971) c 137. O B Shatokhina, D P Alekseeva, V F Peretrukhin, N N Krot, N S Kirin Radiokhimiya 19 824 (1977) c 138. P K Bhattacharyya, R Veeraghavan, R D Saini Radiochim. Acta 30 217 (1982) 139. A V Gogolev, V P Shilov, A K Pikaev Khim. Vys. Energ. 30 255 (1996) e 140. V P Shilov, A V Gogolev, A K Pikaev Radiokhimiya 40 311 (1998) c 141. V P Shilov, N A Budantseva Radiokhimiya 40 546 (1998) c 142. V P Shilov, A V Gogolev, A K Pikaev Khim. Vys. Energ. 32 395(1998) e 143. V P Shilov, V B Nikolaevskii, N N Krot Radiokhimiya 37 (1) 32 (1995) c 144. W D Felix, B L Call, L M Dorfman J. Phys. Chem. 71 384 (1967) 145. V P Shilov, N N Krot Radiokhimiya 11 399 (1969) c 146. A Yu Garnov, A B Yusov, V P Shilov, N N Krot Radiokhimiya 39 506 (1997) c 147. A V Gelis, A V Garnov, V P Shilov Radiokhimiya 40 22 (1998) c 148. A V Gelis, A V Garnov, V P Shilov Radiokhimiya 40 110 (1998) c 149. A B Yusov, A Yu Garnov, V P Shilov, N N Krot Radiokhimiya 39 509 (1997) c 150. V P Shilov, A Yu Garnov, N N Krot, A B Yusov Radiokhimiya 39 513 (1997) c 151. A A Bessonov, A V Gelis, V P Shilov, N N Krot Radiokhimiya 41 516 (1999) c 152. Yu A Komkov, N N Krot, A D Gel'man Radiokhimiya 12 692 (1970) c 153. A Yu Garnov, A V Gelis, A A Bessonov, V P Shilov, N N Krot Radiokhimiya 40 218 (1998) c 154. I G Tananaev Radiokhimiya 31 62 (1989) c 155. I G Tananaev Radiokhimiya 32 (5) 46 (1990) c 156. I G Tananaev Radiokhimiya 32 (2) 7 (1990) c 157. I G Tananaev Radiokhimiya 32 (5) 43 (1990) c 158. I G Tananaev Radiokhimiya 32 (5) 50 (1990) c 159. I G Tananaev Radiokhimiya 32 (1)23 (1990) c 160. I G Tananaev Radiokhimiya 34 (1) 108 (1992) c 161. A Yu Garnov, A V Gelis, A A Bessonov, V P Shilov Radiokhimiya 40 309 (1998) c 162. V I Dzyubenko, V F Peretrukhin Radiokhimiya 19 832 (1977) c 163. V I Silin, A V Kareta J. Alloys Comp. 271 ± 273 803 (1998) 164. V P Shilov, A B Yusov Radiokhimiya 35 (3) 5 (1993) c 165. V I Spitsyn, N N Krot Izv. Akad. Nauk SSSR, Ser. Khim. 4 821 (1982) a 166. G V Ionova, V G Pershina, V I Spitsyn Elektronnoe Stroenie Aktinidov (Electronic Structure of Actinides) (Moscow: Na
ISSN:0036-021X
出版商:RSC
年代:2002
数据来源: RSC
|
3. |
Cryogels on the basis of natural and synthetic polymers: preparation, properties and application |
|
Russian Chemical Reviews,
Volume 71,
Issue 6,
2002,
Page 489-511
Vladimir I. Lozinsky,
Preview
|
|
摘要:
Russian Chemical Reviews 71 (6) 489 ± 511 (2002) Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications V I Lozinsky Contents I. Introduction II. What are cryotropic gelation and cryogels? III. The structure of frozen solutions of low- and high-molecular-weight compounds. The notion of non-frozen liquid microphase (NFLMP) IV. Effects inherent in cryogel formation processes V. Applications of cryotropic gelation processes and materials based on polymeric cryogels VI. Conclusion Abstract. polymeric in gelation cryotropic with deals review This This review deals with cryotropic gelation in polymeric systems, in storing freezing, moderate on place takes which systems, which takes place on moderate freezing, storing in a frozen state and subsequent thawing of solutions or colloidal frozen state and subsequent thawing of solutions or colloidal dispersions precursors. polymeric or monomeric containing dispersions containing monomeric or polymeric precursors.The The polymer materials formed under these conditions are termed as polymer materials formed under these conditions are termed as cryogels. which in gels heterophase macroporous are They cryogels. They are macroporous heterophase gels in which poly- poly- crystals gel during porogens as act solvent frozen the of crystals of the frozen solvent act as porogens during gel formation. formation. The moderately of structure the about knowledge current The current knowledge about the structure of moderately frozen frozen solutions is compounds high-molecular-mass and low- of solutions of low- and high-molecular-mass compounds is consid- consid- ered, gelation cryotropic the of regularities general the ered, the general regularities of the cryotropic gelation processes processes and at gelation from differences and to similarity their and their similarity to and differences from gelation at positive positive temperatures of application of fields the discussed, are temperatures are discussed, the fields of application of cryotropic cryotropic gelation techniques are described, and examples of using poly- gelation techniques are described, and examples of using poly- meric problems applied solving for materials cryogel-based meric cryogel-based materials for solving applied problems are are given.references 433 includes bibliography The given. The bibliography includes 433 references. I. Introduction The applications of processes based on freezing are quite diverse. They include food technology,1 ground freezing for the construc- tion of underground utilities, the formation of cryobanks for storage of biological objects (blood, semen), ice preparation in sports centres, cryoconcentration of brines in chemical technology and fruit juices during processing of agricultural products, freeze- drying (lyophilisation) of drugs and many other. Branches of science such as low-temperature physics, cryobiology and cryo- medicine, geocryology, glaciology, etc., are developing. Cryo- chemistry is in progress,2 including those fields concerned with frost resistance of polymers, low-temperature polymerisation, cryoresistance of latexes, etc.This review is devoted to one line of research in polymer cryochemistry, namely, cryotropic gelation processes. Despite the fact that systematic studies along this line were started only in the 1970s, the number of patented processes and materials has already exceeded several hundred. However, no surveying papers have been published so far. All known reviews V I Lozinsky A N Nesmeyanov Institute of Organoelement Compounds, Russian Academy of Sciences, ul. Vavilova 28, 119991 Moscow, Russian Federation. Fax (7-095) 135 50 85. Tel. (7-095) 135 64 92. E-mail: loz@ineos.ac.ru Received 2 April 2002 Uspekhi Khimii 71 (6) 559 ± 585 (2002); translated by Z P Bobkova #2002 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2002v071n06ABEH000720 489 489 490 494 504 506 are concerned either with particular cases of gelation (for example, radiation-induced polymerisation in frozen solutions giving rise to cross-linked gel materials 3) or with only one type of gel [in particular, the macroporous gels resulting from freezing ± thawing of concentrated solutions of poly(vinyl alcohol) (PVA)4±11].The main goal of this review is to summarise and describe systemati- cally the information on the cryotropic gelation in general and on the specific features that distinguish it from the processes tradi- tionally used to prepare chemical (covalently cross-linked) and physical (non-covalent) gels.The explanation of the main notions and terms such as cryotropic gelation, cryostructuring and cryogel(s), etc., can be found in the literature.6, 9, 10, 12 ± 14 In addition, the author pro- poses that the processes yielding gels of various natures would be termed by words with Greek or Latin roots constructed as follows: `...+tropic' [from tropos (tropos), which implies directed or induced 12], e.g., chemotropic � induced by the formation of chemical bonds between polymer chains and resulting in a three- dimensional covalent network of the gel; ionotropic�induced by ion exchange (the formation of ionic bonds); chelatotropic � induced by chelation; thermotropic � induced by heating; psy- chrotropic � induced by cooling without freezing [from ciwria (psychria), which means chill 12], by analogy with the well-estab- lished terms such as psychrometer, a psychrophilic microorgan- ism, etc.; cryotropic � induced by freezing; solvotropic (or solvatotropic) � induced by a change in the solvent properties (quality).The notions of positive and negative temperatures (referred normally to 0 8C) and the notions of moderately frozen systems also require more accurate definition. As applied to the objects discussed in this review, positive temperatures are those above the freezing point of any solvent, while negative temperatures are those below this point. Systems are considered to be moderately frozen if they retain a fraction of non-frozen solvent (usually, this refers to a temperature region lower than the freezing point of the neat solvent by not more than several tens of degrees).II. What are cryotropic gelation and cryogels? The structured polymeric physical bodies called gels (or lyogels, in colloid chemistry) can be characterised as immobilised solvent ± polymer systems, in which macromolecules are connected in a three-dimensional network by non-fluctuating bonds that are rather stable in time. The nature of these bonds and the network morphology (single-phase and heterophase systems are the 1st-490 and 2nd-kind gels, respectively, according to Papkov's classifica- tion 15) are determined by the chemical structure and the method of preparation of the gel.The solvent retained by the cross-linked polymer plays a very important role, as it prevents the system from collapsing into a compact mass and ensures diffusion of soluble substances from the surrounding liquid medium into the gel and back.16 ± 18 Gels, except for thixotropic ones, are normally able to undergo substantial reversible deformations in the absence of a flow.19 Two basic methods for the preparation of gels are known 15, 20 to exist (Fig. 1): first, swelling of either a polymer (as a block, a film, a fibre or a powder) or a xerogel (cross-linked polymer without a solvent) in an appropriate low-molecular-weight liquid; second, gel formation in a liquid medium (solution or colloid dispersion of the corresponding gel-forming agents).In the former case, a gel is formed due to limited swelling of a non-cross-linked polymer or swelling (exhausting swelling, if the equilibrium is reached) of a xerogel, which, in turn, can be prepared by a solvent-free synthesis or by mere drying of a lyogel. The latter case represents the most widely used route to gel formation. In this case, the initial system can be a solution of monomers in which branching polymerisation or polycondensa- tion is initiated; or a solution of high-molecular-weight com- pounds containing appropriate cross-linking agents (ionotropic and chelatotropic types of cross-linking aor a solution of a polymer converted into a physical gel due to a change in the thermodynamic quality of the solvent (by which, one means its affinity towards the polymer) or to partial crystallisation of the polymer; or colloidal sol, which undergoes a phase transition to a gel-like state.Starting from these precursors, one can prepare chemotropic (covalently cross-linked) gels, ionotropic and chelatotropic gels (in which the three-dimensional network junction knots are stabilised by slightly dissociating salt or coordination bonds, respectively), and also thermotropic and psychrotropic non- covalent (physical) gels. Ageing of colloidal sols usually gives gels of a sophisticated morphology (coagulation and condensa- tion-crystallisation gels, according to Rebinder 21). It was shown 22 that all types of gelation that take place in a solvent can also be realised in a cryogenic version for the corresponding initial systems using appropriate conditions for freezing, storing in the frozen state, and thawing (see Fig.1). Thus, cryotropic gelation is a specific type of gelation that takes place upon cryogenic treatment of the initial systems potentially capable of forming a gel. A necessary condition for the processes resulting in the formation of cryogels is crystallisation (freezing) of Non-cross- linked Swelling polymers Xerogels Gelation of a polymer ± Branching poly- merisation (poly- Bulk poly- merisation Bulk cross- linking of cross-linking agent system polymers condenstion) in solution of mono- mers in solution Figure 1. Generalised chart of the techniques for the formation of polymer gels.V I Lozinsky the main bulk of the low-molecular-weight liquid present in the initial system. The polymer gel phase can be formed in one of the stages of cryogenic treatment: either directly during freezing of the initial system [this refers, for example, to colloid dispersions of gelati- nised starch 23 or to aqueous solutions of polygalactomannan from the locust bean gum (LBG), which is a block polysaccharide giving rise to thermally reversible physical cryogels],24 ± 26 or during storage of samples in the frozen state (this mainly refers to the formation of chemically cross-linked cryogels 14, 27 ± 30), or during thawing of the frozen specimens (this is typical of non- covalent cryotropic gelation of, for example, aqueous solutions of PVA9, 31 ± 34).This will be considered in more detail below in the discussion of the cryotropic gelation kinetics. First, we shall consider the structure and properties of moderately frozen solutions of low- and high-molecular-weight compounds in order to formulate the current views on the medium in which the cryogel formation proceeds. III. The structure of frozen solutions of low- and high-molecular-weight compounds. The notion of non-frozen liquid microphase (NFLMP) The freezing and melting (thawing) of frozen solutions of various substances are fairly complex phenomena dependent on a large number of factors. Although studies in this field have long been carried out, new data providing more precise and more in-depth understanding of the mechanisms of these processes constantly appear in scientific literature.The progress of these studies is promoted to a large extent by the advent of experimental instrumentation and involvement of new methods and theoretical approaches. It is noteworthy that the traditional views according to which frozen solutions are polycrystalline solids at temper- atures below the eutectics cannot explain, for example, the occurrence of chemical reactions with diverse mechanisms in these systems. These views do not allow interpretation of the spectroscopic data (ESR, NMR) concerning the maintenance of the high mobility of some of the solute and solvent molecules down to temperatures tens of degrees lower than the freezing point of the initial system.The cryotropic solidification of neat crystallisable liquids and solutions of low-molecular-weight compounds (LMWC) in such liquids has been studied most extensively, mainly under condi- tions close to equilibrium.35 ± 37 However, freezing of these objects under conditions far from equilibrium brings about specific Lyogels Gelation in a solvent Gelling of a polymer solution upon the loss Gelation of a polyelectrolyte ± Ageing of a colloid of thermodynamic quality of the solvent; cross-linking counter-ions sol system in solution partial crystallisation of the polymer Freezing and thawing CryogelsCryogels on the basis of natural and synthetic polymers: preparation, properties and applications effects, depending on the cooling rate, the specimen geometry, the vessel material, the duration of storage in the frozen state, and lots of other factors.38 ± 41 The processes taking place during the cryogenic treatment of solvent ± polymer, solvent ± polymer ± mo- nomeric additive, and solvent ± swollen gel systems are even more complex.The course of freezing or thawing of such systems is strongly influenced by the polydispersity of the polymers (an exception is represented by narrow homogeneous fractions of molecular-weight standards or purified protein specimens). This gives rise to non-equivalence of the liquid ± solid phase diagrams for the same solvent ± polymer pair characterised by different molecular weights (MW) or molecular-weight distributions (MWD) of the polymeric component.Due to the polymer poly- dispersity, Smith and Pennings 42 proposed that the solvent ± pol- ymer systems subjected to freezing should be called pseudo- binary, to distinguish them from the solvent ±LMWC binary systems. In addition, due to the slow relaxation processes in high-viscosity media, the attained phase states of the frozen specimens depend appreciably on the thermal pre-history during freezing. The capability for association in solutions at low temper- atures, inherent in many polymers, results in the situation that the phase diagram of one system at the same negative temperature would depend in each particular case on the dynamics of cooling of the given specimen and the rate of solvent crystallisation. In addition, high rates of liquid crystallisation in solutions and, especially, in swollen gels bring about the risk of mechanical destruction of polymer chains (so-called cryocracking) under the action of shear or wedging stresses in the zones of movement of the solvent crystallisation fronts (see, for example, Refs 43 ± 46).Below we mainly consider the phenomena that are significant for the cryotropic gelation processes. The anomalous depression of the freezing temperature of cross-linked gels 44, 47 ± 52 or the effect of the polymer MW and concentration on the trend of its solution to undergo either crystallisation or glass transition on cooling 53 are beyond the scope of this review, as well as some other aspects concerning freezing of polymer solutions and gels but not related directly to cryostructuring processes. 1.The liquid microphase in frozen multicomponent solutions The pioneering studies demonstrating the possibility of chemical reactions in moderately frozen solutions (i.e., in systems retaining a fraction of non-frozen solvent) were published in the 1960s.54 ± 60 Grant et al.54 discovered a higher (compared to that at room temperature) degree of hydrolysis of the b-lactam ring in penicil- lin G and 6-aminopenicillanic acid in the presence of alkaline catalysts in frozen aqueous solutions of the reactants. The highest degree of conversion was observed at 718 8C. Study of the reaction products showed 55 that after thawing, the solution of 6- aminopenicillanic acid contained 7 ± 8-unit oligomeric peptides.These were not formed in a similar solution kept at 20 8C for the same period. Thus, oligomerisation took place only on cryogenic treatment. Similar results have also been obtained in the synthesis of oligopeptides by polycondensation of N-carboxyanhydrides of several amino acids in frozen dioxane (Tm=11.4 8C).56 This process was most efficient in the crystallised reaction system at 1 8C rather than at positive temperatures. Butler and Bruice 57, 58 were, apparently, the first to suggest that the systems obtained by moderate freezing of multicompo- nent solutions ofLMWCcontain, after macrosolidification, apart from the solid phase, liquid regions between the solvent crystals in which the reactants are accumulated. Subsequently, this idea was further developed in Pincock's and Kiovsky's works (see, for example, reviews 59, 60).Having studied the kinetics of some usual (i.e., not involving oligomer or polymer formation) reactions of LMWC in frozen solvents (both water and organic solvents), the researchers concluded that these reactions proceed in `liquid holes, present among the crystalline solid, which contain high concen- trations of the reactants'.61 This early stage of investigation resulted in the following conclusions:60 491 (1) moderate freezing of a solution of reactants in a crystallis- able solvent does not terminate the reaction because substances are accumulated in the liquid regions; over a certain range of negative temperatures, this may even accelerate the reaction; (2) if the total volume of the liquid inclusions remains unchanged during the reaction (the temperature does not change and the products and the reactants are in solution), the observed (apparent) rate (vobs) of substance transformation can be described (with good agreement with experimental results) by the usual kinetic equation (1) V vobs= 1 dm dt , where V is the volume of the reaction phase, i.e., the total volume of liquid inclusions; m is the number of moles of reactants in the reaction phase; and t is time.With vobs determined by analytical methods, it is possible to estimate the V value; (3) the dependence of the reaction rate on temperature for moderately frozen solutions does not obey the Arrhenius equation and passes through an extremum; (4) reactions in moderately frozen media differ fundamentally from solid-phase reactions because the reactant molecules in the regions that remain liquid can execute forward motion, which is hardly possible in a solid (at both macroscopic and microscopic levels).A significant contribution to the development of the concept of structural and phase inhomogeneity of multicomponent frozen solutions (in particular, at temperatures below eutectic ones) was made by Sergeev and Batyuk (see reviews 2, 62 ± 64). They confirmed that frozen solutions actually contain regions, the mobility of molecules in which is comparable with that in liquids, and developed kinetic models suitable for quantitative description of reactions proceeding in such systems.Ultimately, it was established quite definitely that moderately frozen solvent ±LMWC systems are not entirely solid not only in the phase diagram region between the liquidus and solidus lines but also well below this region. In non-solidified regions which Sergeev and coworkers called the `non-frozen liquid micro- phase' 65 (NFLMP), the molecules possess high mobility; this shows itself as narrow signals in high-resolution NMR spectra. If the reactants are readily soluble, they are accumulated (in the limiting case, completely) in the NFLMP. From the thermody- namic standpoint, the formation of the liquid microphase in multicomponent frozen systems stems from the fact that `inclu- sion of dissolved substances in the crystal lattice of a solid solvent requires more energy than that spent for the increase in the chemical potentials upon an increase in the concentrations of components in the liquid microphase'.62 The existence of NFLMP in frozen solutions of monomeric substances at temperatures below the eutectic point of the corresponding system can be detected, in particular, by examining the NMR spectra of dissolved substances 66, 67 or ESR spectra of spin probes.68 ± 70 For example, the triplet signal for 2,2,6,6- tetramethyl-4-hydroxypiperidin-1-oxyl in liquid benzene (Tm=5.5 8C) remains such in a frozen sample down to *750 8C and degenerates completely into a singlet only at liquid nitrogen temperatures, due to the fact that the solid microenviron- ment of the nitroxide probe suppresses its rotation.62 It was found 71 that hydrolysis of p-nitrophenyl acetate in aqueous salt systems (solutions of Na2HPO4 or Na2B4O7) with an eutectic composition prepared beforehand occurs at temperatures at least 10 8 lower than the eutectic point; the liquid-phase nature of this process was proved by special experiments.After a thermodynamic equilibrium has been reached, frozen systems in which no reactions take place are characterised by the same NFLMP composition independent of the way of freezing of the initial solution.65 However, during cooling of the initial system to a specified negative temperature, every instant of time is matched by a particular phase state, which depends on the conditions of cooling.The farther the cooling conditions from492 j¡1 j (2) equilibrium, the more pronounced the changes which occur in the frozen system during relaxation to the thermodynamic equili- brium. In solutions of polymers frozen under non-equilibrium conditions, relaxation proceeds much more slowly than that in LMWC systems. It is noteworthy that the liquid ± solid diagrams constructed for such non-equilibrium conditions of freezing (this is almost always the case for chemical reactions carried out in moderately frozen solutions) are called non-equilibrium state diagrams rather than phase diagrams,39 and the former can differ appreciably from the equilibrium diagrams. All these specific features of the phase state of systems subjected to freezing are especially important for those cryochemical processes whose rates are commensurable with the dynamics of crystallisation of the initial solution in which the amount (volume) and composition of the liquid microphase vary continuously during freezing until a phase equilibrium is reached. Thorough experimental research of a number of cryochemical reactions in frozen multicomponent solutions and theoretical analysis of the results made it possible to derive kinetic equation (2), which differs from the above Eqn (1), referring to the simplest case, and allows the calculation of the reaction rates in these systems.2, 62 ± 64 vobs=k0e¡E=RT PjCnj j , Pn 0 ¡ TCi # "D i TX where k0 is the pre-exponential factor; E is the activation energy of the reaction determined for the liquid medium; R is the absolute gas constant; T0 is the crystallisation point for the neat solvent; T is the temperature of the frozen sample; D is the cryoscopic coefficient of the solvent; Ci is the concentration of substances (i components) in the initial solution; Cj is the concentration of a jth reactant in the non-frozen region of the system; nj is the reaction order with respect to the jth reactant. The structure ± kinetic model, which underlies this theory of chemical processes in frozen multicomponent solutions, interprets the non-monotonic pattern of the temperature dependence of the rates of reactions (whose overall order is more than one) in these systems as being due to the competition of the opposite trends, namely, the increase in the reaction rate caused by cryoconcentra- tion and the decrease in the rate following the decrease in the temperature.For the temperature corresponding to the maximum reaction rate Tmax, the following relation was derived:2, 62 ± 64 E2 (3) ¡ ( Tmax= )0:5 R 4R2 j j nj ¡ 1 2 á nj ¡ 1 ET0 X XE ¡ j . nj ¡ 1 X 2 It can be seen from Eqn (3) that Tmax decreases with an increase in the overall reaction order. In the case where all the reaction components are concentrated in the NFLMP, and conditions for fast attainment of Tmax have been found, selective acceleration of one of the several reactions proceeding in parallel becomes possible (and this has been confirmed by experi- ments 63, 64, 72, 73).The deviation of the physicochemical characteristics of the NFLMP (viscosity, density, dielectric constant, heat capacity, etc.) from those of non-frozen solutions caused by the change in the microenvironment has an influence on the reactivity of the solutes and their capacity for association. This, in turn, has an impact on the kinetics of cryochemical reactions in frozen multi- component solutions. The data presented above lead to the conclusion that for a large number of moderately frozen solutions of individual substances and mixtures,NFLMPexists over a fairly broad range of negative temperatures.Both the initial compo- V I Lozinsky nents and the reaction products (if any reactions proceed) are accumulated in this microphase (provided that they are suffi- ciently soluble). These reactions are liquid-phase processes, despite the fact that the frozen specimen as a macroscopic body looks like a solid. The information concerning such cryochemical processes and their mechanisms available from the literature mainly deals with frozen solutions of LMWC. The behaviour of polymer ± solvent systems under these conditions is much more complex; the polydispersity of polymers (see above) is not the only reason for this difference. Another reason is the fact that polymer macro- molecules not only execute translational and rotational motions as a single whole.In addition, they (first of all, flexible-chain polymers) possess a large set of intramolecular conformational degrees of freedom, in particular, segmental mobility. This accounts for the high sensitivity of these systems to freezing regimes. 