Meetings and Conferences

Conference on the Liquid State of Matter

On May 28–30, 1953, a conference on the liquid state of matter was held in Kiev, convened by the Academy of Sciences of the Ukrainian SSR and the T. G. Shevchenko Kiev State University.

The convocation of a broad conference on the liquid state of matter was urgently dictated by the present state of the problem of the molecular structure of liquids. It is known that, at present, in the field of the theory of the liquid state there is a considerable lag in comparison with the situation in the theory of the crystalline state and the kinetic theory of gases. Up to the present time no molecular theory of the viscosity of liquids has been created that would be in satisfactory agreement with experiment. This state of affairs in the theory of the structure of liquids has arisen as a result of the specific difficulties standing in the way of the development of this theory. Whereas in the field of the crystalline state one can model the structure by the “Einstein ideal crystal,” and in the field of the kinetic theory of gases begin the search for a general method with an ideal gas, in the field of the intermediate aggregate state—the liquid—such abstraction is impossible. A liquid has to be studied as it is. In recent years it has become evident that a new approach to the problem is necessary. This new approach has emerged in the study of the molecular structure of a liquid, in the study of the interrelation of physical properties both among themselves and with the molecular structure, and in the study of this relation within certain classes of liquids, followed by generalization within the framework of a general theory of the liquid state. In the postwar years extensive investigations were carried out in Moscow, Leningrad, Kiev, and Kharkov—by various experimental methods—and a genuine need arose to discuss the results obtained and to exchange views on the general direction and plans for further research in the theory of the liquid state.

At the conference devoted to this question, reports were heard and discussed on X-ray and optical work, studies of viscosity and the critical parameters of liquids in connection with their molecular structure, and also work devoted to the relation between the structure of a liquid and its electrical and surface properties.

In the report by V. I. Danilov, “Scattering of X-rays in Liquids and the Structure of Liquids,” the most important achievements in the field of X-ray studies of liquids were presented.

The speaker noted that, despite the comparatively limited possibilities of X-ray structural analysis of liquids in comparison with crystals, with the aid of X-rays important questions of the structure of liquids have been solved, and are at present being successfully solved.

As a result of X-ray studies, a new approach arose to the study of liquids, based on the conception of the liquid state as the state of molecules ordered in a special type of thermal motion, different from that characteristic of gases. The speaker notes the great contribution to the development of the physics of the liquid state made by Soviet scientists, and especially by Ya. I. Frenkel, in whose works modern ideas about thermal motion in liquids, based on X-ray data, were developed.

In studying the scattering of X-rays in liquids it often proves possible not only to establish the presence of a definite order in the arrangement of molecules, but also to draw conclusions about the forces of interaction between molecules, by analyzing the data—obtained radiographically—on the distances between “neighboring” atoms of the liquid, in a manner analogous to that used with crystals.

The latter became possible after the method of integral analysis had been developed and it became possible to determine the statistics of interatomic distances in liquids. In the speaker’s opinion, when studying the scattering of X-rays in liquids one should not neglect the method based on the qualitative comparison of the diffraction pattern of a substance in the liquid and solid states, by comparing “periods of identity” calculated from the Bragg–Wulff equation. The speaker considers it expedient to use both of the above-mentioned methods as mutually complementary.

Consideration of experimental data on the scattering of X-rays in liquid metals and inert gases makes it possible to draw a number of general conclusions about the molecular structure of simple liquids.

The structure of monatomic liquids is characterized by the presence of a definite short-range order for the given liquid, which can be described quantitatively by a distribution function determined radiographically.

Important parameters characterizing short-range order in a liquid are the most probable distance between nearest atoms and the coordination number.

Comparison of the short-range order of monatomic substances in the solid and liquid states makes it possible to conclude that closely packed crystals, after melting, form melts with the same close packing of molecules. Substances possessing a “loose” packing in the solid state sometimes, upon melting, reveal a violation of the packing type with a tendency toward an increase in the coordination number, that is, toward a transition to a type of denser packing. Often, however, the type of packing is preserved even in the case of “loose” substances.

The type of molecular packing in the liquid state is determined to a great extent by the forces of intermolecular interaction.

In the case of molecular liquids, information can be obtained both about the short-range order in the arrangement of molecules and about the structure of the molecules. Considerable success has been achieved in the study of molecular liquids by means of the method of integral analysis. On intensity curves and atomic-distribution curves, features of molecular packing determined by the character of intermolecular bonds are recorded. In many cases, determining the structure of a liquid molecule from distribution curves is carried out with an accuracy not inferior to that of the corresponding measurements on gaseous molecules.

Thus there appears the possibility of using X-ray structural analysis to establish the atomic, molecular, or ionic structure of liquids.

