A large number of small but sharp changes (increases and decreases) of the R 1 fluorescence in ruby are observed as a function of applied magnetic field between 2 K and 77 K. At liquid helium temperatures, these signals have linewidths much narrower than the inhomogeneous linewidth (∼0.1 cm -1 ) of the R 1 line. All of these signals appear at the fields where the Zeeman splittings in the excited state \bar E ( 2 E ) and the ground state 4 A 2 satisfy an energy-matching condition for simultaneous transitions of two or more ions. Two distinct mechanisms are found to be responsible for the signals: spin-spin cross-relaxation between \bar E ( 2 E ) and 4 A 2 , and resonant self-trapping of the R 1 line. From the temperature dependence of the linewidths of the signals, thermal broadening of \bar E ( 2 E ) is measured below 77 K. Similar signals are observed in the N lines due to energy transfer from isolated ions to pairs.
A large number of small but sharp changes (increases and decreases) of the R 1 fluorescence in ruby are observed as a function of applied magnetic field between 2 K and 77 K. At liquid helium temperatures, these signals have linewidths much narrower than the inhomogeneous linewidth (∼0.1 cm -1 ) of the R 1 line. All of these signals appear at the fields where the Zeeman splittings in the excited state \bar E ( 2 E ) and the ground state 4 A 2 satisfy an energy-matching condition for simultaneous transitions of two or more ions. Two distinct mechanisms are found to be responsible for the signals: spin-spin cross-relaxation between \bar E ( 2 E ) and 4 A 2 , and resonant self-trapping of the R 1 line. From the temperature dependence of the linewidths of the signals, thermal broadening of \bar E ( 2 E ) is measured below 77 K. Similar signals are observed in the N lines due to energy transfer from isolated ions to pairs.
Fluorescence studies of the R-line emission of ruby laser rods over the temperature range of 77 to 300 K have disclosed several new features. In particular, for most of the ruby rods, the fluorescent decay time, τ, was not strictly exponential. Effects of this kind have been observed in the fluorescent decay of the 4F3/2 level of Nd3+ in YAG (yttrium aluminum garnet) and in Nd-doped glasses. In both cases, the nonexponential decay was interpreted as evidence for inhomogeneous line broadening effects, among which are strains, ion clustering, and other defects that modify the crystal field at the ion site. Our observations of nonexponential fluorescent decay in ruby are indicative of inhomogeneous line broadening effects in lightly doped (0/05%) ruby. In fact, inhomogeneous line broadening was observed in ruby at liquid helium temperatures, in the spectroscopic studies of A. L. Schawlow.
Fluorescence-line-narrowing and time-resolved studies have been conducted in ruby crystals and powders, in concentrations varying from 0.025 to 0.9 at.%. The resonance fluorescence component of the R1 line at low temperatures is observed to increase dramatically as the laser is tuned from line center toward the low-energy wing of the inhomogeneously broadened absorption. This is presumed to indicate the presence of weakly coupled pairs in the wings of the line. Intraline and interline (${R}_{1}\ensuremath{\rightarrow}{N}_{1,2}$) energy transfer have also been studied, and an elaboration of earlier work is given. Time-resolved selective-excitation scans have revealed numerous channels in the vicinity of the R lines by which the different pairs can be directly excited. These are assumed to be Cr ions near enough to a pair to be in a highly perturbed environment. Various experimental results are shown to be consistent with a model of rapid transfer between resonant Cr ions which feeds the pairs via a weak, nonresonant coupling to these perturbed ions. Macroscopic strain broadening is also shown to be favored over microscopic broadening, and the implications for Anderson localization in ruby are discussed.
