Distribution coefficient, solvent

Mark-Houwink-Sakurada constant Mass transfer coefficient around gel Fractional reduction in diffusivity within gel pores resulting from frictional effects Solute distribution coefficient Solvent viscosity nth central moment Peak skewness nth leading moment Viscosity average molecular weight Number of theoretical plates Dimensionless number... 

The most widely used descriptor for the hydrophobicity term in toxicology is the distribution coefficient between octanol and water, log Pq< - (the environmental scientists would rather call it log The bulk solvent octanol is of course a... 

The constant K is termed the distribution or partition coefficient. As a very rough approximation the distribution coefficient may be assumed equal to the ratio of the solubilities in the two solvents. Organic compounds are usually relatively more soluble in organic solvents than in water, hence they may be extracted from aqueous solutions. If electrolytes, e.g., sodium chloride, are added to the aqueous solution, the solubility of the organic substance is lowered, i.e., it will be salted out this will assist the extraction of the organic compound. 

Hence one extraction with 100 ml. of benzene removes 3 0 g. (or 75 per cent.) of the n-butyric acid, whilst three extractions remove 3 5 g. (or 87-5 per cent.) of the total acid. This clearly shows the greater efficiency of extraction obtainable with several extractions when the total volume of solvent is the same. Moreover, the smaller the distribution coefficient between the organic solvent and the water, the larger the number of extractions that will be necessary. 

The stabiHty criteria for ternary and more complex systems may be obtained from a detailed analysis involving chemical potentials (23). The activity of each component is the same in the two Hquid phases at equiHbrium, but in general the equiHbrium mole fractions are greatiy different because of the different activity coefficients. The distribution coefficient m based on mole fractions, of a consolute component C between solvents B and A can thus be expressed... 

This is the important rule of additivity of resistances. In practice, and are often of the same order of magnitude, but the distribution coefficient m can vary considerably. For solutes which preferentially distribute toward solvent B, m is large and the controlling resistance Hes in phased. Conversely, if the distribution favors solvent A the controlling mass-transfer resistance Hes in phase B. 

Miscellaneous Pharmaceutical Processes. Solvent extraction is used for the preparation of many products that ate either isolated from naturally occurring materials or purified during synthesis. Among these are sulfa dmgs, methaqualone [72-44-6] phenobarbital [50-06-6] antihistamines, cortisone [53-06-5] estrogens and other hormones (qv), and reserpine [50-55-5] and alkaloids (qv). Common solvents for these appHcations are chloroform, isoamyl alcohol, diethyl ether, and methylene chloride. Distribution coefficient data for dmg species are important for the design of solvent extraction procedures. These can be determined with a laboratory continuous extraction system (AKUEVE) (244). 

These variations permit the separation of other components, if desired. Additional data on uranium, plutonium, and nitric acid distribution coefficients as a function of TBP concentration, solvent saturation, and salting strength are available (24,25). Algorithms have also been developed for the prediction of fission product distributions in the PUREX process (23). 

Among the properties sought in the solvent are low cost, avadabihty, stabiUty, low volatiUty at ambient temperature, limited miscibility in aqueous systems present in the process, no solvent capacity for the salts, good solvent capacity for the acids, and sufficient difference in distribution coefficient of the two acids to permit their separation in the solvent-extraction operation. Practical solvents are C, C, and alcohols. For industrial process, alcohols are the best choice (see Amyl alcohols). Small quantities of potassium nitrate continue to be produced from natural sources, eg, the caUche deposits in Chile. 

The equihbrium distribution coefficient can be calculated by material balance, using the weight of the feed F, raffinate R, and extract E, plus the weight-fraction solute in the feed xy and raffinate iv, when the weight-fraction solute in the extraction solvent y, is zero [Eq. (15-8)]. 

For a ternai y system, the phase diagram appears much like that in conventional liquid-liquid equilibrium. However, because a SCF solvent is compressible, the slopes of the tie lines (distribution coefficients) and the size of the two-phase region can vary significantly with pressure as well as temperature. Furthermore, at lower pressures, LLV tie-triangles appear upon the ternary diagrams and can become quite large. 

Extraction from Aqueous Solutions Critical Fluid Technologies, Inc. has developed a continuous countercurrent extraction process based on a 0.5-oy 10-m column to extract residual organic solvents such as trichloroethylene, methylene chloride, benzene, and chloroform from industrial wastewater streams. Typical solvents include supercritical CO9 and near-critical propane. The economics of these processes are largely driven by the hydrophihcity of the product, which has a large influence on the distribution coefficient. For example, at 16°C, the partition coefficient between liquid CO9 and water is 0.4 for methanol, 1.8 for /i-butanol, and 31 for /i-heptanol. 

Although, for most moderators, the surface of a stationary phase in LC can be considered stable at moderator concentrations above about 5%v/v, the results from the same experiments as those carried out by Purnell and his group could still be considered invalid and, at best, would not lead to unambiguous conclusions. Katz et al. [9] avoided this problem by examining liquid/liquid distribution systems using water as one phase and a series of immiscible solvent mixtures as the other and by measuring absolute distribution coefficients as opposed to retention volumes. 

Figure 16. Graphs Showing the Distribution Coefficient of n-PentanoI between Water and Three Binary Solvent Mixtures Plotted against Solvent Composition...