It should be noted that, despite the considerable number of works devoted to the radiographic study of binary liquid systems,

conclusions about the structure of the latter are in most cases qualitative in character.

In conclusion, the speaker dwelt on defining the concept of the structure of a liquid. He pointed to the advisability of carrying out coordinated investigations of the structure of liquids by various methods.

A. F. Skryshevskii in his report on the x-ray study of certain solutions criticized previous x-ray investigations of solutions. Earlier x-ray studies of solutions had been limited to the application of a method based only on a qualitative evaluation of the observed diffraction pattern.

The work presented by the speaker was based for the first time on the assumption that, for binary solutions, the method of integral analysis of intensity curves may be applicable if the concept of an “effective molecule” is introduced. Benzene solutions of ortho- and paradichlorobenzene and aqueous solutions of KOH, NaOH, LiOH, LiCl, and H₂SO₄ were investigated. For these systems radial distribution curves were determined; analysis of these curves made it possible to obtain quantitative data on the structure of the solutions studied.

In the case of benzene solutions, from the distribution curves it proved possible to determine interatomic distances in the dichlorobenzene molecule with an accuracy not inferior to that of the corresponding measurements made on pure components.

On the basis of analysis of the distribution curves for aqueous electrolyte solutions, coordination numbers of ions, the radii of the first coordination sphere, the structure of the ion SO₄⁻⁻, and also data characterizing the influence of various ions on the structure of water were determined. It was found that in solutions the most probable distances between the ions K⁺, Na⁺, Li⁺, OH⁻, Cl⁻ and water molecules in the first coordination sphere, as well as the values of the coordination numbers, are close to those in the corresponding crystalline hydrates; the ion SO₄⁻⁻ in solution has a tetrahedral structure with an S—O distance equal to 1.5 Å.

The results of the investigation of the structure of binary solutions confirm the correctness of the assumption made concerning the possibility of applying the method of integral analysis of intensity curves to the study of the structure of binary liquid systems. Thus, the possibilities of x-ray investigations are substantially broadened.

The reports by V. I. Danilov and A. F. Skryshevskii aroused a lively discussion.

E. A. Porai-Koshits criticized the theoretical foundations of the method of integral analysis of intensity curves and the situation in the field of liquid research that arose as a result of excessive enthusiasm for this method and led to the fact that, beginning in 1942, interest in x-ray investigations of liquids and glasses generally ceased. E. A. Porai-Koshits believes that in studying liquids and glasses it is much better to use the method of comparing intensity curves.

B. V. Deryagin, E. G. Shvidkovskii, O. Ya. Samoilov, and others noted the great importance of x-ray investigations for solving the problem of the structure of liquids. In particular, in the remarks of O. Ya. Samoilov it was noted that x-ray investigations of aqueous electrolyte solutions are important both for the development of the theory of electrolytes and for solving a number of practical problems. The work reported by A. F. Skryshevskii is the first work on the x-ray study of aqueous electrolyte solutions in which radial distribution curves were obtained. O. Ya. Samoilov indicated that E. A. Porai-Koshits assesses the method of integral analysis too skeptically. It is known that this method has provided much that is valuable for elucidating the nature of the distribution of atoms and molecules in a liquid.

A. Z. Golik indicated that, in order to characterize the molecular structure of a liquid, it is necessary to take into account not only the geometry of the arrangement of atoms and molecules, but also the nature of the intermolecular interaction. The immediate task is reduced to a many-sided investigation of the structure and physical properties of liquids.

The point of view expressed by E. A. Porai-Koshits regarding the method of integral analysis also elicited objections from A. S. Lashko and V. I. Danilov.

The report by I. V. Radchenko was devoted to the model study of liquids. The speaker stated in detail the view, available in the literature, on the use of model experiments for studying various physical phenomena and features of molecular motion in liquids and gases, and noted that in earlier experiments on the model study of particle distribution in a liquid, models consisting of particles whose interaction with one another was limited to repulsion upon contact were considered. In such models the effect of attractive forces, naturally, could not be studied.

I. V. Radchenko and F. K. Shestakovskii carried out new models of liquids, on which the influence of central and dipole forces on short-range order in a liquid was clearly shown. It was established that dipole and central forces exert different influence on the mutual arrangement of particles and, consequently, on the diffraction pattern arising when light is scattered by these models. I. V. Radchenko and F. K. Shestakovskii also constructed a model of a liquid consisting of different particles, on which it is possible to trace the influence of interaction forces on the mutual arrangement of particles in such a liquid.

The speaker set forth in detail the methodology of the model investigations of liquids used by the authors.

In the discussion of the report by I. V. Radchenko, B. V. Deryagin, E. G. Shvidkovskii, A. Z. Golik, and others took part.

