Negative linear compressibility

The ideas of Cillery and Bush were taken up again in 1974 by Moya et ai [2.28]. They showed that the negative thermal expansion of /3-eucryptite would require its isothermal linear compressibility (X) parallel to the c axis to be negative if the thermal expansion was dominated by elastic effects. The same group (Hortal et al [2.29]) then measured compressibilities on (3-eucryptite single crystals and found Xy = —(1.13 1) x 10 cm /dyne and Xj = +(22.4 6) X 10 cm /dyne. 

The lattice compression is highly anisotropic, in which the c-axis is the hardest among the three axes. The linear compressibilities are in the order, Kj, > Ka > k. Figure 68 shows the temperature dependence of the linear thermal expansion coefficients along the three crystal axes, a, (f = a, b, and c) at ambient pressure. The a curves are found to be largely different among each crystal axis. The values of and are positive and have a maximum around 105 and 75 K with values of 200 and 73 X 10 K , respectively. On the other hand, aj, is negative below... 

E. P. Bartlett concluded that the compressibility of mixtures of hydrogen and nitrogen is not a linear function of its composition. A maximum positive deviation of 1-9 per cent, was observed in a mixture containing 60 per cent, of hydrogen at 200 atm. press. A maximum negative deviation of 0-55 per cent, appears in a mixture containing 25 per cent, of hydrogen at 1000 atm. press. 

The PEM is made of a piezoelectric transducer that is glued to a ZnSe crystal. The piezoelement converts a periodic voltage to a periodic mechanical (acoustic) wave, which compresses or expands the crystal. This movement changes the refractive index in the x direction and imposes a periodic retardation (or acceleration) of the fix component of the incident linearly polarized wave. The fiy component remains unchanged. The PEM is operated at its resonant frequency (50 kHz). If the optical element is at rest, the polarization of the radiation remains unchanged. If the optical element undergoes compression or expansion, the component fix has a positive (retardation) or negative (acceleration) phase shift relative to the phase component of the component fiy. 

The type of sample most frequently used (thin films, fibers, monofilaments, etc.) on the Rheovibron will not support a compressive load without buckling. This condition would, of course, distort the negative portion of the sine wave making measurements impossible. To alleviate this situation, the sample must be maintained under sufficient tension to eliminate buckling but insufficient to introduce non-linearity or excessive creep into the experiment. The amount of tension is critical to the measurement. 

---