56982 str (80)

i iS    VISCOSITY

duo to the fact that " large groups of _at® togcther with no borę shearing force th^⁸ % be cxpected for one ” (Eyring). Polanyi p^SS that if, owing to strain, the ist and the (_n atom of one row coincides with the ist and Ju Ifc seeond row (the atoms being uniformly spaced in ^lri * row), then one group can be displaced with res^ to the other with i/wth the energy reąuired whe^6ct atoms in one layer coincide with n atoms in the oti? ⁿ The ąuestion of the viscosity of non-Newtoni^’ systems has already been discussed, and it found that it is better to determine their propertj^ in absolute units, even if only at standard arbitrary conditions, than in empirical units, which could not possibly be reproduced with a different type ₀f

instrument. Viscosity is then defined as *¹ and

S

shear modulus as—. We can speak of the viscosity at a point on the flow curve. For plastics like clay

• Hi

and tooth-paste, it is morę convement to use -~

because it happens to be constant over a certain

stress rangę, whereas = is not; but for matenals

which do not give straight linę flow curves, the viscosity is as good a property to quote (under standardised conditions) as any other. The position is unsatisfactory, but it will not be clarified until the laws of flow have been fully worked out, and all the proper constants can be calculated. It must be realised that the distinction between recoverable and non-recoverable deformation is in itself arbitrary, depending on the time taken for

¹ Where v is the ratę of shear.

|i _experimentation. As has already been Inentioned, Bingham describes a vanety of marblc which is solid when examined over a period of hours, but would be observed to flow like a fluid were it examined over a period of years. Tammann points out that glasses are popularly supposed to be supercooled fluids, and not solids; but for all ordinary experimental purposes they behave as solids. Houwink, in his book, discusses the structure of glass at considerable length. The most likely theory seems to be that the network of atoms in cooled glass is neither periodic and symmetrical, nor entirely random. All the interatomic distances will not be eąually probable. Instead of a regular honeycomb structure of atoms, the atoms will be linked together in polygons having various numbers of sides. The breakdown of this type of system is gradual and not sharp. Some resins also show “ slow-yield,” which Houwink shows may cause the loosening of screws or bolts used to hołd pieces of resin together. Tammann and Berger explain the structure of glass as due to primary particles in the form of tiny coin-like discs. These piles of discs offer little resistance to bending, and the volume changes produced by heavy pressure show certain anomalies, which are easily explainable by this hypothesis. A negative coefficient of expansion is explained by a bending of the discs at high tem-peratures, so that a closer packing occurs.¹

BIBLIOGRAPHY

Bastow and Bowden. Proc. Roy. Soc. (A), 1931, CXXXIV., 404; (A), 1935, CLI., 220.

Bouyoucos. Soil Science, I93²< XXXIV., 393-

1 von BuzcLgh: 1 Colloid Systems.” Technical Press Ltd., 1937-