I. V. Radchenko also delivered a communication devoted to the investigation of the scattering of X rays in supercooled diphenyl ether.

The results of studies of the viscosity, density, and other properties of a substance in the liquid state in connection with its molecular structure were reported by A. Z. Golik, S. D. Ravikovich, A. V. Orishchenko, V. P. Solomko, and N. A. Ryndin.

In two reports—“Molecular Structure and Physical Properties of Nonelectrolyte Solutions” and “Structure and Viscosity of Liquid Metals and Alloys”—A. Z. Golik showed that, in order to characterize the structure of a substance in the liquid state, it is insufficient to know the coordination number and the radius of the first coordination sphere; it is necessary to take into account the character of the intermolecular interaction. The speaker showed that intermolecular interaction is closely connected with the critical parameters of the given substance, and on this basis used the critical temperature as an additional parameter (besides the coordination number and the radius of the coordination sphere) characterizing the structure of a liquid.

A. Z. Golik proposed a simple classification of liquids that takes into account the structure of liquids, and showed that among liquids possessing the same structure there are observed simple regularities characterizing the temperature dependence of physical properties and their relationship to one another. He gave a simple physical interpretation of the observed regularities and showed how, on the basis of these physical conceptions, the existence of liquids possessing the same viscosity from the melting point to the critical point had been predicted.

A. Z. Golik further showed that the coordination number in the liquid state changes with temperature, and, in particular, clarified how it ...

varies with temperature for liquids possessing the same structure.

The reports presented numerous data on the temperature and concentration dependence of the viscosity of organic liquids, liquid metals, and solutions.

N. A. Ryndich reported on the results of a study of the viscosity of metallic amalgams. It turned out that, in the case of metallic liquid solutions as well, an increase in viscosity is possible over a wide temperature interval for metallic liquids and solutions. The phenomenon of “negative viscosity” in lead amalgams, discovered by N. A. Ryndich, deserves attention. It also turned out that, at concentrations corresponding to the maximum viscosity on the isotherm \(\nu(c)\), there is a distinct maximum in the activation energy of viscous flow and a minimum of the pre-exponential factor in the formula for the temperature dependence of viscosity.

S. D. Ravikovich reported on the results of an investigation of the influence of the size of the molecular chain and the intensity of intermolecular interaction on the value of the coefficient of viscosity. It was shown that the dependence of viscosity on the size of the molecular chain is linear in character and changes very little with temperature, whereas the dependence of viscosity on the energy of intermolecular interaction is nonlinear and changes sharply with temperature; moreover, viscosity depends to a significantly greater extent on the energy of intermolecular interaction than on the size of the molecular chain. In this connection the structure of isoviscous liquids was analyzed, and an experimental substantiation was given for the concept of molecular structure introduced by A. Z. Golik, together with a method, in good agreement with experimental data, for calculating the effective length of the molecular chain of a solution. The author analyzed certain regularities in the concentration variation of the activation energy of viscous flow of solutions and showed that, considering the question in the plane indicated above, one can find and explain a number of regularities in the properties of the coefficients entering into the formula for the concentration dependence of the activation energy of viscous flow.

V. P. Solomko reported on the results of studying the molecular structure and physical properties of a specific group of solutions whose components chemically interact with one another. The viscosity, density, and critical temperatures were studied for solutions of methyl, ethyl, propyl, and butyl alcohols in formic, acetic, and caproic acids, as well as solutions of the indicated alcohols and acids in water. It turned out that the regularities observed in the indicated group of solutions are considerably more complex than in solutions of a physical type. The viscosity curves of solutions lie both between and outside the curves of the temperature dependence of the viscosity of the components, and as a consequence the viscosity isotherms have a complex form—with a maximum, a minimum, and sometimes with both. In many, especially aqueous, systems maxima were found on the curves of the concentration dependence of the activation energy of viscous flow and minima of the pre-exponential factor at approximately identical concentrations, which, however, do not correspond to the concentrations at which the corresponding maxima on the viscosity isotherms were found. Coincidence of the critical temperatures in these systems does not entail coincidence of the curves of the temperature dependence of viscosity. The author reached the conclusion that in systems with chemically interacting components there may be formation both of indefinite chemical compounds in a state of dissociation and of microregions with a predominant content of molecules of one and the same kind.

A. V. Orishchenko gave a report on the structure and properties of solutions of lithium chloride in primary normal alcohols. It was

it was found that the solubility of lithium chloride in alcohols has a maximum, and as a consequence of this the critical temperature of the solution rises only up to a certain concentration of LiCl (5% in ethanol), and then, within the limits of experimental error, remains constant. The viscosity and density of the indicated solutions were also studied, and the influence of electrostriction effects and of the sizes of ions on a number of physical properties was found; new regularities were also found for the concentration dependence of the activation energy of viscous flow and of the pre-exponential factor in the formula for the temperature dependence of viscosity.