2. Non-frozen solvent in polymer solutions and gels at temperatures below the freezing point of the system. In the subsequent Sections of this review we shall demonstrate, using a number of particular examples, how moderate freezing, storage in the frozen state and subsequent thawing of a gel- forming polymer system result in the formation of a cryogel. Even the mere formation of the gel indicates that the reactants present in a frozen system (including polymers) do not lose mobility after crystallisation of the main bulk of the low-molec- ular-weight solvent.At least, they retain sufficient mobility for participation in intermolecular interactions giving rise to cross- links (either covalent or not) to form the three-dimensional cryogel network. It is clear that this type of motion can occur only in the solvent region that has not crystallised (or vitrified), i.e., in the NFLMP. It is noteworthy that most of the publications dealing with the properties of NFLMP in frozen polymeric systems concentrate on the non-frozen solvent rather than on macromolecular compo- nents because the physical methods used in the studies [most often, NMR and DSC (differential scanning calorimetry)] provide information mainly about the solvent.However, in some recent publications, the behaviour of polymers under these conditions has also been studied (see below). Since the interest in this field is related, first of all, to the development of cryobiology, cryomedi- cine and food technology, aqueous systems were studied most often,1, 9, 22, 37, 39, 63, 64, 67, 70, 74 ± 83 although Oikawa and Mura- kami 52, 84 considered organic solvent ± polymer frozen systems. It was found that, on moderate freezing of polymer solutions or gels, after crystallisation of the main bulk of the solvent, its certain fraction (which is, moreover, much greater than that in sol- vent ±LMWC systems) is not incorporated into the solid phase. Further cooling of samples may result in freezing out of an additional amount of the liquid component but, nevertheless, some remains non-frozen.Physicochemical studies (NMR, ESR, DSC) of these frozen polymeric objects show the presence of non-frozen solvent at temperatures several tens of degrees lower than the freezing point of the system. For example, this is manifested in theNMRspectra as relatively narrow signals of the corresponding nuclei.39, 77, 79, 81, 85 ± 88 The ESR spectra of the spin probes intro- duced into the polymer solution before freezing are typical of the liquid microenvironment of the probe;70, 89 ± 92 calorimetry also shows incomplete crystallisation of the initial liquid.39, 74, 76, 93 ± 111 The fraction of the solvent which remains non-frozen can be rather high.For example, for aqueous polymeric systems (sol- utions of swollen slightly cross-linked networks), the amount of non-crystallised water reaches 0.4 ± 1.0 g per g of the polymer, depending on the temperature, and the layer of non-frozen water firmly fixed by the polymer is*1 nm thick, which corresponds to two or three molecular hydration shells.39 This value is an average because it was obtained by dividing the amount of non-frozen liquid into the polymer weight in the initial system. This did notCryogels on the basis of natural and synthetic polymers: preparation, properties and applications take into account the preferred solvation by water of hydrophilic sites of the polymer chains and the tendency of hydrophobic regions to expel water, which is the case for heteropolymers such as proteins or so-called protein-like synthetic hydrophilic-hydro- phobic copolymers.112 ± 114 Therefore, the real sizes (volumes) of the areas which remain liquid after crystallisation of the main bulk of the solvent in water ± polymer systems appear to be greater.This, in turn, creates conditions for the retention of mobility of chains or chain segments in the non-frozen region of the solvent. This is confirmed by high-resolution NMR spectra of frozen solutions of PVA in D2O, which resembled the usual liquid- phase 1H and 13C NMR spectra of PVA, although the signals were markedly broadened. Thus, under these conditions, the polymer retains, at least, segmental mobility.81, 87, 88 Analogous results were obtained for frozen aqueous solutions of polyacryla- mide.67 The data indicating that some enzymes do not lose activity in moderately frozen solutions 2, 115 ± 125 also attest to the mainte- nance of segmental mobility of polymer chains under these conditions, because the transformation of a substrate into a product normally requires a conformational change in the enzyme macromolecule (especially, in the vicinity of the active site).126, 127 The volume of polymer-containing NFLMP in a particular range of negative temperatures can be estimated using a special procedure proposed by Mikhalev et al.70, 89 ± 92 for processing of the ESR spectra of spin probes (stable nitroxides) introduced into the system subjected to freezing.Hysteresis phenomena were observed,70 that is, the NFLMP volumes measured during freez- ing and thawing were unequal. g For frozen aqueous solutions of polyhydroxy polymers (in particular, polysaccharides), data on the properties of water molecules of the first solvation layer occurring in direct contact with the polymer OH groups have been obtained.These H2O molecules connected to the polymer through hydrogen bonds do not participate in the crystallisation ± melting phase transitions; when a definite negative temperature (so-called T 0 point) is attained, they are vitrified together with the polymer. This fraction of liquid is referred to as non-freezable solvent.93 ± 99 Normally, each hydroxy group of the polymer chain retains approximately one water molecule.108 In the case of polyelectrolytes, the iono- genic groups retain even larger amounts of non-freezable water due to Coulomb forces.99 It should be emphasised that the non- freezable solvent is not identical to the non-frozen solvent (which failed to crystallise under particular conditions).The amount of the latter depends on the temperature and thermal pre-history of a particular specimen. For example, for the frozen D2O±PVA system, the non-frozen solvent is responsible for a relatively narrow signal in the NMR spectrum down to a temperature 55 8 lower than the freezing point of the system. Only on further cooling, is it possible to record a broad signal from the non- freezable solvation water.81 In the general case, the non-freezable solvent is a constituent part of the non-frozen solvent.102 As noted above, the amount of solvating non-freezable solvent depends on the conformation of macromolecules, for example, on the density of packing of the polypeptide chain in the protein globule. Thus it was found by NMR spectroscopy 85 that in a solution of serum albumin frozen at735 8C, the amount of non- freezable water is 0.30 g per g of the protein.If the same macro- molecules have been treated with urea, the packing of the polypeptide chains becomes loose and the quantity of non- freezable water increases to 0.44 g per g of the protein. Thus, the higher the degree of solvation of the polymer in the initial solution, the greater the fraction of non-freezable solvent in the crystallised sample at negative temperatures, because in this case, more groups with high solvent affinity in the polymer become accessible for the solvent molecules.Indeed, determination [by DTA (differential thermomechan- ical analysis), DSC and dilatometry] of the dependence of the non- frozen portion on the initial concentration of poly(N-vinylpyrro- lidone) in aqueous solutions frozen at various temperatures 493 Table 1. Dependence of the weight of water non-frozen at745 8C on the concentration of poly(N-vinylpyrrolidone) in the initial solution.128 Concentration (%) Weight of non-frozen water (%) Relative fraction of non-frozen water /g of H2O per g of the polymer 5.0 15.0 30.9 39.2 46.4 2.18 1.19 0.80 0.77 0.62 10.9 21.0 36.6 48.9 53.6 showed that the overall weight of the liquid that has not crystal- lised increases with an increase in the polymer content but the proportion of this liquid decreases to reach some limiting value, which corresponds to the amount of non-freezable solvent (Table 1).In the case of freezing of swollen cross-linked polymers, the behaviour of the NFLMP has a number of specific features. For example, in cross-linked polyacrylamide hydrogels (PAAG), the amount of non-frozen water obeys an almost linear dependence on the temperature of the crystallised specimen in the 78 to 760 8C range.77, 129 In the subzero temperature range (0 to 78 8C), differences in the mobility of molecules in the liquid part of the solvent depending on whether the temperature of measurements was approached during cooling or heating of the specimen were detected by relaxation NMR spectroscopy.129 In the case of system cooling, the spin ± lattice relaxation time (T1) measured at76 8C was 4 times as long as T1 found for the system obtained by heating of an identical PAAG specimen from a colder state, i.e., the plot for the temperature dependence of T1 has a hysteresis loop.129 Nearly the same picture is observed at subzero temperatures for frozen gelatin or as-casein hydrogels.79 One reason for these hysteresis phenomena may be the fact that during heating, the diffusion of water molecules from the crystalline to liquid phase is hampered due to the presence of the polymer network.77 This demonstrates once again the important role of conditions (cooling or heating) used to prepare frozen specimens for phase equilibration.These effects are inherent in not only aqueous systems. A certain mobility of polymer chains in moderately frozen lyogels can also be detected in cross-linked polymers swollen in organic solvents. For example, measurement of the spin ± spin relaxation times (T2) of protons in a cross-linked rubber swollen in deutero- benzene (Tm=6.8 8C) shows that down to *733 8C, chain segments still execute thermal motion in the medium of the crystallised solvent.130 It has been shown 52 using NMR spectra of benzene-d6 (solid-echo technique) that the mobility of the solvent molecules is still retained at temperatures lower by 15 ± 18 8C.It should be emphasised that various NMR techniques are most useful for investigations of frozen solvent ± polymer systems. To conclude this Section, we would like to summarise the important features of cryotropic gelation. 1. Crystallisation of a monomeric liquid during freezing of polymer solutions is accompanied by a minor temperature depression, which differs only slightly from the crystallisation point of the pure solvent. 2. In moderately frozen multicomponent solutions, some of the low-molecular-weight liquid remains non-crystallised at tem- peratures below the eutectic point. The non-frozen solvent mole- cules retain rather high mobility; these microregions (NFLMP) of the macroscopically frozen solutions accumulate the solutes, which also possess certain molecular (in the case of LMWC) or segmental (in the case of polymers) mobility. 3.The volume of the NFLMP depends on the system temper- ature, concentration of soluble (or swollen) components and, if494 the phase equilibrium has not yet been established, on the instant of the measurement. 4. Due to cryoconcentration effects and some other factors (for example, the increase in the dielectric constant of the medium on cooling), acceleration of chemical reactions can be observed in a definite range of negative (for the used solvent) temperatures. IV. Effects inherent in cryogel formation processes As noted in the previous Section, moderate freezing of sol- vent ±LMWC and solvent ± polymer systems gives rise to phase inhomogeneous systems which include a non-frozen liquid micro- phase.If the initial specimen contains gel-forming agents, the polymer framework of the corresponding cryogel is formed exactly in these non-frozen microregions of the frozen material. This is sketched in Fig. 2.6 a 1 Freezing 2 3 c b Thawing 4 6 5 8 7 Figure 2. Basic diagram of cryotropic gelation.6, 22 (a) initial sample, (b) sample after freezing, (c) cryogel; (1) polymer precursors, (2) solvent, (3) low-molecular-weight compounds or monomer precursors, (4) polycrystals of the frozen solvent, (5) NFLMP, (6) polymer network of the gel phase of the heterophase cryogel, (7) macropores, (8) solvent.The volume of NFLMP depends on numerous factors, in particular, on the nature of the solvent and the initial concen- tration and MW of the solutes (for polymers, also on the chain conformation), the system temperature, the regime of specimen freezing, the presence of soluble or insoluble admixtures, etc. In Fig. 2, all components are shown for clarity on an arbitrary (large) scale. It has been found 67, 70, 80 ± 83 for several gel-forming systems that the NFLMP volume (Fig. 2 b) varies from fractions of a percent to several percent of the solvent volume in the initial specimen (Fig. 2 a). Thawing of the specimen gives rise to a macroporous cryogel (Fig. 2 c) containing large pores with vari- able size and geometry. The polycrystals of the frozen solvent (in some cases, crystal solvates or cryoprecipitate particles if these arise during cooling of the initial specimens) act as pore-forming agents (porogens) in this case.On melting or dissolution of these porogens after thawing, cavities filled with the melted solvent remain in the cryogel bulk. Due to the surface tension forces of the gel phase, the surface of the macropore walls is bent; the faceted shape of the walls, typical of the crystalline porogen particles, is replaced by a round shape. In addition, the wall material posseses its own (gel) porosity. Since this gel is formed at high concen- trations of precursors, the resulting polymer phase is a dense network with a microporous structure. A typical feature of various polymeric cryogels is heterophase and heteroporous (often, supermacroporous, spongy or cellular) morphology. The macropores in the cryogel thus formed are usually interconnected because during freezing, each crystal of the solvent grows, as a rule, from the vessel periphery towards the V I Lozinsky centre (if no directional freezing is involved) until it comes into contact with the face of another crystal. An increase in the precursor concentrations in the NFLMP (with respect to that in the initial system), i.e., cryoconcentration is one of the main reasons for some specific effects typical of cryogel formation.Anumber of examples classified according to the types of gel-forming systems (see Fig.1) are presented below. 1. Cryogels formed by polymerisation or polycondensation The preparation of gel structures by polymerisation or polycon- densation of monomer precursors is one of the most extensively used pathways to solvated polymer networks.15 ± 18, 131 ± 134 Appa- rently, the first description of gels synthesised by polymerisation of unsaturated monomers in frozen media, i.e., polymerisation cryogels, was described in publications 3, 135 ± 138 devoted to the g-radiation-induced gelation of frozen aqueous solutions of 2-hydroxyethyl methacrylate (2-HOEMA).It was found that at 724 8C and at an initial monomer concentration of 10%± 40%, turbid cellular (spongy) cryogels containing interconnected micrometer-size macropores are produced. When the monomer concentration is increased to 50%± 80%, a foam-plastic-like material with isolated macropores is formed, because the frozen system represents a vitrified matrix of the monomer with small ice crystals dispersed in it.When the initial content of water is low (<10%), the process yields a transparent polymeric monolith because cooling of these specimens results in vitrification instead of ice crystallisation,137 while polymerisation, evidently, occurs as a solid-phase process.139 For the irradiation doses used (*1 Mrad) post-polymerisation does not occur in any of the systems, i.e., the polymer is formed only during irradiation. The fact that a monomer with one double bond is converted into a cross-linked gel points to a branching mechanism of polymer- isation, i.e., chain transfer to a side functional group of the monomer takes place.In addition to materials based on 2-HOEMA,3, 4, 138, 140 ± 165 a similar procedure is used to prepare macroporous cryogels based on oligo(ethylene glycol) methyl ether methacrylates 1a ± d 166 and oligo(ethylene glycol) diacrylates 2a and dimethacrylates 2b.167 ± 169 O O O CO(CH2CH2O)nMe CO(CH2CH2)nOC C C CH2 H2C C H2C R R 2a,b 1a±d Me R = H (a), Me (b); n=2 ± 14. n = 9 (a), 14 (b), 23 (c), 50 (d). When the initial concentrations of the structurally related nine-unit mono- 166 and diacrylates 168 are identical, a lower swelling capacity is observed for cryogels based on the oligomer containing two double bonds, pointing to a higher degree of polymer cross-linking.The partial replacement of moderately hydrophilic 2-HOEMA in the initial composition by highly hydrophilic N-vinylpyrrolidone sharply increases the degree of swelling of the resulting cryogels in water and, as noted in a publication,170 makes the polymer sponge `softer'. An interesting way of controlling the microporosity of the polymer phase in heterogeneous cryogels of this type has been reported.171 Oligo(ethylene glycol) with an MW of 1000 Da is introduced as a `molecular' porogen into the initial solution of a mixture of acrylic acid and methyl methacrylate. After radiation- induced cryopolymerisation of the unsaturated monomers, this additive is washed out of the gel phase by a large excess of water. Apart from the usual micropores present in the polymer network, this gel contains additional (a sort of calibrated) pores.In order to synthesise cryogels with reactive aldehyde groups, acrolein is added to a solution of 2-HOEMA [or poly(oxyethylene dimethacrylates)]. Irradiation of the frozen solution of the como- nomers affords a functionalised macroporous carrier suitable forCryogels on the basis of natural and synthetic polymers: preparation, properties and applications covalent coupling of specific antibodies for the preparation of immunosorbents.172 Thus, by varying the concentration and the nature of mono- microbial mer precursors, one can change, over a wide range, the physical and chemical properties of the gels formed in the radiation- induced cryopolymerisation and used as carriers of immobilised enzymes,140, 142, 149 ± 153, 158, 160 ± 164, 171 antibodies,143 non-proteineous biopolymers 165 and cells 141, 144 ± 148, 156, 157, 159, 166 (see Section V).Systematic studies of cryotropic gelation induced by polymer- isation in frozen solutions induced by chemical initiators have been mainly carried out using polyacrylamide cryogels (cryo- PAAG).22, 29, 30, 173 ± 177 The radical copolymerisation of acryl- amide (AAM) with N,N0-methylenebis(acrylamide) (MBAAM) in water was initiated by the redox system used traditionally for this type of comonomer, namely, ammonium persulfate (APS) ±N,N,N0,N0-tetramethylethylenediamine (TMEDA), capable of generating radical ions, which initiate polymerisation even at low temperatures.178 The critical concentration for gelation (CCG) at 20 8C for a mixture of the AAM and MBAAM monomers in the synthesis of polyacrylamide gels (PAAG) amounts to*2% (depending on the monomer ratio).178 However, if the reaction mixture is frozen immediately after addition of the initiator (for example, kept at 710 8C for several hours) and thawed, it is possible to obtain cryoPAAG at a twice lower initial concentration of the same comonomers (Table 2).22, 30 Thus, polymerisation cryogels can also be formed with chemical initiation; in this case, the CCG decreases.In reality, this decrease is only apparent, the main reason being the accumulation of reactants in the polymerisation region, i.e., in the NFLMP.It is significant that this seeming decrease in the CCG is observed for any type of cryotropic gelation. Table 2. Formation of polyacrylamide gels from low-concentration aque- ous solutions of an AAM±MBAAM mixture at 20 and 710 8C in the presence of an initiator.22, 173 The ratio Formation of PAAG or cryoPAAG /mol mol71 Total concentration of AAM and MBAAM [AAM] :[MBAAM] /g dl71 710 8C 20 8C ++++++++++ 50 : 1 40 : 1 30 : 1 20 : 1 10 : 1 50 : 1 40 : 1 30 : 1 20 : 1 10 : 1 77777777+/7 + 1.0 1.0 1.0 1.0 1.0 2.0 2.0 2.0 2.0 2.0 Notes. Initiator: APS, 0.025 g dl71 and TMEDA, 0.05 g dl71. The following designations are used:7 means that the system remains liquid; +/7 means that a very weak gel is formed; + means the formation of a gel or a cryogel.Polyacrylamide cryogels have a spongy morphology, which is mainly dictated by the temperature regime of cryotropic gelation. Figure 3 presents the photomicrographs which show the structure of cryogels synthesised at 710 and 720 8C from identical solutions of comonomers frozen by different procedures: by placing the reactant solution into a cryostat with a specified temperature (freezing from above) (Fig. 3 a, b); and by initial fast freezing of the system in liquid nitrogen to7196 8C followed by transferring the specimens into a cryostat with a specified moderate negative temperature (Fig. 3 c, d). The latter procedure, which minimises the influence of the duration of cooling and 495 a b c d Figure 3.Electronic photomicrographs (scanning electron microscopy) of cryoPAAG specimens formed from aqueous solutions with a total comonomer concentration of 3% and [AAM] :[MBAAM]=30 : 1 at different conditions of cryogenic treatment.22, 176 (a) Freezing from above, 710 8C; (b) freezing from above, 720 8C; (c) low-temperature quenching, 710 8C; (d) low-temperature quenching, 720 8C. Scale: 20 mm to 1 cm. freezing stages on the kinetics of cryochemical reactions, is some- times called low-temperature quenching. When cryogels are prepared by freezing from above, lowering of the synthesis temperature from 710 to 720 8C leads to a greater number of porogen particles (small crystals of ice) having, however, a smaller size; this has a noticeable influence on macro- pore diameter in the cryoPAAG (Fig. 3 a, b).In addition, the lower the temperature, the greater the amount of solvent that freezes out, i.e., the NFLMP volume decreases, while the concen- tration of soluble substances in this phase grows. The walls of the macropores thus formed are thinner; however, according to measurements of the swelling capacity of the network material, they are built of a more concentrated gel.22, 175 When the low- temperature quenching procedure is employed, the cryoPAAG morphology changes (Fig. 3 c).22, 176 The cryogel architecture formed at 710 8C combines the porosity typical of the sample obtained by freezing from above at the same temperature and the structure of macropore walls inherent in the material prepared at 720 8C also by freezing from above. This means that, as the reacting system is warmed from 7196 to 710 8C, the main structural elements of the gel phase have time to be formed during those several minutes when the system is at about 720 8C.Simultaneously, this indicates that polymerisation is rather fast under these conditions. The porous structure of various cryogels can be characterised as an intricate three-dimensional maze. This is observed most often for aqueous systems due to the branched shape of ice polycrystals, which give rise to macropores of complex config- urations in the specimen bulk. To prepare cryogels with a rather regular arrangement of macropores, it is necessary to use direc- tional cooling and a solvent that forms oblong crystals on freezing.Thus formamide (Tm=2.8 8C) crystallises as long thin needles. When cryoPAAG is prepared in this solvent and one end of the tube with the reaction solution is cooled, directional growth of formamide crystals takes place. The resulting cryogel contains long capillary channels oriented along the temperature gra-496 dient.22, 174 Thus, the macroporosity of the polymer cryogels can be `governed' to a certain extent by selecting the solvent and crystallisation regime. The osmotic characteristics (swelling behaviour in various media) of spongy cryogels similar to cryoPAAG are mainly determined by the initial concentrations of precursors, the nature of the solvent used, and conditions of cryogenic treatment.The total volume of the liquid absorbed by a cryogel during swelling consists of two components, namely, the solvating solvent firmly kept by the polymer network and the capillary solvent which fills the macropore space and can be removed rather easily even by slight compression of the cryogel. It can be seen from the photo- micrographs shown in Fig. 3 that the latter component is mark- edly greater than the former one as the internal volume of macropores accounts for the greater part of the cryogel volume. For example, for the cryoPAAG specimen formed at 710 8C from a 3% solution of comonomers (in 40 : 1 molar ratio), the overall swelling capacity in water is*80 ml per g of dry polymer, while the degree of swelling of the macropore wall material is only *4 ml g71 (see Ref.173). The last-mentioned value corresponds to the swelling in water of usual PAAG prepared at positive temperatures from 10%± 20% solutions of the same precursors. This indicates that the polymer network of the cryogel is formed in the NFLMP, i.e., in the medium highly concentrated as compared to the initial solution. In this connection, of interest is one more specific effect inherent in the cryotropic gelation dynamics. Figure 4 shows 22, 176 the variation of the yield of the gel fraction during the PAAG and cryoPAAG formation at temperatures of 25 and 15 8C and at 715 and 725 8C. Cryopolymerisation was carried out using the freezing from above (Fig.4 a) and low-temperature quenching procedures (Fig. 4 b). A mere decrease in the temper- ature of the process in the liquid solvent from 25 to 15 8C decreases the rate of PAAG formation and the yield of the gel Yield of the gel fraction (%) 80 60 40 200 80 60 40 200 10 5 Figure 4. Dynamics of formation of PAAG and cryoPAAG (the temper- ature of synthesis is indicated near the corresponding curve) during freezing of the reacting system from above (a) and using the low-temper- ature quenching technique (b).22, 176 The initial total comonomer concentration is 3 mass %, [AAM] : [MBAAM]=30 : 1. a 715 8C 725 8C 25 8C 15 8C b 715 8C 725 8C 25 8C 15 8C 20 15 t /h V I Lozinsky fraction. Meanwhile, polymerisation in a moderately frozen system, in particular at715 8C, is much faster than, for example, at 25 8C, and the final yield of cryogels is much higher.In other words, at definite negative temperatures, gelation is accelerated compared to the gelation of analogous specimens in a non-frozen medium. This phenomenon is typical of the formation of cryogels. It is also noteworthy that the curves for the corresponding positive temperatures in Figs 4 a, b follow nearly the same pattern. This implies that the low-temperature quenching itself has no influence on the given reacting system; after fast freezing in liquid nitrogen and the subsequent fast warming of the reaction mixture, the usual gelation kinetics does not change (cf., for example, the curves corresponding to 25 8C in Figs 4 a and 4 b).The temperature dependences of the gelation rate in frozen media pass through extrema. The time it takes to reach the gel point (for given systems, this implies an approximately 10% yield of the gel fraction) actually reflects the process rate in the initial stage (Fig. 5). A decrease in the temperature in the positive region entails an increase in the gelation time, whereas in a certain range of negative temperatures in the frozen medium, the time required to reach the gel point sharply diminishes, i.e., the rate of the formation of cryoPAAG increases. In the case of freezing from above, this temperature dependence has a clear-cut extremum, the initial gelation rate being a maximum in a fairly narrow range of negative temperatures.A somewhat different pattern is observed when low-temperature quenching is used, namely, the extremum is less pronounced and a high initial rate of the process is observed over a wider temperature range. This attests to non-equivalence of the conditions of copolymerisation of AAM with MBAAM in specimens brought to the reaction temperature along different routes. In other words, the process of formation of cryoPAAG depends on the method of freezing, although special experiments with a thermocouple embedded in the frozen specimen show 22, 176 that during the fast warming-up from 7196 8C, the system temperature reaches a specified value before the gel point is attained.The non-equivalence of NFLMP characteristics in these reacting systems for the use of these freezing procedures is also indicated by NMR spectroscopy data,67 which reveal the differences between the amounts of the solvent remaining in the liquid state in the specimens frozen by different procedures. Thus, in the case of relatively fast cryotropic formation of chemically cross-linked polymer networks, the regime of freezing of a react- ing system is an important factor which influences, even at the stage of crystallisation, the properties and the formation dynamics of cryogels. Nuclear magnetic resonance spectroscopy is one of the few physical methods that allow direct detection of chemical reactions t /h 765 1 2 1 4321 2 20 10 0 730 T /8C 720 710 Figure 5.Time required for the formation of 10% of gel fraction in the reacting system vs. temperature of PAAG and cryoPAAG synthesis for freezing of the system from above (1) and for the low-temperature quenching version (2).22, 176 The initial total comonomer concentration is 3 mass %, [AAM] : [MBAAM]=30 : 1.Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications 22 19 16 13 10 74 0 5 d /ppm Figure 6. Evolution of the 1HNMRspectra of a solution of a mixture of AAM, MBAAM, TMEDA, and APS in D2O frozen using low-temper- ature quenching at711.2 8C.22, 67 The initial total comonomer concentration is 3 mass %, [AAM] : [MBAAM]=30 : 1. The digits near the curves represent the period of time (min) from the instant the system reaches a specified negative temperature to the instant of recording the spectrum.in moderately frozen systems without thawing the sample. For instance, the evolution of the 1H NMR spectra recorded at definite intervals (Fig. 6) for a solution of reactants (AAM± MBAAM±TMEDA± APS) frozen in D2O proves unambigu- ously that in this case, cryopolymerisation in the frozen medium takes place but neither pre- nor post-polymerisation during freezing or thawing, respectively. For an initial period, the spectra of the frozen solution of the reactants remain identical to the spectra of a similar specimen that does not contain the initiator. Subsequently (starting from about the tenth minute of the process), the intensity of the narrow signals due to the monomer protons [6.15 and 6.03 ppm (CH2), 5.60 ppm (CH)] starts to decrease, and a broad signal [*5 ppm (CH2)] for the protons of the polymer chains appears.67 In addition, NMR spectroscopy shows that the reactant ratio in the NFLMP differs from this ratio in the initial solution due to the partial crystallisation ofAAMand MBAAM at temperatures below the eutectic point; therefore, the reaction medium is enriched in the initiator (the spectra contain an intense common signal for the protons of theCH3 andCH2 groups of TMEDA with a maximum at *2 ppm).As polymerisation advances, the crystallised monomers are again dissolved in the liquid regions of the macrofrozen system and participate in the reaction, resulting ultimately in a high yield of the gel fraction.The measurement of spectroscopic characteristics of the non-frozen fraction of the solvent shows that an initial stage of polymer- isation develops under conditions of increasing volume of the liquid microphase, and only when a certain degree of conversion of the monomers into polymeric products has been attained, does the solvent content reach a constant value.67 Thus, depending on the cryopolymerisation temperature and freezing conditions, the composition and the volume of the reaction solution in the microreactor (non-frozen liquid microphase) can vary, which influences the gelation dynamics. Since the variation of the temperature regimes of cryopoly- merisation allows the macroporous morphology of gel materials to be controlled over certain limits, this approach has been used to prepare cryogels based on so-called stimuli-responsive 79 or sensi- tive 18 polymers.In recent years, gel materials prepared from these polymers have attracted increasing attention both for fundamen- tal research and for applied purposes. In particular, hydrogels that collapse reversibly upon a change in the environment parameters (temperature, pH, ionic strength, thermodynamic quality of the solvent, applied electric field, and so on) are considered to be promising materials for biomedical and biotechnological pur- poses.17, 18, 179 ± 182 For example, in temperature-sensitive hydro- gels in which the chains of the corresponding linear polymer undergo a reversible conformational change on passing through 497 the upper or lower critical solution temperature (UCST or LCST), substantial thermally induced reversible changes in the degree of swelling take place.Hydrogels based on cross-linked poly(N- isopropylacrylamide), poly(N,N-diethylacrylamide) (PDEAAM), poly(N-vinylcaprolactam) and so on are examples of thermally sensitive polymer systems.181 ± 187 Important characteristics of these materials which determine their performance characteristics include the temperature range in which the network collapses and the rate of system response to external treatment. Normally, the larger the specimen under study, the lower this rate, because the kinetics of collapse (or swelling) is controlled by heat transfer within the bulk of the gel and by the solvent diffusion and, hence, depends on the geometric dimensions of the specimen.However, if a macroporous structure is imparted to the material, the intrinsic response time of the macroporous specimen to a change in the temperature would decrease with respect to that observed for a fine-pored material, despite the fact that the weight and heat transfer rates in the gel phase remain the same. In this case, mass and heat transfer processes occur within the thin walls of macropores, i.e., at short distances; as a consequence, the intrinsic time of collapse for this specimen is shorter than that for the microporous gel having the same total volume.One approach to the synthesis of stimuli-responsive gels with enlarged pores is synthesis at temperatures above the collapse point of the cross-linked polymer (i.e., above the LCST for the corresponding non-cross-linked polymer).188 ± 190 In this case, phase segregation results in the formation of a coarser-pored gel (with pores of up to several hundred of A Ê ngsroÈ m) compared to that prepared from the same solution of monomer precursors but in the homogeneous medium at temperatures below the LCST (with pores of several tens ofA). The use of auxiliary porogens, for example foaming agents,191 at the stage of gel formation provides the possibility