摘要:

Russian Chemical Reviews 69 (3) 187 ± 200 (2000) Sorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials S B Savvin, V P Dedkova, O P Shvoeva Contents I. Introduction II. Sorption-spectroscopic methods III. Instruments IV. Test methods V. Conclusion Abstract. the for methods test and sorption-spectroscopic on Data Data on sorption-spectroscopic and test methods for the determination of solid-phase the on ions metal of determination of metal ions on the solid-phase of ion-exchange ion-exchange materials reviewed. are decade past the over published materials published over the past decade are reviewed. The The advantages materials ion-exchange of disadvantages and advantages and disadvantages of ion-exchange materials are are discussed.these of selectivity and limits detection The discussed. The detection limits and selectivity of these techniques techniques are described. The bibliography includes 151 references are described. The bibliography includes 151 references. I. Introduction In recent years, well-known organic reagents have been employed in sorption-spectroscopic and test methods more extensively than in conventional photometry. These techniques, with the use of the same reagent, allow simultaneous gain in sensitivity and selectivity and considerable reduction of the assay time. Remarkable prog- ress in the analytical chemistry of organic reagents has become possible, in particular, owing to the advent of a new generation of instruments for measuring absorption and reflection of solid materials and organic reagents and complexes immobilised on these solids, i.e., the analytical forms of the elements subject to determination, and of compact mini-pumps with controllable flow rates and novel materials as carriers.All these factors have stimulated the development of a great variety of sorption-spectro- scopic techniques for the determination of inorganic compounds using organic reagents. Various materials, e.g., paper, foamed polyurethanes, silica, cellulose, poly(vinyl chloride) and caprone membranes, fabrics, ion-exchange resins, etc., are used as a solid phase in sorption spectroscopy. This review considers the published data on the use of ion-exchange materials as a solid phase.II. Sorption-spectroscopic methods Ion-exchange materials have long been in use as sorbents for the concentration and separation of elements.1± 4 The history of selective sorption of metal ions on natural and synthetic materials has been described in a monograph.1 S B Savvin, V P Dedkova, O P Shvoeva V I Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences, ul. Kosygina 19, 117975 Moscow, Russian Federation. Fax (7-095) 938 20 54. Tel. (7-095) 137 28 78 (S B Savvin), (7-095) 939 70 73 (V P Dedkova, O P Shvoeva) Received 11 November 1999 Uspekhi Khimii 69 (3) 203 ± 217 (2000); translated by R L Birnova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n03ABEH000538 187 187 193 193 197 Strongly acidic and strongly basic ion exchangers possess low selectivities.The main procedure for the enhancement of the selectivity of ion exchange materials is the introduction of com- plexing groups. Complexing sorbents can be synthesised by introducing covalently bound functional groups into the poly- mer 1±4 or by modification of ion-exchange and non-ionic resins with organic reagents based on ion exchange or adsorption.5, 6 The sorbents with pendent chelating groups are well-known and have found wide application, though they have a number of drawbacks, such as difficulties of their synthesis, low reversibility of sorption ± desorption processes, unsatisfactory kinetic charac- teristics, etc.On the other hand, these sorbents are efficient for the concentration and separation of elements with subsequent spec- troscopic determination. Modified ion-exchange resins prepared from conventional ion-exchange resins by sorption of organic reagents have found extensive use in recent years. Some characteristics of complexing sorbents for the concentration and separation of metal ions have been reviewed.5 This review considers various types of ion- exchange resins and modifying reagents, their ligand capacity, methods for the determination of metal ions and areas of application of complexing sorbents. It was shown that the retention of a modifying reagent on an ion-exchange resin depends on the type of the modifying reagent, pH of the aqueous solution of the reagent, the nature of the ion-exchange resin and of the exchangeable anion, the structure of resin grains and the stability of the sorbent in different media.The conditions for desorption of the modifying reagent are also described. Other parameters determining the retention and selectivity of metal ion separation on complexing sorbents include the stability of metal complexes with organic ligands and pH of the test solution; the mechanism of the metal ion sorption on a complexing sorbent is similar to that of complexation in aqueous media. The methods used for the preparation of modified ion- exchangers, their properties and applications are described in a review.6 Such ion-exchangers are generally used for the selective and group concentration of elements, but their most valuable property is that they allow simultaneous concentration and determination of the corresponding elements.This has been employed in the most advanced directions of analytical chemistry, viz., sorption-spectroscopic and test methods. In the choice of optimum conditions for sorption-spectro- scopic methods, one should take into account the fact that the behaviour of a complex compound in solution and on a solid phase may be different. The sensitivity and selectivity of the determination depend on the composition of the complex formed, on conditions of sorption and analysis, on the mode of immobi-188 lisation of the reagent on the carrier, on kinetic features of the process and on many other factors.7 ± 10 Depending on the method used to detect the analytical signal, sorption-spectroscopic methods for the determination of elements with the use of modified ion-exchange resins (solid-phase spec- trometry) are subdivided into sorption-luminescent, sorption- photometric and diffuse reflectance spectroscopy.Sorption-luminescent methods are based on the measurement of parameters of the generation or quenching of luminescence in reactions occurring on a solid phase. These methods are not considered in this review. Sorption-photometric methods (solid-phase photometry) entail the measurement of optical densities of concentrated sub- stances in cells relative to a swollen standard ion-exchange resin having the same degree of granulation, in cuvettes, on neutral glasses or perforated discs, etc., after a complex of the compound to be determined with the reagent has been formed on a solid phase.The methods based on the concentration of the ion to be determined either as such or as its complex on an original or a reagent-modified matrix with subsequent desorption of the com- plex compound and measurement of the optical density of the eluate, also refer to sorption-photometric methods. Diffuse reflectance spectroscopy consists in the measurement of changes in the diffuse reflection of the solid phase containing a complex compound relative to an original or a reagent-modified solid phase. The latter two groups of sorption-spectroscopic techniques are closely interrelated and employ virtually identical terminology.The term `sorption-photometric methods' is often applied to describe the methods involving measurement of the magnitude of diffuse reflection rather than optical density. The absorption spectra of complexes in solution and the spectra recorded on a solid phase by using sorption-photometric methods are usually similar, which considerably simplifies com- parison of their behaviour. The prospects for solid-phase spec- trophotometry are considered in a review 9 which describes its main advantages, such as the possibility of simultaneous concen- tration and preparation of analytical forms of concentrated substances which are best suited for further analysis, enhanced stabilities of compounds in the sorbent phase in comparison with those in solutions, increased selectivity of action of chromogenic reagents due to greater rigidity of the molecule fixed on a polymeric matrix.Other advantages include the possibility of using both water-insoluble organic reagents immobilised on ion- exchange resins and reactions yielding poorly soluble compounds or colloidal systems, fast reaction kinetics, high degree of extrac- tion and high variability of the solution volume to the adsorbent mass ratio. The reproducibility of experimental results depends on the procedures used for measurement of the analytical signal and sampling. 1. Solid-phase photometry Solid-phase photometry (ion-exchange photometry or ion- exchange resin absorptiometry) was originally proposed for detecting trace quantities of elements.11, 12 The ions to be deter- mined can be immobilised on a solid phase in three different ways, viz., (i) by sorption of the final complex, (ii) by sorption on an untreated ion-exchange resin followed by the formation of a complex on a solid phase and (iii) by sorption on an ion-exchange resin pretreated with a reagent.A design of a photometric cell suitable for a transparent ion-exchange resin for measuring optical density at appropriate wavelengths has been proposed. This method is characterised by high sensitivity and permits for measurement of microquantities of ions in complex mixtures. Thus the sensitivity of solid-phase photometry in the determina- tion of bismuth exceeds that of conventional spectrophotometric procedures 320-fold. Solid-phase photometry was recommended for use in the analysis of natural waters without preliminary concentration of the elements to be determined.11, 12 By virtue of higher sensitivity and selectivity which may be achieved by proper choice of a solid phase and a reagent, solid- S B Savvin, V P Dedkova, O P Shvoeva phase photometry is advantageous over spectrophotometry in the determination of nitrite, fluoride and phosphate ions as well as of metal ions, such as zirconium, titanium, copper and uranium.13 In the analysis of geological materials, niobium is adsorbed as a complex with sulfochlorophenol C on the anion-exchange resin AB-17,14 uranium(VI) is adsorbed on AB-17 modified with 2-(2- thiazolylazo)-5-diethylaminophenol.15 The detection limit for uranium and niobium oxide is 0.1 mg ml71.Fe(II), Cu(II), Ni(II), Co(II) and Ti(IV) ions interfere with the determination of ura- nium. This method is time-consuming, since the sorption and masking of the interfering components take 30 min each.15 Vanadium(IV) and vanadium(V) are determined using Erio- chromecyanine R immobilised on an anion-exchanger Sephadex DEAE-25. Vanadium(IV) is sorbed first and then vanadium(V) following its reduction with ascorbic acid. Light absorption of the solid phase is measured at 563 and 585 nm in a 1-mm cell; the detection limits of four- and five-valent vanadium is 1.6 and 1.4 ng ml71, respectively. This method is used for the analysis of natural waters.16 Modification of the anion-exchanger Sephadex QAEA-2 with 5-bromosalicylhydroxamic acid resulted in a sorb- ent in which vanadium(V) forms a complex with an absorption maximum at 560 nm.A procedure has been developed for the determination of vanadium(V) with detection limits of 2.2, 1.3 and 0.5 mg litre71 in samples with volumes of 100, 250 and 1000 ml and with shaking time 20, 30 and 50 min, respectively. This method was used for monitoring crude oil and drinking water for vanadium.17 The use of the first derivative of the absorption spectrum in solid-phase determinations increases the sensitivity of the method and diminishes the background effect. The detection limit in the determination of molybdenum with pyrocatechol violet on a solid phase in the analysis of water, soil and plant tissues is from 2 to 8 ng ml71.The complex (1 : 1) is adsorbed on an anion-exchanger of the dextran type.Acell with an optical path of 1 mmis used and absorption is monitored at 716 nm.18 Phenylfluorone is used as a complexing reagent for the determination of molybdenum(VI) in sea water; the first derivative of the absorption spectrum of the complex on a nitrocellulose membrane filter is recorded at 536 nm. The concentration of molybdenum in sea water near the Japanese coast was found to be equal to 10.80.4 mg ml71 (see Ref. 19). The method of the first derivative in solid phase spectropho- tometry was used for the determination of zirconium on the anion- exchange resin AB-17 modified with Eriochrome Black T,20 and for the determination of lead as an iodide complex on the cation- exchange resin KU-2 modified with Basic Blue K.21, 22 The detection limits for zirconium and lead are 0.01 and 0.09 mg ml71, respectively.Flow-injection analysis makes use of reagents immobilised on a solid phase. For example, in the determination of vanadium(V) in natural water, vanadium is first separated from accompanying components by column chromatography on Sephadex G-25, and then is made to react with 2-[2-(3,5-dibromopyridyl)azo]-5-dime- thylaminobenzoic acid immobilised on the anion-exchange resin AG 1X2 in a flow of a formate buffer. The optical density of the complex formed is measured at 650 nm.The calibration curve is linear within the range of vanadium(V) concentrations from 2 to 10 mg litre71; the productivity of analysis is 5 to 6 samples per hour.23 Flow-injection systems can be used to determine nickel and zinc by solid-phase spectrophotometry owing to different sorption rates. Nickel and zinc complexes with 1-(2-thiazolylazo)-2-naph- thol are adsorbed on silica gel C18. Both complexes are adsorbed at a flow rate of 0.6 ml min71, but only the zinc complex is adsorbed at a flow rate of 1.5 ml min71. Analysis can be carried out at zinc and nickel concentrations from 0.1 to 2 mg litre71. The productivity of the method is 20 samples per hour. The procedure for the determination of zinc and nickel with the help of a flow-injection system was used in the analysis of copper alloys.24Sorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials 2.Diffuse reflectance spectroscopy It can be noted that there are much fewer studies in which changes in optical density are measured as an analytical signal than studies making use of diffuse reflection coefficients. The advantages of immobilisation of analytical reagents on the surface of carriers over photometry of solutions were discussed in a review.10 It was shown that the combination of sorption concentration with measurements of surface analytical signals makes it possible to reduce the detection limit of components that are detected by several orders of magnitude, while the use of adsorbents modified with specific organic reagents increases the selectivity of analysis. The same paper provides a classification of modified sorbents according to the type of the carrier and the modifying reagent and describes various procedures for the immobilisation of analytical reagents as well as different applica- tions of the modified sorbents for analytical purposes.In partic- ular, immobilisation of organic reagents from polar and non- polar solvents and their mixtures on different carriers under both static and dynamic conditions and after treatment of sorbents with high-frequency low-temperature plasma in the presence of the vapour of organic compounds is considered. Modification of sorbents with inorganic reagents is achieved by sorption or precipitation of metal hydroxides, oxides or salts on the surface.Composite materials based on modified sorbents are obtained in the casting of polyacrylonitrile fibres by filling them with finely dispersed sorbents. The problem of what is to be called modified organic reagents, modified sorbents and immobilised reagents was raised in a study 25 where the terms `modification' and `immobilisation' referred to the analytical reagent rather than to the carrier. In their attempt to provide a more exact definition, the authors suggested that organic reagents were considered as modified if a second (or third) modifying (monomeric or polymeric) compo- nent reacts with an organic reagent and changes its properties, but no new chemical compound is formed. Upon immobilisation on a solid matrix, the reagent is modified with a polymeric ion- exchange resin.However, the majority of investigators use the following terminology: a reagent is immobilised on a carrier (solid phase), while the carrier (solid phase) is modified with the organic reagent. The methodology of quantitative measurements in diffuse reflectance spectroscopy is described in the review.26 This study deals with a description of the main factors which affect the accuracy and reproducibility of results of sorption-spectroscopic determination, viz., the particle size, moisture content and magni- tude of the mirror reflection superimposed on the diffuse reflec- tion resulting in the distortion of the spectrum.An attempt to relate the analytical characteristics of reactions occurring in solution to those occurring on a solid phase were undertaken in order to predict the possibility of conducting photometric reac- tions on a solid phase. However, the sensitivity of the sorption- photometric procedure from the optical characteristics of the reagent and the product of the photometric reaction in solution cannot be established a priori. Nevertheless, the authors presume that the known values of molar absorption coefficients of complex compounds in solution and the known sensitivity of the sorption- photometric method can be used to predict, with a lesser than 20% error the values of the diffuse reflection coefficients for other compounds with known molar absorption coefficients for the same matrix.26 A study of formation of mixed-ligand complexes of Zn(II), Cd(II), Pb(II) and Ag(I) on silica modified with thiourea deriva- tives revealed that the rate of formation of such surface complexes depends on the ratio of stability constants of metal ± ligand complexes in solutions.27 Modified silica is used for the determination of osmium(VIII) and palladium(II) in wet dedusting sludge and washing acids 28, 29 and for the determination of gold.30 Gold is determined as the metal after thermal treatment of the sorbent and osmium is determined as the oxide.Analytical potentials of silica and foamed 189 polyurethanes containing N-propyl-N0-[1-(benzothiazolyl-2- thio)-2,2,2-trichloroethyl]ureido or 4-adamantyl-2-(2-hydroxy-1- naphthylazo)thiazolyl (AOT) groups were examined for possible determination of Pt(IV), Pd(II), Au(III), Ag(I), Cd(II) and Hg(II).It was shown that Hg(II) (pH 6.0 ± 8.5) and Cd(II) (pH 9.5 ± 11.0) are quantitatively retained on the AOT-containing sorbent.31 It was demonstrated 32 that many regularities of formation of metal ion complexes with organic ligands on the surface of silica gel and in solution are identical for Fe(III), Cu(II), Co(II), Y(III) and La(III). In the case of easily hydrolysable ions, such as Zr(IV), Ti(IV), V(V), W(VI), Mo(VI) and kinetically inert platinum-group complexes, the irregularities are manifested in increased rates of complexation, enhanced complexing ability and formation of mixed-ligand complexes.The sorption of dithizone (DT), 4,40-bis(dimethylamino)thio- benzophenone, 4-(thiazol-2-ylazo)naphthol (TAN), 1,10-phenan- throline and high-molecular-weight quaternary ammonium salts (QAS) on silica gel (SG) and cellulose (C) was studied aimed at preparing solid-phase reagents for Hg(II), Ag(I), Pb(II), Zn(II) and Fe(III).33 ± 35 It was shown that such sorbents are stable in aqueous solutions in a broad range of pH. The optimum conditions for the formation of Cu(II), Co(II) and Zn(II) complexes with TAN± SG, of Ag(I), Hg(II) and Pb(II) complexes with DT± SG, of Co(II) and Fe(III) thiocyanate and Bi(III) tetraiodide complexes with QAS±SG and QAS±C were established. To this end, a variety of sorption-spectroscopic and test methods as well as reactive indicator papers have been devised for express control of these metal ions in natural, drinking and waste waters and biological fluids.The detection limit for these metal ions lies in the range of 0.2 to 10 mg litre71 at a sample volume of 50 ± 100 ml. Optimum conditions for the sorption of Cd(II), Hg(II) and Pb(II) ions on Silochrom modified with 4-(thiazol-2-ylazo)resor- cinol were found. These ions were sorbed from an acetate buffer pH 6.8 ± 7.5 (15 ml) on the sorbent (0.3 g) upon shaking for 5 ± 10 min. The concentration coefficient for Cd(II) and Pb(II) was 85, that for Hg(II) was 300. This procedure was used for the analysis of soils.36 The cobalt complex with 1-(2-pyridylazo)-2-naphthol (PAN) is adsorbed on an unmodified silica gel and is firmly retained on the sorbent after acidification with up to 0.5 M HCl.This proce- dure allows selective determination of 2 to 10 mg of cobalt in drinking and mineral waters. Ballast components are desorbed upon subsequent acidification. The sorption is performed from 100- and 300-ml sample volumes under static conditions; the mass of the silica gel is 0.2 g.37 1-(2-Pyridylazo)-2-naphthol immobilised on Silochrom C-120 is used for the determination of indium. The concentration coefficient for indium adsorbed from a 150-ml sample is 500.38, 39 Immobilisation of aluminone on silica gels containing differ- ent groups, with different particle diameters and pore sizes was studied as a sensitive layer of an optical sensor for copper.40 Immobilisation produced a favourable effect manifested as a longwave shift of the absorption maximum, a decrease in pH for the complexation and an increase in sensitivity and selectivity. A procedure for the detection of copper in foodstuffs with a detection limit of 0.015 mg ml71 was developed.Diffuse reflectance spectroscopy was used to study the com- pleteness of sorption of Fe(II) and Fe(III) ions as complexes with cathechol-3,5-disulfonic acid on the anion-exchange resin AB-17.41 This study demonstrated the possibility of separate determination of Fe(II) and Fe(III) ions; a test method was proposed for the determination of iron with a detection limit of 0.01 mg litre71.Fe(III) ions are determined at pH 3.5, while a sum of Fe(II) and Fe(III) ions, at pH 6 ± 9. The measurements are performed at two wavelengths (500 and 540 nm); the concentra- tions of Fe ions are calculated from the following equations: Fe(III): F500=(5.70.05) c+(0.830.04), F540=(4.560.05) c+(0.660.04),190 Fe(II): F500=(4.590.28) c+(0.770.33), F540=(3.70.06) c+(0.040.03), where F is the Gourevich ± Kubelka ± Munk function and c is the iron concentration (mg ml71). Iron ions are determined at con- centrations from 0.03 to 6.4 mg ml71. Colorimetric characteristics of coloured sorbates, viz., lumi- nosity (L), colour coordinates (A, B), saturation (S), brightness (Y), colour tone (T), shading (E), etc., can be used for detecting analytical signals in diffuse reflectance spectroscopy in addition to the diffuse reflection coefficient (R) or the Gourevich ± Kubel- ka ± Munk function (F).The use of these colour characteristics as analytical signals makes it possible to determine dissociation constants of organic reagents and increases the sensitivity and accuracy of metal ion determinations. The use of DL and DY instead of DF (where D is the difference between the values of the corresponding coefficients for the complex and the reagent) decreases 4- to 20-fold the detection limit for cobalt, indium and nickel with immobilised reagents. According to sensitivity of determination, these values can be arranged in the following order: DF<DA<DS<DL<DY.42 ± 49 The Gourevich ± Kubelka ± Munk function, which allows the establishment of a correlation between analytical signals in the same photometric reaction occurring both in a solution and on a solid phase, is the most frequently used as the analytical signal.The simplest way to express the analytical signal is to present it as a difference in diffuse reflection coefficients of carriers after the passage of control and test solutions at the optimum wavelength. The intensities of analytical signals depend on the efficiency of sorption of metal ions from solutions and the completeness of conversion of sorbed metals into coloured complex compounds on a solid-phase. The signals, in turn, depend on the acidity of test solutions (Fig.1 a), the mass of sorbent discs (Fig. 1 b) and the composition of the reagent solution. a b DR DR 3 1 0.4 1 0.3 2 2 0.2 0.2 3 0.1 m /mg 5 pH 3 1 40 20 Figure 1. Dependence of the analytical signal of metal complexes on the solid phase of a fibrous sorbent filled with the cation-exchange resin KU-2 on the acidities of solutions (a) and the mass of the sorbent disc (b) in the determination of iron (1), nickel (2) and copper (3) with potassium thiocyanate, dimethylglyoxime and sodium diethyldithiocarbamate, respectively; (a): cM=0.05 (Fe), 0.1 (Ni), 0.1 mg litre71 (Cu); (b): cM= 0.05 mg litre71. 3. Chemical optical sensors Sorption-spectroscopic methods and immobilised organic reagents were used in the design of the so-called chemical optical sensors or sensitive elements.50 ± 85 There are several different interpretations of the term `sensor'.Thus a chemical sensor is defined as a probe which provides direct information about the chemical composition of the environment. It consists of a physical transducer of analytical signals and a chemically selective layer.52 Chemical sensors are the main elements of a new generation of analytical devices, such as sensor analysers, which, together with a sensitive element, comprise sample injectors and various devices for the processing of signals and displaying the results, viz., the S B Savvin, V P Dedkova, O P Shvoeva concentration of the tested substance.53, 54 In other studies,55, 56 a sensor is referred to as a primary device which selectively responds to changes in definite properties of the environment by converting this information into the corresponding electrical signals, e.g., optical signals.We define a sensor as a sensitive element of a sensor analyser. Sensors are supposed to be recoverable systems, i.e., they must easily be reset after the assay is completed. Therefore, the material used as a solid phase should first of all be mechanically and chemically stable. Fibrous materials filled with ion-exchange resins largely meet these requirements. These materials are cast from suspensions of finely dispersed ion-exchange resins in a solution of polyacrylonitrile in dimethyl- formamide. Pressing of suspensions through spinnerets produces thin porous polyacrylonitrile fibres filled with ion-exchanger particles.The diameter of a polyacrylonitrile fibre is 30 to 40 mm, the size of carrier particles is 5 to 10 mm, the degree of filling is 50 mass%± 75 mass %. By virtue of their fibrous struc- ture and high degree of dispersion of filling agents, such materials manifest remarkable kinetic properties. Carriers represent white or light-yellow 0.1 ± 0.2 mm thick fibrous plates having a density of 40 to 80 g m72 which are stable in strongly acidic, neutral and weakly alkaline media. These plates are used to manufacture discs. Depending on their diameter, 10 or 20 mm, thickness and degree of filling, the mass of discs varies from 10 to 50 mg. Prior to use, carriers (in the form of plates or prefabricated discs) are washed with 2 M HCl up to a negative reaction for Fe(III) with thiocyanate ions and then with water up to a negative reaction for chloride ions with silver nitrate and dried in air.Such carriers can be stored at room temperature for an indefinitely long period of time. Prior to use, plates are shaped into discs and swollen in water for at least 24 hours. Immobilisation of analytical reagents is performed in a static or a dynamic regime. In the former case, the discs are kept in aqueous or aqueous ± organic solutions of reagents (107571076 M) for 5 ± 15 min and then washed with water. The concentration of the immobilised reagent varies from 1076 to 1074 mol litre71 g71. In the latter case, the reagent solution is dripped or pumped through the carrier disc placed in a cell with a sintered-glass filter using a peristaltic pump.Filled fibrous sorbents can be used in both modified (i.e., after immobilisation of the organic reagent) and in the original (colour- less) form. Ions to be determined are concentrated in different ways, viz., directly from the test solution with subsequent detection on a solid phase, by concentration of a complex compound formed in the solution and by sorption of the ion in question on a carrier with an immobilised reagent. Apparently, sorption of a preformed complex is the least appropriate solution of the problem of increasing the sensitivity and selectivity of the method. The advantages of discs with immobilised reagents are as follows: full readiness of the so-called sensitive element for the assay, simplicity and lack of necessity to prepare reagent solutions.Of their disadvantages one may relate limited durability and storage life; nor in all cases can they be recovered. The obvious merits of colourless discs are that they are easily recovered and that the same discs can be used for the determi- nation of different elements. Another advantage is that the storage life of carriers is not limited. Their demerits are as follows: the necessity to prepare reagent solutions and the need for an addi- tional procedure, the so-called `development', viz., sorption of a reagent on a carrier after concentration of the ion in question. On the whole, the main attraction of fibres filled with ion- exchange resins is the possibility to vary the degree of their filling and the nature of the ion exchanger and the reagent depending on a concrete selectivity problem.In contrast with other carriers, these materials represent both the sites for colour reactions and the means for concentration of ions to be tested and removal ofSorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials ballast components. They are resistant to mechanical actions and manifest high sorptive and kinetic properties, which ensures a high level of sensitivity, which can be attained by passing any volume of test solutions. The use of fibrous materials ensures easy sorption of reagents and tested ions (e.g., by immersion of the carrier into the solution or by passing the solution through the carrier) and easy separation of the solid phase from the solution (e.g., by decanting under static conditions or by pumping under dynamic conditions).This, in turn, allows easy automation of flow sorp- tion-spectroscopic procedures and, if necessary, test operation without any devices based on visual detection. Studies of fibres filled with ion-exchange resins used as sorbents follow two main lines which include sorption-spectro- scopic determination of metal ions based on changes in their diffuse reflection upon complexation with organic reagents on a solid phase (carrier) and elaboration of test methods. These studies were undertaken with the view of constructing recoverable systems which utilise exclusively fibrous sorbents with immobilised reagents as optical sensors.However, despite its indisputable advantages, this approach significantly restricts the potentials of a new carrier. It was found that organic reagents immobilised on carriers are not always effective in analytical reactions or they change their colour upon immobilisation. Regeneration of optical sensors with immobilised reagents is a complicated task. On repeated regeneration, organic reagents are either destroyed or washed off from the carrier with a regenerating solution. Nevertheless, in some cases sensitive elements can be recovered. Thus in the determination of lead with Xylenol Orange 51 and in the determination of uranium with Arsenazo I,64 the same sensitive element (the optical sensor) can be used for up to 20 assays.Most frequently, optical sensors cannot be regen- erated at all, e.g., a sensitive element for thorium with Xylenol Orange.51 It is economically more advantageous to replace a sensitive element used as a chemical optical sensor than to regenerate it. Two types of sensitive elements or chemical optical sensors are presently being developed, viz., fibrous materials filled with ion- exchange resins with and without organic reagents immobilised on them. In the latter case, the analytical reaction with an organic reagent is accomplished after sorption of the tested ion. In the choice of a particular organic reagent, preference should be given to the most well-studied and accessible reagents possess- ing good analytical characteristics. Organic reagents containing acid groups are immobilised on materials filled with anion- exchange resins; while those containing basic groups are sorbed on cation-exchange resins.In a few studies, sorption was per- formed on ion-exchange fabrics with phosphate or vinylpyridi- nium groups. Fixation of reagents on solid carriers occurs mainly due to the ion exchange mechanism and partially due to adsorption, for- mation of hydrogen bonds and other interactions. Electron and EPR spectroscopy was used to study the proper- ties of 1-(2-pyridylazo)-2-naphthol (PAN) immobilised on poly- acrylonitrile fibres filled with finely dispersed ion-exchange resins and its complexes with copper(II).66 In this study, fibres filled with different ion-exchange resins, viz., KU-2, KB-4, AB-17 and A-5, were employed.The parameters of EPR spectra of copper com- plexes on a solid phase in the pH range of 1 ± 8 were calculated. It was shown that immobilisation of PAN on a solid phase changes its state (viz., the spread of individual ionic forms of the reagent is shifted and widened) and spectroscopic properties. If the concen- tration of the reagent exceeds 1075 mol litre71 g71, this under- goes association on the surface of the carrier. The main contribution to the retention of PAN on these ion-exchange resins is due to ionogenic groups of the carrier; however, under certain conditions, adsorption begins to play a leading role, as is the case with the adsorption of PAN on KU-2 in alkaline media.The role of the nature of the carrier in complexation is rather complicated and ambiguous and is manifested primarily in its competing interaction with metal ions. By using different carriers, 191 one can vary the concentration of the immobilised reagent on the surface and thus control its reactivity. The second type of chemical optical sensors proved to be more advantageous than first type sensors. Their additional advantages are as follows. Since sorption and determination are separated by certain time interval, it is possible to introduce masking complex- ing agents in the concentration and determination steps, which considerably increases the selectivity of the reactions. Also, this particular type of sensor makes use of the original, colourless carriers.This circumstance has stimulated the development of new spectroscopic solid-phase methods (e.g., determination of cations on a solid phase), which consists, for example, in the sorption of metal cations in the form of colourless or weakly coloured anionic complexes on anion-exchange resins and their subsequent detec- tion in the most contrasting colour reaction. Thus the sensitivity of determination of Fe(III) with thiocyanate on the anion-exchange resin AB-17 increases considerably if iron is sorbed in the form of complexes with tartaric or glycolic acids (Fig. 2).67, 68 Compara- tive data on iron determination with thiocyanate following sorption from solutions of glycolic acid on KU-2- or AB-17- containing fibres testify to a significant increase in selectivity.The selectivity factors are equal to 10 and 20 for Zn, 20 and 50 for Cd, 5 and 200 for the uranyl ion, 1 and 10 for Bi and 100 and 3000 for oxalates and phosphates, respectively. 1 DR 2 0.4 0.3 3 0.2 0.1 600 400 500 l /nm Figure 2. Reflection spectra of the iron thiocyanate complex formed upon sorption of Fe(III) from 0.01 M solutions of glycolic (1) and tartaric (2) acids and without addition of anions (3) on a fibrous material filled with the anion-exchange resin AB-17; cFe=0.2 mg ml71. The new methodology has made possible solid-phase deter- mination of thorium with Arsenazo III.The latter is the best reagent for thorium; however, it changed its colour upon sorption on a carrier, which prevented the assay. If thorium in the form of the oxalate complex is sorbed on fibres containing an anion- exchange resin, it can be determined in the form of the complex with Arsenazo III on a solid phase in strongly acidic media. In this approach, the detection limit is 0.2 ng ml71 (the sensitivity of photometric determination of thorium with Arsenazo III is 0.01 mg ml71). Uranium, zirconium, titanium, lanthanum, iron, fluorides and phosphates do not interfere with the determina- tion.69, 70 Fibrous adsorbents filled with ion-exchange resins (AB-17, A-5, KU-2) and the styrene ± divinylbenzene copolymer contain- ing no ionogenic groups were studied as possible carriers for a sensitive element for mercury with 4-phenylazo-3-aminorhoda- nine.79 It was found that the reagent is tightly bound only to carriers containing ionogenic (both cationic and anionic) ion- exchange groups.However, the sensitive reaction of mercury with the immobilised reagent occurred only on fibres with the cation- exchange resin KU-2. A method has been developed for detecting mercury in waters with the detection limit of 0.02 mg ml71. Sensitive optical elements for uranium, mercury and lead have been proposed, which are based on the formation of complexes with Arsenazo III, dithizone and hexaoxacycloazochrome, respectively.80 In these studies, polyamide, poly(ethylene tereph- thalate) and cellulose membranes were used as carriers.192 These membranes are made of transparent or semi-trans- parent materials and are resistant to dilute alkalis and acids.Immobilisation of reagents is carried out in a static or a dynamic mode. In the dynamic mode, the solution is allowed to pass over the membrane surface if the pore diameter is less than 0.45 mm. In the case of larger pores, the solution is filtered through the membrane. The ranges of direct determination in water are 0.01 to 0.10 for uranium, 0.005 to 0.05 for mercury and 0.05 to 0.5 mg litre71 for lead. The possibility of selective determination of uranium(VI) at the level of 261078 mol litre71 in the presence of a 100-fold excess of thorium by diffuse reflectance spectroscopy has been demonstrated.81, 82 In these studies, polycaproamide and nylon Table 1.Some characteristics of sorption-spectroscopic methods for the determination of metal ions using fibrous materials filled with ion-exchange resins. Ion (I) Fibre (fabric) Cu(II) KU-2 KU-2 KU-2 Fe(III) KU-2 Ni(II) KU-2 Ag(I) KU-2 Hg(II) KU-2 Pd(II) A-5 A-5 Pb(II) U(VI) AB-17 V(V) AB-17 Cr(VI) AH-31 Th(IV) F(I) Zn(II) F(II) Co(II) Notes. Abbreviations: PAN, 1-(2-pyridylazo)-2-naphthol; DETC, sodium diethyldithiocarbamate; EDTA, ethylenediaminetetraacetic acid; DMG, dimethylglyoxime; PAAP, 4-phenylazo-3-aminorhodanine; NDEA, p-nitrosodiethylaniline; XO, Xylenol Orange; PAR, 4-(2-pyridylazo)resorcinol; DPC, 1,5-diphenylcarbazide; DT, dithizone; F(I), fabric with vinylpyridine groups; F(II), fabric containing PO3H2 groups.membrane filters were used as matrices. The diffuse reflection coefficient, R, is fixed at 660 nm for uranium and 680 nm for thorium and the uranium concentration is determined from a system of equations: where R is the diffuse reflection coefficient, e is the molar adsorption coefficient, c is the concentration (mol litre71) and l is the membrane thickness. The conditions of sorption of azorhodanines, tyrodine and sulfonitrophenol M on polymeric carriers, e.g., caprone mem- Sorption conditions Reagent Detection limit /mg litre71 0.03 0.002 pH 1, C6H8O6 0.04 M HNO3 PAN DETC+EDTA (pH 8.5) 0.01 KSCN 0.1 M HNO3 0.005 pH 4.5 DMG, 2% NH3 0.005 Rhodazole X pH 3 0.02 PAAP pH 1 0.0005 NDEA, pH 2 ¡À 3 0.01 M HNO3 0.004 0.007 KO Arsenazo I pH 5 pH 6 0.002 PAR, 0.1 M HCl pH 3.5, NaF, H2O2 0.0005 pH 2 DPC, 0.05 M H2SO4 0.0002 H2C2O4, pH 2 Arsenazo III, 4 M HCl 0.02 pH 10, Na2B4O7 DT+DETC+ +Na2B4O7 PAN, 0.05 M HCl 0.0005 pH 5 S B Savvin, V P Dedkova, O P Shvoeva R=eU cU l+eTh cTh l, c=cU+cTh, Selectivity factor MZn, Cd, Hg, Pb, V, Al, Mn, Ni Zn, Cd, U Fe(III), Ni Co Ni Co Cd Zn U(VI) Cu �ºd, Zn 4 Bi Fe(III), Cu, Co Fe, Ni, Co, Zn, Pb, Pd, Ca, Mg Cu Hg Pb, Cr, �º d, Ni, Co, Zn Fe, Cl7 Cu Au, Pt, Ru Rh, Ir Al, Ni, Cu, Zn, Sn Fe(II) Fe(III), La Al, Th Zr Zn, Mn, Cr(III), Al Fe(III), Cr(VI) Cu Mo Ca,W,Fe(III), Cr(III), Hg V, Zn, Co, Ni, Al, Cu, Mo, Mn, Pb, Cd F7, PO3¡¦ Fe(III) La,Ti Zr U(VI) Fe(III), Cu, Co, Ni, Pb Fe(III), Mn Cu, Zn Ni Ref. [M/I] 72 76 >1 100 50 10 76 200 50 20 1052 76 100 105 85 100 102 79 77 >103 >102 50 100 1000 4 51 64 100 20 101 78 >100 50 205 71, 74, 87 >103 >102 70 >103 300 50 105 75 >1 73 100 105Sorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials branes, fibrous materials filled with cation- and anion-exchange resins and fabrics with ionogenic groups, were studied in the development of sensitive elements for Hg(II), Cu(II), Ag(I), Au(III), Pt(IV) and Pd(II).It was shown that azorhodanines and sulfonitrophenol M immobilised on caprone membranes can be successfully used for the determination of these metal ions with the detection limit of 1 to 15 mg litre71. Aprocedure has been developed 83 ± 85 for the determination of silver based on its sorption on discs from polyacrylonitrile fibre filled with the cation-exchange resin KU-2 with subsequent detection with rhodazole X; the detection limit is 5 mg litre71. The main characteristics of methods for sorption-spectro- scopic determination of metal ions which employ fibrous materi- als filled with ion-exchange resins and fabrics carrying ionogenic groups as solid phases are summarised in Table 1.III. Instruments Optical sensors for recording analytical signals are used in mini- devices of the well-known companies `Specord', `Spectroton', `Pulsar' and `Multitest', in mini-photometers `Merck', etc. Recent publications describe two more devices. One of them was devel- oped at the Institute of Analytic Instrumentation of the Russian Academy of Sciences and represents a small-size chemosensor analyser designed for the determination of pH and concentration of metal ions; its operation is based on the principle of reflective photometry of chemosensors made of optically transparent poly- mers containing organic reagents.86 Its designers sought to build a specialised instrument for implementation of one or several similar analytical procedures with the aid of sensors having a universal configuration.This approach permits one to perform the measurements in three different regimes, viz., measurements of light absorption of the tested object (7logR) relative to the diffusely reflecting substrate or an analogous object with the known characteristics. The second and third regimes are intended for procedures utilising definite types of sensors and allow for calibration. Chemosensors for measuring pH and concentrations of Fe(III), copper and rare-earth metal ions at the level of 261073 ± 261075 mol litre71 have been developed. The apparatus (1506130680 mm) has a weight of 1 kg (the volume of the samples varies from 5 to 70 ml) and can be used for measuring reflective properties of coloured reagent papers and multilayer sensitive elements.The measurements are performed at wavelengths from 530 to 700 nm. The duration of the exposure of the sensitive element to the sample is 90 ± 120 s. The apparatus operates on alternating current or utilises the energy of internal accumulators. The second instrument, viz., the optical sensor analyser OSA-TM developed at the Scientific-and-Industrial Corporation `Kvant' (Rostov-on-Don), is intended for environmental moni- toring of natural, drinking and purified waste waters for heavy metals at the MPC and lower levels under both laboratory and field conditions.87 It is designed as a colorimetric densitometer and is used for measuring diffuse reflection coefficients of optical sensors in flow-through cells as well as for measuring optical densities of coloured solutions in cells for liquid samples.This apparatus (37062206100 mm) has a weight of 4.5 kg and operates on alternating current or utilises the energy of accumu- lators. It is supplied with an integral flow-through optical cell, a peristaltic pump, a microprocessor with push-button control and eight light filters operating at 400 ± 750 nm. Its operation is based on the detection of changes in colour (e.g., reflection wich optical sensors or the optical density of the tested solution) as a result of interaction of metal ions with analytical reagents. The analytical signals are shown on a digital display as the transmission (reflection) coefficient, optical density and concentration of metal ions in the tested solution. The potentials of OSA-TM are exemplified 87 in the determi- nation of chromium(III) and chromium(VI) in surface waters.Chromium(III) is measured photometrically with Arsenazo III, 193 while the concentration of chromium(VI) is determined in a flow- through cell with anion-exchanger filled fibres using complexation with 1,5-diphenylcarbazide on the solid phase. A method was proposed for the simultaneous determination of chromium(III) and chromium (VI) in sanitary ± general-purpose water up to 1 : 5 ratio at the concentration of Cr(III)>0.2 MPC and that of Cr(VI)>0.1 MPC. The duration of one assay is 10 to 20 min. A procedure for sorption-spectroscopic assay utilising optical sensors is simple with this instrument.A carrier disc (110 mm) is placed in a flow-through cell and the test solution is pumped through the disc at a rate of 10 ml min71 under optimal sorption conditions. If unmodified carriers are used, the reagent solution is pumped through under optimum assay conditions after which the diffuse reflection coefficient is measured. The main advantage of the OSA-TM analyser is that it allows both routine photometric measurements and measurements in flow-through optical cells where the sorption of the tested ion occurs. The reaction of the ion takes place on a solid phase and is followed by measurement of the analytical signal. A combination of these three processes in a single flow-through cell not only simplifies the procedure but also increases significantly the repro- ducibility of experimental results owing to the omission of one step, i.e., the transfer of the optical sensor from the sorption cell to the measuring cell, which excludes any fluctuations in the sensor wetness.The use of the instrument allows for rapid presentation of experimental results in units of concentration. The construction of the instrument with a flow-through optical cell became possible due to development of optical sensors in which fibrous materials filled with ion-exchange resins or fabrics containing ionogenic groups are used as carriers. IV. Test methods Numerous versions of test methods are being actively devel- oped.88 ± 104 The materials used as carriers for sorption-spectro- scopic assays, such as reagent papers, foamed polyurethanes, silica gel, membranes, ion-exchange resins, etc., are also employed in test methods.Indicator test systems, e.g., from `Merck', are currently known. Some densitometric characteristics for express tests have been obtained which are based on the use of reactive reagent papers 96 and are designed for the determination of certain elements with the help of reflective electronic mini-photometers equipped with light-emitting diodes operating at wavelengths of 56715 and 66010 nm and a modified device for concentration of 1077 mass%± 1075 mass%of components in aqueous media. The main requirements for test reactions are as follows:101 high sensitivity (not lower than MPC); high light-transition contrast; proximity of the optimum acidity of assays to the acidity of the tested solution; stability of external effects for a sufficiently long time and upon exposure to light; the possibility to preserve the test form after positive response for subsequent quantitative assay under laboratory conditions; and selectivity.The methodological aspects of detection and processing of analyticalnals in colorimetry have been reviewed.48, 101 The general colour discrimination criterion (DE) is the main parameter in making a test scale. In order to increase the sensitivity of a test reaction, special parameters, such as whiteness (W) for colourless sorbents and yellowness (G) for blue-coloured complexes have been introduced. The calibration test scale is built up from the known dependence of the general colour discrimination on the concentration of the tested element.One unit of DE corresponds to one colour discrimination threshold; therefore, the concentra- tions of tested elements are selected in accordance with the corresponding magnitude of DE. A reliable sense of difference in light flux intensities is reached if this difference is at least 10%. Therefore, discrete values of colour discrimination with DE>10 are especially preferred in the case of diffuse reflection. If the tested sample loses its colour or is bleached in the course of an194 analytical reaction, its whiteness is used for building up a test scale: W=1007DE.The lower threshold of tested concentrations in any test method is defined as general colour discrimination (10 arbitrary units) between the original test sample (E0) and the test sample following its reaction with the tested element (E): E7E0=10 or W07W=10. In the building up of a test scale, one should not go beyond the limits of saturability (DS) of the tested sample (20% ± 60%) which corresponds to the upper threshold of tested concentrations. The contrast of a colour reaction is enhanced after application of optical bleaching compounds onto the matrix surface; their operation principle is fluorescence. The presence of these com- pounds does not influence chemical and analytical characteristics of immobilised reagents.In their presence, reagents are rapidly and evenly distributed over the matrix surface and are well retained in acid media under both static and dynamic conditions. For example, the detection limit in the determination of ura- nium(VI) with Arsenazo III and application of optical bleaching is 0.001 mg ml71. The colour of the complex compound is bright green, whereas in the absence of the bleaching agent the complex has a grey-blue colour and has a detection limit of 0.1 mg ml71 (see Ref. 102). Aportable field test system for the determination of numerous elements in water, soil and rainfall has been developed.103, 104 This assay includes the following steps, viz., measurement of the lengths of coloured or bleached zones on reactive reagent papers cut into stripes and pasted into a polymeric film one end of which is brought into contact with the tested fluid, visualisation of colour intensity, measurement of the area of the coloured or bleached zone and visualisation of colouring of the tested fluid after application of reagent papers.The range of tested concentrations is 0.001 ± 1000 mg litre71. 