Does ${\mathrm{V}\mathrm{O}}_2$ Host a Transient Monoclinic Metallic Phase?

Ultrafast phase transitions induced by femtosecond light pulses present a new opportunity for manipulating the properties of materials. Understanding how these transient states are different from, or similar to, their thermal counterparts is key to determining how materials can exhibit properties that are not found in equilibrium. In this paper, we reexamine the case of the light-induced insulator-metal phase transition in the prototypical, strongly correlated material ${\mathrm{VO}}_2$, for which a nonthermal Mott-Hubbard transition has been claimed. Here, we show that heat, even on the ultrafast timescale, plays a key role in the phase transition. When heating is properly accounted for, we find a single phase-transition threshold corresponding to the thermodynamic structural insulator-metal phase transition, and we find no evidence of a hidden transient Mott-Hubbard nonthermal phase. The interplay between the initial thermal state and the ultrafast transition may have implications for other transient states of matter.

Popular Summary

Light-induced phase changes in solids offer a new way to manipulate and probe materials. In this technique, ultrashort laser pulses rapidly excite electrons in the solid, triggering a quick localized temperature change. Many such experiments have claimed the emergence of novel transient states of matter, however identifying these phases often requires combining several different techniques. Here, we show that heating differences between techniques can incorrectly suggest different behaviors, thus introducing systematic errors to experiments.

To study these effects, we reexamine a claim of a light-induced transient metallic phase in the material vanadium dioxide. By comparing optical and x-ray dynamics across a wide range of timescales—from femtoseconds to seconds—we determine that heat plays a key role in this transition. When heating is accounted for, we find a single phase transition and no evidence for an alleged “hidden” transient phase.

Given the wide applicability of these techniques to many other solids, our results show that a full understanding of the thermal state is needed to interpret other claims of light-induced transient states of matter.

Figure 1

Cooling dynamics of freestanding ${\mathrm{VO}}_2$ films. (a) Optical response at 50 Hz, room temperature, and atmospheric pressure. Three fluence regimes can be distinguished: below (b) and above (c) the phase-transition threshold, and the saturation regime (d), where the optical properties no longer change with increasing fluence. Much slower relaxation dynamics can be observed when the sample is held in vacuum (e), which leads to lower values for the thresholds of the fluence regimes. Below the phase-transition threshold (f), the system shows a longer recovery time, and $5.2\mathrm{mJ}{\mathrm{cm}}^2$ is sufficient to induce the phase transition (g). The offset present at negative time delays is positive (h) if the sample remains partially metallic when the next pump pulse arrives.

Figure 2

(a) Fluence dependence transmission of the freestanding thin film of ${\mathrm{VO}}_2$ in the microsecond (red) and femtosecond (blue) regimes, measured both at atmospheric pressure (filled) and in vacuum (empty). The shift in threshold fluence due to vacuum demonstrates that accumulated heat strongly modifies the perceived threshold fluence. (b) Volume fraction of the rutile R phase as a function of temperature for the sample in vacuum. Red open circles correspond to the total volume fraction generated at the sample $10\text{−}\mu \mathrm{s}$ after excitation. The blue open circles correspond to the volume fraction that existed before the pump pulse arrived. The dashed line is the difference between these data, and it corresponds to the pump-induced volume fraction.

Figure 3

Simultaneous measurements of the structural and electronic degrees of freedom with time-resolved x-ray absorption spectroscopy. (a) Static, high-resolution x-ray spectra measured at temperatures below and above $T_c$. The absorption peaks observed correspond to the ${\pi }^{*}$, $d_{\parallel }$, and ${\sigma }^{*}$ bands. (b) Fluence dependence of the ${\pi }^{*}$ (529 eV, blue) and $d_{\parallel }$ (530.5 eV, green) states. The insulator-metal threshold is found at $3\mathrm{mJ}{\mathrm{cm}}^2$, and a maximum change is observed at $4.4\mathrm{mJ}{\mathrm{cm}}^2$. High fluences result in a decrease in signal as the material is heated until it is already in the metallic phase upon excitation. At the largest fluences, a new XAS signal is observed, resulting from dynamics of the metallic phase. (c) Comparison of transient and thermal XAS spectra at the oxygen $K$-edge. The differential transient spectrum (solid red line) measured at a delay of 400 ps shows an excellent agreement with that obtained from thermal changes (dashed line). The metallic transient response (purple) has a completely different spectrum. (d) The temporal dynamics of the $d_{\parallel }$ (530.5 eV, green) states show changes rising within the 70-ps experimental resolution and show no subsequent dynamics. Note that this measurement was performed on a different region of the sample, which showed a smaller signal than (a)–(c).

Figure 4

Fluence dependence of the rutile phase volume fraction obtained from the XAS response at 530.5 eV (filled red circles) and 529 eV (blue open circles). (a) Maximum rutile phase volume fraction deduced from the XAS spectra measured 400 ps after photoexcitation. This fraction includes both the transiently induced change and the quasistatic thermal change. (b) Volume fraction obtained 4 ms after excitation, which shows the quasistatic volume fraction of the rutile phase. The dashed lines show the data smoothed. The shaded area corresponds to the region in which a transient phase transition is observed.

Figure 5

(a) Normalized change in transmissivity of a freestanding ${\mathrm{VO}}_2$ sample at 1-ps time delay measured at 100 Hz in air for different probe wavelengths. Below- and above-gap probing show the same threshold fluence, indicating that both the electronic and structural transitions occur at the same critical fluence. (b) Coherent phonon amplitude as a function of fluence measured at 690 nm from a single-crystal sample, compared with the reflectivity change measured at 1 ps, indicating the threshold measured in the visible spectral range corresponds to the structural threshold. The dashed green line is a linear fit to the low-fluence reflectivity.

Figure 6

Fluence dependence of the reflectivity change across the insulator-metal phase transition measured on a single crystal starting at an initial temperature of 78 K (blue circles) and 295 K (red circles), respectively. The difference in threshold can be explained by the thermal energy difference required to heat the sample from 78 K to 295 K.