of producing an even coarser-pored material. In this case, some pores are closed but the rate of collapse or swelling of this gas-filled macroporous stimuli-responsive gel in reply to a change in the external conditions is much higher than that observed for homophase gel materials of the same chemical nature because the gel layer around each gas bubble is very thin.As noted above, cryotropic gelation allows one to form gel systems with open interconnected capillary-sized macropores. Thus redox-initiated copolymerisation of N,N-diethylacrylamide or N-isopropylacrylamide with MBAAM in a frozen aqueous medium affords the corresponding macroporous (with pore size of several tens of micrometers) spongy temperature-sensitive cryo- gels.192, 193 Gel materials of this type, in particular PDEAAM cryogels, collapse on heating much faster (Fig.7 a, curve 2) than usual PDEAAM gels formed in the liquid medium (Fig. 7 a, curve 1).192 During the decrease in the swelling capacity, the main bulk of water retained by the cryogel is separated in the first 10 ± 15 s. This includes almost all the capillary solvent and some fraction of the solvating liquid. The concentration of PDEAAM in the polymer phase of the cryogel (Fig. 7 b) reaches very high values (50% ± 70%), whereas for a collapsed usual gel, these values are lower (17% ± 42%). In other words, during the ther- mally induced decrease in the swelling capacity, a cryogel displaces more liquid from the polymer phase than the usual gel. At temperatures above the LCST and in the shrunk state, PDEAAM cryogels retain not more than 3.5 ± 5 water molecules per mono- mer unit, while the polymer network of a usual gel with the same composition retains up to 10 ± 35 water molecules per unit. A unit of PDEAAM has three lone electron pairs (two at the carbonyl oxygen atom and one at the amide nitrogen atom) able to form hydrogen bonds with three water molecules.Hence, it can be concluded that the degree of hydration of PDEAAM cryogels after the collapse is close to the lowest possible value. Thus, the use of cryotropic gelation techniques allows one to create a new type of gel material combining the properties of a network of stimuli- responsive gels and a supermacroporous structure typical of cryogels.498 Volume ratio (%) 100 80 60 40 20 20 10 0 Content of dry polymer (mass%) 60 40 200 50 : 1 [DEAAM] : [MBAAM] Figure 7.Properties of PDEAAM gels and cryogels.192 (a) dynamics of the thermally induced collapse of the PDEAAM gel (1) and cryogel (2); (b) content of the dry polymer in the gel phase at temperatures below (20 8C, curves 1, 3) and above (60 8C, curves 2, 4) the point of thermally induced collapse of the PDEAAM gel (1, 2) and cryogel (3, 4) for specimens with different initial ratios of the vinyl and divinyl comonomers and a constant total concentration (5.66%). Unlike polymerisation cryogels, polycondensation analogues have been little studied, although actually, the only restriction hampering the synthesis of the latter may be the fact that many polycondensation reactions giving the corresponding polymers are endothermic and require heating of the monomer solution. In the case where the synthetic route chosen uses reactions with positive heats and the reactivity of monomers is rather high, the formation of polycondensation cryogels almost does not differ from the formation of polymerisation cryogels.In particular, freezing is known 194 ± 199 to exert a structuring effect on hydrated sols of weak inorganic hydroxides (for example, iron hydroxide). The transformation of these sols into gels follows a polycondensation mechanism with water evolution to give a three-dimensional network of the inorganic oxide. The formation of silica gel on ageing of silicic acid sols is known 200, 201 to proceed in a similar way; therefore, the cryogenic variant of this process can be regarded as polycondensation cryotropic gelation.Freez- ing of a silicic sol which liberates water during a chemical reaction, i.e., forced dehydration of the system, should apparently shift the equilibrium towards the final product. Preparation of silica cryogel is not an easy experimental task,22 because it requires a rather accurate acid titration of an alkaline solution of a silicate with cooling until a silicic sol is formed. Then the colloid system must be rapidly frozen to ensure that polycondensation (i.e., ageing of the sol) occurs mainly in the NFLMP of the macrofrozen specimen. An interesting feature of the formation of silica cryogel is the fact that after thawing, the inorganic gel is obtained as scales rather than as a lump.The size of particles (2 ± 5 mm) depends on the temperature of the cryo- genic treatment. This is due to the fragility of silica cryogel, which is crushed by the growing ice crystals. In the preparation of elastic a 12 30 t /s 50 40 b 4321200 : 1 150 : 1 100 : 1 V I Lozinsky cryogels based on organic polymers, sample destruction of this type is hardly ever observed. The structure and properties of silica cryogels are mainly determined by the concentration of reactants and the temperature of freezing (Table 3). In particular, when the initial silicate concentration is above 0.5 mol litre71, the specific surface areas of the usual silica gel and silica cryogels formed at710,720, and 730 8C almost do not differ.However, cryogenic treatment of less concentrated systems affords silica cryogels whose specific surface areas are greater than that of the silica gel prepared from the reaction mixture with the same composition at a positive temperature. Silica cryogels formed at 710 8C possess a greater specific surface area than the specimens prepared at 720 and 730 8C. Unfortunately, the experimental data concerning inor- ganic cryogels of this type available to date are inadequate for a well-posed interpretation of this trend. Table 3. Dependence of the specific surface area of silica gels and silica cryogels on the SiO2 concentration in a silicic acid sol and on the gelation temperature.22 Specific surface area (m2 g71) of the dry material formed at different temperatures Concentration of SiO2 in the silicic acid sol /mol litre71 730 8C 720 8C 20 8C 710 8C 94 360 530 645 740 96 370 550 690 810 102 375 600 750 875 0.7 0.5 0.4 0.3 0.25 90 350 510 590 630 a Determined from adsorption of argon. To the best of the author's knowledge, there are no publica- tions dealing with polycondensation cryogels based on organic polymers.These cryogels are still to be prepared and comprehen- sively studied. 2. The formation of cryogels in solvent ± polymer ± cross- linking agent systems Cross-linking in solutions of macromolecular precursors on treat- ment with appropriate cross-linking reagents has long been known and is widely used both in the laboratory and in industry for the preparation of lyogels.Moderate freezing of initial systems of this type, incubation in the frozen state for a required period of time and subsequent thawing lead to macroporous cryogels. It is notable that cryostructurates formed by cross-linking of polymer precursors in frozen media were known a long time ago. Indeed, the porous foodstuff `kori-tofu', rather popular in Japan for several centuries, is nothing more than a cryogel based on soya proteins (so-called 11S-globulins) with covalent cross-linking by intermolecular disulfide bonds.202 Moreover, cross-linking of polypeptide chains at cysteine residues takes place just during the cryogenic treatment.203, 204 Using model thiol-containing polymers (SH derivatives of carboxymethylcellulose and PAAM), it has been shown 205 that atmospheric oxygen dissolved in the aqueous medium acts as the oxidant in these systems.Thiol-containing water-soluble polymers proved to be very convenient objects for the study of the kinetics of cryotropic gelation of this type 83, 206 because the course of the reaction can be monitored by highly sensitive spectrophotometric titration of the SH groups. Figure 8 shows the time variation of the concentration [calculated as the ratio of the current to initial SH titres expressed in per cent (Qrel)] of sulfhydryl groups in liquid and frozen aqueous solutions of the thiol derivative of PAAM at various temperatures. It can be seen that in frozen specimens (at715 and 720 8C), the concentration of the SH groups decreases much faster than in liquid solutions; this markedly accelerates cryo-Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications Qrel (%) 100 80 0 8C 60 15 8C 40 25 8C 720 8C 715 8C 20 t /h 20 15 10 5 0 Figure 8.Kinetics of gelation of the SH derivative of PAAM at various temperatures in a liquid and frozen aqueous solution of the polymer (concentration 2 g dl71, pH 7.8).22, 83 tropic gelation. For example, at 715 8C, a cross-linked gel is formed over a period of approximately 1 h (shown by the black arrow), whereas at 25 8C, this occurs only after 24 h (light arrow).In other words, for the cryostructuring of solvent ± polymer ± cross-linking agent systems, as for the formation of polymer- isation cryogels, gelation in a definite range of negative temper- atures is accelerated compared to that for non-frozen specimens. At75 8C, an aqueous solution of SH-PAAM with a polymer concentration of 2 g dl71 does not become frozen due to under- cooling effects; therefore, the reaction is very slow under these conditions. When low-temperature quenching is used, the system is frozen at 75 8C and the reaction rate is high owing to cryoconcentration (Fig. 9). The temperature dependence of the rate of variation of Qrel in systems frozen by two different procedures passes through an extremum, which is typical of cryotropic gelation. Comparison of the pattern of variation of the SH-group concentration with theNMRdata on the amount of mobile non-frozen solvent in the specimens shows 83 that after a specified negative temperature has been attained, the volumes and properties of liquid microphases in the specimens frozen in differ- ent ways are non-equivalent for a certain period of time.In the case of freezing from above, after crystallisation of the main bulk of the solvent, the reactions proceed initially under conditions of gradually decreasing volume of the liquid component, i.e., with increasing concentrations of reactants. Conversely, in the speci- 103 vinit /mol (litre min)71 8 1 64 2 2 320 10 0 T /8C 710 720 730 Figure 9. Initial rate (vinit) of gelation of the SH derivative of PAAM vs.temperature (concentration 2 g dl71, pH 7.8).22, 83 (1) Low-temperature quenching, (2) freezing from above, (3) variation of vinit for the region of positive temperatures. 499 mens frozen in liquid nitrogen and then heated to a specified negative temperature, the smallest volume of the microreactor and, hence, the greatest degree of substance concentration is attained almost simultaneously when the temperature of the environment (the cryostat refrigerating fluid) becomes equal to the temperature of the frozen reaction mixture. These dynamic differences between the phase states of the specimens brought to the temperature of a cryochemical reaction via different routes account for the non-equivalence of temperature dependences 1 and 2 in Fig.9. Cryotropic gelation in the solvent ± polymer ± cross-linking agent systems, like the synthesis of polymerisation cryogels, is accompanied by an apparent decrease in the CCG. For example, Fig. 10 shows the dependences of the yield of the gel fraction on the polymer concentration in the initial solution (Fig. 10 a) and on the amount of the cross-linking agent (Fig. 10 b) during the formation of normal gels and cryogels of chitosan [poly-b- (1?4)-D-glucosamine] cross-linked by glutaraldehyde at 25 8C and in a moderately (78 8C) frozen medium. Cryogels are formed at substantially lower initial concentrations of both the polymer (Fig. 10 a) and the cross-linking agent (Fig. 10 b).13, 22 Undoubt- edly, this decrease in the CCG is due to the cryoconcentration of the reactants in the NFLMP.The normal gels of cross-linked chitosan are typical homo- phase hydrogels; in the swollen state, they are transparent and rather brittle, especially when the cross-linking degrees are rela- tively high. The whole bulk of the solvent present in the specimens swollen to equilibrium occurs in the bound (solvation) state. As in the case of other chemically cross-linked gels, it cannot be removed by squeezing out without disruption of the gel integrity. Conversely, cryogels formed from the same initial reaction solutions are heterophase spongy non-transparent materials in which the capillary-bound solvent can be separated rather easily by compression even at low mechanical loads. a Yield of the gel fraction (%) 2 80 1 60 40 200 5 3 2 1 4 Polymer concentration /g dl71 b Yield of the gel fraction (%) 2 80 1 60 40 200 15 10 5 [CHO] : [NH2] (mol.%) Figure 10.Yield of the gel fraction vs. the initial concentration of chitosan (a) and vs. the ratio of the number of glutaraldehyde CHO groups to the number of polymer amino groups for a fixed (2.5 g dl71) initial concen- tration of chitosan (b) 13, 22 at (1) 25 8C, (2)78 8C.500 The above examples refer to the synthesis of cryogels in frozen aqueous media. However, there are no obstacles preventing, in principle, the preparation of cryogels from precursors soluble in organic solvents. A necessary condition is that the organic solvent should undergo crystallisation rather than vitrification to ensure the separation of the solid and liquid phases and the formation of an NFLMP with an enhanced concentration of reactants.Figure 11 shows the dependence 22, 30 of the yield of the gel fraction on the polymer concentration in the initial solution (Fig. 11 a) and on the temperature (Fig. 11 b) during the SnCl4- catalysed cross-linking of polystyrene with 1,4-xylylene dichloride according to the Friedel ± Crafts reaction in liquid (13.5 8C) and crystallised (713.5 8C) nitrobenzene (Tm=5.5 8C). These data indicate that processes in organic solvents actually obey the same regularities as gelation in aqueous media, namely, the apparent decrease in the CCG for cryogenic processes (Fig.11 a) and an extremal pattern of dependence of their efficiency on the temper- ature of the frozen system (Fig. 11 b). The polystyrene cryogels formed in crystallised nitrobenzene have a spongy morphology (similarly to cryogels based on hydrophilic polymers considered above) and during swelling absorb substantial amounts of organic solvents compatible with polystyrene, for example, benzene or toluene. The main bulk of the liquid absorbed by the cryogel is the capillary solvent which can be easily squeezed out. Thus, it can be concluded that the type of morphology of cryogels cross-linked by covalent bonds and the effects involved in their formation are rather general and do not depend on the nature of the crystallis- able solvent used.a Yield of the gel fraction (%) 1 90 2 80 70 0.3 0.2 0.1 0 Concentration /mol litre71 b Yield of the gel fraction (%) 2 90 1 80 70 60 0 10 720 730 T /8C 710 Figure 11. Yield of the gel fraction vs. the initial concentration of polystyrene expressed as the molar concentration of monomer units (a) and vs. the temperature (the initial concentration of monomer units of polystyrene is 0.3 mol litre71) (b);22, 30 (1) formation of gels (13.5 8C); (2) formation of cryogels (713.5 8C). 3. Ionic cryogels Ionic, coordination, and coordination ± ionic polymer gels repre- sent special types of gel-like systems. On the one hand, they are thermally irreversible, like covalently cross-linked gels, and on the other hand, the change in the pH or the salt composition of the V I Lozinsky liquid which surrounds the specimen can induce fast dissolution of swollen ionotropic and chelatotropic networks, which are con- verted again into the gel upon an appropriate change in the medium composition; this reversibility makes these systems sim- ilar to physical gels. A specific feature of many ionotropic gels is heterogeneity caused by impediments to the diffusion of new portions of counter-ions into deep regions of the gelling specimen through the gel membrane, which is formed rapidly at the sample surface (periphery).Traditional ionotropic gels, i.e., those formed at positive temperatures, can be divided conventionally into three groups. The first group includes solvated networks based on polyelec- trolytes whose macromolecules are cross-linked to give a three- dimensional structure through low-molecular-weight ions form- ing low-dissociated salt bridges with the ionogenic groups of the polyelectrolyte chains at the network junction knots.When complexing agents are used for cross-linking, a chelate (coordina- tion) network is formed. These ionotropic and chelatotropic gels are capable of substantial repeated swelling after drying. The second group includes polymer gels formed with the change in the charge of polyelectrolyte ionogenic groups, resulting in a decrease in the polymer solubility and phase separation. The gel-like systems formed in this case usually possess low mechanical strengths and, after drying, they are unable to swell to a sub- stantial extent in the same medium as was used for gelation.The third group comprises ionotropic gels formed during preparation of polyelectrolyte complexes (PEC). In essence, these systems represent swollen three-dimensional networks cross-linked by poorly dissociating salt bonds (similarly to the first-group ionic gels). However, in this case, it is often impossible to establish which of the PEC components (either polycation or polyanion) is the gel-forming and cross-linking component, especially if the MW and the initial concentrations of these compounds are similar. Examples of preparation of ionotropic and chelatotropic cryogels that can be attributed (although with some reservations) to the first two groups have now been reported.Cryogels based on PEC have not yet been described. The major obstacle faced during the formation of ionotropic cryogels is a too high rate of gelation, which proceeds even at the stage of component mixing. Indeed, if low-molecular-weight counter-ions inducing gelation are added to a polyelectrolyte solution in a concentration higher than the CCG, a lyogel is formed very rapidly. On subsequent freezing, the solvent crystal- lises in the gel phase, i.e., it is virtually impossible to freeze the system when it still exists in the liquid state. This situation is encountered, for example, in the attempt to prepare a calcium alginate cryogel: the addition of a solution of a calcium salt to an aqueous solution of sodium alginate (the sodium salt of the linear block copolymer of D-mannuronic acid and L-guluronic acid) entails fast formation of a gel network, and the system loses fluidity.This specimen can be frozen, but this would not be accompanied by accumulation of precursors and by chain cross- linking in the NFLMP, peculiar to cryotropic gelation (Fig. 12, pathway a). Preparation of the cryogel by recharging the alginate carboxy groups is impossible either, because acidifying the sol- ution results in fast formation of a thixotropic gel of alginic acid, whose freezing ± thawing does not afford a macroporous material typical of cryogels (Fig. 12, pathway b). Two approaches have been proposed to surmount these obstacles (Fig. 12, pathways c and d ).If a finely powdered salt having a negative temperature factor of solubility is added (as a source of cross-linking counter-ions) to a polyelectrolyte solution (Fig. 12, pathway c) and the resulting suspension is subjected to moderate freezing, gradual dissolution of this salt in the NFLMP and cross-linking of the polyelectrolyte chains take place at low temperature. After thawing, a macro- porous ionotropic cryogel is produced. A crucial point is the choice of an appropriate salt because the solubility of most salts increases with increasing rather than decreasing the temperature.Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications Polyelectrolyte solution Pathway d Pathway c Pathway b Pathway a Lyogel Suspension Lyogel Frozen specimen Cryogel Almost no limitations of this type are involved in the proce- dure shown as pathway d in Fig.12. This strategy allows the preparation of macroporous ionotropic cryogels by both the recharge of ionogenic groups and cross-linking with appropriate counter-ions, although cryotropic gelation does not actually occur in this case.22, 210, 211 To prepare the target material, a polyelec- trolyte solution with any desired concentration limited only by the polymer solubility is merely frozen at a required temperature. This is followed by removal of the crystallised solvent without thawing, Freezing and thawing Solid salt Removal of the solvent ionogenic groups or cross-linking Freezing ??? ??? Dry polymer framework Cryo- structurate Figure 12.Basic diagram of the formation of ionic cryostructurates.22 The degree of salt grinding is also significant. This strategy has been employed to prepare an alginate cryogel 22, 207, 208 using calcium salts with the solubility increasing upon decreasing temperature [in particular, calcium butyrate, pentanoate, succi- nate, glycerophosphate, and some other calcium salts have negative temperature factors of solubility in water (see Ref. 209, p. 1353)]. In this case, as in the formation of covalent cryogels, the properties of the final material depend not only on the precursor concentrations and the temperature of cryogenic treatment but also on the time of keeping the specimens in the frozen state.In particular, when the initial concentrations of sodium alginate and calcium glycerophosphate are 0.5 mass %, keeping frozen speci- mens at 710 8C for 1.5 h gives ultimately (i.e., after thawing, washing and drying) a cryogel with a rupture strength of 250 kg cm72. An increase in the duration of the cryogenic treat- ment to 3 h leads to a material with a rupture strength of 420 kg cm72, and incubation of the sample in the frozen state for 24 h makes this parameter about twice as high (800 kg cm72).22, 208 These data indicate that cross-linking and structuring of alginate require a certain time for dissolution of the solid cross-linking agent and for slow diffusion of calcium ions into the high-viscosity medium of the liquid microphase.However, this strategy for the preparation of ionotropic cryogels is subject to numerous restrictions, related, most of all, to the admissible concentrations of precursors. Above a definite concentration of either the polyelectrolyte or the salt, too fast gelation taking place at positive temperatures prevents the for- mation of a `true' cryogel. Freezing and thawing Cross-linking counter-ions (in solution) Freezing and thawing Recharging of ionogenic groups (in solution) Recharging of 501 for example, by vacuum sublimation (freeze-drying) or cryoex- traction (treatment at a negative temperature with a liquid which acts as a solvent for the crystalline phase but a non-solvent for the polyelectrolyte).Finally, cryogenically structured macroporous polymeric framework is obtained, which is then transformed into an insoluble state either by treatment with cross-linking counter- ions or by recharging ionogenic groups. The last-mentioned operation is also carried out in the polymer non-solvent and this no longer requires low temperatures. For instance, freezing of a solution of sodium alginate and freeze-drying of the specimen thus formed followed by immersion of the porous material into an ethanol solution of calcium acetate results in an ion exchange reaction in which the material structure is fixed by poorly dissociating salt bridges. This yields a calcium alginate cryostructurate 210 which is not dissolved in a physiolog- ical salt solution for several days, whereas a usual alginate gel prepared from the same precursors at a positive temperature loses its integrity in several minutes and is promptly dissolved.When substances capable of forming poorly dissociating coordination (more precisely, coordination ± ionic) bonds with alginate groups, for example, chromium salts, are used as cross-linking agents, this procedure leads to chelatotropic cryostructurates.22 Treatment of plates or grains of a frozen solution of sodium alginate with acidified acetone in the cold (e.g., at 720 8C) induces cryoextraction of ice with simultaneous transformation of the polymer network into alginic acid due to recharge of ionogenic groups. The final macroporous polymer is insoluble in water but is wetted by water and readily swells.210 A similar strategy can be utilised to prepare a macroporous material from a polymeric base.A specimen prepared by freezing a solution of a macromolecular salt followed by removal of the solvent is treated, for recharge of ionogenic groups, with an alkaline reagent in a medium that is a non-solvent for the polymer.211 For example, dropwise addition of a solution of chitosan in dilute acetic acid to cooled (e.g., to 715 8C) octane results in frozen spherical particles, treatment of which with a saturated solution of NaOH in acetone or methanol at a negative temperature affords granulated macroporous chitosan base. This material is stable in alkaline or neutral media and is capable of coordinating heavy metal ions (for example, its capacity with respect to Cu2+ or Co3+ ions equals 5.2 ± 5.5 mmol g71).22, 211 Thus, it is possible to carry out cryogenic structuring of polyelectrolytes, although it requires more sophisticated approaches, and to prepare ionic or coordination polymer materi- als possessing macroporous morphology, typical of cryogels.4. Cryogels with a physical network of the polymer phase As has been noted in the Introduction, non-covalent (physical) cryogels are the best known and investigated gel materials formed by the freezing ± thawing strategy. This refers most of all to PVA cryogels. A detailed analysis of publications (by the end of 1997) devoted to the formation of PVA cryogels, the influence of cryogenic treatment parameters on these gels, and their properties has been given in a review.9 In this Section, we consider only the general fundamental items of non-covalent cryotropic gelation and the data on physical cryogels based on other polymers.However, at the end of the Section, some other results concerning PVA cryogels (published in 1998 ± 2001) are also considered. Generally, non-covalent physical gels form a very extensive class of gel systems produced in both aqueous and organic media (see, for example, the publications 15, 20, 212). In principle, non- covalent gels can be prepared by either cooling (if the solvent ± polymer system has an UCST) or heating (if the system has an LCST).In the former case, one deals with psychrotropic gels, and in the latter case, thermotropic gels are involved. It is clear that the latter cannot be cryogels. But in the case of psychrotropic gel- forming systems, it is not always possible to prepare a physical cryogel by mere freezing. When the polymer concentration is above the critical point and the rate of material solidification on cooling is high, then, as in the case of polyelectrolytes with cross-502 linking counter-ions (Section IV.3), fast gelling takes place. In this case, the system that undergoes freezing is no longer a solution but a gel; therefore, the structure of the polymeric material after thawing differs from that obtained in the case of `standard' cryotropic gelation.An example of such a system is an aqueous solution of agarose (neutral polysaccharide, the major component of agar-agar), which is converted into a gel over a period of several minutes even at 40 ± 30 8C when the polymer concentration is above 1%. An agarose macroporous cryogel can be prepared only from relatively dilute (<0.2%) solutions of the polymer. How- ever, strength characteristics of such a cryogel are poor. Solutions of PVA (in water or DMSO),9, 31 ± 34, 213 amylopec- tin,214 maltodextrin (a mixture of partially depolymerised starch polysaccharides with MW<20 kDa),215 starch,23, 216 ± 219 or LBG23 ± 26 are more convenient for the formation of non-covalent cryogels because they belong to systems gelling rather slowly on decreasing the temperature.They can be frozen without compli- cations, stored for a required period at a specified negative temperature and then thawed to give the corresponding cryogels. The cryogels based on these polymers are thermally reversible: they fuse at elevated temperatures and repeated freezing ± thawing of the solution results again in a cryogel.22, 31, 220 Depending on the type and initial concentration of the polymer and the regimes of cryogenic treatment, this gives either spongy cryostructurates with moderate strength similar to cryo- PAAG, described in Section IV.1, or elastic non-spongy cryogels (which are exemplified by PVA cryogels). Most likely, these distinctions in the morphology of materials are related to the amount of free (unbound) solvent able to crystallise at a chosen negative temperature and to the size of porogen polycrystals thus formed.In some cases, the same solvent ± polymer system with different contents of the macromolecular component can be converted in both spongy and non-spongy cryogels. For example, dilute (0.05 ± 0.75 g dl71), semidilute (0.75 ± 2.0 g dl71) and concentrated (>2.5 g dl71) aqueous sol- utions of LBG remain liquid at room temperature for several months.221 Freezing ± thawing (710 to 730 8C) of these solu- tions gives very weak spongy cryostructurates in the first case, good polymeric sponges in the second case, and elastic non- spongy cryogels in the third case.23 ± 26 Aqueous solutions of maltodextrin containing less than 15 ± 20 g of the polymer per dl (the concentration depends on the MW) cannot be transformed into a gel without cryogenic treat- ment.Concentrated aqueous solutions of PVA containing no residual acetyl groups form gels very slowly, whereas the freez- ing ± thawing procedure always gives cryogels.9, 22, 31 ± 34, 213, 215 Thus, thermally reversible physical cryogels are obtained at lower initial concentrations of the macromolecular precursors and often much faster than without freezing. This indicates that non-covalent cryotropic gelation, like the formation of covalently cross-linked cryogels, is subject to the effects of the apparent decrease in the CCG and accelerated gelation. The chemical nature of the intermolecular bonds stabilising the junction knots of the three-dimensional polymer network in physical cryogels depends on the structure of the gel-forming polymer material.Usually, these are either hydrogen bonds (the predominant type of contacts between the chains in thermally reversible cryogels, similar to cryogels based on PVA, starch, and other polyhydroxy polymers) 222 ± 224 or hydrophobic interactions, which play an important role in the formation of new protein cryogels,82, 225 or various weak (but cooperative) dispersion inter- actions. For PVA cryogels, the formation of microcrystallinity zones acting as junction knots in the supramolecular network has been shown experimentally.226 ± 228 The sizes of microcrystallites are determined by the temperature conditions of cryostructuring and the number of freezing ± thawing cycles.At these knots, syndiotactic sites of PVA chains form numerous hydrogen bonds.9, 22 Apparently, microcrystallinity zones function as the physical network junction knots also in the structure of cryogels formed from starch, maltodextrin, and LBG polysaccharides. V I Lozinsky The macro- and micromorphology and physicomechanical and thermal properties of non-covalent cryogels depend appreci- ably on some characteristics of the polymer precursors, which are not so important in the case where cryogels are formed through covalent cross-linking of macromolecules. For example, in the cryotropic gelation of PVA solutions, apart from essential proper- ties of the polymer such as theMWand chain tacticity, the number of residual O-acyl groups [O-acetyl groups for the PVA prepared from poly(vinyl acetate)] is also a significant factor, as the presence of even *5 mol.