M. S. Tsvet’s Chromatographic Method of Analysis

Z. V. Zhidkova

“Soviet science looks back on its past with pride. In this past it sees great achievements, the conquest of new scientific heights, the discovery of scientific truths, the creation of entire fields in science, and the beginning of the most important branches of technology.”

S. I. Vavilov*)

I. Introduction

Among physicochemical methods of analysis, the chromatographic adsorption method of analysis has recently been acquiring ever greater importance; it is based on the selective adsorption of one or several substances from a solution by one adsorbent or another. The advantage of this method over other physicochemical methods lies in the fact that it is applicable in a number of cases when other methods of fractionating a mixture prove powerless. With its aid, substances very close to one another in composition and properties can be separated from each other, even when they are present in very small amounts. The possibility of separating substances very close in composition and chemical properties constitutes the chief value of the chromatographic method. It should be especially noted that this method of analysis also requires less time and resources because of the comparative simplicity of the necessary technical equipment.

The founder of the chromatographic adsorption method of analysis was the Russian scientist—the botanist Mikhail Semyonovich Tsvet**),

) Vestnik Akademii nauk*, No. 5, p. 5 (1948).

) We give brief biographical data on M. S. Tsvet, taken by us from the article by Academician A. A. Richter and Professor T. A. Krasnoselskaya, published in the USSR Academic edition for 1946 (Selected Works of M. S. Tsvet)[^1].

Mikhail Semyonovich Tsvet was born on May 19, 1872, in Italy (in the city of Asti). His father, Semyon Nikolaevich Tsvet, was Russian; his mother, Maria Dorozza, was Italian. Mikhail Semyonovich Tsvet spent his childhood and youth in French Switzerland, where he graduated from the University of Geneva, and then began...

the first to develop this method for isolating pigments from green leaves. Before him, pigments were isolated from green leaves by means of complex chemical treatment, as a result of which the experimenter obtained products whose properties were far removed from those of the natural pigments in the green leaf. Thus, in the best course on plant physiology of that time, by V. I. Palladin, we read: for separating the component parts of chlorophyll “by Hansen’s method, the alcoholic extract of chlorophyll is mixed with caustic sodium and heated to boiling in a water bath for about three hours,” and so on.

From this there followed either incorrect or primitive conclusions about the composition of chlorophyll, which, in Darwin’s words, is “one of the most interesting substances on the earth’s surface.” In the same course by V. I. Palladin we read: “Chlorophyll is a nitrogenous body, insoluble in water, soluble in alcohol, ether, and oil. The ash of an alcoholic extract of chlorophyll contains iron.” The enormous difficulties in studying chlorophyll lie precisely in its easy alterability, which requires, when extracting it from plants, the use of exclusively neutral liquids.

The structure and composition of chlorophyll have been clarified only in recent times, after M. S. Tsvet developed the chromatographic adsorption method of analysis.

The method itself, which became the basis of chromatographic analysis, was developed by the author as a result of painstaking work. During—

research work in his laboratory of general botany. In 1897 Tsvet moved to Petersburg and took a position in the St. Petersburg biological laboratory, founded a year earlier by P. F. Lesgaft, and then in the academic laboratory of Academician A. S. Famintsyn. On September 21, 1901, at Kazan University, Tsvet defended his dissertation for the degree of Master of Botany, “The Physico-Chemical Structure of the Chlorophyll Grain.” Later Tsvet moved to Warsaw, where he obtained a position first as an extra-staff laboratory assistant, and then as assistant and privat-docent at Warsaw University. From 1907 he was professor of botany and agronomy at the Warsaw Veterinary Institute. After 1908 Tsvet was elected professor of botany and microbiology at the Warsaw Polytechnic Institute, and in 1909 he left his work at the university. In 1910 M. S. Tsvet defended his doctoral dissertation in botany at Warsaw University on the topic: “Chromophylls in the Plant and Animal World.” In the years when Tsvet lived in Warsaw, he traveled several times to Petersburg and Moscow to take part in congresses, where he delivered reports. In 1900, on the proposal of I. P. Borodin, M. S. Voronin, and D. I. Ivanovsky, Tsvet was elected a member of the St. Petersburg Society of Naturalists; in addition, he later was a member of the Warsaw Society of Naturalists. Tsvet was a good lecturer and treated teaching in higher educational institutions with love. He also attracted students to work in his laboratory. In 1914 the imperialist war broke out, and Tsvet, together with the Warsaw Polytechnic Institute, was evacuated from Warsaw to Nizhnii-Novgorod (now the city of Gorky). The difficult life during this period undermined M. S. Tsvet’s already frail health. He died on June 26, 1919.

It was widely known before Tsvet that the “green substance” (chlorophyllins) could not be completely extracted from plant leaves with ligroin. Many authors tried to explain this by the poor solubility of chlorophyllins in ligroin. Tsvet wrote on this subject: “Experimental investigation of the question led me to the following result. The insolubility of the greater part of the chlorophyll pigments from leaves in benzine and ligroin is due not to their insolubility in these liquids, but to the retarding action of the molecular forces of the substrate, i.e., to adsorption absorption”[^1].

And further he formulates with complete precision the foundations of the chromatographic method: “One should likewise expect that all kinds of powdery substances will exert an adsorptive action on chlorophyll pigments in ligroin solutions, and the hope arose that a systematic study of the question would throw some light on the essence of adsorption phenomena and would make it possible, on their basis, to develop a new method of physical separation of substances.”

M. S. Tsvet published his works in 1903 in the Proceedings of the Warsaw Society of Naturalists under the title “On a New Category of Adsorption Phenomena and on Their Application to Biochemical Analysis,” and in a number of other works of a later date[^1].

Almost all modern procedures of the chromatographic method of analysis had already been fully developed and described by Tsvet in those years, and this method is rightly called by his name. Even with all the efforts of a number of foreign scientists to pass over in silence the priority of the great Russian scientist in the creation of this method, they nevertheless are forced to acknowledge it. Thus L. Zechmeister[^45], in an article devoted to the history and methodology of chromatographic analysis, writes: “There is no doubt whatever that the true inventor of chromatography in all its most important features is Tsvet.”