1. Reagent test papers The majority of publications devoted to test methods deal with reagent papers.88 ± 121 Heavy metals on the whole are determined with the help of a device which is similar to that used for air sampling (the detection limit is up to 1077 mol litre71) which employs test papers impregnated with potassium tetraborate and PAN.107 Rapid tests for the determination of toxic elements in water are carried out on papers, films or glasses with the detection limits of 0.05 for Cu(II), 0.2 for Fe(III), 0.5 for Co(II), 0.03 for Mn(II) and 0.03 mg litre71 for V(V) (see Ref.108). Iron ions are determined on reactive test papers with detection limits ranging from 0.02 to 0.2 mg litre71 depending on the method (see Refs 109 ± 111). A tape method was proposed for the determination of Fe(III) (0.5 ± 3.5 mg litre71) on papers impregnated with the anion-exchange resin Amberlite IRA 400 modified with sodium ethylenediaminedi(hydroxybenzylphos- phonate).112 Immobilisation of pyrazole-containing formazanes and hydrazones on the base-paper increases their sensitivity with respect to transition metals.Express tests for determining Cu(II), Ni(II) and Co(II) at the level of 1077 mass%± 1078 mass%have been developed.113 Test paper impregnated with Michler's thio- ketone, potassium dihydrogenphosphate, sodium thiosulfate (in order to reduce copper and remove active chlorine) and Triton X-100 was proposed for the determination of copper in drinking water (the detection limit is 0.05 mg litre71);114 test paper con- taining the nickel complex with 2-nitroso-1-naphthol, diphenyl- guanidine, thiourea and ammonium fluoride was used for the determination of cobalt.115 An express test method for the determination of manganese in drinking water is carried out on paper containing a multicompo- nent reagent. This analysis is based on the measurement of the time interval during which the indicator zone changes its colour as S B Savvin, V P Dedkova, O P Shvoeva a result of oxidation of Malachite Green with potassium periodate (KIO4) catalysed by manganese(II) ions.The detection limit is 0.03 mg litre71 (see Ref. 116). Zinc ions are determined in foodstuff using reagent test paper (FMOPF-6C) at pH 8.0. The effect of copper ions is eliminated by addition of thiourea.117 Reagent test paper was recommended for use as indicators for multicomponent assay of water:118 Ion Ion Concentration range /mg litre71 Concentration range /mg litre71 0.001 ± 0.1 0.01 ± 0.2 0.2 ± 2 Bi(III) Fe(II) Hg(II) 0.001 ± 0.1 0.001 ± 0.1 0.01 ± 0.2 0.1 ± 5 Pb(II) Cu(II) Fe(III) Cr(VI) Lead ions are determined using a reagent test paper (RTP) metal test after its fast separation from Cu, Fe and Zn in aqueous solutions (pH 6.0), emulsions and suspensions using the sensor filter of a pocket indicator ultrafiltration unit.118 Stripes of reagent paper are used for the quality control of water;119 paper stripes impregnated with diphenylthiocarbazone are used for test determination of trace amounts (15 ppb) of lead in drinking water,120 while paper stripes soaked with a covalently bound multidentate chromogenic reagent of the formazane type are used for the determination of mercury with a detection limit of 0.05 mg litre71 (see Ref.121). The detection limit in enzymic determinations of mercury and organomercurials on foamed polyurethanes, paper and chitosan with immobilised peroxidase is 1075 mg ml71, i.e., below the MPC for mercury.122 ± 127 A sorption-catalytic procedure for the determination of manganese 128 is based on preliminary concen- tration of manganese(II) ions on cellulose filter paper with chemi- cally bound diethylenetriaminotetraacetate groups (DETATA) and subsequent catalytic oxidation of 3,30,5,50-tetramethylbenzi- dine with KIO4.The optimum conditions for this reaction in solution and on filter paper were found to be different. Prelimi- nary concentration of manganese on DETATA-paper allows for a nearly 10-fold decrease in the detection limit (up to 661076 mg litre71) and expands the range of tested concentra- tions up to 561076 ± 2.561073 mg litre71.The selectivity of the assay is higher in this case: after preliminary concentration from a 20-ml sample, determination of 1 ng of manganese is insensitive to a 1000-fold molar excess of other ions. This method is applied in the analysis of tap and river water. An express test for the detection and determination of vanadium, which catalyses the oxidation of p-diethylaminoaniline with bromate, on glass or poly(methyl methacrylate) at pH 4 ± 5 has been developed. The detection limit decreases in the presence of ferrone which potentiates the catalytic effect of vanadium. This determination is selective with respect to many metals. The interference of Fe(III) ions is eliminated with fluorides. The detection limit for vanadium is 0.09 mg litre71 (see Ref.129). The approaches used in test methods are very diverse. For example, in one of them the analyte is concentrated in a limited area on a hydrophobic surface by evaporation and the element is detected by colour reaction or fluorescence. This procedure is used to determine aluminium with Xylenol Orange in tap water at pH 8.0 or with lumogallione at pH 7.0 (the detection limit is 2.7 ppb) (see Ref. 130). Molybdenum is determined by an ion-exchange drop reaction with azide with the detection limit of 2 mg in 0.09 ml of the sample using the anion-exchange resin Bio-Rad AG 1X8.131 In the determination of chromium(VI), five to ten beads of the anion- exchange resin Amberlite IR-120 (H-form) are placed onto a white plate and Malachite Green is added (drop reaction).The green colouring of the resin surface observed after 10 min points to the presence of chromium; the detection limit is 1.6 mg.132 Iron ions are determined on the anion-exchange resin AB-17X8Cl with the detection limit of 0.01 mg ml71 (see Ref. 41). Polyester fabric based on poly(ethylene terephthalate)Sorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials is used as a substrate for an array of multipurpose indicators. This procedure permits complete elimination of cross contamination of reagent paper stripes with chemical reagents from adjacent stripes.133 2. Silica gel Test methods on silica gel have attracted attention in the past decade.8, 34, 49, 134 ± 140 It was found 49 that the optical character- istics of Fe(III), Cr(VI), Mn(II), Ni(II), Cu(II) and Zn(II) complexes with PAN on Silochrom C-120 are very similar to those in solution.Thus sorption of 1-(2-pyridylazo-)-2-naphthol and 4-(2-pyridylazo)resorcinol (PAR) from aqueous solutions of var- ious acidities was investigated and the conditions for the prepara- tion of modified adsorbents with predetermined capacity with respect to azo compounds were established. Sorption of PAN and PAR on Silochroms C-80 and C-120 was studied under static conditions.134 A procedure has been developed for the determi- nation of the reagents in the sorbent phase. The indirect assay consists in the measurement of changes in light absorption of solutions after immobilisation.Alternatively, azo compounds are determined upon oxidation with cerium(IV) in the modified sorbent phase. Maximum modification of the silica surface is observed at pH<3. The protonated form of PAR is the most intensively adsorbed on silica hydrated to the greatest degree. The completely dissociated form of PAR is not adsorbed by the silica surface. The degree of adsorption of PAN at pH 4.0 decreases as the proportion of acetone in the tested solution increases. At an acetone concentration of 60 vol. %, PAN is not adsorbed on silica surface. Complete adsorption of PAN with Silochrom C-120 usually takes 20 minutes. It was found that the efficiency of sorption of azo compounds depends on the structure of the silica, its porosity and specific surface.With an increase in specific surface, sorption of the reagent increases. At lower specific surface, the greater part of the reagent diffuses into the sorbent and the pore sizes thus become critical. It was found that adsorbents with a low capacity with respect to the reagent (e.g., <25 mmol g71) are more preferable for use in test methods (see Ref. 134). The colorimetric method was used for separate determination of Cu [(0.4 ± 22.0)61076 mass %] and Zn [(3.8 ± 39.0)61076 mass %] in natural water on PAN-modified silica.48 Cobalt was determined with PAN immobilised on silica gel 135 using a test method based on colour discrimination by luminosity (the detec- tion limit is 0.004 mg ml71); its quantitative determination was performed using the Gourevich ± Kubelka ± Munk function (the detection limit is 0.03 mg ml71).It was found that on going from diffuse reflection coordinates (DF) to sorbate colour coordinates (DL) with a test reagent, the sensitivity and reproducibility of determination of cobalt increase 4- and 1.5-fold, respectively. 4-(2-Thiazolylazo)resorcinol immobilised on silica gel was recom- mended 136 as the test reagent for Co(II), Pd(II) and U(VI) (the detection limits are 0.02, 0.005 and 0.03 mg ml71, respectively). Test methods for the determination of Co(II), Cu(II) and Zn(II) with 1-(2-thiazolylazo)-2-naphthol, for the determination of Ag(I), Hg(II), Pb(II) with dithizone and for the determination of Co(II), Fe(III) and BiI¡4 thiocyanates with quaternary ammonium salts on silica gels and cellulose have been developed.34 The optical and colorimetric characteristics of immobilised 2-(5-bromo-2-pyridylazo)-5-diethylaminophenol and its com- plexes were obtained in the analysis of sorption of indium and uranium(VI) ions on Silochrom C-120.Test methods for the determination of indium with a detection limit of 0.002 mg ml71 (see Ref. 44) and uranium with a detection limit of 0.25 mg ml71 (see Ref. 137) have been developed; subsequent determination of uranium by diffuse reflectance spectroscopy in the sorbate phase reached a detection limit of 3.5 ng ml71. The optimum conditions for the formation of uranium(VI) complexes with PAR immobi- lised on KU-2, foamed polyurethane and Silochrom C-120 were established as pH of 5.8 ± 8.2 and an equilibration time of 20, 20 and 15 min, respectively.An advantage of PAR immobilised on 195 KU-2 is that it elicits analytical signals of greater magnitude and produces a broader plateau on the optical density versus pH of the complex solution curve. This reaction is selective in the presence of cyclohexanediaminetetraacetic acid; the detection limit is 0.1 mg ml71 (see Ref. 138). A linear-colouristic method 141 ± 143 is based on the linear dependence of the length of the coloured zone of the display tube on the concentration of the tested compound. In this procedure, hydrophobic carriers, such as silica gels Diasorb-C16 and Diasorb- phenyl, are modified with known analytical reagents.The dimen- sions of the coloured zone depend on the nature of the carrier, quantity of the adsorbed reagent, internal diameter of the display tube, flow rate and pH of the sample. By varying tube diameter, one can change the concentration range of the tested element. The detection limits for various elements vary from 1078 to 1076 mol litre71. Silica gel modified with lead diethyldithiocarbamate can be used as the indicator powder for the determination of copper ions with a detection limit of 1.561078 mol litre71 in the presence of 1000-fold excess of Ni(II), Zn(II), Cd(II) and Co(II).141 PAN and Xylenol Orange immobilised on silica gel are used for the determination of copper and iron.142 At equal concentrations of these ions (< 561076 mol litre71), two differently coloured zones are formed in the tube filled with silica gel Diasorb- phenyl-PAN, viz., crimson (Cu) and green (Co).142 The possibility of determination of cobalt and Fe(III) in the same sample at the MPC level with the use of a tube filled with silica gel Diasorb-C16 modified with 1-nitroso-2-naphthol was demonstrated.143 This method is characterised by ease of tube preparation and rapidity of analysis.The possibility of direct determination of trace amounts (1079 ± 1075 mol litre71) of heavy metals in water, soil extracts and other natural subjects by the sorption-colouristic method was considered.144 Coloured zones are produced upon frontal chro- matographic sorption of Hg(II), Cd(II), Pb(II), Co(II), Cu(II), Bi(III), Sb(III) and Fe(III) on microcolumns packed with cellulose and polystyrene-based ion-exchange resins modified with polar residues (e.g., sulfides of zinc, cadmium and manganese as well as hexacyanoferrates of calcium and lead) and organic reagents.3. Foamed polyurethanes and membranes The polymeric matrix of foamed polyurethanes, which have a membrane structure, contains both polar groups and hydro- phobic fragments. This makes foamed polyurethanes helpful tools in test methods for the determination of metal cations, polycyclic hydrocarbons, phenols, heteropolyacids, surfactants and other compounds.93, 123, 138, 145 A number of test reactions on foamed polyurethane carriers have been developed for determination of several elements (the colour scale ranges, Dc, are indicated): Co Element Fe 0.004 ± 0.6 Ti 0.04 ± 6 0.01 ± 0.5 Ni 0.1 ± 40 Cr Dc /mg litre71 0.1 ± 1 1,5-Diphenylcarbazide and dimethylglyoxime are used for the determination of chromium and nickel, respectively; potassium thiocyanate is used for the determination of cobalt, titanium and iron.93 Immobilisation of chitosan HRP on foamed polyurethane gives a more active and stable sorbent than on a polystyrene plate or on paper.Oxidation of o-dianizidine proved to be the most efficient indicator reaction which made it possible to decrease the detection limit for mercury to 2561073 ng ml71. The rate of the indicator reaction is monitored visually by fixing the moment of appearance of red-coloured oxidation product.Thiourea was used to enhance the inhibitory effect of mercury. This method is not only highly sensitive, but also highly selective: only 105-fold excess of Cd(II), Pb(II) and Bi(III) interfere with the determination.123 Optimum conditions were established for complexation of uranium(VI) with PAR immobilised on foamed polyurethane,196 KU-2 and Silochrom C-120. The detection limit is 0.1 mg ml71 (see Ref. 138). Test methods employing poly(vinyl chloride) and caprone membranes were recommended for the determination of Hg(II), Ag(I), Cu(II), Au(III), Pd(II) and Pt(IV) in the presence of at least a 10-fold excess of Fe(III), Ni(II), Co(II), Zn(II), Cd(II), Be(II), Al(III) and alkaline-earth metals with detection limits from 1 to 15 mg litre71.Immobilised azorhodanines and Sulfonitrophenol Mwere used as reagents.85 A highly sensitive field test for the determination of chro- mium(VI) in water at the level of <1 mg litre71 is based on the formation of a coloured chromium complex with 1,5-diphe- nylcarbazide in 0.1 M sulfuric acid and its sorption on an ion- exchanger incorporated into the membrane.146 The intensity of coloration is proportional to the chromium concentration in the range from 0.05 to 50 mg litre71. In 50-ml samples, chromium is determined with a detection limit of 10 ng litre71. 4. Fibrous materials filled with ion-exchange resins Fibrous filled materials are perfect carriers for visual assays.They allow for simultaneous concentration and determination of tested elements as well as for variation in the nature of the adsorbent depending on a concrete experimental task. The tested ions may first be adsorbed on original, colourless materials and are then `developed'. Alternatively, adsorbents with immobilised reagents can be used. Carriers can be used under both static and dynamic conditions; they are resistant to mechanical and chemical actions. Fibrous materials filled with ion-exchange resins and fabrics with ionogenic groups are used as carriers in visual determinations of metal ions using known colour reactions with organic reagents.69, 74, 99, 147 ± 151 The tested element is selectively concen- trated on a solid phase; the use of highly sensitive and highly selective reagents facilitates its determination in drinking water at the MPC and lower levels.In addition to remarkable kinetic properties (e.g., ease of sorption and high filtration rates), fibrous materials endowed with Table 2. Some characteristics of test methods for the determination of metal ions which utilise fibrous materials as a solid phase. Ion (I) Fibre (fabric) KU-2 Fe (III) KU-2 Cu(II) KU-2 Ni(II) KU-2 Pd(II) AB-17 V(V) Selectivity factor Detection conditions Sorption conditions M KSCN 0.1 M HNO3 0.01 ± 0.05 M HNO3 DETC+ +EDTA DMG pH 5 ± 6 NDEA 0.01 M HNO3 pH 2.5, 0.05% H2O2 PAR, 1M HCl, 50 ± 60 8C Ni Co Cd Zn U(VI) Cu Cd, Zn, U(VI), Mn(II) Fe(III), Ni Co Bi Cd, Zn Bi Fe(III), Cu, Co Ni, Cu, Zn, Cd Co, Rh, Ir Fe(III), Au, Pt, Ru Ni Cu Mo, Cr(VI) Fe(III) V(IV) ion-exchange properties manifest high stabilities in aggressive media.Therefore they can be used for sorption and determination of metal ions under drastic conditions in the presence of large concentrations of masking substances and in highly acidic media. These materials are distinguished by handling convenience, are easily transferred with pincers and do not lose their shape upon stirring, heating in solution, drying, etc. Express tests are based on colour reactions of metal ions with reagents on a solid phase. Fibrous materials filled with ion- exchange resins used as carriers and fabrics with ionogenic groups in the form of discs 1 ± 2 cm in diameter serve not only as the reaction sites but also as sorbents for the concentration of tested ions (and thus increasing the sensitivity of the reaction) and their separation from ballast substances.The use of carriers containing specific groups allows one to lower the detection limit and to increase the selectivity. In addition, these express methods employ well-known, organic reagents that have proved to be efficient in photometric assays and are commercially available which signifi- cantly simplifies their introduction into routine analytical prac- tice.The analytical procedure is simple and easily accessible. Reagent solutions are added dropwise to a tested solution (10 ml) to ensure optimum sorption conditions after which the carrier disc is introduced.The solution is agitated for 5 ± 7 min and then decanted with the disc held back by a glass rod. One to three drops of the reagent solution are dripped onto a colourless unmodified disc and the colouring of the disc is compared with the calibration colour scale obtained under identical conditions. In some cases, the scale, enclosed in a polyethylene film, is preserved for a sufficiently long period of time. Carrier discs can be regenerated with acids and re-used after washing with water. Some characteristics of modern test methods for visual assessment of metal ions utilising fibrous materials filled with ion-exchange resins and fabrics with ionogenic groups are summarised in Table 2.Change in disc colour [M/I] white?red 200 50 20 1052 white?yellow 100 50 105 white?red 100 105 yellow?red 104 103 yellow?violet 102 500 200 50 202 S B Savvin, V P Dedkova, O P Shvoeva Ref. MPC /mg litre71 Colour scale /mg litre71 99 0.3 0.1 ± 1 99 1 0.05 ± 1 99 0.1 0.1 ± 1 77, 147 0.05 ± 0.5 7 148, 149 0.1 0.1 ± 1Sorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials Table 2 (continued). Ion (I) Sorption conditions Fibre (fabric) AB-17 Cr(VI) 0.01 M H2SO4 AB-17 oxidation of Cr(III) to Cr(VI) Cr(VI)+ +Cr(III) AB-17 Mo(VI) pH 3 ± 5, C6H8O6 AN-31 Th(IV) 0.004 M H2C2O4, pH 2 AN-31 U(VI) pH 6, acetate buffer AN-31 U(VI)+ +Th(IV) 0.01 M NaHCO3, 0.001 M EDTA, pH 9 ± 10 F(I) Zn(II) F(II) Co(II) pH 9 ± 10, borate buffer pH 5 ± 6, acetate buffer Studies with two or three elements or with one element in different oxidation states are being developed during the past decade.For example, both Cr(VI) and Cr(III) ions can be determined following oxidation of Cr(III) by heating with ammo- nium persulfate in 0.05 M H2SO4 followed by decomposition of the excess of the oxidant by boiling with subsequent sorption of Cr(VI) ions on discs prepared from fibres filled with an anion- exchange resin and their detection with 1,5-diphenylcarbazide, which gives a pink coloration. The solution is divided into two parts one of which is used for the determination of only Cr(VI) ions and the other one, for the determination of both Cr(III) and Cr(VI) ions after which the concentration of Cr(III) ions is determined from the difference between the two measurements.99 If uranium(VI) and thorium(IV) are present in solution in 1 : 3 to 3 : 1 ratios, they can be determined in the same sample with Arsenazo III in the presence of a 3-fold excess of zirconium.Sorption is performed from a carbonate medium in the presence of ethylenediaminetetraacetic acid; the assay is carried out after acidification with 6 ± 7 M HCl. The overall concentration of uranium and thorium is determined judging from the coloration of wet discs; thorium is determined judging from the coloration of dried discs.The coloration of the uranium complex with Arsenazo III disappears upon drying due to a decrease in acidity.150 V. Conclusion At present, the detection limits for metal ions attainable using sorption-spectroscopic and test methods reach 1073 ± 1072 mg litre71. Enzymic test methods provide an exception, Selectivity factor Detection conditions M DPC 4 Phenyl- fluorone, 0.5 M HCl 4 6 M HCl Arsenazo I W(VI), Cr(III), H2O2 V(V), Zn Ni, Co, Al, Hg, Fe(III) Pb, Cd Cl7, SO2¡ determination of Cr(VI) against Cr(III) and a sum of chromium ions Fe(III) Cr(VI) Sn, Zr V(V), Th Arsenazo III, F7, PO3¡ Fe(III) La, Ti Zr U(VI) La, Fe(III) Th, Al Ti(IV) Zr Arsenazo III, Th :U 7 M HCl Zr PAN DT+DETC Fe(III), Cu, Co, Ni, Pb Fe(III), Mn Cu, Zn Ni for they allow measurements of up to 1075 mg litre71.However, by reason of high complexity of these reactions, the reproduci- bility of experimental results is rather low. In some cases, sorption-spectroscopic methods allow for a detection limit of 1074 mg litre71. As can be judged from the increasing number of publications, these methods are extensively developing along several different lines. The range of tested elements and carriers is expanding and novel procedures for the detection of analytical signals (e.g., colorimetry, derived photometry, flow-injection analysis, etc.) evolve. New steps are having made on the road to multielement assays.The equipment for sorption-spectroscopic assays is also being improved. The use of small-size instruments allows a more objective estimation of experimental results, on the one hand, though this increases the cost of assays on the other hand. Test methods intended for routine and versatile uses present partic- ularly great interest in those cases where the most reliable tool is the human eye, especially because sometimes it detects differences in colour much better than any device. In this context, the simplest, easiest, highly contrasting and safe systems provide the most efficient test methods. Sometimes, the sensitivity and accu- racy of sorption-spectroscopic methods is one order of magnitude higher than those of test methods.When these methods are used simultaneously, the concentration of toxic elements can be assessed by test methods; more accurate concentration measure- ments are accomplished by sorption-spectroscopic methods. Modern test methods allow detections at theMPC level. Some ions, e.g., mercury, selenium, cadmium, etc., having detection Change in disc colour [M/I] white?pink >103 500 200 100 >104 the same orange?crimson pink?grey-green pink?blue 600 100 20 10 >103 300 50 105 20 1021 1 : 3 ± 3 : 1 pink?green 3 orange?crimson orange?green >1 100 105 197 Ref. MPC /mg litre71 Colour scale /mg litre71 74, 99 0.01 ± 0.25 0.05 99 0.01 ± 0.25 0.05 149 0.25 0.02 ± 0.4 69 0.01 ± 0.1 7 64, 147 0.05 0.1 ± 1 150 0.02 ± 0.24 7 (sum) 0.01 ± 0.12 (thorium) 0.1 ± 0.5 99 1.5 99 0.1 0.05 ± 0.5198 limits below 1073 mg litre71 present a problem.However, the main problem involved in sorption-spectroscopic and test meth- ods is selectivity. In order to increase selectivity, one should make the carrier resistant to aggressive media, such as acids, alkalis, masking complexing compounds, oxidants and reducing agents used to eliminate the interfering effects of ballast components. There must also be a possibility to vary the nature of the carrier in order to increase the sorption of the tested ions and to prevent the sorption of ballast components. In our opinion, ion-exchange materials have indisputable advantages over paper, membranes, silica gel and other materials.They are characterised by high mechanical and chemical stabil- ities, high capacity and good kinetic properties, which allows for a significant decrease in the detection limit by increasing the volume of the tested solution in those cases where the sensitivity of a colour reaction is not high enough. The array of active ionogenic groups is rather broad. The use of fibrous ion-exchange materials eliminates such disadvantage of ion-exchangers as friability and ensures easy drying, washing and other procedures without any detriment to the carrier surface. This increases the reproducibility of exper- imental results and simplifies the automation of analysis. The use of fibrous materials filled with ion-exchange resins permits one to combine selective concentration of an element with its selective determination and thus solves the problem of selectivity, lowering of the detection limit and rapidity of analysis.In this sense, too, these materials hold much promise. Recent trends in the development of sorption-spectroscopic and test methods are towards further improvement of assay methodology and synthesis of novel reagents and updating of the existing ones aimed at their use in combination with the solid phase. Some reagents often lose their ability to form complexes upon immobilisation on the carrier. The broadening of the range of carriers and their potentials, variation of the nature and degree of dispersion and filling of carriers may help to solve numerous problems connected with the increase in the sensitivity and selectivity of analytical methods.And, finally, it is necessary to update the existing mini-instruments and to design new ones including those which utilise fibre optics, flow-through cells and mini-computers providing for rapid analysis in an automated regime. This will favour progress in the design of sorption- spectroscopic and test methods and their application in the analysis of natural and industrial materials and environmental subjects. This work was written with the financial support of the Russian Foundation for Basic Research (Project No. 97-03- 33441). References 1. R Hering Chelatbildende Ionenaustausscher (Berlin: Academie- Verlag, 1967) 2.K M Saldadze, V D Kopylova-Valova Kompleksoobrazuyushchie Ionity (Kompleksity) [Complexing Ionites (Complexites)] (Moscow: Khimiya, 1980) 3. G V Myasoedova, S B Savvin Khelatoobrazuyushchie Sorbenty (Chelating Sorbents) (Moscow: Nauka, 1984) 4. G V Myasoedova, O P Shvoeva, I I Antokol'skaya, L I Bol'shakova, N I Shcherbinina, S B Savvin, in Opredelenie Malykh Kontsentratsii Elementov (Determination of Low Concentration of Elements) (Moscow: Nauka, 1986) p. 78 5. M L Marina, V Gonzalez, A R Rodriguez Microchem. J. 33 275 (1986) 6. T A Belyavskaya, G D Brykina, in Opredelenie Malykh Kontsen- tratsii Elementov (Determination of Low Concentration of Elements) (Moscow: Nauka, 1986) p. 85 7. A A Popov, V K Runov, in Kontsentrirovanie Sledov Organicheskikh Soedinenii (Concentration of Traces of Organic Compounds) (Moscow: Nauka, 1990) p.143 8. R J Hurtubise Anal. Chem. 61 889A (1989) S B Savvin, V P Dedkova, O P Shvoeva 9. G D Brykina, D Yu Marchenko, O A Shpigun Zh. Anal. Khim. 50 484 (1995) a 10. O A Zaporozhets, O M Gaver, V V Sukhan Usp. Khim. 66 702 (1997) [Russ. Chem. Rev. 66 637 (1997)] 11. K Yoshimura, H Waki Talanta 32 345 (1985) 12. K Esimura Bunseki (5) 341 (1986); Ref. Zh. Khim. 6 G 51 (1987) 13. F Capitan, L F Capitan-Vallvey, MC Valencia, JMBosque- Sendra, F Molina, I de Orbe Analysis 19 (6) 177 (1991) 14. G D Brykina,G G Lebedeva, O A Shpigun Zav. Lab. 59 (1) 4 (1993) 15. G D Brykina, G F Agapova, V F Kalinina, L S Krysina Zh.Anal. Khim. 43 1461 (1988) a 16. J Bosque-Sendra, C Said Bondra Fresenius J. Anal. Chem. 360 31 (1998) 17. M I Pascual-Reguera, A Molina-Diaz, N Ramos-Martos, L F Capitan-Vallvey Anal. Lett. 24 2245 (1991) 18. A Molina-Diaz, M I Pascual-Reguera, E Linan-Veganzones, M L Fernandez de Cordova, L F Capitan-Vallvey Talanta 43 185 (1996) 19. I Nakatsuka, A Nishimura, K Ohzeki Anal. Chim. Acta 304 243 (1995) 20. M I Shtokalo, E E Kostenko, I Z Zhuk Zh. Anal. Khim. 46 1093 (1991) a 21. M I Shtokalo, E E Kostenko, I Z Zhuk Zh. Anal. Khim. 47 1827 (1992) a 22. M I Shtokalo, E E Kostenko, in XV Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Minsk, 1993 [The XVth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Minsk, 1993] Vol.4, p. 224 23. S Matsuoka, K Yoshimura, H Waki Mem. Fac. Sci. Kyushu Univ. 18 55 (1991); Ref. Zh. Khim. 11 GD 157 (1992) 24. G T L Sena, C F S Luis, S C A Celso, P R F Rodrigo, K Mauro, F D R Boaventura Chimia 52 368 (1998) 25. S B Savvin, A V Mikhailova Zh. Anal. Khim. 51 49 (1996) a 26. V K Runov, V V Tropina Zh. Anal. Khim. 51 71 (1996) a 27. L N Zhukova, V K Runov, I E Talut', A K Trofimchuk Zh. Anal. Khim. 48 1514 (1993) a 28. V N Losev, M P Bakhtina, I P Bakhvalova, A K Trofimchuk, V K Runov Zh. Anal. Khim. 53 1170 (1998) a 29. N A D'yachenko,A K Trofimchuk, V V Sukhan Zh. Anal. Khim. 54 159 (1999) a 30. V N Losev, I V Maznyak, A K Trofimchuk Zav. Lab. Diagnost. Mater. 64 (6) 11 (1998) 31.N A Dyachenko, E A Karetnikova, A K Trophimchuk, V V Sukhan, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 1, C-54 32. T I Tikhomirova, V I Fadeeva, in International Congress on Analyt- ical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 1, D-24 33. O Zaporozhets,N Petruniok, V Sukhan, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 1, D-26 34. O A Zaporozhets, O Yu Nadzhafova, N I Petrun'ok, V V Sukhan, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Moskva, 1998 [The XVIth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Moscow 1998] Vol. 3, p. 99 35. O Zaporozhets, O Gaver, V Sukhan, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol.1, N-68 36. O V Kuznetsova, T V Savel'eva, V M Ivanov, A A Zheleznova Vestn. Mosk. Univ., Ser. 2, Khim. 36 544 (1995) b 37. L I Savranskii, O Yu Nadzhafova Khim. Tekhnol. Vody 13 425 (1991) 38. V M Ivanov, N I Ershova Vestn. Mosk. Univ., Ser. 2, Khim. 38 396 (1997) b 39. N Ershova, V Ivanov Anal. Chim. Acta 364 235 (1998) 40. L E Zel'tser, A V Bychenko Zh. Anal. Khim. 48 1659 (1993) a 41. O V Kuznetsova, V M Ivanov, N V Kazennov Vestn. Mosk. Univ., Ser. 2, Khim. 38 53 (1997) b 42. N I Ershova, V M Ivanov, S A Morozko, O V Kuznetsova, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Moskva, 1998 [The XVIth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Moscow 1998] Vol.3, p. 108Sorption-spectroscopic and test methods for the determination of metal ions on the solid-phase of ion-exchange materials 43. S A Morozko, O V Kuznetsova, V M Ivanov Zh. Anal. Khim. 52 1146 (1997) a 44. V M Ivanov, N I Ershova Vestn. Mosk. Univ., Ser. 2, Khim. 39 170 (1998) b 45. V M Ivanov, N I Ershova Vestn. Mosk. Univ., Ser. 2, Khim. 39 101 (1998) b 46. O V Kuznetsova, V M Ivanov, in International Congress on Analyt- ical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 2, L-104 47. V M Ivanov, O V Kuznetsova, O V Grineva Zh. Anal. Khim. 54 263 (1999) a 48. S A Morozko, V M Ivanov Zh. Anal. Khim. 52 858 (1997) a 49. S A Morozko, V M Ivanov Zh.Anal. Khim. 50 629 (1995) a 50. S A Momin, R Narayanaswamy Anal. Proc. 26 372 (1989) 51. L M Trutneva, O P Shvoeva, S B Savvin Zh. Anal. Khim. 44 1804 (1989) a 52. Yu A Zolotov Zh. Anal. Khim. 45 1255 (1990) a 53. B F Myasoedov, A V Davydov Zh. Anal. Khim. 45 1259 (1990) a 54. B F Myasoedov Izv. Akad. Nauk, Ser. Khim. 487 (1992) c 55. Yu G Vlasov Zh. Anal. Khim. 47 114 (1992) a 56. Yu G Vlasov, in Proceedings of International Trace Analysis Sympo- sium 94 (The 7th Japan-Russia Joint Symposium in Analytical Chem- istry), Hakodate; Sapporo, 1994 p. 181 57. T Tanata Anal. Chem. 64 196 (1992) 58. W A Decliveira, R Narayanaswamy Talanta 39 1499 (1992) 59. R Narayanaswamy Analyst 118 317 (1993) 60. J Buffle,M L Tercier, N Parthasarathy, K J Wilkinson Chimia 51 690 (1997) 61.A Flamini, A Panusa Sens. Actuators B42 39 (1997) 62. E Pretsch Chimia 51 768 (1997) 63. J Janata, M Josowicz, P Vanysek, DMVaney Anal. Chem. 70 179 (1998) 64. O P Shvoeva, L M Trutneva, S B Savvin Zh. Anal. Khim. 44 2084 (1989) a 65. O P Shvoeva, V P Dedkova, S B Savvin Zav. Lab. Diagnost. Mater. 62 (1) 1 (1996) 66. S B Savvin, L M Trutneva, O P Shvoeva, V K Belyaeva, I N Marov Zh. Neorg. Khim. 36 393 (1991) d 67. O P Shvoeva, V P Dedkova, S B Savvin Zav. Lab. Diagnost. Mater. 62 (8) 21 (1996) 68. O P Shvoeva, V P Dedkova, S B Savvin Zh. Anal. Khim. 52 692 (1997) a 69. O P Shvoeva, V P Dedkova, S B Savvin Zav. Lab. Diagnost. Mater. 62 (8) 13 (1996) 70. O P Shvoeva, V P Dedkova, S B Savvin Zh. Anal.Khim. 53 471 (1998) a 71. S B Savvin, O P Shvoeva, L M Trutneva, E A Likhonina Zav. Lab. 58 (6) 5 (1992) 72. O P Shvoeva, S B Savvin, LMTrutneva Zh. Anal. Khim. 45 476 (1990) a 73. L M Trutneva, O P Shvoeva, S B Savvin Zh. Anal. Khim. 49 473 (1994) a 74. S B Savvin, L M Trutneva, O P Shvoeva Zh. Anal. Khim. 48 502 (1993) a 75. O P Shvoeva, V P Dedkova, S B Savvin Zh. Anal. Khim. 53 1184 (1998) a 76. O P Shvoeva, L M Trutneva, S B Savvin Zh. Anal. Khim. 49 574 (1994) a 77. O P Shvoeva, L M Trutneva, E A Likhonina, S B Savvin Zh. Anal. Khim. 46 1301 (1991) a 78. O P Shvoeva, V P Dedkova, S B Savvin Zh. Anal. Khim. 54 42 (1999) a 79. S B Savvin, L M Trutneva, O P Shvoeva, K A Efendieva Zh. Anal. Khim. 46 709 (1991) a 80.S B Savvin, T G Dzherayan, T V Petrova, A V Mikhailova Zh. Anal. Khim. 52 154 (1997) a 81. A V Mikhailova, N A Smirnova, in Sovremennye Problemy Teoreti- cheskoi i Eksperimental'noi Khimii (Tez. Dokl. Vseros. Konf.) [Current Problems of Theoretical and Experimental Chemistry (Abstracts of Reports of the All-Russian Conference)] (Saratov: Saratov State University, 1997) p. 171 199 82. S B Savvin, A V Mikhailova, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Moskva, 1998 [The XVIth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Moscow, 1998] Vol. 3, p. 221 83. R F Gur'eva, S B Savvin Usp. Khim. 67 236 (1998) [Russ. Chem. Rev. 67 209 (1998)] 84. R F Gur'eva, S B Savvin, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez.Dokl.), Moskva, 1998 [The XVIth Mende- leev Congress on General and Applied Chemistry (Abstracts of Reports), Moscow, 1998] Vol. 3, p. 78 85. R F Gur'eva, S B Savvin Zh. Anal. Khim. 52 247 (1997) a 86. D A Burylov, A A Evstrapov, E D Makarova, A O Petryakov, G E Rudnitskaya, A S Shestakova, V E Kurochkin Zh. Anal. Khim. 52 552 (1997) a 87. S B Savvin, O P Shvoeva, V P Dedkova, S V Rudakov, A V Mikhailova Zh. Anal. Khim. 51 308 (1996) a 88. V M Ostrovskaya Zh. Anal. Khim. 32 1820 (1977) a 89. M S Kravchenko, V F Osyka Unifitsirovannye Metody Issledova- niya Kachestva Vod. Testovye Metody Analiza (Unified Methods of Investigation of Water Quality. Test Methods) (Moscow: Council of Economical Mutual Assistance, 1990) 90.M S Kravchenko, V F Osyka, L N Yurchenko, M Sh Fumarova, in Analiticheskaya Khimiya Ob'ektov Okruzhayushchei Sredy (Tez. Dokl. Vsesoyuz. Konf.), S.-Peterburg, Sochi, 1991 [Analytical Chemistry of Environmental Objects (Abstracts of Reports of the All-Union Conference), St. Petersburg, Sochi, 1991] Vol. 3, p. 195 91. L V Boeva, V A Kimstach, T V Lifintseva Gidrokhim. Mater. 100 152 (1991) 92. V M Ostrovskaya Reaktivnye Indikatornye Sredstva (RIS) dlya Mnogoelementnogo Testirovaniya Vody [Reactive Indicator Means (RIM) for Multielement Testing of Water] (Moscow: IREA, 1992) 93. Yu A Zolotov, in Proceedings of International Trace Analysis Sym- posium 94 (The 7th Japan-Russia Joint Symposium in Analytical Chemistry), Hakodate, Sapporo, 1994 p.7 94. Yu A Zolotov Vestn. Ross. Akad. Nauk 61 (11) 63 (1991); 67 (6) 508 (1997) 95. Yu A Zolotov Zh. Anal. Khim. 49 149 (1994) a 96. V M Ostrovskaya Zh. Anal. Khim. 52 987 (1997) a 97. Yu A Zolotov Ann. Chim. 87 285 (1997) 98. Yu A Zolotov Zh. Anal. Khim. 51 1029 (1996) a 99. O P Shvoeva, V P Dedkova, A G Gitlits, S B Savvin Zh. Anal. Khim. 52 89 (1997) a 100. V M Ostrovskaya, in III Vseros. Konf. `Ekoanalitika-98' (Tez. Dokl.), Krasnodar, 1998 [The Third All-Russian Conference `Ecoanalytics-98' (Abstracts of Reports), Krasnodar, 1998] p. 150 101. V M Ivanov, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Moskva, 1998 [The XVIth Mendeleev Con- gress on General and Applied Chemistry (Abstracts of Reports), Moscow, 1998] Vol.3, p. 107 102. A V Mikhailova, N A Smirnova, in Problemy i Dostizheniya Lyuminestsentnoi Spektroskopii (Tez. Dokl. Vseros. Seminara), Saratov, 1998 [The Problems and Advances in Luminescence Spectroscopy (Abstracts of Reports of the All-Russian Seminar), Saratov, 1998] p. 22 103. V Amelin, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 1, K-19 104. V G Amelin Zh. Anal. Khim. 53 958 (1998) a 105. V G Amelin, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Moskva, 1998 [The XVIth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Moscow, 1998] Vol. 3, p. 6 106. V M Ostrovskaya, in International Congress on Analytical Chem- istry (Abstracts of Reports), Moscow, 1997 Vol.1, D-27 107. R P Pantaler, N B Lebed', L I Avramenko, A B Blank Zh. Anal. Khim. 52 643 (1997) a 108. R P Pantaler, A B Blank, L V Gudzenko, L A Vegorova, T I Ivkova,N B Lebed,A K Timchenko, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 2, K-8 109. V M Ostrovskaya, M S Aksenova, R L Medvedeva Zh. Anal. Khim. 42 2201 (1987) a 110. V Venkata-Ramana, K R Vamuna, I Jaunakais, in Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy (Abstracts of Reports), Chicago, 1994 367PS B Savvin, V P Dedkova, O P Shvoeva 200 111. R P Pantaler, A K Timchenko, L I Avramenko, A B Blank Zh. Anal. Khim. 52 384 (1997) a 112.J Siepak, H Witkowski Chem. Anal. 32 1035 (1987) 113. V M Ostrovskaya, O V Ivanov Vysokoch. Veshch. (4) 178 (1987) 149. V P Dedkova, O P Shvoeva, S B Savvin, in XVI Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Minsk, 1998 [The XVIth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Minsk, 1998] Vol. 3, p. 158 150. V P Dedkova, O P Shvoeva, S B Savvin Zav. Lab. Diagnost. Mater. 64 (12) 7 (1998) 151. V P Dedkova, O P Shvoeva, S B Savin, in Vseros. Konf. `Ekoanalitika-96' (Tez. Dokl.), Krasnodar, 1996 [The All-Russian Conference `Ecoanalytics-96' (Abstracts of Reports), Krasnodar, 1996] p. 107 114. R P Pantaler, L A Egorova, L I Avramenko, A B Blank Zh. Anal. Khim. 51 997 (1996) a 115. R P Pantaler, A K Timchenko, L I Avramenko, A B Blank Zh. Anal. Khim. 53 529 (1998) a 116. L V Gudzenko, R P Pantaler, L I Avramenko, A B Blank Zh. Anal. Khim. 51 993 (1996) a 117. M O Gorbunova, T V Knyazeva, V M Karpenko, in Vseros. Konf. `Ekoanalitika-94' (Tez. Dokl.), Krasnodar, 1994 [The All-Union a�Russ. J. Anal. Chem. (Engl. Transl.) b�Moscow Univ. Bull. (Engl. Transl.) c�Russ. Chem. Bull. (Engl. Transl.) d�Russ. J. Inorg. Chem. (Engl. Transl.) Conference `Ecoanalytics-94' (Abstracts of Reports), Krasnodar, 1994] p. 46 118. V M Ostrovskaya, in Vseros. Konf. `Ekoanalitika-94' (Tez. Dokl.), Krasnodar, 1994 [The All-Union Conference `Ecoanalytics-94' (Abstracts of Reports), Krasnodar, 1994] p. 81 119. J Matheson Water Cond. Purif. 36 (8) 90 (1994) 120. I Jaunakais, W Gaguon, in Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy (Abstracts of Reports), Atlanta, 1997 261P 121. V M Ostrovskaya, M S Aksenova, V F Osyka,M S Kravchenko, S P Bushtets Vysokoch. Veshch. (3) 152 (1992) 122. TNShekhovtsova, SVChernetskaya,NVBelkova, I F Dolmanova Zh. Anal. Khim. 50 538 (1995) a 123. T Shekhovtsova, I Veselova, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 2, P-18 124. T N Shekhovtsova, S V Chernetskaya, E B Nikol'skaya, I F Dolmanova Zh. Anal. Khim. 49 862 (1994) a 125. T Shekhovtsova, S V Muginova, N A Bagirova Mendeleev. Commun. 119 (1997) 126. T N Shekhovtsova, S V Muginova, N A Bagirova Anal. Chim. Acta 344 145 (1997) 127. T N Shekhovtsova, S V Chernetskaya Anal. Lett. 27 2883 (1994) 128. M K Beklemishev, T A Stoyan, I F Dolmanova Analyst 122 1161 (1997) 129. L V Gudzenko, R P Pantaler, L I Avramenko, A B Blank Zh. Anal. Khim. 53 1189 (1998) a 130. A Ishida, Yu Deduchi, E Kaneko, T Yotsuyanagi, in Proceedings of International Trace Analysis Symposium 94 (The 7th Japan-Russia Joint Symposiumin Analytical Chemistry), Hakodate, Sapporo, 1994 p. 253 131. R Kuroda, N Matsumoto Fresenius Z. Anal. Chem. 327 360 (1987) 132. R Parkash, R Bansal, A Kaur, S K Rehani Talanta 38 1163 (1991) 133. US P. 4 532 107; Ref. Zh. Khim. 7 G 140P (1986) 134. S A Morozko, V M Ivanov Zh. Anal. Khim. 51 631 (1996) a 135. V M Ivanov, S A Morozko, S V Kachin Zh. Anal. Khim. 49 857 (1994) a 136. V M Ivanov, O V Kuznetsova Zh. Anal. Khim. 50 498 (1995) a 137. V M Ivanov, S A Morozko, Sabri Massud Zh. Anal. Khim. 50 1280 (1995) a 138. V M Ivanov, Sabri Massud Vestn. Mosk. Univ., Ser. 2, Khim. 35 350 (1994) 139. V M Ivanov, S A Morozko Zh. Anal. Khim. 51 1083 (1996) a 140. V M Ivanov, N V Erokhina Vestn. Mosk. Univ., Ser. 2, Khim. 35 269 (1994) 141. E I Morosanova, I V Pletnev, V Yu Solov'ev, N V Semenova, Yu A Zolotov Zh. Anal. Khim. 49 676 (1994) a 142. I M Maksimova, E I Morosanova, A A Kukhto, N M Kuz'min, Yu A Zolotov Zh. Anal. Khim. 49 1210 (1994) a 143. I M Maksimova, A A Kukhto, E I Morosanova, N M Kuz'min, Yu A Zolotov Zh. Anal. Khim. 49 695 (1994) a 144. S A Mechkovskii, in XV Mendeleevskii S'ezd po Obshchei i Prikladnoi Khimii (Tez. Dokl.), Minsk, 1993 [The XVth Mendeleev Congress on General and Applied Chemistry (Abstracts of Reports), Minsk, 1993] Vol. 2, p. 312 145. S Dmitrienko, L Goncharova, E Myshak, L Pyatkova,AZhigulyev, V Runov, Yu Zolotov, in International Congress on Analytical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 2, K-5 146. W Frenzel Fresenius J. Anal. Chem. 361 774 (1998) 147. V P Dedkova, O P Shvoeva, in International Congress on Analyt- ical Chemistry (Abstracts of Reports), Moscow, 1997 Vol. 2, K-13 148. V P Dedkova, O P Shvoeva, S B Savvin Zav. Lab. Diagnost. Mater. 64 (3) 1