% of these substituents is sufficient for inhibiting the complementary (`side-by-side') interaction of neighbouring chains.213, 229 ± 232 In other words, O-acyl groups act as sort of distance bars preventing the formation of numerous intermolecular hydrogen bonds and thus interfering with the formation of PVA cryogels. In the case of starch cryogels, an important parameter is the ratio of the linear to branched polysaccharides (amylose and amylopectin) in the system sub- jected to freezing, which depends on the type of starch used and, in some cases, on the procedure of its pretreatment.23, 214 ± 219, 233 ± 236 When non-covalent cryogels such as albumin cryogels are formed through hydrophobic interactions, the accessibility of hydropho- bic regions for intermolecular contacts becomes the crucial factor.This is determined by the conformation of chains, in particular, by denaturation-related changes in the macromolecules of these globular proteins, which are specially induced by pretreatment (prior to freezing) such as the addition of chaotropic agents (which destroy the cluster structure of water), heating, or high-pressure treatment.82, 237 ± 243 Thus, as in the case of `usual' psychrotropic gelation, for the formation of physical cryogels, too, the chemical structure and the degree of exposure of the chain groups involved in intermolecular interactions are important factors.The conditions of all stages of cryogenic treatment are highly significant for the objects considered. Whereas in the synthesis of cryogels cross-linked by covalent or ionic bonds, the conditions of thawing of specimens usually have little influence on the properties of final gels, the parameters of non-covalent cryogels are often fairly sensitive to the thawing condi- tions.9, 10, 14, 26, 31 ± 34, 214, 215, 235, 244 For example, Fig. 13 shows the variation of the apparent instantaneous modulus of elasticity (E0) and the fusion temper- ature (Tf) of the LBG cryogels vs. the thawing rate of the specimens (the vth values are laid-off along the abscissa axis on a logarithmic scale).22, 26 The specimens were prepared by freezing aqueous solutions of a polysaccharide (2.0 g dl71) at720 8C for 18 h.It can be seen that the lower the thawing rate, the greater the rigidity and the thermal stability of these physical cryogels. This trend is typical of those thermally reversible cryogels in which hydrogen bonding is the predominant type of intermolecular contacts stabilising the polymer network junction knots (cryogels Tf /8C75 E0 /kPa 2.5 70 2.0 65 1.5 2 1 0.1 0.03 60 3.0 vth /8C min71 1.00.01 1.0 0.3 Figure 13. Properties of LBG cryogels vs. vth .22, 26 (1) Apparent instantaneous modulus of elasticity, (2) cryogel fusion temperature.Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications based on PVA, LBG, starch polysaccharides, gelatin, agar-agar, etc., are examples of such systems).The nature of this effect has been studied in most detail for PVA cryogels.9, 22, 31, 33, 34, 80, 244, 245 The data of Table 4 indicate that the shear modulus (G0) of PVA cryogel specimens prepared from aqueous solutions with different polymer concentrations increases severalfold as the thawing rate decreases from 0.3 to 0.03 8C min71.34, 245 Conversely, in the case of high thawing rates (>5 ± 10 8C min71), PVA cryogels cannot be prepared at all, as, instead of an elastic gel, thawing gives a muddy thick liquid.31, 33, 34 These data indicate that the formation of PVA cryogels takes place exactly at the thawing stage.One can follow the kinetics of gelation and choose the temperature range corre- sponding to the highest intensity of this process (Fig. 14) using the following procedure: aqueous solutions of PVA are first frozen, for example, at 7208C, then the specimens are transferred into a cryostat with a specified subzero temperature and, after incuba- tion over particular periods of time, selected specimens are rapidly thawed (at a rate of*3 8C min71) in order to eliminate the effect of the thawing rate.22, 34 Table 4. Effect of the thawing rate on the rheological properties of PVA cryogels.22, 34, 245 Shear modulus of PVA cryogel specimen /kPa Rate of thawing of frozen specimens /8C.min71 Concentration of PVA in the initial solution /g dl71 <0.3 <0.3 <0.3 1.73 1.80 3.93 4.94 10.6 1.40 3.20 9.40 0.30 0.03 0.30 0.03 0.30 0.03 0.30 0.03 3.0 0.30 0.03 3.0 3.0 5.0 5.0 7.0 7.0 10.0 10.0 10.0 a 10.0 a 10.0 a Note.The materials were prepared by freezing at 720 8C for 18 h a completely deacetylated PVA with MW of 115 kDa. a PVC deacetylated to 99% withMWof 69 kDa. The 0-h points in the plots correspond to the onset of incubation of frozen specimens at a specified subzero temperature (which is written near the corresponding curves in Fig. 14 a). Figure 14 b shows the duration of this incubation. The shear modulus of the PVA cryogels prepared using this technique increases rather weakly whilst keeping the frozen specimens at 710 8C.At higher negative temperatures, gelation becomes much more intense. The observed temperature dependence is non- monotonic, the process being faster in the 72 to 73 8C range than at 71.5 8C. Thus, in this case, the regularities are qualita- tively analogous to the features of covalent cryotropic gelation during the synthesis of cryogels by polymerisation of monomer precursors (see Fig. 5) or by chemical cross-linking of polymer precursors (see Fig. 8 b and Fig. 11 b), i.e., the temperature dependence of the efficiency of formation of non-covalent PVA cryogels at negative temperatures passes through an extremum. Study of frozen solutions of PVA by ESR and NMR spectro- scopy shows 70, 81 a sharp increase in the amount of the mobile solvent above a certain threshold temperature range, which is 476 8C below the crystallisation point of the system.At these temperatures, together with firmly bound non-freezable water, an additional amount of liquid water appears in the solid, frozen specimen. This partial melting of ice increases the NFLMP volume; therefore, the mobility of macromolecules (or their seg- ments) and the efficiency of polymer ± polymer interactions a G0 /kPa 6420 15 10 5 b G0 /kPa 6420 74 76 78 710 Figure 14. Dynamics of formation of PVA cryogels (completely deacety- lated PVA with MW of 115 kDa, initial concentration 7 g dl71, freezing at720 8C over a period of 24 h) at subzero temperatures.22, 34 (a) Time variation of rigidity of PVA cryogels; (b) temperature depend- ences of the rigidity of specimens kept in the subzero region.increase. Finally, this results in the formation of microcrystallites which act as junction knots of the cryogel network. In particular, this mobility can be detected in 13C NMR spectra: *8 8C below the melting point of the main bulk of the solvent, the resonance peaks for the carbon atoms of PVA virtually cannot be detected but when the temperature rises by *3 8C, quite resolved, although weak signals arise. On further increasing the temper- ature (273 8C prior to thawing of the specimen), a real 13CNMR spectrum for the polymer solution is recorded. From this it follows that a temperature 576 8C below the system's thawing point corresponds to a boundary of retardation of the fast (on theNMR time scale) motions of the PVA carbochain backbone.Estimation of the spin ± lattice relaxation correlation times for the carbon nuclei in the polymer, which serve as the criterion for the mobility of kinetic chain segments, shows that reorientation of the 13C nuclei in the PVA methylene groups in a system frozen at subzero temperatures occurs*100 times more slowly than that in the initial liquid solution. This points to a very high viscosity of the NFLMP.81 On the basis of these data, it is possible to interpret the influence of the thawing rate or the duration of incubation of frozen specimens at moderate negative temperatures on the properties of PVA cryogels (see Fig.14). The slower the thawing and the closer its conditions to equilibrium, the longer the period the specimen is at the temperatures optimal for gel network formation. Therefore, slow warming-up during thawing of the preparations is equivalent, in a sense, to storing the specimen for a long period at temperatures where the strongest cryogels are formed. Although a high polymer concentration in the NFLMP is favourable for gelation, this requires a fairly long period of time due to substantial viscosity of the NFLMP and low mobility of chains. The extremal dependence of the physicomechanical prop- erties of PVA cryogels on the cryostructuring temperature (see 503 72.0 8C 72.5 8C 73.0 8C 71.5 8C 75.08C 710 8C t /h 20 24 h 6 h 0.5 h 0 h T /8C 72504 Fig.14 b) is due to the competition of factors favourable and unfavourable for gel formation. On the one hand, at temperatures above the extremum point, the amount of mobile solvent is greater, i.e., the specific concentration of the gel-forming reagent in the NFLMP is lower, so is the probability of formation of effective intermolecular contacts (intensification of thermal motions in the system also counteracts stabilisation of the contacts). On the other hand, at temperatures below the extremum, the amount of mobile solvent decreases, the polymer becomes poorly solvated and retains only the firmly bound water, i.e., the liquid medium in which the motion bringing together the interacting sections of the neighbouring chains takes place is no longer present.Since concentrated aqueous solutions of PVA usually do not freeze in the subzero temperature range, i.e., at temperatures of most intense gelation, due to undercooling effects, the initial specimens always need to be frozen at lower temperatures. The properties of PVA cryogels can be varied by varying the rate of thawing or the time of storage of the frozen materials at subzero temperatures. Figure 15 shows the pattern of porosity of specimens of PVA cryogels thawed at rates of 0.3 and 0.03 8C min71. The photo- micrographs of thin sections of cryogels dyed by Congo Red were taken using optical microscopy; the dark areas represent the gel phase and the light areas are macropores.22, 245 Whereas in the first specimen (Fig.15 a), the macropore size and arrangement are rather irregular, the specimen of the PVA cryogel formed under slow thawing conditions (Fig. 15 b) possesses quite a perfect morphology with a clear-cut alternation of oblong macropores and strands of the gel phase. Undoubtedly, higher rigidity of slowly thawed cryogels of PVA of this type compared to the specimens thawed more rapidly (see Table 4) is due, among other factors, to the more regular macrostructure of the former. b a Figure 15. Photomicrographs (optical microscope) of thin sections of specimens of PVA cryogels formed from an aqueous solution (12 g dl71) of a polymer (MW 69 kDa, degree of deacetylation 99 mol.%) on freezing at 720 8C for 18 h followed by thawing at a rate of 0.3 (a) or 0.03 8C min71 (b) (see Refs 22, 34).Scale: 40 mm to 1 cm. It is worth noting that the morphology of PVA cryogels has been studied most often by electron microscopy techniques, which imply dehydration of specimens,4, 22, 226, 246 ± 253 resulting inevita- bly in some changes in the dimensions and shapes of the structural units of these objects.246, 249, 250, 253 The use of optical microscopy for these studies is more legitimate because it permits one to observe the real structure of a swollen cryogel.226, 245, 253 ± 255 Yet another interesting feature of non-covalent cryogels is an increase in the rigidity upon freezing and thawing repeated many times.4, 5, 9, 11, 25, 32, 226 ± 228, 230 ± 232, 256 ± 258 For example, the dynamic modulus of elasticity of a PVA cryogel prepared from a 15% aqueous solution of the polymer(MW75 kDa) is an order of magnitude higher after four cycles of cryogenic treatment (each cycle includes 24 h freezing at 720 8C and 7 h thawing at 15 8C) than after one such cycle.259 According to the data of X-ray diffraction analysis and DSC,226 ± 228, 260 every subsequent free- zing ± thawing procedure increases the content and the size of V I Lozinsky crystallites in the PVA cryogel. Apparently, repeated occurrence of the system in the temperature of the most intense gelation gives rise to new additional intermolecular contacts and extends those present previously.This is manifested, in particular, as numerous thickened regions observed through an optical microscope.226 In addition, the possible compaction of the supramolecular strands of the cryogel network, which approach one another under the action of growing ice crystals, cannot but be taken into account.Thus, various procedures of cryogenic treatment of the water ±PVA system make it possible to vary the properties and morphology of non-covalent cryogels of a given polymer over wide limits,9, 88, 245, 261 ± 268 the duration of the occurrence of the frozen specimen at moderate negative temperatures being the crucial factor determining the efficiency of non-covalent cryo- tropic gelation. This conclusion is also true for the formation of other physical cryogels.An important role of thawing or incubation at subzero temperatures was demonstrated for the cryotropic gelation of LBG,26 maltodextrin,215 amylopectin,214 and mixtures of amylo- pectin with amylose.235 However, individual features of each of these polymers also make a contribution to the observed regu- larities. For instance, in the case of highly crystalline starch polysac- charides, the primary cryogel is formed immediately during freezing, while thawing conditions influence the further formation (adjustment) of intermolecular contacts.214, 235 This can be clearly seen in a study of dilute systems in which the freezing ± thawing cycle affords a cryoprecipitate rather than a cryogel over the whole specimen bulk. The precipitate can be readily separated from the solution and the yield of the gel fraction can be measured.This parameter shows that a decrease in the rate of thawing results in a substantial increase in the structurisation efficiency compared to that in rapidly thawed preparations.235 To conclude this Section, it should be noted that non-covalent intermolecular interactions play an important role also in the formation of cryogels based on colloidal dispersion systems: suspensions, pastes and colloid sols. Such cryostructurates are often produced on freezing ± thawing of dispersions of food biopolymers; they are formed as coagulates upon cryogenic treat- ment of latexes, and so on. However, subtle physicochemical mechanisms of these processes have not yet been adequately studied, although the need for these studies is obvious.Generally, the formation of non-covalent (physical) polymeric cryogels is characterised by the same effects as covalent cryostruc- turing, i.e., the formation of macroporous gel-like bodies, appa- rent decrease in the CCG, acceleration of gelation over a definite range of negative temperatures and extremal temperature depend- ences of process rates. The main difference between non-covalent cryotropic gelation and the processes of formation of covalently cross-linked cryogels is the important role of the thawing stage (incubation at subzero temperatures) in the former case. V. Applications of cryotropic gelation processes and materials based on polymeric cryogels In recent years, the interest in polymeric cryogels as not only objects of scientific research but also promising materials for the solution of applied problems has constantly increased.The range of applications of the cryotropic gelation techniques and the gel materials is fairly broad. The unique porous structure of cryogels proves to be useful in many cases. An especially large number of publications including reviews 3, 4, 6, 10, 11, 269 have been devoted to the use of various cryogels as materials for biomedical and biotechnological purposes. The development of cryogenic proc- esses for the production of new food forms, new sorbents, filters, mechanochemical gel actuators (manipulator mechanisms), leather-like materials, catalyst systems, and many other things is in progress.The data demonstrating the fields and branches in which diverse cryogels have already been successively used are summarised in Table 5. It should be emphasised that the range ofCryogels on the basis of natural and synthetic polymers: preparation, properties and applications Table 5. Fields of application of cryotropic gelation processes and materials based on polymeric cryogels. Cryogel type Application of a material or cryotropic gelation procedure Gel-formimg reagents 2-HOEMA 2-HOEMA+acrolein IV.1(r) IV.1(r) IV.1(r) 2-HOEMA+gialuronic acid Hydroxy- and methoxyoligoglycol methacrylates Oligoethylene glycol monometh- acrylates and dimethacrylates AAM+MBAAM IV.1(r) IV.1(r) IV.1(c) IV.1(c) AAM+MBAAM+allyl glycidyl ether IV.1(c) IV.1(pc) Cryo(silica gel) Gelatin+formaldehyde Serum albumin+glutaraldehyde IV.2 IV.2 IV.2 IV.2 IV.2 IV.2 IV.2 IV.3 IV.3(c) IV.3(c) IV.3(d ) IV.3(d ) Polystyrene ± polybutadiene block copolymer+Pd or Rh salts Collagen+dialdehyde (diepoxide, epichlorohydrin, diisocyanate, etc.) A dispersion tanned by heavy metal salts IV.2 of leather-processing wastes in the presence of cross-linking agents Sodium alginate+calcium gluconate Sodium alginate+calcium salt with a negative temperature factor of solubility Chitosan+b-tricalcium phosphate Polymeric acids Polymeric bases PVA IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 IV.4 PVA+chitosan or dextran PVA+polyacrylic acid PVA+alginate PVA+polyelectrolyte PVA+gellan PVA+collagen fibres PVA+Fe salts Solutions and dispersions of food biopolymers carriers for enzyme immobilisation carriers for antibody immobilisation and preparation of immunosorbents gel matrices for controlled drug release carriers for immobilisation of microbial cells gel matrices for cultivation of animal cells carriers for immobilisation of microbial cells 22, 270, 271 gel bases for supermacroporous bioaffinity sorbents 269, 272 gel bases for supermacroporous bioaffinity sorbents 269, 272 carriers for immobilisation of microbial cells 22, 273 immobilisation of yeast cells 274 immobilisation of photosynthetically active cell organelles of lettuce 275 ± 278 immobilisation of the submembrane fraction of photosystem II 279, 280 of spinach cells immobilisation of photosynthetically active cyanobacteria 281 immobilisation of thermophilic bacterial cells 282 preparation of heterogeneous catalysts containing 283 zerovalent metal nanoparticles collagen materials for medical purposes leather-like materials for domestic and technical purposes polymeric scaffold of implants containing adsorbed hepatocytes alginate sponges for medical purposes polymer matrix of bone tissue implants highly porous adsorbents for preconcentration of volatile bases highly porous adsorbents for preconcentration of volatile acids gel materials for calibration of NMR tomographs immobilisation of AOT micelles containing enzymes preparation of composites containing supported catalyst particles as fillers gel matrices for controlled drug release carriers for enzyme immobilisation carriers for immobilisation of microbial cells carriers for immobilisation of animal cells gel bases for solid nutritional media for microbe cultivation gel bases for solid nutritional media for plant cell cultivation Support for the preparation of immunosorbents meant for isolation, purification, and modification of viruses chromatographic materials protective coatings for frozen fish and meat composite ice as a structural material thawed soil densification artificial baits for sports and amateur fishing biocompatible materials for medical purposes macroporous membranes biocompatible gel materials gel matrix for controlled drug release support for immobilised microbial cells gel materials for mechanochemical actuators gel matrix for controlled drug release biocompatible gel materials gel dosimetric system production of structured food forms 505 Ref.3, 140, 142, 149 ± 153, 158, 160 ± 164 143, 172 165 141, 144 ± 148, 156, 157, 159, 166 168 22, 284 ± 287 22, 288 ± 290 291, 292 22, 208, 209 293 210, 294 211 266, 295, 296 262 297 ± 299 11, 300 ± 306 307 ± 325 6, 10, 22, 224, 252, 326 ± 381 382 6, 22, 383 6, 22, 384 385, 386 387 388 ± 392 22, 393 22, 394 395 396 ± 405 406, 407 408, 409 168, 261, 410 411 ± 413 7, 414 ± 422 423 248, 424, 425 426 236 ± 243, 427 ± 433506 problems solved using cryogels is permanently extending.In Table 5, types of cryogels are designated according to the numbers of the Sections of the review in which these cryogels are consid- ered: IV.1 are cryogels formed upon polymerisation (r implies radiation initiation, c means chemical initiation) or polyconden- sation (pc); IV.2 are cryogels prepared by covalent cross-linking of poly- meric precursors; IV.3 are ionic cryogels (the letters correspond to the pathways of cryogel formation presented in Fig. 12); IV.4 are non-covalent cryogels with a physical network of the polymeric phase. VI. Conclusion To summarise the discussion on the most important features of cryotropic gelation of polymeric systems, the properties of the gel specimens thus formed and some applications of these materials, several conclusions can be drawn.1. Cryotropic gelation and concomitant chemical reactions (including the non-covalent interactions between the precursor molecules) are liquid-phase processes occurring in the non-frozen liquid microphase of moderately frozen (macroscopically solid) specimens. 2. Cryotropic gelation can take place in both aqueous and organic frozen media provided that the solvent used and the conditions of cooling ensure crystallisation rather than vitrifica- tion of the system. 3. The final products of cryotropic gelation (cryostructuring) are macroporous, often spongy gel-like materials called cryogels (cryostructurates).The role of the porogen in the formation of these polymeric materials is played by crystals of the frozen solvent; therefore, the pattern of porosity of the cryogels is determined by conditions of freezing, incubation in the frozen state, and, in some cases, thawing of the gelling system. The polymeric network of the gel phase (walls of macropores) possess intrinsic microporosity. 4. Due to the accumulation of soluble components in the non- frozen liquid microphase, cryotropic gelation is characterised by an apparent decrease in the CCG with respect to that observed for the gelation in the same initial systems at positive temperatures. 5.As a rule, moderate freezing of the initial system accelerates gelation. 6. The temperature dependences of the rate and efficiency of cryotropic gelation usually pass through an extremum due to the competition between factors that favour (cryoconcentration effects) and hamper (low temperature, high viscosity, and so on) the process. 7. The properties of the polymeric cryogels are determined by the temperature of the cryogenic process, the nature of the solvent used, the type (monomer or polymer precursors) and the concen- tration of gel-forming reagents, the presence of other soluble materials and insoluble fillers, the rate of freezing, the duration of storage in the frozen state, thawing dynamics, and some other factors. 8. Aunique combination of a number of properties, first of all, high porosity and interconnection of macropores and the facility of the process technology for the preparation of gel materials with this type of morphology allow cryogels to be regarded as a new type of polymer systems interesting in both scientific and applied aspects.The cryotropic gelation technique opens up prospects for the perfection of existing materials and development of new materials for various practical applications. The author is grateful to A N Chumaevskaya (Institute of Organoelement Compounds, Russian Academy of Sciences) for performing the patent search and toRV Ivanov andAAKorenev (Institute of Organoelement Compounds, Russian Academy of Sciences) for assistance in the preparation of pictorial data for this review.V I Lozinsky The review was written with partial financial support of the INTAS (Grant, No. 00-0057). References 1. G B Chizhov Teplofizicheskie Protsessy v Kholodil'noi Tekhnologii Pishchevykh Produktov (Thermal Processes in Colloid Technology of Foodstuffs) (Moscow: Pishchevaya Promyshlennost', 1979) 272 p. 2. G B Sergeev, V A Batyuk Kriokhimiya (Cryochemistry) (Moscow: Khimiya, 1978) 295 p. 3. I Kaetsu Adv. Polym. Sci. 105 81 (1993) 4. M Nambu Kobunshi Ronbunshu 47 695 (1990) 5. N A Peppas, Sh R Stauffer J. Contr. Release 16 305 (1991) 6. V I Lozinsky, A V Vakula, A L Zubov Biotekhnologiya (4) 5 (1992) 7. M Suzuki, O Hirasa Adv. Polym. Sci. 110 241 (1993) 8. V I Lozinsky, A L Zubov, in Proceedings of the International Work- shop `Bioencapsulation VI' (Eds F Godia, D Poncelet) (Barcelona: Autonomas University of Barcelona, 1997) p.T1.7 9. V I Lozinsky Usp. Khim. 67 641 (1998) [Russ. Chem. Rev. 67 573 1998] 10. V I Lozinsky, F M Plieva Enzyme Microb. Technol. 23 227 (1998) 11. Ch M Hassan, N A Peppas Adv. Polym. Sci. 153 37 (2000) 12. I V Lekhin, F N Petrov (Ed.) Slovar' Inostrannykh Slov (Vocabulary of Foreign Words) (Moscow: GIINS, 1955) 13. V I Lozinsky, E S Vainerman, S V Rogozhin Colloid Polym. Sci. 260 776 (1982) 14. V I Lozinsky, in Proceedings of the 216th Annual ACS Meeting. Division of Polymeric Materials: Science and Engineering Vol. 79 (Boston: American Chemical Society, 1998) p. 238 15. S P Papkov Studneobraznoe Sostoyanie Polimerov (Jelly-like State of Polymers) (Moscow: Khimiya, 1974) 256 p.16. V Kudela, in Encyclopedia of Polymer Science and Engineering Vol. 7 (New York: Wiley, 1987) p. 783 17. T Tanaka, in Structure and Dynamics of Biopolymers (Ed. C M Nicolini) (Dordrecht: Martinus Nijhoff, 1987) p. 237 18. O E Filippova Vysokomol. Soedin., Ser. C 42 2328 (2000) a 19. S B Ross-Murhy, H McEvoy Br. Polym. J. 18 2 (1986) 20. A A Tager Fiziko-Khimiya Polimerov (Physical Chemistry of Polymers) (Moscow: Khimiya, 1968) p. 427 21. P A Rebinder Izbrannye Trudy. Poverkhnostnye Yavleniya v Dispersnykh Sistemakh. Fiziko-Khimicheskaya Mekhanika (Selected Proceedings. Surface Effects in Dispersed Systems. Physicochemical Mechanics) (Moscow: Nauka, 1979) 381 p.22. V I Lozinsky, Doctoral Thesis in Chemical Sciences, Institute of Organoelement Compounds, Russian Academy of Sciences, Moscow, 1994 23. M Richter, S Augustat, F Schirbaum Ausgewahlte Methoden Der Starkchemie (Leipzig: VEB Fachbuchverlag, 1969) 24. J E Brigham, M J Gidley, R A Hoffmann, C G Smith Food Hydrocolloids 8 331 (1994) 25. R Tanaka, T Hatakeyama, H Hatakeyama Polym. Int. 45 118 (1998) 26. V I Lozinsky, L G Damshkaln, C R T Brown, I T Norton Polym Int. 49 1434 (2000) 27. S V Rogozhin, E S Vainerman, V I Lozinsky Dokl. Akad. Nauk SSSR 263 115 (1982) b 28. USSR P. 1 008 214; Byull. Izobret. (12) 131 (1983) 29. V I Lozinsky,M N Korneeva, E S Vainerman, S V Rogozhin Dokl. Akad. Nauk SSSR 270 101 (1983) b 30.V I Lozinsky, G F Korotaeva, E S Vainerman, S V Rogozhin Colloid Polym. Sci. 262 617 (1984) 31. L V Domotenko, V I Lozinsky, E S Vainerman, S V Rogozhin Vysokomol. Soedin., Ser. A 30 1661 (1988) a 32. Y Mori, H Tokura,M Yoshikawa J. Mater. Sci. 32 491 (1997) 33. L G Damshkaln, I A Simenel, V I Lozinsky J. Appl. Polym. Sci. 74 1978 (1999) 34. V I Lozinsky, L G Damshkaln J. Appl. Polym. Sci. 77 2017 (2000) 35. V A Kireev Kurs Fizicheskoi Khimii (Course of Physical Chemistry) (Moscow: Khimiya, 1975) 36. Yu I Aleksandrov Tochnaya Kriometriya Organicheskikh Veshchestv (Precise Cryometry of Organic Substances) (Leningrad: Khimiya, 1975) 161 p. 37. H T Meryman, in Cryobiology (Ed. H T Meryman) (London; New York: Academic Press, 1966) p. 1Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications 38.D Fox,M M Labes, A Weissberger (Eds) Physics and Chemistry of the Organic Solid State (New York; London: Interscience, 1965) 39. F Franks (Ed.) Water and Aqueous Solutions at Subzero Temperatures (New York; London: Plenum, 1982) 40. J Hallett Am. Sci. 72 582 (1984) 41. K R Diller Cryobiology 22 268 (1985) 42. P Smith, A J Pennings Polymer 15 413 (1974) 43. A A Berlin, E A Penskaya Dokl. Akad. Nauk SSSR 110 585 (1956) b 44. W Kuhn, H Majer Angew. Chem. 68 345 (1956) 45. H H G Jellinek, S Y Fok Makromol. Chem. 104 18 (1967) 46. N K Baramboim Mekhanokhimiya Vysokomolekulyarnykh Soedinenii (Mechanochemistry of High-Molecular-Weight Compounds) (Moscow: Khimiya, 1978) 47.W Kuhn, E Peterli, H Majer J. Polym. Sci. 16 539 (1955) 48. G Kanig Kolloid-Z. Z. Polym. 173 97 (1960) 49. W Kuhn, R Bloch, P Moser Experientia 18 197 (1962) 50. B B Boonstra, F A Hechman, G L Taylor J. Appl. Polym. Sci. 12 223 (1968) 51. D J Solms, A M Rijke J. Phys. Chem. 75 2623 (1971) 52. H Oikawa, K Murakami J. Macromol. Sci., B Phys. 28 187 (1989) 53. M A Masse, E Kiran, A L Fricke Polymer 27 619 (1986) 54. N H Grant, D E Clark, H E Alburn J. Am. Chem. Soc. 83 4476 (1961) 55. N H Grant,D E Clark,H E Alburn J. Am. Chem. Soc. 84 876 (1962) 56. N H Grant, D E Clark, H E Alburn J. Am. Chem. Soc. 88 4071 (1966) 57. A R Butler, T C Bruice J. Am. Chem. Soc. 86 313 (1964) 58. T C Bruice, A R Butler J. Am.Chem. Soc. 86 4104 (1964) 59. R E Pincock, T E Kiovsky J. Chem. Educ. 43 358 (1966) 60. R E Pincock Acc. Chem. Res. 2 97 (1969) 61. R E Pincock, T E Kiovsky J. Am. Chem. Soc. 87 2072 (1965) 62. G B Sergeev, V A Batyuk Usp. Khim. 45 793 (1976) [Russ. Chem. Rev. 45 391 (1976)] 63. V A Batyuk, Doctoral Thesis in Chemical Sciences, Moscow State University, Moscow, 1985 64. V A Batyuk, in Khimiya Nizkikh Temperatur i Kriokhimicheskaya Tekhnologiya (The Chemistry of Low Temperature and Cryochemical Technology) (Moscow: Moscow State University, 1987) p. 163 65. G B Sergeev, V A Batyuk,M B Stepanov, B M Sergeev Dokl. Akad. Nauk SSSR 213 891 (1973) b 66. V I Kvlividze, M B Pylova Kolloidn. Zh. 39 1167 (1977) c 67. D G Gusev, V I Lozinsky, V I Bakhmutov Eur.Polym. J. 29 49 (1993) 68. O Ya Grinberg, A T Nikitaev, K I Zamaraev, A Ya Lebedev Zh. Strukt. Khim. 10 230 (1969) d 69. E M Mil', A L Kovarskii, A M Vasserman Izv. Akad. Nauk SSSR, Ser. Khim. 2211 (1973) e 70. O I Mikhalev,M Serpinski, V I Lozinsky, P V Kapanin, I I Chkheidze,M V Alfimov Cryo-Lett. 12 197 (1991) 71. G B Sergeev, B M Sergeev, I A Kalnina Zh. Fiz. Khim. 58 2431 (1984) f 72. G B Sergeev, V A Batyuk, T Vegvart,M B Stepanov Vestn. Mosk. Univ., Ser. 2, Khim. 19 156 (1978) g 73. G B Sergeev, BMSergeev, I A Kalnina Kinet. Katal. 