140 BRIDGMAN. 



and then opened, it will be found that the change from red to black 

 Sb2S3 has not taken place completely; the mass is mostly transformed, 

 but there is still an appreciable quantity of the red left throughout the 

 mass, and particularly on the surfaces. The complete transition from 

 red to black Sb2S3 (which of course is irreversible) takes place in two 

 stages; the first stage is as just described at a pressure of 12000 kgm. 

 somewhere between 150° and 200°, and the second stage takes place on 

 releasing pressure from 12000 kgm. at the constant temperature of 200° 

 at a pressure of about 10000 kgm. This second stage in the transition 

 is also accompanied by a drop of pressure, that is, by a decrease of 

 volume. I have verified by trial that this second stage in the decom- 

 position is irreversible, as indeed it must be. This second drop is 

 remarkable; if the second stage in the transition were purely a pres- 

 sure effect, thermodynamics shows that it must be accompanied by 

 an increase of volume, instead of a decrease. The observed effect 

 must be due to friction; the highest pressures so increase the frictional 

 resistance to the second stage of the transition that it cannot run, but 

 as pressure decreases the frictional resistance falls rapidly until it is 

 low enough for the transition to proceed. 



In addition to the irreversible change to the black Sb2S3 at high 

 temperatures, red Sb2S3 shows a reversible transition at lower tempera- 

 tures. Several measurements of this reversible transition were made. 

 The transition line runs from 7800 kgm. at 0° to 12000 kgm. at 32°, 

 and the change of volume is approximately 0.010 cm.^ per gm. It was 

 not possible, however, to make any very accurate measurements of 

 this transition. The reaction is sluggish, and there is some impurity 

 present, as showTi by the rounding of both corners of the Av curves, 

 an unusual feature. It seems that it is not possible to obtain Sb2S3 

 entirely pure, but it always contains some free S and Sb2 Ss and other 

 sulfides. The result above shows that this impurity is soluble in both 

 phases. It is not likely, therefore, that any chemical refinement would 

 give a much more satisfactory material, and I did not try for further 

 more accurate measurements on the transition curve. 



The importance of the existence of this second modification of red 

 Sb2S3 is to be insisted on, however, because it shows that ordinary 

 red Sb2S3 cannot be amorphous, but must be crystalline. No sub- 

 stance is known with a transition of an amorphous phase. The 

 universal opinion that Sb2S3 is amorphous is doubtless due to its 

 appearance, which is as a precipitate, too fine to show any crystal- 

 line structure. Tammann, however, states that amorphous red Sb2S3 

 may be obtained by rapidly cooling the melt of black Sb2S3. We see 