However, this truly outstanding method did not receive broad support among the contemporaries of its author. To some extent the reason for this may have been the unfavorable review given by the famous German chemist Willstätter, who at that time had worked for many years on the investigation of chlorophyll. It would be wrong to say that the method was not appreciated at all—in 1911 the Russian Academy of Sciences awarded Tsvet, for his book Chlorophylls in the Plant and Animal World, the Grand Prize named after A. P. Akhmatov. But after this, “Professor Tsvet’s remarkable chromatographic method, which in importance is justly compared with spectral analysis, was long forgotten in Tsvet’s homeland”1.

The “seed” sown by the modest Russian scientist gave “shoots” only about 20 years later in the classical works of N. A. Shilov², L. K. Lepin³, M. M. Dubinin⁴, and others on the study of the dynamics of sorption of vapor-like substances.

The works of Kuhn, Winterstein, and Lederer⁵ on the analysis of carotenes confirmed the full power of this method.

Over the last decade the method has become widely used in organic and inorganic chemistry, in biochemistry, medicine, and other sciences and branches of industry. Simultaneously with the introduction of this method for the analysis or isolation of various products, the theory of chromatographic analysis and various methods of carrying it out have been developed. This includes the works of M. M. Dubinin⁶, A. A. Zhukhovitskii⁷, E. N. Gapon⁸,⁹, B. N. Nikol’skii¹⁰, M. A. Konstantinova-Shlezinger¹¹, O. M. Todes¹², M. V. Radushkevich¹³, A. N. Kharin¹⁴, and others.

The large amount of material accumulated on various questions of chromatography led to the necessity of convening a conference, which took place in November 1950. At the conference, new directions in chromatography were noted (ion-exchange, precipitation, etc.), together with their broad application in various areas of the national economy, as well as major successes in the synthesis of absorbents.

II. PHYSICOCHEMICAL FOUNDATIONS AND METHODOLOGY OF CHROMATOGRAPHIC ANALYSIS

The classical method of chromatographic adsorption analysis, developed by M. S. Tsvet, is based on the selective absorption of substances from a solution by some adsorbent, depending on the ratio of the values of the adsorption coefficients of the given substances on the given adsorbent. Let us imagine that a solution of a mixture of the substances under study, whose adsorption coefficients with respect to the given adsorbent are different, is filtered through a layer of adsorbent. The individual components of the solution will be absorbed in the adsorbent layer successively, forming zones (or bands) of different coloration, if the given substances have a characteristic coloration in the visible region of the spectrum (see insert, Fig. 1, p. 392). Initially, in the very upper layer, the component of the solution with the greatest adsorption affinity for the given adsorbent is adsorbed; in the very lowest layer, the component with the smallest adsorption affinity is adsorbed. Between them, from top to bottom, all the other components are arranged in the order of decreasing adsorption affinity for the adsorbent.

The order of absorption of the given components from the given solution on the given adsorbent is entirely definite and does not depend on how many times and in what manner the filtration is repeated. Let us take, for example, a solution of three previously known components (substances) $a$, $b$, $c$ and pass this solution through an adsorbent. Let

if the adsorptive affinity of the components of the solution decreases from \(a\) to \(b\) and to \(c\), then the order of arrangement of the zones \(A, B, C\), in which the corresponding components are absorbed, will be as shown in Fig. 1. During subsequent filtration of a solution consisting of only component \(a\), the zone corresponding to this component (zone \(A\)) will expand somewhat, which is connected with an increase in the adsorbed amount of the added component, while zones \(B\) and \(C\), being displaced, will move downward by the corresponding amount. When a solution consisting of only component \(b\) is added, zone \(B\), corresponding to this component, expands; zone \(A\) remains in its former place; zone \(C\) shifts by the amount of the increase in zone \(B\), etc. In short, as M. S. Tsvet said, “Each member of the adsorption series, possessing a greater adsorptive affinity than the next, displaces it from its compound and in turn is displaced by the preceding one.”

Practically, the zones corresponding to the individual components of the solution passed through the adsorbent overlap in most cases. Therefore it is sometimes difficult to determine where the absorption zone of one component ends and the zone of another begins. In order to obtain a clear chromatographic picture, some pure solvent, called in chromatography a “developer,” is additionally filtered through the adsorbent. Each zone, under the action of the stream of developer, descends; the displacement of each zone is the greater, the smaller the adsorptive affinity of the component (corresponding to the given zone) with respect to the given adsorbent. It is possible, however, that not all zones move under the action of the stream of the given “developer.” In this case it is necessary to select other developer-solvents until the aim is achieved. As a result of “development,” the individual zones become separated from one another and, in most cases, white intervals appear between them (see Fig. 1). In some cases it is not possible to select a sufficiently good “developer,” and the zones in the chromatogram remain closely adjacent to one another, without separating white bands.

Fig. 1

Replacing the “developer” in obtaining a chromatogram sometimes leads to a reversal of the relative arrangement of the zones in it. Fig. 2 gives an example of such a reversal of zones in the adsorption of a mixture of two substances (meta-nitrophenol and para-nitromonoethyl-aniline) on silica gel when developing the chromatogram with a ligroin solution in the first case—benzene, and in the second—ether.

An analogous phenomenon may also be observed when the adsorbent and solvent are changed. One of the examples of this was described by Lero-

with benzene,^15 who studied the arrangement of the zones of two substances: lycopene \(C_{40}H_{56}\) and cryptoxanthin \(C_{40}H_{56}O\), during their adsorption on various adsorbents; he found that upon adsorption on aluminum oxide and on calcium carbonate the zones are arranged in the sequence: cryptoxanthin, lycopene; whereas upon adsorption on calcium oxide hydrate the reverse sequence of zones is observed: lycopene, cryptoxanthin.