After the reports by A. Z. Golik and his co-workers, an extensive discussion unfolded on the results of the research.

I. M. Rudenko considers it correct to seek general regularities for all liquids. A. G. Shvidkovskii noted the great importance of the work of A. Z. Golik and his co-workers and the interest in these works among specialists in the field of molecular physics. He dwelt on experimental questions of the viscosimetry of metals, pointing out possible errors due to insoluble impurities and corrections for kinetic energy.

B. V. Deryagin indicated that the path by which regularities are sought within groups of liquids possessing a similar molecular structure promises decisive success in the creation of a molecular theory of the liquid state. B. V. Deryagin considers the reported work a major step forward toward the creation of such a theory. He also dwelt on the influence of electrification on the viscosity of metallic liquids.

S. D. Ravikovich showed that the methods of viscosimetry used by A. Z. Golik and N. A. Ryndich are free from those errors whose possibility was indicated by A. G. Shvidkovskii.

O. Ya. Samoilov devoted his remarks to the change in the coordination number of a liquid in connection with D. I. Mendeleev’s periodic law. He then dwelt on questions connected with the hydration of ions and complex formation in a liquid, giving one of the possible variants of the interpretation of the structure of solutions whose components interact very strongly with one another.

V. M. Chulanovskii, V. I. Danilov, and D. S. Kamenetskaya devoted their remarks to the question of the influence of molecular size and of the intensity of intermolecular interaction on the value of the viscosity coefficient of liquids.

V. M. Chulanovskii believes that the sharp increase in viscosity in the series of polyatomic alcohols is explained by the emergence of spatial formations connected with one another by hydrogen bridges, whereas in monatomic alcohols the character of the formation of linked systems is different—linear.

V. I. Danilov dwelt on the question of separating the influence of the hydrogen bond and the van der Waals bond on the value of the viscosity of liquids, noting that at high temperatures the van der Waals bond may exert a greater influence than the hydrogen bond, and expressed the wish that the indicated investigations be coordinated with X-ray diffraction studies.

D. S. Kamenetskaya pointed to the possibility of establishing a connection between the activation energy and the state diagrams of binary systems, and analyzed the physical meaning of the coefficients entering into the formula for the concentration dependence of the activation energy of viscous flow.

A. P. Brynza spoke about investigations of thermodiffusion in binary liquid systems and showed that the results obtained agree with the conclusions made in the reports of A. Z. Golik and his co-workers.

S. S. Urazovskii dwelt on the importance of clarifying the question of the presence or absence of groupings of like atoms in determi—

MEETINGS AND CONFERENCES

in the salt composition of a liquid solution. In his opinion, an answer to the question posed could be provided by studies of the differential properties of liquid systems—such, for example, as differential heats of dissolution—and the advisability of planning and carrying out such work is beyond doubt.

B. V. Deryagin gave an extensive report on joint work with V. V. Karasev and Z. M. Zorin on the topic “On a special aggregate state of liquids in layers bordering the surface of a solid body.”

In the case of liquid crystals, a solid wall is capable of causing an ordered orientation of molecules, an ordering that extends deep into the liquid without a gradual transition to a disoriented, disordered arrangement of the molecular axes. The zone of oriented arrangement of the axes of molecules in liquid crystals extends over distances several orders of magnitude greater than the thickness of a monolayer. In ordinary liquids, the depth of propagation of the orienting influence of a solid wall is, for monomers, of the order of \(10^{-5}\) cm. The work of B. V. Deryagin and his co-workers is devoted to elucidating the law according to which the influence of the wall on the orientation and other properties of the liquid dies away with increasing distance from the wall. To solve this problem the authors developed two different methods of investigation, by means of which a number of important and interesting results were obtained.

B. V. Deryagin and his co-workers established that boundary films form special boundary phases, separated from the remaining volume of the liquid by an interface at which the properties of the substance undergo a sharp jump. The authors showed that the phase difference between the adsorbed layer and the bulk liquid phase is connected with differences in their molecular structure, thereby confirming the statement made by P. P. Lazarev concerning the anisotropic structure of the surface layers of a liquid. The reason for the appearance of boundary phases lies in the fact that the polymolecular adsorbed layers have a special structure, different from the structure of an ordinary liquid; moreover, the role of the substrate on which the boundary phase is formed is reduced to the orientation of the molecules of the boundary phase.

Summing up the results of the investigations carried out, B. V. Deryagin concludes that one may speak of a special phase state in a layer of thickness \(10^{-6}\) cm, corresponding to which, at a certain distance from the solid wall, the properties of the liquid change by a jump corresponding to a first-order phase transition.

