407 Raman Effect and Temperature I—Ammonium Chloride, Bromide, Iodide y A. C. Menzies and H. R. Mills,* Physical Laboratory, University College, Southampton Communicated by Professor R. Whiddington, F.R.S.—Received October 2, 1934) This work was begun by one of us at University College, Leicester, hree years ago (when preliminary results were obtained for solid nitrogen •eroxide at about —80° C using C02 snow from gas-cylinders) f but wing to the great difficulty in obtaining refrigerants there, had to be ibandoned, and the work was recommenced at Southampton last year.
408 A. C. Menzies and H. R. Mills particular difficulty in the ultra-violet region. It can be used with any refrigerant, and even when charged with liquid oxygen can be left un attended for 30 hours.
Raman Effect and Temperature 409 t the refrigerant by the insertion between the head and the rest of the hr of washers of materials having suitable thermal conductivities. For given refrigerant the bottom of the bar will have the same temperature hatever the washer introduced, but the head will attain different fixed mperatures.
Raman Effect and Temperature 411 rainless steel head was tried, but scrapped on account of its poor con- activity. It is used in the high temperature apparatus.) If solid at ordinary temperatures, the material is first crushed until it is the form of grains of about 1 mm across, though the best size of grain upends upon the material. It is then sandwiched between plungers 'hich fit into the tunnel, and the whole is compressed in a vice, so that e powdered material forms a tablet across the middle of the tunnel, he best thickness varies with the material, the more transparent it is, le thicker must be the tablet. An average size would be about 5 mm lick, corresponding to a little more than 1/3 cc of material.
412 A. C. Menzies and H. R. Mills gated in the course of the present work. The particular interest attaching to the ammonium salts is that the ammonium ion NH4+ is thought to be capable of rotation in the solid at ordinary temperatures, but that this rotation settles down at temperatures of the order of —35° C (chloride -30° C; bromide -38° C; iodide -43° C).* It was thought that some evidence of this might be found in the Raman spectra of salts containing this ion. The most suitable would clearly be the halides, since the anion is monatomic, and so incapable of giving rise to internal vibrations. Thus any lines obtained with the chloride would be due to the NH4+ group, or to NH4C1, while if the sulphate, for example, were used the SO"4 internal vibrations might render the inter pretation more difficult.
Raman Effect and Temperature 413 Raman lines found by previous workers* for ammonium chloride ;! room temperature only are given in Table II; they are to be compared ith the first column of figures in Table I.
414 A. C. Menzies and H. R. Mills instrument, taking exposures at —26° C, —28° C, —32° C, —34° q The first two were similiar, and also the last two, but the two pairs were dissimilar, showing that the transition occurred between —28° C and -32° C.
Raman Effect and Temperature 415 tiding: its origin in the vibration of ammonium against chlorine atoms. / corresponding frequency has not been observed, however, in the bomide or iodide. This does not mean that they do not exist—the eidence is entirely negative, for they might be too weak to appear with t; exposure given.
416 A. C. Menzies and H. R. Mills to possess the caesium chloride structure, each ammonium tetra hedron is surrounded by eight halogens in a cubic structure. The most reasonable arrangement is to put the ammonium tetrahedron within the cube of halogen atoms as in fig. 2, where it will be seen that four halogens are near to corners of the tetrahedron, four to faces. We can now form a lattice by a simple translation so that the tetrahedra in all the cubes are similarly situated. It will now be found that any halogen is not symmetrically placed; a corner of a tetrahedron will be on one side, a face Fig. 2—Cube of eight chlorine atoms with NH4+ tetrahedon at centre, face towards observer of another tetrahedron on the other side. If now the lattice vibrates, there will be an asymmetric restoring force, and the corresponding frequency might be expected to appear in the Raman effect. This, then, may provide a rough picture of the arrangement in ammonium chloride below —30° C.