26 1069 (1985) h 74. E L Andronikashvili Biofizika 17 1068 (1972) i 75. I D Kuntz, in Proteins at Low Temperatures (Ed. O Fennema) Washington, DC: American Chemical Society, 1979) p. 27 76.J Pouchly, J Biros, S Benes Makromol. Chem. 180 745 (1979) 77. S Katayama, S Fujiwara J. Phys. Chem. 84 2320 (1980) 78. F Franks, in Effects of Low Temperatures on Biological Membranes (Eds G J Morris, A Clarke) (London: Academic Press, 1981) p. 3 79. E Suzuki, N Nagashima Bull. Chem. Soc. Jpn. 55 2730 (1982) 80. V I Lozinsky, E S Vainerman, L V Domotenko, A L Blyumenfel'd, V V Rogov, E N Barkovskaya, E I Fedin, S V Rogozhin Kolloidn. Zh. 51 685 (1989) c 81. D G Gusev, V I Lozinsky, E S Vainerman, V I Bakhmutov Magn. Reson. Chem. 28 651 (1990) 82. N R Konstantinova, V I Lozinsky Food Hydrocolloids 11 113 (1997) 83. V I Lozinsky, T O Golovina, D G Gusev Polymer 41 35 (2000) 84. H Oikawa, K Murakami Polymer 27 1563 (1986) 85. I D Kuntz J.Am. Chem. Soc. 93 514 (1971) 86. K Ogino, H Sato J. Polym. Sci., Polym. Phys. 33 445 (1995) 87. F Horii, K Masuda, H Kaji Macromolecules 30 2519 (1997) 88. K Masuda, F Horii Macromolecules 31 5810 (1998) 507 89. O I Mikhalev, I V Yakovleva, V I Trofimov, A B Shapiro Cryo-Lett. 6 245 (1985) 90. O I Mikhalev, A M Kaplan, V I Trofimov, V L Tal'roze Dokl. Akad. Nauk SSSR 287 385 (1986) b 91. O I Mikhalev, I N Karpov, E B Kazarova, M V Alfimov Chem. Phys. Lett. 164 96 (1989) 92. M Serpinski, Candidate Thesis in Physico-Mathematical Sciences, Institute of Chemical Physics, Russian Academy of Sciences, Moscow, 1992 93. K Nakamura, T Hatakeyama, H Hatakeyama Polymer 21 871 (1983) 94. T Hatakeyama, K Nakamura, H Hatakeyama Makromol.Chem. 21 1265 (1983) 95. K Kamide, K Okajima, T Matsui, H Kobayashi Polym. J. 16 259 (1984) 96. H Kumagai, K Nakamura, J Fujiwara Agric. Biol. Chem. 49 3097 (1985) 97. T de Vringer, J G H Joosten, H E Junginger Colloid Polym. Sci. 264 623 (1986) 98. Y H Ross J. Food Sci. 51 684 (1986) 99. T Hatakeyama, H Yoshida, H Hatakeyama Polymer 28 1282 (1987) 100. T Lovric, V Pilizota, A Janekovic J. Food Sci. 52 772 (1987) 101. M Jochem, C KoÈ rber Cryobiology 24 513 (1987) 102. O Miyawaki, T Abe, T Yano Agric. Biol. Chem. 52 2995 (1988) 103. S Charoenrein, D S Reid Thermochim. Acta 156 373 (1989) 104. D Simatos,G Blond, in Water Relationships in Food (EdsH Levine, L Slade) (New York: Plenum, 1991) p. 139 105. W L Kerr, J Ju, D S Reid J.Food Sci. 58 675 (1993) 106. H D Goff Pure Appl. Chem. 67 1801 (1995) 107. R Tanaka, T Hatakeyama,H Hatakeyama, in Gums and Stabilizers for the Food Industry (Eds P A Williams, G O Phillips) (London: The Royal Society of Chemistry, 1998) p. 43 108. H Hatakeyama, T Hatakeyama Thermochim. Acta 308 3 (1998) 109. M Tanaka, T Motomura, N Ishii, K Shimura,M Onishi, A Mochizuki, T Hatakeyama Polym. Int. 49 1709 (2000) 110. E A Podorozhko, E A Kurskaya, V K Kulakova, V I Lozinsky Food Hydrocolloids 14 111 (2000) 111. L Huang, K Nishinari J. Polym. Sci., Polym. Phys. 39 496 (2001) 112. A R Khokhlov, P G Khalatur Physica A 249 249 (1998) 113. E A Zheligovskaya, P G Khalatur, A R Khokhlov Phys. Rev. E 59 3071 (1999) 114. V I Lozinsky, I A Simenel, E A Kurskaya, V K Kulakova, V Ya Grinberg, A S Dubovik, I Yu Galaev, B Mattiasson, A R Khokhlov Dokl.Akad. Nauk 375 637 (2000) b 115. N S Pushkar', A M Belous (Eds) Aktual'nye Problemy Kriobiologii (Topical Problems of Cryobiology) (Kiev: Naukova Dumka, 1981) 116. G B Sergeev, V A Batyuk, B M Sergeev Dokl. Akad. Nauk SSSR 214 1354 (1974) b 117. G B Sergeev, A I Gudima, B M Sergeev, A A Taran, V A Batyuk Biokhimiya 46 986 (1981) j 118. H Meos, V ToÄ ugu,M Haga, A Aaviksaar,M Schuster, H-D Jakubke Tetrahedron: Asymmetry 4 1559 (1993) 119. S Gerisch, H-D Jakubke, H-J Kreuzfeld Tetrahedron: Asymmetry 6 3039 (1995) 120. A JoÈ nsson, P Aldercreutz, B Mattiasson Biotechnol. Bioeng. 46 429 (1995) 121. M Haensler, G Ullmann, H-D Jakubke J.Peptide Sci. 1 283 (1995) 122. M Haensler, H-D Jakubke J. Peptide Sci. 2 279 (1996) 123. M Haensler, S Gerisch, J Rettelbusch, H-D Jakubke J. Chem. Technol. Biotechnol. 68 202 (1997) 124. G Ullmann, M Haensler,W Gruender, M Wagner, H-J Hofmann, H-D Jakubke Biochim. Biophys. Acta 1338 253 (1997) 125. M Haensler, S Thust, P Klossek, G Ullmann J. Mol. Catal., B Enzym. 6 95 (1999) 126. N A Preobrazhenskii, R P Evstigneeva (Eds) Khimiya Biologiche- ski Aktivnykh Prirodnykh Soedinenii (The Chemistry of Biologically Active Natural Compounds) (Moscow: Khimiya, 1970) Vol. 1 127. Yu A Ovchinnikov Bioorganicheskaya Khimiya (Bioorganic Chemistry) (Moscow: Prosveshchenie, 1987) 128. H H G Jellinek, S Y Fok Kolloid Z. Z. Polym. 220 122 (1967) 129. S Katayama, S Fujiwara J.Am. Chem. Soc. 101 4485 (1979) 130. H Oikawa, K Murakami Polymer 25 225 (1984)508 131. A A Shits, in Entsiklopediya Polimerov (Encyclopedia of Polymers) Vol. 1 (Moscow: Sovetskaya Entsiklopediya, 1972) p. 594 132. V I Irzhak, B A Rozenberg, N S Enikolopyan Setchatye Polimery (Network Polymers) (Moscow: Nauka, 1974) 248 p. 133. L Z Rogovina, G L Slonimskii Usp. Khim. 43 1102 (1974) [Russ. Chem. Rev. 43 503 (1974)] 134. P J Flory, in Encyclopedia of Polymer Science and Engineering Vol. 10 (New York: Wiley, 1987) p. 95 135. I Kaetsu,H Okubo,A Ito,K Hayashi J. Polym. Sci., Polym. Chem. 10 2203 (1972) 136. I Kaetsu, A Ito, K Hayashi J. Polym. Sci., Polym. Chem. 11 1811 (1973) 137. M Yoshida,M Kumakura, I Kaetsu Polymer 20 3 (1979) 138.M Yoshida,M Kumakura, I Kaetsu Polymer 20 9 (1979) 139. V A Kargin, V A Kabanov Zh. Vsesoyuz. Khim. O-va im. D I Mendeleeva 9 602 (1964) k 140. I Kaetsu,M Kumakura, M Yoshida Biotechnol. Bioeng. 21 847 (1979) 141. T Fujimura, I Kaetsu Z. Naturforsch., C Biosci. 37 102 (1982) 142. M Kumakura, I Kaetsu J. Appl. Biochem. 5 165 (1983) 143. M Kumakura, I Kaetsu J. Appl. Biochem. 5 348 (1983) 144. T Fujimura, I Kaetsu Appl. Biochem. Biotechnol. 8 145 (1983) 145. M Kumakura, I Kaetsu J. Chem. Technol. Biotechnol. 33V 95 (1983) 146. M Kumakura, I Kaetsu J. Appl. Polym. Sci. 28 3759 (1983) 147. I Kaetsu,M Kumakura, S Kikuchi, S Adachi, M Suzuki Z. Naturforsch. C, Biosci. 38 812 (1983) 148. T Fujimura, I Kaetsu Int.J. Appl. Radiat. Isot. 34 929 (1983) 149. M Kumakura, I Kaetsu Int. J. Biochem. 16 1159 (1984) 150. M Kumakura, I Kaetsu Polym. J. 16 113 (1984) 151. M Kumakura, I Kaetsu J. Appl. Polym. Sci. 29 2713 (1984) 152. M Kumakura, I Kaetsu, T Kobayashi Enzyme Microb. Technol. 6 23 (1984) 153. M Kumakura, I Kaetsu Biosci. Rep. 4 181 (1984) 154. M Kumakura, I Kaetsu, K Nisizawa Biotechnol. Bioeng. 26 17 (1984) 155. M Kumakura, I Kaetsu Process Biochem. 19 124 (1984) 156. T Fujimura, I Kaetsu Z. Naturforsch., C Biosci. 40 576 (1985) 157. M Kumakura, I Kaetsu, K Nisizawa Biotechnol. Appl. Biochem. 8 195 (1986) 158. M Kumakura, I Kaetsu Process Biochem. 23 51 (1988) 159. M Tamada, N Kasai, I Kaetsu Biotechnol. Bioeng. 32 386 (1988) 160. M Kumakura, I Kaetsu J.Appl. Polym. Sci. 37 2229 (1989) 161. V Ya Kabanov Usp. Khim. 67 861 (1998) [Russ. Chem. Rev. 67 783 (1998)] 162. I Kaetsu Radiat. Phys. Chem. 47 419 (1996) 163. O Z Higa, M Kumakura Biomaterials 18 697 (1997) 164. M Kumakura Polym. Adv. Technol. 12 415 (2001) 165. K Sutani, I Kaetsu, K Uchida Radiat. Phys. Chem. 61 49 (2001) 166. T Fujimura, I Kaetsu Biotechnol. Bioeng. 29 171 (1987) 167. M Kumakura, I Kaetsu Makromol. Chem. 184 1831 (1983) 168. F Yoshi, I Kaetsu Appl. Biochem. Biotechnol. 8 115 (1983) 169. M Kumakura, I Kaetsu J. Mol. Catal. 23 1 (1984) 170. M Kumakura, I Kaetsu J. Chem. Technol. Biotechnol. 34V 39 (1984) 171. K Piskin, E Arca, E Piskin Appl. Biochem. Biotechnol. 10 73 (1984) 172. M Kumakura, I Kaetsu Colloid Polym.Sci. 262 450 (1984) 173. V I Lozinsky, E S Vainerman, E F Titova, E M Belavtseva, S V Rogozhin Colloid Polym. Sci. 262 769 (1984) 174. E M Belavtseva, E F Titova, V I Lozinsky, E S Vainerman, S V Rogozhin Colloid Polym. Sci. 262 775 (1984) 175. V I Lozinsky, E S Vainerman, S A Ivanova, E F Titova, M I Shtil'man, E M Belavtseva, S V Rogozhin Acta Polym. 37 142 (1986) 176. V I Lozinsky, S A Morozova, E S Vainerman, E F Titova, M I Shtil'man, E M Belavtseva, S V Rogozhin Acta Polym. 40 8 (1989) 177. B J Tighe, in Proceedings of the 3rd International Conference on Diffusion in Polymers, New York, 1991 p. 11 178. P G Righetti, B S W Brost, R S Snyder J. Biochem. Biophys. Methods 4 347 (1981) 179. R Dagani Chem. Eng. News 75 26 (1997) 180.H G Schild Prog. Polym. Sci. 17 163 (1992) 181. S Hirotsu Adv. Polym. Sci. 110 1 (1993) V I Lozinsky 182. I Yu Galaev Usp. Khim. 64 505 (1995) [Russ. Chem. Rev. 64 471 (1995)] 183. Y H Bae, T Okano, S W Kim J. Polym. Sci., Polym. Phys. 28 923 (1990) 184. L I Valuev, O N Zefirova, I V Obydennova, N A Plate J. Bioactive Compatible Polym. 9 55 (1994) 185. P-Y Yeh, P KopecÆ kova, J KopecÆ ek J. Polym. Sci., Polym. Chem. 32 1627 (1994) 186. E E Makhaeva, L T M Thanh, S G Starodubtsev, A R Khokhlov Macromol. Chem. Phys. 197 1973 (1996) 187. I V Bakeeva, I I Pashkin, Yu E Kirsh, V A Tverskoi, V P Zubov Vysokomol. Soedin., Ser. A 42 1016 (2000) a 188. B G Kabra, S H Gehrke Polym. Commun. 32 322 (1991) 189. X S Wu, A S Hoffman, P Yager J.Polym. Sci., Polym. Chem. 30 2121 (1992) 190. Q Yan, A S Hoffman Polymer 36 887 (1995) 191. J Chen, K Park J. Contr. Release 65 73 (2000) 192. V I Lozinsky, E A Kalinina, V Ya Grinberg, N V Grinberg, V V Chupov, N A Plate Vysokomol. Soedin., Ser. A 39 1972 (1997) a 193. X-Z Zhang, R-X Zhuo Macromol. Chem. Phys. 200 2602 (1999) 194. N N Lyubavin Zh. Russ. Fiz.-Khim. O-va, Khim. Otd. 21 397 (1889) 195. S I D'yachkovskii Zh. Obshch. Khim. 1 964 (1931) l 196. A Buzagh Kolloid Z. 146 133 (1956) 197. V L Zolotavin, V V Vol'khin, V V Rezvushkin Kolloidn. Zh. 22 305 (1960) c 198. E I Ponomarev, B I L'vovich, S A Kolesova Izv. Sib. Otd. Akad. Nauk, Ser. Khim. Nauk 11 (3) 57 (1965) 199. V V Vol'khin, E I Ponomarev, V L Zolotavin Kolloidn.Zh. 35 144 (1973) c 200. R K Iler The Chemistry of Silica (New York: Wiley-Interscience, 1979) 201. P W J G Wijnen, T P M Beelen, K P J Rummens, H C P L Saeijs, J W De Haan, L J M Van de Ven, R A Van Santen J. Colloid Interface Sci. 145 17 (1991) 202. D Fukushima J. Am. Oil Chem. Soc. 58 346 (1981) 203. T Watanabe J. Am. Oil Chem. Soc. 51 111A (1974) 204. K Hashizume, T Watanabe Agric. Biol. Chem. 35 449 (1971) 205. E S Wainerman, V I Lozinsky, S V Rogozhin Colloid Polym. Sci. 259 1198 (1981) 206. V I Lozinsky, T O Golovina, E S Vainerman, S V Rogozhin Vysokomol. Soedin., Ser. A 31 334 (1989) a 207. USSR P. 1 171 474; Byull. Izobret. (29) 105 (1985) 208. USSR P. 1 171 476; Byull. Izobret. (29) 106 (1985) 209. A Seidel Solubilities of Inorganic and Metal Organic Compounds Vol. 1 (New York: Van Nostrand, 1953) 210.Russ. P. 2 035 476; Chem. Abstr. 128 13 896 (1998) 211. Russ. P. 2 078 099; Chem. Abstr. 124 58 298 (1996) 212. H Berghmans, in Integration of Fundamental Polymer Science and Technology (Eds P J Lemstra, L A Kleintjens) (Amsterdam: Elsevier ,1988) p. 296 213. S V Rogozhin, V I Lozinsky, E S Vainerman, L V Domotenko, A M Mamtsis, S A Ivanova,M I Shtil'man, V V Korshak Dokl. Akad. Nauk SSSR 278 129 (1984) b 214. V I Lozinsky, L G Damshkaln, R Brown, I T Norton J. Appl. Polym. Sci. 75 1740 (2000) 215. V I Lozinsky, L G Damshkaln, R Brown, I T Norton J. Appl. Polym. Sci. 83 1658 (2002) 216. W E Schulze, D T Yu,M M MacMasters Staerke 16 41 (1964) 217.W S Chan, T R Toledo J. Food Sci. 4 301 (1976) 218. A C Eliasson, H R Kim J. Texture Stud. 23 279 (1992) 219. A S Navarro,M N Martino, N E Zaritzky J. Food Eng. 34 411 (1997) 220. V I Lozinsky, L V Domotenko, E S Vainerman, A M Mamtsis, S V Rogozhin Polym. Bull. 15 333 (1986) 221. P H Richardson, A H Clark, A L Russell, P Aymard, I T Norton Macromolecules 32 1519 (1999) 222. V I Lozinsky, E V Solodova, A L Zubov, I A Simenel J. Appl. Polym. Sci. 58 171 (1995) 223. V I Lozinsky, L V Domotenko, A L Zubov, I A Simenel J. Appl. Polym. Sci. 61 1991 (1996) 224. V I Lozinsky, A L Zubov, E F Titova Enzyme Microb. Technol. 18 561 (1996) 225. Russ. P. 2 058 083; Chem. Abstr. 126 18 107 (1997)Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications 226.F Yokoyama, I Masada, K Shimamura, T Ikawa K Monobe Colloid Polym. Sci. 264 595 (1986) 227. C M Hassan, N A Peppas Macromolecules 33 2472 (2000) 228. C M Hassan, J H Ward, N A Peppas Polymer 41 6729 (2000) 229. M Watase Nippon Kagaku Kaishi (7) 973 (1983) 230. M Watase Nippon Kagaku Kaishi (9) 1254 (1983) 231. M Watase, K Nishinari Makromol. Chem. 190 155 (1989) 232. F Urushizaki, H Yamaguchi, K Nakamura, S Namajiri, K Sugibayashi, Y Morimoto Int. J. Pharm. 58 135 (1990) 233. R V Golovnya, T A Misharina, M B Terenina Nahrung 42 380 (1998) 234. A G Filatova, I O Volkov, N I Krikunova, T A Misharina, R V Golovnya Izv. Akad. Nauk, Ser. Khim. 312 (2000) e 235. V I Lozinsky, L G Damshkaln, R Brown, I T Norton J.Appl. Polym. Sci. 78 371 (2000) 236. H S Guraya, C James, E T Champague Starch/Staerke 53 64 (2001) 237. M Fuchigami, N Kato, A Teramoto, in High Pressure Bioscience and Biotechnology (EdsR Hayashi,C Balny) (Amsterdam: Elsevier, 1996) p. 379 238. M Fuchigami, A Teramoto, in High Pressure Bioscience and Bio- technology (Eds R Hayashi, C Balny) (Amsterdam: Elsevier, 1996) p. 411 239. M Jegouic, V Ya Grinberg, A Guingant, T Haertle J. Protein Chem. 15 501 (1996) 240. M Fuchigami, A Teramoto J. Food Sci. 62 828 (1997) 241. M Fuchigami, A Teramoto, N Ogawa J. Food Sci. 63 1054 (1998) 242. M Shimoyamada, K Toà matsu, K Watanabe Food Sci. Technol. Res. 5 284 (1999) 243. M Shimoyamada, K Toà matsu, S Oku, K Watanabe J.Agric. Food Chem. 48 2775 (2000) 244. V I Lozinsky, A L Zubov, I N Savina, F M Plieva J. Appl. Polym. Sci. 77 1822 (2000) 245. V I Lozinsky, L G Damshkaln J. Appl. Polym. Sci. 82 1609 (2001) 246. V I Lozinsky, L V Domotenko, E S Vainernan, A M Mamtsis, E F Titova, E M Belavtseva, S V Rogozhin Colloid Polym. Sci. 264 19 (1986) 247. V I Lozinsky, A L Zubov, V K Kulakova, E F Titova, S V Rogozhin J. Appl. Polym. Sci. 44 1423 (1992) 248. P Giusti, L Lazzerri, N Barbani, P Narducci, A Bonaretti, M Palla, L Lelli J. Mater. Sci., Mater. Med. 4 538 (1993) 249. H H Trieu, S Qutubuddin Colloid Polym. Sci. 272 301 (1994) 250. H H Trieu, S Qutubuddin Polymer 36 2531 (1995) 251. M Kobayashi, I Ando, T Ishii, Sh Amiya Macromolecules 28 6677 (1995) 252.V I Lozinsky, A L Zubov, E F Titova Enzyme Microb. Technol. 20 182 (1997) 253. P J Willcox, D W Howie Jr, K Schimdt-Rohr, D A Hoagland, S P Gido, S Pudjijanto, L W Kleiner, S Venkatraman J. Polym. Sci., Polym. Phys. 37 3438 (1999) 254. F Fergg, F J Keil, H Quader Colloid Polym. Sci. 279 61 (2001) 255. F Fergg, F J Keil Chem. Eng. Sci. 56 1305 (2001) 256. K Nishinari, M Watase, F Tanaka, J. Chim. Phys. 93 880 (1996) 257. R Tanaka, T Hatakeyama,H Hatakeyama, in Gums and Stabilizers for the Food Industry (Eds P A Williams, G O Phillips) (London: The Royal Society of Chemistry, 1998) p. 43 258. Ch M Hassan, N A Peppas J. Appl. Polym. Sci. 76 2075 (2000) 259. M Watase, K Nishinari Makromol. Chem. 186 1081 (1985) 260. M Nagura,M Nagura, H Ishikawa Polym.Commun. 25 313 (1984) 261. H S Shin, S Y Kim, Y M Lee, K H Lee, S J Kim, Ch E Rogers J. Appl. Polym. Sci. 69 479 (1998) 262. Y E Shapiro, E G Pykhteeva, A V Levashov J. Colloid Interface Sci. 206 168 (1998) 263. Y E Shapiro J. Colloid Interface Sci. 212 453 (1999) 264. Y E Shapiro, T I Shapiro J. Colloid Interface Sci. 217 322 (1999) 265. L Galantini, A A D'Archivio, S Lora, B Corain J. Mol. Catal., B Enzym. 6 505 (1999) 266. M J Gordon, K C Chu, A Margaritis, A J Martin, C R Ethier, B K Rutt Biotechnol. Bioeng. 65 459 (1999) 267. Y E Shapiro Colloids Surf. A, Physicochem. Eng. Aspects 164 71 (2000) 268. D Dibbern-Brunelli, T D Z Atvars J. Appl. Polym. Sci. 75 815 (2000) 509 269. V I Lozinsky, F M Plieva, I Yu Galaev, B Mattiasson Bioseparation 10 (4 ± 6) (2002) 270.K A Lusta, N G Starostina, N B Gorkina, B A Fikhte, V I Lozinsky, E S Vainerman, S V Rogozhin Prikl. Biokhim. Mikrobiol. 24 504 (1988) m 271. O I Slabova, D I Nikitin, V I Lozinsky, E S Vainerman, S V Rogozhin Mikrobiologiya 60 23 (1991) n 272. P Arvidsson, F M Plieva, I N Savina, V I Lozinsky, S Fexby, L Bulow, I Yu Galaev, B Mattiasson J. Chromatogr. (2002) 273. O I Slabova, D I Nikitin, V I Lozinsky, V K Kulakova, E S Vainerman, S V Rogozhin Mikrobiologiya 57 940 (1988) n 274. P Dhulster, P Parascandola, V Scardi Enzyme Microb. Technol. 5 65 (1983) 275. M F Cocquemcot, D Thomas, M L Champigny, A Moyse Eur. J. Appl. Microbiol. Biotechnol. 8 37 (1979) 276. M F Cocquemcot, B Thomasset, J N Barbotin, G Gellf, D Thomas Eur.J. Appl. Microbiol. Biotechnol. 11 193 (1981) 277. M F Cocquemcot, V Larreta-Garde, D Thomas Biochimie 62 615 (1980) 278. M F Cocquemcot, J F Hervagault, D Thomas Enzyme Microb. Technol. 8 533 (1986) 279. R Carpentier, S Lemieux Appl. Biochem. Biotechnol. 15 107 (1987) 280. R Carpentier, S Lemieux, M Mineault Biotechnol. Lett. 10 133 (1988) 281. G C Papageorgiou, T Lagoyanni Appl. Microbiol. Biotechnol. 23 417 (1986) 282. F Estival, C Burstein Enzyme Microb. Technol. 7 29 (1985) 283. E Yu Baukova, L M Bronshtein, E Sh Mirzoeva, E S Vainerman, A L Blyumenfel'd, S Yu Panov, O L Lependina, P M Valetskii Vysokomol. Soedin., Ser. A 35 233 (1993) a 284. L H H Olde-Damink, P J Dijkstra,M J A van Luyn, P B van Wachem, P Nieuwenhuis, J Feijen J.Mater. Sci., Mater. Med. 6 429 (1995) 285. T V Gnezdilova, Candidate Thesis in Chemical Sciences, Institute of Organoelement Compounds, Russian Academy of Sciences, Moscow, 1994 286. Russ. P. 2 053 796; Chem. Abstr. 126 37 131 (1997) 287. Russ. P. 2 116 801; Chem. Abstr. 133 109 927 (2000) 288. T V Gnezdilova, E F Titova, L G Radchenko, V I Lozinsky, in Tez. Dokl. 5-i Vsesoyuz. Konf. po Khimii Nizkikh Temperatur, Moskva, 1991 (Abstracts of Reports of the 5th All-Union Confer- ence on the Chemistry of Low Temperatures, Moscow, 1991) p. 55 289. Russ. P. 2 008 362; Chem. Abstr. 123 290 337 (1995) 290. Russ. P. 2 116 350; Chem. Abstr. 133 106 553 (2000) 291. L Shapiro, S Cohen Biomaterials 18 583 (1997) 292.R Glicklis, L Shapiro, R Agbaria, J C Merchuk, S Cohen Biotechnol. Bioeng. 67 344 (2000) 293. Y Zhang,M Zhang J. Biomed. Mater. Res. 55 304 (2001) 294. Russ. P. 2 066 449; Chem. Abstr. 126 338 212 (1997) 295. I Mano, H Goshima, M Nambu,M Iio Magn. Reson. Med. 3 921 (1986) 296. K C Chu, B K Rutt Magn. Reson. Med. 37 314 (1997) 297. U Prusse, S Horold, K-D Vorlop Chem. Ing. Tech. 69 100 (1997) 298. U Prusse, V Morawsky, A Dierich, A Vaccaro, K-D Vorlop, in Preparation of Catalysts Vol. VII (Eds B Delmon, P A Jacobs, R Maggi, J A Martens, P Grande, D Poncelet) (Amsterdam: Elsevier, 1998) p. 137 299. U PruÈ sse, M Hihnlein, J Daum, K-D Vorlop Catal. Today 55 79 (2000) 300. N A Peppas, J E Scott J. Contr.Release 18 95 (1992) 301. S R Stauffer, N A Peppas Polymer 33 3932 (1992) 302. A Takamura, F Ishii, H Hidaka J. Contr. Release 20 21 (1992) 303. B J Ficek, N A Peppas J. Contr. Release 27 259 (1993) 304. M G Cascone, L Di Silvio, B Sim, S Downes J. Mater. Sci., Mater. Med. 5 770 (1994) 305. M G Cascone, B Sim, S Downes Biomaterials 16 564 (1995) 306. C M Hassan, J E Stewart, N A Peppas Eur. J. Pharm. Biopharm. 49 161 (2000) 307. O Ariga, T Yamakawa, H Fujimatsu, Y Sano J. Ferment Bioeng. 68 293 (1989) 308. W-I Cha, S-H Hyon, Y I Ikada Kobunshi Ronbunshu 48 425 (1991) 309. E Kokufuta, E Jinbo Macromolecules 25 3549 (1992) 310. T Yamauchi, E Kokufuta, Y Osada Polym. Gels Networks 1 247 (1993)510 311. T Hayashi, S-H Hyon, W-I Cha, Y I Ikada J.Appl. Polym. Sci. 49 2121 (1993) 312. S M Gautier, P E Michel, L J Blum Analyt. Lett. 27 2055 (1994) 313. V I Lozinsky, F M Plieva, A L Zubov Biotekhnologiya (1 ± 2) 32 (1995) 314. C Cristallini, L Lazerri,M G Cascone, G Polacco, D Lupinacci, N Barbani Polym. Int. 44 510 (1997) 315. L Doretti, D Ferrara, P Gattolin, S Lora Talanta 44 859 (1997) 316. L Doretti, D Ferrara, P Gattolin, S Lora, F Schiavon, F M Veronese Talanta 45 891 (1998) 317. F M Veronese, C Mammucari, P Caliceti, F Schiavon, S Lora J. Bioact. Compat. Polym. 14 315 (1999) 318. M I Shtil'man, S N Zhavoronkov,M Yu Yarmysh, P S Voskanyan, A M Tsatsakis Zh. Prikl. Khim. 72 1001 (1999) o 319. Yu N Belokon', K A Kochetkov, F M Plieva, N S Ikonnikov, V I Maleev, V S Parmar, R Kumar, V I Lozinsky Appl.Biochem. Biotechnol. 88 97 (2000) 320. F M Plieva, K A Kochetkov, I Singh, V S Parmar, Yu N Belokon', V I Lozinsky Biotechnol. Lett. 22 551 (2000) 321. A V Bacheva, F M Plieva, E N Lysogorskaya, I Yu Filippova, V I Lozinsky J. Peptide Sci. 6 (Suppl.) 113 (2000) 322. B Orlich, E Kempin, K-H Reichert, R Schomacker Chem. Ing. Tech. 72 1209 (2000) 323. A V Bacheva, F M Plieva, E N Lysogorskaya, I Yu Filippova, V I Lozinsky Bioorg. Med. Chem. Lett. 11 1005 (2001) 324. I Yu Filippova, A V Bacheva, O V Baibak, F M Plieva, E N Lysogorskaya, E S Oksenoit, V I Lozinsky Izv. Akad. Nauk, Ser. Khim. 1811 (2001) e 325. E Efremenko, V Lozinsky, V Sergeeva, F Plieva, T Makhlis, G Kazankov, A Gladilin, S Varfolomeyev J.Biochem. Biophys. Methods 51 195 (2002) 326. USSR P. 1 351 971; Byull. Izobret. (42) 89 (1987) 327. R Zh Manolov, V I Lozinsky, I M Tavobilov, E S Vainerman, S I Bezborodova, S V Rogozhin, A M Bezborodov Biotekhnol. Biotekh. 2 3 (1987) 328. O Ariga, H Takagi, H Nishizawa, Y Sano J. Ferment. Technol. 65 651 (1987) 329. V I Lozinsky, E S Vainermzn, S V Rogozhin, in Tez. Dokl. 4-go Vsesoyus. Simpoziuma `Inzhenernaya Enzimologiya', Kiev, 1983 (Abstracts of Reports of the 4th All-Union Symposium `Engineer- ing Enzymology', Kiev, 1983) p. 54 330. USSR P. 1 405 310; Byull. Izobret. (32) 279 (1990) 331. R Zh Manolov, I M Tavobilov, V I Lozinsky, E S Vainerman, E F Titova, E M Belavtseva, S I Bezborodova, S V Rogozhin, A M Bezborodov Prikl.Biokhim. Mikrobiol. 24 514 (1988) m 332. V I Lozinsky, N G Faleev, A L Zubov, S B Ruvinov, T A Antonova, E S Vainerman, V M Belikov, S V Rogozhin Biotechnol. Lett. 11 43 (1989) 333. USSR P. 1 505 972; Byull. Izobret. (33) 119 (1989) 334. L V Belousova, V I Lozinsky, E S Vainerman, S V Rogozhin, S N Egorov, N S Egorov, in Fermenty Mikroorganizmov (Enzymes of Microorganisms) (Moscow: VNIISENTI, 1989) Pt. 2, p. 224 335. V I Lozinsky, F M Plieva, in Proceedings of the International Workshop `Bioencapsulation V', Potsdam, 1996 p. T3 336. G P Alabyan, Doctoral Thesis in Chemical Sciences, Institute of Organic Chemistry, National Academy of Sciences of Armenia, Erevan, 2000 337. S D Varfolomeev, E I Rainina, V I Lozinsky, S B Kalyuzhnyi, A P Sinitsyn, T A Makhlis, G P Bachurina, I G Bokova, O A Sklyankina, E V Agafonov, in Physiology of Immobilized Cells (Eds J A M de Bont, J Visser, B Mattiasson, J Tramper) (Wageningen: Elsevier, 1989) p.325 338. V I Lozinsky, E S Vainerman, A L Zubov, V K Kulakova, S V Rogozhin Biotekhnologiya (5) 32 (1990) 339. G P Alebyan, E N Arzumanov, M B Mkrtchan, P V Tozalakyan, V I Lozinsky, E S Vainerman, S V Rogozhin Biotekhnologiya (6) 29 (1990) 340. A L Zubov, V I Lozinsky, E S Vainerman, S V Rogozhin, in Fermenty Mikroorganizmov i Degradatsiya Biopolimerov (Enzymes of Microorganisms and Degradation of Biopolymers) (Ed. V G Debabov) (Moscow: VNIISENTI, 1990) p. 111 341. M N Filimonova, G K Gabdulina, I F Zabbarova Prikl. Biokhim. Mikrobiol.26 706 (1990) m V I Lozinsky 342. A P Sinitsyn, E I Rainina, V I Lozinsky, S D Spasov, in Immobi- lizovannye Kletki Mikroorganizmov (Immobilized Cells of Micro- organisms) (Sofia: Sofia University, 1991) 343. H Myoga,H Asano,Y Nomura,H Yoshida Water Sci. Technol. 23 1117 (1991) 344. M A Pusheva, A N Ryabokon',M V Kevbrina, N P Bolotina, E I Rainina Mikrobiologiya 60 616 (1991) n 345. S J Velizarov, E I Rainina, A P Sinitsyn, S D Varfolomeev, V I Lozinsky, A L Zubov Biotechnol. Lett. 14 291 (1992) 346. S D Varfolomeev, E I Rainina, V I Lozinsky Pure Appl. Chem. 64 1193 (1992) 347. A L Simonian, E I Rainina, V I Lozinsky, I E Badalian, G A Khachatrian, S Sh Tatikian, T A Makhlis, S D Varfolomeev 353. O Ariga,M Kato, T Sano, Y Nakazawa, Y Sano J.Ferment Appl. Biochem. Biotechnol. 36 199 (1992) 348. K D Vorlop, P Remmers DECHEMA Biotech. Conf. 5 939 (1992) 349. H Asano, H Myoga,M Asano,M Toyao Water Sci. Technol. 26 1037 (1992) 350. H Asano, H Myoga,M Asano,M Toyao Water Sci. Technol. 26 2397 (1992) 351. R J Manolov Appl. Biochem. Biotechnol. 33 157 (1992) 352. O A Nikitina, S S Zatsepin, S V Kalyuzhnyi, E I Rainina, S D Varfolomeev, A L Zubov, V I Lozinsky Mikrobiologiya 62 477 (1993) n Bioeng. 76 203 (1993) 354. K Furukawa, A Ike, S L Ryu,M Fujita J. Ferment. Bioeng. 76 515 (1993) 355. R Zh Manolov World J. Microbiol. Biotechnol. 9 29 (1993) 356. E I Rainina,M A Pusheva, A M Ryabokon', N P Bolotina, V I Lozinsky, S D Varfolomeev Biotechnol. Appl. Biochem. 19 321 (1994) 357.V V Fokina, A Yu Arinbasarova, A L Zubov, V I Lozinsky, K A Koshcheenko Prikl. Biokhim. Mikrobiol. 31 213 (1995) m 358. Russ. P. 2 036 095; Byull. Izobret. (15) (1995) 359. A N Ryabokon',M A Pusheva, E N Detkova, E I Rainina Mikrobiologiya 64 776 (1995) n 360. S Ledakowicz, U Becker, W-D Deckwer Catal. Today 24 65 (1995) 361. V Fokina, N Susina, A Arinbasarova, A Zubov, V Lozinsky, K Koshcheenko, in Immobilized Cells: Basics and Applications (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p. 90 362. A M Ryabokon',M V Kevbrina, M A Pusheva, A L Zubov, V I Lozinsky, E A Rainina, in Immobilized Cells: Basics and Applications (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p.106 363. V I Lozinsky, A L Zubov, T A Makhlis, in Immobilized Cells: Basics and Applications (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p. 112 364. E J T M Leenen, V A P MDos Santos, J Tramper, R H Wijffels, in Immobilized Cells: Basics and Applications (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p. 205 365. S Bielecki, R Bolek, in Immobilized Cells: Basics and Applications (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p. 472 366. W M Rostron, D C Stuckey, in Immobilized Cells: Basics and Applications (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p. 703 367. T Willke, K-D Vorlop, in Immobilized Cells: Basics and Applica- tions (Eds R H Wijffels, R M Buitelaar, C Bucke, J Tramper) (Amsterdam: Elsevier, 1996) p.718 368. E J T M Leenen, V A P MDos Santos,K C F Grolle, J Tramper, R H Wijffels Water Res. 30 2895 (1996) 369. E I Rainina, E N Efremenko, S D Varfolomeev, A L Simonian, J R Wild Biosens. Bioelectron. 11 991 (1996) 370. E I Rainina, I E Badalian, O V Ignatov, A B Fedorov, A L Simonian, S D Varfolomeev Appl. Biochem. Biotechnol. 56 117 (1996) 371. Russ. P. 2 080 388; Byull. Izobret. (15) (1997) 372. J-W Kim, E I Rainina, E N Efremenko, C R Engler, J R Wild Biotechnol. Lett. 19 1067 (1997) 373. C Buchtmann, U Kies, W-D Deckwer, V Hecht Biotechnol. Bioeng. 56 295 (1997) 374. Russ. P. 2 104 866; Byull. Izobret. (5) (1998)Cryogels on the basis of natural and synthetic polymers: preparation, properties and applications 375. S Gough, N Barron, A L Zubov, V I Lozinsky, A P McHale Bioprocess Eng. 19 87 (1998) 376. E Leidig, U PruÈ sse, K-D Vorlop, J Winter Bioprocess Eng. 21 5 (1999) 377. J-W Kim, E I Rainina, C R Engler, J R Wild Can. J. Chem. Eng. 77 883 (1999) 378. Y-P Ting, G Sun J. Chem. Technol. Biotechnol. 75 541 (2000) 379. A Yu Fedorov, E V Volchenko, I N Singirtsev, V I Korzhenevich, G M Shub Prikl. Biokhim. Mikrobiol. 36 59 (2000) m 380. M Szczesna, E Galas, S Bielecki J. Mol. Catal., B Enzym. 11 671 (2001) 381. M Szczesna-Antczak, E Galas Biomol. Eng. 17 55 (2001) 382. M G Cascone,M Laus, D Ricci, R S del Guerra J. Mater. Sci., Mater. Med. 6 71 (1995) 383. USSR P. 1 213 069; Byull. Izobret. (7) 146 (1986) 384. L V Domotenko, Candidate Thesis in Chemical Sciences, Institute of Organoelement Compounds, Academy of Sciences of the USSR, Moscow, 1986 385. F M Plieva, E I Isaeva, V I Lozinsky Biotekhnologiya (5) 32 (1998) 386. Russ. P. 2 130 069; Chem. Abstr. 133 101 928 (2000) 387. R Murakami, H Hachisako, K Yamada, Y Motozato J. Chromatogr. 678 180 (1994) 388. USSR P. 683 705; Byull. Izobret. (33) 15 (1979) 389. L P Zuikova, E I Izyumova, G G Korneva Rybnoe khoz-vo (1) 70 (1977) 390. V E Shapovalov, E V Yakush,M N Vakhrusheva, V F Mikhaleva, N A Gradov Rybnoe khoz-vo (11) 66 (1982) 391. USSR P. 971 209; Byull. Izobret. (41) 16 (1982) 392. L I Bulatnikova, E V Plisko, O I Belyatskaya Plast. massy (7) 52 (1983) 393. USSR P. 1 649 219; Byull. Izobret. (18) 145 (1991) 394. USSR P. 1 705 500; Byull. Izobret. (2) 131 (1992) 395. Russ. P. 2 054 254; Byull. Izobret. (5) 8 (1996) 396. J A Stammen, S Williams,D N Ku,R E Guldberg Biomaterials 22 799 (2001) 397. Y Honda, S Y Kim, M Nambu, N Naoi, E Sakaue Doc. Ophthalmol. 63 205 (1986) 398. M Kuriaki, K Nakamura, J Mizutani Kobunshi Ronbunshu 46 739 (1989) 399. K Morimoto, A Nagayasu, S Fukanoki, K Morisaka, S-H Hyon, Y Ikada Drug Develop. Ind. Pharm. 16 13 (1990) 400. C H Lee, K J Lee, A J Park, Y A Shin Arch. Pharm. Res. 16 43 (1993) 401. L Lelli, N Barbani, A Bonaretti, P Guisti, P Marchetti, R Giannarelli, R Navalesi, P Viacava J. Mater. Sci., Mater. Med. 5 887 (1994) 402. K Ito, N Minoura, S Aiba, M Nagura, M Kodama J. Biomed. Mater. Res. 28 459 (1994) 403. R S del Guerra,M G Cascone, N Barbani, L Lazzeri J. Mater. Sci., Mater. Med. 5 613 (1994) 404. L Lazzeri, N Barbani, M G Cascone, D Lupinocci, P Giusti, M Laus J. Mater. Sci., Mater. Med. 5 862 (1994) 405. H S Shin, S Y Kim, Y M Lee J. Appl. Polym. Sci. 65 685 (1997) 406. T Hirai, K Hanaoka, T Suzuki, S Hayashi Kobunshi Ronbunshu 46 613 (1989) 407. M Tasaki, T Kida, T Homma Kobunshi Ronbunshu 46 733 (1989) 408. M G Cascone, S Maltini J. Mater. Sci., Mater. Med. 10 301 (1999) 409. M G Cascone, S Maltini, N Barbani, M Laus J. Mater. Sci., Mater. Med. 10 431 (1999) 410. P Martens, K S Anseth Polymer 41 7715 (2000) 411. M Okazaki, T Hamada, H Fujii, A Mizobe, S Matsuzawa J. Appl. Polym. Sci. 58 2235 (1995) 412. M Okazaki, T Hamada, H Fujii, O Kusuodo, A Mizobe, S Matsuzawa J. Appl. Polym. Sci. 58 2243 (1995) 413. M C Doria-Serrano, F A Ruiz-Trevino, C Rios-Arciga, M Hernandez-Esparza, P Santiago Biomacromolecules 2 568 (2001) 414. T Shiga, Y Hirose, A Okada, T Kurauchi Kobunshi Ronbunshu 46 709 (1989) 415. T Shiga, Y Hirose, A Okada, T Kurauchi J. Appl. Polym. Sci. 44 249 (1992) 416. M Yonese, K Baba,M Kondo, H Kishimoto Chem. Lett. 199 (1992) 417. O Kolosov,M Suzuki, K Yamanaka J. Appl. Phys. 74 6707 (1993) 511 418. T Shiga, Y Hirose, A Okada, T Kurauchi J. Intell. Mater., Syst. Struct. 4 553 (1993) 419. T Shiga, Y Hirose, A Okada, T Kurauchi J. Appl. Polym. Sci. 47 113 (1993) 420. N Tsunemoto, M Suzuki Polym. Gels Networks 2 247 (1994) 421. Y Mori, H Tokura,M Yoshikawa Kagaku Kogaku Ronbunshu 20 625 (1994) 422. S Y Kim, H S Shin, Y M Lee, Ch N Jeong J. Appl. Polym. Sci. 73 1675 (1999) 423. M G Cascone, N Barbanti, S Maltinti, L Lazzeri Polym. Int. 50 1241 (2001) 424. L Lazzeri Trends Polym. Sci. 4 249 (1996) 425. K C Chu, K J Jordan, J J Battista, J Van Dyk, B K Rutt Phys. Med. Biol. 45 955 (2000) 426. R Lawrence, P Jelen Food Microstructure 1 91 (1981) 427. S V Rogozhin, G L Slonimskij, L Z Rogovina, E S Wainerman, P P Pivovarov Nahrung 25 391 (1981) 428. V B Tolstoguzov, E E Braudo J. Text. Stud. 14 183 (1983) 429. S V Rogozhin, E M Belavtseva, E S Vainerman, L G Radchenko, P P Pivovarov, T O Golovina, F V Pertsevoi Nahrung 28 165 (1984) 430. P J Lillford, in Properties of Water in Foods (Eds D Simatos, J L Multon) (Dordrecht: Martinus Nijhoff, 1985) P. 543 431. R Lawrence, F Consolacion, P Jelen Food Technol. 40 77 (1986) 432. N R Konstantinova, V I Lozinsky, in Proceedings of the 2nd International Conference on Low Temperature Chemistry, Kansas City, KS, 1996 p. 195 433. Russ. P. 2 118 495; Chem. Abstr. 133 16 693 (2000) a�Polym. Sci. (Engl. Transl.) b�Dokl. Chem. (Engl. Transl.) c�Colloid. J. (Engl. Transl.) d�J. Struct. Chem. (Engl. Transl.) e�Russ. Chem. Bull., Int. Ed. (Engl. Transl.) f�Russ. J. Phys. Chem. (Engl. Transl.) g�Moscow Univ. Bull. (Engl. Transl.) h�Kinet. Catal. (Engl. Transl.) i�Biophysics (Engl. Transl.) j�Biochemistry (Engl. Transl.) k�Mendeleev Chem. J. (Engl. Transl.) l�Russ. J. Gen. Chem. (Engl. Transl.) m�Appl. Biochem. Microbiol. (Engl. Transl.) n�Microbiology (Engl. Transl.) o�Russ.