1) benzene + ligroin

2) ether + ligroin

Fig. 2. Reversal of the relative arrangement of zones upon changing the solvent.

The apparatus needed to carry out chromatographic adsorption analysis is very simple. It consists of a glass tube in which, at some distance from its lower end, a porous partition is placed (porous glass or a metal mesh with a cotton plug). The tube is filled with a powdered adsorbent (such a glass tube filled with a powdered adsorbent is called an adsorption column) and a solution of the substances under study is filtered through the layer of adsorbent. In order that absorption (adsorption) take place uniformly along the entire length of the tube, uniform packing of the adsorbent is necessary. The most favorable method for uniformly filling the tube is one based on suspending the adsorbent in a neutral solution. In this case filling of the tube is usually carried out under slight pressure, for which purpose the adsorption column is connected by means of a stopper to a Büchner flask (Fig. 3).*

* When filling a tube with a dry adsorbent, the latter is poured in small portions and tamped with the aid of a metal or glass disk, or a cork stopper, the diameter of which is only slightly smaller than the diameter of the tube. But the process of such filling is very laborious, and, most importantly, rarely gives good results; in most cases the compaction of the adsorbent occurs nonuniformly over the height of the column, which leads to differing adsorptive capacity of the adsorbent in different parts of the column and to the formation of so-called “false zones.”

** During filling of the tube and during preparation of the chromatogram, if the work is being carried out under pressure, care must be taken, in order to avoid drying and the consequent cracking of the column, that solvent or solution always be present on the surface of the adsorbent. Otherwise, the stream of solution or solvent will rush into the cracks that have formed; upon subsequent washing, the process of washing or adsorption will be disturbed, and the chromatographic picture spoiled.

The tubes used may differ both in size and in shape, depending on the amount of adsorbent required for the experiment; the latter, in turn, depends on the problem at hand. For microchemical work, tubes with a diameter of 1–2 mm are used; for spectroscopic work, of the order of 10 mm; for preparative work, tubes of still larger size, holding several kilograms of adsorbent (Fig. 4).

Plant installations are designed for hundreds of kilograms and tons of adsorbent.

When selecting a tube, one takes into account, of course, also the adsorption capacity of the adsorbent with respect to the substance under investigation and the amount of the substance being investigated. Thus, if a substance is adsorbed poorly, it is desirable to have as high and wide a tube as possible; but tubes that are too wide are undesirable, since it is more difficult to obtain a good chromatogram in them because of the nonuniform distribution of the solution over the surface at the moment it is poured into the tube. Sometimes, when

Fig. 3.

a  b

Adsorption tube for large quantities

Fig. 4. Types of adsorption tubes.

there is a large amount of the substance under investigation, in order to avoid too great an increase in the dimensions of the column, the work is carried out on several tubes simultaneously. If it is necessary, after obtaining the chromatogram, to divide the column of adsorbent mechanically according to the zones obtained, it is extremely convenient to use a tube with a ground joint[^16], shown in Fig. 4, b. When the experiment is окон

...completed, the ground joint is opened and the column of adsorbent is pulled out of the tube together with the partition.

Of essential importance in chromatographic analysis is the choice of adsorbent. Any powdered substances that do not react with the solvent used and do not decompose the adsorbed substances may serve as adsorbents. M. S. Tswett tested about 100 different adsorbents, but only a few of them proved suitable. The following substances may serve as inorganic and mineral adsorbents: aluminum oxide \((\mathrm{Al}_2\mathrm{O}_3)\), calcium oxide \((\mathrm{CaO})\), calcium oxide hydrate \((\mathrm{Ca}(\mathrm{OH})_2)\)—slaked lime, calcium carbonate \((\mathrm{CaCO}_3)\), zinc oxide \((\mathrm{ZnO})\), calcium sulfate \((\mathrm{CaSO}_4)\), magnesium oxide \((\mathrm{MgO})\), franconite, silica gel, fuller’s earth, powdered talc. Among organic adsorbents known in practice are: powdered sugar, inulin, milk sugar, activated charcoal from coal, and bone charcoal.

The adsorption activity of one or another adsorbent with respect to a given substance may be different depending on the method of preparation of the given adsorbent and on the presence in it of adsorbed moisture. Indeed, depending on the above-mentioned conditions, the specific surface of the adsorbent changes, which is the decisive factor in physical adsorption. For example, different conditions for preparing an adsorbent may lead to the formation of different pore sizes in it, and this will lead to a change in its active surface, since the latter may be expressed by the function

\[ S=\int_{r_0}^{\infty} \varphi(r)\,dr, \]

where \(r_0\) is the smallest radius of pores passable for the molecules of the substance. It must be borne in mind that the specific surface accessible to molecules of one substance (for example, water) does not necessarily coincide with the surface accessible to larger molecules (for example, benzene). As regards the influence of adsorbent moisture, with removal of the adsorbed moisture, all other conditions being equal, the free active surface of the adsorbent increases, i.e., its adsorption activity with respect to the given substance increases.

The particle sizes of the adsorbent undoubtedly affect its adsorption capacity; therefore the adsorbent with which the tube is filled should not be too finely divided, since smaller particles may also prove more active. Some authors recommend using finely dispersed powders as adsorbents; for example, E. Lederer\(^{17}\) indicates the desirable sizes of the adsorbent particles used as from \(1.5\) to \(10\,\mu\). However, with such a particle size of the adsorbent, filtration is very greatly slowed down.

solution, and therefore the most favorable particle size lies within the range of 40–60 μ. The use of particles of considerably larger size leads to a blurred appearance of the chromatogram.

The choice of one adsorbent or another is ultimately decided by experiment, but in preliminary trials one may be guided by the following regularity: adsorbents of a basic character adsorb substances possessing acidic properties, and vice versa.

One of the most commonly used adsorbents in the adsorption of both organic and inorganic substances is aluminum oxide, whose favorable properties are determined not only by its amphoteric character, but also by its comparatively easy regeneration. The adsorptive capacity of aluminum oxide can be increased or decreased.

