Saturation Shake-Flask Methods

Solubility measurement at a single pH [37-39] under equilibrium conditions is largely a labor-intensive procedure, requiring long equilibration times (12h-7 days). It s a simple procedure. The drug is added to a standard buffer solution (in a flask) until saturation occurs, indicated by undissolved excess dmg. The thermostated saturated solution is shaken as equilibration between the two phases is established. After microfiltration or centrifugation, the concentration of the substance in the supernatant solution is then determined using HPLC, usually with UV detection. If a solubility-pH profile is required, then the measurement needs to be performed in parallel in several different pH buffers. 

---

Avdeef, A., Berger, C. M., Brownell, C. pH-metric solubility. 2. Correlation between the acid-base titration and the saturation shake-flask solubility-pH methods. Pharm. Res. 2000, 17, 85-89. 

Although Poct sometimes can be related to the more easily measured capacity factor in HPLC, the most reliable values still are obtained from traditional shake-flask methods or the slow-stir technique (de Bruijn, 1990). Although slow and tedious, methods which equilibrate the solute directly between the mutually saturated phases can cover a log P... 

The shake-flask method is based on the phase solubility technique that was developed 40 years ago and is still the most reliable and widely used method for solubility measurement today (Higuchi and Connors, 1965). The method can be divided into hve steps sample preparation, equilibration, separation of phases, analysis of the saturated solution and residual solid, and data analysis and interpretation (Yalkowsky and Banerjee, 1992, Winnike, 2005). 

The most common method for determining partition and distribution coefficients is the shake flask method. In this technique, the candidate drug is shaken between octanol (previously shaken together to presaturate each phase with the other) and water layers, from which an aliquot is taken and analyzed using UV absorption, HPLC or titration. In terms of experimental conditions, the value of the partition coefficient obtained from this type of experiment is affected by such factors as temperature, insufficient mutual phase saturation, pH and buffer ions and their concentration, as well as the nature of the solvents used and solute examined (Dearden and Bresnen 1988). 

In our laboratory, saturation shake-flask solubility is routinely performed at pH 6.8 and pH 1.0 (for acids only) 0.5 mL of buffer is added to about 2 mg of sample in a small borosilicate vial (2 mL) and sonicated for 3 minutes using a sonicating bath. The suspension is then incubated at 25.0°C for 20 hours in a thermostated water bath at a shaking speed of200 cycles/min.The phases are separated by decantation/centrifugation.The supernatant is carefully collected and the solute quantified using the analytical method defined previously. Before each injection, a blank is performed to avoid phantom peaks due to carryover or impurities present in the system. After the centrifugation it is advisable to re-check the pH of the medium. 

Another common source of uncertainty in solubility determinations is when the compound is highly ionized (>99%) at the pH considered. This usually happens when the pH is more than 2 units above the pKa for an acid or than 2 log units lower than the pKa for a base. In this case the solubility pH profile is very steep around that pH, which results in a higher uncertainty on the result. This is the case of Diclofenac at pH 6.8, where we observed some differences within our measurements and between our data and literature values. This limitation is, however, valid for both dissolution template titration method and the saturation shake-flask. 

The solubility of the compound will typically be studied by creating a saturated solution, adding an excess of material under study to a known amount of solvent until dissolution ceases. The sample will then be shaken under controlled room temperature (20-25 °C) typically for 24-48 h. After this time, the solution will be assumed to have achieved a thermodynamic equilibrium, and the liquid will be saturated. It is important to be certain that some solute remains, to ensure that the liquid is completely saturated. The liquid is then assayed to determine the exact amount of material that went into solution. This value is the thermodynamic solubility of the molecule. This is typically called the shake flask method. ... 

The experimental approaches are similar to those for solubility, i.e., employing shake flask or generator-column techniques. Concentrations in both the water and octanol phases may be determined after equilibration. Both phases can then be analyzed by the instrumental methods discussed above and the partition coefficient is calculated from the concentration ratio Q/Cw. This is actually the ratio of solute concentration in octanol saturated with water to that in water saturated with octanol. 

Methods have been proposed to miniaturize, speed up and automate the shake-flask approach. The main difficulties in this challenge are the number of time-consuming steps which cannot be totally eliminated and the persistence of well known drawbacks. For example, the mutual saturation and decantation of organic and aqueous phases, or the crucial separation of the two phases after shaking which multiplies the manipulations. Automation of the process is also difficult due to several compound-dependent parameters which have to be rigorously controlled, such as the volume ratio between organic solvent and aqueous phase according to the estimated log P, or the sample concentration. 

The traditional shake-flask technique remains the method of reference for the direet measurement of log P ranging from 3 to - -6 with great accuracy and high precision if temperature control is maintained. In this method the pure solute is partitioned between two immiseible but mutually pre-saturated liquid phases sueh as distilled/deionized/Milli-Q water and highly pure... 

Few measurements of important parameters are as simple to make as the partition coefficient of a solute between water and octanol. The principles of the shake-flask measurements remain the same as in the Berthelot and Jungfleisch reports of 1872 and the further analysis by Nemst. Yet the interplay of solvation forces in both the water and octanol phases are so involved and complex that current molecular mechanics and quantum chemical calculations can dispel very little of the empiricism that now dominates logP estimation. It is difficult to predict how soon ab initio calculations will give us dependable information regarding the conformation, the tautomeric form, and the electron distribution of complex solutes, not just as they exist in a vacuum or crystal, but also as they exist in the aqueous phase as well as the water-saturated octanol phase. At present the quantum chemical methods that look the most promising for the complex solutes seem to fail when applied to the simplest hydrophobic organic structures of all the aliphatic hydrocarbons. 

Method I.—10 gms. freshly distilled aniline are added to a solution of 30 gms. cone, hydrochloric acid in 75 c.cs. water and diazotised with 8 gms. sodium nitrite in 30 c.cs. water, the temperature being kept about 0°. 30 gms. common salt are added with shaking, and the solution cooled in a freezing mixture. 60 gms. stannous chloride in 25 gms. cone, hydrochloric acid are then added, and after standing for some hours the hydrochloride of phenylhydrazine separates, is filtered oft and washed with a little saturated salt solution. It is transferred to a flask and treated with excess of caustic soda solution, when the free base is extracted with ether. The ethereal solution is dried with caustic potash, and the ether removed by evaporation. The phenylhydrazine may be purified, if desired, by freezing or by distilling in vacuo. 

---