The shifts of the $N_1$ (3A2), $N_2$ (4A1), {4C2,4D2,4E2}, and 4F1 near-neighbor ${\mathrm{Cr}}^{3+}$-pair emission lines in ruby [see P. Kisliuk et al., Phys. Rev. 184, 367 (1969), for notation] have been measured under hydrostatic pressure up to 11 GPa. With increasing pressure, all near-neighbor pair lines shift to lower energy at rates between -7.4 and -10.3 ${\mathrm{cm}}^{\mathrm{-}1}$/GPa. These shifts are larger than, or approximately equal to, that of the isolated-${\mathrm{Cr}}^{3+}$-ion R lines (-7.53 ${\mathrm{cm}}^{\mathrm{-}1}$/GPa), and therefore near-neighbor pair lines are not expected to overlap the R lines used in the calibration of hydrostatic high-pressure experiments. Our results demonstrate that hydrostatic pressure shifts of the near-neighbor pair lines are not accurately modeled by an elastic-constant analysis of shifts measured under uniaxial stress. We evaluate our data in terms of Munro’s scaling-theory model, and find our results to correspond to a pressure shift of the orbital exchange parameter of dJ/dP=3×${10}^{\mathrm{-}2}$ ${\mathrm{cm}}^{\mathrm{-}1}$/GPa.

• Received 31 October 1984

DOI:https://doi.org/10.1103/PhysRevB.31.7449