TECHNIQUES FOR ASSESSING RHEOLOGICAL PROPERTIES 463 of shear. One normally finds, for this very low rate of shear, if plotting the viscosity of the system against time, that initially the viscosity will be a minimum but it then will gradually build up with time to a steady maximum value. In the case of a simple w/o emulsion of medium concentration this takes something like 30 sec, but nevertheless one can get quite a number of readings in between. With pigment and other materials present in the continuous phase it takes rather longer. One then applies an equation to this data to find out what apparent volume fraction of material is actually present, i.e. one has the original particles which will once again link up in the stationary state. I am assuming that all the particles are approx. the same size but this need not be the case-and in between the particles one gets a certain amount of liquid which is bound to be held within the particles therefore the effective volume fraction disperse phase is higher than the actual volume of particles present. This apparent volume fraction can be determined from the equilibrium viscosity at the low rate of shear. This method was originally introduced by Mooney (26), who gave an empirical equation for determining the theoretical relationship between viscosity and volume fraction. All he did was to take this equilibrium viscosity measurement, transfer it to this graph, and find the apparent corresponding volume fraction. This was always higher than the theoretical value, and it gives some idea of the amount of liquid held within the aggregates. But unfortunately Mooney's equation was a very empirical one, he did not appreciate that particle size has a great influence. Because his equation held for some old data, he assumed that it could be applied to systems with quite different particle sizes, which is not true. I have discussed this question (7) and given an alternative equation which can be used. MR. N. ]. VAN ABBI•: It seems to me that rotational viscometers, which are quite elaborate and expensive instruments, are basically dependent upon delicate mechan- isms such as torsion springs. How often should calibration checks be made ? THE LECTURER: Most of the instruments described are quite simple, possibly apart from the commercial viscometer. The one described in p. 447 operates on a coaxial cylinder principle with a simple torsion wire. Obviously with any instru- ment one must periodically check whether the wire or spring is functioning correctly. It can be replaced and one can still get reproducible results. This raises just one point. Very often it is extremely difficult to reproduce a particular consistency. Even within a single batch of material there might be a wide range of consistencies, e.g. in a simple, spherical, baked cake, consistency measurements at different points reveal a wide variation from the centre to the . outside of the cake. This complicates matters because one is never certain that a stating panel and the instrument are testing the same product. MR. R. CLARK: Certain toilet preparations are now being consumed in sufficient quantities to merit their consideration for production by continuous methods, toothpaste for instance. Which of the various techniques mentioned would you recommend as a continuous viscometer for simple process control? TH• LECTURER: This is a very old problem for which there is no solution at present. The only instrument I know of which will function satisfactorily in a pipeline is the Ultraviscoson viscometer, but unfortunately this is only suitable for (26) Mooney, M. y. Colloid Sci. 1 195 (1946).

464 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS measuring Newtonian viscosity. As far as I know there is nothing available for non- Newtonian measurement. DR. A. W. M•DDL•TO•: In page 458 you refer to a correlation between palate assessment of gum solutions and their rheological properties. Can you explain how this correlation was derived ? Ts• L•CTUR•R: As far as I can remember people were asked to estimate the degree of sliminess of the gum solutions presented to them, and rheological studies were simultaneously carried out on these solutions. They did find by statistical correlation that there seemed to be a relationship between the degree of sliminess and the degree of non-Newtonian behaviour of the material. The number of materials examined at the time, however, was quite limited, so I do not know whether this reported opinion would have very wide application. DR. A. W. MXD•L•TO•: Would the errors be likely to occur in the assessment of the subjective relations ? THE LECtUReR: Yes. From the little experience I had with gum solutions and similar things, they are very difficult to reproduce in consistency, unless one very carefully follows a standardized procedure and allows the solutions to age for a certain given time before testing. This certainly applies with gelatine. DR. A. R. RoGeRs: What methods do you use for studying distribution of sizes in the 1/• and less range ? Ts• L•C•UR•R: At the moment we are able to see those which are just below 0.5/• under the microscope, but we have no accurate method of determination I understand that Dr. Otterwill developed a method based upon light scattering principles which enable him to do this for polystyrene latex particles. These, of course, are different, but we are hoping to examine something along these lines next year. One of the major difficulties, even when you are able to measure these sizes, is how to find the average size. In all methods of calculating average size, apart from doing a pure arithmetical mean, one gets involved in summing the powers of the particle size, and all these submicroscopic sized particles would just cancel out. I have been thinking about this subject of late, because it is very important to our rheological studies. Possibly in order to show the effect of these submicroscopic sized particles one should calculate not a mean, but a reciprocal mean diameter, and then the very small particles would play a much larger part than the larger ones, which is the right order. DR. J. J. MAus•R: What would be the best method for measuring the spread- ability of thixotropic gels ? THE LEC•URSR: Any method which will enable you to measure viscosity ag very high rates of shear. DR. J. J. MAUS•ER: Is that a matter of spreadability ? TnE L•CTUR•R: I imagine that one would get a drastic breakdown of any strucgure which is present in the material, and this would be the same as the conditions pre- vailing at high rate of shear. The problem with all the rheological methods of measurement is that one does not know what shearing conditions one is using in the practical process. As yet no one has been able to calculate these. One or two