B. V. Deryagin’s report gave rise to many questions and a lively discussion.

A. R. Regel gave a report on the connection between the electrical properties of liquids and their structure.

For solid bodies, the connection between electrical properties and structure is unquestionable if one proceeds from the electronic theory. With respect to liquids, however, this cannot be said. Proceeding from experimental material on the electrical properties of liquids, Academician A. F. Ioffe pointed to the absence of any fundamental difference in the electrical properties of solid and liquid bodies, which makes it possible to think that short-range order plays the determining role for electrical properties. The results of the work carried out by the speaker and his co-workers confirmed the correctness of the conclusions drawn by A. F. Ioffe.

Consideration of the character and features of short-range order in liquids makes it possible to carry out a general analysis of the tendencies in the change of bonds under the influence of a substance being introduced and heated by it in the liquid state, as well as of the changes in electrical properties that may occur in this case.

The speaker emphasized that, for analyzing the connection between the structure of liquids and their electrical properties, a decisive role is played by taking into account the periodicity

the Mendeleev law and the basic principles of physicochemical analysis established by N. S. Kurnakov, and demonstrated this with a number of convincing examples.

Considering experimental data on the change in the electrical properties of metals upon melting, and also analyzing a number of attempts to generalize them, A. R. Regel showed that taking into account only the initial structure in the solid state is insufficient even for a qualitative analysis of the effect of melting on electrical properties, and that for the analysis of the electrical properties of melts it is necessary to take into account the structural features of liquids.

The report presented very interesting data obtained in analyzing the relation between electrical properties and the structure of substances with sufficiently pronounced homopolar bonds—with spatial and linear systems of homopolar bonds.

The first group includes substances with a diamond-like lattice: Si, Ge, GaSb, InSb, HgSe, HgTe. As shown in the work of the speaker, carried out jointly with N. P. Mokrovskii and A. N. Blum, the change in the electrical conductivity of this group of substances upon melting is a characteristic example of a clear connection between electrical properties and structure both in the solid and in the liquid state. In the present case, electrical conductivity proves to be a property for which the melting temperature manifests itself as a quite definite type of singular point in Kurnakov diagrams.

Comparison of the temperature dependence of electrical conductivity and density, as well as data on structure (in those cases where such data are available) for the substances listed above, shows the existence of an undoubted correlation between them. It turns out, moreover, that substances with the same structure in the solid state may, upon melting, change their electrical properties and density (structure) in quite different ways. These changes have a regular character associated with Mendeleev’s periodic law.

The melting of Si, Ge, GaSb, and InSb leads to a sharp increase in electrical conductivity, which is evidently connected with a rearrangement of the short-range order from the “diamond structure” to denser packings characteristic of the metallic state. The melting of HgSe, on the contrary, leads to a sharp decrease in electrical conductivity and density. Here, probably, molecular liquid is formed, as may be confirmed to a certain degree by examples of related HgSe compounds.

The transition from HgSe to HgTe corresponds to a weakening of the chemical bond; this leads to a sharp change in the character of the temperature dependence of electrical conductivity, which is qualitatively quite understandable if one assumes substantial dissociation of HgTe upon melting and the destruction of the HgTe residue upon further heating of the liquid.

The connection between electrical conductivity and structure for substances with “linear homopolar bonds” was considered using the examples of Se, Te, and the Se—Te system, which forms a continuous series of solid solutions. In this case as well, analyzing the connection between changes in electrical conductivity, viscosity, and density and changes in structure, the speaker convincingly showed that, for a profound understanding of the electrical properties of a substance both in the solid and in the liquid state, a detailed analysis of its structural features and of the chemistry of the elements constituting the substance is necessary on the basis of Mendeleev’s periodic law, i.e., a detailed analysis of the statistics and character of the short-range order is necessary. On the other hand, the study of the electrical properties of a substance over a wide temperature interval, encompassing both the solid and the liquid state, can help in the proper analysis of the structure.

In the report by M. F. Vuks, “Some Data on the Investigation of the Structure of Liquids by the Light-Scattering Method,” results obtained ...

valuable in the study of Rayleigh scattering in liquids (pure liquids) and solutions.

The main attention in the work is devoted to the investigation of the influence of the temperature of the liquid on the integral intensity of anisotropic scattering of light.

The experiments showed that, for different groups of liquids, the intensity of anisotropic scattering depends on temperature in different ways: in some it increases with an increase in the temperature of the liquid, in others it decreases, and in still others it remains unchanged.