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (3) 201 ± 220 (2000) Naphthyridines. Structure, physicochemical properties and general methods of synthesis V P Litvinov, S V Roman, V D Dyachenko Contents I. Introduction II. Physicochemical properties III. The reactivity of naphthyridines IV. General methods for the synthesis of naphthyridines V. Conclusion Abstract. applica- and properties chemical synthesis, the on Data Data on the synthesis, chemical properties and applica- tions years 15 last the over mainly published naphthyridines of tions of naphthyridines published mainly over the last 15 years are are described systematically and analysed. The bibliography includes described systematically and analysed. The bibliography includes 238 238 references. references. I.Introduction Naphthyridines (pyridopyridines, diazanaphthalenes) represent a group of six isomeric heterocyclic systems containing two fused pyridine rings with different mutual arrangements of nitrogen atoms. They include two groups of compounds � N(1),N(i )- naphthyridines (i=5, 6, 7, 8) 1 ± 4 and N(2),N( j )-naphthyridines ( j=6, 7) 5, 6. N N N N N N 3 1 2 1,5-naphthyridine 1,8-naphthyridine 1,6-naphthyridine N N N N N N 6 4 5 1,7-naphthyridine 2,7-naphthyridine 2,6-naphthyridine The first derivative of the cyclic naphthyridine system was obtained in 1893 by Reissert,1 who proposed this name for the new class of heterocyclic compounds. The first representatives of unsubstituted naphthyridines � 1,5-naphthyridine 1 2 and 1,8- naphthyridine 2 3 �were described only in 1927.Naphthyridines containing nitrogen atoms at positions 1,6 (3),4 1,7 (4),5 and 2,7 (5) 6 were prepared in 1958, and 2,6-naphthyridine (6)7, 8 was synthesised only in 1965. Since then, researchers have shown ever increasing interest in the chemistry of naphthyridines. Indeed, the bibliography of a review 9 published in 1950 included 75 references, while that of a review 10 which appeared in 1970 V P Litvinov N D Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky prosp. 47, 117913 Moscow, Russian Federation. Fax (7-095) 135 53 28. Tel. (7-095) 135 88 37. S V Roman, V D Dyachenko T G Shevchenko Lugansk State Pedagogical Institute, ul. Oboronnaya 2, 348011 Lugansk, Ukraine. Fax (38-064) 255 33 68.Tel. (38-064) 253 83 94 (V D Dyachenko) Received 16 September 1999 Uspekhi Khimii 69 (3) 218 ± 238 (2000); translated by Z P Bobkova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069nABEH000559 201 201 202 207 218 comprised 242 references. More recent reviews 11, 12 (1983) present 223 studies. Several reviews dealing with more specific aspects were also published before 1983.13 ± 22 In recent years, the number of publications devoted to various aspects of naphthyridine chemistry has sharply increased. More than 1000 publications have appeared during the last 15 years, 40% of them being patents. This interest in naphthyridine derivatives is due to the excep- tionally broad spectrum of their biological activities.They are used for the diagnostics and therapy of diseases of humans including AIDS; for combating exo- and endo-parasites in agri- culture and in cattle breeding; as preservatives and ingredients of cutting fluids, as ligands in analytical chemistry, etc. This review mainly surveys the data published over the last 15 years; however, earlier fundamental studies are also invoked in the discussion of structures, physicochemical properties and reactiv- ities of naphthyridines. II. Physicochemical properties 1. Structure X-Ray diffraction study was first performed for unsubstituted 1,5- (1) and 2,6-naphthyridines (6).10, 23, 24 It was found that transition from pyridine to naphthyridines is accompanied by shortening and lengthening of C7C bonds approximately equal to that observed on passing from benzene to naphthalene.In addition, the N7C bonds in naphthyridines were found to be shorter than these bonds in pyridine. X-Ray diffraction analysis was also used to determine the structures of several naphthyridine derivatives, 1, 2, 4 and 6.25 ± 37 2. Quantum-chemical calculations Since the 1950s, quantum chemistry methods have been used to elucidate structure ± property relationships in the naphthyridine series.38 ± 57 The intensities of bands in the UV spectra of naph- thyridines have been found to be correlated with the energies of lower unoccupied molecular orbitals;38 ± 40 quantum-chemical calculations have also been used in the 1H and 13C NMR spectroscopy 47 ± 51 and for comparison of the total p-energy and delocalisation energy for naphthyridines and naphthalene.41, 43 Other results include determination of p-electron distribu- tion;43, 44 elucidation of correlations of the total p-electron energy with specific features of electrophilic 43 and nucleophilic 43, 44 substitution and with the hyperfine splitting constants in the EPR spectra.47 In recent years, quantitative structure ± biological activity relationships for 6-[2-(4-arylpiperazino)ethyl]-5,6,7,8-tetrahydro-202 1,6-naphthyridines have been established 53 and complexes of antitumour antibiotics � naphthyridinomycins � with DNA have been studied by molecular modelling methods.54 3.Spectroscopy Spectroscopic techniques (UV, IR, 1H and 13C NMR) are widely used not only to identify naphthyridines but also to elucidate fine points concerning their structures.10, 12 Thus it has been found that the UV spectra of 1,5-, 1,6-, 1,7- and 1,8-naphthyridines 1 ± 4 are fairly similar to one another and contain three separate groups of bands, analogous to the bands displayed in the spectra of quinoline and isoquinoline. However, the spectra of 2,7- and 2,6-naphthyridines 5 and 6 differ both from each other and from the spectra of any other naphthyridines.10, 58 The UV spectra of 4-hydroxy-1,5-naphthyridine 59 and of a large number of substituted 1,8-naphthyridines 60 have been analysed in detail.In order to disclose the influence of substituents on the absorption spectra, UV spectra of 1,8-naphthyridines monosub- stituted at position 2 (Me, Cl, Br, OMe, OEt, SMe, NH2 , OH, SH), position 3 (Me, Cl, Br, NO2, NH2) and position 4 (Me, Cl, Br, OMe, OEt, OH, SH) have been studied.61 Transmission and absorption spectra of 1,8-naphthyridine 2 in the range of 50 000 ± 20 000 cm71 have been investigated in solutions and in various mixed crystals at 4.2 and 300 K.52 The pattern of the band corresponding to the p?p* transition was found to be similar to that observed in the spectrum of the related naphthalene molecule with allowance for perturbations induced by the replacement of carbon by nitrogen. The position of the long-wavelength p?p* band at about 27 000 cm71 complies with the theoretical predictions based on extended HuÈ ckel calcu- lations.The bands at about 31 600 ± 31 900 cm71 were assigned empirically to the second n?p* transition.52 All bands in the IR spectra of several substituted 1,5-naph- thyridines have been assigned; the spectrum of 1,5-naphthyridine 1 was compared with the spectrum of 5-substituted quinoline.62 IR spectroscopy was used to study the keto ± enol tautomerism of 4-hydroxy-1,5-naphthyridine in solution and in the solid state; it was found that the keto form predominates in polar solvents and the enol form, in nonpolar solvents.59 Study of the IR and Raman spectra of 1,6- and 1,8-naphthyr- idines permitted the assignment of all 42 fundamental vibrations of these two heterocyclic systems.63 Data on the 1H NMR spectra (chemical shifts and spin ± spin coupling constants) for the naphthyridines 1 ± 4 and 6 can be found in the literature.7, 8, 10, 12, 42, 43, 49 ± 51, 64 ± 77 The chemical shifts observed in the 13C NMR spectra of the naphthyridines 1 ± 6 were compared with those for quinoline and isoquino- line.12, 49 ± 51 The resonance signals in the 1HNMR spectrum of 2-(20-pyridyl)-1,8-naphthyridine were completely assigned using homonuclear 2D NMR spectroscopy.75 The behaviour of derivatives of the isomeric naphthyridines 1 ± 4 under electron impact has been stthe main fragmentation pathways have been identified.10, 78 ± 80 The struc- tures and destruction pathways of the isomeric ions [M7HCN]7.and [M72 HCN]7., resulting from fragmentation of the naph- thyridines 1 ± 4 and benzazines, were studied using collisional activation.81 It was found that the [M72 HCN]7.ions can exist as two isomers, one of them being characteristic of naphthyridines and the other being characteristic of benzazines. High-resolution photoelectron spectra of the naphthyridines 1 ± 6 have been described.82 The ionisation potentials for the n and p orbitals are listed in Table 1. Polarised phosphorescence spectra of 1,5-naphthyridine 1 83 ± 85 and the EPR spectrum of 1,6-naphthyridine 3 were recorded.86 A number of publications deal with the capacity for ionisation (including the pKa values) of the naphthyridines 1 ± 4 and their derivatives.10, 58, 70, 87 ± 91 Table 1. Ionisation potentials (eV) of naphthyridines 1 ± 6.82 Compound 1,5-Naphthyridine 1 1,8-Naphthyridine 2 1,6-Naphthyridine 3 1,7-Naphthyridine 4 2,7-Naphthyridine 5 2,6-Naphthyridine 6 III.The reactivity of naphthyridines Reactions of naphthyridines including electrophilic substitution (bromination, nitration), nucleophilic substitution (amination, metallation), reduction, oxidation, complexation, etc., have been considered fairly comprehensively in reviews. 10, 12 In yet another review, attention is focussed on the reactivity of naphthyridines towards N-nucleophiles.11 Here we survey the data on the reactivity of naphthyridines published over the last 15 years. Electrochemical reduction of the naphthyridines 1 ± 6 has been reported.92, 93 On exposure to UV radiation in the presence of sodium nitrite, 3-amino-1,6-naphthyridine-4(1H)-one eliminates a nitrogen molecule being thus converted into 5-azaindole-3- carboxylic acid.94, 95 Nucleophilic substitution of bromine at position 4 in 3,4-dibromonaphthyridines 7a ± c on treatment with a saturated alcoholic solution of ammonia in an autoclave affords 4-amino-3- bromo-N(1),N(i )-naphthyridines 8a ± c.96 Br N(i ) N 7a ± c i = 5 (a), 6 (b), 7 (c).The naphthyridine derivatives 9a ± c, possessing fungicidal activity, have been synthesised using nucleophilic substitution of the halogen atoms in isomeric halo naphthyridines. These prod- ucts can also be employed for combating exo- and endo-parasites in agriculture and cattle breeding, for wood protection, and as preservatives for paints, varnishes and cutting fluids during metal working.97 XYR N(i) N 9a ± c i = 5 (a), 6 (b), 7 (c); X =O, S, SO, SO2; Y is alkenyl or a single bond; R is cycloalkyl or piperidin-2-yl.In recent years, nucleophilic addition reactions have been widely used for the synthesis of various derivatives of isomeric naphthyridines. For example, nucleophilic addition of indole to 1,5-, 1,6- and 1,8-naphthyridine derivatives in the presence of V P Litvinov, S V Roman, V D Dyachenko n-Orbital 9.20 10.40 1.20 9.20 10.10 0.90 9.50 9.90 0.40 9.30 10.00 0.70 9.35 10.10 0.75 9.40 10.00 0.75 Br NH3, EtOH 4±6 h, 130 ± 150 8C p-Orbital 11.05 11.33 9.07 11.10 8.99 11.14 8.98 8.87 NH2 Br N(i ) N 8a ± c203 Naphthyridines. Structure, physicochemical properties and general methods of synthesis CN N CN N + Hal¡3Hal2 S N S N 25 CHCl3 Ha Hb Hal C C Hc Hal Hal29 27 benzoyl chloride in toluene or DMF yields, depending on the temperature and reactant ratio, hydrogenated mono- (10 ± 12) or bis-(3-indolyl)naphthyridines 13 ± 18.98 ± 100 Hetarylation includes the in situ formation of N-acyl heteroaromatic cations and the addition of nucleophiles to them.Aromatisation of theN-acylated hydrogenated naphthyridine derivatives 13 and 14 on treatment with tetrachloroquinone affords the same compound, naphthyr- idine 19. Hal=Br, I. R N R Bz N CN N I2 R N N R N + 26 I¡3N CHCl3 12 11 10 Bz Bz S N R 28 ICH2 R R HN HN HN R R R N 15 H N 14 H N 13 H R R N R R R R N N NH NH NH Bz Bz Bz 17 18 16 N R R is 3-indolyl.N R 19 X-Ray diffraction analysis has shown that the pyridine nitro- gen atom in the compound 25 has a short non-valence contact with the C=C bond of the cyclohexene fragment. The close arrangement of these reaction sites is favourable for synchronous cyclisation in intermediate 29 involving the Ha, Hb and Hb, Hc systems of protons. The high stereoselectivity of this reaction is apparently due to the synchronism of the action of the donor (i.e. the lone electron pair of the naphthyridine nitrogen atom) on the p-electrons of the cyclohexene double bond and electrophilic rupture of the double bond, which are followed by quaternisation.Thus the Hb and Hc protons occupy trans-positions. This intra- molecular electrophilic quaternisation of the compound 25 can be regarded as synchronous trans-quaternisation with cis-annela- tion.108 Refluxing of N(1),N(i )-naphthyridin-4(1H)-ones 20a ± d in DMSO gives rise to 1-methylthiomethyl naphthyridine deriva- tives 21a ± d, which depress microbial growth.100 ± 103 Similar methylthiomethylation of naphthyridin-2(1H)-ones 22a ± d has resulted in N-substituted N(1),N(i )-naphthyridin-2(1H)-ones 23a ± d.104 2-Cyanomethylthio derivatives 30, prepared by alkylation of the compounds 24 with chloroacetonitrile, cyclise under condi- tions of the Thorpe ± Ziegler reaction to give the corresponding thieno[2,3-b]pyrido[3,2-e]quinuclidines 31, containing a 1,5-naph- thyridine fragment.107 O O R R2 R2 CN N DMSO R1 R1 a b, c D, 8±36 h 24 N N N(i ) N(i ) HN SCH2CN 30 20a ± d 21a ± d CH2SMe R R NH NH2 DMSO N N N(i ) N(i ) D, 8 h CN CN O N O S S N N N 22a ± d H 23a ± d CH2SMe 31 i = 5 (a), 6 (b), 7 (c), 8 (d); R1=H, Me; R2=H, CO2H, CO2Et.R=H, Ph, 4-FC6H4; (a) ClCH2CN, KOH, DMF, 25 8C, 3 h; (b) OH7; (c) H+. Similar treatment with sodium ethoxide of thioethers 32, prepared by the reaction of 1,2,5,6,7,8-hexahydro-1,6-naphthyr- idine-2-thiones 33 with halo ketones, esters, amides or nitriles, resulted in the synthesis of thieno[2,3-b]-1,6-naphthyridines 34.109 Ph Ph Regioselective alkylation of quinuclidinopyridine-2(1H)-thi- ones 24, containing a 1,5-naphthyridine fragment, with 3-bromo- cyclohexene or allyl bromide in the presence of a base affords the corresponding 2-(cyclohex-2-enylthio)- (25) and 2-allylthio-deriv- atives 26.Halogen-induced quaternisation of these products is a synchronous process yielding salts 27 and 28 with high regio- and stereoselectivity.105 ± 108 CN N CN CN a PriN PriN RCH2Hal EtONa R S N N S SCH2R CN N NH 25 OH7 32 33 Ph Ph Ph R NH2 S 24 HN CN N PriN b R S N N CH2 SCH2CH 26 Br 34 Ph (a) , R =H; (b) CH2=CHCH2Br, R=H, Ph, 4-FC6H4. R=MeCO, PhCO, EtO2C, CONH2, CONHPh, CN, CONHC6H4Me-4, CONHC6H4OMe-4, CONHC6H4COMe-4.204 The reaction of 6,7,8a-trimethyl-1-oxo-3-thio-1,2,3,5,6,7,8,8a- octahydro-2,7-naphthyridine-4-carbonitrile (35) with 4-bromophe- nacyl bromide or N-cyanochloroacetamidine in the presence of a base gives thieno[2,3-c]-2,7-naphthyridines 36.110 Me Me MeN MeN NH2 CN+HalCH2Z KOH DMF Me Me Z S N O S O NH 36 35 Hal=Br, Z=4-BrC6H4CO; Hal=Cl, Z=C(NH2)NCN.Cyclocondensation of 3-ethoxycarbonylmethylthio-2,7-naph- thyridine 37 on heating in anhydrous ethanol in the presence of an equimolar amount of sodium ethoxide affords thienonaphthyr- idine 38. The dihydrochloride of the compound 38 possesses a mutagenic activity.111 Me Me Me Me MeN MeN NH2 CN EtONa 55 ± 60 8C, 2 h CO2Et S Mor Mor SCH2CO2Et N38 N37 Mor is morpholin-1-yl. Treatment of 1-methyl-N(1),N(i )-naphthyridinium iodides 39a,b with liquid ammonia in the presence of potassium perman- ganate gives rise to mixtures of imino (40a,b) and oxo (41a,b) derivatives of dihydronaphthyridines and vicinal methylamino- pyridinecarboxamides 42a,b.112 X NH3 KMnO4 I7 N + Me Y 39a,b X X X CONH2 + + NHMe O NH NMe NMe Y 42a,b Y 41a,b Y 40a,b X=N, Y=CH (a); X=CH, Y =N (b).Unlike compounds 39a,b, 7-methyl-1,7-naphthyridinium iodide 43 and 6-methyl-1,6-naphthyridinium iodide 44 react with liquid ammonia and KMnO4 with ring contraction as the pre- dominant reaction pathway yielding 4-azaindole derivatives 45 ± 48 together with the oxo derivatives of the corresponding dihydronaphthyridines 49 and 50.Study of the 1H NMR spectra showed that on treatment with liquid ammonia in the absence of potassium permanganate, the compounds 43 and 44 are converted into amino derivatives 51 and 52.113 a, b I7 + MeN N O NH2 43 N N NMe + NMe + MeN N O O O 46 45 49 V P Litvinov, S V Roman, V D Dyachenko MeN+ a, b I7 N 44O N MeN + N 50 c 43 MeN N NH2 51 NH2 MeN c 44 N 52 (a) NH3 (liq), KMnO4,733 8C; (b) OH7; (c) NH3 (liq). When 3,6-dinitro-1,8-naphthyridine 53 or 6-R-1-ethyl-3- nitro-1,8-naphthyridin-2(1H)-ones 54 are made to react with a solution of potassium permanganate in liquid ammonia at 7338C, amination proceeds giving rise to the corresponding 4-amino derivatives 55 and 56. The intermediate formation of s-adducts 57 and 58 was detected by 1H NMR spectro- scopy.114, 115 It was also shown that 1-ethyl-3-nitro-1,8-naphthyr- idin-4-one is not aminated under these conditions.115 However, treatment of 1,8-naphthyridine derivatives 59a,b with liquid ammonia at 745 8C affords 4-amino-1,4-dihydro-3,6-dinitro-2- R-1,8-naphthyridines 60a,b.The compound 59c was converted into a mixture of the naphthyridine 60c and isomeric 5-amino-2- ethoxy-3,6-dinitro-5,8-dihydro-1,8-naphthyridine 61.116 NO2 O2N NH3 (liq) 733 8C N N 53 + H NH3NO2 O2N 7 N N 57 R NO2 NH3 (liq) 733 8C O N Et N 54 + H NH3NO2 R KMnO4 7 OH N NEt 58 R=H, NO2. NO2 O2N NH3 (liq) 745 8C R N N 59a ± c O O N +49 NMe NMe+ 48 47 NH NH2 NH2 NO2 O2N KMnO4 N N 55 NH2 R NO2 O N Et N 56Naphthyridines.Structure, physicochemical properties and general methods of synthesis H NH2NO2 O2N R=OH, NH2 R HN N 60a,bH NH2 H NH2 NO2 NO2 O2N O2N R=OEt + OEt N OEt NH HN N60c 61 R=OH (a), NH2 (b), OEt (c). 2-Nitro-N(1),N(i )-naphthyridines 62a ± c undergo replace- ment of the nitro group rather than amination on treatment with a solution of potassium permanganate in liquid ammonia; this gives the corresponding 2-aminonaphthyridines 63a ± c, although their yields are relatively low. These products react with DMSO and trifluoromethanesulfonic anhydride giving rise to S,S- dimethyl-[N(1),N(i )-naphthyridin-2-yl]sulfylimines 64a ± c.115 a b N(i ) N(i ) NH2 NO2 N 63a ± c N 62a ± c N(i ) N SMe2 N 64a ± c i=5 (a), 6 (b), 8 (c); (a) NH3 (liq), KMnO4,733 8C; (b) DMSO, (CF3SO2)2O.The kinetics of N-methylation of benzonaphthyridines 65a ± d with methyl iodide in DMSO have been studied.117 It was shown that the salt 66, formed from the naphthyridine 65b, slowly dimerises in a 10% aqueous solution of DMSO to give compound 67. MeN+ a b N(i ) I7 N N66 65a ± d Me Me MeN+ N 2 I7 Me MeN+ N 67 Me i = 5 (a), 6 (b), 7 (c), 8 (d); (a) i=6, MeI, MeCO2Et, D, 0.5 h; (b) DMSO±H2O (1 : 9), 10 h. The methyl group in benzo[2,3]naphthyridines 65a ± d can be easily modified. Thus refluxing of these compounds with SeO2 in chloro-, 1,2-dichloro- or 1,2,4-trichlorobenzene gives aldehydes 68a ± d, which serve as the starting compounds for the synthesis of diverse functional derivatives of benzonaphthyridines.118 ± 120 Oxidation of the compounds 65a ± d to the aldehydes 68a ± d and further oxidation to the corresponding acids are the key stages of the synthesis of compounds 71 ± 75, which are the aza analogues of the potential antitumour preparation, N-[2-(dimethylamino)- ethyl]acridine-4-carboxamide.118 ± 120 SeO2 N(i ) 65a ± d N 68a ± d CHO O a N(i ) HN 69b COOH b or c N(i ) N 70a,c COOH O d + N(i) N HN 69d COOH i = 5 (a), 6 (b), 7 (c), 8 (d); (a) i=6, NaClO2; (b) i =7,Ag2O; (c) i=5; (d ) i=8, NaClO2.The aldehydes 68a ± d can be oxidised using Ag2O, NaClO2, K2Cr2O7, etc.; oxidation can proceed to different extents depend- ing on the structure of the initial compound and the reagent employed.For instance, oxidation of the aldehyde 68b with NaClO2 involves the central ring and gives oxo acid 69b. The oxidation of the aldehyde 68c with Ag2O affords acid 70c as the only product. The aldehyde 68a is oxidised by Ag2O to 70a but the reaction does not proceed to completion, up to 10% of the starting compound being recovered unchanged. The reaction of the aldehyde 68d with NaClO2 yields a mixture of acids 69d and 70d. The oxo acids 69a ± d are also produced upon oxidation of 68a ± d with NaClO2 in ButOH. Further modification of the acids 69 and 70 includes reactions of the carboxy group with 2-(N,N-dimethylamino)ethylamine and nitration of the benzene ring.Thus compounds 74 and 75 were prepared by oxidation of the methyl group in nitro derivatives 76 followed by modification of the acids 77 and 78 formed (Scheme 1). Dipolar 1,3-cycloaddition of various dienophiles to ben- zo[3,4]naphthyridinium ylides 79 ± 81, generated in situ by dehy- drobromination of the corresponding quaternary salts, has been studied; this gives rise to benzo[h]pyrrolonaphthyridines 82 ± 88 (Scheme 2).121 ± 126 The compounds 82 ± 84 exhibit bactericidal activities.124, 125 The dipolar 1,3-cycloaddition of N-dichloromethylides 89 ± 91, generated in situ from the corresponding benzo[h]naph- thyridines 92 ± 94 and dichlorocarbene 127 (produced from CCl3CO2Na on heating or from CHCl3 in the presence of a base and Et3BnN+Cl7), to dimethyl acetylenedicarboxylate is less selective; only the ylide 89 is converted into the corresponding adduct 95 in 52% yield.In the case of 90 and 91, the reaction gives a mixture of dipolar 1,3-addition products 96 and 97 and bisadducts 98 and 99 (Scheme 3). Quaternisation of naphthyr- idines 92 ± 94 128 ± 130 (the compounds 92 and 93 were prepared by the Skraup method 128 and 94 was synthesised by photocyclisation of diazastilbene 131) induced by alkyl, aryl or benzyl halides gave 205 N 70d COOH206 O N a HN 71 69b O N b, a HN 72 a 70c N N 73 N(i ) a 65a ± d N 76a ± d NO2 N(i ) c N 74a ± d CONH(CH2)2NMe2 O NO2 N(i ) d NH 78a ± d COOH (a) KNO3, H2SO4; (b) K2Cr2O7, H2SO4; (c) Me2N(CH2)2NH2; (d ) NaClO2 .quaternary salts 100 ± 102. The salts 101, 102 exhibit bactericidal and fungicidal activities.128 N X7 +NR 100 R=2,4-(NO2)2C6H3, H2C CHCH2, Bn, Me(CH2)9; X=Cl, I. In order to elucidate the influence of the structure of benzo[h]- naphthyridines on their reactivity, charge distributions, excitation energies, and the oscillator strengths for the molecules 92 ± 94 and their amino derivatives have been calculated. The calculation was carried out by a semiempirical self-consistent field method in the Pariser ± Parr ± Pople p-electron approximation with allowance for the configurational interaction between singly excited states and by the HuÈ ckel method.The results are in satisfactory agree- ment with the data of the UV spectra.132 The populations of the atomic p-orbitals, the order of p-bonds and the frequency and oscillator strengths for three electron transitions in the molecules of the naphthyridines 92 ± 94 and their salts 100 ± 102 were also calculated. By comparing the calculated and experimental spectra, correlation coefficients for the relation ncalc=anexp+b were determined. 133 CONH(CH2)2NMe2 NO2 CONH(CH2)2NMe2 CONH(CH2)2NMe2 NO2 N(i ) b N 77a ± d Me O NO2 N(i ) c HN 75a ± d CONH(CH2)2NMe2 N X7 101 102 N + R Scheme 1 NO2 COOH NX7 NR + N + N 7CHR1 79 a N b +N 7 80 CHR1 c N b N + 7 81 CHCO2Et R1=COPh, CO2Et; R2=CN, CO2Et, Ac; O; (b) CH2 CHR2; (c) MeO2CC CCO2Me.(a) O O The formation ofN-ylides of the benzonaphthyridinium series upon dipolar 1,3-addition of benzonaphthyridine N-oxides to dimethyl acetylenedicarboxylate at room temperature has been reported.134 Treatment ofN-oxides 103a,b with POBr3 affords a mixture of 1,5- (1) or 1,8-naphthyridines (2) with their brominated deriva- V P Litvinov, S V Roman, V D Dyachenko Scheme 2 N a N R1 82 N b N R1 83 R2N c N R1 MeO2C CO2Me 84 NN 85 R1N R2 N 86 R1N CO2Me N R1 CO2Me 87 NH N R2 H 88 EtO2CNaphthyridines. Structure, physicochemical properties and general methods of synthesis 92 93 94 tives.A similar mixture of reaction products is also formed when POBr3 reacts with 1,5-naphthyridine bis(N-oxide).135 NO N(i ) 103a,b i = 5 (a), 8 (b). Carboxylic acid derivatives 104 of the pyrrolonaphthyridine series � potential preparations for treatment of degenerative, ischemic, and autoimmune diseases � have been synthesised using intramolecular cyclisation of ethyl (3-ethoxycarbonylnaph- thyridin-4-ylamino)acetates 105 induced by potassium tert-but- oxide in a toluene ±ButOH mixture. 136 N(i ) (a) ButOK, PhMe, ButOH, 20 h. The reaction of naphthyridinecarboxylic acids 106 with gua- nidine in the presence of carbonyldiimidazole in anhydrous THF results in hetaroylguanidines 107, which can be used for the therapy of heart diseases, for surgical operations and organ transplantation, for the diagnostics and treatment of hypertension and proliferative diseases.137 R2 R1 YR4 N(i ) 106 R1, R2=H, Hal, Alk, alkenyl, alkynyl, CN, AlkF; R3, R4=H, Hal, Alk, AlkF, AlkO, Alk2N, CON=C(NH2)2 or none; X=N,Y=C; X = C, Y = N.N N MeO2CC CCO2Me CCl2 + N N 7CCl2 89 N N MeO2CC CCO2Me CCl2 +N N 7 90 CCl2 N N MeO2CC CCO2Me CCl2 +N N 7 91 CCl2 Br Br POBr3 1 or 2+ + N N N(i ) N(i ) CO2Et NHCH2CO2Et HN CO2Et a OH N(i ) N N 105 104 R2 NH2 R1 HN XR3 NH2 XR3 NH2 C(O)N CO2H YR4 N(i ) NH2 107 207 Scheme 3 N N Cl MeO2C CO2Me 95 N N + CO2Me CO2Me N N CO2Me MeO2C Cl CO2Me 98 96 CO2Me N N + CO2Me CO2Me N N CO2Me MeO2C Cl CO2Me 99 97 CO2Me IV.General methods for the synthesis of naphthyridines General synthetic methods used to prepare various types of naphthyridine include the Skraup, Friedlander and some other name reactions; cyclisation, cyclocondensation, dimerisation reactions, etc. 1. The Skraup method and its modifications The Skraup reaction was first employed to synthesise unsaturated 1,5-naphthyridine 1 from 3-aminopyridine in the late 1920s.2, 138, 139 Subsequently this reaction and its modifications have been successfully used to prepare other naphthyridines and their derivatives.9, 10, 12 Recently 1,6-naphthyridine 3 has been prepared by the Skraup method, namely, by heating (135 8C, 48 h) of 4-aminopyridine with glycerol, fuming sulfuric acid and nitrobenzene.140 CH O H2SO4 CHCHO CH2 N N NH2 NH HO H N N N PhNO2 7H2O N 3 HN HN In recent years, a modified Skraup reaction (20% oleum, nitrobenzene, FeSO4 .7H2O, H3BO3) has been used to synthesise fused naphthyridines.Thus 3-aminobenzo[ f ]quinoline 108 was converted in this way into naphtho[2,1-b]-1,5-naphthyridine 109.141 Similarly 2-aminobenzo[h]quinoline 110 was transformed into a mixture of naphthonaphthyridines 111 and 112 (yields 6.8% and 4.5%, respectively), and 3-aminobenzo[g]quinoline 113 gave rise to naphthonaphthyridine 114 (yield 4.2%). The naphtho- naphthyridines 111 and 114 were also prepared from hydro- genated benzoquinolines 115 and 116. Intermediates 117 and 118 were dehydrogenated by treatment with dichlorodicyanoquinone and palladium supported on carbon, respectively.In this case, the overall yields of products were substantially higher.142208 H2N N108 NH2 N 110 N 111 (6.8%) NH2 N 113 NH2 N 115 N 117NH2 N 116 118 A modified Skraup method, namely, the reaction of 4-amino- isoquinoline 119 with methyl vinyl ketone in the presence of As2O5 and concentrated sulfuric acid, was used to prepare 4-methylben- zo[c]-1,5-naphthyridine 120 in 36% yield.143 A similar reaction of 3-aminoquinoline 121 with crotonaldehyde or methyl vinyl ketone results in 2-methyl- (122) or 4-methyl-benzo[ f ]-1,7-naphthyridine 123. Oxidation of the compounds 120, 122 and 123 on treatment with SeO2 in dioxane yields aldehydes 124 ± 126.NH2 a As2O5 N 119 N CH2=CHCHO 130 8C, 5 h N 109 (12.5%) CHCHO CH2 N N N + 112 (4.5%) N CHCHO CH2 N 114 (4.2%) CHCHO CH2 O Cl NC Cl NC N O 111 (24%) CHCHO CH2N Pd/C 114 (20%) N N N SeO2 N N 124 120 CHO Me N121 (a) CH2=CHCOMe; (b) MeCH=CHCHO. 2. The Friedlander reaction in the synthesis of naphthyridines The condensation of vicinal aminopyridinecarbaldehydes 127 with malononitrile in ethanol in the presence of a base (according to the Knoevenagel reaction pattern) furnished 2-amino-3-cyano- N(1),N(i )-naphthyridines 128a ± d; some of them were found to be effective diuretics.144, 145 The condensation of the aldehydes 127 with creatinine in ethylene glycol affords imidazo[4,5-b]- N(1),N(i )-naphthyridines 129a ± d.146 The reaction of the com- pounds 127 with 2-methylcyclohexanone in ButOH in the pres- ence of ButOK (1.5 h, refluxing) followed by dehydrogenation of the products 130a,c,d thus formed (refluxing with Pd/C in Ph2O) gives rise to benzo[b]-N(1),N(i )-naphthyridines 65a,c,d.Using the Friedlander reaction, naphthyridines 131 have been synthesised; they are active components of herbicide formulations.147, 148 Benzo[b]-1,6-naphthyridine 65b has been prepared by condensa- tion of the amino acid salt 132 (obtained from 7-methylisatin 133) CHO N127 i = 5 (a), 6 (b), 7 (c), 8 (d). 127 O NH (a) NH ; (b) Me N V P Litvinov, S V Roman, V D Dyachenko b SeO2 N N CHO N Me N 122 (21%) 125 CHO Me NH2 a SeO2 N N N N 123 (40%) 126 CN CH2(CN)2 NH2 CN NH2 N(i ) CN CN NH NH2 HN N N(i ) 128a ± d N(i ) NMe a NH2 N N(i ) N 129a ± d Pd/C b N N N(i ) N(i ) 65a,c,d Me 130a,c,d Me CO2Et c N CO2Et N(i ) 6 stepsCO2Et HN O N N(i ) 131 N ; (c) EtO2CCH2C(O)CO2Et.O MeNaphthyridines. Structure, physicochemical properties and general methods of synthesis with N-benzylpiperidone (KOH, 0.5 h, <40 8C; refluxing, 20 h) followed by dehydrogenation of the reaction product 134.117 ± 119 CO2K O NBn O KOH O O NH2 HN Me Me 132 133 CO2H N NBn 10% Pd/C Ph2O, D N N Me 65b 134 Me 3. Condensation of aminopyridines with diethyl ethoxymethylidenemalonate As in the previous cases, the use of condensation of amino- pyridines with diethyl ethoxymethylidenemalonate for the syn- thesis of naphthyridines attracts attention because it provides the possibility of synthesising diverse naphthyridines exhibiting a broad range of biological activities.Thus the condensation of substituted aminopyridines 135 with diethyl ethoxymethylidene- malonate 136 and subsequent cyclisation on refluxing of the resulting diesters of N-(pyridyl)aminomethylidenemalonic acids 137 gives derivatives of naphthyridinecarboxylic acids 138, pos- sessing antibacterial activity or serving as intermediates in the synthesis of compounds with this type of activity.130, 149 ± 161 EtO O C CO2Et R1nR1nCO2Et D + EtO RN2 CO2Et R2HN N 135 136 N 137 O7 EtO O R1nH R1n CO2Et CO2Et 7EtOH + RN2 N(i ) NR2 N(i ) 138 n=1, 2; R1=Me, EtO, EtS, F, 4-pyridyl; R2 =H, cyclopropyl.Naphthyridines 138 (R2=H) without substituents at nitro- gen have served as the starting compounds en route to pyra- zolo[3,4-c]naphthyridines 139, which act as modulators of benzodiazepine receptors and possess sedative and antispasmodic activities.162 ± 164 The reactions of chloronaphthyridines 140 with amino acid amides afford compounds 141, which possess anxio- lytic, anticonvulsive, sedative and hypnotic activities, antimicro- bial and analgesic properties; they are capable of reducing blood pressure and are used as pharmaceuticals.165 ± 167 OH Cl R1n R1nCO2Et CO2Et R2=H a b 138 N N N(i ) N(i ) 140 HN NR3 N NR3 R1n R1nO O N 139 HN N(i ) N(i ) n=1, 2; R1=Me, MeO; R2=Ph, pyridyl, quinolyl, isoquinolyl, pyrimidyl, thiazolyl; (a) (COCl)2, MeCN, DMF, from730 to 20 8C; (b) R3NHNH2, xylene, D.209 R4 R5NCH(CH2)mCONR2R3 R1nCO2Et a 140 N N(i ) 141 (a) R2R3NC(O)(CH2)mCH(R4)NHR5; n=1, 2; m=0±2; R1=Hal, Alk, AlkO, NO2, CF3; R2, R3=H, Alk, Ph, Bn; R27R3=(CH2)x (x=4 ± 8); R4=H, Alk, Bn; R5=H, Alk. The condensation of diethyl ethoxymethylidenemalonate 136 with 5-amino-2,3-dihydrofuro[3,2-b]pyridine 142 154 or 7-amino- oxazolo[4,5-b]pyridine 143 168 results in annelated naphthyridines 144 and 145. 2,3-Dihydrofuro[3,2-b]-1,8-naphthyridine 144 shows antibacterial activity.154 O O C(CO2Et)2 136 Ph2O, D N N NH2 HN 142 OH CO2Et O N N 144 CO2Et NH2 NHCH O 136 CO2Et Ph2O, D O NEt2 NEt2 N N N 143 N OH CO2Et N N N O 145 Et2N 4.Condensation of aminopyridinecarboxylic acids and their derivatives with compounds containing an active methylene group The condensn of vicinal aminopyridinecarboxylic acids or esters with ethyl hexanoate 146 in the presence of ButOK under a nitrogen atmosphere gave N(1),N(i )-naphthyridin-2(1H)-ones 147a ± c, exhibiting antiallergic activities.169 OH CO2R Bu N(i ) +BuCH2CO2Et 146 O N NHPh N 147a ± c Ph R=H, Me, Et; i = 5 (a), 7 (b), 8 (c). Naphthyridones 148, which are of interest as pharmacological preparations, have been synthesised by the reaction of vicinal aminopyridinecarbonitriles 149 with diethyl malonate.Alkylation of these products with ethyl monobromoacetate results in N-sub- stituted compounds 150.170 NH2 NH2 CN CO2Et CO2Et b a O N O N NH2 NH N(i ) N(i ) 149 148 CH2CO2Et 150 (a) CH2(CO2Et)2, EtONa; (b) BrCH2CO2Et, BuOH, K2CO3; i =5±8.210 When diethyl malonate is made to react with ethyl 3-amino- picolinate, substituted 1,5-naphthyridine 151 is formed; it is used in the synthesis of potential antimalarial remedies.171 NH2 EtONa +CH2(CO2Et)2 N CO2Et 1,8-Naphthyridine-3-carboxamides 152 � starting com- pounds for the preparation of products with antihypertensive, antiphlogistic and antiaggressive activities � have been synthes- ised from 2-aminonicotinic acid and amido esters 153.The syn- thesis includes the formation of the chloride and a salt of 2-aminonicotinic acid upon the reaction of 153 with POCl3 . The subsequent reaction of these intermediates results in the C-acy- lated product, the condensation of which gives the compound 152.172 CO2H+R2NC(O)CH2CO2Et N 153 NH2 Cl O + POCl3 NR2 N CO2Et NH2 R2=Me2, Et2, Pri2, (CH2)5. 5. Acylation of a-hetarylacetonitriles with halo-substituted nicotinic and isonicotinic acid chlorides Fused azahetareno-1,7-naphthyridines 154a ± 158a, containing a nitrogen atom in the bridgehead position, were obtained by the reaction of 3-bromoisonicotinic acid chloride 159 with 2-cyano- methyl derivatives of pyridine, 1-methylbenzoimidazole, benzo- 1,3-thiazole, quinoline or 4-methyl-1,3-thiazole followed by cyc- lisation of the resulting C-acylated derivatives on heating to the melting point or on refluxing in N-methyl-2-pyrrolidone (as shown for compound 154a taken as an example).173 N N CN N(i ) O N(i ) 154a,b O 155a,bMeN N CN N(i ) N(i ) O O 158a,b 157a,b i = 7 (a), 6 (b).COCl Br + N CH2CN N 159 OH N CO2Et OH N 151 POCl3 Cl CONR2 O N N 152 H NMe S N CN CN N(i ) O 156a,b SCN CN NH O N Br V P Litvinov, S V Roman, V D Dyachenko CN N 154a HO N Br A similar reaction of 2-cyanomethyl derivatives of hetarenes with 4-chloronicotinoyl chloride yielded annelated azahetareno- 1,6-naphthyridines 154b ± 158b.173 In this case, the intermediate C-acylated derivatives were not isolated because they easily cyclised to give 154b ± 158b.4,6-Dichloronicotinoyl chloride was used as the initial compound in this synthesis.174 The product of interaction of 2-cyanomethylbenzo-1,3-thia- zole with 2-chloronicotinoyl chloride cyclises on refluxing in DMF to give benzothiazolo[3,2-a]-1,8-naphthyridin-4-one deriv- ative 156c.173 COCl N + CH2CN S Cl N S CN NH N OCl CN SN D N H OCl 6. Photocyclisation of aminopyridines with chlorides of ortho-chlorinated hetarenecarboxylic acids The reaction of 3-chlorobenzo[b]thiophene-2-carboxylic acid chloride 160 with 2-, 3- or 4-aminopyridine 161 in benzene in the presence of triethylamine has resulted in the synthesis of amides 162, which undergo photocyclisation (UV irradiation by a medium-pressure mercury lamp, 450 W), giving rise to naphthyr- idines 163.175 Deoxygenation of these products on treatment with POCl3 followed by dechlorination of compounds 164 results in [1]benzothieno-N(1),N(i )-naphthyridines 165a ± d.H2N Cl Et3N + COCl S N 161 160 Cl hn S CONH N 162 N(i ) POCl3 NH S 163 O S N N CN 156c ONaphthyridines. Structure, physicochemical properties and general methods of synthesis N(i ) Pd/C N S 164 Cl i = 5 (a), 6 (b), 7 (c), 8 (d). Similarly, the reaction of chloride 160 with 4-amino-2-methyl- quinoline gave 1-methylbenzo[h][1]benzothieno[2,3-c]-1,6-naph- thyridine 166.176 NH2 a, b, c, d +160 Me N (a) Et3N; (b) hn; (c) POCl3; (d ) Pd/C.Photocyclisation is also one of the stages in the synthesis of naphthothieno-N(1),N(i )-naphthyridines 167a ± c and 168 from the chlorides of vicinal chloronaphthothiophenecarboxylic acids 169a ± c and 170.177 H2N Cl a, b + N COCl 161 S 169a ± c i = 5 (a), 6 (b), 7 (c). Cl NH2 + COCl S N 170 N N S 168 (a) PhH, D, 4 h; (b) hn (Hg lamp, 450 W), 1% MeOH in PhH, 3 h; (c) POCl3, D, 4 h; (d ) 10% Pd/C, KOH, MeOH±PhH (1 : 1), 20 8C, 24 h. 7. Synthesis of naphthyridines fused to a thiophene ring A one-stage procedure has been proposed 178 ± 181 for the prepara- tion of isomeric N(1),N(i )-naphthyridines; the procedure includes Pd(PPh3)4-catalysed cross-coupling of thiopheneboronic acids 171, 172 or 173 containing an ortho-formyl group with o-amino- hetaryl halides 174 ± 176.The cross-coupling products cyclise spontaneously during the reaction to give thiophenonaphthyr- idines 177 ± 185 with all possible types of ring fusion. The effects of the amount of the catalyst, the nature of the base and the reaction time on the yield of the naphthyridines 178 and 181 have been studied.180 N(i ) N S 165a ± d N 174a ± c Me NN S 166 NHR N(i) N 175b,c NH S 167a ± c O R=H: X=Cl (a), Br (b); R=Ac, X=Br (c); (a) Pd(PPh3)4, Na2CO3, DME or DMF. a, b, c, d Me N 176 (a) Pd(PPh3)4, Na2CO3, DME or DMF.When 2-bromo-3-nitropyridine 186 is employed in the cross- coupling with formylthiopheneboronic acids 171 ± 173, thieno- 1,5-naphthyridine 5-oxides 188 ± 190 are produced. However, in this case, cyclisation of the cross-coupling products 191 ± 193 proceeds on treatment with FeSO4 in NH4OH. Isomeric thieno- 1,5-naphthyridine 9-oxides 194 ± 196 are formed upon cross- coupling of the formylboronic acids 171 ± 173 with 3-acetyl- amino-2-bromopyridine oxide 187. 