Ruggli and Jensen^18 describe a method for activating aluminum oxide by washing it three times with tap water. In this process the adsorbent takes up a small amount of lime, as a result of which its adsorptive activity is increased. Gellbron and Fipers^19 lowered the activity of aluminum oxide by washing it with methanol and subsequently drying it in air.

Another most commonly used adsorbent is silica gel. The adsorptive properties of silica gel are especially strongly affected by the method of preparation, as may be seen from the works of M. O. Kharmandaryan and S. A. Kapelevich^20, A. V. Kiselev^21, and others.

Solvents are chosen according to the criteria of best dissolution of the substances under investigation and least adsorption of them by the adsorbent. The most widespread solvents are organic substances: petroleum ether, acetone, carbon disulfide, benzene, gasoline, methanol, ethanol, chloroform, etc. In inorganic chemistry water is very widely used as a solvent. Sometimes mixtures of two or several solvents are employed.

The same solvents are used as “developers.” The choice of “developer” is made, in most cases, empirically.

As a result of carrying out all stages of chromatographic adsorption analysis (filtration of the solution through the column and its development), we obtain a finished chromatogram. In this connection, as Tsvet^1 describes: “Like light rays in a spectrum, the different components of a complex pigment arrange themselves according to a regular law one after another in the column of adsorbent and become accessible to qualitative and quantitative determination.”

The individual components of the chromatogram are either mechanically separated from one another, as was indicated earlier, or gradually

are transferred into the filtrate. After this they are analyzed by one or another method.

For a visual examination of the composition of the substance, one may use the rapid, so-called “circular” method. For this it is necessary to have two glass plates, in one of which a hole 4–6 mm in diameter has been made at the center. The size of the plates depends on the size of the chromatogram. The adsorbent is poured onto the plate without a hole and, by lightly rocking the plate in a horizontal plane, is distributed over it in an approximately uniform thin layer. Then the plate with the hole is placed on top; the latter, when lightly pressed with the fingers, is moved back and forth in the horizontal plane over the adsorbent. In this way the air present between the plates and the adsorbent is removed, and the uniformity of the distribution of the adsorbent between the plates is increased. Then, through the hole, with the aid of a pipette, the substance under investigation is introduced; with uniform compaction of the adsorbent it spreads at the same rate in all directions from the hole. Upon subsequent “development” we obtain a chromatogram in which the individual zones are arranged in the form of concentric circles, as shown in Fig. II (see insert).

III. ANALYSIS OF INDIVIDUAL ZONES OF THE CHROMATOGRAM CORRESPONDING TO INDIVIDUAL COMPONENTS OF THE MIXTURE UNDER INVESTIGATION

As soon as the chromatogram has been obtained, its separate zones are analyzed. Recently several methods have been developed for analyzing the components of a chromatogram that may be recommended.

1. Chemical analysis of individual components after they have been eluted into solution or directly on the adsorbent.

2. Spectrophotometric, carried out in two variants:

a) One of the methods of analyzing components directly in the adsorbed state consists in determining the spectral absorption of the adsorbed substance by measuring the diffuse reflection from the pure adsorbent and from the adsorbent with the substance adsorbed on it²³. By comparing two beams of light we can judge the total absorption of light by the adsorbed substance.

However, the data obtained do not give exact values of the absorption coefficients because of the specific character of the propagation of light in a turbid medium. The maxima on the spectral transmittance curve prove to be expressed somewhat more weakly—the substance will appear somewhat more whitish than in solution. The general form of the transmittance curve, however, is preserved.

b) Another variant consists in measuring the absorption of the investigated components after they have been transferred into solution. This method of analysis has the advantage that it can be successfully carried out in the case of a small amount of substance. Thus, the composition of one of the dyes, which is a sensitizer, was not established in any attempts at chemical analysis. By chromatographic means it was found that this dye is complex, since its chromatogram contains five rings (see Fig. 1). True,

Fig. 5.

some zones of the chromatogram contain very little substance, in accordance with the fact that there is very little of it in the dye itself; as is seen from Fig. 1, such is the uppermost zone (the dark band). However, by the spectrophotometric method an analysis of all the components was carried out, including the component corresponding to the upper zone. In Fig. 5 are shown the spectral absorption curves of the individual components. The zones on the chromatogram are counted from top to bottom.

1. Recently, the so-called “liquid chromatogram” method has become very widespread. It consists

in observing the separation process by directly measuring the concentration of the solution flowing out of the column, for which purpose the solution emerging from the column is fed in small portions into a cuvette. The solution concentrations measured in the cuvette are plotted graphically as a function of the volume of solution that has passed through the adsorbent, or of time. The curve expressing the observed dependence is called the “effluent curve.” For colorless substances the concentration of solutions is determined by measuring the refractive index with an interferometer. Such a method of analysis has been described in detail by S. Claesson²³.

IV. FORMS OF CARRYING OUT CHROMATOGRAPHIC ANALYSIS

If one considers the physicochemical factors that underlie the chromatographic separation of substances, three principal types of chromatographic analysis may be distinguished: adsorption (molecular and ion-exchange), partition, and precipitation chromatography.

Recently the development of chromatographic analysis has proceeded in the direction of creating new forms of carrying it out, differing in the method of obtaining chromatograms and in the manner of movement of the individual zones along the adsorbent column. We shall dwell on the characteristics of the individual forms of carrying out chromatographic analysis that have at present become more or less widespread.