Katz et al. tested the theory further and measured the distribution coefficient of n-pentanol between mixtures of carbon tetrachloride and toluene and pure water and mixtures of n-heptane and n-chloroheptane and pure water. The results they obtained are shown in Figure 17. The linear relationship between the distribution coefficient and the volume fraction of the respective solvent was again confirmed. It is seen that the distribution coefficient of -pentanol between water and pure carbon tetrachloride is about 2.2 and that an equivalent value for the distribution coefficient of n-pentanol was obtained between water and a mixture containing 82%v/v chloroheptane and 18%v/v of n-heptane. The experiment with toluene was repeated using a mixture of 82 %v/v chloroheptane and 18% n-heptane mixture in place of carbon tetrachloride which was, in fact, a ternary mixture comprising of toluene, chloroheptane and n-heptane. The chloroheptane and n-heptane was always in the ratio of 82/18 by volume to simulate the interactive character of carbon tetrachloride. 

In the experiments of Katz et al., that validated the relationship given in equation (10), the distribution coefficients (K) were referred to the solvent phase (mobile... 

It should be recalled that the distribution coefficients are referenced to the solvent mixture and not the stationary phase and are thus the inverse of the distribution coefficient employed in the chromatography elution equation. 

When the relationship between the distribution coefficient of a solute and solvent composition, or the corrected retention volume and solvent composition, was evaluated for aqueous solvent mixtures, it was found that the simple relationship identified by Purnell and Laub and Katz et al. no longer applied. The suspected cause for the failure was the strong association between the solvent and water. As a consequence, the mixture was not binary in nature but, in fact, a ternary system. An aqueous solution of methanol, for example, contained methanol, water and methanol associated with water. It follows that the prediction of the net distribution coefficient or net retention volume for a ternary system would require the use of three distribution coefficients one representing the distribution of the solute between the stationary phase and water, one representing that between the stationary phase and methanol and one between the stationary phase and the methanol/water associate. Unfortunately, as the relative amount of association varies with the initial... 

Testing the applicability of equation (10) to liquids where the solvent components associate with themselves is experimentally difficult. Katz et al. attempted to do this by measuring the distribution coefficients of some solutes between hydrocarbon and... 

Katz et al. also plotted the distribution coefficient of n-pentanol, benzonitrile and vinyl acetate against the concentration of unassociated methanol in the solvent mixture and the results are shown in Figure 32. It is seen that the distribution coefficient of all three solutes is predominantly controlled by the amount of unassociated methanol in the aqueous solvent mixture. In addition, the distribution coefficient increases linearly with the concentration of unassociated methanol for all three solutes over the entire concentration range. The same type of curves for anisole and benzene, shown in Figure 33, however, differ considerably. Although the relationship between distribution coefficient and unassociated methanol concentration is approximately linear up to about 50%v/v of unassociated methanol, over the entire range the... 

In contrast molecular interaction kinetic studies can explain and predict changes that are brought about by modifying the composition of either or both phases and, thus, could be used to optimize separations from basic retention data. Interaction kinetics can also take into account molecular association, either between components or with themselves, and contained in one or both the phases. Nevertheless, to use volume fraction data to predict retention, values for the distribution coefficients of each solute between the pure phases themselves are required. At this time, the interaction kinetic theory is as useless as thermodynamics for predicting specific distribution coefficients and absolute values for retention. Nevertheless, it does provide a rational basis on which to explain the effect of mixed solvents on solute retention. 

This equation, although originating from the plate theory, must again be considered as largely empirical when employed for TLC. This is because, in its derivation, the distribution coefficient of the solute between the two phases is considered constant throughout the development process. In practice, due to the nature of the development as already discussed for TLC, the distribution coefficient does not remain constant and, thus, the expression for column efficiency must be considered, at best, only approximate. The same errors would be involved if the equation was used to calculate the efficiency of a GC column when the solute was eluted by temperature programming or in LC where the solute was eluted by gradient elution. If the solute could be eluted by a pure solvent such as n-heptane on a plate that had been presaturated with the solvent vapor, then the distribution coefficient would remain sensibly constant over the development process. Under such circumstances the efficiency value would be more accurate and more likely to represent a true plate efficiency. 

The distribution coefficient for n-heptane (solute i) distributed between ethylene glycol (solvent 1) and benzene (solvent 2) at 25°C is given as the ratio of mass fractions... 

This may be illustrated by the following example. Suppose that 50 mL of water containing 0.1 g of iodine are shaken with 25 mL of carbon tetrachloride. The distribution coefficient of iodine between water and carbon tetrachloride at the ordinary laboratory temperature is 1 /85, i.e. at equilibrium the iodine concentration in the aqueous layer is 1 /85th of that in the carbon tetrachloride layer. The weight of iodine remaining in the aqueous layer after one extraction with 25 mL, and also after three extractions with 8.33 mL of the solvent, can be calculated by application of the above formula. In the first case, if x, g of iodine remains in the 50 mL of water, its concentration is x,/50 gmL 1 the concentration in the carbon tetrachloride layer will be (0.1 —x1)/25gmL 1. 

The effect of molecular interactions on the distribution coefficient of a solute has already been mentioned in Chapter 1. Molecular interactions are the direct effect of intermolecular forces between the solute and solvent molecules and the nature of these molecular forces will now be discussed in some detail. There are basically four types of molecular forces that can control the distribution coefficient of a solute between two phases. They are chemical forces, ionic forces, polar forces and dispersive forces. Hydrogen bonding is another type of molecular force that has been proposed, but for simplicity in this discussion, hydrogen bonding will be considered as the result of very strong polar forces. These four types of molecular forces that can occur between the solute and the two phases are those that the analyst must modify by choice of the phase system to achieve the necessary separation. Consequently, each type of molecular force enjoins some discussion. 

It is the solvent, in this case methanol, that is responsible for reducing the distribution coefficients of the solutes with respect to the stationaiy phase (and consequently, their retention) by increasing the solute-solvent interactions in the mobile phase. Bearing this in mind, then the curves shown in figure 11 can explain some of the unique characteristics of methanol water mixtures when used as the mobile phase in reversed phase LC. 

---