2-Tributylstannyl-3-formyl- thiophene 197 can be used in the cross-coupling with arenyl halides 186 and 187 instead of arylboronic acids. A mixture of N-oxide isomers 190 and 196 is also formed upon oxidation of thienonaphthyridine 179 with m-chloroperbenzoic acid.181 B(OH)2 S CHO 171 OHC B(OH)2 NHR S 172 a X CHO S B(OH)2 173 N 171 N 180 Br N 172 a N 181 S N 173 N 182 Me 171 Me Br 172 a NH2 Me 173 211 S N N 177 S N N 178 S N N 179 S S S N N183 S N N184 S N N 185212N 186 NO 187 (a) Pd(0), Na2CO3, DME or DMF; (b) FeSO4, NH4OH.The reaction of vicinal trimethylstannyl derivatives of tert- butoxycarbonylaminothiophenes 198 and 199 with halo-substi- tuted pyridinecarbaldehydes 200 ± 203 in the presence of PdCl2 , CuO and bis(diphenylphosphino)butane results in fused thieno- naphthyridines of all the possible types.182 N 171 CHO NO2 191 OHC NO2 N 172 a Br NO2 192 OHC N S 173 NO2 193 ON 171 N 194 O NHAc N 172 a Br N 195 O S N 173 N 196 b 186 CHO 193 a 187 196 SnBu3 S 197 NHCO2But SnMe3 S 198 SnMe3 I a NHCO2But S CHO 199 N 200 S S N b N 188 O S N S b N 189 OS N b N 190 O S S 190 S N N S N N 198 I CHO a 199 N 201 198 CHO Br a 199 N 202 198 CHO a Br 199 N 203 (a) PdCl2, CuO, Ph2P(CH2)4PPh2.The photoelectron spectra of all the thienonaphthyridine isomers have been analysed using semiempirical methods for the calculation of the ionisation energies of molecular orbitals.183, 184 Cross-coupling catalysed by zerovalent palladium complexes has been employed to obtain other annelated naphthyridines.Thus the reaction of bromoaminopyridine 176 with 2-formylphe- nylboronic acid provided 9-methylbenzo[c]-1,8-naphthyridine 204.178 B(OH)2 176, Pd(PPh3)4 Na2CO3, (MeOCH2)2 CHO Cross-coupling of o-bromoacetanilides 205 with trimethyl- stannylpyridyl ketones 206 in the presence of PdCl2 and CuO in DMF affords benzo[c]-2,7-naphthyridines 207.185 Benzo[c]-2,7- naphthyridines 208 were prepared by refluxing 2-tert-butoxycar- bonylaminophenylboronic acid 209 with substituted iodonicotin- aldehydes 210 in the presence of Pd(PPh3)4 and K2CO3 in a mixture of ethanol and toluene.186 NHAc SnMe3COR2 Br+ R3 N 206 R1 205 R1=H, Me, OMe; R2=H, Me; R3=H, OMe. I B(OH)2 + NHCO2But N 210 209 R=F, Cl, CONPri2 . V P Litvinov, S V Roman, V D henko S N N S N N S N N S N NS N N S N NMe PdCl2, CuO DMF N CHO Pd(PPh3)4 K2CO3 R N N204 R1 N R3 R2207 N R N 208Naphthyridines.Structure, physicochemical properties and general methods of synthesis 8. b-Enaminocarbonyl compounds in the synthesis of naphthyridines b-Enaminocarbonyl compounds have long been used successfully in the synthesis of isomeric naphthyridines. Thus pyridone 211, prepared by the reaction of b-enamino ketone 212 with 4,4- dicyano-3-aminobut-2-enenitrile 213, cyclises to give substituted 1,2,7,8-tetrahydro-1,8-naphthyridin-2-one 214. On treatment with concentrated hydrochloric acid at 100 8C, compound 214 undergoes disproportionation giving rise to a mixture of hexahy- dro (215) and dihydro derivatives of 1,8-naphthyridin-2-ones 216.187, 188 CN H2N + PhCOCH CHNR1R2 212 CN NCCH2 213 Ph NH2 Ph NH2 CN CN conc.HCl CN 100 8C O O NH NH HN 214 211 Ph NH2 NH2 H Ph CN CN+ O N O NH NH NH 216 215 The reaction of 4-dialkylaminopyridones 217, containing a b-enaminoamide fragment, with N-methylcyanoacetamide (218) is accompanied by replacement of the endocyclic amino group followed by cyclisation to give 2,7-naphthyridine 219.189 O NR2R3 NMe PriONa NH2 PriOH CN+NCCH2CONHMe 218 O R1 O R1 N 217 Me N 219 Me R1=Ph, 4-ClC6H4, 4-MeOC6H4; R2=Et, Ph, CH2Ph, CH2CH(OEt)2; R3=H, CH2CH2OCH2CH2. One approach to the construction of fused 1,6-naphthyridines 220 includes the reaction of 4-amino-2,6-dioxo-1,2,5,6-tetrahy- dro-3-pyridinecarbonitrile 221 with cyclic b-enamino ketones 222. The pyridinedione 221 reacts with 3-dimethylamino-1-phenyl- propan-1-one to give 1,5,6,7-tetrahydro-1,6-naphthyridine-5,7- dione 223.190 ± 192 OH NH2 HN O CN CN O + N CHMor (CH2)n O O 222 HN (CH2)n 221 220 n=1±3.221+PhCOCH2CH2NMe2 NH Ph NH2 CN CN Ph Me2N D 7Me2NH O O O NH HN O223 213 Enamino amides 226 and 227, containing a b-enaminocar- bonyl fragment, are promising starting compounds in the syn- thesis of hydrogenated naphthyridines 224 and 225.193 ± 204 Study of the kinetics of these reactions demonstrated that the rate of cyclisation of the compounds 226 is much higher than that for the amidine 227, due to the more pronounced delocalisation of the positive charge in the amidine system compared to the enamine system.195, 203, 204 AlkNH D 7Me2NH CON C(Alk)NMe2 O CN CN NAlk 226 NAlk 224 NMe2 CN NC D 7Me2NH O O NH NMe NMe225 N 227 9.Cyclisation of 1,5-dinitriles and 1,5-amino(amido)nitriles of the pyridine series Acid hydrolysis of 3- or 4-cyanopyridylacetonitriles or their reaction with sodium ethoxide affords isomeric amino derivatives of naphthyridines 228a ± c and 229a ± c. The 1,6-naphthyridines 228d,e have been prepared by hydrolysis of 2-cyanomethylnicoti- nonitriles.144, 205 R NH2 a N(i ) N CH(R)CN 228a ± c Br CN R N OEt b N(i ) N 229a ± c NH2 R=H: i=6, 7;R=Me, i=6; (a) HBr; (b) EtONa. R R N N NH2 a CN N CN 228d,e Br R=H, 4-FC6H4; (a) 30% HBr ± AcOH.2-Cyanophenyl- (230a) or 2-cyano-3-pyridylacetonitriles 230b react with methyl crotonate to give adducts 231a,b. Subsequent acid-catalysed cyclisation of compounds 231a,b gives rise to the corresponding pyrido- and benzo[c]-1,8-naphthyridines 232a,b� potential means for treating heart weakness, stenocardia and hypertension.206, 207 CH2CN 5% MeONa ±MeOH +MeCH CHCO2Me CN X 230a,b Me Me HBr X X O Br N NH CN CN CO2Me 231a,b 232a,b X=CH (a), N (b).214 O R1 NH a R2 CN CN O O R1 R1 b N NH R2 OMe R2 CN CN235 233 O R1 c N R2 CN236 R1, R2=H, Me, Ph; (a) CH2(CN)2; (b) NCCH2CO2Me; (c) NCCH2CONH2; (d ) HX, X=Cl, Br; (e) MeONa, MeOH, X =OMe; ( f ) HBr; (g) OH7, H2O orRO7, ROH.The methoxy group in 5-cyano-6-methoxy-3,4-dihydro- 2(1H)-pyridones 233 is replaced by malononitrile, methyl cya- noacetate or cyanoacetamide. The intermediates 234 ± 236 cyclise in acid or basic media to give 3,4-dihydro-1,6-naphthyridin- 2(1H)-ones 237 ± 239 (Scheme 4).208, 209 Cyclisation of piperidone 240 yields 1,6-naphthyridine mono- hydrate 241. An X-ray diffraction study showed that the C(2)7N(1) bond (1.362A) is longer than these bonds in other cyclic amides due to less efficient conjugation with the carbonyl group.210 CN CN HN H2N 1 O HN NC HBr N NC Me 240 Me Br241 Dinitriles 242 have been used to perform regioselective syn- thesis of 7-amino-8-cyano-1,6-naphthyridines 243.144, 211, 212 X R2 MeOH or AcOH ± HBr CN R1 R1 HN CN 242 R1, R2=Me, Et; X=Br, OMe.Naphthyridinediones 244 and 245 have been synthesised using reactions of 3-amino-4,4-dicyanobut-2-enenitrile (213) with a-diketone monophenylhydrazones. Cyclisation of intermediate dinitriles 246 and 247 was performed on treatment with a mixture of AcOH and HCl.213 CN H2N AcONH4 CN NCCH2 213 O R1 CN R2 CN 234 R1 H O f R2 OMe CN R1 H O g R2 NH2 CN O 2R2 N N NH2 CN 243 V P Litvinov, S V Roman, V D Dyachenko Scheme 4 O O R1 R1 NH NH NH d or e CN CN + R2 R2 CN X X H2N NH2 CN N 237b N 237a O O R1 NH NH CO2Me CO2Me + R2 Br NH2 Br H2N N 238a N 238b O NH CN O HN H2N239 O O PhNHN NH PhNHN NH c a CN Me CN Me O HO 246 CN CN N 244 NH2 PhNHN NH2 N PhNHN CN N c b Me CN Me O O 247 CN CN NH 245 (a) PhNHN=C(Ac)CO2Et; (b) PhNHN=C(Ac)CN; (c) AcOH, HCl.Enamino nitriles of 2,7-naphthyridinone(or -thione) 248 have been synthesised by cyclisation of amido(or thioamido) nitriles 249, which are prepared from enamine 250 or substituted malo- nonitrile 251.214, 215 CN NC PhNCX CH2(CN)2 Mor Me N Me Me Me251 Mor X Me N 250 Me PhNCX CH2(CN)2 NHPh Me Me NMeNaphthyridines. Structure, physicochemical properties and general methods of synthesis CN NC CN X Me NH NHPh Me MeMeN NPh Me Me N X 249 CN Me NH2 MeMeN NPh 248 X X=O, S.On treatment with sodium ethoxide in ethanol, 1-methyl-2- dicyanomethylidenepiperidine-3-carbanilides 252 cyclise to give partially hydrogenated 1,6-naphthyridin-5(6H)-ones 253.216 O O NAr NHAr EtONa, EtOH CN NH2 Me N Me N CN CN 253 252 Ar=Ph, 4-ClC6H4, 3-MeC6H4. Ammonolysis of enamines 254, obtained from 1-aryl-3-(2,2- dicyanovinyl)-4-piperidones affords 2-amino-6-aryl-3-cyano- 5,6,7,8-tetrahydro-1,6-naphthyridines 255, employed in the syn- thesis of pteridines 256.211 CN NH3 4-R1C6H4N MeOH CN NR22254 NH2 . HCl HN CN 4-R1C6H4N NMe2 EtONa, DMF N NH2 255 NH2 N 4-R1C6H4N N N NH2 256 R1=Me, Et; R22 =(CH2)4. 10.Thorpe dimerisation of cyanopicolines Dimerisation of 2-methylnicotinonitrile (257) induced by potas- sium tert-butoxide and including intramolecular addition of the imine to the nitrile at the cyclisation stage gives rise to 5-amino-7- (2-methyl-3-pyridyl)-1,6-naphthyridine (258).130 N CNH CN ButOK N Me N 257 Me N NH2 N N N 258 Me 215 8-Amino-1,7-, 1-amino-2,6- and 1-amino-2,7-naphthyridines 259 ± 261, possessing efficient bactericidal and fungicidal activ- ities, were prepared from the corresponding cyanopicolines in a similar way.144 Me NH2 N N N(i ) N N N 259, 260 261 Me NH2 i=6, 7. 11. Condensation of piperidones with cyanoacet(or thioacet)amides The condensation of cyanothioacetamide 262 with the sodium salt of 2-formyl-3-quinuclidone 263 in ethanol in the presence of AcOH has resulted in the synthesis of 1,5-naphthyridine-2(1H)- thione 24a.106 CN N CHONa N NCCH2C(S)NH2 262 S O NH 24a 263 Cyanothioacetamide 262 condenses with 2-arylmethylidene- 3-oxoquinuclidines 264 in the presence of piperidine to yield relatively stable salts 265; in an acid medium, these products are converted into 3,4-dihydro-1,5-naphthyridinethiones 266.Upon refluxing in an HCl ± EtOH mixture, the compounds 266 are transformed into naphthyridinethiones 24, which are also formed directly on refluxing the salts 265 in an ethanolic solution of HCl. R H R CN N N 10% HCl PiH+ 262 Pi O S7 HN 265 264 R CN N 24 HCl EtOH S HN 266 R=H, Ph, 4-FC6H4; Pi is piperidine.The partially hydrogenated 1,6-naphthyridinethione 33 has been synthesised by the reaction of 3,5-bis(phenylmethylidene)-1- isopropyl-4-piperidone 267 and the amide 262.107 Ph Ph CN PriN PriN 262 S O HN33 267 Ph Ph The reaction of the salt 268 with the amide 262 in ethanol in the presence of AcOH follows the Knoevenagel reaction route and affords adduct 269, which recyclises when refluxed in ethanol in the presence of a catalytic amount of piperidine, giving rise to the the 2,naphthyridine 35. The compound 35 is the product of a cascade type reaction comprising acidic cleavage according to Claisen, the Michael intramolecular addition, the interaction between the thioamide and ester groups and dehydrogenation.110216 ONa O Me 262 AcOH, EtOH Me Me N 268 Me O N EtO Me Me Me MeN CN EtO2C Me CSNH2 Me MeN Me O HN 35 The cyclocondensation of piperidone 270 with cyanoaceta- mide in the presence of diethylamine gives 1,6-naphthyridinone 271, which reacts with POCl3 to give chloro derivative 272.217 The compound 272 was also prepared by condensation of enamine 273 with ethoxymethylidenemalononitrile followed by cyclisation of intermediate 274.217 CHONa NCCH2CONH2 Et2NH, H2O MeN Me Me O 270 CN MeN Me Me O NH 271 CN Mor CN EtO THF, N2 Me Me Me Me Me N 273 Refluxing of 1,3-dicarbonyl compounds 275 with cyanoacet- amide in ethanol in the presence of piperidine gives 2,7-naphthyr- idine derivatives 276, used as components of formulations for treatment of diseases of blood circulation organs.218 CN O NCCH2CONH2 EtOH, Pi RN COMe 275 R=H, Alk, Ph, Ac, Bn, 4-pyridyl.12. Cyclocondensation of piperidones with aminopyrimidines When aminopyrimidines 277 are heated (100 8C, 3 h) with the salts of isomeric hydroxymethylidenepiperidones 278 and 279 in 85% orthophosphoric acid, substituted tetrahydropyrimido- [4,5-b]-1,6- (280a,b) 219 and -1,7-naphthyridines 281a,b are formed; 220 they are potential antitumour remedies. O C(CN)CSNH2 Me EtOH, Pi Me Me N 269CSNH2 CN Me MeN CN 7H2 Me S O NH CN S CN POCl3 MeN Me Me Cl N 272 Mor CN HCl (gas) 272 PriOH CN 274 Me N O NH RN 276 Me V P Litvinov, S V Roman, V D Dyachenko ONa BnN OH O N BnN 278 OH R N ONa N 280a,b N O BnN R H2N O N 277a,b 279 NH BnN R N N 281a,b R=NH2 (a), OH (b).Refluxing of aminopyrimidines 277a,b and 282 with methyl 1-benzyl-4-oxopiperidine-3-carboxylate hydrochloride 283 in gla- cial acetic acid provided the synthesis of pyrimido[4,5-c]-2,7- naphthyridin-6-ones 284a,b and 285. The compounds 284b and 285 present in concentrations of 1075 and 261076 mol litre71, respectively, inhibit in vitro the growth of malignant leukemia cells by 50%. 221 NBn O HN 277a,b AcOH O R N N 284a,b NH2 NBn. HCl N O NBn NH2 H2N NH2 283 CO2Me N N 282 AcOH O H2N N N 285 H A similar condensation of the aminopyridine 277a with hydrogenated 3-formylpyridine 286 in the presence of HCl in ethanol results in tetrahydropyrimido[4,5-c]-2,7-naphthyridine 287 mixed with its 7-benzoyl derivative.219, 222 NH O NCOPh 277a, HCl HN D N H2N N 287 CHO 286 13.Other general methods for the synthesis of naphthyridines Below we present the syntheses of naphthyridines starting from pyridine derivatives containing an amino (or protected amino) group and a carbonyl group or its synthetic equivalent in vicinal positions. Cyclisation is accomplished by treatment with basic or acidic reagents. CO2Et EtONa, EtOH R R O N NH N(i ) NH2 288 289 R=H, Me; i=5±8.O C(O)NH2 EtONa, EtOH NH Me Me N C Ph N(i ) 291 CPh 290Naphthyridines. Structure, physicochemical properties and general methods of synthesis O C(O)NH2 TsOH NH N(i ) N CH2CH(OMe)2 293 292 CH(OH)CH2CO2But b a R1 R1 O N NH N(i ) NHCOBut 294 295 c R1 Cl N N(i ) R1 N XC6H4OCH(Me)CO2R2-4 N(i ) 296 i=6±8; R1=H, Me, Cl; R2=Alk; X=O, S; (a) 3M HCl; (b) POCl3, D; (c) 4-HXC6H4OCH(Me)CO2R2, K2CO3 or NaH, DMSO. Me Me OHPh Cl Cl Me c a, b N N X CH(OEt)2 NHCOBut 297 299 Me Ph Cl N N 301 O O O O CHO OHPh Ph e d Me a, b N N N OH OH X CH(OEt)2 NHCOBut NH 298 300 CHO CHO Ph Ph f N N OH OMe N HN 302 (a) BunLi; (b) (EtO)2CHCOPh; (c) HCl; (d) H2SO4; (e) MeOH, H3PO4; ( f ) diazabicycloundecene, dioxane, D.Cyclisation of ethyl aminopyridineacrylates 288 in ethanol in the presence of sodium ethoxide has served as a route to N(1),N(i )-naphthyridin-2(1H)-ones 289.223 Cyclisation of ethy- nylpyridinecarboxamides 290 under similar conditions yields 1,6-, 1,7-, 2,6- and 2,7-naphthyridones 291.224 Naphthyridones 292 were prepared by refluxing vicinal (2,2-dimethoxyethyl)pyridine- carboxamides 293 with toluene-p-sulfonic acid in benzene.224 N(1),N(i )-Naphthyridin-2(1H)-ones 294, prepared by refluxing vicinal 2-tert-butoxycarbonyl-1-hydroxyaminopyridines 295 in dioxane in the presence of 3M HCl, have been used to obtain naphthyridine derivatives 296, which possess herbicidal and fungicidal properties and are efficient for the protection of cotton, soy-bean and sugar-cane crops.225, 226 Intramolecular cyclisation of substituted tert-butoxycarbonylaminopyridines 297 and 298, synthesised from compounds 299 and 300 via a two-stage proce- dure, gives 1,7- (301) and 1,6-naphthyridines 302.227 2-(3-Cyanopropyl-oxy or -thio)pyridine-3-carbonitriles 303 undergo an intramolecular nucleophilic rearrangement, similar to the Smiles rearrangement, being thus converted into fused 1,6- naphthyridines 304.228 ± 230 A similar cyclisation of 2-pyridinecar- bonitrile 305 gives rise to dihydrofuro[3,2-f ]-1,7-naphthyridine 306.231N R X(CH2)3CN ButOK CN 303 R X N 7R CN CN H X R N N N7 X N R N 304 NH2 X=S, R=H; X=O, R=Me.O(CH2)3CN ButOK CN N 305 Substituted 1,2,4-triazines decompose on thermolysis with nitrogen evolution.Thermolysis of compounds 307 and 308, having an ethynyl substituent in the side chain, yields substituted naphthyridines 309 232 and 310.233 N NH(CH2)3C CH MeO2C MeO2C N N 307 CONH(CH2)2C CH F3C NN N Ph 308 N-Hydroxyethyl-substituted 1,6- (311) and 2,7-naphthyri- dones 312 were prepared by refluxing the pyranopyridine isomers 313 and 314 with ethanolamine.234 Ph N H2NCH2CH2OH O 313 O Ph H2NCH2CH2OH O N 314 O The synthetic routes to 1,8- (315) and 1,6-naphthyridones 316 and 317 and annelated systems 318 and 319 make use of the reactions of pyrans 320 and 321 with malononitrile or its deriva- tives. The reactions are carried out in ethanol in the presence of piperidine (Scheme 5).235 ± 237 217 7 N R X(CH2)2CHCN CNX7 N N CN CH X N R N NHO N N 306 NH2 HN N MeO2C D 7N2 MeO2C 309 HN O D F3C 7N2 N Ph 310 Ph N N(CH2)2OH 311 O Ph N N(CH2)2OH 312 O218 R1 NC CO2Et Me H2N O 320 R1=2-ClC6H4, 4-ClC6H4, 4-MeC6H4; (a) NCCH2CONH2; (b) CH2(CN)2; (c) R2CH=C(CN)2; R2=Alk, Ar.Me CN a O 321 NHNC HNO318 CH2 CN 321 O NH2 (a) NCCH2CONH2 . 3-Acetyl-2-cyanomethylidenepyridine 322 and 3-acetyl-4-cya- nomethylidenepyridine 323 cyclise on heating (130 8C, 30 ± 45 min) in 85% orthophosphoric acid to give 1,6- (324) and 2,7-naphthyridinone derivatives 325 (Scheme 6).238 Me Me Me O H3PO4 CN Me NPh R1 322 R1 CNO H3PO4 Me SMe Me NR2 323 R1=CN, CO2Et; R2=Ph, 4-MeOC6H4.V. Conclusion The data surveyed here demonstrate that general and facile methods for the synthesis of diverse substituted naphthyridines are now available. Various naphthyridines can be synthesised Scheme 5 NH2 R1 NC CO2Et a Me N O HN 315 R1 O CO2Et HN b Me N H2N CN 316 R1 O CO2Et N c Me N R2 NC CN 317 CN NC Me a NH2 NH O CNCN NHNH2 CN CH2 NH2 NH2 CN a N O NH2 N 319 Scheme 6 Me N D O Me NPh R1 324OH R1 N D Me Me NR2 O 325 V P Litvinov, S V Roman, V D Dyachenko under the same conditions using similar equipment. Various types of biological activity exhibited by many naphthyridine derivatives together with the ease of their preparation make these products attractive objects for the search for new highly effective medicinal preparations, which can be done using combinatorial synthesis.This review was supported by the Russian Foundation for Basic Research (Project No. 99-03-32965). References 1. A Reissert Berichte 26 2137 (1893) 2. B Bobranski, E Sucharda Berichte 60 1081 (1927) 3. G Koller Berichte 60 1918 (1927) 4. N Ikekawa Chem. Pharm. Bull. 6 263 (1958) 5. N Ikekawa Chem. Pharm. Bull. 6 401 (1958) 6. N Ikekawa Chem. Pharm. Bull. 6 269 (1958) 7. G Giacomello, F Gualtieri, F M Riccieri,M L Stein Tetrahedron Lett. 1117 (1965) 8. R Tan, A Taurins Tetrahedron Lett.2737 (1965) 9. C F H Allen Chem. Rev. 47 275 (1950) 10. W W Paudler, T J Kress Adv. Heterocycl. Chem. 11 123 (1970) 11. H C van der Plas,M Wozniak, H J W van den Haak Adv. Heterocycl. Chem. 33 95 (1983) 12. W W Paudler, R M Sheets Adv. Heterocycl. Chem. 33 147 (1983) 13. M Weiss, H Hauser, in Heterocyclic Compounds (Ed. R C Elderfield) (New York: Wiley, 1961) 14. J Pomorski Wiad. Chem. 24 773 (1970) 15. D G Wibberley, in Heterocyclic Compounds. Ser. 1 Vol. 4 (New York: Interscience, 1973) p. 167 16. Y Hamada, I Takeuchi J. Synth. Org. Chem. Jpn. 32 602 (1974) 17. W Czuba Wiad. Chem. 32 91 (1978) 18. V Chuba Khim. Geterotsikl. Soedin. 3 (1979) a 19. J A Bajgrowicz,W Czuba Wiad. Chem. 33 235 (1979) 20. W Czuba Wiad. Chem. 34 263 (1980) 21.W Czuba Wiad. Chem. 34 593 (1980) 22. W Czuba Wiad. Chem. 35 441 (1981) 23. M Brufani, D Duranti, G Giacomello Gazz. Chim. Ital. 89 2328 (1959); Chem. Abstr. 55 5081h (1961) 24. M Brufani, W Fedeli, G Giacomello, F M Riccieri, A Vaciago Atti Acad. Nazl. Lincei, Rend., Classe Sci. Fis., Mat. Nat. 40 187 (1966); Chem. Abstr. 65 11 464a (1966) 25. Y Besidsky, K Luthman, A Claesson, C J Fowler, I CsoÈ regh, U Hacksel J. Chem. Soc., Perkin Trans. 1 465 (1995) 26. PWCodding,MB Szkaradzinska, AWRoszak, L J Aha, T J Hagen, J M Cook Can. J. Chem. 66 2981 (1988) 27. B Pniewska, A Rykowski, R Anulewicz Acta Crystallogr., Sect. C 46 2198 (1990) 28. B Pniewska, R Anulewicz Acta Crystallogr., Sect. C 43 2198 (1987) 29. M Datta, S S Hannan, A N Talukdar Acta Crystallogr., Sect.C 51 978 (1995) 30. T Kurihara, M Katsuda, K Nasu, MInoue, M Doi Chem. Pharm. Bull. 33 5551 (1985) 31. H Bock, T T H Van, H SchoÈ del Monatsh. Chem. 127 391 (1996) 32. WO PCT 96 39406; Chem. Abstr. 126 117 960 (1997) 33. C J Fahrni, A Pfaltz,M Neuburger,M Zehnder Helv. Chim. Acta 81 507 (1998) 34. MMaekawa,MMunakata, S Kitagawa, T Kuroda-Sowa,Y Suenaga, M Yamamoto Inorg. Chim. Acta 271 129 (1998) 35. T Siener,U Mueller,M Janes,U Holzgrabe Pharmazie 53 442 (1998) 36. K Chinnakali, H-K Fun, I A Rezak, P Murugan, V T Ramakrishnan Acta Crystallogr., Sect. C 54 781 (1998) 37. F Zamora,M Sabat, B Lippert Inorg. Chim. Acta 282 237 (1998) 38. T E Peacock J. Chem. Soc. 1946 (1960) 39. S F Mason J. Chem. Soc.493 (1962) 40. S C Wait Jr , J W Wesley J. Mol. Spectrosc. 19 25 (1966) 41. T J Kress Ph D Thesis, Ohio University, Athens, 1967 42. A H Gawer, B P Dailey J. Chem. Phys. 42 2658 (1965) 43. W W Paudler, T J Kress J. Org. Chem. 33 1384 (1968) 44. V Oakes, H N Rydon J. Chem. Soc. 204 (1958) 45. H C Longuet-Higgins, C A Coulson J. Chem. Soc. 971 (1949) 46. S Basu, R Bhattacharya Proc. Natl. Inst. Sci. India A23 1 (1957); Chem. Abstr. 52 864i (1958) 47. J C M Henning J. Chem. Phys. 44 2139 (1966)Naphthyridines. Structure, physicochemical properties and general methods of synthesis 48. A C Boicelli, R Daniel, A Mangini, L Lunazzi, G Placucci J. Chem. Soc., Perkin Trans. 2 1024 (1973) 49. C L Khetrapal, A C Kunwar Mol. Cryst. Liq. Cryst. 15 363 (1972) 50. C L Khetrapal, A Saupe, A C Kunwar Mol.Cryst. Liq. Cryst. 17 121 (1972) 51. R Danieli, L Lunazzi, C A Veracini J. Chem. Soc., Perkin Trans. 2 19 (1976) 52. P Morwell, A R Lacey Chem. Phys. 118 71 (1987) 53. A Shiozawa, Y Kogo, Y Ichikawa, C Komuro,MIshikawa, S Kurashige, H Miyazaki, H Yamanaka, T Sakamoto Chem. Pharm. Bull. 33 5332 (1985) 54. M B Cox, P Arjunan, S K Arora J. Antibiot. 44 885 (1991) 55. AGueiffier,AViols,YBlache, J P Chapat,OChavignon, J C Teulade, F Fauvelle, G Grassy, G Dauphin J. Heterocycl. Chem. 34 765 (1997) 56. C G Mohan, P C Mishra Indian J. Biochem. Biophys. 34 429 (1997); Chem. Abstr. 128 200 547 (1998) 57. M Wozniak,M Grzegozek, P Suryolo Liebigs Ann. Recl. 2601 (1997); Chem. Abstr. 128 102 020 (1998) 58.A Albert J. Chem. Soc. 1790 (1960) 59. D N Bailey, D M Hercules, T D Eck Anal. Chem. 39 877 (1967) 60. W Skoda, H Bayzer Monatsh. Chem. 89 5 (1958) 61. M Wozniak,M Tomula Zesz. Nauk. UJ. Pr. Chem., 75 (1987); Ref. Zh. Khim. 12 B 1209 (1988) 62. W Czuba Bull. Acad. Polon. Sci., Ser. Sci. Chim. 11 423 (1963); Chem. Abstr. 60 2454f (1964) 63. J T Carrano, S C Wait Jr J. Mol. Spectrosc. 46 401 (1973) 64. W W Paudler, T J Kress Chem. Ind. (London) 1557 (1966) 65. W W Paudler, T J Kress J. Heterocycl. Chem. 2 393 (1965) 66. W W Paudler, T J Kress J. Org. Chem. 31 3055 (1966) 67. W W Paudler, T J Kress J. Heterocycl. Chem. 4 284 (1967) 68. W W Paudler, T J Kress J. Heterocycl. Chem. 5 561 (1968) 69. W L F Armarego, T J Batterman J.Chem. Soc., B 750 (1966) 70. W L F Armarego J. Chem. Soc., C 377 (1967) 71. W W Paudler, T J Kress J. Org. Chem. 32 832 (1967) 72. W W Paudler, T J Kress J. Org. Chem. 32 2616 (1967) 73. E M Hawes, D G Wibberley J. Chem. Soc., C 315 (1966) 74. E M Hawes, D G Wibberley J. Chem. Soc., C 1564 (1967) 75. ELEzell,RPThummel,GEMartinJ. Heterocycl.Chem. 21 817 (1984) 76. M Prabhakaram, Q Ching,M S Feather, B J Ortwerth Amino Acids 12 225 (1997); Chem. Abstr. 127 205 845 (1997) 77. M Munakata, S-G Yan, M Maekawa,M Akiyama, S Kitagawa J. Chem. Soc., Dalton Trans. 4257 (1997) 78. W W Paudler, T J Kress J. Heterocycl. Chem. 4 547 (1967) 79. M Wozniak, B Szpakiewicz Zesz. Nauk. UJ. Pr. Chem., 87 (1987); Ref. Zh. Khim. 11 B 1151 (1988) 80. M Wozniak,M Tomula Zesz.Nauk. UJ. Pr. Chem., 83 (1989); Ref. Zh. Khim. 23 B 1158 (1990) 81. M S Bobyleva, N S Kulikov Teor Eksp. Khim. 25 626 (1989) b 82. D M W van den Ham, D van der Meer Chem. Phys. Lett. 12 447 (1972); Chem. Abstr. 76 119 668 (1972) 83. R Mueller, F Doerr Z. Electrochem. 63 1150 (1959) 84. G Fischer Chem. Phys. Lett. 21 305 (1973) 85. A D Jordon, G Fischer, I G Ross J. Mol. Spectrosc. 87 345 (1981) 86. R Bramley, B J McColl Mol. Phys. 29 649 (1975) 87. A Albert, J N Philips J. Chem. Soc. 1294 (1956) 88. KAAllen, JCymerman-Craig,AADiamantis J. Chem. Soc. 234 (1954) 89. A Albert, A Hampton J. Chem. Soc. 505 (1954) 90. A Albert, W L F Armarego J. Chem. Soc. 4237 (1963) 91. A Albert, C W Rees Spec. Lect. Biochem. Univ. Coll.London 96 (1954); Chem. Abstr. 52 8281i (1958) 92. R Roullier, E Laviron Electrochim. Acta 421 (1976) 93. R Roullier, E Laviron Electrochim. Acta 773 (1978) 94. K MoÈ ller, O SuÈ s Ann. Chem. 612 153 (1958) 95. T K Adler, A Albert J. Chem. Soc. 1794 (1960) 96. M Wozniak, H Jackiewicz Pol. J. Chem. 57 587 (1983) 97. BRD P. 4 308 014; Chem. Abstr. 122 9884 (1995) 98. A K Sheinkman, T N Nezdiiminoga, T S Chmilenko, in Novoe v Khimii Azinov (Tez. Dokl. 2-i Vsesoyuz. Konf.), Sverdlovsk, 1985 [New in Chemistry of Azines (Abstracts of Reports of the Second All- Union Conference), Sverdlovsk, 1985] p. 149 219 99. A K Sheinkman, T N Nezdiiminoga, T S Chmilenko, in Khimiya, Biokhimiya i Farmakologiya Proizvodnykh Indola (Tez. Dokl. 1-i Vsesoyuz.Konf.), Tbilisi, 1986 [The Chemistry, Biochemistry and Pharmacology of Indole Derivative (Abstracts of Reports of the First All-Union Conference), Tbilisi, 1986] p. 179 100. A K Sheinkman, T N Nezdiiminoga, T S Chmilenko, N A Klyuev Khim. Geterotsikl. Soedin. 1218 (1986) a 101. Pol. P. 125 298; Chem. Abstr. 105 191 054 (1986) 102. Pol. P. 125 299; Chem. Abstr. 101 151 826 (1984) 103. Pol. P. 125 310; Chem. Abstr. 105 191 055 (1986) 104. W Czuba, T Kowalska, H Poradowska, P Kowalski Pol. J. Chem. 58 1221 (1984) 105. AMShestopalov, Doctoral Thesis in Chemical Sciences, Institute of Organic Chemistry, Acad. Sci. of the USSR, Moscow, 1991 106. A M Shestopalov, V N Nesterov, Yu A Sharanin, V P Litvinov, V Yu Mortikov,V E Shklover, Yu T Struchkov Khim.Geterotsikl. Soedin. 557 (1989) a 107. A M Shestopalov, V Yu Mortikov, Yu A Sharanin, A V Turov, V P Litvinov Zh. Org. Khim. 25 1980 (1989) c 108. V P Litvinov, V K Promonenkov, Yu A Sharanin, AMShestopalov, in Organicheskaya Khimiya. (Itogi Nauki i Tekhniki) [Organic Chemistry (Advances in Science and Engineering Series)] (Moscow: Izd. VINITI, 1989) Vol. 17, p. 145 109. A A Geies, A A Abdel-Hafez, J C Lancelot, H S El-Kashef Bull. Chem. Soc. Jpn. 66 3716 (1993) 110. V A Artemov, A M Shestopalov, V P Litvinov Khim. Geterotsikl. Soedin. 512 (1996) a 111. Russ. P. 1 640 984; Byull. Izobret. (29) 321 (1995) 112. M Wozniak, D J Buurman, H C van der Plas J. Heterocycl. Chem. 22 765 (1985) 113. MWozniak, H C van der Plas, S Harkema J.Org. Chem. 50 3435 (1985) 114. MWozniakMonogr.Prak. 25 (1987); Ref. Zh.Khim. 10Zh 250 (1988) 115. M Wozniak,M Tomula Liebigs Ann. Chem. 471 (1993) 116. M Wozniak, H C van der Plas,M Tomula, A van Veldhuizen J. Heterocycl. Chem. 22 761 (1985) 117. Q Chen, L D Deady, M F Mackay Aust. J. Chem. 46 1909 (1993) 118. Q Chen, L W Deady J. Heterocycl. Chem. 29 1197 (1992) 119. Q Chen, L W Deady Aust. J. Chem. 46 987 (1993) 120. Q Chen, L W Deady, B C Baguley,W A Denny J. Med. Chem. 37 593 (1994) 121. B Bachowska,W Sliwa Monatsh. Chem. 115 1101 (1984) 122. T Padzikowska,W Sliwa J. Prakt. Chem. 329 529 (1987) 123. B Bachowska,W Sliwa Acta Chim. Hung. 125 491 (1988) 124. Pol. P. 149 332; Chem. Abstr. 115 8769 (1991) 125. Pol. P. 149 212; Chem.Abstr. 114 101 971 (1991) 126. T Girek, T Zujewska, W Sliwa Acta Chim. Hung. 127 711 (1990) 127. B Bachowska Monatsh. Chem. 126 227 (1995) 128. G Matusiak, W Sliwa Acta Chim. Hung. 125 267 (1988) 129. Pol. P. 144 778; Chem. Abstr. 112 178 938 (1990) 130. Pol. P. 165 956; Chem. Abstr. 125 275 674 (1996) 131. C F Nutaitis,M L Grawley, J O Baza-Nutaitis Org. Prep. Proced. Int. 30 481 (1998) 132. B Mianowska,W Sliwa Acta Chim. Hung. 128 93 (1991) 133. G Matusiak, A Nowek,W Sliwa Acta Chim. Hung. 126 187 (1989) 134. T Zujewska, B Bachowska Aust. J. Chem. 49 523 (1996); Chem. Abstr. 125 195 461 (1996) 135. W Czuba,M Grzegozek Zesz. Nauk. UJ. Pr. Chem. 63 (1991); Ref. Zh. Khim. 8 Zh 249 (1993) 136. Eur. P. 587 473; Chem. Abstr. 121 108 763 (1994) 137.BRD P. 3 508 816; Chem. Abstr. 105 191 059 (1986) 138. B Bobransk1, E Sucharda Rocz. Chem. 7 192 (1927); Chem. Abstr. 22 777 (1928) 139. Germ. P. 507 637; Chem. Abstr. 25 716 (1931) 140. Eur. P. 321 191; Chem. Abstr. 112 77 161 (1990) 141. I Takeuchi, Y Hamada, K Okamura Heterocycles 29 2109 (1989) 142. I Takeuchi, Y Hamada,M Hirota Chem. Pharm. Bull. 41 747 (1993) 143. L Chrazastek, B Mianowska,W Sliwa Aust. J. Chem. 47 2129 (1994) 144. F S Babichev, Yu A Sharanin, V K Promonenkov, V P Litvinov, Yu M Volovenko Vnutrimolekulyarnoe Vzaimodeistvie Nitril'noi i Aminogrupp (Intramolecular Interaction of Nitrile and Amino Groups) (Kiev: Naukova Dumka, 1987)220 145. Yu A Sharanin, V K Promonenkov, V P Litvinov, in Organi- cheskaya Khimiya.(Itogi Nauki i Tekhniki) [Organic Chemistry (Advances in Science and Engineering Series)] (Moscow: Izd. VINITI, 1991) Vol. 20 146. S Grivas, E Ronne J. Chem. Res. (S) 268 (1994) 147. BRD P. 3 601 688; Chem. Abstr. 107 217 628 (1987) 148. S. Afr. P. 85/04 792; Chem. Abstr. 105 208 769 (1986) 149. Pat 58614 Evropa; Chem. Abstr. 98 59 888 (1983) 150. J Matsumoto, T Miyamoto, A Minamida, Y Nishimura, H Egawa, H Nishimura J. Heterocycl. Chem. 21 673 (1984) 151. Eur. P. 387 802; Chem. Abstr. 114 101 973 (1991) 152. G B Barlin, W-L Tan Aust. J. Chem. 37 1065 (1984) 153. Jpn. P. 1 112 279; Chem. Abstr. 130 124 998 (1999) 154. I Hayakawa, N Suzuki, K Suzuki, Y Tanaka Chem. Pharm. Bull. 32 4914 (1984) 155. Fr. P. 2 531 084; Chem.Abstr. 101 55 088 (1984) 156. G B Barlin, W-L Tan Aust. J. Chem. 38 459 (1985) 157. Eur. P. 172 651; Chem. Abstr. 105 97 485 (1986) 158. Eur. P. 146 243; Chem. Abstr. 103 215 151 (1985) 159. J Chattopadhyay, S K Basu Indian J. Chem. B29 98 (1990) 160. S Radi, P Hradil Collect. Czech. Chem. Commun. 56 2240 (1991) 161. V Cecchetti, S Clementi, G Cruciani, A Fravolini, P G Pagella, A Savino, O Tabarrini J. Med. Chem. 38 973 (1995) 162. US P. 4 560 691; Chem. Abstr. 104 207 249 (1986) 163. Eur. P. 115 469; Chem. Abstr. 101 211 137 (1984) 164. Eur. P. 168 350; Chem. Abstr. 104 207 258 (1986) 165. Eur. P. 346 208; Chem. Abstr. 113 6315 (1990) 166. Eur. P. 346 207; Chem. Abstr. 112 216 904 (1990) 167. Fr. P. 2 567 887 Frantsiya; Chem. Abstr.105 191 056 (1986) 168. R K Russell, O Lever, J William Synth. Commun. 23 2931 (1993) 169. M H Sherlock, J J Kaminski, W C Tom, J F Lee, Sh-Ch Wong, W Kreutner, R W Bryant, A T McPhail J. Med. Chem. 31 2108 (1988) 170. A D Dunn Z. Chem. 30 20 (1990) 171. G B Barlin, W-L Tan Aust. J. Chem. 37 2469 (1984) 172. M Di Braccio, G Roma, A Balbi, E Sottofattori, M Carazzone Farmaco 44 865 (1989) 173. F S Babichev, YuMVolovenko, A G Nemazanyi, S I Tyukhtenko Ukr. Khim. Zh. 52 506 (1986) 174. YuMVolovenko, A G Nemazanyi, G L Vesel'skaya, F S Babichev Dokl. Akad. Nauk Ukr. SSR, Ser. B (6) 36 (1986) 175. H Kudo, K Takahashi, R N Castle,M L Lee J. Heterocycl. Chem. 24 1009 (1987) 176. M J Musmar, R N Castle J. Heterocycl. Chem. 28 203 (1991) 177.R N Castle, S Pakray, G E Martin J. Heterocycl. Chem. 28 1997 (1991) 178. S Gronowitz, A-B HoÈ rnfeldt, Y-H Yang Chem. Scr. 26 311 (1986) 179. S Gronowitz Chem. Scr. 27 535 (1987) 180. S Gronowitz, J Malm, A-B HoÈ rnfeldt Collect. Czech. Chem. Commun. 56 2340 (1991) 181. J Malm, A-B HoÈ rnfeldt, S Gronowitz Heterocycles 35 245 (1993) 182. P BjoÈ rk, T Aakermann, A-B HoÈ rnfeldt, S Gronowitz J. Heterocycl. Chem. 32 751 (1995) 183. A L Marzinzik, P Rademacher, J Malm, S Gronowitz Acta Chem. Scand. 49 907 (1995) 184. A L Marzinzik, P Rademacher, P BjoÈ rk, S Gronowitz Acta Chem. Scand. 49 913 (1995) 185. J Malm, P BjoÈ rk, S Gronowitz, A-B HoÈ rnfeldt Tetrahedron Lett. 35 3195 (1994) 186. C Cochennec, P Rocca, F Marsais, A Godard, G Queguiner J.Chem. Soc., Perkin Trans. 1 979 (1995) 187. L A Rodinovskaya, V K Promonenkov, Yu A Sharanin, V P Litvinov, A M Shestopalov, in Organicheskaya Khimiya. (Itogi Nauki i Tekhniki) [Organic Chemistry (Advances in Science and Engineering Series)] (Moscow: Izd. VINITI, 1989) Vol. 17, p. 1 188. H Junek, G Stolz Monatsh. Chem. 101 1234 (1970) 189. V Aggarwal, G Singh, H Ila, H Junjappa Synthesis 214 (1982) 190. H Junek, W Remp Monatsh. Chem. 104 433 (1973) 191. W Remp, H Junek Monatsh. Chem. 104 1101 (1973) 192. L A Perez-Medina, R P Mariella, M S McElvain J. Am. Chem. Soc. 69 2574 (1947) 193. V G Granik, N B Marchenko, T F Vlasova, R G Glushkov Khim. Geterotsikl. Soedin. 1509 (1976) a V P Litvinov, S V Roman, V D Dyachenko 194.V G Granik, O Ya Belyaeva, R G Glushkov, T F Vlasova, O S Anisimova Khim. Geterotsikl. Soedin. 1106 (1977) a 195. V G Granik, A B Grigor'ev,M K Polievktov Khim. Geterotsikl. Soedin. 1523 (1977) a 196. V A Azimov, V G Granik, R G Glushkov, L N Yakhontov Khim. Geterotsikl. Soedin. 355 (1978) a 197. V G Granik, E O Sochneva, N P Solov'eva Khim. Geterotsikl. Soedin. 416 (1980) a 198. V G Granik, E O Sochneva, N P Solov'eva, E F Kuleshova, O S Anisimova Khim. Geterotsikl. Soedin. 1120 (1980) a 199. V A Azimov,N N Bychikhina, L N Yakhontov Khim. Geterotsikl. Soedin. 215 (1981) a 200. V A Azimov,N N Bychikhina, L N Yakhontov Khim. Geterotsikl. Soedin. 1283 (1981) a 201. L V Ershov, V G Granik Khim. Geterotsikl. Soedin. 929 (1985) a 202. V G Klimenko, A M Zhidkova, F M Stoyanovich, V G Granik Khim. Geterotsikl. Soedin. 364 (1986) a 203. VGGranikUsp.Khim. 51 207 (1982) [Russ.Chem. Rev. 51 119 (1982)] 204. VGGranikUsp.Khim. 52 669 (1983) [Russ.Chem. Rev. 52 377 (1983)] 205. MB Sommer,MBegtrup,K P Bogeso J. Org. Chem. 55 4822 (1990) 206. BRD P. 3 423 003; Chem. Abstr. 104 207 247 (1986) 207. BRD P. 3 623 533; Chem. Abstr. 109 122 478 (1988) 208. P J Victory, J Teixido, J I Borrell Heterocycles 34 1905 (1992) 209. P Victory, J Teixido, J I Borrell, N Busquest Heterocycles 35 1 (1993) 210. D G Andrez, J R Helliwell, E J Dodson, J F Pinella, G German, A Alvarez-Larena, J Teixido, P Victory Acta Crystallogr., Sect. C 48 104 (1992) 211. E C Taylor, J S Scotnicki, S R Fletcher J. Org. Chem. 50 1005 (1985) 212. G Koitz, B Thierrichter, H Junek Heterocycles 20 2405 (1983) 213. I A El-Sakka J. Chem. Res. (S) 434 (1996) 214. E G Paronikyan, A S Noravyan, S A Vartanyan Arm. Khim. Zh. 40 587 (1987) 215. E G Paronikyan, G V Mirzoyan, A S Noravyan, S A Vartanyan Khim. Geterotsikl. Soedin. 989 (1987) a 216. H Takahata, T Suzuki, T Yamazaki Heterocycles 24 1247 (1986) 217. S G Pilosyan, V V Dabaeva, B D Enokyan, E A Abgaryan, A S Noravyan Arm. Khim. Zh. 41 687 (1988) 218. Jpn. P. 01/279 884; Chem. Abstr. 112 198 358 (1990) 219. A Cangjee, K A Chemeng J. Heterocycl. Chem. 24 123 (1987) 220. A Cangjee, I O Donkor J. Heterocycl. Chem. 26 705 (1989) 221. A Cangjee, J K O'Donnell, T J Bardos, T I Kalman J. Heterocycl. Chem. 21 873 (1984) 222. A Cangjee, K A Kwasi J. Heterocycl. Chem. 22 1153 (1985) 223. T Sakamoto, Y Kondo, H Yamanaka Chem. Pharm. Bull. 33 4764 (1985) 224. T Sakomoto, Y Kondo, H Yamanaka Chem. Pharm. Bull. 33 626 (1985) 225. J A Turner J. Org. Chem. 55 4744 (1990) 226. US P. 4 472 193; Chem. Abstr. 102 6456 (1985) 227. A Straub Synth. Commun. 23 365 (1993) 228. K Sasaki, R A S Shamsur, S Kashino, T Hirota J. Chem. Soc., Chem. Commun. 1767 (1994) 229. K Sasaki, A S S Rouf, S Kashino, T Hirota Heterocycles 41 1307 (1995) 230. K Sasaki, A S S Rouf, T Hirota J. Heterocycl. Chem. 33 49 (1996) 231. T Hirota, T Matsushita,K Sasaki, S Kashino Heterocycles 41 2565 (1995) 232. G Seitz, J Richter Chem.-Ztg. 113 252 (1989) 233. G Seitz, R John Chem.-Ztg. 114 381 (1990) 234. J Epsztajn, M W Plotka, J Scianowski Synth. Commun. 22 1239 (1992) 235. A-F A Harb, A-H M Hesien, S A Metwally,M H Elnagdi Liebigs Ann. Chem. 585 (1989) 236. C N O'Callaghan, T B H McMurry, J E O'Brien, S M Draper, D J Wilcock J. Chem. Soc., Perkin Trans. 1 1067 (1996) 237. C N O'Callaghan, T B H McMurry, J E O'Brien J. Chem. Res. (S) 490 (1995) 238. F M Asaad, J Becher, J Moller, K S Varma Synthesis 301 (1987) a�Chem. Heterocycl. Compd. (Engl. Transl.) b�Theor. Exp. Chem. (Engl. Transl.) c�Russ. J. Org. Chem. (Engl. Tr

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (3) 221 ± 230 (2000) Sulfamides in the synthesis of heterocyclic compounds G A Gazieva, A N Kravchenko, O V Lebedev Contents I. Introduction II. Structure and physicochemical properties of sulfamide III. Synthesis of heterocyclic compounds containing the sulfamide fragment IV. Properties of cyclic sulfamides Abstract. physico- and structures the of analysis comparative A comparative analysis of the structures and physico- chemical properties of sulfamides and ureas has been performed. chemical properties of sulfamides and ureas has been performed. New heterocyclic of synthesis the for procedures New procedures for the synthesis of heterocyclic compounds compounds containing the and surveyed are fragment sulfamide the containing the sulfamide fragment are surveyed and the properties properties of The described.are compounds resulting the of the resulting compounds are described. The bibliography bibliography includes 112 references includes 112 references. I. Introduction The chemistry of sulfamides has received sufficient attention in the literature. The synthesis, physicochemical properties and reac- tions of these compounds have been considered many times in reviews which were partially or completely devoted to sulfa- mides.1±13 However, the latest review 13 was published more than 10 years ago. During the last decade, abundant experimental data have accumulated, which call for further generalisation. Sulfamides attract permanent interest because acyclic and cyclic products containing sulfamide fragments exhibit a broad spectrum of physiological activities.Thus sulfamoylamidines possess antiulcerous 14 and diuretic 15 activities, 4-phenyl-1,2,5- thiadiazolidin-3-one 1,1-dioxide exhibits antispasmodic activ- ity,16 5-alkyl-2-phosphinyloxymethyl-1,2,5-thiadiazolidin-3-one 1,1-dioxides are proposed for the treatment of rheumatoid arthri- tis,17 and 1,2,5-thiadiazolidine 1,1-dioxides containing an indole substituent at position 2 are used for the treatment of migraine.18, 19 5-Alkyl-2-fluoromethyl-1,2,5-thiadiazolidin-3-one 1,1-dioxides inhibit enzymes, viz., human leucocyte elastase and cathepsin G.20 6,7-Dihydro-5H-cyclopenta[c]- and 5,6,7,8-tetra- hydrobenzo[c]-2,1,3-thiadiazine 2,2-dioxides act as myorelax- ants,21 and 3,5-dioxo-1,2,6-thiadiazine 1,1-dioxides are used as antiinflammatory agents.22 Aryl-substituted seven- and eight- membered cyclic sulfamides inhibit HIV-1 protease.23, 24 Many sulfamides possess antibacterial properties.25, 26 Sulfamides are also used in photography,27 they are compo- nents of fungicide 28 and insecticide 29 mixtures and are employed as detergents.30 3,5-Dioxo- and 3,5-diimino-1,2,6-thiadiazine 1,1- dioxides are used for the preparation of azo dyes.31 3-Amino- 1,2,6-thiadiazine 1,1-dioxides exhibit fungistatic activities.32 G A Gazieva, A N Kravchenko, O V Lebedev N D Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, Leninsky prosp.47, 117913 Moscow, Russian Federation. Fax (7-095) 135 53 28. Tel.(7-095) 938 35 79 Received 25 November 1999 Uspekhi Khimii 69 (3) 239 ± 248 (2000); translated by T N Safonova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n03ABEH000562 221 221 222 227 S,S-Dioxides of benzothiadiazine, thiadiazine, and thiatriazine as well as some acyclic sulfamides can serve as herbicides.33 ± 35 In the 1980s, a new class of chiral oxidising agents based on sulfamides was found. These are 2-sulfamoyloxaziridines which exhibit high enantioselectivity (40% ± 91%) in asymmetric oxida- tion of sulfides to sulfoxides.36 ± 38 It is believed 1, 7, 14, 39 ± 42 that the increased interest in sulfa- mide derivatives stems from the fact that these compounds are structurally similar to ureas.In this connection, we thought that it was worthwhile performing a comparative analysis of the struc- tures and physicochemical properties of sulfamide and urea. However, the major aim of the present review is to systematise procedures for the synthesis of heterocyclic compounds based on sulfamide. Wherever possible, a comparative analysis of the properties of cyclic sulfamides and of the corresponding cyclic ureas was carried out. II. Structure and physicochemical properties of sulfamide Whereas the electronic structure of the carbonyl group has been studied comprehensively and described in detail, the data on the electronic structure of the sulfonyl group are ambiguous. The character of the S7O bond remains unclear, and the question of whether this bond is purely polar, semipolar, covalent or `inter- mediate' is still open.The question about the nature of the sulfonyl groups is closely related to the question about the participation of the d orbitals of the sulfur atom in bonding with other elements and to the question of whether the sulfur atom can exhibit `hypervalence' thus violat- ing the Lewis octet rule.43 Quantum-chemical calculations with the 3-21G* and 6-31G* basis sets (i.e., taking into account the contribution of the d orbitals) for a series of model compounds containing the SO and SO2 groups demonstrated that the contribution of the d orbitals to the formation of the S7O bond is insignificant.44, 45 Thus Reed and von Rague' Schleyer 44 concluded that the character of the S7O bond is determined primarily by the partially ionic s bond and the p ± s* `negative hyperconjugation'.However, studies of the electronic structures of sulfamides and heterocyclic com- pounds containing the SO2 group by calculation methods dem- onstrated that pp ± dp interaction with the participation of the d orbitals of the sulfur atom contributes to the formation of the multiple S7O bond.45 ± 50 According to these calculations, the S7O bond has two components of the s and p types, viz., the s bond between the corresponding atomic orbitals of the oxygen and sulfur atoms and the p bond between, on the one hand, three lone electron pairs of the oxygen atom polarised toward the sulfur222 atom, and, on the other hand, the p and d orbitals of the sulfur atom.However, the S7O bond is not, strictly speaking, a double bond because the p type of interaction is pp ± dp bonding with the participation of all three electron pairs of the oxygen atom. Nevertheless, as in Ref. 43, we shall arbitrarily denote these bonds as O=S=O. It should be remembered that these `double' bonds differ substantially from true double bonds formed by elements of the second row (for example, from C=O bonds) and are characterised by a substantial contribution of the ionic state.43 ± 50 The specificity of the structure of the sulfo group is reflected in the spatial configuration of the atoms in the sulfamide molecule as well as in the electron density distribution.Recently, the crystal structure of sulfamide has been studied by X-ray and neutron diffraction analysis (Table 1).49, 51, 52 According to the results of these studies, the S and N atoms in the sulfamide molecule have tetrahedral and pyramidal bond configurations, respectively, whereas the nitrogen atom in amides and urea is characterised by sp2 hybridisation and a planar bond configuration.52 The urea molecule is planar and symmetrical. The bond angles in this molecule differ little from 120 8.53 This structure is consistent with the notion of conjugation between the p electrons of the C=O double bond and the lone electron pairs of the nitrogen atoms. This conjugation cannot occur in the sulfamide molecule because the sulfur atom does not possess unoccupied orbitals for participation in electron delocalisation.It has been demon- strated 45, 47 that the d orbitals of the sulfur atom are not involved in bonding with the S andNatoms in sulfamides. It was also found that sulfonamides cannot adopt the limiting resonance structure 2 analogous to the structure 4 for carboxamides.57 R1 R1 + R2 R2 N N H H S S 7 O O O O 2 1 R1 R1 + R2 R2 N N H H C C 4 3 O7 O The electron-withdrawing effect of the SO2 group is stronger than that of the carbonyl group, and hence the acidity of the amino groups in sulfamide is higher than that in urea. Thus the amino groups in urea exhibit a weakly basic character, whereas Table 1. Comparative characteristics of the sulfamide and urea molecules.Urea Sulfamide Parameter 1.276 1.356 1.046 1.070 1.434 1.614 1.034 1.038 C7O C7N N7H N7H0 Bond lengths /A S7O S7N N7H N7H0 Bond angles /deg O7S7O0 118.0 O7S7N 106.2 O07S7N 107.0 N7S7N0 111.7 H7N7H0 112.5 O7C7N 120.9 N7C7N0 118.2 C7N7H 122.5 C7N7H0 120.5 H7N7H0 117.0 Dihedral angle /deg OSO/NSN 89.39 N7H 3342 ± 3449 1681 Stretching vibrations /cm71 SO2 3210 ± 3372 N7H 1315 ± 1369, C=O 1145 ± 1155 870 ± 955 1004, 1466 CN SN2 Note. The geometric parameters for the sulfamide molecule were taken from Refs 51 and 52, the parameters for the urea molecule were taken from Ref. 53, and the stretching vibration frequencies were published in Refs 53 ± 56. G A Gazieva, A N Kravchenko, O V Lebedev these groups in sulfamide are weakly acidic.Urea forms salts with acids 58 and sulfamide gives salts with heavy metals (for example, AgNHSO2NHAg) and alkali metals.1, 12, 59 These data agree with the Taft induction constants s1 for the electron-withdrawing substituents in dioxane (s0) and in weakly acidic solvents (s00).60 1 1 Group 00 s s1 1 0 0.11 0.38 CONH2 SO2NH2 0.210.02 0.460.03 Therefore, sulfamide can be considered as an analogue of carbamide only bearing in mind that the N7SO2 conjugation is actually not realised in sulfamide by virtue of electronic and steric factors and that, in addition, the NH bond in sulfamide is substantially more acidic, and, correspondingly, the nitrogen atom is less basic than those in urea.The results of ab initio calculations 45 ± 50 of the geometry, dipole moments, atomic charges and bond orders in sulfamides and heterocyclic compounds containing the NSO2N fragment agree well with the experimental data. In the crystal, the sulfamide molecules, like the urea molecules, are linked together through strong hydrogen bonds,51, 52, 54, 58 which may be responsible for the rather high melting points of these compounds. In solutions of sulfamides, these hydrogen bonds are substantially weaker and are observed only in water, which is confirmed by the data from IR and Raman spectro- scopy.55 The 15N, 14N and 33S NMR spectra of sulfamide have also been reported.61, 62 III. Synthesis of heterocyclic compounds containing the sulfamide fragment Since heterocyclic systems containing the NSO2N fragment are highly diversified, we believe that it is more convenient to classify procedures for their synthesis according to the initial compounds, as in Ref.12, rather than according to the resulting products, as in Ref. 13. Sulfamide and urea are not as strong nucleophiles as ammonia or amines; nevertheless, they can also react with electrophilic reagents. Reactions of sulfamides with carbonyl compounds, diamines, dinitriles and other bifunctional reagents are considered in this Section. 1. Reactions of sulfamides with carbonyl compounds One of the most general procedures for the synthesis of hetero- cyclic compounds based on sulfamides involves their condensa- tion with carbonyl compounds 39, 63 ± 72 or with their acetals.41, 42, 73 Formaldehyde,7, 12, 13, 63, 65 paraformaldehyde 64 and triox- ane 63 are most commonly used in the synthesis of heterocyclic compounds containing methylene groups.N SO2 N N conc. HCl or 60% H2SO4 O2S N5 N N N NH2 H2SO4 SO2 SO2 SO2 SO2+CH2O [or (CH2O)3] N N N NH2 6 N NH3 or (CH2)n N N H2N(CH2)2NH2 N O2S 7 (n=1, 2)Sulfamides in the synthesis of heterocyclic compounds Reactions of alkylsulfamides with paraformaldehyde in tri- fluoroacetic acid occur through the intermediate formation of 1,5- dithia-2,4,6,8-tetraazocine 1,1,5,5-tetroxides 8 to yield 2,6-dithia- 1,3,5,7-tetraazabicylco[3.3.1]nonane 2,2,6,6-tetroxides 9.64 O O O O S S NH2 N AlkN NH AlkN (CH2O)n SO2+(CH2O)n NAlk NAlk N HN NHAlk S S O O O O 9 8 Condensation of sulfamides with formaldehyde and primary amines afforded tetrahydro-1,2,4,6-thiatriazine 1,1-dioxides 10 in 60%± 86% yields.65 NH2 RN2 SO2+CH2O+R2NH2 HN NHR1 S NR1 O O 10 R1=H, OAlk, OAr; R2=Alk, Ar.Ureas also react with formaldehyde to form a wide variety of products (both polymeric and heterocyclic) including compounds analogous to compounds 8 and 10.7, 74, 75 3,7-Bis(ethoxycarbonyl)perhydro-1,5-dithia-2,4,6,8-tetraazo- cine 1,1,5,5-tetroxide (11) was prepared from sulfamide and glyoxylic acid followed by esterification.66 H2NSO2NH2+OHCCOOH H2O 50 8C, 5 h O O S NH HN COOH HOOC EtOH H2SO4 H H HN S NH O O O O S NH HN COOEt EtOOC H H HN S NH 11 (29%) O O 3-Imino-4-substituted 1,2,5-thiadiazolidine 1,1-dioxides (12) were formed upon boiling of solutions of aldehydes, sodium (or potassium) cyanide and sulfamide in aqueous ethanol.39 Presum- ably, the reaction proceeds as follows: SO2NH2 O NH2 N M+CN7, H+ RCH+ SO2 C H R NH2 O O SO2NH2 S HN NH HN C N C R H NH RH 12 (12% ± 62%) R=n-C6H13, PhCH2CH2, Ph, 4-MeOC6H4, 3,4,5-(MeO)3C6H2, , , , .S N This reaction did not take place with paraformaldehyde, trioxane, acetone and acetophenone. 