1. Adsorption chromatography

a) Analysis by washing. After a chromatogram has been obtained, its individual zones are transferred into solution by washing with some solvent, and sometimes with several solvents. In this process the width of the zones increases as they move, i.e. the zones, as it were, spread out. The rate of movement depends on the adsorption affinity of the components of the chromatogram corresponding to the given zones for the given adsorbent, on the solvent used for washing, and on the temperature at which the experiment is conducted (the temperature of the adsorption column). A case may occur in which all possible selections of solvent do not lead to complete extraction of the zones into solution, i.e. one or several components of the solution under investigation are adsorbed by the adsorbent so strongly that their zones cannot be displaced from their initial position. This is one of the disadvantages of analysis by washing. In order nevertheless to wash out strongly adsorbed zones, it is necessary to raise the temperature of the adsorption column, whereby the adsorptive capacity of the adsorbent decreases and the given zones begin to yield to elution. Another disadvantage of the method is that, during washing, a very large quantity of solvent is consumed.

the amount of solvent. But these shortcomings are offset, since in the described method each component of the mixture leaves the column in a pure state, which is very convenient for preparative purposes. From the appearance of the “effluent curve” in elution analysis one can judge whether the separation of the mixture is incomplete or complete. Fig. 6 shows effluent curves for both the first and the second cases. With complete separation of the mixture, the concentration of the dissolved substance in the cuvette remains equal to zero throughout the entire experiment, except for those intervals of time when the chromatogram zones pass through the cuvette; these portions are represented on the graph by peaks. The area lying under such a peak determines the amount of substance present in the zone.

Fig. 6. Effluent curves in elution analysis:
I — incomplete separation, II — complete separation.

b) Thermal desorption. This form of chromatographic analysis is based entirely on shifting the adsorption equilibrium toward its decrease when the temperature is raised and toward its increase when the temperature is lowered. The actual process of analysis consists in the following: after the solution of the substance under study has been filtered in the usual way through a column with adsorbent, the temperature of the latter is gradually raised, as a result of which its adsorption affinity with respect to the components of the solution changes, and the individual zones begin to move along the column. The temperature of the column can be raised in various ways, for example: by winding several turns of wire around the column and gradually increasing the strength of the current passing through them, or by slowly changing the depth of immersion of the tube with the adsorbent into a refrigerator or into a furnace.

c) Frontal analysis^23. This form of chromatographic analysis is the simplest. Usually, after the adsorbent in the column has been wetted with a small portion of pure solvent, the solution of the substances under investigation is continuously filtered through the column. Until the adsorbent has become saturated with the given—

substance completely, pure solvent flows out of the column (the concentration of the dissolved substance when measuring the output curve is equal to zero), but as soon as the adsorbent becomes saturated, the concentration rises sharply. The output curve for a solution

Fig. 7. Output curves in frontal analysis:
a — for one component, b — for two components.

with one component has one step, as shown in Fig. 7, a; with two components, two steps (Fig. 7, b). The first step contains only the first component in pure form, the second—component 1 and component 2. Thus, we do not obtain a pure “liquid” chromatogram for each component, as was the case in elution analysis. However, if the adsorption isotherm is known for each pure component, then from the output curve the composition of the mixture being analyzed can be determined. The principal advantage of the frontal method is that it can be applied in cases where the substance is adsorbed on the adsorbent irreversibly, or where the substances differ very little from one another in their adsorption. The frontal method was, for example, widely used in the analysis of a mixture of fatty acids. The output curve for one such analysis is shown in Fig. 8.

Fig. 8. Frontal analysis of a mixture of fatty acids.

d) Displacement development. The difference between this method and the elution method consists in the fact that, after passing the solution of the mixture under analysis through the column, a solvent is passed through it to which a substance is added that is called in the literature the “displacer”; the adsorption affinity of the latter with respect to the given adsorbent must be greater than that of any component of the mixture. Thus, the added substance (the “displacer”), being taken up by the adsorbent in its very upper layers, moves all the lower-lying layers downward and displaces them from the column in the order of their vertical position. All components of the mixture are thereby obtained in pure

states and are characterized by the corresponding step in the output curve, as shown in Fig. 9; the height and width of the peaks, with preliminary calibration, make it possible to determine the qualitative and quantitative composition of the mixture being analyzed. In the described form of conducting chromatographic analysis it is possible to use adsorbents with a high adsorption capacity, or, as they say, capacious adsorbents, since the displacement is carried out by means of elution. This increases the efficiency of the process, for it is possible to analyze a large number of substances in the presence of many components. However, there may be cases of irreversible adsorption, when the zones cannot be displaced by anything. In this case the only suitable method proves to be frontal analysis.

Fig. 9. Output curve of a multicomponent mixture under displacement development.

d) Ion-exchange chromatogram. Recently, in inorganic chemistry, the ion-exchange chromatogram has become very widespread; it is based on the exchange of ions between a solution and an ion-exchange adsorbent. To carry out this form of chromatography it is necessary that the adsorbent used contain ions capable of being exchanged for the ions of the components of the solution; at the same time no other reactions, apart from exchange adsorption, should occur. Such adsorbents are customarily called permutites in the literature on chromatography. The arrangement of the components of a mixture in the adsorption column depends on the degree of adsorption of their ions by the given adsorbent, and their removal from the column is effected by ordinary washing. This form of chromatography has been successfully applied in the separation of the rare earths24 and radioactive elements25.

In Fig. 10 the output curve is given for the separation of the elements of the yttrium group. The radioactivity of the solutions flowing out of the column was measured with a Geiger–Müller counter. The theory of ion-exchange chromatography for various cases is described in the works of E. N. Gapon and T. B. Gapon8. They also carried out work on the separation of cobalt and nickel ions on a specially prepared permutite38.

2. Partition chromatography

This form of chromatographic analysis, proposed by Martin and Synge in 194133, has recently become widely used in the analysis of many substances. It is based on differences in the magnitude of the partition coefficients of the components

mixtures between two immiscible solvents. At the beginning of the analysis the adsorbent in the column is impregnated with some solvent, which is called the “stationary” solvent, and then a solution of the mixture being analyzed in another solvent is passed through the column.

