Anomalous behavior of nonequilibrium excitations in $\mathrm{U}{\mathrm{O}}_2$

Ultrafast optical pump-probe studies of uranium dioxide ($\mathrm{U}{\mathrm{O}}_2$) under pressure were performed in order to better understand the material's response to ionizing radiation. Photoexcitation generates oscillations in the time-resolved reflectivity at two distinct GHz-scale frequencies. The higher-frequency mode is attributed to a coherent longitudinal acoustic mode. The lower-frequency mode does not correspond to any known excitation under equilibrium conditions. The frequency and lifetime of the low-frequency mode are studied as a function of pressure. Abrupt changes in the pressure-dependent slopes of these attributes are observed at $\sim 10$ GPa, which correlates with an electronic transition in $\mathrm{U}{\mathrm{O}}_2$. Variation of probe wavelength reveals that the low-$k$ dispersion of the low-frequency mode does not fit into either an optical or acoustic framework. Rather, we propose that this mode is related to the dynamical magnetic structure of $\mathrm{U}{\mathrm{O}}_2$. The implications of these results help account for the anomalously small volume of damage known to be caused by ionizing radiation in $\mathrm{U}{\mathrm{O}}_2$; we propose that the existence of the low-frequency mode enhances the material's transient thermal conductivity, while its long lifetime lengthens the timescale over which energy is dissipated. Both mechanisms enhance damage recovery.

Figure 1

(a) Time-resolved reflectivity of $\mathrm{U}{\mathrm{O}}_2$ at 800 nm at a variety of pressures (listed on the right in GPa) following ultrafast photoexcitation at 400 nm. (b) Fourier transforms of the traces in (a) following background subtraction. The HFM is attributed to a well-characterized LA phonon, while the LFM results from photogenerated species.

Figure 2

Frequency of the two observed modes as a function of pressure for 400-nm pump and 800-nm probe pulses. The slight hardening of the HFM is consistent with increased sound velocity at high pressure. The LFM hardens dramatically with pressure up to $\sim 10$ GPa, then levels off as pressure continues to increase. Scatter in frequency values as a function of pump fluence is within the size of the data points. No variation in the frequency is observed as a function of pump fluence.

Figure 3

LFM lifetime as a function of pressure measured with 800-nm probe following ultrafast photoexcitation at 400 nm. Lifetime decreases with increasing pressure up to $\sim 10$ GPa, at which point it levels off as pressure continues to increase. LFM lifetime represents the lower bound of the lifetime of the photogenerated species. No systematic trend is observed in the LFM lifetime as a function of pump fluence, only random scatter.

Figure 4

Phonon dispersion at low $k$. Probe wavelengths used were 700, 800, and 900 nm. Data in blue (filled) were collected at 2.2 GPa. Data in black (empty), at 12.7 GPa. Error in the data is within the size of the symbols. The frequency of the HFM varies linearly with $k$ through the origin, as expected for an acoustic mode. The LFM frequency varies nonlinearly with $k$. Wavelength-dependent refractive indices are taken from Ref. [33].