Application of the formula for the intensity of anisotropic scattering, obtained for gases, to the scattering of light by liquids reveals that in some cases the measured intensity is smaller than the calculated one, and in others larger. This deviation of the intensity of anisotropic scattering for liquids from the value calculated according to the “gas” theory is due, in the author’s opinion, to the presence in a liquid of the orientational interaction of molecules (short-range orientational order), characterized by the ratio \(\gamma_{\text{eff}}^2/\gamma^2\), where \(\gamma_{\text{eff}}^2\) is the effective optical anisotropy of molecules in the liquid, and \(\gamma^2\) is the true anisotropy of the molecules. For some liquids \(\gamma_{\text{eff}}^2 > \gamma^2\), for others \(\gamma_{\text{eff}}^2 < \gamma^2\). Upon heating the liquid and disrupting the short-range orientational order, the effective anisotropy should approach the true anisotropy of the molecules,

\[ \text{and } \frac{\gamma_{\text{eff}}^2}{\gamma^2} \text{—to unity.} \]

These considerations are confirmed by the investigation of light scattering by benzene and nitrobenzene.

The different behavior of anisotropic scattering under a change in temperature is explained by the author on the basis of the relation between the structure of the liquid and the structure of the corresponding crystal. Substances crystallizing in the cubic system (cyclohexane) or noncubic, but with weak anisotropy (water, benzene), give in the liquid state substantially weaker anisotropic scattering than with a disordered orientation of the particles \((\gamma_{\text{eff}}^2 \ll \gamma^2)\). Consequently, upon heating such a liquid, the intensity of anisotropic scattering should increase.

Substances possessing, in the crystalline state, large optical anisotropy give in the liquid state intense anisotropic scattering, considerably greater than with a disordered orientation of the particles \((\gamma_{\text{eff}}^2 > \gamma^2)\). Upon heating such a liquid, the intensity of anisotropic scattering should decrease. This is observed for nitrobenzene.

The stated propositions are also confirmed by investigations of anisotropic light scattering by dilute solutions (solutions of benzene, nitrobenzene, carbon disulfide in carbon tetrachloride, cyclohexane, heptane, alcohol, ether, and acetone).

Separation of molecules by dissolution leads to the same result as heating the liquid—the approach of the intensity of anisotropic scattering to the value of the intensity for gases.

Thus, the optical anisotropy of molecules does not change upon transition from vapor to liquid. Each change in it should be attributed to the orientational interaction of molecules and to short-range orientational order in the liquid.

The investigation of light scattering in solutions also makes it possible to calculate the optical anisotropy of molecules.

In the report of M. N. Padenkov, on work carried out jointly with M. F. Vuks, the results are presented of an experimental investigation of anisotropic light scattering for several groups of liquids. These studies with still greater persuasiveness confirm the proposed

M. F. Vuks set forth the idea of a connection between the intensity of anisotropic scattering of light in liquids and the optical anisotropy of the crystal.

Studies of anisotropic scattering of light were carried out primarily with liquids crystallizing in the cubic system (cyclohexane, cyclohexanol, camphene), and also with water, which has an isotropic tetrahedral coordination of molecules.

In parallel with the study of anisotropic scattering, the authors also carried out studies of isotropic scattering. In water an unusual dependence of the intensity of isotropic scattering on temperature was observed. As is known, the intensity of isotropic scattering increases with increasing temperature in accordance with the Smoluchowski–Einstein formula. In water, however, as the experiments showed, the intensity of isotropic scattering remains practically unchanged when heated from 20 to 108° C.

This is explained by the fact that the compressibility, on which the scattering intensity depends, has an anomalous dependence on temperature in water.

The authors also carried out investigations of the anisotropic and isotropic scattering of light for monatomic alcohols, paraffins, and ketones. In all cases the intensity of anisotropic scattering increases with rising temperature, fastest of all in alcohols, then in paraffins, and slowest of all in ketones.

In these investigations, as the authors note, the method of investigation proposed by A. Z. Golik proved extremely expedient; in it liquids are grouped according to the similarity of their molecular structure. For liquids with a similar molecular structure, a similar picture is obtained for the temperature dependence of the intensity of scattered light.

For benzene derivatives—toluene, ethylbenzene, aniline, xylene, diphenyl ether, orthodichlorobenzene, metadichlorobenzene, metabromobenzene—the magnitude of the intensity of anisotropic scattering and the temperature behavior are different for different liquids. The different temperature behavior is explained by the authors by the different short-range orientational order in these liquids.

In order to verify the influence of orientational interaction in liquids on light scattering, studies were carried out with binary liquid solutions (phenol–water, solutions of chlorobenzene in bromobenzene, chlorobenzene in iodobenzene, nitrobenzene in hexane, oleic and acetic acids in water), which fully confirmed the above considerations.

Following the report of M. F. Vuks and the communication of M. F. Vuks and M. N. Dadenkova, a lively discussion developed.

