Absorption transitions between two excited optical states in ruby have been studied by comparing the absorption coefficients measured under optically pumped and unpumped conditions. The spectrum has been determined in the π-and σ-polarizations in the photon energy range from 5,500 to 45,000 cm -1 . Several absorption bands have been found which arise from the transitions from the excited states t 2g 3 2 E g and t 2g 3 2 T 1g of the Cr 3+ ion. From the locations and intensities of these bands, the terminal states of the observed transitions are assigned to the charge transfer state, the t 2g 3 2 T 2g state, and the doublet states belonging to the electron configuration t 2g 3 e g . Calculated energies between these states and the metastable states, t 2g 3 2 E g and t 2g 3 2 T 1g , are in good agreement with the peak energies of the observed absorption bands. The present experiment gives new information inaccessible by the ordinary ground state absorption measurement, such as the locations of several doublet states with the configuration t 2g 2 e g and the intensities of doublet-doublet transitions in ruby.
Many existing advanced laboratory experiences associated with the metastable 2E term of Cr3+ in ruby, which gives rise to the R-lines at 692.7 and 694.3 nm, focus on a room-temperature measurement of the radiative lifetime of the 2E term. In our local work developing a laboratory experience in atomic phosphorescence, we noted a lack of consistency in the literature for the lifetime of the Cr3+ 2E term. These projects typically use commercially available ruby spheres for which the manufacturer(s) only state an ∼2% Cr concentration. The uncertainty in chromium concentration represents one major source of systematic error for this laboratory experience, as does the neglect of reabsorption of R-line photons within the respective sample. We present our results and corrections for systematic issues that make this project a more rewarding experience for students. Our result for the room temperature radiative-lifetime for the 2E term is 3.3 ± 0.1 ms; which, unlike some more recent reports1, compares favorably with that of Nelson and Sturge3.
A spatially periodic distribution of excited electronic states is produced in ruby which is placed into the interference region of two light beams derived from a pulsed laser (wavelength 530 nm). The spatially modulated population density acts like an optical grating and is detected by diffraction of a quasi-cw laser beam (488 nm). The lifetimes of the population density gratings are measured for ruby crystals with 0.05 to 5% Cr-doping and compared with the fluorescence lifetimes of the R-lines (694 nm). For the 0.05% ruby both lifetimes are nearly equal. At higher Cr-concentrations the grating decay is considerably faster than the fluorescent decay. The grating lifetimes are independent of the grating period A = 1 to 100 μm. From this observation an upper limit of the diffusion constant for energy transfer between excited electronic states in ruby is estimated.
Direct relaxation within the optically excited
E(2E) level of 700-ppm ruby is investigated in magnetic fields up to 65 kG making an angle to the
c axis. At low pump intensities, the experimental relaxation rates are found to be in excellent agreement with the theory by Blume et al., predicting a fifth-power dependence on field and a ${cos}^{3}\ensuremath{\theta}{sin}^{2}\ensuremath{\theta}$ dependence on orientation. The observed proportionality constant of 4.7 \ifmmode\times\else\texttimes\fi{} ${10}^{\ensuremath{-}19}$ ${\mathrm{G}}^{\ensuremath{-}5}$${\mathrm{s}}^{\ensuremath{-}1}$ is close to the theoretical estimate based on static experiments in ruby. When increasing optical pumping, the phonon-bath conduction is demonstrated to become the limiting process due to resonant trapping of the phonons. Following pulsed excitation, a prolonging of the decay time by a factor up to 13 is attained at excited-state concentrations ${N}^{*}\ensuremath{\sim}2\ifmmode\times\else\texttimes\fi{}{10}^{17}/{\mathrm{cm}}^{3}$. In stationary experiments, a nonlinear increase with
N∗ is observed of the equilibrium population of the upper Zeeman level of
E to over a decade above the nonbottlenecked value, corresponding to bottlenecking factors of over \ensuremath{\sim}200. Under the assumption of a constant phonon lifetime of the order of the flight time through the illuminated region, rate equations appear to consistently account for the experiments. A Gaussian-type resonance line, fully homogenized supposedly by magnetic interactions with the surrounding Al nuclei and
Cr3+ ions, explains the lattice-bath equilibration. Finally, the prospect is discussed of using the
E direct transition in ruby as a tunable phonon generator and detector with bandwidth 0.002 ${\mathrm{cm}}^{\ensuremath{-}1}$ in the practical frequency range of 2-15 ${\mathrm{cm}}^{\ensuremath{-}1}$.