352 



THE BELL SYSTEM TECHNICAL JOURNAL, MARCH 1952 



The polystyrene solutions discussed above were in benzene, a good 

 solvent. Here, the situation is converse to that for polyisobutylene ; for 

 polystyrene, cyclohexane is a poor solvent and benzene, good. Hence, if 

 the previous interpretation of reduced single chain quasi-confignrational 

 (fj-'i) stiffness is general for solvents of more endothermic mixing, the 

 "plastic" molecule polystyrene shoukl show it in cyclohexane. This is 

 indeed evident in Fig. 30, showing one of the same polystyrenes of Fig. 

 29, measured at 20 kc (normalized to 1 per cent concentration). Also, on 

 Fig. 30 are shown the inherent viscosity (practically, the intrinsic vis- 

 cosity, in this case) and the absolute viscosity of the 1 per cent solution 



2 1200 



(0 



^ 1000 



(0 20 30 40 50 60 70 80 90 100 



FREQUENCY IN KILOCYCLES PER SECOND 



Fig. 29 — Change of shear stiffness, ^xb , with frequency, for 1 per cent solutions 

 of polystyrene in benzene. 



under steady flow, rjs . These are all plotted against temperature down 

 to phase separation, at about 26° to 27°C. 



The marked positive slope of the ^nrjr/c curve denotes the large con- 

 traction in molecular coil volume preceding phase separation or msolu- 

 bility. The absolute viscosity, rjs , however, rises with declining tem- 

 perature because it is dommated by solvent viscosity, but when the 

 polymer phase comes out, rjs abruptly falls off. 



The Mb values are consistent with this steady flow behaviour, except 

 that the rise of hb at the tiu'bidity point seems to be because a layer of 

 swollen polymer-rich phase forms on the torsional crystal surface. This 

 condition is seen in Fig. 30 to coincide nicely with the abrupt changes 

 in steady flow \'iscosity. 



The slight maximum m the mb curve at about 35°C may not be real. 