A. Z. Golik drew attention to the fact that the effects observed by the speakers depend, and moreover in a very complex way, on the molecular structure of liquids, and therefore the further development of the indicated method for studying the structure of liquids in cooperation with other investigations (compressibility, density, viscosity, electro-optical phenomena, etc.) is extremely useful. On the other hand, without taking into account the molecular structure of liquids it is impossible to understand the nature of the processes occurring in it, in particular optical effects.

A. Z. Golik noted that the examples of optical investigations under discussion once again confirm that the grouping of objects of investigation according to the criterion of similar molecular structure is expedient. The regularities reported by the speakers for isotropic and anisotropic scattering, which apply to a group of liquids with similar molecular packing and a related type of bonding, have a very clear character.

On the basis of investigation of the scattering effect in groups of liquids of the same type, while at the same time using data obtained by other methods, one can penetrate more deeply into the nature of this interesting effect.

S. S. Urazovskii noted that the twofold form of the temperature dependence of anisotropic scattering for various liquids is correctly interpreted by the speakers, and expressed the wish to coordinate work on light scattering with investigations of changes in the symmetry of molecules during the transition from one aggregate state to another.

G. A. Porkosh and Kokin also noted the advisability of applying the methodology under discussion to the study of the structure of liquids, and expressed the wish not to limit oneself to investigations of scattering at an angle of \(90^\circ\) to the incident ray, but to study the indicatrix of scattering, which will make it possible to estimate the sizes of inhomogeneities in liquid solutions.

The report by V. M. Chulanovskii presented a brief review of work on the vibrational spectroscopy of liquids. Molecular interactions manifest themselves in spectra distinctly and in various ways, and therefore spectral methods can be used to study the liquid state. As early as 1949 the author drew attention to the possibility of studying molecular interactions through their manifestations in the region of ordinary vibrational absorption spectra, i.e., in the comparatively easily accessible infrared region.

Intermolecular interactions affect first of all the contour of the absorption band. Since combination bands are readily observed in the vibrational contour of an absorption band, one may expect the appearance (symmetrically with respect to the principal intramolecular bands) of combination bands whose wave numbers are equal to the sum or difference of the wave numbers of intramolecular and intermolecular vibrations.

The speaker believes that a method for verifying whether bands of small frequencies belong to intermolecular vibrations can be the study of those changes in the spectrum which occur upon dissolving the pure substance or upon replacing one solvent by another. The different environment of the absorbing molecule should be reflected in the strength of the intermolecular bonds and, consequently, in the position and form of the corresponding spectral formations. This influence will be stronger and different than in the case of bands of intramolecular origin.

The relative intensity of bands of small frequencies, situated near various intramolecular bands, should reflect the degree and character of the coupling of both types of vibrations. The expected phenomena should be especially pronounced in the spectra of those groups which include hydrogen.

As objects of investigation, substances containing one group \( \mathrm{C-H} \), \( \mathrm{N-H} \), or \( \mathrm{O-H} \) were chosen.

The study of the valence vibration of the \( \mathrm{C-H} \) group was carried out on chloroform and bromoform. In both cases one rather narrow band near \(3020\ \mathrm{cm}^{-1}\) is characteristic. Around this principal band symmetrically arranged weak bands are observed. It is shown that they owe their origin to combinations of intermolecular vibrations with intramolecular ones. The probability of formation of combination bands from intra- and intermolecular vibrations is different for different intramolecular vibrations. The speaker gave a formula describing the distribution of the intensity in a simple band.

The results of investigations with chloroform and bromoform make it possible to establish that in these cases the intermolecular bond is quantized. A comparison of the low-frequency spectra of liquid and microcrystalline bromoform reveals a noticeable difference between them, due, in the speaker’s opinion, to the different character of the bond in the two cases.

The report also presented studies of secondary amines having one \( \mathrm{N-H} \) group. They are intermediate in the character of their bond between the previously considered substances and substances contain-

MEETINGS AND CONFERENCES

...the hydroxyl group. Here two bands of valence vibration of the N—H group are observed. The broader band, 3315 cm\(^{-1}\), belongs to complexes bound by a hydrogen bond. The second band, 3682 cm\(^{-1}\), is determined by another type of weaker molecular bond. Both bonds are quantized. Around each of these bands there are bands of low frequencies.

When secondary amines are dissolved in various solvents, the band 3682 cm\(^{-1}\) is displaced relative to its position in the pure amine, changes its half-width and the value of the wave number of the intermolecular vibration. These changes occur in parallel and can be used to characterize the intermolecular bond.

The change in the low-frequency spectrum when the concentration of an amine dissolved in a neutral solvent is changed allows the speaker to conclude that in solutions there is not a uniform distribution of the dissolved substance among the solvent molecules, but the presence of separate weakly bound systems. The character of the complexes in the solution depends not only on the nature of the solvent, but also on its concentration.