223 O O 3-Oxo derivatives 13 are prepared from the corresponding 3-imino-1,2,5-thiadiazolidine 1,1-dioxides. SO2NH2 S HN OEt NH HN EtOH/HCl MeOH/MeONa 12 C C C C O RH O RH 13 Unlike sulfamide derivatives 13, 2,4-dioxoimidazolidines 14 are readily formed from ureas and a-dicarbonyl compounds in an acidic medium.7, 76 O NR RN O 14 Over many years, the reactions of sulfamides with a- and b-dicarbonyl compounds have also been widely used in the synthesis of heterocyclic compounds.5, 12, 13, 67 ± 71 Thus the reac- tion of N,N0-dimethylsulfamide with glyoxal in an acidic medium (pH 3 ± 4) afforded the first representative of bis(sulfamides) of the bicyclooctane series, viz., 2,4,6,8-tetramethyl-3,7-dithia- 2,4,6,8-tetraazabicylco[3.3.0]octane 3,3,7,7-tetroxide (15), (the yield was 23%).67 R3 Me Me R1 R5 O O N N N N S S O O N N N N O O Me R4 Me R2 R6 16 15 It should be noted that analogous bicyclic bisureas 16 have been prepared long ago from ureas and a-dicarbonyl compounds in 91% yield.Over the past century, tens of representatives of this class of compounds have been synthesised,7, 77 whereas sulfamide analogues have been previously unknown. With the aim of increasing the yield of compound 15, we performed the reactions of N,N0-dialkylsulfamides with glyoxal monohydrate in 36.7% hydrochloric acid. However, new hetero- cyclic compounds, viz., 3,30-bis(6,8-dialkyl-2,4-dioxa-7-thia-6,8- diazabicyclo[3.3.0]octane 7,7-dioxides) (17, the yields were 54%± 85%), were obtained instead of the expected products 15. The structures of the products 17 were confirmed by X-ray diffraction analysis. R R O O N O N O S S N O O N O O 17 R R R=Me, Et, Pr, Bu.The reaction of N,N 0-dimethylurea with glyoxal afforded another product, viz., 4,6,12,14-tetramethyl-2,8,10,16-tetra- oxa-4,6,12,14-tetraazatetracyclo-[7.7.0.03,7011,15]hexadecane-5,1- 3-dione (18).78 Me Me O O N N O O N N O O 18 Me Me Reactions of sulfamide with b-diketones yielded various 1,2,6- thiadiazine 1,1-dioxide derivatives.70, 71224 NNHC6H4SO2NHR1 HCl R3 H2NSO2NH2+ R2 O O N NC6H4SO2NHR1 R2 R3 HN S N O O 19 (50% ± 68%) R1=H, CH2CONH2, C(NH2)=NH, Ar; R2=Me, Ph; R3=Me, Ph. C(O)Me MeOH, HCl H2NSO2NHR+ (H2C)n O Me Me R N N (H2C)n (H2C)n + SO2 SO2 N NR 21 (15% ± 43%), 23 (47%) 20 (7%), 22 (41% ± 74%) R=H, Me; n = 1 (20, 21), 2 (22, 23).Reactions of sulfamide with symmetrical b-diketones afforded one type of products (19, R2=R3); however, the reactions with unsymmetrically substituted b-diketones gave mixtures of tauto- mers (20 ± 23, R=H). Moreover, the reactions of unsymmetri- cally substituted b-diketones with monosubstituted sulfamides afforded mixtures of regioisomers (20 ± 23, R=Me).70 ± 72 The reactivity of b-diketones with respect to sulfamide depends on both the electronic and steric characteristics of the substituents in the diketones. Thus 3,5-diphenyl-2H-1,2,6-thia- diazine 1,1-dioxide and its 2-butyl derivative were prepared in 45% and 39% yields, respectively, whereas 3,5-di-tert-butyl-2H- 1,2,6-thiadiazine 1,1-dioxide has not been synthesised yet.72 It was demonstrated 72 that the reactions of b-amino- and b-chloro-a,b-unsaturated ketones with sulfamide occurred regio- specifically under milder conditions than the reactions with the corresponding b-diketones.In anhydrous ethanol in the presence of HCl, these compounds gave a series of 2H-1,2,6-thiadiazine 1,1- dioxides 24 and 25. R R R R H2NSO2NH2 N 60 8C, 3 ± 6 h O NH2 S NH O O 24 (83% ± 87%) R=Me, Ph, But. R2 R1 60 8C R2 R1 3 ± 24 h O NH2 H2NSO2NHBz NBz N S R2 R1 O refluxing 2± 3 h Cl O O 25 (47% ± 85%) R1=Alk, Ar; R2=H, Me. In the reactions of ureas with b-dicarbonyl compounds, bicyclic products 26 were formed more readily than 2-oxodihydro- pyrimidines 27, whereas the reactions with b-chloro-a,b-unsatu- rated aldehydes afforded only derivatives 27.7, 79 R2 R1 N N NR N O O O N N 27 26 R4 R3 G A Gazieva, A N Kravchenko, O V Lebedev The use of acetals of carbonyl compounds instead of these compounds as such in the reactions with sulfamides resulted in the formation of six- and eight-membered N,S-containing heterocyc- lic compounds analogous to those described above along with new unexpected a-sulfamidoalkylation intramolecular prod- ucts.41, 42, 73 Presumably,42, 73 the mechanism of these reactions is analogous to the mechanism of a-ureidoalkylation.7 Reactions of sulfamides with acetals in an acidic medium afforded iminuim ions 28 and 29, H+ R1HNSO2NHR2+R4CH(OR3)2 7H+ + + R1NHSO2NR2+R1NSO2NHR2 29 28 CHR4 CHR4 R1=H:R2=H, (CH2)nPh, CH2C6H4OMe-3; R1=R2=(CH2)nPh (n=0 ± 4); R3=Me, R4=H; R3=Et: R4=H, CO2Et, CH2CO2Et; which underwent intramolecular or intermolecular a-sulfamidoal- kylation depending on the nature of the substituent in the sulfamides and the reaction conditions.In the case of a-sulfamidoalkylation of the aromatic rings in the iminium ions, annelation products 30, 35 ± 38 and 41 were formed. If the aromatic substituent in sulfamide is either remote from the generated iminium ion or absent, intermolecular dimer- isation occurs to form eight-membered heterocyclic compounds 11, 34 and 40. Under certain conditions, these products can react with acetals to form compounds 5 and 31 ± 33. Symmetrical diarylsulfamides react with acetals (EtO)2CH(CH2)mCOOEt (m=0 or 1) to form new sulfamide 39 as a result of intramolecular a-sulfamidoalkylation.O HN S R1=Ph, O R4=CO2Et NH H CO2Et 30 (56%) O O S NR1 N R1=Ph, CH2Ph, (CH2)4Ph, R4=H R1N S N O + O 31 ± 33 R1NHSO2NH O O CHR4 S 28 (R2=H) NR1 HN CO2Et EtO2C R1=H, CH2Ph, R4=CO2Et H H R1N S NH O O 11, 34O N S R1=R4=H O N N O N S 5 (95%) O Yield (%) R1 Compound Ph CH2Ph (CH2)4Ph 46 72 56 H 7574 31 32 33 11 34 CH2PhSulfamides in the synthesis of heterocyclic compounds + R1NSO2NHR2 29 CHR4 R1 = (CH2)nPh (n=2, 3), R2=H, R4=H, CO2Et R1=R2=(CH2)3Ph, R4=H R1=H, R2=CH2C6H4OMe-3, R4=CH2CO2Et n R4 Compound 2 H 3 H 35 36 37 38 39 88 2 CO2 Et 8085 3 CO2 Et 11 3 CO2 Et 68 The reactions of unsubstituted sulfamide and arylsulfamides with two equivalents of an acetal in an acidic medium afforded 1,2,6-thiadiazines 42 ± 46, while the reaction with one equivalent of an acetal yielded 1,2,6-thiadiazine 47.2 (EtO)2CHCH2CO2Et, R=H, (CH2)mPh (m=0± 3) RHNSO2NH2 (EtO)2CHCH2CO2Et, Compound R 42 43 44 45 46 H 38 Ph 42 CH2Ph 68 (CH2)2Ph 37 (CH2)3Ph 79 2. Reactions of sulfamides with diamines An efficient procedure for the synthesis of heterocyclic com- pounds containing the sulfamide fragment involves condensation of sulfamide with diamines.12, 13, 23, 24, 80 This procedure allows one to prepare heterocyclic compounds with five, six or more atoms in the ring.Note that this procedure is the most convenient method for the synthesis of the latter compounds. RHNSO2NH2 refluxing B H2NCHCH2CHCH2CHNH2 OR PhH2C CH2Ph (CH2)n N (CH2)n NSO2NH2+ SO2 2 NH2 H2N Ph Ph H R4 39 H R4 35 ± 38 O O O O S N(CH2)3Ph Ph(CH2)3N Me Me N(CH2)3Ph Ph(CH2)3N S (CH2)nNH2 O O 40 (80%) (CH2)nNH2 CH2 O Me O (n= 0, 1) O NHS NH 41 CH2CO2Et Yield (%) NH2 NH2 O O B is an organic base. O O Heating of ureas with diamines is also used in the synthesis of heterocyclic compounds.81 S RN NH 3. Reactions of sulfamides with dinitriles CH2CO2Et CO2Et 42 ± 46 O O S NH HN The reactions of sulfamides with dicyan and malonodinitriles find wide application.13, 82 ± 86 Numerous examples of the preparation of diamino and diimino derivatives of 1,2,5-thiadiazole 1,1-diox- ides and 1,2,6-thiadiazine 1,1-dioxides (52 ± 55) are documented (Scheme 1).Acid hydrolysis of compounds 52 and 54 afforded the corre- sponding 3-oxo derivatives.85, 86 R=H O 47 4. Condensation of sulfamides with diols and other bifunctional reagents Yield (%) Until recently, condensation of dihydroxy compounds with sulfa- mides have not found wide use, whereas these reactions with ureas have long been known 7 and they are successfully used, for example, in the synthesis of bicyclo[3.3.0]octanediones (16). Procedures for the preparation of bi-, tri- and tetracyclic compounds (57, 59 and 61, respectively) containing sulfamide fragments from sulfamides and the corresponding dihydroxy compounds 56, 58 and 60, respectively, have been developed in recent years.67, 87 R1 NHR2 OH N H+ O SO2 + N OH NHR2 56 R1 R1=Alk; R2=H, Alk.225 O O S NH HN Ph Ph 48 OR O O S NH HN Ph Ph O OMe Me 49 O CH2 (CH2)n (CH2)n NH HN S 50 O O O O NH HN S 51 O O R2 R1 O N N S O N N O R1 R2 57 (9% ± 63%)226 H2NSO2NHR1 R1=H, Alk, Ar, ArAlk; R2=R3=H, Hal, Ar, Alk, ArAlk. HOH2CN O N Me 58 HOH2CN O N HOH2C 60 A method for the synthesis of tetrahydro-3,5-dioxo-1,2,6- thiadiazine 1,1-dioxides (62) from malonyl chloride and N-sub- stituted sulfamides was also proposed.35, 88 O NHR2 Cl SO2 + Cl C NHR1 R1=(CH2)nPh (n=0 ± 2), C6H11; R2=H, CH2Ph.An interesting procedure for the preparation of 5-methyl-3- oxo-3,6-dihydro-2H-1,2,6-thiadiazine 1,1-dioxide (63) was described.89, 90 HN NC CN R1NO HCl R2 R3 C R1N CN NC H2N CH2OH NH2 N H+ SO2 O + N NH2 O Me CH2OH NHMe N H+ O +2 SO2 N NHMe CH2OH O O S C R1N NR2 O O O 62 (70% ± 90%) HN NH R1N NH S O O 52 (62% ± 76%) R1=H, O O R2=R3=H, Hal, Ar, Alk, ArAlk N S NH2 R2 R3 R1=H, Alk, Ar, ArAlk, R2=R3=H O O S NH HN N N O N N Me Me 59 (80%) O O S NMe MeN N N O O N N MeN NMe S O O 61 (72%) G A Gazieva, A N Kravchenko, O V Lebedev NH2 H2N H2/Pd N N R1=CH2Ph S NH2 O O 53 (96%) N S O NH2 O HCl H2O/EtOH R1N N S O O O O N S N NH2 H2N R2 R3 55 O O H2/Pd R1=CH2Ph R1N N S NH2 H2O, H+ H2N54 (91% ± 95%) O NPhH NaOH PhN SO2 + O NH2 H2C O O NH2 A method for the synthesis of 3-oxo-3,6-dihydro-2H-1,2,6- thiadiazine 1,1-dioxides (64) from sulfamides and 5-alkanoyl-1,3- dioxane-4,6-diones was covered by a patent.91 OC O R1 SO2 + R2 O O NHR4 R3 R1=R2=R3=Alk; R4=H, Alk, Ar.The reactions of derivatives of 1,3-dioxane-4,6-diones with urea and thiourea proceed analogously.92 The use of a new synthon, viz., diphenyl N-sulfamoyl- carboimidate, which was prepared from sulfamide and dichloro- diphenoxymethane, opens a new way to the synthesis of 1,2,6- thiadiazine 1,1-dioxide (65) and 1,2,4,6-thiatriazine 1,1-dioxide (66).93 O O Cl Cl MeCN S + OPh PhO NH2 H2N O O K2CO3 PhO +CH2(CN)2 S N NH2 PhO O O O O S S HN N NH2 NH2 CN CN PhO PhO CN CN Scheme 1 O ON N S NH2 H2N O O N S R1N NH2 O(84% ± 91%) O O S NHMe 63 (63%) O O S NR4 HN O R1 64 O O PhO S N NH2 PhO O O S NH N HCl PhO NH2 CN 65 (53% ± 65%)Sulfamides in the synthesis of heterocyclic compounds O O O O S S NH3 N N NH2 NH2 +R2NH OPh OPh PhO R2N O O O O S S N N N NH2 HC(OEt)3 Ac2O, DMF R2N R2N NH2 HN 66 (22% ± 57%) ., N R2N= N Of the currently available approaches to the construction of cyclic sulfamides, the reactions of sulfuryl halides 94 ± 98 and chlorosulfonyl isocyanate 20, 99 with amino compounds should also be mentioned.The preparation of 3-oxo-1,2,5-thiadiazolidine 1,1-dioxides (13) from chlorosulfonyl isocyanate and amino acids is an example.99 OCN SO2Cl HN CO2R2 PhCH2O1) PhCH2OH, 2) Et3N, CH2Cl2, 3) HCl .H2NCH(R1)CO2R2 H S N 5% Pd/C, MeOH 60 atm H2 O O O R1 O O S MeONa, MeOH HN NH H2N CO2R2 O H S NO R1 O R1 13 R1=H, Me, Bu, CH2CH2SMe. The above-considered experimental data demonstrate that sulfamides are rather often similar in their chemical behaviour to ureas. Sulfo-analogues of many urea-containing compounds have been prepared. IV. Properties of cyclic sulfamides Recently, the aromaticity of heterocyclic compounds has been the subject of many theoretical studies.47, 49, 50 The results of ab initio quantum-chemical calculations of the electronic properties and the geometry of these molecules, which were confirmed by X-ray diffraction analysis,47, 49, 50, 100, 101 demonstrated that the rings in 1,2,5-thiadiazole 1,1-dioxide and 1,2,6-thiadiazine 1,1-dioxide, unlike those in 2-pyrimidinones, are nonaromatic.47 The lack of the aromaticity results from the specificity of the structure of the SO2 group (see Section II).The sulfur atom cannot be involved in p delocalisation because the oxygen and sulfur atoms in thiadi- azines (67) deviate form the plane passing through the remaining atoms. NH N N S NH O O O 67 68 The tautomerism 39, 47, 71 and rotational isomerism 40, 102 of cyclic sulfamides attract considerable interest.The difference in the stability of the tautomers of 1,2,6-thiadiazine 1,1-dioxide (67) and 2-pyrimidinone (68) results from the aromaticity of the compound 68 and nonaromaticity of the compound 67.47 According to the data from X-ray diffraction studies and NMR spectroscopy, 3,5-dialkyl- and 3,5-diaryl-substituted 1,2,6- thiadiazine 1,1-dioxides exist in solutions predominantly as the isomers A, whereas the tautomers B of 3(5)-hydroxy- and 3(5)- amino-substituted derivatives are more stable.47 In no case was 227 the tautomer C experimentally observed. In the case of substituted 2-pyrimidines, the stabilities of tautomers A0 ±C0 are related in a different way. R1 R2 R1 R2 R1 R2 > N N N N N S S S NH OH O O O O O A B C R1 R2 R2 R1 R2 R1 > N N N N NH N O OH O B0 C0 A0 Tautomeric forms of 3-oxo- (13) and 3-imino-1,2,5-thiadi- azolidine 1,1-dioxides (12)39 as well as those of 6,7-dihydro-5H- cyclopenta[c]- (20 and 21) and 5,6,7,8-tetrahydrobenzo[c]-2,1,3- thiadiazine 2,2-dioxides (22 and 23)71 were studied.O O O O S S NH HN NH N H H HO O R R 13 130 O O O O S S NH N NH HN H H HN H2N R 12 120 R Me Me H N N (H2C)n (H2C)n O O S S O O NH N 21, 23 20, 22 n=1, 2. In solution, 3-oxo- and 3-imino-1,2,5-thiadiazoles occur as equilibrium mixtures of the tautomers 12 and 120 and the tautomers 13 and 130, respectively, whereas 6,7-dihydro-5H- cyclopenta[c]-2,1,3-thiadiazine 2,2-dioxide and 5,6,7,8-tetrahy- drobenzo[c]-2,1,3-thiadiazine 2,2-dioxide exist predominantly in the forms 21 and 22, respectively.The rotational isomerism of glycosylated derivatives of 1,2,6- thiadiazine 1,1-dioxides at the N-glycosidic bond was studied by NMRspectroscopy. Thus 6-b-D-glucopyranosyl-2-methyl-3-oxo- 3,6-dihydro-2H-1,2,6-thiadizaine 1,1-dioxides (69) exist in solu- tions as mixtures of syn and anti rotamers.40 O Me AcO O S N O O N AcOAcO OAc sin-69 Me AcO Me O O N AcOAcO S N OAc O Me O anti-69 It was found that the energy barriers to rotation about the glycosidic bond for glycosylated thiadiazine derivatives are lower than those for the corresponding uracils and pteridines.40, 102 This is attributed to the fact that the SO2 group deviates from the plane of the heterocycle and hinders the rotation about the C7N bond to a lesser extent.228 Under the action of various reagents, six- and seven-mem- bered cyclic sulfamides are transformed with elimination of the NSO2N,103, 104 SO2 (see Ref. 105) or CO fragments.88 Thus derivatives of 1,2,6-thiadiazine 1,1-dioxides were converted under the action of nucleophiles into pyrazoles 70 103 and 2-pyr- idones 71.104 O O R2 Me S NR1 PhN H2O, D R1NHSO2NH2+N +NH2NH2 Me N Me Me R2 H 70 O O S NH HN +RCH2NH2 CH2CO2Et 42 CO2Et CH2RO N +EtOH H2NSO2NH2 + EtO2C 71 Under the action of oxidising agents, seven-membered cyclic sulfamides, which do not contain substituents at the nitrogen atoms were converted into pyridazine derivatives 72 with elimi- nation of the SO2 fragment,105 MeO MeO O N NaOCl aq.NaOH N O NHS NH R R 72 whereas successive nitrosation and acidification of N-substituted tetrahydro-3,5-dioxo-1,2,6-thiadiazine 1,1-dioxides afforded 2-substituted 4-amino-2,3-dihydro-3-oxo-1,2,5-thiadiazole 1,1- dioxides (73) with elimination of the CO group.88 O O O O O O S S S NR HN NR HN AcOH/NaNO2 H+ NR N O O O O O H2N 73 NOH R=Ar, cyclo-C6H11. Substitution at the nitrogen atoms is typical of cyclic sulf- amides. It was demonstrated that heterocyclic sulfamides undergo alkylation,20, 24, 41, 69, 73, 93, 106, 107 acylation,69 nitration,94 hydr- oxymethylation,20 silylation 40, 41, 102 and glycosylation.40, 41, 102 Diazomethane,106 dimethyl sulfate 107 and alkyl and aryl halides 20, 24, 41, 69, 73, 93, 107 are used as alkylating agents.Proce- dures for alkylation with methyl iodide in acetone in the presence ofK2CO3 or with benzyl chlorides and bromides in the presence of sodium hydride are most generally used. O O H S N N HCOOEt H MeI, K2CO3 H MeCOMe N H EtOOCH S NO O O O Me S N N Me COOEt H H N S N Me EtOOC Me O O G A Gazieva, A N Kravchenko, O V Lebedev O O H H S N N NaH, DMF Bz Bz R=4-MeOC6H4CH2. 1,2,5-Thiadiazolidine 1,1-dioxides were acylated with acyl chloride or dimethylcarbamoyl chloride in an alkaline media,69 nitrated with nitric acid 94 and hydroxymethylated with a meth- anolic solution of formaldehyde.20 1,2,6-Thiadiazine 1,1-dioxides were silylated with hexame- thyldisilazane.Their silyl derivatives underwent glycosylation under the action of penta-O-acetyl-b-D-glucopyranose and 1,3- oxathiolane.40, 41, 102 O Me N NH(SiMe3)2 O2S Me N AcO H O Me N O2S Me NSiMe3 O R H N NH(SiMe3)2 O2S Py NH PhCO2 O H R N O2S NSiMe3 Halogenation and arylation of 1,2,6-thiadiazine 1,1-dioxides 25 (R1=R2=H) at the carbon atom using organometallic reagents have been reported.108 Amine exchange in 3,4-diamino- 1,2,5-thiadiazole 1,1-dioxides 53 has been described.83 O O H2N S NBz N N S O 25 53 The sulfo group is stable to reducing agents.5, 43 The SO2 group also remains intact upon electroreduction.Thus electro- reduction of 3,4-diphenyl-1,2,5-thiadiazole 1,1-dioxide in aceto- nitrile afforded the corresponding dihydrothidiazole 1,1-dioxide (74),109 i.e., only the C=N bond was reduced. Ph Ph Ph N N N S O O O All the above-described reactions are also typical of cyclic ureas. The latter are alkylated with alkyl halides and alkyl sulfates O O R R S N N Bz Bz RCl O Me N OAc O OAc O2S Me N AcO AcO OAc O OAc AcO OAc 69 O H R N OAc O O2S N PhCO2 S O S NH2 NO HPh NH S O 74Sulfamides in the synthesis of heterocyclic compounds and undergo acylation, hydroxymethylation, nitration and glyco- sylation at the nitrogen atoms.7, 110 ± 112 * * * The examples considered in this review demonstrate that the chemistry of cyclic sulfamides is progressing successfully.Proce- dures for the preparation of heterocyclic compounds which contain different numbers of carbon atoms in the ring and bear different substituents at the nitrogen and carbon atoms have been developed. One would expect further progress in this field of chemistry. New possibilities of the reactions of sulfamides with carbonyl compounds yielding bi- and polycyclic sulfamides would be expected to be opened. Various useful properties of sulfamides give promise that new approaches to their directed synthesis will be developed. References 1. L F Audrieth,M Sveda, H H Sisler,M J Butler Chem.Rev. 26 49 (1940) 2. A Dorlars, in Methoden der Organischen Chemie (Houben Weyl) Vol. XI/II (Stuttgart: Georg Thieme Verlag, 1958) p. 643 3. F L Scott,W J Spillane Mech. React. Sulfur Compd. 2 133 (1968) 4. W J Spillane, F L Scott Mech. React. Sulfur Compd. 5 59 (1970) 5. A Lawson, R B Tinkler Chem. Rev. 70 593 (1970) 6. W J Spillane Int. J. Sulfur Chem. 8 469 (1973) 7. H Petersen Synthesis 243 (1973) 8. E Fluck,W Haubold, in Gmelin Handbuch der Anorganische Chemie. New Supp. Ser Vol. 32 (EdsMBecke, H Bitterer) (Berlin: Springer, 1977) 9. R Appel, J Kohnke, in Methods of Chemistry Vol. 7, Part B (Ed. F Korte) (New York: Academic Press, 1978) p. 744 10. K K Andersen Comprehensive Organic Chemistry Vol. 3 (Ed. J D Neville) (Oxford: Pergamon Press, 1979) p.363 11. G A Benson,W J Spillane Chem. Rev. 80 151 (1980) 12. S D McDermott,W J Spillane Org. Prep. Proc. Int. 16 49 (1984) 13. V J Aran, P Goya, C Ochoa Adv. Heterocycl. Chem. 44 81 (1988) 14. Y Yanagisawa, Y Hirata, Y Ishii J. Med. Chem. 30 1787 (1987) 15. P Ahuja, J Singh, M B Asthama, V Sardana, N Anand Indian J. Chem., Sect. B 28 1034 (1989) 16. C-H Lee, H Kohn J. Pharm. Sci. 79 716 (1990) 17. US P. 5 541 168; Ref. Zh. Khim. 3 O 198P (1999) 18. Eur. P. 501 568; Chem. Abstr. 118 6977 (1993) 19. J L Castro, R Baker, A R Guiblin, S C Hobbs,M G N Russell, M S Beer, J A Stanton, K Scholey, R J Hargreaves,M I Graham, V G Matassa J. Med. Chem. 37 3023 (1994) 20. W C Groutas, R Kuang, R Venkataraman Biochem.Biophys. Res. Commun. 198 341 (1994) 21. A Castro, A Martinez, I Cardelus, J Llenas, in The VIIIth Spanish Symposium on Therapeutical Chemistry (Abstracts of Reports), Salamanca, 1993 p. 43 22. P Goya, R Nieves, C Ochoa Rev. R. Acad. Sci. 78 373 (1984); Chem. Abstr. 104 207 234 (1986) 23. US P. 5 506 355; Chem. Abstr. 125 34 038 (1996) 24. P K Jadhav, F J Woerner Tetrahedron Lett. 36 6383 (1995) 25. P O Chastenet, J Buxeraud J. Lyon Pharm. 48 305 (1997) 26. V Ordeanu,M Ionica, A Andries Farmacia 45 (5) 63 (1997); Chem. Abstr. 128 228 383 (1998) 27. Jpn. P. 62 183 450; Chem. Abstr. 108 29 358 (1988) 28. Chin. P. 1 113 758; Chem. Abstr. 124 325 421 (1996) 29. Jpn. P. 10 203 902; Chem. Abstr. 129 171 855 (1998) 30. US P. 5 753 599; Chem. Abstr.129 17 272 (1998) 31. BRD P. 4 208 535; Chem. Abstr. 118 61 519 (1993) 32. I Alkorta, P Goya, C Nombela, R Medina, C Perez Martin Arzneim.-Forsch. 41 264 (1991); Chem. Abstr. 115 8744 (1991) 33. G Hamprecht, K-H KoÈ nig, G Stubenrauch Angew. Chem. 93 151 (1981) 34. Eur. P. 549 079; Chem. Abstr. 120 298 645 (1994) 35. Br. P. 2 213 145; Chem. Abstr. 112 55 921 (1990) 36. F A Davis, J P McCauley Jr , M E Harakal J. Org. Chem. 49 1465 (1984) 229 37. F A Davis, ODStringer, J P McCauley Jr Tetrahedron 41 4747 (1985) 38. F A Davis, J P McCauley Jr, S Chattopadhyay,M E Harakal, J C Towson,W H Watson, I Tavanaiepour J. Am. Chem. Soc. 109 3370 (1987) 39. C-H Lee, J D Korp, H Kohn J. Org. Chem. 54 3077 (1989) 40. P Goya, A Martinez,M L Jimeno J.Chem. Soc., Perkin Trans. 2 783 (1990) 41. T Breining, A R Cimpoia, T S Mansour, N Cammack, P Hopewell, C Ashman Heterocycles 41 87 (1995) 42. C-H Lee, H Kohn J. Org. Chem. 55 6098 (1990) 43. E N Prilezhaeva, in Poluchenie i Svoistva Organicheskikh Soedinenii Sery (Synthesis and Properties of Sulfur Organic Compounds) (Ed. L I Belen'kii) (Moscow: Khimiya, 1998) p. 115 44. A E Reed, P von Rague' Schleyer J. Am. Chem. Soc. 112 1434 (1990) 45. O Mo, J L G De Paz,M Yanez, I Alkorta, J Elguero, P Goya, I Rozas Can. J. Chem. 67 2227 (1989) 46. J G Angyan, C Bonnelle, R Daudel, A Kucsman, I G Csizmadia THEOCHEM 42 273 (1988) 47. J Elguero, P Goya, A Martinez, I Rozas, O Mo, J L G De Paz, M Yanez J. Phys. Org. Chem. 3 470 (1990) 48.I Alkorta Theor. Chim. Acta 89 1 (1994) 49. I Alkorta, C Garcia-Gomez, J A Paez, P Goya J. Phys. Org. Chem. 9 203 (1996) 50. I Rozas Int. J. Quantum Chem. 62 477 (1997) 51. F Belaj, C Kratky, E Nachbaur,A Popitsch Monatsh. Chem. 118 427 (1987) 52. R M Ibberson J. Mol. Struct. 377 171 (1996) 53. K S Sulaimankulov, in Soedineniya Karbamida s Neorganicheskimi Solyami (Carbamide Compounds with Inorganic Salts) (Frunze: Ilim, 1971) p. 7; 224 54. V de Zea Bermudez, G Lucazeau, L Abello, C Poinsignon J. Mol. Struct. 297 185 (1993) 55. V de Zea Bermudez, G Lucazeau, L Abello, C Poinsignon J. Mol. Struct. 301 7 (1993) 56. A Popitsch Monatsh. Chem. 113 529 (1982) 57. J Elguero, P Goya, I Rozas, J Catalan, J L G De Paz THEOCHEM 53 115 (1989) 58.C D Nenizescu Chimie Organica Vol. 1 (Bucuresti: Editura Tehnica, 1960) 59. E Nachbaur, A Popitsch Angew. Chem., Int. Ed. Engl. 12 339 (1973) 60. V A Pal'm Osnovy Kolichestvennoi Teorii Organicheskikh Reaktsii (Foundation of Quantitative Theory of Organic Reactions) (Leningrad: Khimiya, 1967) p. 164 61. P Goya, C Ochoa, I Rozas, A Alemany,M L Jimeno Magn. Reson. Chem. 24 444 (1986) 62. P S Belton, J D Woollins Magn. Reson. Chem. 24 1080 (1986) 63. T Esser, A E Karu, R F Toia, J E Casida Chem. Res. Toxicol. 4 162 (1991) 64. J Dusemund, T Schurreit Arch. Pharm. 319 826 (1986) 65. J Dusemund, T Schurreit Arch. Pharm. 320 534 (1987) 66. C H Lee, H Kohn Heterocycles 27 2581 (1988) 67. G A Orekhova, A N Kravchenko, O V Lebedev, in The IXth Conference of Organic Chemistry and Biochemistry for Young Scien- tists (Abstracts of Reports), Liblice, 1995 p.111 68. J Elguero, C Ochoa,MStud, C Esteban-Calderon,M Martinez- Ripoll, J-P Iayet,M C Vertut J. Org. Chem. 47 536 (1982) 69. J B Wright J. Org. Chem. 29 1905 (1964) 70. R N Goyal, S Bhargava Curr. Sci. 58 287 (1989) 71. A Castro, A Martinez J. Chem. Soc., Perkin Trans. 2 1561 (1994) 72. A Alberola, J M Andres, A Gonzalez, R Pedrosa,M Vicente Synthesis 355 (1991) 73. C-H Lee, H Kohn J. Heterocycl. Chem. 27 2107 (1990) 74. P Christjanson, K Siimer, A Laht, H Oja Tallinna Tehnikaulik. Toim 731 10 (1992); Chem. Abstr. 120 299 913 (1994) 75. K Siimer,H Oja, P Christjanson, A Laht VTT Symp. 133 332 (1992); Chem. Abstr.118 170 890 (1993) 76. G Schwenker, H Guo Arch. Pharm. 326 45 (1993) 77. AABakibaev,RRAkhmedzhanov,AYuYagovkin,TPNovozheeva, V D Filimonov, A S Saratykov Khim.- Farm. Zh. 27 (6) 29 (1993) a 78. R H Barker, S L Vail,G B Barcelo J. Heterocycl. Chem. 3 354 (1966) 79. F M Abd El Latif Asian J. Chem. 5 184 (1993); Chem. Abstr. 118 124 488 (1993) 80. R Glinka, K Walczynski,M I Grabowski, A Stepien Pol. J. Chem. 65 2053 (1991)G A Gazieva, A N Kravchenko, O V Lebedev 230 81. US P. 5 206 362; Chem. Abstr. 119 160 330 (1993) 82. Span. P. 545 696; Chem. Abstr. 107 39 870 (1987) 83. V J Aran, J R Ruiz, E Davila, I Alkorta, M Stud Liebigs Ann. Chem. 337 (1988) 84. Span. P. 547 443; Chem. Abstr. 108 6031 (1988) 85. I Alkorta, V J Aran, A G Bielsa, M Stud J.Chem. Soc., Perkin 86. I Alkorta, V J Aran, E Davila, J R Ruiz, M Stud Liebigs Ann. 87. G A Gazieva, A N Kravchenko, K Yu Chegaev, Yu A Strelenko, Trans. 1 1271 (1988) Chem. 1135 (1989) O V Lebedev Mendeleev Commun., 28 (2000) 88. V J Aran, A G Bielsa, J R Ruiz, M Stud J. Chem. Soc., Perkin Trans. 1 643 (1986) 89. Span. P. 553 863; Chem. Abstr. 109 129 066 (1988) 90. J Elguero, P Goya, R Nieves, C Ochoa, C Rodellas,M Martinez- Ripoll, S Garcia-Blanco J. Chem. Res. (S) 94 (1988) 91. Jpn. P. 6 242 978; Chem. Abstr. 107 39 852 (1987) 92. Jpn. P. 6 242 975; Chem. Abstr. 107 39 853 (1987) 93. M Haake, B SchuÈ mmelfelder Synthesis 753 (1991) 94. O V Anikin, L G Gareeva, I E Chlenov, V A Tartakovskii, Yu T Struchkov, V S Kuz'min, Yu N Burtsev Izv. Akad. Nauk SSSR, Ser. Khim. 1812 (1989) b 95. G A Hanewacker, Z Mester, B Unterhalt Arch. Pharm. 326 497 (1993) 96. M KnollmuÈ ller, R Faub Monatsh. Chem. 116 1027 (1985) 97. M KnollmuÈ ller, P Kosma Monatsh. Chem. 116 1141 (1985) 98. M KnollmuÈ ller, P Kosma Monatsh. Chem. 116 1321 (1985) 99. G W Muller, G E DuBois J. Org. Chem. 54 4471 (1989) 101. A Herrero, C Ochoa, M Stud, F Florencio, I Hernandes-Fuentes, 100. P Goya, A Martinez, C Ochoa,M Stud, M L Jimeno, C Foces-Foces, F H Cano, M Martinez-Ripoll Tetrahedron 41 3105 (1985) C Abdradelo, J L G De Paz J. Org. Chem. 54 5025 (1989) 102. P Goya, A Martinez,M L Jimeno,W Pfleiderer Liebigs Ann. Chem. 961 (1987) 105. C S Lee, S H Kim, C-H Lee J. Korean Chem. Soc. 41 677 (1997); 103. J Elguero, P Goya, A Martinez Heterocycles 29 245 (1989) 104. C H Lee, Y S Chung, B Y Chung Bull. Korean Chem. Soc. 14 592 (1993); Chem. Abstr. 120 163 925 (1994) Chem. Abstr. 128 88 869 (1998) 106. G Faleschini, E Nachbaur, F Belaj Phosphorus Sulfur Silicon Relat. Elem. 65 147 (1992) 109. M V Mirifico, J A Caram, E J Vasini Electrochim. Acta 36 167 107. I Alkorta, P Goya, A J Paez Heterocycles 32 279 (1991) 108. L R R Wang, T Benneche, K Undheim Acta Chem. Scand. 44 726 (1990) (1991) 110. L I Suvorova, V A Efes'ko, L V Epishina, O V Lebedev, L I Kh- mel'nitskii, S S Novikov, M V Povstyanoi, V D Krylov, G V Korotkova, L V Lapshina,A F Kulik Izv. Akad. Nauk SSSR, Ser. Khim. 1306 (1979) b 111. A S Ermakov, S A Serkov, V A Tartakovskii, T S Novikova, L I Khmel'nitskii Khim. Geterotsikl. Soedin. 1129 (1994) c 112. V A Timoshchuk Usp. Khim. 63 730 (1994) [Russ. Chem. Rev. 63 695 (1994)] a�Pharm. Chem. J. (Engl. Transl.) b�Russ. Chem. Bull. (Engl. Transl.) c�Chem. Heterocycl. Compd. (Engl. Tr

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (3) 231 ± 260 (2000) Heterophase mechanisms of thermal oxidation of polymers. New horizons Yu A Mikheev, G E Zaikov Contents VIII. The model of oxidation of thermofluctuationally active objects I. Introduction II. The influence of the thermomechanical state of a polymer on the oxidation kinetics III. The microreactor model and kinetic regimes of oxidation IV. Kinetic model of oxidation of polyolefins V. Oxidation of molten polystyrene VI. Oxidation of molten poly(ethylene oxide) VII. Characteristic features of initiated oxidation of polymers IX. Cross-linking of polyolefin macromolecules X. Oxidation of fibrillised polyolefins XI. Conclusion Abstract. numerous of analysis kinetic of results the on Based Based on the results of kinetic analysis of numerous experimental and heterogeneous of advantages data, experimental data, advantages of heterogeneous and heterophase heterophase mechanisms polymers non-crystalline of autooxidation of mechanisms of autooxidation of non-crystalline polymers over over homogeneous liquid-phase mechanisms are demonstrated. The homogeneous liquid-phase mechanisms are demonstrated.The supermolecular to according chains polymer of organisation supermolecular organisation of polymer chains according to the the pattern non-uniform incorporating micelles spongy of pattern of spongy micelles incorporating non-uniform porous porous areas is chosen as the model structure. It is established that, areas is chosen as the model structure. It is established that, depending the polymer, the of state physical the on depending on the physical state of the polymer, the oxidation oxidation kinetics correspond either to a mechanism in which reaction kinetics correspond either to a mechanism in which reaction chains to or nanophases non-uniform over distributed are chains are distributed over non-uniform nanophases or to a mechanism one in located mainly are they which in mechanism in which they are mainly located in one nanophase.nanophase. The stages macroscopic the for obtained equations kinetic The kinetic equations obtained for the macroscopic stages of of oxidation the of dependence the account into take oxidation take into account the dependence of the chemical chemical reactivity of mobility thermomechanical the on reactants of reactivity of reactants on the thermomechanical mobility of the the molecular sponge.It is demonstrated that the fine structure of molecular sponge. It is demonstrated that the fine structure of molecular the in role crucial a plays micelles spongy molecular spongy micelles plays a crucial role in the chemical chemical physics 94 includes bibliography The polymers. of physics of polymers. The bibliography includes 94 references. references. I. Introduction Oxidation of block polymers is often considered within the framework of homogeneous oxidation schemes designed for liquid hydrocarbons.1 ±13 This is formally based on the view, accepted long ago, that a non-crystalline polymer is an amorphous homogeneous material. The evident physical difference between polymeric and liquid phases is attributed in its simplified form to the functioning of unstable structures continuously formed and degraded and having rather long lifetimes and low rates of the relaxation response to an external action.Meanwhile, the kinetics of polymer oxidation usually do not fit in the framework of homogeneous reactions. In fact, when analysing the mechanisms of several macroscopic stages (for Yu A Mikheev, G E Zaikov N M Emanuel Institute of Biochemical Physics, Russian Academy of Sciences, ul. Kosygina 4, 117977 Moscow, Russian Federation. Fax (7-095) 137 41 01. Tel. (7-095) 939 73 62 (Yu A Mikheev). E-mail: chembio@glas.apc.org (G E Zaikov) Received 2 August 1999 Uspekhi Khimii 69 (3) 249 ± 282 (2000); translated by Z P Bobkova #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v069n03ABEH000549 231 232 235 236 243 245 248 253 257 258 259 example, induction period 5 and self-accelerated stage 4±6, 8± 15) in terms of homogeneous schemes, researchers were forced either to change the essence of these schemes or to use mathematical simplifications.1 ±17 The results were often at variance with experimental data.The contradictions caused by the use of the homogeneous reaction model could not fail to cast doubt 16 upon the validity of the mechanisms of inhibited oxidation and `critical' concentra- tions of antioxidants discussed in the literature.4, 6 ± 8 Nevertheless, the generally accepted view of the liquid-phase homogeneous structure of a non-crystalline polymeric matrix delayed for a long period the development of an adequate oxidation mecha- nism, even after the dependence of the reaction rate on the thermomechanical state of the polymeric matrix had been estab- lished.4±6, 18±26 Instead of diverging from the old-fashioned homogeneous model, the researchers spent effort on the improve- ment of this model.This was done by invoking ideas of poly- chromatism of kinetic parameters 5± 7 and by assuming the presence of so-called `disordered areas' 4, 5 and `rigid cages' 8 in the matrix being oxidised. The kinetic `polychromatism' assumes that reactions which are elementary in homogeneous media cannot be characterised by a single rate constant when they occur in polymers. It is postulated that the reactivity of particles depends on their closest environ- ment. However, the physical grounds for this dependence usually remain unknown.The ideas of the molecular-segment `disordered areas' and `rigid cages' make the views on a polymer process somewhat more detailed. It cannot be ruled out that the schemes based on each concept would describe different particular situations. They are unable to serve as the common ground for the simulation of reactions in a non-crystalline polymeric matrix because they assume a priori that the `areas' and `cages' arise in an essentially homogeneous medium. The same is true for the known theoretical approach resorting to the model of propagation of infectious diseases.14 This approach is based on the statements that oxidation of a partially crystalline polymer develops through multiplication of `pest- holes', which successively involves amorphous regions dispersed inside the crystalline phase, and that the `infection' is transmitted by monomeric oxidation products acting as initiators.232 Since the above approaches are modifications of the same homogeneous reaction model, none of them discloses the physical reasons for, for example, the fact that oxidation of polymer samples in different physical states often follows the same kinetics.Thus the temperature dependence of the rate constant of non- inhibited oxidation of low-density polyethylene (LDPE) in the induction period is described by an Arrhenius equation in the temperature range of 40 ± 200 8C; this implies the absence of changes in the oxidation mechanism on passing from the solid state to the molten state (Tm*115 8C).27 The inapplicability of homogeneous schemes (both the classi- cal and modified ones) to the description of the oxidation of polymers is also indicated by the fact that the same polymer often behaves (in the reaction) in different ways depending on the method used to prepare the sample.In addition, the macroscopic stages of polymer oxidation are described by kinetic laws different in kind. The reasons for these specific kinetic features are hidden in the supermolecular organisation of polymer chains in spongy micelles incorporating non-homogeneous porous areas, i.e. in the real heterogeneity of the `amorphous' polymer phase.15, 16, 28, 29 The molecular-chain sponge is actually a system of connected pendula (oscillators), the oscillations of which can become cooperated; this makes the units of macromolecules move apart to form nanopores with different widths.It was emphasised in a series of studies 15, 16, 18 ± 22 that the initiating activity of admixed initiators and polymeric hydro- peroxides and the routes of transformation of free radicals depend on which particular site in the nanoporous structural zone of the molecular spongy matrix contains the reactants. Due to the high degree of heterogeneity and the high elasticity with respect to mechanical treatment, typical of this matrix, stretching of polymer chains and the amplitudes of beating of nanopores of each size can vary over wide limits. Therefore, the kinetics of oxidation of samples existing in different thermomechanical states, depending on the preparation method, can be different in kind, although the chemical structure of the polymeric molecules remains the same.II. The influence of the thermomechanical state of a polymer on the oxidation kinetics 1. Characteristic features of the process under structural relaxation conditions The thermomechanical state of a polymer subjected to oxidation largely depends on the character of thermal and mechanical treatment of the polymer sample prior to the reaction. Oxidation of polymer samples can be described by qualitatively different kinetics depending on the previous history of the samples. This refers, for example, to samples of isotactic polypropylene (IPP), studied by Pudov et al.1, 2, 30 and Monakhova et al.24, 31 The data on accumulation of polymeric hydroperoxide (POOH) in IPP samples obtained by Pudov et al.1, 2, 30 (Fig. 1) differ from the results of Monakhova et al.24, 31 (Fig.2) by a longer period of accumulation and a lower rate of decomposition of POOH. The plots for the oxygen absorption rate vs. the concentration of accumulated POOH also follow different patterns in these two cases. Pudov et al.1, 2, 30 attribute this plot to one macroscopic stage and describe it as a sum of two terms close in magnitude u=a [POOH]0.5+b [POOH], where a and b are constants. On the contrary, the curves shown in Fig. 2 24, 31 actually represent three macroscopic stages of oxida- tion, characterised by different types of variation of u vs. [POOH]. In fact, successive transitions from a more or less extensive in- duction period to a short self-accelerated process and then to a stage with a virtually constant rate of oxygen absorption can be distinguished in curves 3 and 4.At the third stage of oxidation, the concentration of POOH first increases and then starts to decrease. Yu A Mikheev, G E Zaikov [POOH]/ mol kg71 21 42 40 80 0 t/ min 120 Figure 1. Variation in the concentration of hydroperoxide in films pressed from non-melted IPP powder during oxidation at 130 8C and oxygen pressures of 26.7 (1 ) and 53.3 kPa (2).1, 2, 30 [POOH]/ mol kg71 O N 2 / mol kg71 5 3 6 6 6 4 4 4 2 2 2 1 t/ min 200 0 100 Figure 2. Calculated curves for hydroperoxide accumulation (1, 2) and experimental (3, 4) and calculated (5, 6) curves for the oxygen absorption during oxidation at 130 8C and PO2 =40 kPa of IPP films prepared from the melt and containing small (1, 3, 5) and large (2, 4, 6) spherulites.24, 31 The dots correspond to experimental concentrations of POOH.These kinetic differences are apparently due to nonequivalent methods used to prepare the samples in question. In one study 30 films were prepared by short-term (2 ± 3 min) pressing of the IPP powder at 125 ± 130 8C without bringing the crystals to melting, while in the other studies 24, 31 the initial IPP was first melted at 220 8C and then the melt was cooled to give films with small spherulites. On additional heating, large spherulites were formed but the degree of crystallinity did not change. It is obvious that pressed films employed by Pudov et al.1, 2, 30 were characterised by enhanced internal mechanical stress, caused by deformation of the framework of non-melted crystallites and by biaxial orientational stretching of polymer chains which fill the space between the crystallites.Comparison of curves 1, 2 in Fig. 1 with the corresponding curves in Fig. 2 leads to the conclusion that in the concentration range of 0 ± 0.2 mol kg71, accumulation of POOH in the films prepared by pressing occurs at almost the same rate as in the films with small spherulites. However, the yield of POOH in the subsequent oxidation stages is much higher in the former type of films. This indicates that the reaction rate increases in samples with internal stresses. Apparently, at a certain stage of the process, efficient stress relaxation occurs in `stressed' films, which is accompanied by the appearance of local centres of thixotropic excitation of the matrix.32 In these centres, the frequency and amplitude of beats of macromolecular units increase; this changes the dimensions of nanopores and, conse- quently, increases the rate of generation of free radicals.15 ± 22, 33, 34 The IPP films prepared by cooling a melt 24, 31 differed from those made of a pressed conglomerate of non-melted particles by the presence of a much more equilibrated (as regards the thermo- mechanical state) framework of crystallites.The residual internalHeterophase mechanisms of thermal oxidation of polymers. New horizons stress in the crystallites is much lower. They are distributed more uniformly in the material and do not bring about any thermome- chanical excitation of the matrix or substantial shear deforma- tions during oxidation.The increase in the induction period associated with enlarge- ment of spherulites due to recrystallisation (see Fig. 2, curve 2 ) can be explained by assuming that some of the units of macro- molecules are incorporated into crystallites and thus induce orientational drawing of the polymer chains in the intercrystallite regions. The stretched polymer chains are fixed rather firmly on the crystallite framework; this accounts for the extension of the induction period. A similar effect is involved in the oxidation of highly crystal- line films of IPP, high-density polyethylene (HDPE) and poly- methylpentene subjected to uniaxial orientational drawing. The greater the degree of drawing, the greater the increase in the induction period of oxidation.5, 35 The rate of oxidation of LDPE is affected by orientational drawing in a different way.The low degree of crystallinity of this polymer does not ensure sufficient resistance to the internal stress and the fast relaxation increases the oxidation rate from the very beginning of the process.35 Thus, despite the identical treatment of the polyolefin films (orientational drawing), the rates of the initial stage of their oxidation differ depending on the strength of the supermolecular framework. Yet another specific feature of oxidation of uniaxially stretched IPP films is that the molecular mass of the polymer increases substantially at the initial oxidation stages, whereas in the case of isotropic films, fast destruction of macromolecules is observed.5 Calculations have shown 35 that the influence of drawing is due to the change in the physical structure of the IPP matrix subjected to oxidation rather than to the change in the solubility and the oxygen diffusion coefficient. It should be noted that the draw ratio also influences the parameters of the self-accelerated stage of oxidation.This stage becomes longer for high draw ratios;5, 26, 35 in some cases, the rate of oxygen absorption increases according to the law u * exp(Ft) rather than in proportion with time.26 The state of the spongy molecular matrix can change due to different types of treatment and thus influence the oxidation kinetics. Thus, it has been reported 20 that previous heat treatment of a non-stabilised commercial sample of IPP in vacuo decreases the rate of its high-temperature oxidation by increasing the induction period; however, the rate of low-temperature (20 8C) oxidation substantially increases (possibly, due to the structural relaxation under the influence of absorbed moisture 29).Simultaneously, a new specific feature appears which cannot be interpreted within the framework of the homogeneous reaction model, namely, different induction periods are observed for identical contents of POOH in the polymer and, conversely, identical induction periods correspond to hydroperoxide concen- trations that differ by an order of magnitude.20 Note that these regularities (observed in the kinetic regime of oxygen absorption) reflect a general feature of heterogeneous micro- and nanoporous materials associated with the lability of their physical structure.In many cases, this lability is undesirable because it accounts for the poor reproducibility of the useful properties of materials.36 The molecular sponge, which is an important structural material of a non-crystalline polymer matrix, exhibits clear-cut susceptibility to thermomechanical treatment and has a fairly labile structure. Owing to these properties of the molecular sponge, non-crystalline polymers differ fundamentally from homogeneous fluids, in which reaction rates do not change upon shear mechanical treatment (stirring). Moreover, parame- ters of elementary chemical steps proceeding in homogeneous media are usually unambiguous. 