ISSN:0036-021X
出版商:RSC
年代:2002
数据来源: RSC
|
4. |
Synthesis of 2'-O-alkylnucleosides |
|
Russian Chemical Reviews,
Volume 71,
Issue 6,
2002,
Page 513-534
Timofei S. Zatsepin,
Preview
|
|
摘要:
Russian Chemical Reviews 71 (6) 513 ± 534 (2002) Synthesis of 20-O-alkylnucleosides T S Zatsepin, E A Romanova, T S Oretskaya Contents I. Introduction II. Alkylation of nucleosides with diazomethane and trimethylsulfonium hydroxide III. Alkylation of nucleosides with methyl iodide in the presence of silver(I) oxide IV. Alkylation of nucleosides in the presence of bases V. Preparation of 20-O-alkylnucleosides by glycosylation of heterocyclic bases VI. Alkylation of the activated cis-diol group of nucleosides in the presence of transition metal compounds VII. Synthesis of pyrimidine 20-O-alkylnucleosides from O2,20-anhydronucleosides VIII. Other methods for the synthesis of 20-O-alkylnucleosides IX. Conclusion Abstract. of synthesis the on data published The The published data on the synthesis of 20-O-alkylnucleo- sides are reviewed.The known methods are systematised and their sides are reviewed. The known methods are systematised and their advantages and disadvantages are discussed. The bibliography advantages and disadvantages are discussed. The bibliography includes 174 references includes 174 references. I. Introduction Synthesis of modified oligonucleotides is a popular trend in the modern organic chemistry of nucleic acids. The introduction of modified fragments into the oligonucleotide chain allows one to solve diverse problems, such as the enhanced transport of DNA fragments through cell membranes, the increase in their resistance against nucleases and the increase in thermodynamic stabilities of various NA duplexes.The introduction of chemically reactive groups into oligonucleotides allows the preparation of various conjugates with peptides, dyes, etc. Post-synthetic random modi- fications or modifications at the 30- and/or 50-ends of oligonucleo- tides are possible in principle, but direct incorporation of a modified unit into a predetermined position of an oligomeric chain in the course of oligonucleotide synthesis is the most versatile approach. This review surveys the synthesis of modified nucleosides containing alkoxy groups at positions 20 of the carbohydrate moieties. These compounds are used as precursors of monomeric components in oligonucleotide synthesis. Preparation of now commercially available building blocks for the oligoribonucleo- tide synthesis, of conformationally locked nucleic acids (LNA) T S Zatsepin, T S Oretskaya Department of Chemistry,MV Lomonosov Moscow State University, 119992 Moscow, Leninskie Gory, Russian Federation. Fax (7-095) 939 31 81.Tel. (7-095) 939 31 48. E-mail: tsz@yandex.ru (T S Zatsepin) Tel. (7-095) 939 54 11. E-mail: oretskaya@belozersky.msu.ru (T S Oretskaya) E A Romanova A N Belozersky Institute of Physicochemical Biology, M V Lomonosov Moscow State University, 119992 Moscow, Leninskie Gory, Russian Federation. Fax (7-095) 939 31 81. E-mail: romanova@belozersky.msu.ru Received 13 March 2002 Uspekhi Khimii 71 (6) 586 ± 608 (2002); translated by R L Birnova #2002 Russian Academy of Sciences and Turpion Ltd DOI 10.1070RC2002v071n06ABEH000714 513 514 515 515 522 527 530 531 532 (their synthesis is not confined to the alkylation of the 20-hydroxy group) and 20-O-(b-D-ribofuranosyl)nucleosides (for reviews see Refs 1 and 2) are not considered here.Modified oligonucleotides containing an alkoxy group at the 20-position of the carbohydrate moiety cannot be prepared by modification of natural oligonucleotides. The only possible approach is the synthesis of a modified monomer and its incor- poration into the oligonucleotide chain in the course of automated oligonucleotide synthesis. 20-O-Methyladenosine was isolated for the first time from a natural source in 1959;3 approaches to the synthesis of this class of compounds were developed soon thereafter (in 1965 ± 1966).4±6 In the past decade, modified oligonucleotides containing an alkoxy group in position 20 of the carbohydrate fragment are actively employed as antisense probes.The results of these studies are summarised in several books 7±9 and reviews.10 ± 13 It was demonstrated that 20-O-methyloligonucleotides form duplexes with RNA which are kinetically and thermodynamically more stable and simultaneously more sensitive to mismatches than non- modified oligoribonucleotides,14 which makes them indispensable tools in antisense biotechnology. Besides, 20-alkyloligonucleotides are used in the design of ribozymes 15 which are more resistant to nuclease digestion than their natural counterparts, and in studies of the processing of genetic material.16 ± 19 20-O-(3-Aminopropyl)- oligonucleotides are especially resistant against 30-exonucleases due to the interaction of the amino group with the metal-binding domain.This was demonstrated by X-ray diffraction analysis using the Klenow fragment from E. coli as an example.20 20-O-(2- Aminoethyl)oligonucleotides are the most potent inhibitors of the in vivo gene expression according to the antigenic mechanism, viz., the formation of a triplex with genomic DNA, in comparison with other known modified analogues of nucleic acids.21, 22 Oligonu- cleotides containing chemically reactive groups in position 20 of the carbohydrate fragment are used for conjugation with various compounds, e.g., fluorophores 23 (these conjugates are used for structural analyses of various RNAs), electrochemically active compounds 24 (design of DNA analysers), metal ion complexes 25 (chemical nucleases), DNA and RNA duplexes with covalently linked chains.26, 27 The alkylation of the 20-hydroxy group of nucleosides usually follows a nucleophilic substitution mechanism.514 XO B O OH OY X, Y are H or protective groups; B is a heterocyclic base; AlkZ is an alkylating reagent.Blocking of reaction centres of heterocyclic bases, 30- and 50-hydroxy groups to avoid side reactions is the central problem in the synthesis of 20-O-alkylnucleosides. The choice of suitable protective groups still remains a topical problem. The majority Table 1. Alkylation of the 20-hydroxy group of nucleosides with diazo- methane in the H2O±DME system.NucleosideNH2N N N HO (Ado) NO OH OH NH2 N N HO (Cyd) OO OH OHNH2N N N N Cl HO (1a) O OH OHOMe N N HO (1b) OO OH OH OH N HO (1c) OO OH OH O HN N TrO O O (2) OH TrO Note. Hereinafter, Tr=Ph3C. a The alkylation by diazoethane gave a mixture of ethyl and methyl derivatives (*1 : 1); b the yield of 20-O,3-N- dimethyl-30,50-di-O-tritylarabinouridine is indicated. XO B AlkZ O OY OAlk Ref. Yield (%) 40 27 38 20 22 a 4 28 29 30 33 28 25 32 30 ± 40 31 37 35 90 b0 34 Cyd Cyd Ado Guo Ado Cyd Ado Cyd Ado Cyd T S Zatsepin, E A Romanova, T S Oretskaya of currently known alkylation procedures imply the use of strong organic or inorganic bases, therefore classical protective groups are not always compatible with the reaction conditions.The review considers recent advances in the synthesis of 20-O- alkylnucleosides and historically significant methodologies. Such a choice of literature sources is determined by the fact that the main approaches to the synthesis of 20-O-alkylnucleosides have been known since the 1970's ± 1980's. However, modified 20-O- alkyloligonucleotides came into use only in the early 1990's. Studies in this field have led to the development of novel methods for the synthesis of 20-O-alkylnucleoside and optimisation of the existing methods. II. Alkylation of nucleosides with diazomethane and trimethylsulfonium hydroxide The alkylation procedures of nucleosides with diazomethane and trimethylsulfonium hydroxide have been proposed at different times, but here we consider them together because these reactions follow similar mechanisms.In addition, both these methods have only a historical significance at present. + + 7 Nu7 N Me N NuH+ N N CH2 7N2 7Nu7 MeNu +7 +7 D NuH Me3SOH Me2SCH2 7H2O 7Me2S Broom and Robins 4 have synthesised 20-O-methyladenosine by treatment of adenosine with diazomethane in aqueous dime- thoxyethane (Table 1) and these results were reproduced by different authors.28 ± 35 Later, diazomethane was used for the alkylation of other nucleosides (Cyd, 1a ± c, see Table 1). The reaction with diazoethane proceeds in a similar way,33 but a mixture of methyl and ethyl derivatives is formed in a *1 : 1 ratio.The formation of methyl ethers is rationalised by solvent participation. Alkylation of 30,50-di-O-tritylarabinouridine (2) under similar conditions led to the dialkylation product, viz., 20-O,3-dimethyl-30,50-di-O-tritylarabinouridine.35 With 3-de- azauridine (1c), the reaction proceeded selectively with the heter- ocyclic base (see Table 1).34 The use of trimethylsulfonium hydroxide as an alkylating reagent in the synthesis of 20-O-methylnucleosides was proposed and developed by Kinoshita et al. in 1978 ± 1980 36 ± 38 (Table 2). Table 2. Alkylation of the 20-hydroxy group of nucleosides with oxonium derivatives and trimethyl phosphate in DMF at 110 8C.Ref. Yield (%) Alkylating reagent Nucleoside ++++ 30 28 20 11 36, 37 38 38 38 6 3939 40 40 26 22 28 ++ Me3SOH7 Me3SOH7 Me3SOH7 Me3SOH7 Me3PhN+OH7 Me3PhN+OH7 (MeO)3PO (MeO)3PO Me3SOH7 (see a) 5±10 41 Me3SOH7 (see a) 5±10 41 O N HN N H2N N Note. . Guo=HO O OH OH a The reaction was carried out in DMSO with B(OH)3 addition at 110 8C.Synthesis of 20-O-alkylnucleosides This alkylating reagent is less reactive than diazomethane; there- fore, the reaction is carried out in DMF at 110 8C. The alkylation with trimethylanilinium hydroxide 39 and trimethyl phosphate 40 gave similar results. An attempt to use trimethylsulfonium iodide 38 resulted in compounds in which only the heterocyclic base was alkylated.The addition of boric acid to trimethylsulfo- nium hydroxide also led to the alkylation of heterocyclic bases.41 The main disadvantage of both methods is their low selectiv- ity, since the alkylation products of both the sugar residue and the heterocyclic base are formed. These possess similar chromato- graphic mobilities, which impedes separation of the reaction products. The yields of the target products are usually low. III. Alkylation of nucleosides with methyl iodide in the presence of silver(I) oxide Historically, the alkylation of nucleosides with methyl iodide in the presence of Ag2O was the first method for the preparation of 20-O-methylnucleosides. Furukawa et al.6 performed the alkyla- tion of 30,50-di-O-trityluridine (3a) devoid of protective groups in the heterocyclic base; therefore, the reaction partly involved the uracil residue (Table 3).This method was not further used until the early 1980's when this was rekindled for the alkylation of nucleosides with 30- and 50-hydroxy groups blocked by the 1,1,3,3- tetraisopropyldisiloxane-1,3-diyl protective group (TIPDS, the Markiewicz protection) and the protected heterocyclic base (Table 3, compounds 3b ± f, 4a,b, 5). Beigelman et al.51 made an attempt to synthesise 20-O-meth- ylcytidine from 50-O-dimethoxytrityl-30-O-tert-butyldimethyl- silyl-N4-isobutyrylcytidine (4c), which is an intermediate product in the preparation of building blocks for the oligoribo- nucleotide synthesis.This attempt was unsuccessful, since the alkylation was accompanied by the 30?20 migration and partial removal of the silyl protective group; 20,30-di-O-methylcytidine Table 3. Alkylation of the 20-hydroxy group of nucleosides with methyl iodide in the presence of Ag2O. Nucleoside R1 O Tr R3N R1O N* OO OH R2O TIPDS TIPDS TIPDS TIPDS TIPDS TIPDS TIPDS R3 N O N* TIPDS TIPDS TIPDS DMTr DMTr DMTr R3 TIPDS TIPDS N NN N* O TIPDS BzN S N* Note. Hereinafter, DMTr=(4-MeOC6H4)2PhC, TBDMS=ButMe2Si. a The yields of the alkylation products with methyl, ethyl, propyl, allyl and pentyl iodides are 93%, 83%, 75%, 79% and 50%, respectively; b the product was not isolated from the reaction mixture. R3 R2 Tr HCl3CCMe2OC(O) Bz Bz Bz 2,4-(NO2)2C6H3S 2-NO2C6H4S 4-MeOC6H4C(O) 2-NO2C6H4O BzNH BzNH TBDMS PriCONH BzNH 4a 4b 4b 4c 4d 4-ButC6H4OCH2C(O)NH 4f HH Cl Cl 7 515 and 30-O-methylcytidine were formed as side products.Hodge and Sinha 52 have synthesised 20-O-methyl, -ethyl, -allyl, -propyl and -pentyl nucleoside derivatives 4d,f (the reaction was carried out in the presence of 0.1 ± 0.2 equiv. of pyridine). The yields of 20-O- alkylnucleosides decreased with an increase in the size of the alkyl group but remained at a sufficiently high level. Special mention should be made of the fact that this approach is suitable for large- scale (tens of grams) syntheses (see Table 3, compound 4d).52 Today, the alkylation with methyl iodide in the presence of silver(I) oxide has proved to be a fairly efficient approach to the synthesis of pyrimidine 20-O-alkylnucleosides. At the same time, it is seldom used for the synthesis of purine derivatives.It is of note that the methylation of partially protected 2-thiouridine (6) 53 gave a complex mixture, which may be due to the interaction of Ag2O with the thioketone group and subsequent alkylation of the heterocyclic base. IV. Alkylation of nucleosides in the presence of bases 1. The use of alkali metal hydroxides There are two approaches to the alkylation of nucleosides in the presence of alkali metal hydroxides: the reaction may be carried out in non-aqueous or in aqueous media.In the mid-1960's, Reese et al.5 performed the alkylation of 30,50-di-O-trityluridine (3a) with benzyl chloride in the presence of potassium hydroxide in dry dioxane, which resulted in protected 20-O-benzyluridine (7) (Table 4). Ura TrO Ura TrO BnCl, KOH O O dioxane OH OBn OTr OTr3a 7 Ref. Reaction conditions Yield (%) Nucleoside : MeI :Ag2O Com- pound No. 6 42 Me2CO, 20 8C, 72 h 40 8C, 7 h 3a 3b 3c 3c 3c 3d 3e 3f 50 93 Me2CO, 20 8C 86 43 85 40 94 92 100 1 : 15 : 5 1 : 100 : 5 1 : 15 : 10 1 : 26 : 2.9 1 : 26 : 2.9 1 : 10 : 8 1 : 10 : 8 1 : 7 : 8 44, 45 46 47 47 48 PhH, 40 8C, 16 h PhH, 40 8C, 16 h Me2CO, 20 8C, 16 h Me2CO, 20 8C, 16 h 40 8C, 5 h Me2CO, 20 8C, 72 h PhH, 40 8C, 4 h PhH, 20 8C, 2 h 49 44, 45 50 51 52 52 81 69 98 26 see.a see.a 1 : 20 : 10 1 : 26 : 2.9 1 : 20 : 3 7 7 1 : 25 : (1.5 ± 2.5) 1 : 25 : (1.5 ± 2.5) PhMe, 0 8C, 5 ± 17 h PhMe, 0 8C, 5 ± 17 h 44, 45 46 66 40 1 : 26 : 2.9 1 : 26 : 2.9 PhH, 40 8C, 1 h PhH, 40 8C, 4 h 556 53 see.b 7 7516 Table 4.Alkylation of the 20-hydroxy group of nucleosides in the presence of alkali metal hydroxides. Nucleoside Compound Alkylating reagent No. Ura 3a TrO O OH TrO Cyd Ado 8 Cyt HO OH O HO Ade 9 HO OH O HO Ade 10a O O OH O 7 Cyt 10b O P OO O O P OH O 7O Note: Hereinafter, Ura is the uracil residue, Cyt is the cytosine residue, Ade is the adenine residue.a The yield over two steps (alkylation and detritylation) is indicated; b the yield over two steps (protection of 50- and 30-hydroxy groups and alkylation) is indicated. The advantage of this method is that the heterocyclic base is not alkylated; however, the yields of the steps of the synthesis of the starting compound (30,50-di-O-trityluridine) do not exceed 30%± 45%. Later, in the 1990's, the alkylation was performed with more complex alkyl halides.54, 55 In the early 1970's, it was proposed 56 to use dialkyl sulfates as alkylating reagents in the presence of alkali metal hydroxides in aqueous media (see Table 4, compounds 8, 9). It was shown 56 that at pH>13, no alkylation of the heterocyclic base occurs.The alkylation of adenosine and cytidine 30,50-cyclophosphates (10a,b) with alkyl halides in water ± organic media has been described.63 No side alkylation products at the 50- and 30-hydroxy groups were formed, which resulted in higher yields and facilitated the chro- matographic isolation of the reaction products. As can be seen from Table 4, the alkylation of the nucleoside 3a in non-aqueous media gives fairly good yields even in the case of rather complex alkyl halides, but drastic reaction conditions significantly restrict the use of this method. Alkylation by dialkyl sulfates and alkyl iodides in aqueous and water ± organic media proceeds as a rule with low yields and is of low practical value. 2. The use of organic bases Alkylation of nucleosides in the presence of organic bases was first proposed by Sproat et al.64 ± 70 (Table 5).The alkylation of BnCl CH2Cl CH2Cl (MeO)2SO2 (MeO)2SO2 (MeO)2SO2 (MeO)2SO2 (MeO)2SO2 (EtO)2SO2 (MeO)2SO2 (MeO)2SO2 MeI EtI MeI EtI Nucleoside : alkylating Reaction conditions reagent : sodium hydroxide 1 : 1.2 : 5 1 : 1.1 : 10 1 : 1.1 : 10 1 : 30 : 100 1 : 10: 50 1 : 6 : 30 1 : 30 : 100 1 : 2 : 45 1 : 2 : 45 1 : 6 : 30 1 : 6 : 30 1 : 10 : 3 1 : 10 : 3 1 : 10 : 3 1 : 10 : 3 partially protected nucleosides with alkyl halides was carried out in the presence of 2-tert-butylimino-2-diethylamino-1,3-dimethyl- perhydro-1,3,2-diazaphosphorine (BDDDP) (see Ref. 71). This reaction and subsequent extraction are not accompanied by the formation of hydroxides; therefore, the protective groups of heterocyclic bases are preserved.It is of note that in addition to the known groups, e.g., compounds 11a,b,d, 12, novel protective groups were introduced, viz., the pivaloyloxymethyl group (com- pound 13) for the protection of the N3-position of uracil and the tert-butyldiphenylsilyl group (compound 11c) for the blocking of the O6-position of guanine (see Table 5). The pivaloyloxymethyl protective group is more stable than the conventional benzoyl group. In addition, the N3-pivaloyloxymethyl derivative 13 is obtained in a higher yield than the N3-benzoyl derivative. The use of the tert-butyldiphenylsilyl protective group made it possible to increase the total yield and to reduce the number of steps in the synthesis of 20-O-propargylguanosine, since the 30- and 50-hydroxy groups and the O6 atom of the heterocycle are deprotected simultaneously under the action of tetrabutylammonium fluo- ride.64 The alkylation of compound 14a was accompanied by a side reaction at the exocyclic amino group; compound 14b did not undergo alkylation under these conditions.64 T S Zatsepin, E A Romanova, T S Oretskaya dioxane, D benzene ± dioxane, D, 2 h benzene ± dioxane, D, 2 h H2O, 20 8C, 5 h H2O, 20 8C, 5 h H2O±DME, 20 8C, 7 h H2O, 20 8C, 5 h H2O, 20 8C, 10 h H2O, 20 8C, 10 h H2O±DME, 20 8C, 7 h H2O±DME, 20 8C, 7 h H2O±DMF, 20 8C, 3 h H2O±DMF, 20 8C, 3 h H2O±DMF, 20 8C, 3 h H2O±DMF, 20 8C, 3 h Ref.Yield (%) 60 65a 5 54 55 48a 56 57 22 12 58 25 59 60 60 61 58 b 6 b 20 b 62 10 63 63 50 40 63 63 62 30Synthesis of 20-O-alkylnucleosides Table 5. Alkylation of the 20-hydroxy group of nucleosides in the presence of BDDDP. Nucleoside R2 N N N N R1 O O TIPDS OH O 2,5-Cl2C6H3O N N O O O TIPDS O OHO ButC(O)OCH2 N N O O O TIPDS O OH Note. The yield of compound 11c is indicated over two steps (alkylation and desilylation). Hereinafter, TBDPS is ButPh2Si; a the yields over two steps (alkylation and reduction of the crotyl residue to the butyl residue) are indicated. R2 N R1 N O O TIPDS O14a,b R1=NH2, R2=Ph2NCO2 (a); R1=PriCONH, R2=Ph2NCO2 (b).It was also shown 64 that the introduction of the carbamoyl protective group at the exocyclicN2 atom of guanosine decelerates the alkylation reaction and strongly decreases the yields. Nearly all the works from Sproat's laboratory are methodo- logically flawless: as a rule, all reactions proceed with high yields and result in the formation of single products. The syntheses are multistep, but the total yields are rather high. The main disad- vantage of this method is the high cost of BDDDP; therefore, the preparation of compounds in gram quantities using this method is hardly possible. These works are also interesting with respect to R2 R1 HHHHCl Cl Cl NH2 NH2 NH2 NH2 NH2 NH2 NH2 Cl Cl Cl Cl Cl Cl 2,5-Cl2C6H3O 2,5-Cl2C6H3O 2,5-Cl2C6H3O 2,5-Cl2C6H3O 2,5-Cl2C6H3O 2,5-Cl2C6H3O 2,5-Cl2C6H3O TBDPS7O TBDPS7O TBDPS7O TBDPS7O TBDPS7O TBDPS7O TBDPS7O 2-NO2C6H4O 2-NO2C6H4O 2-NO2C6H4O 2-NO2C6H4O 2-NO2C6H4O 2-NO2C6H4O NN OH Compo- Alkylating reagent und No.11a 11a 11a 11a 11b 11b 11b 11c 11c 11c 11c 11c 11c 11c 11d 11d 11d 11d 11d 11d MeI MeI HC:CCH2Br MeCH=CHCH2Br MeI MeI MeCH=CHCH2Br MeI MeI CH2=CHCH2Br CH2=CHCH2Br HC:CCH2Br N:CCH2Br EtO2CCH2Br MeI CH2=CHCH2Br HC:CCH2Br N:CCH2Br N:CCH2Br EtO2CCH2Br 12 12 12 12 MeI HC:CCH2Br N:CCH2Br EtO2CCH2Br 13 13 HC:CCH2Br N:CCH2Br the application of novel protective groups for the heterocyclic bases.3. The use of alkali metal hydrides With alkali metal hydrides as the bases, the alkylation may be carried out with both the unprotected and the partially protected nucleoside. In the former case, the selectivity of the reaction involving the 20-hydroxy group is due to its higher acidity in comparison with the 30- and particularly the 50-hydroxy group.72 Nevertheless, this is accompanied by the formation of side products including the 30-isomer with similar chromatographic mobilities; therefore, protected adenosine derivatives are some- times used (Table 6). In the case of other (especially, pyrimidine) nucleosides, the alkylation partly involves the heterocyclic base. The use of alkali metal hydrides as the bases demands careful selection of protective groups, since some of them are unstable under the reaction conditions.In addition, post-synthetic workup results in the formation of alkali metal hydroxides, which pre- cludes the application of certain standard protective groups. Nevertheless, at present this approach is one of the most popular ones. The method proposed by Robins et al.32 back in 1966 and recently employed by Yano et al.73 in which 2-amino-6-chloro- Nucleoside : alkyl halide :BDDDP 1 : 1.2 : 1.2 1 : 1.2 : 1.2 1 : 1.25 : 1.2 1 : 1.2 : 1.2 1 : 1.2 : 1.2 1 : 1.2 : 1.2 1 : 1.2 : 1.2 1 : 2.5 : 5 1 : 2 : 2 1 : 2.5 : 5 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 2 : 2 1 : 1.2 : 1.2 1 : 1.25 : 1.2 1 : 2 : 2 1 : 2 : 1.5 1 : 1.25 : 1.2 1 : 2 : 2 517 Ref.Reaction Yield (%) time /h 65 66 68 68 65 66 66 67 64 67 64 64 64 64 64 64 64 64 69 64 63 757 70 ± 75 a 68 70 70 ± 75 a 48 49 51 51 51 47 48 61 51 71 70 61 62 66266663333333333333 66 68 69 70 761 64 90 6231 68 69 48 65 23Table 6. Alkylation of the 20-hydroxy group of nucleosides in the presence of sodium hydride. NucleosideR4 N N N N R3 R1O O R2O OH R4 R3 R2 R1H H H 2 NH H H H 2 NH MeI H H H 2 NH H H H 2 NH EtI H H H 2 NH AllBr H H H 2 NH H H H 2 NH EtO2CCH2Br H H H 2 NH EtO2CCH2Br H H H 2 NH PhthN(CH2)3Br H H H 2 NH H H H 2 NH PhthN(CH2)2± 4Br H H H 2 NH Me(CH2)8Br H H H 2 NH Im(CH2)3Br H H H 2 NH H H H 2 NH Ph(CH2)2OTs H H H 2 NH H H H 2 NH H H H 2 NH H H 2 NH NH2 H H 2 NH NH2 H H 2 NH NH2 H H 2 NH NH2 H H 2 NH NH2 H H 2 NH NH2 H H 2 NH NH2 Alkylating reagent a MeI MeI AllBr PhthN(CH2)5Br TsS(CH2)6Br N Br NFe Br Fe Br NMe2 O AllBr BnO(CH2)nBr (n=2, 3) PhthN(CH2)3Br C4H9Br C7H15Br C10H21Br C16H33Br Nucleoside : alkylating reagent : sodium hydride 1 : 1.15 : 0.75 1 : 1.03 : 1.2 1 : 1.1 : 1.6 1 : 21 : 1 7 7 1 : 1.1 : 1 7 7 71 : 1.15 : 1.5 