Fig. 10. Separation of the yttrium group.

solvent, called the “mobile” solvent. The role of the adsorbent in the form of analysis described is reduced to the fact that it serves as a simple mechanical carrier of the “stationary” solvent; it is assumed that it does not interact in any way with the adsorbed substance. Not all adsorbents meet these requirements. Only a few of those known up to the present time can be successfully used in partition chromatography; these include: silica gel, cellulose, and starch. The individual components of the mixture are washed out of the column with the aid of the mobile solvent. In this way analyses have been carried out for the separation of amino acids, technical penicillin^26, and other substances on silica gel and starch. Paper partition chromatography exhibits certain distinctive features^27, and therefore we shall dwell on it separately. To carry out the analysis it is sufficient to have only a small strip of filter paper. Near one of its ends a drop of the solution under investigation in the “mobile” solvent is applied; the “stationary” solvent is water sorbed by the fibers of the paper. The strip of paper is then introduced into an atmosphere satu-

...saturated with vapors of water and solvent, and with one end of it (near which a drop of the solution has been applied) is brought into contact with the same solution containing water. Under the action of capillary forces the solution begins slowly to move along the strip of paper, and after several hours we obtain on the paper a series of spots arranged along the flow of the solvent and corresponding (with complete separation) to the number of components of the mixture. But there may be cases when, with the use of only one solvent, complete separation of the mixture does not occur; in this case the strip of paper is turned through 90° and its other end (near which the drop is located) is placed in a mixture of another solvent with water. A chromatogram obtained by development with one solvent is called one-dimensional, while one obtained by development with two solvents (in two mutually perpendicular directions) is called two-dimensional. Each component in a two-dimensional chromatogram is characterized by two coordinates, the magnitude of which is established by preliminary experiments with the individual components (Fig. 11).

Fig. 11. Two-dimensional partition chromatogram on paper.

3. Precipitation chromatography

A very promising method of precipitation chromatography, proposed by E. N. Gapon and T. B. Gapon^9, is still in the stage of development, and therefore its field of application is as yet small. Precipitation chromatography is based on the formation of precipitates when a solution is passed through a column; for this purpose an inactive powdered carrier is mixed, by one means or another, with a solid precipitant—a reagent which gives sparingly soluble compounds with the components of the mixture. On the basis of their investigations the authors conclude that “any substance (precipitant) capable of interacting with a solution of a series of other substances (the substances being chromatographed), when deposited on any highly dispersed substance (carrier), forms a precipitation chromatogram if the solubilities of the precipitates differ.” The sequence of formation of precipitates is determined by the magnitude of the solubility of the sparingly soluble compounds. The zones obtained for most precipitation chromatograms are separated by sharp boundaries.

V. CHROMATOGRAPHIC ANALYSIS OF COLORLESS SUBSTANCES

Tsvet also pointed out that “adsorption analysis, applicable primarily to pigments, also extends to colorless, i.e., ‘invisibly colored,’ substances.” All the above-mentioned forms of obtaining chromatograms for colored substances are also applicable when working with colorless substances.

The difficulty lies only in fixation and, consequently, in the analysis of the individual zones of the chromatogram corresponding to the individual components of the solution under investigation. At present several methods are known for analyzing chromatograms of colorless substances:

1. Empirical method. This method is, perhaps, the most primitive of all those described below. It consists in the following: the solution under investigation is filtered through an adsorption column, and then, after the column has been developed with a “developer,” the column of adsorbent is pushed out of the tube and, after being cut into several arbitrary parts, the adsorbed substances are transferred into solution. The individual substances are analyzed by a physical or chemical method.

2. Method of color reaction, carried out in various variants:

a) Conversion of colorless substances into colored ones before chromatographic separation. A substantial shortcoming of this method, as M. M. Senyavin^29 points out, is that the conversion of colorless substances into colored ones is associated with an increase in size and complication of their molecules, and this often leads to a decrease in the difference in adsorbability of the components of the mixture being separated and, thereby, hampers the chromatographic separation.

b) Obtaining a color reaction with adsorbed components, or the so-called “brush” method, proposed by Zechmeister, Cholnoky, and Ujhelyi.^30 It consists in applying a smear of reagent to the column of adsorbent after its removal from the adsorption tube. In this case the reagent gives different coloration with the adsorbed components of the chromatogram. Thus, for the separation of α- and β-naphthols, as reagent one may take a mixture of solutions of sulfanilic acid and sodium nitrite, which gives a violet coloration for α-naphthol and an orange coloration for β-naphthol.

c) Obtaining a color reaction with solutions of components flowing out of the column. Ruttgli and Jensen^18 separated naphtholsulfonic acids by adsorbing them with alumina from aqueous solutions. On washing with water, the indicated acids passed into solution successively one after another and could be detected with the aid of diazo solutions, which give different dyes with these acids.

3. Illumination of the chromatographic column with ultraviolet light.

The method described may be applied:

a) when the adsorbed substances themselves give bright luminescence bands upon illumination with ultraviolet light. In this case the color of the luminescence for one and the same adsorbed substance may differ, depending on the choice of adsorbent[^31];

b) when adsorbents are used that form a strongly fluorescent column. The method makes it possible to detect chromatogram components that are capable of quenching the fluorescence of the adsorbent and therefore appear as dark zones[^32].

4. Method of color transformation.

This method was developed by E. M. Brumberg[^34,^35] and is based on the fact that most substances that are colorless in the visible region have strong absorption bands in the ultraviolet region, i.e. are “colored” in that region of the spectrum which is not perceived by our eye. One of the fields of application of this method is the distribution chromatogram on paper. The method consists of the following. After a distribution chromatogram of some substance has been obtained on paper, as shown in Fig. 11, it is photographed in ultraviolet rays.

