M Szyrej, B Marciniak, I Kotula - Turbidimetry as a method of solubility and metastable zone width determination in organic systems - страница 1



Серія фізична. 2000. Bun. 33. С. 299-304

VISNYK LVIV UNIV. Ser.Physic. 2000. № 33. P. 299-304


PACS number(s): 82.90.+j






M.Szyrej, B.Marciniak, I.Kotula

Institute of Chemistry, Pedagogical University.Armii Krajowej 13/15, 42-200 Czqstochowa, Poland

1. Introduction

A knowledge of some physical properties of solutions such as solubility, density, viscosity, refractive index and their ability to supercooling, play an important role in analysing the crystal growth mechanism, modelling of the solute cluster formation, controlling of the solution supersaturation during crystals growth and, as in a case of the solubility, in analysis of processes such as crystallization and extraction, of an enormous economic importance.

The crystallization process, which includes the preparation of supersaturated solution, formation of nuclei and the growth of crystals is closely connected with the above ability to supercooling of solution, which is the most often expressed by the width of the metastable zone.

A measure of this last quantity, which allows description of the nucleation process, is the maximum supercooling of solution i.e., difference between solubility and nucleation, usually detected as first symptoms of precipitation. Observation of this symptoms is however rather complicated, due to the fact that the sensitivity of the experimental set-up should allow the detection of nuclei as small as possible.

The aim of the present work was to assess the usefulness of a turbidimetric method in the investigation of organic solutions in their metastable state and especially in determination of the width of the metastable zone. Our investigation concerns solutions of pyrene in benzene (B), 1,1-dichloroethane (1,1-dCHLE) and 1,2-dichloroethane (1,2-dCHLE). The solubility temperatures obtained in this way were compared with solubility data evaluated by classical, dynamic method.

2. Experimental

2.1. Materials

Pyrene p. (Loba, Austria), was recrystallized several times from benzene and anhydrous methanol, giving a purified sample that melted at TmfJK = 425.15 ± 0.5.

The solvents used were purchased from Fluka AG. These chemicals were dehydrated with anhydrous sodium sulphate and distilled with a packed column of 1 m height at large reflux ratios, just before each experiment.

© Szyrej M., Marciniak В., Kotufa I, 2000

2.2. Turbidimetric measurements

The turbidimetric measurements were carried out using apparatus schematically shown in Fig. 1.

A sample of solute-solvent mixture of predetermined composition was prepared by weight (weighed out to 10~8 kg).

The sample (1.5 cm3) and magnetic stirrer were placed in a glass cuvette (3), closed by teflon cork equipped with platinum resistance thermometer (4). The temperature of the cell was controlled using the Peletier thermostat (6), which allowed to change the temperature of the sample, in the range of 288 К - 318 К, with accuracy of 0.01 K. Dur­ing measurement, the sample was stirred with a magnetic stirrer, at constant rate. Photo­diode (1) emits the light X = 620 - 800 nm, passing through the sample under investiga­tion, and incident on the receiving photodiode (2). The turbidimeter measures the receiv­ing photodiode potential (U) versus temperature (T) of the sample. Both the temperature and potential were registered by PC (9) every 3 s.

The accuracy of measurements of the potential was 0.2 mV and of the temperature 0.01 K. Measurements were performed at the heating of the samples with the rate 2 K/min and cooling with the rate 0.04 K/min. Each experiment was made several times, to ensure reproducibility and to produce satisfactory values (for Ts ±0.1% and for Tp ±0.4%).

The solubility of pyrene in benzene, 1,1-dCHLE and 1,2-dCHLE was also deter­mined in the temperature range corresponding to temperatures of the turbidimetric meas­urement by a classical, dynamic method, based on the measurement of the so called tem­perature at which the last trace of crystals disappeared, described in our former works [1].

3. Results and discussion

The investigations involved the determination of a receiving photodiode potential (U) changes versus temperature (T) during a slow increase and decrease of temperature of the samples of investigated solutions. In Fig. 2, the typical, for the used apparatus, run of above dependences, obtained for solution of pyrene in 1,1-dCHLE, is presented.

U [mV]

700 600 500 400 300 2001 100-




Fig. 2. The dependences of receiving photodiode potential on temperature for the pyrene solution in 1,1-dCHLE (x = 0,0901) for heating (A) and cooling (B).

The characteristics obtained under appropriate experimental conditions had a re-peatable shape observed during numerous heating and cooling experiments.

As shown in the Fig. 2 the photodiode potential recorded at the growing tempera­ture at first does not change and then decreases, receiving a minimal value at the tem­perature (7'i). During the next heating of the sample the potential dramatically increases up to the maximum constant value at the temperature (Ts). Our interpretation of the above dependences is based on the known relation of the intensity of the transmitted light and the solution turbidity

/ = /0ехр(-т/), (1) where / is intensity of transmitted light; I0 is intensity of incidence light, x is a turbidity and / is thickness of solution layer. Values of /0 and / are constant under our experimental conditions, thus the intensity of transmitted light is turbidity-dependent. The turbidity is defined as a total intensity of scattered light in all directions.

Taking advantage of Reyleigh's [2] formula and assuming that particles have globu­lar shape of the radius r, we may describe turbidity by the equation

і = kNr\ (2) where к is constant for experimental conditions, N is the number of particles in the given volume and r is the radius of a particle.