Further, it was noted in the report that in the formation of a hydrogen bond two cases should be distinguished: the first—when the hydroxyl is contained in both interacting molecules, and the second—when only one molecule contains a hydroxyl group, while the other contains only an atom of oxygen, chlorine, nitrogen, or sulfur. Between the two indicated cases there is an essential quantitative and qualitative difference. On the basis of experiments with alcohols and water it was shown that only in the first of the cases indicated above can one speak of a bond that is exceptional in character and strength, which the author proposes to call a hydrogen bond. In connection with this, the generally accepted formula O—H...O is unsuitable for describing the hydrogen bond and must be replaced by another.

The work shows a close connection between the formation of a hydrogen bridge and the phenomenon of exchange of hydrogen atoms.

Using examples of the mixing of water with acetone, diethyl ether, and carbon tetrachloride, it is shown that the complexes of a substance containing hydroxyl are preserved undestroyed in the environment of the solvent.

Those who took part in the discussion of the report by V. M. Chulanovskii (E. A. Porai-Koshits, I. V. Radchenko, S. S. Urazovskii) noted that a whole series of other investigations—such as, for example, X-ray studies of acetone–water mixtures, glasses, mica, bismuth compounds, and others, as well as studies of the dielectric permittivity of liquid mixtures and studies of dissolved polymorphic modifications of monochloroacetic and glycolic acids in various solvents—indicate the preservation, when liquids are mixed, of small groups consisting of molecules of the separate components of the mixture.

The speakers cited various methods for determining the sizes of these regions of inhomogeneity and pointed out that the study of the microheterogeneous structure of liquids is of considerable interest.

I. G. Polotskii presented the results of a joint study with Z. L. Khodovyi of the rate of propagation of ultrasound and the adiabatic compressibility for several one-component and binary liquid systems. Binary liquid systems with different types of intermolecular interaction were studied: benzene—carbon tetrachloride, water—methyl alcohol, phenyl mustard oil—diethylamine. Isotherms of the ultrasound propagation velocity and of the coefficient of adiabatic compressibility were constructed for each of the systems, for temperatures of 20 and 40° C. For the system benzene—carbon tetrachloride the compressibility isotherms are smooth curves. For the system methyl alcohol—water the compressibility isotherms have a minimum. For the system phenyl mustard oil—diethylamine, in the region of concentrations corresponding to the formation...

for the formation of a chemical compound, the compressibility isotherms have a very sharp minimum. With increasing temperature the minima become somewhat smoothed. The work also compares data on the velocity of ultrasound and compressibility with data on the viscosity of the corresponding systems.

An investigation was also carried out of the temperature dependence of the velocity of propagation of ultrasound and of the coefficient of adiabatic compressibility for melts of salol, thymol, and orthochloronitrobenzene in the temperature interval 30–90° C. It was found that the temperature dependence of the velocity of propagation of ultrasound for these three substances has a linear character. Changes in the adiabatic compressibility of these same systems are represented graphically by smooth curves passing through the entire temperature interval studied. This continuity of the curves upon transition from the stable to the metastable state is an argument in favor of the fact that in this transition there is no abrupt change in the forces of intermolecular interaction, and consequently no disruption of the short-range order in the mutual arrangement of the molecules.

In closing the conference, A. G. Shvidkovskii noted that Russian science has glorious traditions in the field of molecular physics. The conference showed that these traditions are being successfully continued and multiplied. The conference demonstrated significant successes on the path toward creating a theory of the liquid state, the broad scope of experimental and theoretical work in this field, and there is every reason to suppose that modern physical science in the USSR will continue successfully and multiply the foremost traditions of Russian science in the complex and important problem of the liquid state of matter.

A. G. Shvidkovskii noted the businesslike character of the conference and pointed out that the principled and benevolent criticism with which the entire work of the conference was imbued contributed to a businesslike and free exchange of opinions.

V. M. Chulanovskii noted the good organization of the conference and, on behalf of the out-of-town participants, expressed gratitude to the Organizing Committee.

The conference showed that the general direction of the work reported at the conference is fruitful and promising. In the course of the discussion it was noted that the general opinion of the conference participants was that there is a fundamental possibility of constructing a quantitative theory of the liquid state, and a new approach to this problem was approved, connected with the study of the interrelation of the physical properties of a liquid with its molecular structure—an aspect that was constantly at the center of the conference’s attention. The conference participants emphasized the advisability of studying various physical properties of a liquid on the same objects of investigation.

The conference adopted a resolution approving the general plan and direction of research in the field of the liquid state.

A decision was made to prepare and hold the next All-Union Conference on the Liquid State, scheduled for 1954.

S. D. Ravikovich, G. P. Roshchina, A. F. Skryshevskii