Owing to the selectivity of absorption of different components in the ultraviolet region of the spectrum, the spots formed by the different components will give images with different degrees of blackening. At the same time, the picture obtained when photographing in rays of different wavelengths will be different. Photography is carried out in rays of three different wavelengths, and the resulting photographs are projected onto one place on a screen by means of three projectors, in front of each of which is placed a light filter transmitting one of the three primary colors: red, green, and blue (other light filters may also be used, but they must be fully fixed when carrying out a given analysis). If all three images are superimposed, then on the screen we obtain a colored spot, the color and shade of which correspond only to the given component and will be different for different components of the mixture under investigation. The choice of the three wavelengths for photographing is determined by the nature of the absorption of the components of the substance under investigation. This method may, however, be simplified if it is replaced by the simple projection of a spot illuminated by three wavelengths from the ultraviolet region of the spectrum onto a fluorescent screen consisting of a glass plate on which, from the side of the object being viewed, three thin layers of fluorescent substances excited by these wavelengths have been deposited. E. M. Brumberg constructed a special instrument (ultrache-

microscope) to carry out chromatographic analysis by the method described. For a detailed description of the method, see46.

1. Measurement of the dielectric constant. This method was proposed by G. V. Troitskii36 and consists in measuring the dielectric constant directly along the column with the aid of a ring-shaped capacitor connected to the oscillatory circuit of a generator. The capacitor is fitted onto the column and can be moved along it.

2. The method of the liquid chromatogram, described by us in the present article somewhat earlier.

VI. PRACTICAL APPLICATIONS OF THE CHROMATOGRAPHIC ADSORPTION METHOD OF ANALYSIS

This method can be applied to the analysis of many objects where the adsorption process does not lead to a chemical change in the substance. We cannot examine here all cases of its application; therefore, dividing them according to their intended purpose*), we shall confine ourselves to only a few examples.

1. Determination of the degree of purity of individual products and purification from impurities

It is important to note that, in determining the degree of purity of a substance, one may assert that the substance is reliably pure if, in chromatographic analysis with several solvents on several adsorbents, the substance proves to be homogeneous, for, as Tsvet points out, “it is altogether improbable that the adsorption isotherms of two substances should change in exactly the same way under all variations of the medium and the adsorbent.” As an example of determining the degree of purity of a substance, one may cite the analysis, carried out by us, of a crystalline violet dye in acetone solution on aluminum oxide. As a result it was found that this dye, judging by its name, could have been thought to be perfectly pure, but in the chromatogram it revealed three rings, i.e., it contained as impurities two substances forming two narrow rings in the upper part of the column. The spectrophotometric characteristic of these fractions is shown in Fig. 12. M. V. Pronina37 applied this method for the complete removal of semicarbazones from their solutions in oils in the isolation of aldehydes and ketones from a shale-tar fraction. Other methods, for example the method of repeated crystallization upon cooling, did not give the desired result. The author37 indicates that the method of chro-

*) Such a division has been made by many authors, in particular in the article by B. Ya. Sveshnikov1.

matographic analysis on silica gel may be applied to tar, petroleum, and other products in which the semicarbazones are readily soluble. Yu. Yu. Lur’e and N. A. Filippova.^39

Fig. 12.

applied this same method for the determination of sulfur, phosphorus, and arsenic in nickel and copper.

2. Decomposition of a Complex Mixture into Components and Analysis of the Individual Components

For this purpose, adsorption analysis was first applied by Koval’skii^38 to glycogen*) from the liver and muscles of rabbit, dog, guinea pig, white rat, and frog. Chromatograms obtained on calcium carbonate, when developed with distilled water and iodine solution, contained from 2 to 11 rings. As a result of the analysis it was established that glycogens are complex mixtures of individual chemical substances—polysaccharides of the glycogen type.

In different periods of the year and under different functional states of the organism, glycogens gave a number of changes in the chromatogram, expressed in the fact that individual zones underwent quantitative changes.

*) Glycogen (animal starch) is a complex carbohydrate in the form of which reserves of carbohydrates are deposited in the human or animal body.

The chromatographic method, in the analysis of complex mixtures, can successfully replace other kinds of analysis: crystallization, distillation, or sublimation; and in some cases it is simply the only one.

Thus, in determining alkaloids in a belladonna tincture, usually the entire sum of alkaloids is found quantitatively, and only qualitatively—the presence of one or another of them.

By the chromatographic method, on a column with silica gel, from an aqueous-alcoholic solution of belladonna, after development and subsequent washing of the column with alcohol, atropine, scopolamine, and hyoscyamine were isolated in separate fractions; they were quantitatively determined by the method of luminescent titration^39.

3. Isolation of substances from very dilute solutions

This case is of great importance in determining a substance from a mixture when the given substance is present in the latter in negligible concentration.

Thus, with appropriate treatment of the roots and stems of Cinchona, quinette (a mixture of extracted alkaloids) is obtained with a quinine content averaging about 4–6%. Since quinette is a mixture of isomeric compounds and compounds close to quinine, the process of isolating quinine from quinette is complex and laborious. However, by the chromatographic method on a column of specially prepared SiO₂, this problem is solved within an hour^40.

4. Testing and control of products^41

The chromatographic method of analysis, applied for this purpose, is purely analytical in character and finds application chiefly in the analysis of food, pharmaceutical, and other products. For the purposes of practical control, it is not even necessary to obtain a chromatogram with which an exact chemical analysis could be established; it is sufficient to obtain such a chromatogram with which possible falsifications could be detected (for example, falsification of wines, oils).

For some objects, deviation from the norm is easily noticed with the aid of ultraviolet light.

5. Determination of molecular structure

This method is especially valuable for separating both structural and geometric isomers, and also for clarifying the relation between the adsorption capacity of organic substances and the structure of the molecule. For example, on aluminum oxide

can be separated into boletol and isoboletol43. These substances are isomers, differing only in the position of the hydroxyl group:

boletol

isoboletol

The adsorption capacity for certain linear polycyclic compounds increases with an increase in the number of benzene rings, i.e., the more highly colored compounds are found at the top of the column43.

In the series of fatty acids the adsorption capacity increases with the number of double bonds. Thus, we have the following series in order of increasing adsorbability: stearic, oleic, linoleic, and linolenic acids44.

The chromatographic method of analysis has been developing especially rapidly in recent times; the limits of its application and improvement are continually expanding.

