We have investigated the process of quenching of fluorescence from the
3P0 state of
Pr3+ in
La0.72Pr0.25Nd0.03F3 due to cross relaxation to Pr and Nd ions. The fluorescence decay of the ensemble of Pr ions after pulsed excitation is measured as a function of temperature in the (2-70)-K region. Cross relaxation occurs either directly or through a donor-donor transfer within the inhom-geneous line. Knowledge of the dynamics occurring within the inhomogeneous line derived from previous fluorescence line-narrowing experiments is used to analyze the cross-relaxation process from a microscopic viewpoint according to models for fluorescence decay in the presence of traps. In general the decay is consistent with an electric dipole-dipole transfer mechanism. At 2 K the decay can be well described by a direct transfer to Pr and Nd traps without donor-donor transfer. The temperature dependence of the decay at long times can be described for
35<T<70 K by the hopping model. The agreement with theory is much better if the cross relaxation to Nd ions is assumed to occur mainly via the lowest Stark level of the
4I29 manifold.
We have investigated the process of quenching of fluorescence from the
3P0 excited state of
Pr3+ in La
F3 due to cross relaxation among the Pr ions. The fluorescence decay of the ensemble of Pr ions after pulsed excitation is measured as a function of temperature for Pr concentration of 20 mol%. About 12% of the ions are in nonquenching sites and contribute an exponential component to the decay. The remainder of the decay originates from the bulk of the ions in quenching sites. Cross relaxation occurs directly in steps after transfer within the inhomogeneous line. Knowledge of the dynamics occurring within the inhomogeneous line derived from fluorescence line-narrowing experiments is used to analyze the cross-relaxation process from a microscopic viewpoint based on a recent model for fluorescence in the presence of traps. At 2 K the decay is consistent with an electric dipole-dipole transfer mechanism with a nearest-neighbor quenching rate of 8.9 \ifmmode\times\else\texttimes\fi{}
104 ${\mathrm{s}}^{\ensuremath{-}1}$ which is independent of temperature for
T<32 K and which successfully predicts the observed decay rate in Pr
F3. The temperature dependence of the decay for $8\ensuremath{\le}T\ensuremath{\le}32$ K is well described by a hopping model for excitation trapping. For
T>35 K an activation process becomes dominant consistent with the onset of cross relaxation involving the
1D2(5) and
3H6(1) states in which a 190-${\mathrm{cm}}^{\ensuremath{-}1}$ phonon is created.
Time-resolved fluorescence line narrowing is used to study spectral energy transfer in Pr
F3 and Pr
Cl3. The excitation spectrum of Pr
F3 in the region of the ${(^{3}H_{4})}_{1}\ensuremath{\rightarrow}^{3}P_{0}$ absorption indicates that several strong satellite lines accompany the main transition, their origin being ascribed to different
Pr3+ ions lying in strongly perturbed lattice sites. The spectral transfer within the inhomogeneously broadened main transiton is measured, where the energy mismatch is about 2 ${\mathrm{cm}}^{\ensuremath{-}1}$. This process is characterized by a strong ($\ensuremath{\sim}{T}^{4.3}$) dependence of the transfer probability on sample temperature. The growth of the full inhomogeneous emission profile accompanies a uniform decay of the donor line shape. A transfer process between the lowest-energy satellite ions is also measured. Here the energy mismatch is 32 ${\mathrm{cm}}^{\ensuremath{-}1}$ resulting in a weak temperature dependence for the transfer rate. Both the intraline and intersatellite transfer are characterized by simple exponential rate equations. Recent calculations of phonon-assisted transfer probabilities are compared to our measurements, with the intersatellite transfer ascribed to a "one-phonon direct" process whereas the transfer within the main transition occurs by a "one-phonon second-order" mechanism. In Pr
Cl3 the transfer times are faster than the 15-nsec response time of our detection system indicating a much stronger ion-ion coupling.
We present optical linewidth and line-shape studies of transitions of
4f impurity ions in doubly doped systems and show that interactions which involve excitation transfer between the impurity systems produce homogeneous broadening and line-shape changes in the optical transitions. In particular, we have studied the behavior of the 5985-\AA{} line of
Pr3+ in doubly doped La
F3:
Pr3+,
Nd3+ as a function of
Nd3+ concentration. The 5985-\AA{}
Pr3+ transition at 4.2\ifmmode^\circ\else\textdegree\fi{}K involves two long-lived electronic states of the
Pr3+ ion; it follows that in the singly doped La
F3:
Pr3+ the transition shows a Gaussian shape which is characteristic of inhomogeneous strain and imperfection broadening. As
Nd3+ is added to the system, a broader Lorentzian component begins to appear in the 5985-\AA{}
Pr3+ transition and eventually (at
Nd3+ concentrations of \ensuremath{\sim}1% or above) completely dominates the width and shape of the transition. The Lorentzian component of the transition line shape was successfully separated using a Fourier-transform technique and the width of this component was studied as a function of the
Nd3+ concentration. The width is found to increase linearly with increasing
Nd3+ concentration up to concentrations of 2%; at higher concentrations, the linewidth saturates and approaches an asymptotic value of \ensuremath{\sim}2 ${\mathrm{cm}}^{\ensuremath{-}1}$. We construe this homogeneous component to be evidence of a rapid short-range energy transfer mechanism between the
Pr3+ and
Nd3+ systems at the terminating
(3H6)1 level of the 5985-\AA{}
Pr3+ transition. Various mechanisms are investigated as causes of the energy transfer process; it is concluded that the spin-exchange interaction is the most probable cause for the transfer. Results of other spectroscopic measurements on the doubly doped La
F3:
Pr3+,
Nd3+ system are also presented.
Data are reported which show that the temperature dependence of fluorescence linewidths in Pr
F3 are qualitatively like those in La
F3:
Pr3+. This result is interpreted as supporting the explanation of German and Kiel for the anomalous temperature dependence of fluorescence linewidths in Pr
Cl3.