Inhomogeneous Photosusceptibility of ${\mathrm{VO}}_2$ Films at the Nanoscale

Pump-probe nano-optical experiments were used to study the light-induced insulator to metal transition (IMT) in thin films of vanadium dioxide (${\mathrm{VO}}_2$), a prototypical correlated electron system. We show that inhomogeneous optical contrast is prompted by spatially uniform photoexcitation, indicating an inhomogeneous photosusceptibility of ${\mathrm{VO}}_2$. We locally characterize temperature and time dependent variations of the photoexcitation threshold necessary to induce the IMT on picosecond timescales with hundred nanometer spatial resolution. We separately measure the critical temperature $T_L$, where the IMT onsets and the local transient electronic nano-optical contrast at the nanoscale. Our data reveal variations in the photosusceptibility of ${\mathrm{VO}}_2$ within nanoscopic regions characterized by the same critical temperature $T_L$ where metallic domains can first nucleate.

Figure 1

Nano-optical study of the light induced IMT in a ${\mathrm{VO}}_2$ film. All data in Fig. 1 were obtained with an initial temperature $T_i=330\mathrm{K}$. (a) Schematic of the experiment. A horizontally polarized pump beam (red pulses) initiates the IMT. We use a vertically polarized $8\mathrm{\mu }\mathrm{m}$ wavelength pulsed probe (purple pulse) focused on an AFM tip to monitor the local near-field amplitude, $S$, at a variable time delay, $\mathrm{\Delta }\mathrm{t}$, after photoexcitation. In our experiments, temperature dependent data is obtained by changing the initial (equilibrium) temperature of the sample, $T_i$, with a heat stage. (b)–(f), Images of $S$, using 5.3 nJ of photoexcitation power, obtained (b), before photoexcitation: $\mathrm{\Delta }\mathrm{t}=-6.4\mathrm{ps}$, and (c)–(f), after photoexcitation: (c) $\mathrm{\Delta }\mathrm{t}=23.6$, (d) $\mathrm{\Delta }\mathrm{t}=53.6$, (e) $\mathrm{\Delta }\mathrm{t}=203.6$, (f) $\mathrm{\Delta }\mathrm{t}=303.6\mathrm{ps}$. (g) Dynamics of the pump induced change of the near-field amplitude $\mathrm{\Delta }S$ obtained at two representative locations, Location 1 and location 2, where values of $S$ are near the extremes of high and low contrast observed in panel (f), respectively. The green, blue, and gray shaded regions mark the first, second, and third stages, respectively, of the three-step response to photoexcitation as described in the main text. (h) Data of $\mathrm{\Delta }S$ are shown against the photoexcitation energy at the same two locations investigated in panel (g). Data collected at $\mathrm{\Delta }\mathrm{t}=250\mathrm{fs}$ are shown with open circles, while data obtained at $\mathrm{\Delta }\mathrm{t}=4\mathrm{ps}$ are shown with filled diamonds. Note that systematic inhomogeneity observed at $\mathrm{\Delta }\mathrm{t}=250\mathrm{fs}$ and above $Q_c$ likely derives from carriers that have abruptly transitioned to the metallic state (see Supplemental Material [19], Sec. 2.4 for details and discussion).

Figure 2

Spatially resolved and temperature dependent photo-excitation energy thresholds. All data were obtained at the time delay $\mathrm{\Delta }\mathrm{t}=600\mathrm{ps}$. (a) and (b) Raw data showing the pump-induced change of the near-field amplitude, $\mathrm{\Delta }S$, as a function of the applied photoexcitation energy, $Q$, at two locations in our film, $L_1$ (red) and $L_2$ (blue), which are separated by about $1\mathrm{\mu }\mathrm{m}$ in real space (inset) for a series of initial temperatures, $T_i$. (a) The data are offset vertically for clarity with the dashed black line indicating $\mathrm{\Delta }S=0$ at each initial temperature. The red and blue solid lines are fits used to determine the thresholds, which are marked by the red and blue arrows below each trace (see main text). The inset shows an image of the examined area, obtained at $T_i=329\mathrm{K}$ $\mathrm{\Delta }\mathrm{t}=600\mathrm{ps}$, $Q=5\mathrm{nJ}$. (b) The photoexcitation threshold, $Q_c$ is plotted against $T_i$. The solid lines are guides to the eye. The dashed lines show extrapolations. (c) The measured trend of ${\mathrm{c}}^{*}(T_i)$. The gray shaded areas indicate a temperature range of 333–355 K, where the transition region is observed in independent resistivity measurements (Fig. S1 [19]).

Figure 3

Speed limits of the IMT. (a) Dynamics of ΔS obtained at a single location for a series of photo-excitation energies, $Q$, at $T_i=332\mathrm{K}$. The values of $Q$ used in each measurement are indicated on the right-hand side of the plot. The green, blue, and gray shaded regions mark the first, second and third stages, respectively, of the three-step response following photoexcitation described in the main text. (b) Data from (a) are plotted against $Q$ at selected time delays. The data are vertically offset for clarity; the dashed black lines indicate $\mathrm{\Delta }S=0$ for each time delay. (c) The bottom axis indicates the values of $Q_c(x,\mathrm{\Delta }t)$ extrapolated as indicated in panel (b) at each location, $x$, and time delay, $\mathrm{\Delta }t$. The gray shaded region marks delays where nonthermal effects (attributed to carrier injection and relaxation—i.e., first two stages marked by the green and light blue regions in panel (b) obscure observations of $Q_c$. The solid red and blue lines are calculations from nucleation theory described in Sec. S1.2 [19].