In the present article we have not set ourselves the task of encompassing all the wealth of material that has accumulated at the present time, and have confined ourselves to a brief survey, indicating a small list of works in the various branches of chromatography.

CITED LITERATURE

1. M. S. Tsvet, Classics of Science, Publ. Acad. Sci. USSR (1946).
2. N. A. Shilov, B. V. Nekrasov, ZhRKhO 60, 103 (1928). A complete list of the works of N. A. Shilov was printed in the journal Uspekhi Khimii for 1948.
3. L. K. Lepin, Zh. obshch. khim. 1, 388 (1931); Zh. fiz. khim. 5, 276 (1934).
4. M. M. Dubinin, ZhRKhO 62, 683, 1627, 1829 (1930); Zh. obshch. khim. 1, 289 (1931).
5. R. Kuhn, A. Winterstein und E. Lederer, Zeits. f. physiolog. Chem. 197, 141 (1931).
6. M. M. Dubinin, Zh. prikl. khim. 4, 283 (1931).
7. A. A. Zhukhovitskii, Zh. fiz. khim. 10, 412 (1937); A. A. Zhukhovitskii, Ya. D. Zabezhinskii, D. S. Sominskii, Zh. fiz. khim. 12, 308 (1939).
8. E. N. Gapon, Zh. obshch. khim. 3, 144, 153, 159, 660, 667 (1933); A. N. Ivanov, E. N. Gapon, Zh. fiz. khim. 15, 659 (1941); E. N. Gapon, Zh. fiz. khim. 15, 665 (1941); Zh. fiz. khim. 20, 237 (1946); E. N. Gapon, T. B. Gapon, Usp. khim. 17, 452 (1948).

1. E. N. Gapon, T. B. Gapon, DAN 60, 401 (1948); DAN 60, 118, 817 (1948).

2. B. N. Nikol’skii, Zh. fiz. khim. 5, 266 (1934); B. N. Nikol’skii, V. I. Paramonova, Usp. khim. 8, 1535 (1939).

3. M. A. Konstantinova-Shlezinger, N. A. Gorbachëva, Zh. analit. khim. 3, 213 (1948).

4. O. M. Todes, Ya. M. Bikson, DAN 75, 727 (1947).

5. M. V. Radushkevich, DAN 57, 471 (1947).

6. A. N. Kharin, P. N. Protasov, Zh. fiz. khim. 22, 1219 (1948); A. N. Kharin, L. M. Voitko, Zh. prikl. khim. 22, 835, 1191, 1237 (1949).

7. A. L. Le Rosen, J. Am. Chem. Soc. 64, 1905 (1942).

8. L. Zechmeister u. L. Cholnoky, Ann. d. Chem. 516, 30 (1935).

9. E. Lederer, Bull. Soc. Chem. 5, 897 (1939).

10. P. Ruggli, P. Jensen, Helv. Chem. Acta 18, 625 (1935); 19, 64 (1936).

11. Heilbron, Phipers, Biochem. J. 29, 1369 (1935).

12. Kharmandaryan M. O., Kapelevich S. A., Zh. khim. promyshl. 7, 1484 (1930).

13. A. V. Kiselev, Zh. fiz. khim. 13, 452 (1949).

14. T. P. Kravets, Kh. L. Pesykina, Z. V. Zhidkova, Izv. AN SSSR (ser. fizich.) 14, 493 (1950).

15. S. Classon, Arkiv. Kem. Min. Geol. 15A, 9 (1941).

16. B. N. Ketelle, G. E. Boyd, J. Am. Chem. Soc. 69, 2800 (1947).

17. E. N. Topkins, I. K. Knym, W. E. Cohn, J. Am. Chem. Soc. 69, 2769 (1947).

18. R. R. Goodall, A. A. Levi, Analyst 72, 277 (1947).

19. Williams, Kirby, Science 107, 481 (1948).

20. T. B. Gapon, E. N. Gapon, Zh. anal. khim. 4, 131 (1949).

21. M. M. Senyavin, Usp. khim. 18, 183 (1949).

22. L. Zechmeister, L. Cholnoky, Unjhelyi, Bull. Soc. Chem. biol. 18, 1885 (1936).

23. A. Winterstein, K. Shön, Zeits. f. physiol. Chem. 230, 139 (1934).

24. G. W. Sease, J. Am. Chem. Soc. 69, 2242 (1947).

25. A. J. Martin, K. Z. Synge, Biochem. J. 35, 1368 (1941).

26. E. M. Brumberg, DAN 25, 473 (1939); Izv. AN SSSR (ser. fizich.) 6, 32 (1942).

27. E. M. Brumberg, I. N. Berezhnaya, V. V. Dutkinskii, S. E. Manuilov, DAN 74, 747 (1950); E. M. Brumberg, S. A. Gershgorin, DAN 69, 801 (1949).

28. G. V. Troitskii, Biokhimiya 5, 375 (1940).

29. M. V. Pronina, Zav. labor. 14, 1463 (1948).

30. V. V. Kovalevskii, Biokhimiya 13, 131 (1948).

31. V. S. Krasnova, Zh. prikl. khim. 18, 284 (1945).

32. V. S. Krasnova, Zh. prikl. khim. 18, 86 (1945).

33. I. K. Kozlov, Probl. pitaniya 7, 26 (1938).

34. Kogl, Deijs, Ann. d. Chem. 545, 23 (1935).

35. Winterstein, Vetter, Zeits. f. physiol. Chem. 230, 169 (1934).

36. Tярре, Biochem. Zeits. 305, 150 (1940); 306, 316 (1940); 307, 97 (1941).

37. Collection of articles on chromatography (edited by M. M. Dubinin); No. 1, 9 (1949).

38. E. M. Brumberg, UFN 42 (1951).

Fig. I.

Fig. II.

1. S. I. Vavilov, Bulletin of the Academy of Sciences of the USSR, No. 2, p. 9 (1949). ↩