Bichromatic excitation of the R1 line in ruby at liquid-helium temperature is considered. The frequency difference of the spectral components of the driving field is resonant with the energy gap between the anticrossed spin sublevels +1/2 and +3/2 in the ground state 4A2. The condition of the spin coherence trapping in the nonabsorbing state is analyzed. Population trapping is revealed in the reduced spontaneous emission from the excited state and in large spin magnetization of the ground state Cr3+ precessing with the frequency difference of the spectral components of the driving field. This magnetization can be detected by a pickup coil. The lifetime of the induced magnetization is strongly dependent on the ratio of the lifetime of the excited state $\ifmmode \bar{E}\else \={E}\fi{}$ and the spin coherence decay time (T2)spin.
We have found that the amplitudes of the optical FID, two-pulse photon echoes and stimulated photon echoes in ruby measured as a function of applied magnetic field exhibit sharp increases at the magnetic fields where two Zeeman splittings in the ground state 4 A 2 or in the excited state \bar E ( 2 E ) are harmonically related. The phenomena are confirmed to be inherent in the Stark (or frequency) switching experiment where the sample is subjected to the cw irradiation, and the origin of the sharp increases be the change of the steady state population in 4 A 2 of the nonresonant Cr ions caused by a combined action of the cw optical pumping and the spin-spin cross relaxation. The mechanism involved is successfully explained by the analysis based on the rate equations.
The theory of imprisoned radiation developed by Holstein and recently extended by Scherr is adapted to apply to the luminescence decay in a solid. In addition to including a correction for losses from the excited level, it is necessary to consider reflections at the sample surfaces. Reflections are calculated approximately for a slab. The single Gaussian absorption function used in the reabsorption theory in lieu of an actual absorption function for the imprisoned radiation has the same integrated absorption I as the actual function and has a height equal to 2I times the integral of the square of the actual function. Calculations are described for the ruby R lines, including the case where the Zeeman components are separated in a magnetic field. Experimental measurements of decay times at 77\ifmmode^\circ\else\textdegree\fi{}K of the ruby R lines are presented for samples having a Cr+3 concentration of approximately 1019 ${\mathrm{cm}}^{\ensuremath{-}3}$ and thickness from 0.6 to 13 mm. Magnetic fields up to 20 kG are used. The various decay times, lying between 5 and 12 msec, are predicted within about 5% by the theory, with the help of the approximate absorption corrections and of Zeeman component intensities given by Sugano and Tanabe. The R1 line is used for most comparisons. Intensity increases of up to 100% in the steady-state R-line emission in a 20-kG magnetic field are shown to be accounted for by approximate treatments related to the decay-time theory. No change of transition probabilities with magnetic field is indicated by the foregoing results. Further observations suggest that the channels leading from the excitation levels for both single ions and pairs are affected by magnetic field, however. These observations include changes in the relative intensities of various pair lines when the field is applied, and a field-dependent difference in the emission intensities produced with blue and green excitation. Field-induced changes in pumping light absorption, if present, are insufficient to account for these variations.
Studies of the resonant interaction of radiation from a liquid-nitrogen-cooled ruby laser with a ruby crystal in the temperature range 4.2°–100°K are described. The interaction was measured by observation of the change which occurs in the Cr3+ ground state EPR absorption on laser irradiation. At low radiation intensities, the temperature dependence of this change is shown to be determined by the well-known thermal tuning behavior of the R lines. Precise measurement of the relative laser and ruby-sample absorption frequencies by this technique of double-resonance spectroscopy is described. Under optical saturation conditions, additional temperature-dependent pumping effects occur at low temperatures. In particular, as the ruby temperature is increased from 4.2° to 40°K, the laser-induced change in EPR absorption is observed to undergo a large increase. This increase cannot be explained by a relative thermal shift of the laser and absorption frequencies which is known to be very small in this temperature range. The phenomena are attributed in part to the temperature-dependent homogeneous linewidth which determines the widths of the holes burnt into the inhomogeneous strain-broadened R lines. Calculations indicate that the related temperature-dependent thermalization process which redistributes excited Cr3+ among the 2E levels quantitatively accounts for most of the increase.