II.—An Apparatus for the Determination of the Melting Point of Fats.

By FRANK T. SHUTT, M.A., F.C.S., F.I.C.,

AND

H. W. CHARLTON, B.A.Sc.

(Read May 29, 1900.)

In the course of an investigation recently undertaken to ascertain the effects of certain feeding stuffs upon the quality of the pork produced, it became necessary to determine the melting point of a large number of samples of fat. Since the work would include some twelve hundred determinations, a method was desired that would be fairly rapid and at the same time give uniform and comparative results.

The various methods cited by authors are all more or less unsatisfactory, as is acknowledged generally by chemists, and it is extremely doubtful whether any of them, save perhaps that which notes the temperature at which a particle of fat assumes the spheroidal condition in a fluid of like specific gravity, give the true melting point. This latter method is an extremely tedious determination, and its use, when a large number of samples is to be examined, quite impracticable.

While it may be possible with unmixed glycerides to obtain fairly concordant readings by observing the disappearance of opacity in drawn-out glass tubes, such does not seem to be the case with mixed fats. Pork fat, it may be stated, consists in proportions that vary slightly, according to the part of the animal from which the sample is taken, of olein, palmitin, and stearin-the first, fluid; the two last, solid, at ordinary temperatures.

At the outset we made a trial of methods involving (a) the coating of the thermometer bulb with the fat, and (b) the melting of the fat in drawn-out glass tubes, both open and sealed, using in some cases air, and in others, water, as the medium for conveying the heat. With these tubes, bound to a thermometer, we observed the temperature at which the fat became transparent and also at which the small cylinder of fat rose in the tube, if water was the medium, and that at which the first drop fell, if air was the medium employed. After giving these plans an exhaustive trial, we came to the conclusion that they were not reliable, for it was impossible, even with most careful attention to details to get duplicate readings nearer than one or two degrees Celsius.

Our attention was next directed to the method of Christomanos, in which mercury is used as the medium to convey the heat to the fat

in the drawn-out glass tubes, and the electric current, with a bell in the circuit, is employed to note the moment of fusion. (For details of process, see Journal of the Society of Chemical Industry, Sept., 1890.) In using this apparatus we found it practically impossible to obtain concordant results throughout a series of determinations owing to the extreme difficulty of introducing the same amount of mercury into the tube containing the fat. Thus, necessarily, the pressure upon the fat varies somewhat with each determination. Since slight differences of pressure affect materially the apparent melting point, we decided, if possible, to materially modify the method, avoiding the use of the drawn-out glass tube, and eliminating, as far as practicable, the factor of pressure. In this we were, in a large measure, successful.

The apparatus is depicted in the accompanying illustration and may be described as follows:-(a) water-bath, (b) porcelain dish hold ing mercury, (c) small porcelain capsule also containing mercury, (d) thermometer reading to one-fifth degree, (e) cone, through which a wire closing the circuit passes. The disc of fat rests on the plane surface of the cone and insulates the exposed part of the connecting wire, (f) battery, (g) bell, (h) metal mould for making fat discs.

The cone is conveniently made by passing a copper wire about ten inches in length through the stem of a small funnel and then pouring in melted sealing wax, keeping the wire central. The surface of the

wax, after solidification, is smoothed with a file, the end of the copper wire projecting, say, 2 millimetres, beyond the plane.

The circular mould (h) may be made from a piece of sheet lead. Its diameter should correspond with that of the mouth of the cone (inverted funnel) and have a thickness of, say, 4 millimetres.

Having obtained the weight of the cone (e) with its projecting wire, a calculation is made to ascertain the diameter of a disc of fat of the thickness of the mould necessary to displace a volume of mercury equivalent to the weight of that of the filled cone. A circle of metal of such a diameter is then cut from the mould, which now presents the appearance of a washer. Its surfaces are amalgamated to prevent adhesion of the fat.

Modus operandi: Fill the water-bath so that the surface of the porcelain dish (b) below the ring is in contact with the water. Pour mercury in (b) to the depth of about half an inch, and insert thermometer, as shown in illustration. Arrange the height of the capsule (c) so that the end of the copper wire from the cone dips about half an inch under the surface of the mercury, but taking care that it does not rest upon the bottom of the capsule.

The disc of fat is made by placing the mould (h) on the base of the cone (conveniently held inverted in the left hand) and filling the central space with the fat to be tested. This is most easily done by using

a spatula. The surface of the fat should be left even with that of the mould, which is then removed.

After placing the cone with its adhering disc of fat on the surface of the mercury (b), the temperature of the water-bath is uniformly and slowly raised. We have obtained the best results from using a rising temperature practically equivalent to 1 degree per minute.

After the melting point has been noted (which is indicated by the ringing of the bell), the melted fat is wiped from the surface of the mercury by a piece of filter paper. In making a number of determinations, the mercury will be found, after a time, to become sluggish. should then be shaken with gasoline, together with a few drops of nitric acid, if necessary.

It

The following data include the highest and lowest melting point determinations in the series under investigation. Two other results (C and D) taken without any selection from our records have been added, in order to show the extent to which duplicate determinations may vary:

The use of this apparatus seems to present the following advantages: (1) There is no necessity to re-melt the fat, as when drawn-out glass tubes are used, and thus one source of error is avoided; (2) The factor of pressure is practically eliminated; (3) Rapidity; The method appears to allow of about twice as many determinations being made per hour as by that of Christomanos. In conclusion we may say that the variations due to a too rapid rise of temperature do not appear to be any greater than those of other electrical methods.