233 2. Specific features of oxidation in the steady state of the polymer matrix The notion of structural steady state as applied to a polymer being oxidised is somewhat arbitrary.Oxidation is inevitably associated with the formation of oxygen-containing groups in macromole- cules and with rupture or cross-linking of polymer chains. These processes change the initial balance of elastic forces in the spongy molecular matrix and cause more or less fast structural evolution. However, if the chemico-mechanical perturbation is relatively slight and the matrix quickly returns to the thermomechanical equilibrium, this process can be regarded as being structurally stationary. The structurally steady-state oxidation includes several mac- roscopic stages which obey different kinetic laws.Thus the induction period is characterised by a low rate of oxygen con- sumption (u1). This period ends after a small amount of oxygen, usually less than 0.1 mol kg71, has been absorbed. The u1 value does not depend on the concentration of hydroperoxide accumu- lated during the induction period but increases with an increase in the pressure of oxygen without, however, reaching a limiting value 20 1PO2 2 Ü1=2 , a u1 à Öb á PO where a1 and b are constants. The induction period is followed by a short stage of self- accelerated oxygen absorption. The reaction rate at this stage (u2) is most often proportional to the oxidation time. In this case, reaction chains are initiated upon dissociation of POOH and are terminated upon recombination of macroperoxyls. As oxidation develops, the increase in the rate of oxygen absorption becomes less pronounced. After some period, the rate reaches a maximum and no longer depends on the POOH concentration.Whereas in the first stage (u1), the rate of oxygen absorption does not reach any limiting value upon an increase in the pressure of oxygen, in the third stage (u3) this rate does have a limit. It is significant that the third stage of oxidation is established earlier than the maximum hydroperoxide concentration has been attained and continues even after the POOH concentration has considerably diminished. This pattern (which can be seen, for example, in Fig. 2) is typical of reactions having the zero order with respect to hydroperoxide.20 ± 22 This distinguishes high- molecular-mass block polymers from their solutions.The difference between the kinetics of oxidation of block and dissolved LDPE is demonstrated in Fig. 3 (data of Ref. 37). Although curves 1 0 ± 3 0 are qualitatively similar, the kinetics of accumulation of POOH in the two systems are substantially different. It can be seen that the variation of the hydroperoxide concentration in the melt is described by a curve with a sharp maximum (curve 1 0), while that in solution is described by monotonic curves with relatively low limiting values (curves 2 0, 3 0). According to a study by Iring et al.,37 oxidation of LDPE in solutions can be described by a kinetic equation for liquid-phase oxidation (1) dâPOOHä=a u7k[POOH], dt =0, a ulim=k[POOH]lim?const. dâPOOHä dt lim where a is the yield of POOH, u is the rate of oxygen absorption and k is the rate constant for the decomposition of POOH.The reaction reaches a steady-state regime at those degrees of con- version at which the consumption of the CH groups is still insignificant and the following conditions hold: 234 a 103 Z 300 1 200 100 160 0 b 103 Y 105 1 0 200 0 Figure 3. Amount of absorbed oxygen (Z, moles of O2 per mole of molecular units) vs. oxidation time (a) and amount of accumulated hydroperoxide (Y, moles of POOH per mole of molecular units) vs. amount of absorbed oxygen (b) during the oxidation of a melt (1, 1 0) and a solution of LDPE (the LDPE concentration is 0.56 mol litre71) in trichlorobenzene (2, 2 0, 3, 3 0).37 Temperature/ 8C: 1, 1 0, 2, 2 0 �160, 3, 3 0 �170. Pressure of O2/ kPa: 1, 1 0 �100; 2, 2 0, 3, 3 0 �75.Oxidation in the melt also reaches a steady-state regime at a relatively low consumption of CH groups (Fig. 3, curve 1); however, the pattern of variation of the POOH concentration (Fig. 3, curve 1 0) differs from that for the liquid-phase reaction. Oxidation of high-molecular-mass block polyolefins in the steady-state regime is characterised by the following fea- tures.15, 20 ± 22 (1) hydroperoxide is not the only molecular product of the reaction (other oxygen-containing compounds are also formed); (2) decomposition of POOH is due to its reaction with an oxidation product and is not accompanied by the formation of free radicals; (3) oxidation has the zero order with respect to hydroper- oxide. It should be emphasised that the zero order with respect to hydroperoxide in the steady state has been repeatedly found by numerous researchers; however, this fact has not received adequate attention (apparently, in the heyday of the theory of liquid-phase oxidation of hydrocarbons, researchers have not yet faced the problem of structure-kinetic modelling).Since this finding was undoubtedly significant, it was specially verified in relation to the oxidation of molten polyethylene oxide (PEO),15, 18, 19 a high-elasticity copolymer of ethylene with propy- lene and solid-like IPP.21 These experiments confirmed simulta- neously yet another fact which had been repeatedly described previously � admixed initiators present in polymers accelerate the attainment of the highest reaction rate but do not influence the rate itself over a broad range of concentrations.Figure 4 presents experimental results on the oxidation of a PEO melt in air.15 It can be seen that the constant rate u3 is established earlier (curve 1) than the maximum POOH concen- tration is attained (curve 2). The curves for the consumption of the polymer CH groups (curve 1 0) and for the accumulation of aldehyde reaction products (curve 2 0) also indicate that constant 103 Z 120 3 80 40 2 t/ min 400 0103 Y 2 3 0 1 2 0 50 103 Z 0 Yu A Mikheev, G E Zaikov a O N 2 , [POOH]/ mol kg71 b 1 4.0 DCO, DCH 3 0 0.8 3.0 2 0 123456 0.6 2.0 1 0 4 ±6 0.4 3 1.0 2 0.2 0 10 5 0 1 5 0t/ h Figure 4. Curves for oxidation of molten poly(ethylene oxide) at 90 (1, 2, 1 0 ± 3 0) and 80 8C (3 ± 6); (a) oxygen absorption (1) and hydroperoxide accumulation (2 ± 6); (b) variation of the optical density of the CH (1 0) and CO (2 0, 3 0) absorption bands in the IR spectra.Concentration of dibenzoyl peroxide /mol kg71: 0 (1 ± 3, 1 0, 2 0), 0.06 (3 0), 0.0165 (4), 0.039 (5) and 0.12 (6). The IR spectra were recorded using chloroform solutions of the polymer (10 mg ml71) (data from Refs 15, 18). oxidation rates are rapidly attained and that they do not depend on the content of POOH in the melt. The introduction of a foreign initiator, dibenzoyl peroxide (BP) in concentrations of up to 0.12 mol kg71, into the PEO melt eliminates the self-accelerated stage but does not influence the steady-state rates of accumulation of aldehyde (curve 3 0 in Fig. 4 is parallel to the straight section in curve 2 0) and hydroperoxide (curves 4 ± 6 are parallel to the straight section in curve 3).The fact that the steady-state rate of oxidation does not depend on the content of a foreign initiator indicates that the polymeric matrix has some structural filter (`bottle-neck'), which restricts the transfer of free radicals from area the initiator dissociates to other areas in which chain oxidation occurs. It is noteworthy that dibenzoyl peroxide present in a relatively high amount in a PEO melt depresses the reaction; as its concen- tration increases from 0.12 to 0.38 mol kg71, the rate u3 dimin- ishes 1.4 fold at 90 8C,19 which is inconsistent with the homogeneous model. The `bottle-neck' effect is clearly manifested in the experi- ments on oxidation of solid-like IPP in the presence of dibenzoyl peroxide.21 Figure 5 shows how the total concentrations of BP being consumed and POOH being accumulated vary during the [BP]/ mol kg71 [POOH]/ mol kg71 2 0.15 0.15 1 0.10 0.10 0.05 0.05 3 4 0 10 8 6 2 t/ h Figure 5.Variation of the total concentration of hydroperoxide and dibenzoyl peroxide (1, 2) and the concentration of only dibenzoyl peroxide (3) during oxidation of IPP films in air at 100 8C.21 [BP]0/ mol litre71: (1) 0.025, (2, 3)*0.8.Heterophase mechanisms of thermal oxidation of polymers. New horizons [POOH]/ mol kg71 NO2 / mol kg71 1 3 5.0 4.0 2.0 3.0 2 2.0 1.0 4 1.0 200 0 t/ h 400 Figure 6. Amount of the absorbed oxygen (1, 3) and accumulated hydroperoxide (2, 4) vs. time of oxidation of low-molecular-mass APP withMn=4500 (1, 2) and 5-fold orientationally drawn IPP films (3, 4) at T=120 8C and PO2 =103 kPa.26 oxidation of IPP films in air at 100 8C.In this particular case, the presence of a structural filter for the radicals arising is quite obvious: the higher the content of BP in the polymer, the greater fraction of BP decomposes unproductively with respect to the accumulation of POOH. In addition, the steady-state rate of the process does not depend on the initial concentration of BP or on the current concentration of POOH. Yet another result obtained in a study of oxidation of a low- viscosity melt of atactic polypropylene (APP) with the molecular mass Mn= 4500 and orientationally drawn IPP films (Fig. 6) is worth noting.26 The rate of oxidation at 120 8C increases for a long period in direct proportion to the concentration of the accumulating POOH; simultaneously, the amount of absorbed oxygen increases with time according to an exponential law and u*exp(Ft).The same pattern is followed at 110 and 130 8C. Meanwhile, oxidation of solutions of these polymers in inert trichlorobenzene (concentration 0.56 mol kg71 per monomer unit) obeys kinetic laws of homogeneous reactions, i.e. the dependences u*t and u * [POOH]0.5 hold. Thus, the kinetics of oxidation of polymer samples in which the molecular sponge exists in the state of enhanced thermal fluctuation mobility differ from the kinetics of oxidation of liquid hydrocarbons or high- molecular-mass polyolefins in the structural steady state. To summarise, it can be noted that oxidation of the same (from the chemical viewpoint) polymer is described, depending on its thermomechanical state, by a whole range of kinetic laws different in kind, which is inconsistent with the model of homoge- neous reactions. Thus a new model taking into account the nanosize heterogeneity of the non-crystalline polymeric phase needs to be developed.III. The microreactor model and kinetic regimes of oxidation In several studies,15, 16, 18 ± 22, 28, 29, 33, 34, 38, 39 non-crystalline poly- mers are considered as systems of molecular-chain spongy micelles which fill the cells of an infinite three-dimensional framework (Fig. 7 a). This framework is an open-work lattice of paracrystal- line domains, each of them containing up to a dozen parallel segments. The size of spongy micelles is of the order of tens of nanometers.The three-dimensional framework in the matrix of partially crystalline polymers incorporates a more or less shaped carcass of lamellar crystallites (Fig. 7 b). The molecular spongy micelle, which acts as a chemical microreactor, is a heterophase aggregate, the chains of which enter the paracrystalline domains at one end and enter the spongy grains at the other end. The chains within the grain are tightened by cohesion forces but are not completely collapsed, being 235 a b s-zones v-zones Figure 7. Scheme of the molecular spongy-micellar nanosized reactor for non-crystalline (a) and partially crystalline (b) polymers. stabilised as a sponge filled with fairly narrow micropores (more precisely, nanopores).The sizes of the micropores range from a chemical bond length to a polymer chain thickness (0.1 ± 0.5 nm). Under conditions of balanced elastic forces between the spongy grain and the domains of the continuous framework, a layer is formed consisting of radially arranged chain segments, which constitute the walls of relatively wide nanopores (super- nanopores). As a consequence, two types of nanosized zones arise in the micelle, namely, s-zones consisting of radial chain segments (the size of a thermodynamic segment is 2 ± 3 nm), which form supernanopores, and v-zones consisting of grains with narrow nanopores. The chain segments of these two zones are characterised by different motion dynamics. The segments in the s-zones play a role of mechanical strings fixing the micellar grain; the s-zones themselves can be compared to a porous adsorbent with a rigid framework.Av-grain is close in thermomechanical properties to a mobile drop of a liquid.Such a drop hanging from a domain of the continuous three-dimensional framework can execute, to a certain extent, its own vibrational or rotational motions and volume beats, thus transferring the kinetic momentum to more or less flexible polymer chains composing the drop. Due to their spongy structure, the micelles can rapidly rearrange and capture even those admixtures where the molecules are larger than the diameter of empty nanopores. In addition, since the molecular sponge is, in essence, a set of coupled pendula, it is subject to interference, i.e., damping or synchronisation of oscillations; this creates conditions for a fluctuation increase in the size of v-nanopores or, correspondingly, for more or less fast migration of admixtures over the sponge bulk.The nanosized heterogeneity of the molecular spongy micelles influences the chemistry of chain reactions. Firstly, foreign com- pounds (oxygen, initiators, inhibitors) are distributed non-uni- formly between the zones, being mainly accumulated in the v-grains. Secondly, due to different sizes of the s- and v-nano- pores, the reactivities of macromolecules and admixtures absorbed in them are appreciably dissimilar. Indeed, a reacting species in a narrow v-micropore resides most of the time under constrained conditions �either on a nanopore wall (if it belongs to the macromolecule) or in the pore bulk (if it is a foreign species).A foreign molecule is expelled to the pore bulk under the influence of dispersion electromagnetic forces.28, 29 The situation in an s-nanopore is different: most of the time, a foreign molecule is adsorbed on the walls of a pore. The difference between the chemical reactivities of a species caused by the steric and energetic nonequivalence of s- and v-nanopores affects all the intermediates of a chain reaction. The quantitative changes in the reactivity of the same reactant can be very pronounced; the pathways of transformations in the two zones can differ even qualitatively. Chain reactions in spongy micelles can proceed in different modes; kinetic modelling can be performed most easily for the steady-state heterophase and homophase modes, which are real- ised in the pure form when the rate of structural evolution is negligibly low, i.e.when the volumes and the properties of the zones are virtually invariant.236 By definition, the scheme of a heterophase process should include separate charts for the reactions in each zone and the chart for the transfer of the reacting species from one zone to another. The primary reactions occur usually in s-zones because these zones provide more favourable conditions for the dissociative generation of free radicals, which subsequently migrate to v-zones. If a chemical reaction is localised in one zone, it can formally be referred to as a homophase reaction. This situation is observed, for example, when free radicals are generated in v-zones when nanopores undergo high-frequency fluctuations and thus reach the size of s-nanopores.In this case, the initiating activity of the POOH groups captured by v-nanopores increases and the fluctu- ation initiation of a chain reaction starts to prevail. The heterophase mode, involving coupling of non-uniform reaction chains, is more frequently encountered in the absence of clear-cut structural evolution of high-molecular-mass polyolefins. Oxidation of a polymer without an admixed initiator starts upon direct reaction of oxygen adsorbed on s-nanopore walls with the CH groups of the polymer s-segments. The greater part of the generated free radicals enter into subsequent reactions and decay without leaving the s-nanopore. When a small portion of radicals has passed from s-zones to the bulk of v-grains and has induced there reaction v-chains, the process acquires a heterophase char- acter. The sequence of s,v-transitions (s,v-translations) of free rad- icals described by the corresponding chart means physically the passage of free valences through the structural `bottle-neck'.The `bottle-neck' capacity depends on the thermomechanical state of the polymer sample and on the presence of impurities able to promote or, conversely, suppress the s,v-translation of free valences. The secondary reaction chains induced by this transfer (in v-zones) differ fundamentally from the primary chains (in s-zones) in the character of transformations. In particular, oxygen is mainly accumulated in v-zones. In addition, the rate of rupture of polymer chains with participation of alkyl macroradicals and the rate of termination of oxidation chains in v-zones are lower than those in s-zones.Consequently, the reaction v-chains are longer. Thus, the amount and composition of the oxidation products are mainly determined by the transformations of poly- mer units of v-zones, which constitute the main bulk of the molecular spongy micelles. The steady-state oxidation mode includes the steps of transfer of hydroperoxide groups from v-zones to s-zones. The hydro- peroxide formed in narrow v-nanopores dissociates into radicals very slowly due to unfavourable steric and energetic conditions. However, as hydroperoxide is accumulated, it starts to occupy the stationary s-zones, which have supernanopores favourable for dissociation.Thus the process passes from the induction period to the self-accelerated stage. IV. Kinetic model of oxidation of polyolefins 1. Induction period In accordance with the foregoing, the overall scheme of an initial period of autooxidation of polyolefins includes two charts for reactions in s- and v-zones and a chart for the inter-zone chemical induction, responsible for coupling of inhomogeneous reaction chains. In the absence of a foreign initiator, free radicals are generated in s-nanopores upon the reaction of oxygen with theCHgroups of the polymer. The scheme of the primary reaction chain (in the s-zone) can be written in the following general form: k00s PHs+O2,s d1Ps (generation of radicals), kd Rs +R= (rupture of polymer chain), Ps Yu A Mikheev, G E Zaikov k1s R1s (isomerisation), Rs Rs +R1s recombination and disproportionation products, s and R1s for PP have been considered previously 21).where PHs is a polymer unit in zone s, which includes reactive hydrogen atoms, Ps is the macroradical with a free valence on an internal unit of the macromolecule [for polyethylene (PE), this is *CH2CH* and for polypropylene (PP), these are *CH2C(CH3)* and *CHCH(CH3)*], d1 is a factor character- ising the yield of free radicals, Rs and R= are a terminal radical and a terminal unsaturated group, respectively, and R1s is the product of isomerisation of the terminal radical Rs (the structures of RThe pattern of the inter-zone transfer of free valence includes migration of the radical Rs from the supermolecular framework structure to the s-nanopore bulk (with transition to the Rm state) and subsequent oxidation of Rm: km Rm , Rs k¡m ktr R1mO+HO RmOO.m. R1mOOH Rm +O2,s The subscript `m' implies belonging to the nanopore bulk and `tr' stands for belonging to the inter-zone chain transfer process. As noted above, the rate of the s,v-transfer is negligibly low compared to the rate of radical transformations in the s-zone. It is determined by the thermomechanical ability of the radicals Rs to oscillate with some frequency between the rigid framework of s-segments and the s-nanopore bulk and to pass to the state Rm, more convenient for the reaction with oxygen. The terminal radical RmOO. thus formed undergoes a-isomerisation 40 to a hydroperoxyalkyl radical, which decomposes with elimination of free hydroxyl HOm.The free hydroxyl migrates to the v-zone and induces there a chain reaction HOm HO.v, k01 H2O+P.v , HO.v+PHv k1 POO.v P.v+O2OO. POO.= *C CH2*. X The POO. radical thus formed can be converted according to three pathways to give hydroperoxides, epoxides, or chain termination products. OOH k02, PH . *C CH2* + *C CH2* X X OO. OOH O *C CH2* *C CH.* CH*+HO. *C X X X k k3 b-isomer- isation t , POO. E chain termination products Epoxides E, which are highly reactive towards free radicals, undergo subsequent transformations giving rise to carbonyl compounds (the corresponding reactions are considered below). When calculating the s-scheme, we employed the condition of steady-state concentration of free radicals and did not take into account the loss of s-radicals by the inter-zone transfer; however, this transfer needs to be taken into account (as an initiation step) in the calculation of the v-scheme. The concentration of macro- radicals responsible for the inter-zone chemical induction can be described by the expression237 Heterophase mechanisms of thermal oxidation of polymers.New horizons . CO2 + *CH O sa1 a k00 0saO2;sa 2k1s aR *CHXC X O O. *CHXCH2* (k00 0s = d1 k00s [PHs], the subscript 1 at the values of radical *CHXC CH2* . + *C OH X concentrations and reaction rates implies the first stage of oxidation), the rate of the s,v-transfer is given by the expression trkmk00 0saO2;sa2 . or *CH CH* X k utr;1 a 2k1sOk¢§m a ktraO2;saU , O *C. OOH and the concentration of the peroxyl radicals in the v-zones at the first stage of oxidation is determined by the equation *C +HO .OO.O2 X X H *CH *C. *CHXCH2* (2) trkmk00 0sU0:5aO2;sa X X CH2* . + *C *CH OOH aPOO a1 a a4ktk1sOk¢§m a ktraO2;saUa0:5 . Ok X X . *CH CH* or Now we use Eqn (2) to express the rates of the absorption of X O OO. (3) *C *CH2CH daO2a dt a k1aP aaO2a a Ok02aPHa a k3UaPOO a1= 7HO. 1 u1 a ¢§ OO. *CXCH2* 7*CXCH2* OOH . *CH CH OOH X oxygen (u1), accumulation of hydroperoxide (u[POOH],1), and evolution of water (uaq,1) CH X H X O O O O O O O =Ok2 a k3UaPO2 Ob a P 2 U0:5 , O2 C O. *C +XC C. *C . *CHC *C u daPOOHa dt aPOOHa;1 a 1 O H X O. 2 U0:5 , a k0 k2aPO2 2aPHaaPOO a1 a Ob a P H X O O O2 H X O *C CHOO. CH *C O *C.2 U0:5 , uaq;1 a k01aHO aaPHa a k3aPOO a1 a Ob a P k3aPO2 . CH X X X O O O O 2 a k02[PH], a=Okmk00 0sgs=4ktk1sU0:5, b a k¢§mOktrgsU¢§1 s a [O2,s] P¢§O 12 ).The u values found from these expressions are where k (g in good agreement with experimental data.15, 20 ¡¾ 22 XC +*C According to the heterophase oxidation model (unlike the H O. *C CH O. X homogeneous model), water is mostly formed via a chain reaction involving free hydroxyls rather than upon decomposition of POOH. Isomerisation of the terminal *C.HX radical ensures migra- tion of the free valence along the macromolecule and the for- mation of ketone groups at some distance from the chain end. O2 *CH.CHCH2CH2 X X *CH2CHCH2CH. XOO. X OOH O *CHCHCH2CH2 *CCHCH2CH2+HO. Oxidation of the intermediate epoxides is accompanied by additional absorption of oxygen and the formation of POOH and water but these transformations do not influence the form of Eqn (2).The products of epoxide oxidation appear somewhat later than POOH;15, 21 however, they are formed at a noticeable rate even in the induction period; after that, the rate rapidly increases and becomes stabilised for a long period.41 ¡¾ 43 . *CCHCH2CH2 X X X X X X Further oxidation of these ketones results in the formation of larger fragments with aldehyde, ketone and acid groups The steady-state concentration of the epoxides is relatively low because they efficiently react with free radicals (Scheme 1). The carbonyl compounds formed are fairly diverse (they include polymeric and volatile aldehydes, ketones, acids); their formation can be described by the following reactions: O O O O2 O2 *CCHCH2CH2 *CC.CH2CH2 *C O O *C.C* X X X X OO. *CXCH2*(k5) 7*CXCH2* OOH OC* O. O *C. C* X O X X A. O O O. X O O *C + H2CCH2C *C CCH2CH2 *C +*C X O. O O X X X O. O Scheme 1 k04 *C C.CHX* O *C. CCHX* X X O O O. 7*CX7CH2* OOH A. CH2* *C *C CHCHX*+ O O. O X X k00 4 E CH *C *C C* CH C*+X. *C * . CH CX X X X X 7*CX7CH2* OOH O. CH C C* *CHCH2 O *CHCH2.+ C CH C* X X X X X X B. Here and later X=H, Me.238 O H O. *C C7CHCH2CH2 *CH.CCHCH2CH2 O X O O X X X 7*C H O *CHXCH2* H2CCH2CHC O H2CCH2CHC X X OH O. X X *C.XCH2* ( or*CHXCH.*) The presence of some compounds such as methyl ethyl ketone, methyl propyl ketone, butyraldehyde, pentanone, hexanone, etc. among the products of oxidation of polyolefins proves the occurrence of the above reactions.41 ¡¾ 43 The transformations of the terminal radical B are shown in Scheme 2. This set of reactions, which is far from being complete, is still sufficient to demonstrate the pathways to the final products of polyolefin oxidation and to state the fact that transformations of radicals A and B and carbonyl compounds have no influence on the steady-state concentration of peroxyl radicals [Eqn (2)].The concentration of carbonyl compounds, which are oxidised and formed again in the chain process, continuously increases. (4) In view of the foregoing, one can write the following condition for the steady-state concentration of epoxide daEa dt a k3[POO ]17k4[E][POO ]1=0, where k4=(k04 +k00 4). It follows from Eqn (4) that [E]=k3 / k4= uCO,1= = (a1+a2)k4[E][POO ]1=ak3[POO ]1 , daCO dt a 1 const. The total rate of accumulation of carbonyl compounds (CO) in the epoxide branch of the reaction is equal to where a1 and a2 are the yields of polymeric and volatile com- pounds, respectively, and a is their sum.Thus, after condition (4) has been reached, the rate uCO,1 depends on the pressure of oxygen in the same way as the rates u1 and u[POOH],1 and the content of CO groups increases with time in accordance with the equation (5) [CO]1=uCO,1 t. The same is true for the growth of the concentration of broken polymer chains (n) in the v-zones of spongy micelles detected by viscometry. The total rate of chain rupture is composed of two fractions proportional to the constants k04 and k00 4 : un,1= =(n1k04+n2k00 4)[E][POO ]1=const[POO ]1, (6) dt 1 dn where n1 and n2 are coefficients.The direct proportionality between the amount of oxidation products and the amount of oxygen absorbed in the induction O O2 *CHC *CHCH2OO. *CH 7HO. H X X X OO. *CXCH2* 7*CXCH2* OOH . CH2 B. OO. . O2 CCH3 *CH2 *CH2 *CH2 . CCH3 X XCHCH2 B. X O X O. X O O2 *C +CH3C , *C *C CCH3 CCH3 . O. O O O O O2 CHC *CH2 CHC *CH. OOH X X OO. . *CH2CHC X O C. Yu A Mikheev, G E Zaikov period, which follows from the scheme in question, is also retained for some time after the induction period has been completed, this time being much longer for the processes of accumulation of carbonyl compounds and destruction of macromolecules than for accumulation of POOH.41 ¡¾ 43 It should be emphasised that Eqn (5) refers to the overall process of formation of carbonyl compounds accompanied by abstraction of monomeric fragments.These destruction steps change the viscosity of the polymer solution only slightly com- pared to the primary scission steps in the epoxide branches, the rates of which are described by Eqn (6). 2. Hydroperoxide-initiated process The induction period in the autooxidation of polyolefins usually ends when 1072 mol kg71 of oxygen has been absorbed due to the occupation of the s-zones of the spongy matrix by the hydro- peroxide from v-zones. This occupation is provided by the transfer of free valence and migration of OOH groups along the polymer chain and by the fluctuation steps of rearrangement of the molecular sponge units.15, 20, 21, 33 By expressing the total frequency of the v,s-transfer of the OOH groups in terms of the effective constant kvs, taking into account the fact that activation of these processes requires a fluctuation of the sponge with opening of narrow v-micropores, we can express the scheme of the hydroperoxide v,s-transfer as kf kvs POOHf POOHv POOHs , k¡¾ f where the subscript `f ' refers to a fluctuationally excited v-nano- pore with a trapped OOH group.Having got into the s-nanopores, the OOH groups become able to dissociate to give free radicals k0s PHs (HO.+PO. )s POOHs dPs and, as their concentration increases, to enter into the bimolecular reaction kf 2 POH+O2 POOHs+POOHf giving no free radicals. This scheme of the chemical inter-zone feedback permits us to express the steady-state concentrations of the OOH groups in the fluctuationally active and s-states by the equation daPOOHaf dt a kfaPOOHav ¢§ k¢§faPOOHaf ¢§ ¢§ kvsaPOOHaf ¢§ kfaPOOHasaPOOHaf a 0 .Scheme 2 . *CH+CO , *CHC. X X O C. O O CCH3 CCH3 *C *C.7HO. X X H OO. *CXCH2* 7*CXCH2* OOH OX O+HO.. C O *CHC H XHeterophase mechanisms of thermal oxidation of polymers. New horizons In view of the high rate of deactivation of fluctuations, the last two terms can be neglected and the resulting expression [POOH]f=kfaPOOHav k¢§f can be used in the equation for the rate of the reaction involving s-hydroperoxide daPOOHas =kvs[POOH]f7kf[POOH]s[POOH]f7 dt 7k0s[POOH]s=0. The concentration [POOH]s found in this way, (7) k [POOH]s=k¢§fk0s a kfkfaPOOHav , vskfaPOOHav can be used to express the overall rate of radical generation in s-zones by the equation uis=k00 0 [O2,s] + dk0s[POOH]s&dk0s[POOH]s .The contribution of the direct reaction betweenCHbonds and oxygen soon becomes negligibly small and the expression for utr,1 derived previously is replaced by the equation utr;2 a ktrkmdk0saPOOHasaO2;sa 2k1sOk¢§m a ktraO2;saU (the subscript 2 denotes the second stage of the oxidation process). The equation describing the concentration of peroxyl radicals also changes (8) tk1sOk¢§m a ktraO2;saUa0:5 . [POO.]2=Oktrkmdk0saPOOHasaO2;saU0:5 a4kk , trkmaO2;sa During an initial period of the accelerated stage of autoox- idation, the final reaction products are accumulated synchro- nously with the hydroperoxide. Being unable to dissociate in the v-zones of the spongy matrix, hydroperoxide is accumulated for some period almost without reacting with other oxidation prod- ucts because their concentrations are too low.The rate of hydro- peroxide accumulation during this period is described by the equation u[POOH],2=k2[POO ]=A(dk0s[POOH]s)0.5 , where A a k2 4k

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC

摘要:

Russian Chemical Reviews 69 (3) 261 ± 278 (2000) Relaxation properties of polymers and the physical network model V I Irzhak Contents VIII. The network of physical bonds as a dynamic factor I. Introduction II. The role of topological factors. The reptation model III. Correlation between viscosity and viscoelasticity IV. Regularities of viscous flow V. Diffusion of macromolecules VI. Stress relaxation VII. Relaxation of chain orientation IX. On the nature of a physical network nodes X. Theoretical approaches other than the reptation model XI. Conclusion Abstract. of dynamics the in network physical the of role The The role of the physical network in the dynamics of macromolecules is considered. It is shown that experimental data macromolecules is considered.It is shown that experimental data are often inconsistent not only quantitatively but also qualita- are often inconsistent not only quantitatively but also qualita- tively with the results obtained in terms of the widespread tively with the results obtained in terms of the widespread reptation model based on the concept of network entanglement. reptation model based on the concept of network entanglement. Evidence is given indicating that the network of physical bonds is Evidence is given indicating that the network of physical bonds is the governing factor in the dynamics of macromolecules. The the governing factor in the dynamics of macromolecules. The bibliography includes 218 references. bibliography includes 218 references. I.Introduction The condensed state of substances is determined by the nature of intermolecular interactions. This statement appears trivial; how- ever, in the case of polymers the situation is not that simple. Indeed, below the glass transition temperature, the properties of a high-molecular body virtually do not differ from those of a low- molecular body (the latter may be brittle), whereas in the high- temperature range they differ rather substantially. This is primar- ily due to the chain nature of the molecules owing to which polymers display properties which allow one to speak about their specific (highly elastic) state; network structures are only possible in polymeric bodies; it is in the polymers that such a structural level as the topological one is the most important.Thus, inter- molecular interactions in high-molecular-mass compounds can- not be reduced only to the formation of more or less strong bonds of an energetic nature. In order to understand and describe properties of polymeric substances, it is necessary to take into consideration topological features of their structures. The use of the concept of physical networks 1 allows a more adequate description of polymers: energetic interactions are assumed to be localised at certain points regarded as nodes, whereas the topological aspects of the structure are interpreted V I Irzhak Institute for Problems of Chemical Physics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region, Russian Federa- tion.Fax (7-096) 515 35 88. Tel. (7-096) 517 16 90. E-mail: irzhak@icp.ac.ru Received 17 September 1999 Uspekhi Khimii 69 (3) 283 ± 301 (2000); translated by V D Gorokhov #2000 Russian Academy of Sciences and Turpion Ltd DOI 10.1070/RC2000v06903ABEH000554 261 262 262 264 267 268 271 272 272 273 275 with the aid of terminology used to describe covalent networks (e.g., internodal chains, tails, gel and sol fractions, etc.).2, 3 Any polymeric system can be described as a network whose nodes have a finite lifetime. The concentration of nodes depends on external conditions: temperature, strain, the presence of a solvent. Physical networks, usually called gels, are fairly well studied subjects.4± 7 According to Flory,8 they can be classified as follows: well-ordered lamellar structures, including gel-like mesophases; totally disordered polymeric networks; polymeric networks formed by physical aggregation which are mostly disordered but comprise some ordered regions; and specific disordered struc- tures.This classification comprises covalent network polymers but does not include non-crosslinked polymers. This is due to the fact that in the cases under consideration the nodes of physical networks are represented by certain microphase structures. Com- mon linear or branched polymers devoid of such structures, though not being called gels, exhibit, in definite temperature and time intervals, properties of gels, which are in particular exhibited in the relaxation of the elasticity modulus.Consequently, when considering all polymers, one can make recourse to the above- mentioned Flory classification complemented with disordered polymeric systems comprising the nodes whose nature differs from that of microphases. At present, the most popular viewpoint is that the nodes of a physical network are entanglements, i.e., units of topological and entropic rather than energetic origin.9 ±13 However, it should be noted that in this case too, the concentration of nodes depends on temperature, although the heat of node formation is rather small, *9 kJ mol71.14 The polymers containing functional groups capable of form- ing sufficiently strong energetic bonds (of the hydrogen bond type) can be regarded as an intermediate case, if these groups are distributed along the chain in a manner, which prevents the formation of `zones of physical bonds'.15 According to the Flory classification, the first group of physical networks is macro- molecular in essence.Therefore, if the network polymers are excluded, then all the polymeric systems regarded as physical networks may be classified as follows: (1) networks with nodes of a microphase type which have ordered or disordered structures;262 (2) networks with nodes formed by strong intermolecular bonding of the hydrogen bond type; (3) networks with nodes of entropic (topological) nature. The first group comprises systems which are commonly regarded as physical gels. Numerous papers and reviews have been devoted to the thermodynamic aspects of such systems and their phase structure.4 ± 8, 15 The role of energetic intermolecular interactions in these systems has been established.All other polymers may be included into the second and mostly the third groups. An important point here consists in estimating the contribution of the energetic component and intermolecular interactions to the properties of polymeric systems of these types. II. The role of topological factors. The reptation model A distinctive feature of polymeric systems consists in the existence of the highly elastic state which is manifested in that they exhibit properties of an elastic body over definite time and temperature intervals: above the glass transition temperature and below the flow temperature, one observes a high-elasticity plateau in the relaxation curves.In this range, the real part of the elasticity modulus is slightly dependent on the temperature and time (frequency of measurements) of the experiment; consequently, the plateau is of the entropic rather than energetic origin. It is noteworthy that the above-mentioned weak dependence is rela- tive, because in the interval of low temperatures and high frequencies (short times) the high-elasticity plateau is superim- posed on the glass transition region. For this reason, it is some- times difficult to establish whether the changes in the elasticity modulus are associated with this circumstance, or this parameter depends per se on the temperature and time of the experiment in the plateau region.Evidence in favour of the entropic nature of intermolecular interactions which determine properties of the polymeric systems is provided by the dependences of the range of the high-elasticity plateau and the viscosity of melts and concentrated solutions of polymers on their molecular masses (MM). The theories which substantiate the above dependences (see, e.g., Refs 10, 13, 16 ± 18) are based on the concepts of crucial role of the network of physical (topological) nodes, while the network connectivity (critical conditions of the network appearance) depends on the length of the chains that are cross-linked the correlation between theoretical predictions of basic regularities of diffusion and relaxation in such systems and experimental results requires that some refinements of the type of `constrain release' and `tube relaxation or tube renewal' are introduced.19, 20 The simplest version 13 of the reptation model describes the dynamic properties of a polymer melt in terms of the Rouse theory (R) ifMMis smaller than a certain critical value Mc.In this case, the viscosity Z should be proportional and the diffusion coefficient D inversely proportional to MM; hence, the product ZD is independent of MM. The dynamics of a macromolecule is determined by two characteristic scaling parameters: the segment length l and the mean square radius of the macromolecule coil hR2i1/2. The dependence of the squared displacement of the central segment of the chain on time gm(t) obeys the following relations: (1) gm(t) ! t if gm<l 2, t1/2 if l 2<gm< hR2i, t if gm> hR2i.At MM>Mc , the crucial role is played by the third scaling parameter, i.e., the mean square radius of the chain between entanglements or the `tube' radius hR2e i1/2. The dependence of viscosity and maximum relaxation time on MM should be exponential with the exponent equal to 3. The diffusion coefficient decreases proportionally to the squared MM; consequently, the product ZD is proportional to MM. The squared displacement of V I Irzhak the central segment of the chain as a function of time gm(t) is determined by the following relations: (2) gm(t) ! t if gm<l 2, t1/2 if l 2<gm<hR2e i , t1/4 if hR2e i<gm<hR2i , t1/2 if hR2i N >gm>R2 , Ne t if gm>hR2i N .Ne where N is the chain length and Ne is the chain length between entanglements. Predictions of the dynamics of macromolecules have been the subject of both numerous experimental studies 10, 14 ± 23 and com- puter simulations.24 ± 38 In the latter case, the energy factor was not, as a rule, taken into account, since it was believed that only the excluded volume (the `repulsive' part of the interaction potential) and impermeability of chains are significant. Unfortu- nately, there are virtually no studies (except for the investiga- tions 29 ± 32 on the behaviour of low-molecular chains of the type of paraffin oligomeric chains) in which the `attractive' part of the interaction potential has been estimated.In short, `...the reptation hypothesis and models built thereon give an extremely good qualitative description of a broad range of experimental observa- tions; in many cases, the agreement is quantitative. ...If it is more important for a good model to explain a wide variety of phenom- ena reasonably well than to be in precise agreement with a restricted set of observations, then reptation-based models must be viewed as extremely successful'.19 The existence of topological nodes in the structure of polymers is proved by their fixation upon cross-linking of linear polymers.33 At a sufficiently high degree of cross-linking, the total number of nodes as determined from physical measurements exceeds the number of nodes calculated from the conditions of a chemical experiment by an amount which presumably corresponds to the number of entanglements (topological nodes) in a non-cross- linked system.2 For a number of polymeric systems 2, 39, 40 this finding was confirmed experimentally and by computer simulation.41 ± 44 However, there are data (see, e.g., Ref. 45) which disprove this view.Thus, it would seem that there are no reasons for casting doubts on the existence of entanglements as structural units and their role in the formation of relaxation properties of polymers. However, a detailed analysis of experimental results shows that such reasons exist. III. Correlation between viscosity and viscoelasticity It is usually assumed that the character of the dependence of viscosity on MM, a kink in the curve in the vicinity of certain `critical' value of Mc 46 and the appearance of a high-elasticity plateau { are determined by the same factor, viz., by the presence of topological nodes or entanglements of polymeric chains.14 There- fore, it is natural to suggest if not equality, at least a correlation between the parameters Me and Mc.However, back in 1974, Graessley 17 noticed that such a correlation does not exist and, what is more, theMc values found from the results of viscosity and compliance measurements are conflicting. Having analysed 70 polymers, Aharoni 49 showed that despite the absence of a correlation between Mc and Me definite correspondence can be revealed for small groups of similar polymers.50, 51 However, it should be noted that the values of both viscosity at finite concentrations of polymers and their MM exceed critical values and the elasticity moduli in the high-elasticity plateau region { Based on the concepts of classical theory of network elasticity, the chain length between the network entanglements (Me) may be calculated from the magnitude of the elasticity modulus in this region.9, 47, 48Relaxation properties of polymers and the physical network model depend on the size of the polymeric coil.This circumstance was apparently first brought to light by Fox 52 and this fact was repeatedly confirmed in later studies.48, 53, 54 It has been shown that the Mc values for many polymers approximately correspond to one size of the macromolecular coil: the characteristic viscosity [Z]c corresponding to theMc value of different polymers amounts to*13 dl g71 (26 polymers were analysed).This conclusion was confirmed by Chee 53 (35 polymers) and He and Porter 54 (37 polymers). The latter study has shown that the mean square radius of the coil of the critical size of a macromolecule hR2e i1/2 is constant Table 1. Parameters of a macromolecular coil under critical conditions. Polymer PE PPr PB PIB cis-PB cis-PI PDMS PS PVA PVC PAA PAM PAN PMMA PBMA PHMA POMA PDDMA PEBMA PEO P-e-CL PMA 1,2-PB PTMO PPO P-a-MS P-p-MS P-m-MS PTFE PEIP PETP PDMP PC PDeMS PTMA PDeMA PDeS PHMS PVCz PPS PVN 0.85a 0.85b 70.86b 0.92c 0.906c 0.970d 1.06c 1.108d 1.38c 771.154c 1.19c 1.05c 1.007b 0.971b 0.98a 71.13b 0.985d 1.11 0.899 0.98a 1.000 1.04 1.04a 771.34a 777771.02a 77771.19a 1.04a 1.19a aData from Ref. 53; b data from Ref.48; c data from Ref. 57; d data from Ref. 58; e data from Ref. 54. Note. PE, polyethylene; PPr, polypropylene; PB, poly(1-butene); PIB, polyisobutylene; cis-PB, poly(cis-1,4-butadiene); cis-PI, poly(cis-1,4-isoprene); PDMS, polydimethylsiloxane; PS, polystyrene; PVA, poly(vinyl acetate); PVC, poly(vinyl chloride); PAA, poly(acrylic acid); PAM, polyacrylamide; PAN, polyacrylonitrile; PMMA, poly(methyl methacrylate); PBMA, poly(n-butyl methacrylate); PHMA, poly(n-hexyl methacrylate); POMA, poly(n- octyl methacrylate); PDDMA, poly(dodecyl methacrylate); PEBMA, poly(2-ethylbutyl methacrylate); PEO, poly(ethylene oxide); P-e-CL, poly(e- caprolactam); PMA, poly(methyl acrylate); 1,2-PB, poly(1,2-butadiene); PTMO, poly(tetramethylene oxide); PPO, poly(propylene oxide); P-a-MS, poly(a-methylstyrene); P-p-MS, poly(p-methylstyrene); P-m-MS, poly(m-methylstyrene); PTFE, polytetrafluoroethylene; PEIP, poly(ethylene isophtha- late); PETP, poly(ethylene terephthalate); PDMP, poly(2,6-dimethyl-1,4-phenylene); PC, poly(bisphenol A carbonate); PDeMS, poly(decamethylene succinate); PTMA, poly(tetramethylene adipate); PDeMA, poly(decamethylene adipate); PDeS, poly(decamethylene sebacate); PHMS, poly(hexa- methylene sebacate); PVCz, polyvinylcarbazole; PPS, poly(propylene sulfide); PVN, poly(2-vinyl naphthalate).Mc (see Ref. 49) 3500a 6890 716020 5940 7650 24500 35000 24540 6250 771330 31530 60435 91900 114000 18640a 74410 5020 24100 4110 2530a 7750 40800 5010a 76000b 3270a 7773000b 4000b 6000a 6000b 4500b 4000b 727100a 20 000a 14700a 104Ky /dl g71 22.5a 15 b 710 b 17 b 11.9 c 7.8 d 8 b 8.6 d 12.5 b 7722.5b 6.5b 2.95c 4.3c 5.2b 3.5a 713c 25.3c 7.2d 12.5d 23.1a 11.6d 7.4c 6.9a 717.8b 21.0a 77721.4b 19.6b 19.0a 19.0b 19.8b 20.2b 77.6a 6.0a 6.5a and equal to about 4 nm (the spread of values from 1.3 nm for poly(oxy-2,6-dimethyl-1,4-phenylene) to 5.9 nm for poly(dodecyl methacrylate). An illustrative example of the invariance of the coil critical size was given in a study of Baird and Ballman.55 The dependence of viscosities of the rigid-chain poly(p-phenylene terephthalate) (the persistence length A=130 nm,Mc=1180) and the flexible-chain nylon-6,6 (A=9.4 nm,Mc=5300) onMMis the same, if not the MM proper but the polymer coil size (characteristic viscosity) is used as the argument.Table 1 contains the values of characteristic [Z]c /dl g71 70.125 70.127 0.131 0.104 0.122 0.150 0.134 0.099 770.082 0.115 0.073 0.130 0.062 770.086 0.179 0.112 0.080 70.102 0.1495 770.14b 77770.12b 0.12b 70.15b 0.13b 0.13b 7777 [Z]cr r /g ml71 710.6 710.9 12.0 9.4 11.0 15.9 14.5 13.7 779.5 13.7 7.7 13.1 6.0 779.7 17.6 12.4 7.2 710.2 15.5 77777777777777777 263 hR2i1/2 /nm 2.67e 2.73 4.73e 3.64 2.64 2.66 4.14 4.99 4.28 2.45 4.44e 2.47e 1.37 4.42 4.70 6.57 7.80 5.89e 3.45e 2.08 2.77 4.80 1.98 2.01e 2.98 5.25 74.48e 2.69e 71.63e 2.50e 1.28e 1.88e 2.54e 772.53e 2.57e 2.55e 777264 viscosity [Z]c for different polymers.The coil radius was calculated by the Flory ± Fox formula 56 hRc2i1/2=ÖâZäyMcÜ1=3 , F where [Z]c is the characteristic viscosity of the y-coil and F=2.8761021. It is seen that the values of the critical radii of the coils of different polymers are close.Antonietti et al.59 have shown that macromolecules of poly- styrene with different topologies (linear, cyclic and star-like with different numbers of branches and microgel spheres) also exhibit the same dependence of viscosity on MM if the argument is the polymer coil size. An analogous result was also obtained for other types of polymers, viz., rigid-chain,60 liquid-crystalline,61 comb- shaped,62 and branched 63, 64 ones. The use of characteristic viscosity allows unification of the data on viscosity of different polymer systems over a wide concentration range.65 Such a similarity of the behaviour of macromolecules (includ- ing highly cross-linked globules) which differ greatly in their nature and topological type is often interpreted from a purely geometrical viewpoint: entanglements can arise only when the coils overlap, i.e., when the total volume of coils (the product of the coils concentration by the macromolecular coil volume) is larger than the system's volume.The criterion ratios proposed by different authors 49, 52 ± 54, 66, 67 differ in minor details. However, the fact that dense globules 59 and rigid rods 60 obey the same regularity casts doubts on the validity of the topological approach. The data on the correlation between the elasticity moduli Ge in the high-elasticity plateau region and the molecular characteristics of block polymers have been analysed.13, 49, 54, 57, 68 The value of Table 2.Characteristics of the chains of a network of entanglements. Polymer Me PE PEO PPr PMMA PS PDMS cis - trans-PB cis - trans-PI PEB PPB PBB PIPB PIBB cis-PB cis-PI PMB PPSKS-25 BPR EPR SKF-260 SKF-260 MP PPO PTMO PIB 828a 1624a 4623a 10013a 13309a 12293a 2300b 3900b 5500b 8000b 10800b 8000b 1080b 3500b 4500b 4400b 3600b 3450b 2400b 2100b 6200b 10500b 4500b 2000b 8000b aData from Ref. 57; b data from Ref. 68. Note. PEB, polyethylbutadiene; PPB, polypropylbutadiene; PBB, poly- butylbutadiene; PIPB, polyisopropylbutadiene; PIBB, polyisobutylbuta- diene; PMB, polymethylbutadiene; PP, polypiperylene; BPR, butadiene ± propylene rubber; EPR, ethylene ± propylene rubber; SKS-25, SKF-260 and SKF-260 MP, synthetic rubber brands.