777771 : 1 : 1.2 1 : 1.2 : 1.3 1 : 1.1 : 0.5 1 : 0.9 : 0.5 1 : 1.1 : 1 71 : 1.15 : 1.5 7777 Reaction conditions DMF, 0 8C, 4 h DMF, 0 8C, 3.5 h DMF, 0 to 20 8C, 10 h DMF, 0 to 20 8C, 10 h DMF, 20 8C, 18 h DMF DMF, 30 8C, 17 h DMF, 0 ± 5 8 b 81 C 21 DMF DMF DMF DMF DMF, 0 8C, 3 h; 60 8C, 16 h DMF,75 8C, 24 h DMF, 20 8C, 18 h DMF, 35 8C, 48 h DMF, 20 8C, 18 h DMF DMF, 30 8C, 17 h DMF, 5 h DMF, 5 h DMF, 5 h DMF, 5 h Ref.Yield (%) 73 42 74 23 25 b 75 3.5 b 75 76 766 c 77 78 722 d 79 80 17 7 82 7 82 7 82 83, 84 23 85 15 86 18 87 45 88 30 77 c 77 89 48, 50 77777 80 90 90 90 90Table 6 (continued). NucleosideR4 N N N N R3 R1O O R2O OHR3 N N O R1O O R2O OHO R3 N N O R1O O R2O OH R2 R1H H 2 NH NH2 H H 2 NH OH H H N Cl H H H 2 NH 2-NO2C6H4(CH2)2O MeI H H 2 NH 2-NO2C6H4(CH2)2O EtI H Tr 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 7 H H 2 NH H H 2 N7 H H H 2 N7 H H H N 7 H TIPDS Tr Tr H DMTr 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 R4 R3 2 OMe H OMe H OBn N NMe2 NHCOPri 77 NHBz 7 NC(CH2)2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 7 BnOCH2 Alkylating reagent a MeO((CH2)2O)n(CH2)2I (n=0, 1, 2, 4) AllBr MeI BnCl MeO(CH2)2Br MeO(CH2)2Br MeI MeI EtI BnCl MeI BnCl MeI MeI AllBr PrBr PriBr EtO2CCH2Br TsO O O TsO O R-isomer O TsO O S-isomer PhthN(CH2)4Br MeO(CH2)2Br Nucleoside : alkylating reagent : sodium hydride 77 7 71 : 1.9 : 0.9 1 : 8 : 0.9 1 : 42 : 1.5 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.03 : 1.2 1 : 1.1 : 1.6 1 : 21 : 1 1 : 30 : 5 71 : 2 : 2 1 : 8 : 1.6 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 1 : 1.15 : 1 Reaction conditions DMF, 5 h DMF,720 8 e C 4 91 5 DMF,745 to 20 8C, 3 h DMF,750 to715 8C, 6 h PhH ± dioxane, 20 8C 4 THF, D THF, D DMF, 0 8C, 3.5 h DMF, 0 to 20 8C, 10 h DMF, 0 to 20 8C, 10 h DMF, 20 8C 12 94 DMF PhH ± dioxane, 20 8C 4 DMF,750 to715 8C, 6 h THF, D THF, D THF, D THF, D THF, D THF, D THF, D THF, D THF, D THF, D Ref.Yield (%) 7 90 76 7 75 22 75 131 9293 78 93 93 7 74 25 b 75 14 b 75 95 798 9675 33 93 98 93 88 93 67 93 20 93 97 93 72 93 34 93 35 93 64 93 83Table 6 (continued). R1 Nucleoside O R3 N N O R1O O R2O OHThy BnO O O OH BnO O Thy BnO O OH BnO Thy BnO O OH BnO Ura THPO OHO THPO Ade TBDMSO OHO TBDMSO R4 R3 R2 Alkylating reagent a 7 MeO((CH2)2O)n(CH2)2Br BnOCH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 (n=1, 2, 3) 7 C10H21O((CH2)2O)3(CH2)2Br BnOCH2 2,4-Cl2C6H3CH2 2,4-Cl2C6H3CH2 7 BnOCH2 TIPDS MeI 7 BnOCH2 TIPDS EtO2CCH2Br ButO2CCH2Br 7 TrS TIPDS DMF,745 to 20 8C 7 7 O O BnOCH2 H H O S O O 7 BnCl H Tr Tr 7 BnCl H Tr Tr7 7 7 MeI 7 7 7 7 MeI 7 7 7 7 MeI 7 7 7 7 AllBr 7 7 7 7 AllBr 7 Nucleoside : alkylating reagent : sodium hydride 1 : 1.15 : 1 1 : 1.15 : 1 1 : 10 : 1.2 1 : 5 : 2.2 1 : 4 : 4 O S O O , 1 : 20 : 10 1 : 20 : 0.9 1 : 7.4 : 2 1 : 2.6 : 7.5 1 : 5 : 3 1 : 2 : 2 1 : 2.5 : 2.5 Reaction conditions THF, D THF, D DMF, 0 8C, 2 h DMF,75 8C, 1.5 h THF, 18 h DMF, 20 8 g 102 C 33 DMF, 20 8C CH2Cl2, 36 8C, 23 h CH2Cl2, 20 8C, 48 h THF, 0 8C, 2 h THF, 20 8C, 17 h THF, 20 8C, 16 h Ref.Yield (%) 93 89, 76, 24 93 61 71 f 97 98 98 99 78 100, 101 63, 50 103 60 104 62 105 62 106 84 107, 108 40 107, 108 32Table 6 (continued). R1 Nucleoside O BzN N SO O TIPDS OH O R1 Me N N O HO O OH HO Note. a Abbreviations: Im, imidazolyl, Phth, phthaloyl; b The yield for a three-step synthesis (protection of the heterocyclic base, dimethoxytritylation, alkylation) is indicated; c the yield of a mixture of 20- and 30-isomers is indicated; d the yield over two steps (Markiewicz protection and alkylation) is indicated; e the total yield over two steps is indicated; f the yield over two steps (alkylation and desilylation) is indicated; g the yield of 20-O- benzyl-30,50-di-O-trityl-3-benzyluridine is indicated.R4 R3 R2 Alkylating reagent a 7 7 7 MeI 7 7 7 AllBr 7 NH2 7 7 AllBr 7 OH Reaction conditions Nucleoside : alkylating reagent : sodium hydride 7 PhH 7 7 7 7 Ref. Yield (%) 7 53 76 7 76 7522 purine riboside was used as the starting compound seems to be a very promising approach to the large-scale synthesis of 20-O- alkylguanosine derivatives. The total yield of alkylation and hydrolysis to 20-O-alkylguanosine is *45%; the use of expensive reagents is avoided. V. Preparation of 20-O-alkylnucleosides by glycosylation of heterocyclic bases The data on the synthesis of 20-O-alkylnucleosides by glycosyla- tion of heterocyclic bases are summarised in Tables 7 and 8.This involves two steps, viz., the alkylation of a partially protected carbohydrate and the glycosylation itself. In some cases, the alkylation of the 2-hydroxy group of the carbohydrate A is followed by a series of transformations in order to obtain the glycosyl donor B. X1O X1O alkyl- ation O OZ1 OH OY1A X2O BTMS, Cat O OZ2 OAlk OY2B BTMS is a silylated heterocyclic base; X1, X2, Y1, Y2, Z1, Z2 are protective groups. Table 7. Synthesis of protected 2-O-alkylcarbohydrates as precursors of 20-O-alkylnucleosides. Compound No. Blocked carbohydrate 15 OBn O O O OH OBz 16 BzO O OBz OH O 17 O OMe TIPDS OH O 18 OO O OH OO 2,4-Cl2C6H3CH2O 19 O OMe OH 2,4-Cl2C6H3CH2O O OH 20 O OMe TIPDS O 21 O O OBn TIPDS OH O a The yield for the alkylation of the b-anomer is indicated; b the yield for the alkylation of the a-anomer is indicated.The synthesis of 20-O-alkylnucleosides by glycosylation of heterocyclic bases was proposed in the early 1970's by Haines.109 20-O-Methyluridine was synthesised starting from the partially protected b-D-ribopyranoside 15, the total yield being as low as 5%.In the early 1980's, Imbach et al.110, 111 performed the syn- thesis of pyrimidine 20-O-methylnucleosides (total yield 50%± 55%) starting from the accessible 1,3,5-tri-O-benzoyl-b-D- ribofuranose (16). Later, this approach was used by Cook et al.112 for the synthesis of 20-O-methyluridines containing various sub- stituents at C5.The use of CF3SO3SiMe3 instead of SnCl4 as a glycosylation catalyst made it possible to increase the yields of target nucleosides in some cases (see Table 8). O OZ1 Keller and Haner 113 carried out the alkylation of the partially protected b-D-ribofuranoside (17) with tert-butyl bromoacetate in the presence of sodium hydroxide under conditions of phase- transfer catalysis. The desilylation with methanol and sulfuric acid was accompanied by transesterification of the tert-butyl ester and partial anomerisation. The glycosylation of thymine and adenine derivatives was stereoselective, the total yield was 15%. OAlk OY1 Parmentier et al.114 have succeeded in obtaining methyl 2-O- B X2O O methyl-D-ribofuranoside in four steps without the isolation of intermediate products (yield*75%).If the intermediate products had to be isolated in each step, the total yield decreased to 8%. Cook et al.78, 79 have performed the alkylation of the partially OAlk OY2 protected methyl b-D-riboside (19) (see Table 7) with ethyl bromoacetate. Subsequent glycosylation according to VorbruÈ g- gen 121 gave the target nucleoside (see Table 8). Alkylating reagent CH2N2 CH2N2, BF3 . Et2O CH2N2, BF3 . Et2O BrCH2CO2But, NaOH, BnBu3N+Cl7 MeI, NaH BrCH2CO2Et, NaH, DMF, 20 8C, 3 h MeI, NaH 1) HC:CCH2Br, NaH; 2) B10H14 BrCH2CO2Et, NaH, DMF MeI, BDDDP MeI, NaH MeI, NaH CH2=CHCH2OC(O)Cl, DMAP, Pd(OAc)2, Ph3P T S Zatsepin, E A Romanova, T S Oretskaya Ref. Yield of the alkylation product (%) 109 88 110, 111 112 74 71 113 114 115 67 a 75 74 42 116 80 55 78, 79 95 117, 118 117, 118 42 b 96 a 119 119 70 87Table 8.Glycosylation of heterocyclic bases with 2-O-alkylcarbohydrates. R1 Heterocyclic base OTMS R2 R3 N N R1 OTMS OTMS OTMS OTMS OTMS OTMS OTMS R3 R2 H OTMS H OTMS H OTMS H OTMS H OTMS H OTMS Me OTMS Me OTMS Glycosyl donor OAc BzO O OMe BzO OBz BzO O OMe BzO OBz BzO O OMe BzO 4-ButC6H4C(O)O O OAc OMe AcO BnO O OAc OMe BnO BzO O OAc O AcO B10H10 OBz BzO O OMe BzO OMe AcO O OAc AcO Reaction conditions SnCl2, (CH2Cl)2, 20 8C, 24 h SnCl4, 20 8C TMS± OTfl, MeCN, 20 8C, 0.5 h 20 8C, 16 h TMS± OTfl, MeCN, 20 8C, 2 ± 3 h TMS± OTfl, MeCN, 20 8C, 4 h TMS± OTfl, MeCN, 20 8C, 0.5 h TMS± OTfl, (CH2Cl)2, 20 8C, 9 days Yield of the b-nucleoside (%) (a : b ratio) 24 74 70 75 55 ± 70 65 (1 : 4) 71 55 37 (2 : 3) Ref.109 110 111 112 42 114 116 112 117, 118Table 8 (continued). R1 Heterocyclic base OTMS R2 R3 N N R1 OTMS OTMS OTMS OTMS OTMS OTMS OTMS R3 R2 Me OTMS Me OTMS Me OTMS CH2N3 OTMS CH2NPhth OTMS CF3 OTMS F OTMS NO2 OTMS Glycosyl donor NH O O CCl3 O TIPDS OMe O OMe BzO O OCH2CO2But BzO 2,4-Cl2C6H3CH2O O OMe OCH2CO2Et 2,4-Cl2C6H3CH2O NH O O CCl3 O TIPDS OMe O NH O O CCl3 O TIPDS OMe O OBz BzO O OMe BzO OBz BzO O OMe BzO OBz BzO O OMe BzO Reaction conditions TMS± OTfl, MeCN, 20 8C, 5 min TMS± OTfl, (CH2Cl)2, 20 8C (22) TMS± OTfl, 20 8C, 30 min BSA, TMS± OTfl, (CH2Cl)2, D TMS± OTfl, MeCN, 20 8C, 5 min (22) TMS± OTfl, MeCN, 20 8C, 5 min (22) TMS± OTfl, MeCN, 20 8C, 30 min TMS± OTfl, MeCN, 20 8C, 30 min TMS± OTfl, MeCN, 20 8C, 30 min Yield of the b-nucleoside (%) Ref.(a : b ratio) 119 94 121 93 113 79 78, 79 44 120 62 120 61 112 62 112 68 112 56Table 8 (continued). R1 Heterocyclic base OTMS R2 R3 N N R1 OTMS OTMS OTMS OTMS STMS H R2 N N N N R1 TMS H R3 R2 CN OTMS I OTMS H NHAc H NHBz Me NHBz Me OTMS 7 NHBz 7 NHBz Glycosyl donor OBz BzO O OMe BzO AcO O OAc O AcO NPhth 4-ButC6H4C(O)O O OAc OMe AcO OBz BzO O OMe BzO NH O O CCl3 O TIPDS OMe O NH O O CCl3 O TIPDS OMe O 4-ButC6H4C(O)O O OAc OMe AcO 2,4-Cl2C6H3CH2O O OMe OCH2CO2Et 2,4-Cl2C6H3CH2O Reaction conditions TMS± OTfl, MeCN, 20 8C, 30 min TMS± OTfl, (CH2Cl)2,720 to 0 8C, 3 h 20 8C, 16 h SnCl4, 20 8C TMS± OTfl, MeCN, 20 8C, 5 min (22) TMS± OTfl, MeCN, 20 8C, 5 min (22) TMS± OTfl, MeCN, D, 3 ± 5 h SnCl4 Yield of the b-nucleoside (%) (a : b ratio) 88 82 (1 : 2) 55 ± 70 70 95 71 55 ± 70 63 Ref. 112 115 42 110, 111 119 120 42 113Table 8 (continued).R1 Heterocyclic base NHC(O)CH2OPh R2 N N N N R1 TMS NHC(O)CH2OPh NHC(O)But 7 7 7 O H2N NH NH O Note.Abbreviations: TMS± OTfl, trimethylsilyl trifluoromethanesulfonate; BSA, N,O-bis(trimethylsilyl)acetamide. R3 R2 7 NH2 7 H 7 4-NO2C6H4(CH2)O Glycosyl donor 4-ButC6H4C(O)O O OAc OMe AcO NH O O (22) CCl3 O TIPDS OMe O NH O O (22) CCl3 O TIPDS OMe O NH O O (22) CCl3 O TIPDS OMe O Reaction conditions TMS± OTfl, MeCN, D, 3 ± 5 h TMS± OTfl, MeCN, 20 8C, 5 min TMS± OTfl, MeCN, 20 8C, 5 min TMS± OTfl, MeCN, 20 8C, 5 min Yield of the b-nucleoside (%) (a : b ratio) 55 ± 70 77 86 80 Ref. 42 119 119 119Synthesis of 20-O-alkylnucleosides Both anomers of 1-N-(2-O-methyl-D-arabinofuranosyl)thy- mine were prepared from the blocked carbohydrate 20 by Got- fredsen et al.117, 118 (see Table 7).It should be noted that no alkylation of the a-anomer of the monosaccharide derivative occurred in the presence of sodium hydride, whereas the yield of the alkylation product in the presence of BDDDP was only 42%. In the case of the b-anomer, the reaction with sodium hydride as a base was virtually quantitative. In 1994, Chanteloup et al.119 published a paper, which described conversion of compound 21 into the O-trichloroacet- imidate 22. The latter served as a stereoselective and a high- OOBzO HO H10B10 BzO B=Ura. j 18 4-ButC6H4COO 4-ButC6H4COO 4-ButC6H4COO 4-ButC6H4COO B=Ura, CytAc, AdeBz, GuaPheAc. (a) NaH, ( f ) Py, H2O; (g) Ac2O, Py; (h) BTMS, TMS7OTfl; (i ) MeONa, MeOH; ( j ) MeI, NaH; (k) 4-ButC6H4COCl, Py; (l ) NaOH.OO a, b O O O O O OH 18 B10H10 BzO O e, f OH OH O HO O OAc h, i O OAc B10H10 HO HO OO O c O OMe O HO O OH OMe OH O OH OMe OHO OAc OMe OAc B O l OAc OMe Br; (b) B10H14; (c) H+; (d ) BzCl; (e) NaIO4; Scheme 1 c, d O OO g OH O OH B10H10 B O O OH B10H10 23 O k OH OMe OH e, f g h B HO O OMe OH24 527 yielding donor (see Table 8). The total yield was 35%± 45%. Later, the glycosylation of compound 22 was studied in more detail.120 It is known 122 that glycosylation with b-D-ribofuranose O-trichloroacetimidates follows the SN2 mechanism with inver- sion of configuration of the C1 atom. This suggests that the synthesis of the target b-nucleoside requires the a-anomer of the O-trichloroacetimidate. However, the thermodynamically and kinetically more stable b-anomer [yields 39% (a : b=7 : 93) and 77% (a : b=1 : 99), respectively] is predominantly formed in the presence of NaH and DBU (diazabicycloundecene).The a-anomer was predominantly formed only with K2CO3 as the base (total yield 71%, a : b=93 : 7). Hence, this approach can be used for the synthesis of both a- and b-nucleosides. An elegant synthesis of 20-O-alkylnucleosides 23 and 24 (total yields 35%± 40%) from the commercially available allofuranose derivative 18 has been described (Scheme 1).51, 116 Although glycosylation is rather tedious and is seldom used, it is appropriate for the synthesis of 20-O-alkylated ribothymi- dine 78, 79, 112, 113, 118 ± 120 and large-scale syntheses of modified 20-O-alkylnucleosides.Glycosylation is the limiting step. It often proceeds non- stereoselectively and yields a mixture of a- and b-anomers. In addition, it strongly depends on the nature of the substituent at the C(20) atom and does not always give satisfactory yields of target products in the case of purine bases. VI. Alkylation of the activated cis-diol group of nucleosides in the presence of transition metal compounds In the alkylation of the cis-diol group of nucleosides with alkyl halides or diazoalkanes in the presence of transition metal com- pounds (Tables 9 ± 11), the transition metal is activated due to complex formation.In this case, the reaction may be carried out with nucleosides with non-protected 20- and 30-hydroxy groups. The main advantage of this approach is that the reaction proceeds in the absence of a base, which eliminates the necessity to protect the imino groups of the heterocyclic bases and functional groups of alkyl halides. However, the alkylation by this method yields considerable amounts of 30-isomers which are hardly separable from the target 20-O-alkylnucleosides owing to their similar chromatographic mobilities. In some cases, the isomeric mixture can be separated by crystallisation; more often, the modified nucleoside is first subjected to 50-O-dimethoxytritylation and then to chromatography. The best ratio of the 20- and 30-isomers was achieved by alkylation of nucleosides in the presence of tin chloride 123 ± 134 (see Table 9).In this reaction, it is the tin derivative A initially formed that undergoes alkylation. B XO B XO O a O b HO OH HO OH SnCl Cl A B XO O HO OCH2R (Me) (a) SnCl2; (b) RCHN2 or MeI. Yet another alkylation procedure in the presence of tin derivatives was proposed by Wagner et al.135 The starting com- pound is treated consecutively with dibutyltin dichloride and dibutyltin oxide and alkyl halide with 20,30-O-dibutylstannylene nucleosides B as intermediates. Later, this method was used for528 Table 9. Alkylation of the activated cis-diol group of nucleosides with diazomethane in the presence of SnCl2. X Nucleoside Ado Urd Cyd Guo 1a 1c B XO O H OH HO TBDPS DMTr HDMTr DMTr DMTr DMTrH R R2 1=NH2; R2=OEt H R 1 : 3 : 0.2 1=R2=NH2 DMTr DMTr Tr Note.a Alkylation with PhCHN2; b alkylation with MeI; c the yield over two steps (alkylation and amination) is indicated; dDMA, dimethylacetamide; eNPE=4-NO2C6H4(CH2)2; f alkylation with Me3SiCHN2. the synthesis of various 20-O-alkylnucleosides 135 ± 145 (see Table 10). XO B O HO OH B O N HN N N Gua Ura NH2NN N Ade O N HN N N 4-ButC6H4CONH R N R=NHBz R=NPEOC(O)NH e O NO R=H RN R=Bz N O N N N N R1 R1=H; R2=NHBz R1=NPEOC(O)NH; R2=ONPE R1=H; R2=NHTr B XO a b O O O SnBu Bu B Nucleoside : alkylating reagent : SnCl2 1 : 3 : 0.18 1 : 3 : 0.06 1 : 3 : 0.02 1 : 3 : 0.06 1 : 3 : 0.02 1 : 3 : 0.18 1 : 3 : 0.06 1 : 3 : 0.02 1 : 3 : 0.2 1 : 3 : 0.06 1 : 3 : 0.02 71 : 3 : 0.02 1 : 3 : 0.11 1 : 3 : 0.02 1 : 3 : 0.18 1 : 3 : 0.2 1 : 3 : 0.2 1 : 1.6 : 0.04 1 : 3 : 0.2 1 : 3 : 0.2 1 : 1.6 : 0.04 1 : 1.6 : 0.04 1 : 1.6 : 0.04 1 : 3 : 0.2 1 : 1.6 : 0.04 1 : 1.1 : 0.25 7 7DMF, 50 8C, 24 h DMF, 50 8C, 8 h DMA, DMSO, CH2Cl2, dioxane,710 8C, 45 min DMF, 0 8C, 18 h XO (a) Bu2SnCl2, Pri2NEt or Bu2SnO; (b) AlkHal.The intermediate tin derivative B is much more stable than the derivative A. In addition to tin salts, other metal salts were T S Zatsepin, E A Romanova, T S Oretskaya Ref. Reaction conditions Yield of the 20-isomer (%) (20 : 30 ratio) 38 (1.7 : 1) 30 (1 : 2) 27 a 30 a 74 (2 : 1) 35 (5 : 1) 34 a 67 (6.2 : 1) b 15 (1 : 1) 16 ac MeOH±DME, 20 8C MeOH±DME, 20 8C DME, 20 8C, 8 h MeOH±DME, 20 8C 40(*2 : 1) DME, 20 8C, 8 h MeOH±DME, 20 8C MeOH±DME, 20 8C DME, 20 8C, 8 h MeOH±DME, 20 8C MeOH±DME, 20 8C DME, 20 8C, 8 h EtOH ±DME, 20 8C 34 DME, 20 8C, 8 h MeOH±DME, 20 8C 123 124 125 124 125 123 124 125 126 124 125 127 125 35 21 a 40 (5 : 1) 131 21 a DME, 20 8C, 8 h 128 129 22 (1 : 2) 41 MeOH±DME, 50 8C DMF, 50 8C, 8 h 130 MeOH±DME, 20 8C, 8 h 20 (2 : 1) 131 63 (2 : 1) DMAd, DMSO, CH2Cl2, dioxane,710 8C, 45 min 129 43 DMF, 50 8C, 8 h 129 131 49 87 (14 : 1) DMF, 50 8C, 8 h DMA, DMSO, CH2Cl2, dioxane,710 8C, 45 min 131 74 (4.3 : 1) 131 65 (2.2 : 1) DMA, DMSO, CH2Cl2, dioxane,710 8C, 45 min DMA, DMSO, CH2Cl2, dioxane,710 8C, 45 min 132 133 129 131 7 51 (*1 : 1) 36 79 (4 : 1) 134 21 f B O HO OAlkSynthesis of 20-O-alkylnucleosides Table 10.Alkylation of the activated cis-diol group of nucleosides in the presence of equimolar amounts of dibutyltin dichloride or dibutyltin oxide. Nucleoside Urd Cyd NHAc N N N N* HO O OH HO O HN AcHN N NHAc N O * N O BzN O * N a The nucleoside :AllBr : Bun4 NBr ratio is 1 : 1.2 : 1.1; b PhthN= quinone : CsF ratio is 1 : 2 : 2.5. Table 11. Alkylation of the activated cis-diol group of nucleosides by trimethylsulfonium hydroxide in the presence of acetylacetonates.Nucleoside Cyd Ado Guo Urd Note. The reactions were carried out with the nucleoside :Me3S+OH7: M(acac) ratio of 1 : 2 : 0.5 in DMF (70 8C, 1 h). a The nucleoside :Me3- S+OH7: Sr(acac) ratio in DMF (75 8C, 30 min) is 1 : 1.3 : 1.3; b the nucleoside :Me3S+OH7: Ag(acac) ratio in DMF (70 8C, 2 h) is 1 : 1.3 : 1.3. Alkylating reagent MeI MeI H2C=CHCH2Br HC:CCH2Br PhthN(CH2)3Br b PhthN(CH2)3Br b ClCH2O(CH2)3Br BnCl PhthN(CH2)3Br b ClCH2O(CH2)3Br N ClCH2O(CH2)3Br * N ClCH2O(CH2)3Br 2-(BrCH2)AQ c 2-(BrCH2)AQ c 2-(BrCH2)AQ c M(acac) Yield of the 20-isomer (%) (20 : 30 ratio) Cu(acac) Mg(acac) Cu(acac) Mg(acac) Sr(acac) Cu(acac) Mg(acac) 60 (*3 : 1) 43 (10 : 1) 45 (1 : 1) 21 (*2 : 1) 75 (10 : 1) 18 (*1.2 : 1) 21 (20 : 1) Nucleoside : alkylating reagent 71 : 1.3 1 : 2 1 : 2 1 : 2 see a 1 : 27 7DMF, 130 8C (CH2Cl)2, 70 8C, 15 min DMF, 100 8C, 1 h 7 7 1 : 1.3 1 : 1.3 1 : 1.3 see d see d 1 : 2 ON O Nucleoside Ref.HO 148 149 148 149 91 a 148 149 HO ( N H *6.6 2 :1) Reaction conditions DMF, 20 8C, 18 h DMF, 20 8C, 18 h DMF, Bun4 NBr, 20 8C DMF, 120 8C, 6 h (CH2Cl)2, 70 8C, 15 min (CH2Cl)2, 70 8C, 15 min (CH2Cl)2, 70 8C DMF, CsF, 20 8C, 48 h DMF, CsF, 20 8C, 48 h DMF, 20 8C, 48 h O . d the nucleoside : 2-(bromomethyl)anthra- . cAQ= O O RN O N* O R=Me R=Me R=Bn OH O N BnN N N* 529 Ref.Yield of the 20-isomer (%) (20 : 30 ratio) 135 136 137 138, 139 80 140, 141 142 135 80 35 (2 : 1) 7 49 (1.8 : 1) *50 (*1.5 : 1) 16.8 14 38 (1.25 : 1) 31 (2 : 1) 16.8 142 51 (2.5 : 1) 142 60 (1.25 : 1) 142 45 (1 : 0.4) 143 144 145 31 (1.8 : 1) 30 (1.8 : 1) 35 (1.4 : 1) Ref. M(acac) Yield of the 20-isomer (%) (20 : 30 ratio) 148 149 149 Cu(acac) Mg(acac) Mg(acac) 18 (*1.2 : 1) 65 (20 : 1) 62 Ag(acac) 70 91 b530 tested,131, 146 but the degree of conversion in the alkylation and the ratio of the 20- and 30-isomers could not be increased. It should be noted that sometimes the alkylation of the cis-diol group in the presence of cesium fluoride 147 affords larger amounts of the 20-isomer.However, the addition of CsF in the alkylation of 20,30-O-dibutylstannylene-N3-benzoyluridine slightly increased the selectivity, but the total yield of the reaction products decreased 144, 145 (see Table 10). The methylation of nucleosides with trimethylsulfonium hydroxide in the presence of transition metal acetylacetonates (Cu, Co, Zn, etc.) (see Table 11) is less popular.148, 149 + XO XO B B Me3SOH7 O O M(acac) OMe HO OH HO acac is the acetylacetone residue, Mis the transition metal. This approach eliminates the alkylation of heterocyclic bases (except for uracil) and provides high yields in certain cases. Later, it was shown 91 that the maximum selectivity of nucleoside alkylation is achieved in the presence of silver acetylacetonates (for protected guanosine), strontium acetylacetonate (for adeno- sine) and magnesium acetylacetonate (for cytidine and protected uridine).149 Nevertheless, this procedure is hardly suitable for the introduction of other alkyl groups into nucleosides.VII. Synthesis of pyrimidine 20-O-alkylnucleosides from O2,20-anhydronucleosides In addition to direct alkylation of ribonucleosides, pyrimidine 20- O-alkylnucleosides are synthesised starting from O2,20-anhydro- cyclonucleosides. This approach is attractive because the synthesis of the anhydrocycle results in both the activation of the C(20) atom and elimination of the possibility of the reaction of theN3 atom of the heterocyclic base. The synthesis of the now commercially available O2,20-anhy- drouridine and its derivatives is performed using different meth- ods.150, 151 Most often, this involves heating of uridine and its derivatives with diphenyl carbonate in the presence of sodium hydrogen carbonate as base.The O2,20-anhydrouridine deriva- Table 12. Synthesis of pyrimidine 20-O-alkylnucleosides from O2,20-anhydronucleosides. O O R1 R1 N HN N N (R2O)nE XO XO OO O O OR2 HO HO R1 X (R2O)nE Nucleoside : (R2O)nE ratio (MeO)2Mg (EtO)2Mg (PrnO)2Mg (AllO)2Ca (MeO)2Mg (MeO)3B (MeO)3B [PhthN(CH2)3O]3B [PhthN(CH2)2O]3B 1 : 4 1 : 6 1 : 6 1 : 40 1 : 4 1 : 2 77 7 1 : 2 [PhthN(CH2)3O]3B 1:2 1 : 3 1 : 5.8 DMTr DMTr DMTr DMTr HHHHHTBDPS HDMTr [MeO(CH2)2O]3Al PhthN(CH2)2OH HHHHHHMe HMe Me HH T S Zatsepin, E A Romanova, T S Oretskaya tives 25 are easily purified by recrystallisation; the yields are *90% (Table 12).O O R R HN N O N N O XO XO (PhO)2CO, NaHCO3, DMF, 100 ± 150 8C O O 25 HO OH HO X is H or protective groups; R=H, Me. It has been shown 152 that no alkoxylation of nucleosides with magnesium alkoxides occurs in the presence of blocked 30-hydroxy groups, presumably due to the necessity of its involve- ment in the coordination of magnesium. Similar conclusions were made by Ross et al.154 for the alkoxylation with trialkyl borates. It is suggested that if magnesium alkoxides are used, the 50-hydroxy group was protected with the dimethoxytrityl group. Later,153 the alkoxylation of anhydrouridine containing a non-protected 50-hydroxy group with magnesium methoxide was carried out under similar conditions. If aluminium alkoxides and trialkyl borates are used, the alkoxylation at the C(20) atom occurs with a free 50-hydroxy group.154, 157 Blommers et al.158 did not synthesise alkoxides beforehand, but carried out a reaction with an alcohol in the presence of tetra(isopropoxy)titanium.The alkoxylation of O2,20-anhydrouridine