When the solution supersaturation is extremely high, the number of all precipitates formed at the earlier and later stages of the process is approximately constant, therefore the investigated system may be considered as a monodispersed one. Thus turbidity is dependent only on the radius of particles.

The decrease of the potential value observed at the temperature {Tx), indicating the maximal solution turbidity, may be connected with the monodispersion of particles in the whole solution volume resulting from the stirring.

Under our experimental conditions we suggest, that a rapid increase of photodiode potential during heating, proportional to the intensity of transmitted light, is connected with the change of the size of dissolved particles. The lowest temperature, at which pho­todiode potential achieves a maximum value, indicates the moment, when scattering particles exist no more in the solution. We take this temperature as the saturation tem­perature (Ті).

As is shown in the Fig. 2, during the sample cooling, at the temperature (Tp) the rapid potential decrease to the value near zero at the temperature (T2) takes place. This potential decrease may be connected with the appearance of considerable number of detectable nuclei. The temperature of the first precipitation (Tp) was evaluated at the beginning of the potential decrease. Values of the metastable region width (АГтах) for different concentrations were calculated as differences between temperatures of satura­tion and precipitation, defined as above. An average accuracy of solubility temperature measurements by both methods was estimated as 0.3 K, while accuracy of measurements of precipitation temperature was about 0.6 K.

In order to assess usefulness of turbidimetric method in determination of equilib­rium temperature of a solid-liquid solubility, the measured solubility data for pyrene in benzene, 1,1-dCHLE and 1,2-dCHLE are compared with those obtained by a dynamic method, (Fig. 3), as logarithmic dependences, lnxs=J{l/T,). The dashed line in those figures indicates the ideal solubility of the solute calculated from pyrene melting point and melting enthalpy. The full lines, defining experimental solubility of pyrene, were obtained as a result of their fitting to the experimental points, by the last-squares method.

a) b)

It can be seen from Fig. 3 that the investigated solutions are non-ideal with positive deviations from the ideal solubility calculated from the melting point (424.4 K) and the melting enthalpy (17 111 J/mol) of pyrene.

In Table 1, the values A and В of the solubility equation, in general form \nxs=A- BITS_ are listed.

The values of solubility temperatures (Ts), measured by both methods, and the first precipitation temperature (Tp), measured by turbidimetric method only, for concentration range investigated by turbidimetry, are given in Table 2.

It can be seen from Table 2 (columns 2, 3), that the solubility temperatures of pyrene, obtained by turbidimetric method are lower than those measured by dynamic method. The differences of the solubility temperature, calculated at the same concentration are on the average 0.8 К for benzene, 2.8 К for 1,1-dCHLE and 4.7 К for 1,2-dCHLE.

In Table 2 (columns 3, 4 and 5) the values of metastable zone width, estimated as a difference of turbidimetrically measured (7",) and (Tp) are presented. Average values of АГтах are 7.4 К for benzene, 5.3 К for 1,1-dCHLE and 7.2 К for 1,2-dCHLE. 4. Conclusions

On the basis of obtained results it was established that:

1.  The values of the solubility temperatures, determined by turbidimetric method are in every case lower, than those obtained by dynamic method. Average differences are: 0.8 K; 2.8 K; 4.7 К for benzene, 1,1-dCHLE and 1,2-dCHLE solutions, respectively. Mean percentage relative error of the measurements of solubility temperatures by both methods is 0,9%. It was also established that the solubility temperatures of pyrene in benzene, determined by turbidimetric method deviate from literature values for ca. 0.4 % [3].

2.  Mean metastable zone width values estimated by the turbidimetric method are 7,4 К for benzene, 5.3 К for 1,1-dCHLE and 7.2 К for 1,2-dCHLE.

3.  Performed measurements indicate that the turbidimetric method may be used for the investigation of the solubility of solids in liquids and for the estimation of the metastable zone width. The advantage of this method is the short measurement time, this fact being of importance in investigation of solutions of highly volatile solvents.


1.  I.Kotula, B.Marciniak, in International Conference on Solid State Crystals'98: Single Crystal Growth, Characterization, and Applications, eds. A.Majclirowski, J.Zielinski, Proc. SPIE 3724, 75 (1999),

2.  K.A. Stacey "Light Scattering in Physical Chemistry", Academic Press, New York 1956,

3.  E. McLaughlin, Н.Л. Zainal J.Chem.Soc. 863 (1959).


Інститут хімії і захисту оточуючого середовища Педагогічного Університету
Ченстохова,                                  13/15, 42-200 Ченстохова, Польша

У цій статті подані результати спроб використання турбідиметрії як нового мето­ду визначення розчинності та ширини метастабільної зони в органічних системах. На основі дослідження деяких систем "пірен - органічний розчинник" виявлено, що ви­значені турбідиметрично температури розчинності цієї розчинної речовини перебува­ють у добрій відповідності з температурами,отриманими класичним, динамічним мето­дом. Відхилення між ними становлять 0,8 К. для бензолу; 2,8 К для 1,1 -дихлоретану і 4,7 К для 1,2-дихлоретану. Виявлено також, що турбідиметричний метод може бути використаний для оцінки ширин метастабільних зон таких систем. Ключові слова: турбідиметрія, метод, система.

Стаття надійшла до редколегії 29.05.2000 Прийнята до друку 15.09.2000


Похожие статьи

M Szyrej, B Marciniak, I Kotula - Turbidimetry as a method of solubility and metastable zone width determination in organic systems