[Z]y r [Z]y /cm3 g71 5.1 5.6 8.0 7.4 8.9 7.7 77777779.3 7.9 77777777.8 10.1 8.8 6.47 5.24 10.20 6.50 9.23 8.60 9.8 9.9 9.6 9.8 10.2 9.8 10.2 10.1 8.7 10.5 10.0 10.4 10.1 8.9 8.7 6.15 7.8 10.3 10.2 V I Irzhak the effective volume of the network of entanglements ([Z]y) { was found to be constant irrespective of the magnitude ofMc , which is determined by the nature (rigidity) of a polymeric chain (Table 2). However, the product [Z]y r is somewhat smaller (*8 cm3 g71, on the average, for the polymers studied) than that derived from viscosimetric measurements (13 cm3 g71). At the first glance, this result does not contradict the simple network model according to which Mc à 2Me .However, it should be noted that the same regularities are characteristic of both rigid rods 60 and dense highly cross-linked globules 59 as well as latex dispersions,69 viz., at definite concen- trations and/or MM of polymers, the polymeric systems exhibit viscoelastic properties. Evidently, the elastic properties are deter- mined by the network structure resulting from intermolecular interactions which are apparently of an energetic nature.61, 70 The viscoelastic properties of comb-like polymers are presumed to be also determined by intermolecular interactions due to the side groups.IV. Regularities of viscous flow It is believed that the viscosities of polymers strongly depend on MM (MM>Mc) and concentration of the polymers (exponents of 3.4 and*5, respectively) and are determined by the formation of interchain topological nodes called entanglements (see, e.g., Refs 9, 10, 12 ± 14, 16 ± 19, 46, 71). The entanglements arise in the concentration region where mutual penetration of polymer coils resulting in a cooperative flow of polymer chains occurs. It is presumed 10, 12, 13 that in the frame of the reptation model, the translation diffusion of polymer chains during such a flow is restricted by the `tube', i.e., it becomes substantially anisotropic. Changes in the size of a macromolecular coil upon concen- tration of a solution are highly important in the studies of the dependence of the viscosity of a polymer solution on the concen- tration and MM of this polymer.The dependence of the size of a macromolecular coil on the polymer concentration in the range of semidilute and concentrated solutions was investigated both theoretically 10, 13, 72, 73 and experimentally.74 ± 76 Flory 77 was the first to notice that the coil changes its dimensions upon transition from dilute to concentrated solutions. In the isolated state (in a dilute solution), the conformation of macromolecules is deter- mined by the competition of two factors, viz., the effect of excluded volume and the energy of interaction of monomeric units with the medium (in an athermal solution where the interaction energy does not play any role and the coil swells owing to the effect of excluded volume), while, according to Flory, the coil should have the y-dimensions in the bulk.Edwards et al.13, 73 proposed a theory based on the idea of screening of the energy of repulsion of monomeric units of the same chain by the units of the neighbouring chains. With such a screening, the effect of excluded volume is neutralised: as in the case of dilute solutions, the energy of repulsion of monomers of the same chain is compensated by the energy of their attraction owing to the interaction with the monomeric units of the neigh- bouring chains. In this case, the principal role is played by the entropy factor the nature of which has been clearly explained by de Gennes in a simple example.10 The entropic nature of this factor is manifested in that the temperature factor influences the size of the y-coil in a melt to a substantially lesser extent than in a dilute solution.78 For the same reason, the concentration dependences of the swelling coefficient shown in Fig.1 comprise two regions: in the first region, the shape of the curve is substantially dependent on the nature of the solvent and the curve may be monotonic or have { As in this case one deals with the bulk polymers, the macromolecular coil has the y-dimensions.Relaxation properties of polymers and the physical network model a271 123 0 0.5 u0r 45 Figure 1.Dependence of the swelling coefficient a on the concentration of polymer segments r in a solution at different temperatures (differing solvent quality).72 (17y/T)N: (1) >0.75, (2) 0.75, (3) 0; |17y/T|N: (4) <0.75, (5) >0.75. u0, segment volume; y, Flory temperature; N, chain length. extrema (the region of dilute solutions); in the second region, the shape of the curve is the same for solvents of any type. Variation of the size of a macromolecular coil as a function of the polymer solution concentration is described by power depend- ences, for example:13 2n¢§1=2O1¢§3nU , hR2i ¢§ 1!c1=2 or Rg a R0g c c hR2iy where Rg is the hydrodynamic radius of the polymer coil, c* is the critical concentration (the boundary between dilute and semi- dilute solutions), n is the power index in the relation hR2i1=2!Mn, whereMis the molecular mass of a polymer.Usually, experimental data are also presented in the form of power dependences.74, 79 It should be noted that the above dependences can serve as estimates. As is mentioned in the monograph of Doi and Edwards,13 `the scaling argumentation gives only a qualitative picture of the dependence on basic parameters'. The sought dependence of the macromolecular coil size on the polymer concentration may be derived by analysing computer simulation models.79 ¡¾ 81 As a rule, these models take into consid- eration only the factor of the excluded volume, i.e., the repulsive part of the interaction potential. This means that the computer simulation techniques investigate athermal systems. The results of such simulations indicate 79, 80 ¡¾ 82 that the following simple relation is obeyed over the entire concentration range (Fig.2 a). (3) Vc V0 a 1 a gc , 1 where Vc and V0 are the coil volumes at the concentration c and in an extremely dilute solution, respectively; g is a coefficient. The coefficient g depends on the chain length N (Fig. 3) g=Na , where the exponent a is equal to 1 for short chains and a ^ 0.34 for long chains. Such a character of the dependence may be explained qualitatively by the relationship between the coil size and the length of the chain contour: the linear dependence in the first case and the exponential dependence with the exponent 1/2 in the second case.The validity of relation (3) is illustrated in the experiments showing the effect of `thickening' (increase in concentration) of a polymer solution on the diffusion of macromolecular coils,83 viscosity 2, 84, 85 and characteristic viscosity of the polymer.86 ¡¾ 90 For corresponding analysis of the experimental results, see Ref. 82. a hR2i3=2 0 =hR2i3=2 2.4 2.0 1.6 1.2 0.2 0 b 0 hR2i3=2=hR2i3=2 54321 0 0.4 Figure 2. Dependence of the size of a macromolecular coil on the polymer concentration in the coordinates of Eqn (3); (a), data from computer simulation,81 the numerals at the straight lines indicate the chain length; (b), system PS7carbon disulfide.74 log g 0.4 1 0 2 70.4 70.8 1.5 1.0 Figure 3.Dependence of the coefficient g from Eqn (3) on the chain length according to the data reported by de Vos and Bellemans 79 (1) and Paul et al.81 (2). By comparing changes in the diffusion coefficients of a rigid sphere (chloroform as the solvent) and a macromolecular PS coil (toluene as the solvent) upon concentration of solutions in the PMMA concentration range from 0% to 5%, Brown and Rym- den 83 have concluded that the size of the coil of the PS macro- molecule depends on the `thickener' concentration. As the molecular masses of PS and PMMA are high (ranging from 1.266106 to 156106 for the former and from 1.016105 to 4.456105 for the latter), the solutions in the above-mentioned concentration range are regarded, if not as concentrated, at least as semidilute ones for which the entropy factor is decisive.In the experiments 86 on measuring the characteristic viscosity of PS (MM=3.56106) in a mixture of bromoform with PS with MM=0.466105 or PS with MM=1.16106 as the `thickener', the interval of the system's volume `packing' {parameter [Z]c} is overlapped almost 10 times which is undoubtedly indicative of the transition to the region of concentrated solutions. The results of the studies on the variation of the coil size as a function of concentration of polymers obtained by the neutron scattering method (Fig. 2 b) 74 are not described by relation (3). Possibly, in this system the break in the straight line corresponds to the critical concentration of transition from a dilute solution to semidilute one, i.e., the energy factor plays the decisive role in the initial stage.Daud et al.74 used carbon disulfide which is a good 265 200 100 50 20 0.4 w /(v/v) c /g cm73 0.8 2.0 logN266 solvent. There are no accurate data on the characteristic viscosity of PS in this solvent, although it may be assessed using the Mark ± Kuhn equation [Z]=KMa ; if it is assumed that K=161074 dl g71 and a=0.72,58 then [Z]=0.44 dl g71, while the critical concentration is *0.02 g ml71. Of course, these values are lower than the value of characteristic viscosity corresponding to the inflection point on the graph, though they are close in their orders of magnitude. It is also possible that a definite role in the above discrepancy is played by some additional factors, e.g., the sample prehistory.91 Thus, the analysis of experimental data and results of com- puter simulation make it possible to conclude that formula (3) may be used for the description of the concentration dependence of the coil size in the range of semidilute and concentrated solutions of polymers.(4) It is noteworthy that relation (3) is a consequence of the dependence established for dilute solutions 84, 85 Vc=V0 (17g[Z]c+...) . For low concentrations of polymers (5) Vc7V0=7kV2 0 c , since[Z] !V0 . If the latter expression is assumed to hold for finite values of concentration c, then Eqn (5) may be presented in form of the differential equation (6) dV=7kV2dc, from which Eqn (3) is derived by integration.It is seen that the coefficient g=kV0 is a function of MM. The fact that the principal quantity in Eqn (3) is the product [Z]c means that the scaling relations are virtually fulfilled,10, 13 i.e., the macromolecular coil size depends on the dimensionless parameter c/c*. Equation (6) is valid over the entire range of concentrations for both dilute and concentrated solutions of polymers (the differing k values, see Fig. 2 b). This means that a polymeric solution of any concentration acts as a continuous medium for a macromolecular coil and that there is no difference in principle between the solvent and the polymer solution: the difference is purely quantitative due to different energetic interaction of the polymer chain with the medium and viscosity, which is confirmed by diffusion experiments.83 This may also mean that macromolecular coils behave inde- pendently during the flow of a polymeric solution and/or a melt viz., interactions of the entanglement type do not affect the viscosity.In other words, Eqn (3) holds over the entire range of polymer concentrations (up to its melt). This hypothesis formed the basis for the analysis of the viscosity dependence on the concentration and MM of a poly- mer.2, 92 ± 95 It turned out that over a certain concentration interval depending on the glass transition temperature of a system, all polymers obey the following law: (7) ln Zrel=K71ln[1+K[Z]c (c7c*)] , where Zrel=Z/Zc , Zc is the viscosity of a polymer solution at the critical point, [Z]c is the characteristic viscosity of a polymer at the critical point, K=0.125 for all the polymers studied.As an example, Fig. 4 shows the data on the viscosity of polydimethylsiloxane solutions in siloxane pentamer.96 The curves plotted in the coordinates of Eqn (7) are straightened over the entire range of concentrations and MM, the straight line slope {the product K[Z]c} being dependent on MM in conformity with a Z /P 100 101 0.1 0.01 0.1 b Z0.125 2.0 1.5 1.0 0.5 0.6 0.4 0.2 0 Figure 4. Dependence of the viscosity of polymethylsiloxane on itsMM and concentration in the siloxane oligomer with the chain length equal to 5 plotted in the logarithmic coordinates (a) and the coordinates of Eqn (7) (b);96 103Mw: (1) 490; (2) 152; (3) 88; (4) 36; (5) 11.2; (6) 2.9.rather narrow molecular-mass distribution (M the Mark ± Kuhn equation. In this case, the exponent a is equal to 0.54, which is somewhat larger than under y-conditions but smaller than in an athermal solution. Analogous conclusions have been drawn from other studies.89, 94 The corresponding data are listed in Table 3. According to the data reported by Kataoka and Ueda,96 the Mc of the bulk polymer is 27 000 (30 000, according to Ref. 48). However, there is no kink in the dependence of viscosity on MM plotted in the coordinates of Eqn (7) (Fig. 5 a). Naturally, such a kink will certainly be absent at lower polymer concentrations.The results obtained may be suggested to be due to that the polysiloxane fractions had a broad molecular-mass distribution (Mw=Mn&2). However, the data on the viscosity of PS 59 with a w=Mn51.1) and of polybutadienes 102 also with a narrow molecular-mass distri- bution over a very wide MM range (from 103 to 107) point to the preservation of the above regularity (Fig. 5 b). The linear law Table 3. Parameters of the Mark ± Kuhn equation for the critical point.94 104K /dl g71 Polymer 5.8 9.7 72.95 3.2 7 PDMS PIB a PIB b PVA PVC P-a-MS a Solvent, xylene. b Solvent, carbon tetrachloride. c The references com- prise the studies the results of which were used to calculate parameters of the Mark ± Kuhn equation.V I Irzhak 123456 w /(w/w) 123456 0.8 w /(w/w) a Ref. c 96 97 98 99 100 101 0.54 0.50 0.53 0.56 0.66 0.4Relaxation properties of polymers and the physical network model a Z0.125 2 loga b3 0.4 10 0 1 5 70.4 70.08 0 3 M0.54 1000 logM 5 4 Figure 5. Dependence of Eqn (7) parameters onMMof polydimethylsi- loxane 96 (a) and bulk polydimethylsiloxane 96 (1), PE 59 (2) and polybu- tadiene 102 (3) (b); a is the slope of the straight lines in Fig. 4 b. holds up to MM of the order of 106. Deviation from the linearity observed for samples with larger MM may be due to their higher polydispersity. It should be noted that an analogous deviation from linearity is observed in double logarithmic coordinates.The conclusion about the independence of macromolecular coils in concentrated solutions was confirmed by the data on the relationship between the solution viscosity and the longest relax- ation time (t) determined from the critical gradient rate in the longitudinal flow of solution between two coaxial capillaries.87 It was shown that the following relation holds over a wide range of polymer concentrations t a AMaZaZ0 , RT where A is a coefficient and Z0 is the solvent viscosity. In this equation which is valid for dilute solutions,56 the role of the solvent viscosity is played by the solution viscosity. The above results make it possible to conclude that the critical conditions under which a kink in the plot of the viscosity depend- ence onMMand concentration of a polymer is observed, are not associated with the transition of the individual flow of coils into a cooperative one.This artefact is determined by the representation of the sum of two magnitudes in logarithmic coordinates [see Eqn (7)]. Apparently, the reptative character of the displacement of chains in the flow of polymer solutions is manifested at the MM and concentrations of polymers which are appreciably larger than the critical ones. Thus it was suggested that 13 the true reptation manifests itself in the range of MM of about 800 Mc, i.e., the values of the order of tens of millions. V. Diffusion of macromolecules Various aspects of the translational diffusion of polymers have been reviewed by Green.20 In experiments, the criterion of transition from the R-dynamics to the reptation dynamics was the dependence of the diffusion coefficient on the molecular mass of the chain, which is linear in the R-model and quadratic in the reptation model.Various methods (ranging from radioactive labelling NMR spectroscopy) and different polymers (PS, PE, PDMS, etc.) were used for deriving quantitative data. Having analysed the available experimental results concerning the diffu- sion of polymers in bulk and in mixtures, Green 20 concluded that the diffusion coefficient depends on the mixture composition, the dependence being often non-monotonic. In many cases, the validity of the dependence characteristic of the reptation mechanism was revealed.D ! M7a, where a=2. It should however be emphasised that in the graphs plotted from the results of Green's study 20 all the points fall onto the same straight line for bothMM>Mc andMM<Mc. At the same time, this author points to numerous data which show that 267 the exponent a for the polymers with MM substantially higher than Mc is markedly smaller than 2 (e.g., for PDMS103). In the case of polyisoprene,20 this exponent is, on the contrary, close to 3, while in the 60% solution of PS in dibutyl phthalate 104 a=2.60.2. Comparative investigation of the dependences of viscosity and diffusion of hydrogenated polybutadiene has been carried out by Pearson et al.,105 who have shown that at the MM much higher than Me (MM>9Me) the following relation is respected , (DZ)rep=(DZ)R e M M where the subscripts `rep' and `R' denote the reptation and Rouse models, respectively.Similar results were obtained for PS and hydrogenated poly- isoprene. It was proposed to describe diffusion in the general form using the expression , DZ=(DZ)rep+(DZ)R=(DZ)R 1 a di M M h whereM*=(9.51.1)Me and d=1.50.1. In order to describe the dependence of diffusion coefficient on MM and concentration of a polymer, more complex expressions are used, in particular the `stretched' exponent:106, 107 D ! exp(7kMb) , where b=1 for low MM and b<1 for high MM. Such a character of the dependence is related 108 to the Kohlrausch ¡¾ Williams ¡¾ Watts law.109 This expression is empirical but this does not prevent its use because the power dependences fit experimen- tal results poorly.Indeed, the results reported by Pahl et al.110 indicate that the dependence of the diffusion coefficient of PDMS (with MM ranging from 118 000 to 716 000) on MM is an exponential function rather than a power one. These investigators revealed a trend towards the transition of the translational diffusion to the reptation regime; however, the reptation regime was not achieved because of the time limit (*1 s, with the use of NMR in a gradient field).110 Experimental studies of diffusion were carried out in dilute and semidilute solutions of polymers using different methods, in particular NMR spectroscopy in a gradient field,22, 29, 111, 112 dynamic light scattering 113 and neutron spin echo.23, 114 The dependence of dynamic (including also diffusive) charac- teristics of macromolecules on the solution concentration is determined by the characteristic parameter x, viz., the correla- tional length of the excluded volume.10, 13 The dependence of this parameter on concentration, for example, for a semidilute sol- ution is as follows: x!c¢§v=O3v¢§1U !c¢§3=4 .With the consideration of this formula, the diffusion coeffi- cient of a polymeric chain in the range of concentrations corre- sponding to semidilute solutions is expressed as D ! N72 c71.75 . As noted above, experimental results are more likely to indicate an exponential function.106, 107, 115 Thus depending on the nature of the solvent and the polymer, the value of power n in the `stretched' exponent varied from 0.42 to 2.21.115 In addition, n depends onMM n!M70.25 as results from the analysis of more than 40 systems show.The validity of specific models is confirmed by the results of computer simulation of the dynamics of polymers116, 117 using molecular dynamics 30, 32 and Monte Carlo methods.118 ¡¾ 124 Despite its somewhat artificial character, the Monte Carlo method is often used in studies of physical properties of poly- mers,31 and it was this method that provided basic information on the dynamics of macromolecules. As the investigation of dense268 (bulk) systems by computer simulation techniques is a rather complex problem, the concentration factor was found to play a substantial role: basic results were obtained for the systems in which the polymer concentrations did not exceed 50%.With the Monte Carlo method, the main attention is paid to the investigation of the dependence of the chain diffusion coef- ficient on the concentration and MM.118 In this context, the following trends have been brought to light: (i) as the concentration increases, the power index a in the dependence of maximum relaxation time t onMM t ! Na varies from 2 to 3, which corresponds to the change in the conditions of chain displacement, i.e., from the R-regime to the reptation regime; (ii) at low concentrations, the temporal dependence of the displacement of the macromolecule mass-centre obeys the relation gm ! t 0.54 ; (iii) as the concentration increases, the power index approaches 0.50, which is related to the above-discussed compres- sion of the polymer coil to the y-dimensions; (iv) the squared shear of the inner monomeric chain depends over some time interval on the current time (t) gm ! t 0.3 , which is interpreted as manifestation of the reptation mechanism in which the power index should be equal to 0.25.It should be noted that the extension of the results obtained to the range of polymer concentrations close to 100% leads to a paradoxical conclusion that the dependences of diffusion coeffi- cient and maximum relaxation time onMMbecome stronger than those predicted by the reptation model: the power indexes increase to 2.4 and 3.6, respectively.120 Furthermore, restriction of the lateral shift of the segregated chain was revealed solely in the case on `frozen' environments.If in a mobile system, all chains are equally mobile, the `tube' effect, expressed in the anisotropy of the differentiated chain motion, is not manifested. a t/N2 80 60 40 200 2000 1000 1/(DhR2i1=2) b log(t/N2) 2.0 1.5 4 6 N0.54c 2 0 Figure 6. Dependence of the minimum time of R-relaxation t/N2 on the macromolecule size:81, 119 (a) in the Stokes equation coordinates; (b) con- centration dependence. On the other hand, the analysis of the dependences of the diffusion coefficient D and maximum relaxation time t reported by Paul et al.81, 119 shows that the relationship between t and D71 is linear over the entire range of concentrations studied for all lengths of chains and for all lengths of chains except for 20 (i.e., 50, 80, 100, 200), a single dependence has been established (Fig.6 a) Nt2 à 2:6 á 0:027 . DhR2i1=2 This formula expresses essentially the Stokes law in which the minimum time of R-relaxation t/N2 plays the role of viscosity, whereas the concentration is implicitly accounted for in the values of t and hR2i. Here, the minimum relaxation time } should not depend on the concentration and MM of a polymer due to the effect of hydrodynamic screening.13 However, as one can see from the data shown in Fig. 6 b, such a dependence does exist: the minimum relaxation time is functionally related to the length of the polymeric chain and its solution concentration.It should be emphasised that the dependences observed are linear in the corresponding coordinates and that they have no kinks, which could suggest the transition from the R-mode of relaxation to the reptation regime. Consequently, as in the case of maximum viscosity, cooperativity is not manifested in slow processes of macromolecular diffusion. VI. Stress relaxation In the curves of the stress relaxation of linear polymers with a narrow molecular-mass distribution, one can differentiate several zones corresponding to different states of the system:125 glassy, transient, highly elastic (plateau of high elasticity) and terminal states (see, e.g., Fig. 7 a).It is seen that MM has no effect on the shape of relaxation curves in the glassy and transient states, this effect being manifested only at the end of the high-elasticity plateau, viz., in the terminal zone. This fact is well known. The high-elasticity modulus exhibits rather small variation in the plateau region. A drastic drop of this modulus (by 90% and } This magnitude corresponds to the friction coefficient of a bead of the R-chain. According to the R-model, the polymeric chain is a system of beads (friction elements) connected by elastic elements. a logG(t) (Pa) 753 b log tef 6420 72 0 74 Figure 7. Curves of relaxation of the stress of PS with a narrow molecular-mass distribution;126 (a) elasticity modulus; (b) effective time of relaxation; 1073M: (1) 80; (2) 125; (3) 193; (4) 239; (5) 267. V I Irzhak 12345 logt (s) 4Relaxation properties of polymers and the physical network model more) occurs in the terminal zone, and the kinetics of its decrease strictly obeys the exponential law.All the zones corresponding to the above-listed ranges are clearly seen in the curves of the dependence of effective relaxation time tef on the current time (Fig. 7 b) 126 ¢§1 (8) . tef a ¢§dlnGOtU dt The relaxation time is approximately proportional to the square root of the time in the region of the transient state, this dependence is almost linear (the power index is slightly larger than 1) in the high-viscosity region, then it weakens in the region where MM begins to have its effect and, finally, the relaxation time becomes invariant with time, i.e., tef=tmax.The data reported by other investigators 127 ¡¾ 129 confirm the above-noted regularities of changes in the stress relaxation. The maximum time of relaxation is MM-dependent (Fig. 8) tmax ! M4 . However, this law cannot always be revealed. Indeed, as is seen from the data shown, on the one hand, if the MM interval is not too large, the regularity may be distorted. On the other hand, relaxation measurements of polymers in a wide MM range are usually performed at strongly differing temperatures, and the use of the temperature ¡¾ time superposition principle reduces them to a certain selected temperature. Minor inaccuracies in such nor- malisation may lead to some errors (Fig.8). The analysis of the whole body of available data provides evidence that it is the above-mentioned law that is observed. Let us emphasise that the data shown in Figs 7 and 8 refer to polymers with a narrow molecular-mass distribution; therefore, the peculiarities of the stress relaxation noted above are charac- terised namely by the dependence on the chain length. It is presumed that the relaxation of relatively short chains is well described by the R-model, while the relaxation of fairly long chains is fitted by the Doi ¡¾ Edwards (DE) model. Herewith, the R-relaxation occurs in a small segment of the transient zone, which adjoins the high-elasticity plateau and is characterised by the exponential dependence of relaxation on the current time with the exponent 1/2.13 However, a thorough analysis 130 of the shape of the relaxation curves corresponding to the R- and DE-models has shown (Figs 9 and 10) that one can differentiate three seg- ments (in the logarithmic coordinates) in the dependence of the elasticity modulus relaxation on time , (9) ¢§ p2t tmax GO0U a 1n GOtU pa1 exp Xn where 6n (10) GOtUdt GO0U a p2tmax !¢§1 ?O0 log tmax (s) 6420 1234 72 logM 5 4 Figure 8.Dependence of the maximum relaxation time on MM of PE according to the data of Akovali 127 (1), McGrory and Tuminello 128 (2), Lin 129 (3) and Aklonis and Tobolsky 126 (4). Normalised temperature/8C: (1 and 4) 129; (2) 160; (3) 127.5.269 a log[G(t)/G(0)] 0 72 74 1 2 3 4 5 6 logt 6 4 2 0 b logtef 6 4 3 4 5 1 2 20 logt 6 4 2 0 Figure 9. Curves of the stress relaxation (a) and the temporal depend- ence of the effective time of R-relaxation (b) according to Eqns (8) ¡¾ (10). The chain length: (1) 11; (2) 21; (3) 51; (4) 101; (5) 201; (6) 501. ¢§1=2 , where tmax is the maximum relaxation time and n is the R-chain length. These are (i) the linear segment for the interval of relative relaxation *0.9<G(t)/G(0)<1 and t/t041 (t0 is the minimum time of R-relaxation); (ii) the segment of the power dependence with the power index 1/2 for the interval 0.1<G(t)/G(0)<0.9 (depending on the chain length, the left limit may be smaller) t GO0U ! t0 GOtU this relaxation interval corresponds to the time t of variation of the relaxation time from its minimum to maximum value; (iii) the segment of the exponential decay of the relaxation modulus GGOO0tUU % 1n e¢§t=t. This segment of the curve corresponds to the region in which the elasticity modulus amounts to fractions of a percent of the initial value, while the time t is longer than the maximum relaxation time: t/tmax>1.a log [G(t)/G(0)] 0 72 1 2 3 4 5 6 74 4 6 2 8 logt 0 b log tef 6 5 4 3 2 1 642 4 6 2 8 logt 0 Figure 10. Curves of the stress relaxation (a) and the temporal depend- ence of the effective time of DE-relaxation (b) according to Eqns (8), (11) and (12).The chain length: (1) 11; (2) 21; (3) 51; (4) 101; (5) 201; (6) 501.270 Characteristically, regardless of the time, each of the relaxa- tion curve segments refers to a strictly definite range of values of the relaxation degree G(t)/G(0). Thus the criterion of R-relaxation may be the observance of the power law with the power index equal to 1/2 in the region of the degree of relaxation 0.8 ¡¾ 0.1. The interval corresponding to this relaxation is the longer the higher is theMMof a polymer, which is also inherent in the R-model. It is also typical that the dependence of the effective time of R-relaxation on time tef proves to be proportional to time, namely, in the wide range (where the elasticity modulus decreases proportionally to the square root of the time) (see Fig.9 b). According to the R-model, tmax ! M2. According to the DE-model, temporal changes in the relaxa- tion modulus obeying the relations (11) p¢§2exp ¢§p2t , t GO0U a p82 GOtU pa1;3::: Xn ¢§1 , (12) GOtUdt GO0U a 12 p2t ! ?O0 are characterised by substantially farther extension of the region of high-elasticity plateau than in the R-model (Fig. 10). In this case, the drop of the modulus in the plateau region is insignificant (see Fig. 10 a). The plateau passes smoothly to the exponential decay; the transition interval is rather narrow, while its position in the time scale depends on MM. The temporal dependence of the effective time of DE-relaxa- tion has its characteristic features, too (Fig.10 b). In the R-model, the minimum value of tef is independent ofMMand its maximum value is proportional to the square of MM, whereas in the DE- model the minimum and maximum values of tef are proportional to the square and the cube of MM, respectively. The temporal dependence of the effective relaxation time in the transient range obeys the law tef ! t1/2 . Comparison of the features of the relaxation curves corre- sponding to the models considered with experimental results shows that none of these models fit the experimental results. b logG00(o) (Pa) a logG0(o) (Pa) 5 5 3 4 1 2 1 2 4 5 6 3 3 3 4 5 1 0 72 74 72 74 76 76 logo (s71) Figure 11. Frequency dependences of the real (a) and imaginary (b) parts of the elasticity modulus and of the effective time of PS relaxation (c).131 1073M: (1) 3000; (2) 770; (3) 275; (4) 128; (5) 70; (6) 39. a b logG00(o) logG0(o) 1 0 1 2 72 2 3 72 3 76 4 4 74 5 5 6 710 6 76 0 75 75 logo Figure 12.Frequency dependences of the real (a) and imaginary (b) parts of the elasticity modulus and of the effective relaxation time (c) in terms of the R-model equations. The chain length: (1) 501; (2) 201; (3) 101; (4) 51; (5) 21; (6) 11. V I Irzhak According to the experimental data, the elasticity modulus decreases by *30% in the region of high-elasticity plateau, whereas the models studied suggest a substantially smaller drop, particularly in the case of the DE-model. The experimental curves of relaxation comprise no segment corresponding to the exponential dependence with the power index 1/2, which contradicts the R-model.A segment of the exponential dependence of tef on t, which may be ascribed to the DE-relaxation, can be found only in individual curves in the transient range (transition from the plateau to the exponential drop) (see, e.g., curve 5 in Fig. 7 b). Though, such a segment is virtually absent on the curves describing the relaxation of PS over a widerMMinterval. However, one cannot assume that the high- elasticity plateau corresponds totally to the R-relaxation for the following reasons. Firstly, the dependence of tef on t is non-linear: in the logarithmic coordinates, tan a>1 (*1.3) in the maximum dependence region.Secondly, as noted above, in the R-relaxation the segment corresponding to the transient state covers a consid- erable interval of the relaxation degree which is the longer the larger is the MM. In the experimental curves, this segment is little dependent on the chain length and covers the range of *90% relaxation, which, of course, is substantially more than follows from the DE-model. Finally, the dependence of the maximum relaxation time onMMis a function with the exponent 4 and is not a quadratic function as in the case of R-relaxation, nor is it a cubic function as in the case of reptation model. The frequency (o is the frequency of impact) dependence of the relaxation of the real and imaginary parts of the elasticity modulus for narrow PS fractions is shown in Fig.11.131 The corresponding dependences for the R- and DE-models are pre- sented in Figs 12 and 13. It is seen that the relaxation curves cannot be described by any of the models discussed. Thus the curve reflecting the frequency dependence of the relaxation of the real part of the elasticity modulus has virtually no segment of exponential dependence with the exponent 1/2 characteristic of the R-model. However, in contrast to the curves typical of the DE- model, the experimental curves exhibit a substantial decay (by almost one order of magnitude) of the elasticity modulus in the region of the high-elasticity plateau. The decrease in the imaginary part of the elasticity modulus in the high-frequency region occurs much more slowly than those predicted by the R- and DE-models: the power index in the dependence of elasticity modulus relaxation is equal to 1 for the c log tef (s) 3 1 2 0 3 4 6 73 5 6 0 0 72 74 76 2 logo (s71) logo (s71) log tef c 1 2 3 4 4 5 2 6 0 0 0 75 logo logoRelaxation properties of polymers and the physical network model b a logG00(o) logG0(o) 1 2 3 4 5 6 0 0 72 72 1 2 3 4 5 6 74 74 76 710 76 710 logo 72 Figure 13. Frequency dependences of the real (a) and imaginary (b) parts of the elasticity modulus and of the effective relaxation time (c) in terms of the DE-model equations.The chain length: (1) 501; (2) 201; (3) 101; (4) 51; (5 21; (6) 11. R-model and to 1/2 for the DE-model, whereas according to experimental data, it makes up 0.2 ± 0.3, the latter dependence being the more clear-cut the larger is the MM of polymers.The position of the maximum on the curves reflecting the frequency dependence of the imaginary part of the elasticity modulus depends on MM, though it is not cubic, as follows from the DE- model, but has the exponent 4, i.e., it is analogous to the maximum relaxation time dependence. Variation of the effective time of relaxation (13) tef à G0ÖoÜ oG00ÖoÜ also points to an appreciable distinction from the R- and DE- models, on the one hand, and to a discrepancy between the predictions based on these models and the experimental results, on the other: as in the case of the temporal dependence, on the curve reflecting the frequency dependence of the effective relaxa- tion time one can differentiate two segments with constant values of minimal and maximal times and a transition region where tef is proportional to the frequency (R-model) or to the square root of the frequency (DE-model).According to the experimental data, this dependence is expressed by the relation tef !o a, where 0.5<a<1. Hence, the analysis of experimental data shows that the regularities of relaxation of such narrow fractions of linear polymers cannot be described even qualitatively by either the R-model or the DE-model. The quantitative description of the relaxation of polymers based on the models considered faces serious difficulties; there- fore, numerous semi-empirical and empirical approaches which recommend the use of exponential laws or dependences of the `stretched' exponent type have appeared.For example, it is proposed 125, 132, 133 to express the elasticity modulus relaxation with sums [see, e.g., Eqns (9) and (11)] by analogy with the R-model, though with a different spectrum of relaxation times and a small number of the sum members. The physical meaning of such presentation is unclear. There are equations of the type 134 G(t)&t a e7bt, which can be used for deriving the dependences of the elasticity modulus relaxation on the frequency of measurements of the type of the Havriliak ± Negami dependence 135 G(o)&[1+(ot)2]7a , as well as the Kohlrausch ± Williams ± Watts equation G(t)=a1exp(7k1t b1)+a2exp(7k2t b2) .Equations of this type often give a satisfactory description of the regularities of relaxation processes over a certain time or frequency interval. Attempts are being made to provide them with a theoretical base (e.g., some studies 122, 136, 137 were devoted to the development of coupling theory). It should however be noted that 271 c logtef (c) 1 2 34 65 8642 76 710 72 logo 72 logo no substantial progress concerning the region of highly elastic state has been achieved. VII. Relaxation of chain orientation The adequacy of a specific model is judged from the results of experimental studies on the orientation of non-cross-linked poly- mer chains in a concentrated solution, melt or a network poly- mer.10, 13 According to current theoretical concepts, the orientation of chains in a non-cross-linked polymer is possible if the chain length exceeds the magnitude of Me, i.e., if in a given time interval the polymer chain is a network with its nodes formed by entanglements.However, many experimental studies 138 ± 145 have shown that oligomeric and short chains with the MM much lower than Me can also be oriented in a polymeric matrix composed of the chains the MM of which is substantially larger thanMe or which form a network polymer. In this case, the motive force of the orientation is the so-called nematic interaction, i.e., the manifestation of intermolecular forces of non-topological character.140, 141 The nature of this interaction remains the subject of discussions (see, e.g., Refs 142, 143).It was speculated that the important role is played by the free volume 9 or configurational entropy in the implementation orientation.146 ± 148 It has been proposed 142, 143, 149 that the coefficient e deter- mined by the relation (14) hui6ujiprobe à ehui6ujimatrix , be used as the measure of nematic interaction where ui and uj are the components of the vector of orientation of monomeric chains. The value of e depends on the concentration andMMof short chains: as the concentration of long chains increases, the relaxa- tion of the orientation of short chains is decelerated.140 For example, the value of coefficient e reaches 0.9 for short polybuta- diene chains(MM41000) and decreases to 0.4 as the chain length increases (see Ref.142). Approximately the same value of e is characteristic of the nematic interaction in polybutadiene, ethyl- ene ± propylene 140, 150 and polystyrene 151 block copolymers, as well as of the orientation of short molecules in the polydimethyl- styrene 152 and polyisoprene 153, 154 matrices. The value of e equal to 1 was found for the polydimethylsiloxane matrix ± oligomer system (see Refs 155, 156). Most studies have not revealed any temperature dependence of e, though the opposite data are also available (see, e.g., Refs 157 ± 159). Sotta 160 studied the phenomenon of a free segment orienta- tion in a polymer network by the Monte Carlo method and found that e=0.397. The only factor whose action can explain the occurrence of orientation is the effect of excluded volume.The theoretical approach to the estimation of e magnitude developed by Merrill et al.161 consists in a modification of the DE- model and bears a semi-empirical character. This approach proposes the use of the notion of `nematic interaction' of the end unit of a test chain with the matrix, though its physical meaning is not explained. Upon introduction of groups capable of strong intermolecular interaction, e.g., formation of hydrogen bonds, into polymeric chains, the coefficient e increases substantially and becomes temperature-dependent.149 This points to an important role272 played in this phenomenon by the intermolecular interaction of energetic origin.However, an attempt at theoretical consideration of this factor 162 did not go beyond the frame of empirical incre- ment to the diffusion coefficient which takes into account the lifetime of the arising hydrogen bond. It should be noted that the temperature dependence of the equilibrium constant of the reaction of hydrogen bond formation in macromolecular systems (polybutadienes with theMMranging from 26 000 to 48 500) is markedly weaker than that in low- molecular systems (squalene):163 as the temperature is increased from 20 to 100 8C, the equilibrium constant is changed 4-fold in the polymer and more than 10-fold in squalene. In contrast with the low-molecular model, the entropy of hydrogen bond forma- tion decreases with temperature (provided the enthalpy remains constant, 28.6 kJ mol71).It may be presumed that weak inter- molecular bonds (of the van der Waals type) will exhibit a still weaker temperature dependence. The effect of groups capable of forming hydrogen bonds that have been specially introduced in the polymeric chains on the relaxation properties of polymers is significant:163 ± 169 the region of high-elasticity plateau is widened by several orders of magnitude with regard to frequency, the maximum of the G00(o) is shifted to the low-frequency region also by a few orders; in addition, a shoulder appears on the low- frequency branch.163, 164, 167 Such changes occur upon addition of as little as 1%±4% of phenylurazole groups modifying polybu- tadiene.The observed effect of hydrogen bonds proves to be significantly stronger than the reptative deceleration of the relaxation of polymeric chains.165 VIII. The network of physical bonds as a dynamic factor The effect of fluctuating physical entanglements on the dynamics of a macromolecule was studied by the Monte Carlo method 170 using a classical chain model on the Verdier ± Stockmayer cubic lattice with two types of kinetic units.116, 117 The chain length was varied from 5 to 1000 monomeric units within the observation time intervals which made it possible to follow the motion of the chain as of an integral system. The behaviour of the chain was typical of the R-model, i.e., the displacement of a random monomeric unit gm(t) obeyed the law expressed by relations (1).a log hDR2ÖtÜi 43210 2 0 71 Figure 14. Dependence of displacements of a random monomeric unit and of the chain mass-centre (dashed lines) on time 170 for chains with the unit number N=100 for different quantities of fixed monomeric units (pf) with a lifetime (tl) of 66103 (a) and 5% of fixed monomeric units (b) with different lifetimes; (a): pf=0% (1), 5% (2), 10% (3), 20% (4), 30% (5) and 60% (6); (b): tl=0 (1), 36103 (2), 66103 (3), 36104 (4), 105 (5). For comparison, the Roman numerals I ± III designate the lines with the slopes corresponding to t, t1/2, t1/4. I 123 II 4 6 5 III 4 6 logt If the chain was fixed on the lattice by a virtual network of physical bonds, the number of bonds and their lifetimes being set randomly, the dynamic behaviour of the chain had the features characteristic of the reptation diffusion, i.e., at certain lifetimes and concentrations of nodes, at the distance equal to the medium- length chain scale a regime appeared determined by the relation (Fig.14) gm(t) ! t a , where a was 0.25 (the value characteristic of the reptation regime) or even smaller. The motion of the chain mass-centre is charac- terised by the same dependence [see also relation (2)]. Consequently, the long-living fluctuating physical nodes can provide a polymer system with dynamic properties the manifes- tation of which is usually related to the availability of a network of topological nodes. However, it is unclear whether the appearance of such type of fluctuating long-lived nodes in polymeric systems is possible above the glass transition temperature.IX. On the nature of a physical network nodes It is known that polymeric systems are characterised by rather strong intermolecular interactions of energetic origin above the glass transition temperature.1 This is manifested, for instance, as the ll-transition in their melts.171 This transition temperature is related to the glass transition temperature by the simple expres- sionTll&(1.20.05)Tg. In other words, supramolecular structural units are preserved up to the temperatures 20 ± 60 8C higher than the glass transition temperature: in the opinion of Bershtein and Egorov,171 the liquid ± liquid transition temperature Tll represents the upper temperature limit of the cooperative segmental motion.Bartenev and Barteneva 172 developed concepts of the l-tran- sition, viz., the relaxation process characteristic of the highly elastic and liquid state of polymers. They argue that the l-relax- ation is determined by the existence, at temperatures above the glass transition temperature, of microvolume physical entangle- ments which have linear sizes ranging from 10 to 100 nm and log hDR2ÖtÜi 43210 2 0 71 71 V I Irzhak b I 1 5 234 II III 4 6 logtRelaxation properties of polymers and the physical network model represent long-lived (*104 ¡¾ 105 s) fluctuations of the local ori- entation of segments.Numerous data point to the formation of structural (including microphase) long-lived fluctuations. Thus the application of the techniques of static and dynamic light scattering has shown that 173 that large-scale density fluctuations occur in poly- (methyl-p-toluylsiloxane (Tg=717 8C) up to 90 8C. Poly- (methyl methacrylate) was found to preserve its `memory' about the glass structure (Tg=65 8C) 174 up to 132 8C (13C NMR). The use of EPR and NMR revealed 175 structural heterogeneities in a number of linear and network polyurethanes at temperatures 30 ¡¾ 70 8C higher than the glass transition temperatures. The characteristic sizes of heterogeneities are *10 nm, while their lifetimes are longer than 100 ms. The use of double-quantum NMR spectroscopy 176 revealed an anomalously high degree of orientation of monomeric units in the PB melt.It should be emphasised that the stability of heterogeneities (associates) is strongly dependent on the MM of polymers.176, 177 Naturally, the trend towards the increase in the concentration of long-lived associates in polymeric systems is enhanced as the temperature is decreased and the solvent concentration is increased.178 Simulation of properties of polymer solutions by the Monte Carlo method has shown 179 that there is a trend towards formation of dynamically fluctuating local domain structures with a parallel arrangement of segments if a polymeric chain is locally rigid. Mutual attraction of monomeric units enhances this effect. Analogous results were obtained in a study of polymeric liquids of the n-alkane type by the method of molecular-dynamic simulation.180 In this case, use was made of the Lennard ¡¾ Jones potential with the energy of attractive interaction of non-bound monomeric units equal to 500 J mol71.The extent of ordering decreases with the increase in temperature, although its total collapse is observed at a temperature approximately twice as high as Tg. Experimental studies 181, 182 indicate that associates of reactive oligomers substantially influence the kinetics of their polymer- isation and the rate of relaxation processes in oligomer ¡¾ polymer systems.181 We considered several clear examples of the existence of long- lived structures in polymeric melts, and numerous other examples can be found in recent reviews.1, 182, 183 The role of intermolecular bonds in the manifestation of relaxation properties of polymers is great.It has been shown 163, 165 that even a small number of hydrogen bonds formed in a polymeric chain lead to an appreciable extension of the high- elasticity plateau. The presence of functional groups capable of association substantially influences the rheological properties of polymer systems.184, 185 A characteristic property of polymers is their ability to fix multicentre entanglements.186, 187 Relevant computations 187 indi- cate that the lifetime of a complex formed by several bonds, with the energy of each bond reaching a value of the order of kT, increases exponentially as a function of the number of bonds, the exponent growing substantially as the energy increases (Fig.15). In this case, the lifetime of the complex exceeds by many orders of magnitude the lifetime of a single bond. For example, the lifetime of a complex formed by three bonds (10RT) at 27 8C is 1 s, whereas the lifetime of each bond is*1079 s.189 Such is the effect of connectivity! Even in the absence of energetically strong physical bonds, the monomeric units localised within the same volume element can remain in contact for a long time (which is at least twice as long as the time of release from the `tube').32, 190 In this situation, the number of such `contacting' units is fairly large: it can amount to a half of the total number of units.190 These results were obtained by the methods of molecular dynamics applied to systems in which only the impermeability of chains and the effect of excluded volume were taken into consideration.Ben-Naim et al.190 believe that such long-lived contacts are formed due to the presence of entanglements, viz., topological 273 logta 4 5 3 6 2 4 1 20 1.5 logN 1.0 0.5 Figure 15. Dependence of the lifetime of associates ta on the chain length N at different values of the bond energy;188 E/kT: (1) 0; (2) 0.5; (3) 1.0; (4) 1.5; (5) 2.0. nodes. However, in our opinion, in this case the nature of the nodes is not significant: the presence of some long-lived nodes leads to the appearance of others in long chains, and it is their ensemble that determines the dynamics of chains in concentrated systems.X. Theoretical approaches other than the reptation model Most theoretical approaches proposed for the description of dynamic properties of polymeric systems take into consideration the existence of a network of topological nodes.10, 13, 17 ¡¾ 20, 117 For this reason, there is no need to discuss them here. Instead, let us dwell briefly upon some new ideas. 1. The coupling theory A series of reports (see the bibliography in Ref. 137) discuss various aspects of the coupling theory, the fundamentals of which were formulated by Ngai and Plazek.191 A modification of the R-model is used for describing the dynamics of low-molecular polymeric systems with the MM of a test chain exceeding certain critical MM which corresponds to the chain length between entanglements. The relaxation times tp of all modes increase proportionally to the magnitude (t/tc)np, where np is the coupling parameter, tc is the time of transition of the test chain from the R-motion to a decelerated motion due to its interaction with the surrounding macromolecules.As a result, a simple exponent describing relaxation at short time intervals (t<tc) is transformed into the `stretched' exponent, fOtU a exp ¢§ 1¢§ni h t t where t* is the effective relaxation time . t a tnc 1=O1¢§nU t0 The latter relation takes into account the cooperative motion of the `test' chain and the surrounding chains with a relaxation time characterising the motion of the non-interacting chain (t0).The coupling parameter n varies from 0 to 1; it is dependent on the character of intermolecular interaction, and in many cases n=0.4. The proposed approach made it possible to describe to a good approximation a broad spectrum of dynamic experiments (see, e.g., Ref. 137). To our mind, the coupling theory is of a semi-empirical character, though its authors argue that this theory is based on fundamental ideas of classical mechanics with a chaotic structure determined by the non-harmonic (non-linear) nature of the interaction between molecules.137 The authors do not provide any clear-cut physical justification of the relations used and confine themselves solely to claiming that any cooperative process274 (be it segmental motion or motion of a polymeric chain) is described by the `stretched' exponent.2. The model of lateral diffusion , GOtU a GN Some studies 192 ¡¾ 196 consider diffusion of contacts of a given chain with surrounding chains instead of the diffusive motion of the chain along the `tube'. Since the diffusion occurs over the chain contour, the friction coefficient determined by the presence of contacts is time-dependent. Taking into account this fact and also the lifetimes of contacts measured by Herman 192, 193 the following expression for the relaxation modulus was proposed: !#2 ¢§j 2At2=3 N2 8 p2j 2 exp " ja1;3;::: X where A and N are constants. This model describes well the dependence of self-diffusion coefficient on MM in melts of PE, PEO and PDMS.195 A characteristic feature of the model is the isotropic character of the motion of macromolecules in space, despite the fact that the coefficients of lateral and longitudinal (along the chain) diffusion differ, though not as substantially as predicted by the reptation model, which prohibits the lateral displacement of the chain.The physical meaning of the approach proposed consists in that the friction elements (beads) of the R-model interact with the medium in a more complex manner; in such interactions, the friction coefficient depends on time and the composition of the system (concentration andMMof the matrix). 3. The monomolecular model A generalised linear form of the equation of macromolecule dynamics in the monomolecular approximation has been pro- posed (see the bibliography in Refs 197, 198) jOsUOrai ¢§ oijrjaUt¢§sds¢§ bOsUOrai ¢§ nijraj Ut¢§sds ¢§ md2rai dt2 a O? O? 0 0 agrig a fai OtU , ¢§2mTA (15) where m is the mass of a Brownian particle which is associated with a segment of the macromolecule; raij is the coordinate of the point in which the particle having the number i, j and belonging to the macromolecule `a' is located; b(s) and j(s) are the memory functions; nij and oij are the symmetrised and antisymmetrised components of the velocity gradient; m and A are the coefficient and the matrix characterising the equilibrium form of the macro- molecular coil; t ¡¾ s is the time interval; T is temperature and f is the random force.This approach allowed consideration of various hypotheses relative to the form of the memory functions b(s) and j(s) and comparison of computational and experimental results. In partic- ular, in a simple case of the single relaxation time t: bOsU a 2xdOsU a xt Be¢§s=t , where x is the liquid viscosity and d(s) is the delta-function. In this expression, the relaxation time of the environment proves to be the characteristic relaxation time of the macromolecular coil. The increase in the parameter B, which is the measure in the increase of the particle friction coefficient due to the fact that during slow motions the probe chain pulls the tail from the surrounding macromolecules, obeys the exponential law:197 B ! Mb, where b>2.The first term in the right hand part of Eqn (15) reflects the strength of the interchain interaction, while the second term expresses the strength of the internal resistance associated with changes in the macromolecular coil shape and determined by the V I Irzhak presence of surrounding molecules bound to the chain under consideration. Equation (15) describes essentially the motion of an isolated macromolecule in a continuous medium possessing viscoelastic properties. Thus, the approach proposed takes into consideration the effect of the environment, i.e., the theory does not go beyond the semi-empirical framework. 4. The mode-coupling theory Over the last decade, following the publication of Schweizer's paper,199 many studies have been reported which develop the model of local interaction of monomeric units of polymeric chains.The theory of polymer mode-coupling makes an attempt to describe the chain motion at the microlevel under the action of fluctuating intermolecular forces.200 Microscopic description of the dynamics of a single polymeric chain in a condensed medium is performed using a non-Marko- vian generalisation of the Langevin equation with consideration of the fluctuating force of the intermolecular interaction. The central point of this theory is the conclusion about the memory function, which takes into account the local four-particle (two particles belonging to the probe chain and two others, to the neighbouring chains) contact; in this case, the effect of excluded volume, i.e., the repulsion energy, is regarded as essential.The cell (contact) in which the above-mentioned interaction occurs behaves as an entanglement. These long-lived contacts are pre- sumed to be distributed along the probe chain and their concen- tration to be proportional to the square root of the chain length (see the theory of local diffusion). It is also assumed that the existence of these contacts ensures high cooperativity of the motion of all segments of the probe chain, contrary to what is predicted by the simple R-model. Thus, according to the mode-coupling theory, the relaxation of a polymeric system results from the highly cooperative but isotropic motion of the probe chain and the surrounding chains.It is still unclear whether this picture is realistic or the provided averagings overestimate the `cellular' effect.201 Schweizer et al.202, 203 have thoroughly reviewed and general- ised the theoretical studies and the results of their comparison with the experimental data. It was shown that the conclusions drawn from the theory of polymer mode-coupling and the reptation theory are qualitatively analogous. The mode-coupling theory makes it possible to reveal the sensitivity of diffusion constants to various factors, such as fluctuation of concentrations, degree of polymerisation, composition of a mixture, temperature, concen- tration of a solution and proximity to the interface.201 A disadvantage of this theory consists in a definite arbitrari- ness of the choice of the memory function mode (cf.the mono- molecular model). Furthermore, a detailed analysis has shown 204 that the renormalised procedure does not result in the convergence of results and, consequently, is physically incorrect. 5. The physical network model: relaxation of stress A new approach has been proposed recently for considering the extent and character of the intermolecular interaction in poly- meric systems within the frame of the physical network model.130 Sufficiently long chains of networks with both covalent nodes and entanglements, as well as suspended `tails' were found to exhibit some features of the R-relaxation (see, e.g., Refs 205 ¡¾ 207).It is therefore natural to represent a macromolecule as if it contained k R-subchains of length m connected by physical nodes. In this case, relaxation may be described in the frame of the R-model taking into account that the relaxation of the rigidity of the `springs' connecting the nodes follows the R-mechanism, whereas the nodes themselves are characterised by a friction coefficient, which is determined by the interchain interaction and hence depends on the concentration andMMof a polymer. The initial system of differential equations which describes the relaxation of the R-chain is expressed as 71Relaxation properties of polymers and the physical network model (16) Z dyi dt a GO2yi ¢§ yia1 ¢§ yi¢§1U , where yi is the coordinate of the ith bead and G is the elasticity modulus of the spring connecting the friction nodes.Solution of this system provides an expression for the tempo- ral dependence of the elasticity modulus relaxation [see Eqn (9)]. The physical network model assumes that G is time-depend- ent. Introduction of the time function j into Eqn (16): GOxUdx , jOtU a Ot 0 , GOtU a 1 ¢§p2j tt and its solution yield the equation analogous to that for the R-model, though the expression of time in it is more complex: k pa1 exp Xk where , (17) jOtU a gt a c ¢§q2 t c m qa1 h q¢§2 1 ¢§ exp i Xm g is the elasticity modulus of the R-springs, c=c0m2 is the maximum time of R-subchain relaxation and c0 is the minimum relaxation time.Evidently, the total chain length is n=mk , where k is proportional to the chain length. According to the R-model, the relaxation time t is proportional to the squared chain length and to the friction coefficient. But the latter param- eter per se should depend on n. This dependence is rather complex and its determination is a separate problem, though it may be presumed that the second power is a fairly reasonable magnitude. Thus, in conformity with the experimental results, we derive the general dependence t!n4. The relaxation curves calculated from Eqn (17) are qualita- tively similar to the experimental curves.130 Thus changes in the relaxation of elasticity modulus in the region of high-elasticity plateau are rather significant, and the dependence G(t)!t7a, where a ^ 0.25, is observed to hold over a certain interval.The segment of the exponential dependence with the exponent 1/2 is absent if the number of nodes k is not too large. At longer times, the exponential decay is determined by the relaxation time which is proportional to the chain length raised to the fourth power (more precisely, to k4); extrapolation of the curve to the ordinate axis gives a segment whose magnitude is dependent on the total length of the chain and the number of physical nodes: lnGO0U a ¢§lnk ¢§ c0n g t0k5 b ¢§ t0k4 t , p¢§2&1:6 for sufficiently large m; at m??, where b a pa1 Xm b=p2/6=1.645. A detailed analysis of experimental results (including studies of diffusion of macromolecules, rheology of solutions and melts, frequency dependence of relaxation modulus, etc.) with consid- eration of physical bonds is a matter for the future, but even now one can discern good prospects for this approach for studying the dynamics of polymeric systems.For example, Walkenhorst 208 showed that the process of physical network rearrangement is responsible for the long-term relaxation. The concept of the network of physical nodes was used by Soloviev et al.209 ¡¾ 211 for the description of the relaxation proper- ties of elastomers. In these studies, analysis of the deformation curves differentiated two types of processes differing in their relaxation times: fast diffusive displacements of `free' segments 275 which are not involved in the formation of intermolecular nodes of the physical network and a slow diffusive process of rearrange- ment of the network itself.Using basic postulates of non-equili- brium thermodynamics and considering the destruction of physical nodes as a reversible chemical reaction, these authors derived a system of equations which describe viscoelastic proper- ties of elastomers over a wide range of deformation rates. Inves- tigation of the stretching of copolymers of acrylamide with nonyl acrylate at the rates ranging from 1 to 0.01 s71 and in the temperature range 20 ¡¾ 120 8C has demonstrated that experimen- tal data correlate with theoretical predictions. The approach of Soloviev et al. represents, in its essence, the development of Bartenev's idea 172 about the occurrence of a fast (a-process) a slow (l-process) stages of the relaxation of elastomers.Wientjes et al.212 made an attempt to employ the physical network model for describing rheological properties of polymers. XI. Conclusion This review has analysed the available experimental data on the dynamics of polymers with consideration of the network of physical bonds. It has been found that these data are at variance with the reptation model and that this model is insufficient for understanding the relaxation behaviour of polymers. The authors of this theory agree that the classical variant of the reptation theory yields results which are inconsistent in many respects with the experimental data.13 The attempts to refine the reptation model through consideration of the `tube' relaxation and the disappearance of restricting entanglements 13, 18 have resulted in a certain enhancement of the dependence of maximum relaxation time and viscosity on MM.These and analogous amendments `facilitate' the relaxation process and decrease the relaxation times compared to those given by the simple non-modified theory. At the same time, the review demonstrates that the theoretical relaxation times are underestimated compared to those obtained experimentally, whereas the reptative behaviour is characteristic of the chains with theMMsubstantially larger thanMe. The existence of a network of physical bonds confirmed experimentally is indisputable. However, its role in determining a broad spectrum of dynamic phenomena has not yet been elucidated. Unfortunately, the physical network model has not been developed to the extent reached, e.g., by the reptation model.The investigators have still to clarify whether the intermolecular interactions play a role in the formation of relaxation properties of polymeric systems or these interactions merely make some con- tribution to the network of entanglements.213, 214 The existence of the network of physical bonds is taken into account explicitly or implicitly in the consideration of dynamic properties of polymers. It underlies many approaches: for instance, some aspects of the problem of physical nodes and their effect on relaxation processes in polymers were discussed by Bartenev and Barteneva.172 As a matter of fact, the theory of polymer mode coupling also uses the physical network model (the network nodes are assumed to be areas with high local packing).The results reported by Sotta 160 point to an important role played by the network: orientation of a `free' segment is brought about under the action of the oriented matrix without introduction of some energetic factors due only to the excluded volume effect. Apparently, the fact that the temperature position of the beginning of the high-elasticity plateau (Tg) corresponds not to MM but rather to Me , which is much smaller in order of magnitude than the corresponding size of the kinetic segment,215 should be interpreted from the same viewpoint.This may imply that an important, if not decisive, significance in the physical network formation belongs to the high-density fluctuations, which seem to play the role of nuclei of the glassy state 216 or of the regions of local order in terms of the cluster model of the structure of amorphous polymers.217 Only the existence of long-lived nodes of the network, and not their nature, is important for the manifestation of `reptative'276 properties; therefore, the physical network model may be used for describing dynamic properties of a large number of systems ranging from physical gels to polymers with intermolecular interactions of the dispersive type. It should be mentioned in conclusion that about three decades ago a vivid discussion was underway on whether or not such a structural formation as topological entanglements exists in poly- meric systems.The discussion was initiated by the appearance of the `packet' model of the structure of polymers 218 which was the antithesis of the `felt' structure of Flory.77 This discussion resulted in the triumph of the idea of entanglements not only in the structural respect; it also laid the foundation for the subsequent models of the dynamics of macromolecules. Now, the time has come for a new discussion on whether the entanglements do play the role ascribed to them in the dynamics of polymers or this role is at least overestimated. Preparation of this review was supported by the Russian Foundation for Basic Research (Project No.98-03-33358a). References 1. V I Irzhak, G V Korolev, M E Solov'ev Usp. Khim. 66 179 (1997) [Russ. Chem. Rev. 66 167 (1997)] 2. V I Irzhak, B A Rozenberg, N S Enikolopyan Setchatye Polimery (Cross-Linked Polymers) (Moscow: Nauka, 1979) 3. R T F Stepto (Ed.) Polymer Networks. Principles of Their Formation, Structure and Properties (London; Weinheim; New York; Tokyo; Melburn; Madras: Blackie Academic and Professional, 1998) 4. J P Cohen Addad (Ed.) Physical Properties of Polymeric Gels (Chichester; New York; Brisbane; Toronto; Singapure: Wiley, 1996) 5. WBurchard, S B Russ-Murphy (Eds) Physical Networks. Polymers and Gels (London; New York: Elsevier, 1990) 6. K Te Nijenhuis Adv. Polym. Sci. 130 1 (1997) 7. A H Clark, S B Ross-Murphy Adv.Polym. Sci. 83 57 (1987) 8. P J Flory Farad. Disc. Chem. 57 7 (1975) 9. J D Ferry Viscoelastic Properties of Polymers (New York: Wiley, 1980) 10. P-G De Gennes Scaling Concepts in Polymer Physics (Ithaca; London: Cornell University Press, 1979) 11. G M Bartenev, S Ya Frenkel' Fizika Polimerov (The Physics of Polymers) (Leningrad: Khimiya, 1990) 12. A Yu Grossberg, A R Khokhlov Statisticheskaya Fizika Makromo- lekul (Statistical Physics of Macromolecules) (Moscow: Nauka, 1989) 13. M Doi, S F Edwards The Theory of Polymer Dynamics (Oxford: Clarendon Press, 1986) 14. R S Porter, J F Johnson Chem. Rev. 66 1 (1966) 15. S B Ross-Murphy, in Polymer Networks. Principles of Their Forma- tion, Structure and Properties (Ed.R T F Stepto) (London; Weinheim; New York; Tokyo; Melburn; Madras: Blackie Academic and Professional, 1998) p.288 16. F Bueche Physical Properties of Polymers (New York; London: Interscience, 1962) 17. W W Graessley Adv. Polym. Sci. 16 3 (1974) 18. W W Graessley Adv. Polym. Sci. 47 67 (1982) 19. T P Lodge, N A Rotstein, S Prager Adv. Chem. Phys. 79 1 (1990) 20. P F Green, in Diffusion in Polymers (Ed. P Neogi) (New York; Basel; Hong Kong: Marcell Decker, 1996) p. 251 21. J Klein Nature (London) 271 143 (1978) 22. G Fleischer, O E Zgadzai Colloid Polym. Sci. 266 208 (1988) 23. B Ewen, U Maschke, D Richter, B Farago Acta Polym. 45 143 (1994) 24. D Richter, L Willner, A Zirkel, B Farago, L J Fetters, J S Huang Macromolecules 27 7437 (1994) 25. T Kotaka, K Adachi Macromol.Symp. 101 123 (1996) 26. J D Wilson, R F Loring J. Chem. Phys. 103 1641 (1995) 27. P T Callaghan, E T Samulski Macromolecules 31 3693 (1998) 28. D Yavich, D W Maed, J P Oberhauser J. Rheol. 42 671 (1998) 29. E Fischer, R Kimmich, N Fatkulin J. Chem. Phys. 104 9174 (1996) 30. B DuÈ nweg, G S Grest, K Kremer IMA Vol. Math. Its Appl. 102 159 (1998) V I Irzhak 31. J J De Pablo, F A Eskobedo, in Encyclopedia of Computational Chemistry Vol. 3 (Ed. P von R Schleyer) (Chichester; New York; Weinheim; Brosbane; Singapure; Toronto: Wiley, 1998) p. 1763 32. S W Smith, C K Hall, B D Freeman, J A McCormic IMA Vol. Math. Its Appl. 102 203 (1998) 33. K Kremer, G S Grest, in Monte Carlo and Molecular Dynamics Simulation in Polymer Science (Ed.K Binder) (Oxford: Oxford University Press, 1995) p. 194 34. K Kremer, G S Grest, in Computer Simulation of Polymers (Ed. R J Roe) (Cliffs, NJ: Prentice Hall Englewood, 1991) p. 167 35. G D Smith, D Y Yoon, R L Jaffe Macromolecules 28 5897 (1995) 36. W Paul, G D Smith, D Y Yoon Macromolecules 30 7772 (1997) 37. W Tschoep, K Kremer, J Batoulis, T BuÈ rger, O Hahn Acta Polym. 49 61 (1998) 38. W Tschoep, K Kremer, J Batoulis, T BuÈ rger, O Hahn Acta Polym. 49 75 (1998) 39. L Mullins J. Polym. Sci. 19 225 (1956) 40. C G Moore, W F Watson J. Polym. Sci. 19 237 (1956) 41. E R DuÈ ring, K Kremer, G S Grest Phys. Rev. Lett. 67 3531 (1991) 42. G S Grest, K Kremer, E R DuÈ ring Europhys. Lett. 19 195 (1992) 43. E R DuÈ ring, K Kremer, G S Grest Macromolecules 26 3241 (1993) 44.E R DuÈ ring, K Kremer, G S Grest J. Chem. Phys. 101 8169 (1994) 45. M A Sharaf, J E Mark, A S Alshamsi Polym. J. 28 375 (1996) 46. G C Berry, T G Fox Adv. Polym. Sci. 5 261 (1968) 47. G V Vinogradov, E A Dzyura, A Ya Malkin, V A Grechanovskii J. Polym. Sci., A-2 9 1153 (1972) 48. D W Van Krevelen Properties of Polymers Correlations with Chemical Structure (Amsterdam: Elsevier, 1972) 49. S M Aharoni Macromolecules 19 426 (1986) 50. R S Porter, J F Johnson Rheol. Acta 7 332 (1986) 51. A Ya Malkin, E A Dzyura, G V Vinogradov Dokl. Akad. Nauk SSSR 188 1328 (1969) 52. T G Fox, V R Allen J. Chem. Phys. 41 344 (1964) 53. K K Chee J. Appl. Polym. Sci. 33 1067 (1987) 54. T He, R S Porter Macromol.Chem., Theory Simul. 1 119 (1992) 55. D C Baird, R L Ballman J. Rheol. 23 505 (1979) 56. V N Tsvetkov, V E Eskin, S Ya Frenkel' Struktura Makromolekul v Rastvore (The Structure of Macromolecules in Solution) (Moscow: Nauka, 1964) 57. L J Fetters, D J Lohse, D Richter, T A Witten, A Zirkel Macromolecules 27 4639 (1994) 58. Yu S Lipatov, A E Nesterov, T M Gritsenko, R A Veselovskii Spravochnik po Khimii Polimerov (Handbook on the Chemistry of Polymers) (Kiev: Naukova Dumka, 1971) 59. M Antonietti, T Pakula, W Bremser Macromolecules 28 4227 (1995) 60. H Rehage, H Hoffmann Disc. Farad. Soc. 76 363 (1983) 61. V G Kulichikhin Mol. Cryst. Liq. Cryst. 169 51 (1989) 62. V Fourmaux-Demange, A Brulet, J P Cotton, L Hilliou, P Martinoty, P Keller, F Boue Macromolecules 31 7445 (1998) 63.L A Uttracki J. Polym. Sci., Polym. Lett. Ed. 11 717 (1973) 64. P G Santaneglo, C M Roland J. Non-Cryst. Solids 235 ± 237 709 (1998) 65. V E Dreval', Doctoral Thesis in Chemical Sciences, Ural State University, Sverdlovsk, 1974 66. W W Graessley, S F Edwards Polymer 22 1329 (1981) 67. T A Kavassalis, J Noolandi Macromolecules 22 2709 (1989) 68. E A Sidorovich Vysokomol. Soedin., Ser. A 37 2020 (1995) a 69. R de Rooij, A A Potanin, D van den Ende, J Mellema J. Chem. Phys. 99 9213 (1993) 70. W G Miller J. Polym. Sci., Part C 65 91 (1978) 71. G V Vinogradov, A Ya Malkin Reologiya Polimerov (Rheology of Polymers) (Moscow: Khimiya, 1977) 72. I E Erukhimovich, V I Irzhak, V G Rostiashvili Vysokomol.Soedin., Ser. A 18 1470 (1976) a 73. M Muthukumar, S F Edwards J. Chem. Phys. 76 2720 (1982) 74. M Daud, J P Cotton, B Farnoux, G Jannink, G Sarma, H Benoit, R Duplessix, C Picot, P G de Gennes Macromolecules 8 804 (1975) 75. A A Dadali, L I Kuzub, L M Bogdanova, V I Irzhak Vysokomol. Soedin., Ser. A 19 1393 (1977) a 76. Yu B Grebenshchikov, V I Irzhak, L I Kuzub, P P Kushch, N S Enikolopyan Dokl. Akad. Nauk SSSR 210 1124 (1973) b 77. P J Flory Principles of Polymer Chemistry (Ithaca, NY: Cornell University Press, 1953)Relaxation properties of polymers and the physical network model 78. A T Boothroyd, A R Rennie, G D Wignal J. Chem. Phys. 99 9135 (1993) 79. E De Vos, A Bellemans Macromolecules 7 812 (1974) 80. O F Olaj,W Lantschbauer Makromol. Chem., Rapid Commun.3 847 (1982) 81. W Paul, K Binder, D W Heermann, K Kremer J. Phys. II 1 37 (1991) 82. V I Irzhak Vysokomol. Soedin. 42 (2000) a 83. W Brown, R Rymden Macromolecules 21 840 (1988) 84. V P Budtov Vysokomol. Soedin., Ser. A 9 765 (1967) a 85. S R Rafikov, V P Budtov, Yu B Monakov Vvedenie v Fizikokhi- miyu Rastvorov Polimerov (Introduction to Physical Chemistry of Polymer Solutions) (Moscow: Nauka, 1978) 86. V G Baranov, Yu V Brestkin, S A Agranova, V N Pinkevich Vysokomol. Soedin., Ser. B 28 841 (1986) a 87. V G Baranov, D Kh Amribakhshov, Yu V Brestkin, S A Agranova, I S Saddikov Vysokomol. Soedin., Ser. A 29 1190 (1987) a 88. V G Baranov, Yu V Brestkin, V I Poddubnyi, S A Agranova, V N Pinkevich Vysokomol.Soedin., Ser. A 29 2516 (1987) a 89. V G Baranov, D Kh Amribakhshov, S A Agranova, S Ya Frenkel' Vysokomol. Soedin., Ser. B 30 384 (1988) a 90. A N Krasovskii, D N Polyakov, V Baranov, S S Mnatsakanov Vysokomol. Soedin., Ser. A 33 1221 (1991) a 91. D K Potter, A Rudin Macromolecules 24 213 (1991) 92. V I Irzhak, L I Kuzub, N S Enikolopyan Dokl. Akad. Nauk SSSR 214 1340 (1974) b 93. V I Irzhak, in Karbotsepnye Polimery (Carbon-Chain Polymers) (Moscow: Nauka, 1977) p. 192 94. V I Irzhak, in Mashiny i Tekhnologiya Pererabotki Kauchukov, Polimerov i Rezinovykh Smesei (Machines and Technology of Processing of Caoutchoucs, Polymers and Rubber Mixtures) Yaroslavl': Yaroslavl' Polytechnic Institute, 1978) No. 2, p. 29 95. V I Irzhak Int.J. Polym. Mater. 8 201 (1980) 96. T Kataoka, S Ueda J. Polym. Sci., A-2 5 973 (1967) 97. M E Johnson,W W Evans, J Iordan, J D Ferry J. Coll. Sci. 7 498 (1952) 98. A A Tager, V E Dreval', F A Khasina Vysokomol. Soedin. 5 432 (1963) a 99. J Schurz, K MuÈ ller Makromol. Chem. 121 58 (1969) 100. G Pezzin, N Gligo J. Appl. Polym. Sci. 10 1 (1966) 101. M Sakai, T Fujimoto, N Nagasawa Macromolecules 5 786 (1972) 102. R H Colby, L J Fetters,W W Graessley Macromolecules 20 2226 (1987) 103. T Cosgrove, P C Griffiths, J Hollingshurst, R D S Richards, J A Semleyn Macromolecules 25 6761 (1992) 104. N Nemoto, T Kojima, T Inoue, M Kishine, T Hirayama, M Kurata Macromolecules 22 3793 (1989) 105. D S Pearson, L J Fetters,W W Graessley, G V Strate, E von Meerwall Macromolecules 27 711 (1994) 106.G D J Phillies J. Phys. Chem. 96 10061 (1992) 107. G D J Phillies Macromolecules 20 558 (1987) 108. B Nystrom, J Roots Macromolecules 24 184 (1991) 109. G Williams, D C Watts Trans. Farad. Soc. 66 80 (1970) 110. S Pahl, G Fleischer, F Fujara, B Geil Macromolecules 30 1414 (1997) 111. G Fleischer, O E Zgadzai, V D Skirda, A I Maklakov Colloid Polym. Sci. 266 201 (1988) 112. J-P Cohen Addad, A Guillermo J. Chem. Phys. 111 7131 (1999) 113. P Stepanek, W Brown Macromolecules 31 1889 (1998) 114. B Ewen, D Richter Adv. Polym. Sci. 134 1 (1997) 115. G D J Phillies Macromolecules 19 2367 (1986) 116. K Binder (Ed.) Monte Carlo Method and Molecular Dynamics Simulations in Polymer Science (Oxford: Oxford University Press, 1995) 117. Yu Ya Gotlib, A A Darinskii, Yu E Svetlov Fizicheskaya Kinetika Makromolekul (Physical Kinetics of Macromolecules) (Leningrad: Khimiya, 1986) 118. K Binder,W Paul J.Polym. Sci. B, Polym. Phys. 35 1 (1997) 119. W Paul, K Binder, D W Heermann, K Kremer J. Chem. Phys. 95 7726 (1991) 120. J Skolnick, A Kolinski, R Yaris J. Chem. Phys. 86 1567 (1987) 121. A Kolinski, J Skolnick, R Yaris J. Chem. Phys. 86 7164 (1987) 122. K L Ngai, S L Peng, J Skolnick Macromolecules 25 2184 (1992) 277 123. M Wittkop, S Kreitmeier, D GoÈ ritz Comput. Theor. Polym. Sci. 6 41 (1996) 124. J Baschnagel,W Paul, V Tries, K Binder Macromolecules 31 3856 (1998) 125. S Matsuoka Relaxation Phenomena in Polymers (Munich; Vienna; New York; Barcelona: Hansen Publishers, 1992) 126.J J Aklonis, A V Tobolsky J. Appl. Phys. 36 3483 (1965) 127. G Akovali J. Polym. Sci. A-2 5 875 (1967) 128. W J McGrory,W H Tuminello J. Rheol. 34 867 (1990) 129. Y-H Lin Macromolecules 19 159 (1986) 130. T F Irzhak, L I Kuzub, O V Nikitina, V I Irzhak Vysokomol. Soedin. 42 (2000) 131. A Schausberger, G Schindlauer, H Janeschitz-Kriegl Rheol. Acta 24 220 (1985) 132. G Williams J. Non-Cryst. Solids 131-133 1 (1991) 133. M BaumgaÈ rtel, A Schausberger, H H Winter Rheol. Acta 29 400 (1990) 134. L C E Struik Physical Ageing of Amorphous Polymers and Other Materials (Amsterdam: Elsevier, 1978) 135. S Havriliak, S Negami Polymer 8 161 (1967) 136. K L Ngai, A SchoÈ nhals, E Schlosser Macromolecules 25 4915 (1992) 137.K L Ngai, D J Plazek, R W Rendell Rheol. Acta 36 307 (1997) 138. J F Tassin, L Monnerie Macromolecules 21 1846 (1988) 139. J F Tassin, L Monnerie, L Fetters Macromolecules 21 2404 (1988) 140. J A Kornfeld, G G Fuller, D S Pearson Macromolecules 22 1334 (1989) 141. H Watanabe, T Kotaka, M Tirrell Macromolecules 24 201 (1991) 142. C M Ylitalo, J A Zawada, G G Fuller, V Abetz, R Stadler Polymer 33 2949 (1992) 143. A K Rizos, T Jian, K L Ngai Macromolecules 28 517 (1988) 144. C Hayes, L Bokobza, F Boue, E Mendes, L Monnerie Macromolecules 29 5036 (1996) 145. A K Kalkar, F Pfeifer, W Siesler Macromol. Chem. Phys. 199 667 (1998) 146. G Adam, J H Gibbs J. Chem. Phys. 43 139 (1965) 147. E A Di Marzio J.Chem. Phys. 36 1563 (1962) 148. T Tanaka, G Allen Macromolecules 10 426 (1977) 149. U Seidel, R Stadler, G Fuller Macromolecules 28 3739 (1995) 150. C M Ylitalo, J A Kornfeld, G G Fuller, D S Pearson Macromolecules 24 749 (1991) 151. C M Ylitalo, G G Fuller, V Abetz, R Stadler, D S Pearson Rheol. Acta 29 543 (1990) 152. H Toriumi, B Deloche, J Herz, E T Samulski Macromolecules 18 304 (1985) 153. J-P Queslel, B Erman, L Monnerie Polymer 29 1818 (1988) 154. B Erman, J-P Jarry, L Monnerie Polymer 28 727 (1987) 155. P Sotta, B Deloche, J Herz, A Lapp, D Durand, J-C Rabadeux Macromolecules 20 2769 (1987) 156. P Sotta, B Deloche, J Herz Polymer 29 1171 (1988) 157. B Deloche, E T Samulski Macromolecules 14 575 (1981) 158. E W Thulstrup, J Michl J.Am. Chem. Soc. 104 5594 (1982) 159. P Schmidt, B Schneider Makromol. Chem. 184 2075 (1983) 160. P Sotta Macromolecules 31 8417 (1998) 161. W W Merrill, M Tirrell, J F Tassin, L Monnerie Macromolecules 22 896 (1989) 162. L Leibler, M Rubinstein, R H Colby Macromolecules 24 4701 (1991) 163. L L de Lucca Freitas, R Stadler Macromolecules 20 2478 (1987) 164. R Stadler, L de Lucca Freitas Colloid Polym. Sci. 266 1102 (1988) 165. R Stadler, L de Lucca Freitas Macromolecules 22 714 (1989) 166. L de Lucca Freitas, C Auschra, V Abertz,R Stadler Colloid Polym. Sci. 269 566 (1991) 167. M MuÈ ller, U Seidel, R Stadler Polymer 36 3143 (1995) 168. M MuÈ ller, R Stadler, K Kremer, G Williams Macromolecules 28 6942 (1995) 169. M MuÈ ller, A Dardin, U Seidel, V Balsamo, B Ivan, H W Spiess, R Stadler Macromolecules 29 2577 (1996) 170.S E Varyukhin, V I Irzhak Dokl. Akad. Nauk 366 749 (1999) b 171. V A Bershtein, V M Egorov Differentsial'naya Skaniruyushchaya Kalorimetriya v Fizikokhimii Polimerov (Differential Scanning Calorimetry in the Physical Chemistry of Polymers) (Leningrad: Khimiya, 1990)V I Irzhak 278 172. G M Bartenev, A G Barteneva Relaksatsionnye Svoistva Polimerov (Relaxation Properties of Polymers) (Moscow: Khimiya, 1992) 173. T Kanaya, A Patkowski, E W Fischer, J Seils, H GlaÈ ser, K Kaji Acta Polym. 45 137 (1994) 174. A S Kulik, D Radloff, H W Spiess Macromolecules 27 3111 (1994) 175. B E Krisyuk, G I Sandakov Vysokomol. Soedin., Ser. A 37 615 216. V G Rostiashvili, V I Irzhak, B A Rozenberg Steklovanie Polimerov (Vitrification of Polymers) (Leningrad: Khimiya, 1987) 217. G V Kozlov, D S Sanditov Angarmonicheskie Effekty i Fiziko- Mekhanicheskie Svoistva Polimerov (Anharmonic Effects and Physical Mechanical Properties of Polymers) (Novosibirsk: Nauka, 1994) 218. V A Kargin, G L Slonimskii Kratkie Ocherki po Fizikokhimii Poli- merov (Essays on Physical Chemistry of Polymers) (Moscow: Khimiya, 1967) a�Polym. Sci. (Engl. Transl.) b�Dokl. Chem. Technol., Dokl. Chem. (Engl. Transl.) c�Physics-Uspekhi (Engl. Transl.) (1995) a 176. R Graf, A Heuer, H W Spiess Phys. Rev. Lett. 80 5738 (1998) 177. G Szamel, T Wang J. Chem. Phys. 107 10793 (1997) 178. M Heckmeier, M Mix, G Strobl Macromolecules 30 4454 (1997) 179. A Kolinski, J Skolnick, R Yaris Macromolecules 19 2550 (1986) 180. D Rigby, R-J Roe J. Chem. Phys. 89 5280 (1988) 181. S M Mezhikovskii Fizikokhimiya Reaktsionnosposobnykh Oligomerov (Physical Chemistry of Reactive Oligomers) (Moscow: Nauka, 1998) 182. S M Kurmaz, Candidate Thesis in Chemical Sciences, Institute for Problems of Chemical Physics, Russian Academy of Sciences, Cherogolovka, 1998 183. R BoÈ mer Curr. Opin. Solid Mater. Sci. 3 378 (1998) 184. M Shibayama Macromol. Chem. Phys. 199 1 (1998) 185. L Z Rogovina, V G Vasil'ev Macromol. Symp. 106 299 (1996) 186. A K Lele, R A Mashelkar J. Non-Newtonian Fluid Mech. 75 99 (1998) 187. V I Irzhak Rol' Fizicheskikh Uzlov v Protsessakh Relaksatsii Oligo- mernykh i Polimernykh Sistem (The Function of Physical Nodes in Processes of Relaxation of Oligomeric and Polymeric Systems) (Chernogolovka: Institute of Chemical Physics in Chernogolovka, Russian Academy of Sciences, 1997 188. S E Varyukhin, V I Irzhak Dokl. Akad. Nauk 369 (6) (1999) b 189. R F Fox Energy and Evolution of Life (New York:WH Freeman, 190. E Ben-Naim, G S Grest, T A Witten, A R C Baljon Phys. Rev. E 1988) 53 1816 (1996) 191. K L Ngai, D J Plazek J. Polym. Sci., Polym. Phys. Ed. 23 2159 (1985) 192. M F Herman Macromolecules 25 4925 (1992) 193. M F Herman Macromolecules 25 4931 (1992) 194. M F Herman, B V Panajotova Comput. Theor. Polym. Sci. 7 101 (1997) 195. B V Panajotova, M F Herman J. Chem. Phys. 108 5122 (1998) 196. M F Herman, B V Panajotova J. Chem. Phys. 110 8792 (1999) 197. V N Pokrovskii Usp. Fiz. Nauk 162 (5) 87 (1992) c 198. V N Pokrovskii, Y A Altukhov, G V Pyshnograi J. Non-Newtonian Fluid Mech. 76 153 (1998) 199. K S Schweizer J. Chem. Phys. 91 5802 (1989) 200. K S Schweizer, G Szamel J. Chem. Phys. 103 1934 (1995) 201. H Tang, K S Schweizer J. Chem. Phys. 105 779 (1996) 202. K S Schweizer,M Fucks, G Szamel, M Guenza, H Tang Macromol. Theory Simul. 6 1037 (1997) 203. M Fucks, K S Schweizer J. Chem. Phys. 106 347 (1997) 204. V G Rostiashvili,M Rehkopf, T As J. Phys. Chem. 110 639 (1999) 205. D A Vega, M A Villar, J L Alessandrini, E M Valles J. Polym. Sci., Polym. Phys. Ed. 37 1121 (1999) 206. T Nicolai, F Prochazka, D Durand Phys. Rev. Lett. 82 863 (1999) 207. T P Kulagina, Doctoral Thesis in Physicomathematical Sciences, Institute of Chemical Physics in Chernogolovka, Russian Academy of Sciences, Chernogolovka, 1995 208. R Walkenhorst, J C Selser, G Piet J. Chem. Phys. 109 11043 (1998) 209. M E Solov'ev, A B Raukhvarger, L I Makhonina, G V Korolev, V I Irzhak Vysokomol. Soedin., Ser. A 31 485 (1989) a 210. M E Solov'ev, A B Raukhvarger, A R Basaev, A N Privalov, V I Irzhak, G V Korolev, L I Makhonina Vysokomol. Soedin., Ser. A 34 127 (1992) a 211. M E Solov'ev, Doctoral Thesis in Physicomathematical Sciences, N N Semenov Institute of Chemical Physics, Russian Academy of Sciences, Moscow, 1994 212. R H W Wientjes, R J J Jongschaap,M H G Duits, J Mellema J. Rheol. 43 375 (1999) 213. S M Chitanvis Phys. Rev. E 58 3469 (1998) 214. A Eckstein, J Suhm, C Friedrich, R D Maier, J Sassmannshausen, M Bochmann, R Mulhaupt Macromolecules 31 1335 (1998) 215. Yu A Ol'khov, S M Baturin, V I Irzhak Vysokomol. Soedin., Ser. A 38 849 (1996) a

ISSN:0036-021X    

出版商:RSC    

年代:2000

数据来源: RSC