with trimethyl borate was successfully employed by other authors.91, 159 Besides, the 20-O-methylated product was obtained by the reaction of the fluorescent uridine analogue 1-(b-D-ribofuranosyl)-2,20-anhydro- benzo[g]quinazoline-2,4(3H)-dione 26 with a twofold excess of (MeO)3B [(MeO)3CH, NaHCO3, MeOH].160 O N O N HO O HO 26 Yield (%) Ref.Reaction conditions DMF, 100 8C, 4 h DMF, 100 8C, 16 h DMF, 100 8C, 16 h DMF, 100 8C, 16 h MeOH, D, 5 h (MeO)3CH, NaHCO3 MeOH 152 152 152 152 153 154 154 154 155 77, 156 157 158 88 61 62 42 92 86 ± 97 93 18 21 50 91 32 THF, 150 8C, 24 h BH3 . THF, (CH2OH)2, NaHCO3, 150 8C, 16 h MeO(CH2)2OH, D, 48 h Ti(OPri)4, NaHCO3 (cat), THF, 140 8C, 48 hSynthesis of 20-O-alkylnucleosides VIII. Other methods for the synthesis of 20-O-alkylnucleosides A modified procedure for a two-step methylation of nucleosides involving alkylation with 1,3-benzodithiolium tetrafluoroborate and subsequent reduction of the intermediate product 27 over Raney nickel was proposed by Sekine et al.161 This excludes the necessity of protecting the heterocyclic base and the amino groups additionally introduced into the heterocycle, but the overall yield is only 35% ± 45%.O R NH S N + O O O S TIPDS O OH O R NH N O O O TIPDS S O O S 27 O R NH N O O O TIPDS OMe O R=H, CH2N(Bn)CH2CO2Et. Sproat et al.162 have succeeded in developing an efficient procedure for the synthesis of 20-O-alkylnucleosides by alkylating nucleosides with 30- and 50-hydroxy groups protected (the Mar- kiewicz protective group) with allyl ethyl carbonate in the pres- ence of a Pd(0) complex. A 2,5-dichlorobenzyl protective group was used to protect the OH group of the heterocycles.66 The degree of conversion in the allylation of compounds 11a,b and 12 163 was 90% in 30 min and no side products were formed.However, this reaction can only be used for the synthesis of allyl ethers, since it proceeds via the intermediate allylpalladium(0) complexes. Later, this approach was used by Froehler et al.164 for the preparation of 20-O-allyl derivatives of 5-(propyn-1-yl)uridine and 5-(propyn-1-yl)cytidine. Sproat et al.165 have designed novel protective groups for heterocyclic bases which made it possible to reduce the number of steps in the synthesis of the 20-O-allyl derivatives 28a ± c and to increase the overall yields of the target products, especially, in the case of the guanosine derivative 28a.B O EtOC(O)OAll, [(PhCH=CH)2CO]3Pd2 BuPh2P O TIPDS O OH 29a ± c BF¡4 Ni/Ra, EtOH 531 B O O TIPDS O OCH2CH CH2 28a ± c NBui2 O N TBDMSO N Me N N N N N B=TBDMSO N (c). (b), (a), N N N N N NH O Direct allylation of the TIPDS-protected cytidine and adeno- sine derivatives (29b,c) is also possible, whereas earlier alkylation of uridine and inosine derivatives followed by transamination had to be performed. A similar allylation procedure in which tetrakis(triphenyl- phosphine)palladium was used as a catalyst was proposed by Kumar et al.166 This procedure allowed the alkylation of nucleo- sides bearing only the 50-hydroxy protected group. With a 2.5-fold excess of allyl ethyl carbonate, the yields of the target allylation products of 50-O-dimethoxytrityluridine and 50-O-dimethoxytri- tyl-N4-benzoylcytidine were *60%, whereas in the case of the stoichiometric ratio of the reactants the reaction involved pre- dominantly the heterocyclic base (90%).Subsequent modification of the allyl derivatives leading to the introduction of the aldehyde groups was repeatedly employed in different laboratories.167 ± 170 The alkylation of compounds 30 with dimethyl sulfide in the presence of dibenzoyl peroxide (yield 55%± 70%) has been carried out.171, 172 Later,173 compounds 31 were prepared by the reaction with DMSO and acetic anhydride in the presence of acetic acid. B B O O O O a or b TIPDS TIPDS O OH O OCH2SMe 30 31 B=Ura, CytBz, AdePhOAc, GuaBui.(a) Me2S, Bz2O2, 2,6-lutidine; (b) DMSO, Ac2O, AcOH. This approach was used 51 in a two-step synthesis of the 20-O- methyluridine derivative 32 by alkylation of compound 33 with methyl phenyl sulfide and subsequent desulfurisation. Ura Ura O O O O a b TIPDS TIPDS OH O O OCH2SPh 33 Ura O O TIPDS O OMe 32 (45 ± 50%) (a) MeSPh, Bz2O2, dimethylaminopyridine; (b) Bu3SnH, Bz2O2. Matsuda et al.174 have synthesised 20-O-trifluoromethyl- adenosine 34 via the intermediate 20-methyl xanthate 35.532 Ade Ade O O O O c, d a, b TIPDS TIPDS OC(S)SMe O OH O 35 AdeBz O O TIPDS O OCF3 34 (a) BuLi, CS2,778 8C; (b) MeI, 778 8C; (c) BzCl, Py; (d) HF, Py, 1,3-dibromo-5,5-dimethylhydantoin. The yield calculated over two steps was as low as 24% due to the formation of side products and partial removal of the Markiewicz protective group under the reaction conditions.IX. Conclusion The data considered above suggest that currently the methods for the synthesis of 20-O-alkylnucleosides are well-developed, espe- cially for the simplest representatives of this type of modified nucleoside, viz., 20-O-methylnucleosides. Presumably, the main application of 20-O-alkylnucleosides consists in their incorpora- tion into oligonucleotides. The majority of modified oligonucleo- tides prepared from 20-O-alkylnucleosides meet the major requirement, viz., the ability to form stable duplexes with the complementary regions of nucleic acids.At present, the synthesis of 20-O-alkylnucleosides is performed using two ideologically different approaches. One of them entails the maximum protection of all reactive centres of the molecule and, as a consequence, involves many steps. The other, in contrast, is aimed at the reduction of steps, which inevitably results in side reactions and, correspondingly, complicates the chromatographic separation of the reaction mixture. It should be noted that the total yield of the target product in the former approach is higher, as a rule. The advantage of the latter approach consists in the reduction of the time and cost of the synthesis. Currently, the methods of nucleoside alkylation with diazo- methane and trimethylsulfonium hydroxide, as well as with dialkyl sulfates in aqueous media, have only a historical signifi- cance.The most efficient procedure for the preparation of pyrimidine 20-O-alkylnucleosides is the synthesis via O2,20-anhy- dronucleosides and treatment with alkyl iodides in the presence of silver(I) oxide.43 The alkylation of pyrimidine nucleosides with alkyl halides in the presence of transition metal derivatives can be recommended only for those cases where the reactants contain groups sensitive to strong bases, since this procedure yields a large amount of the 30-isomer. Alkylation in the presence of alkali metal hydrides is the method of choice for adenosine and its derivatives. It is of note that this method excludes the protection of 30- and 50-hydroxy groups as well as the functional groups of the heterocyclic base.The use of cyclic sulfates as alkylating reagents is especially attractive, since the alkylation product can be subjected to further nucleophilic substitution.91, 92 The alkylation of uridine and thymidine in the presence of alkali metal hydrides affords considerable amounts of the 30-iso- mers. Therefore, the 30- and 50-hydroxy groups as well as the heterocyclic base should be protected. If this approach is chosen, 20-O-alkylcytidine is prepared preferably by transamination of 20- O-alkyluridine. The optimum procedure for the synthesis of 20-O-alkylguano- sine and its derivatives proposed by Sproat et al.64 is rather costly and cost reduction is hardly possible.The alkylation of 9-(b-D- ribofuranosyl)-6-chloro-2-aminopurine and subsequent conver- sion of the reaction product into guanosine derivatives is a promising alternative to this method.91 However, the conditions of this reaction were not optimised and the possibility of its T S Zatsepin, E A Romanova, T S Oretskaya application in large-scale syntheses was not investigated. The use of transglycosylation is another promising approach;91 however, this reaction is extremely sensitive to both the structure of the aglycon and the substituent at the C(20) atom of the carbohydrate fragment and will hardly ever be used as a versatile procedure. References 1. O M Gritsenko, E S Gromova Usp. Khim. 68 267 (1999) [Russ. Chem. Rev. 68 241 (1999)] 2.B S Ermolinskii, S N Mikhailov Bioorg. Khim. 26 483 (2000) a 3. J D Smith, D B Dunn Biochim. Biophys. Acta 31 573 (1959) 4. A D Broom, R K Robins J. Am. Chem. Soc. 87 1145 (1965) 5. B E Griffin, C B Reese, G F Stephenson, D R Trentham Tetrahedron Lett. 4349 (1966) 6. Y Furukawa, K Kobayashi, Y Kanai,M Honjo Chem. Pharm. Bull. 13 1273 (1965) 7. B S Sproat, A I Lamond, in Antisense Research and Applications (Eds S T Crooke, B Lebleu) (Boca Raton, FL: CRC Press, 1993) Ch. 18 8. P D Cook, in Current Protocols in Nucleic Acids Chemistry (New York: Wiley, 2000) Ch. 4 9. P D Cook Annu. Rep. Med. Chem. 33 313 (1998) 10. P D Cook Anticancer Drug Res. 6 585 (1991) 11. M Manoharan Biochim. Biophys. Acta 1489 117 (1999) 12. P D Cook Nucleosides, Nucleotides 18 1141 (1999) 13.D D Ma, T Rede, N A Naqvi, P D Cook Annu. Rev. Biotechnol. 5 155 (2000) 14. M Majlessi, N C Nelson, M M Becker Nucleic Acids Res. 26 2224 (1998) 15. F Burlina, A Favre, J-L Fourrey Bioorg. Med. Chem. 5 1999 (1997) 16. S M L Barabino, B J Blencone, U Ryder, B S Sproat, A I Lamond Cell 63 293 (1990) 17. V J Bardwell, M Wickens, S Bienroth, W Keller, B S Sproat, A I Lamond Cell 65 125 (1991) 18. A I Lamond, B S Sproat FEBS Lett. 325 123 (1993) 19. A I Lamond Biochem. Soc. Trans. 21 1 (1993) 20. M Teplova, S T Wallace, V Tereshko, G Minasov, A M Symons, P D Cook,M Manoharan,M Egli Proc. Natl. Acad. Sci. USA 96 14 240 (1999) 21. A M StuÈ tz, J Hoeck, F Natt, B Cuenoud, M WoisetschlaÈ ger J. Biol. Chem.276 11 759 (2001) 22. N Puri, A Majumdar, B Cuenoud, F Natt, P Martin, A Boyd, P S Miller,M M Seidman J. Biol. Chem. 276 28 991 (2001) 23. K Yamana, H Zako, K Asazuma, R Iwase, H Nakano, A Murakami Angew. Chem., Int. Ed. 40 1104 (2001) 24. N A Whittemore, A M Mullenix, G B Inamati,M Manoharan, P D Cook, A A Tuinman, D C Baker, J Q Chambers Bioconjugate Chem. 10 261 (1999) 25. J Hall, D Husken, R Haner Nucleic Acids Res. 24 3522 (1996) 26. G Glick Biopolymers (Nucleic Acids Sci.) 48 83 (1998) 27. S I Antsypovich, T S Oretskaya Usp. Khim. 67 274 (1998) [Russ. Chem. Rev. 67 245 (1998)] 28. D M G Martin, C B Reese, G F Stephenson Biochemistry 7 1406 (1968) 29. J B Gin, C A Dekker Biochemistry 7 1413 (1968) 30. F Rottman, K Heinlein Biochemistry 7 2634 (1968) 31.M J Robins, S R Naik Biochemistry 10 3591 (1971) 32. T A Khwaja, R K Robins J. Am. Chem. Soc. 88 3640 (1966) 33. L M Pike,M K A Khan, F Rottman J. Org. Chem. 39 3674 (1974) 34. M J Robins, A S K Lee J. Med. Chem. 18 1070 (1975) 35. J F Codington, R J Cushley, J J Fox J. Org. Chem. 33 466 (1968) 36. K Yamauchi, T Nakagima,M Kinoshita J. Org. Chem. 43 1593 (1978) 37. K Yamauchi, T Tanabe,M Kinoshita J. Org. Chem. 44 638 (1979) 38. K Yamauchi, T Nakagima,M Kinoshita J. Chem. Soc., Perkin Trans. 1 2787 (1980) 39. G R Pettit, K Yamauchi, J J Einck Synth. Commun. 10 25 (1980) 40. K Yamauchi,M Kinoshita J. Chem. Soc., Perkin Trans. 1 762 (1978) 41. Y Hisanaga, T Tanabe, K Yamauchi,M Kinoshita Bull. Chem. Soc. Jpn. 54 1569 (1981) 42.T Kamimura, T Masegi, T Hata Chem. Lett. 965 (1982)Synthesis of 20-O-alkylnucleosides 43. C J Welsh, J Chattopadhyaya Acta Chem. Scand., Ser. B 37 147 (1983) 44. H Inoue, Y Hayase,M Asaka, S Iwai, K Miura, E Ohtsuka Nucleic Acids Res. Symp. Ser. 16 165 (1985) 45. H Inoue, Y Hayase, A Imura, M Asaka, S Iwai, K Miura, E Ohtsuka Nucleic Acids Res. 15 6131 (1987) 46. A G Ven'yaminova, Z A Kosolapova,M N Repkova Bioorg. Khim. 16 635 (1990) a 47. C J Welsh, H Bazin, J HeikillaÈ , J Chattopadhyaya Acta Chem. Scand., Ser. B 39 203 (1985) 48. T Ariyama, H Nishimoto, S Ozaki Bull. Chem. Soc. Jpn. 63 3356 (1990) 49. A Nyilas, J Chattopadhyaya Acta Chem. Scand., Ser. B 40 826 (1986) 50. M M Vaghefi, R I Hogrefe Nucleosides, Nucleotides 12 1007 (1993) 51.L Beigelman, A Karpeisky, J Matulic-Adamic, N Usman Nucleosides, Nucleotides 14 421 (1995) 52. R P Hodge, N D Sinha Tetrahedron Lett. 36 2933 (1995) 53. K Shohda, I Okamoto, T Wada,M Sekine Nucleic Acids Symp. Ser. 42 161 (1999) 54. K Yamana, Y Ohashi, K Nunota,M Kitamura, H Nakano, O Sangen, T Shimidzu Tetrahedron Lett. 32 6347 (1991) 55. K Yamana, R Aota, H Nakano Tetrahedron Lett. 36 8427 (1995) 56. J T KusÆ mierek, D Shugar Acta Biochim. Pol. 18 413 (1971) 57. J T KusÆ mierek, J Giziewicz, D Shugar Biochemistry 12 194 (1973) 58. Z Kazimierczuk, E Darzynkiewicz, D Shugar Biochemistry 15 2735 (1976) 59. E Darzynkiewicz, J T KusÆ mierek, D Shugar Biochem. Biophys. Res. Commun. 46 1734 (1972) 60. E Darzynkiewicz, J T KusÆ mierek, D Shugar Acta Biochim.Pol. 21 305 (1974) 61. J Giziewicz, D Shugar Acta Biochim. Pol. 20 73 (1973) 62. E Darzynkiewicz, Z Kazimierczuk, D Shugar Cancer Biochem. Biophys. 1 203 (1976) 63. I Tazawa, S J Tazawa, J L Alderfer, P O P Ts'O Biochemistry 11 4931 (1972) 64. M Grùtli, M Douglas, B Beijer, R Gmimil Garcia, R Eritja, B Sproat J. Chem. Soc., Perkin Trans. 1 2779 (1997) 65. B S Sproat, B Beijer, A M Iribarren Nucleic Acids Res. 18 41 (1990) 66. B S Sproat, A I Lamond, R GuÈ imil Garcia, B Beijer, U Piels, M Douglas, K Bohmann,M Carmo-Fonseca, S Weston, S O'Loughlin Nucleic Acids Res. Symp. Ser. 24 59 (1991) 67. M Grùtli, M Douglas, B Beijer, R Eritja, B Sproat Bioorg. Med. Chem. Lett. 7 425 (1997) 68.M Grùtli,M Douglas,R Eritja, B Sproat Tetrahedron 54 5899 (1998) 69. MGrùtli, B Beijer, B Sproat Tetrahedron 55 4299 (1999) 70. M E Douglas, B Beijer, B S Sproat Bioorg. Med. Chem. Lett. 4 995 (1994) 71. R Schwesinger Chimia 39 269 (1985) 72. I Velikyan, S Acharya, A Trifonova, A FoÈ ldesi, J Chattopadhyaya J. Am. Chem. Soc. 123 2893 (2001) 73. J Yano, L S Kan, P O P Ts'O Biochim. Biophys. Acta 629 178 (1980) 74. P J LMQuaedflieg, A P van der Heiden, LHKoole, A J JMCoenen, S van der Wal, E M Meijer J. Org. Chem. 56 5846 (1991) 75. E Wagner, B Oberhauser, A Holzner, H Brunar, G Issakides, G Schaffner,M Cotten,M Knollmuller, C R Noe Nucleic Acids Res. 19 5965 (1991) 76. T P Prakash,A M Kawasaki,G Vasquez, A S Fraser,M D Casper, P D Cook,M Manoharan Nucleosides, Nucleotides 18 1381 (1999) 77.T P Prakash, AMKawasaki, A S Fraser, G Vasquez,MManoharan J. Org. Chem. 67 357 (2002) 78. A M Kawasaki,M D Casper, T P Prakash, S Manalili, H Sasmor, M Manoharan, P D Cook Nucleosides, Nucleotides 18 1419 (1999) 79. A M Kawasaki,M D Casper, P T Prakash, S Manalili, H Sasmor, M Manoharan, P D Cook Tetrahedron Lett. 40 661 (1999) 80. R H Griffey, B P Monia, L L Cummins, S Freier, M J Greig, C J Guinosso, E Lesnik, S M Manalili, V Mohan, S R Owens, B R Ross, H Sasmor, E Wancewich, K Weiler, P D Weeler, P D Cook J. Med. Chem. 39 5100 (1996) 81. M Manoharan, C J Guinosso, P D Cook Tetrahedron Lett. 32 7171 (1991) 82. C J Guinosso, G D Hoke, S M Freier, J F Martin, D J Ecker, C K Mirabelli, S T Crooke, P D Cook Nucleosides, Nucleotides 10 259 (1991) 533 83.M Manoharan, L K Johnson, K L Tivel, R H Springer, P D Cook Bioorg. Med. Chem. Lett. 3 2765 (1993) 84. M Manoharan, K L Tivel, B S Ross, P D Cook Gene 149 147 (1994) 85. H M Deshmukh, A D Broom Bioconjugate Chem. 7 108 (1996) 86. J K Bashkin, J Xie, A T Daniher, U S Sampath, J L-F Kao J. Org. Chem. 61 2314 (1996) 87. C J Yu, H Wang, Y Wan, H Yowanto, J C Kim, L H Donilon, C Tao,M Strong, Y Chong J. Org. Chem. 66 2937 (2001) 88. C J Yu, Y Wan, H Yowanto, J Li, C Tao,M D James, C L Tam, G F Blackburn, T J Meade J. Am. Chem. Soc. 123 11 155 (2001) 89. C W Gundlach IV, T R Ryder, G D Glick Tetrahedron Lett. 38 4039 (1997) 90. D Werner, H Brunar, C R Noe Pharm.Acta Helv. 73 3 (1998) 91. L Beigelman, P Haeberli,D Sweedler,A Karpeisky Tetrahedron 56 1047 (2000) 92. H-U Blank, D Frahne, A Myles, W Pfleiderer Liebigs Ann. Chem. 742 34 (1970) 93. P Martin Helv. Chim. Acta 78 486 (1995) 94. K Kikugawa, F Sato, T Tsuruo, N Imura, T Ukita Chem. Pharm. Bull. 16 1110 (1968) 95. K-i Shohda, I Okamoto, T Wada, K Seio,M Sekine Bioorg. Med. Chem. Lett. 10 1795 (2000) 96. W Hutzenlaub, W Pfleiderer Chem. Ber. 106 665 (1973) 97. P von Matt, T Lochmann, R Kesselring, K-H Altmann Tetrahedron Lett. 40 1873 (1999) 98. B Cuenoud, F Casset, D HuÈ sken, F Natt, R M Wolf, K-H Altmann, P Martin, H E Moser Angew. Chem., Int. Ed. 37 1288 (1998) 99. K Seio, T Wada, K Sakamoto, S Yokoyama,M Sekine J. Org. Chem.63 1429 (1998) 100. A S Fraser, AMKawasaki,ME Jung,MManoharan Tetrahedron Lett. 41 1523 (2000) 101. T P Prakash, M Manoharan, A S Fraser, A M Kawasaki, E A Lesnik, S R Owens Tetrahedron Lett. 41 4855 (2000) 102. N Imura, T Tsuruo, T Ukita Chem. Pharm. Bull. 16 1105 (1968) 103. H-U Blank,W Pfleiderer Liebigs Ann. Chem. 742 1 (1970) 104. A A Koshkin, J Wengel J. Org. Chem. 63 2778 (1998) 105. V K Rajwanshi, R Kumar,M Kofod-Hansen, J Wengel J. Chem. Soc., Perkin Trans 1 1407 (1999) 106. H M Pfundheller,A A Koshkin,C E Olsen, J Wengel Nucleosides, Nucleotides 18 2017 (1999) 107. S Manfredini, P G Baraldi, R Bazzanini, D Simoni, J Balzarini, E De Clerq Bioorg. Med. Chem. Lett. 7 473 (1997) 108. S Manfredini, P G Baraldi, E Durini, S Vertuani, J Balzarini, E De Clerq, A Karlsson, V Buzzoni, L Thelander J.Med. Chem. 42 3243 (1999) 109. A H Haines Tetrahedron 29 2807 (1973) 110. F Dumont, R H Wightman, J C Ziegler, C Chavis, J L Imbach Tetrahedron Lett. 3291 (1979) 111. C Chavis, F Dumont, R H Wightman, J C Ziegler, J L Imbach J. Org. Chem. 47 202 (1982) 112. B S Ross, R H Springer, G Vasquez, R S Andrews, P D Cook, O L Acevedo J. Heterocycl. Chem. 31 765 (1994) 113. T H Keller, R Haner Helv. Chim. Acta 76 884 (1993) 114. G Parmentier, G Schmitt, F Dolle, B Luu Tetrahedron 50 5361 (1994) 115. M Sollogoub, B Dominguez, K R Fox, T Brown Chem. Commun. 2315 (2000) 116. W Tjaks, A K M Anisuzzaman, A H Soloway Nucleosides, Nucleotides 11 1765 (1992) 117. C H Gotfredsen, J P Jacobsen, J Wengel Tetrahedron Lett.35 6941 (1994) 118. C H Gotfredsen, J P Jacobsen, J Wengel Bioorg. Med. Chem. 4 1217 (1996) 119. L Chanteloup, N T Thuong Tetrahedron Lett. 35 877 (1994) 120. K Shohda, T Wada, MSekine Nucleosides, Nucleotides 17 2199 (1998) 121. H VorbruÈ ggen, K Krolikiewicz Angew. Chem., Int. Ed. 15 689 (1976) 122. R R Schmidt Angew. Chem., Int. Ed. 25 212 (1986) 123. M J Robins, S R Naik Biochim. Biophys. Acta 246 341 (1971) 124. M J Robins, S R Naik, A S K Lee J. Org. Chem. 39 1891 (1974) 125. L F Christensen, A D Broom J. Org. Chem. 37 3398 (1972)534 126. W T Markiewicz,MWiewiorowski Nucleic Acids Res. 2 951 (1974) 127. M J Robins,M MacCoss, A S K Lee Biochem. Biophys. Res. Commun. 70 356 (1976) 128. G Ekborg, P J Garreg J.Carbohydr. Nucleosides, Nucleotides 7 57 (1980) 129. J HeikillaÈ , S BjoÈ rkman, B OÈ berg, J Chattopadhyaya Acta Chem. Scand., Ser. B 36 715 (1982) 130. F Seela, H Debelak, H Reuter, G Kastner, I A Mikhailopulo Nucleosides, Nucleotides 17 729 (1998) 131. H Cramer,W Pfleiderer Helv. Chim. Acta 79 2114 (1996) 132. J T KusÆ mierek, D Shugar Nucleic Acids Res. Spec. Publ. 4 s73 (1978) 133. M J Robins, F Hansske, S E Bernier Can. J. Chem. 59 3360 (1981) 134. T E Leonard, P Bhan, P S Miller Nucleosides, Nucleotides 11 1201 (1992) 135. D Wagner, J P H Verheyden, J G Moffat J. Org. Chem. 39 24 (1974) 136. B S Ross,G Vasquez, S Manalili, E Lesnik,R Griffey Nucleosides, Nucleotides 16 1547 (1997) 137. V Gopalakrishnan, V Kumar, K N Ganesh Nucleosides, Nucleotides 11 1263 (1992) 138.A K M Anisuzzaman, F Alam, A H Soloway Polyhedron 9 891 (1990) 139. W Tjaks, A K M Anisuzzaman, L Liu, A H Soloway, R F Barth, D J Perkins, D M Adams J. Med. Chem. 35 1628 (1992) 140. M Manoharan, K L Tivel, L K Andrade, P D Cook Tetrahedron Lett. 36 3647 (1995) 141. M Manoharan, K L Tivel, M Zhao, K Nafisi, T L Netzel J. Phys. Chem. 99 17461 (1995) 142. X Wu, S Pitsch Helv. Chim. Acta 83 1127 (2000) 143. K Yamana, Y Nishijima, T Ikeda, T Gokota, H Ozaki, H Nakano,O Sangen, T Shimidzu Bioconjugate Chem. 1 319 (1990) 144. H M Deshmukh, S P Joglekar, A D Broom Bioconjugate Chem. 6 578 (1995) 145. D Ly, L Sanii, G B Schuster J. Am. Chem. Soc. 121 9400 (1999) 146. M J Robins Ann. N.Y. Acad. Sci. 255 104 (1975) 147. N Nagashima, M Ohno Chem. Pharm. Bull. 39 1972 (1991) 148. K Yamauchi, T Nakagima,M Kinoshita J. Org. Chem. 45 3865 (1980) 149. K Yamauchi, T Nakagima,M Kinoshita Bull. Chem. Soc. Jpn. 59 2947 (1986) 150. J G Moffat, in Nucleoside Analogues. Chemistry, Biology and Medical Applications Vol. A26 (Eds R T Walker, E De Clercq, F Eckstein) (New York; London: Plenum, 1979) P2. 92 151. T Ueda, in Chemistry of Nucleosides and Nucleotides Vol. 1 (Ed. L B Townsend) (New York, London: Plenum, 1988) p. 50 152. D P C McGee, Y S Zhai Nucleosides, Nucleotides 15 1797 (1996) 153. S K Roy, J Y Tang Org. Process Res. Dev. 4 170 (2000) 154. B S Ross, R H Springer, Z Tortorici, S Dimock Nucleosides, Nucleotides 16 1641 (1997) 155. M Manoharan, T P Prakash, I Barber-Peoc'h, B Bhat,G Vasquez, B S Ross, P D Cook J. Org. Chem. 64 6468 (1999) 156. T P Prakash, M Manoharan, A M Kawasaki, E A Lesnik, S R Owens, G Vasquez Org. Lett. 2 3995 (2000) 157. U Legorburu, C B Reese, Q Song Tetrahedron 55 5635 (1999) 158. M J J Blommers, F Natt, W Jahnke, B Cuenoud Biochemistry 37 17714 (1998) 159. A Kittaka, T Kuze, T Asakura, K Ito, T Miyasaka, J Inoue Bioorg. Med. Chem. Lett. 8 3207 (1998) 160. F Godde, J-J Toulme, S Moreau Nucleic Acids Res. 28 2977 (2000) 161. M Sekine, L S Peshakova, T Hata, S Yokoyama, T Miyazawa 162. B S Sproat, A M Iribarren, R G Garcia, B Beijer Nucleic Acids 163. B S Sproat, A M Iribarren, B Beijer, U Piels, A I Lamond 164. B C Froehler,R J Jones,X Cao, T J Terhorst Tetrahedron Lett. 34 165. B Beijer, M Grotli, M E Douglas, B S Sproat Nucleosides, 166. V Kumar, V Gopalakrishnan, K N Ganesh Bull. Chem. Soc. Jpn. 167. J T Goodwin, S E Osborne, E J Scholle, G D Glick J. Am. Chem. J. Org. Chem. 52 5060 (1987) Res. 19 733 (1991) Nucleosides, Nucleotides 10 25 (1991) 1003 (1993) Nucleotides 13 1905 (1994) 65 1665 (1992) Soc. 118 5207 (1996) T S Zatsepin, E A Romanova, T S Oretskaya 168. E J Maglott, G D Glick Nucleic Acids Res. 26 1301 (1998) 169. A V Kachalova, T S Zatsepin, E A Romanova, D A Stetsenko, M J Gait, T S Oretskaya Nucleosides, Nucleotides, Nucleic Acids 19 1693 (2000) 170. T S Zatsepin, A V Kachaloova, E A Romanova, D A Stetsenko, M J Gait, T S Oretskaya Bioorg. Khim. 27 45 (2001) a 171. S Zavgorodny,M Polianski, E Besidsky, V Kriukov, A Sanin, M Pokrovskaya, G Gurskaya, H LoÈ nnberg, A Azhayev Tetrahedron Lett. 32 7593 (1991) 172. A Karpeisky, C Gonzalez, A B Burgin, N Usman, L Beigelman, Nucleosides, Nucleotides 16 955 (1997) 173. A E Pechenov, S G Zavgorodnii, V I Shvets, A I Miroshnikov Bioorg. Khim. 26 362 (2000) a 174. N Nishizono,Y Sumita,Y Ueno,A Matsuda Nucleic Acids Res. 26 5067 (1998) a�Russ. J. Bioorg. Chem. (Engl. Transl
ISSN:0036-021X
出版商:RSC
年代:2002
数据来源: RSC
|
|