Fig. 3: Dynamics of photoinduced second-harmonic generation.

a, c Experimental investigations of photoinduced SHG processes for (a) the leading and (c) trailing cases. They correspond to the scenarios where the generated SH power decreases and increases with the pump tuning into its resonance, respectively. Top: effective detunings of the pump and SH resonances measured by a vector network analyzer; Middle: Measured pump transmission and generated SH power; Bottom: Measured frequency offset as a function of the pump wavelength. b,d, Numerical simulations of photoinduced SHG processes for the leading (b) and trailing (d) cases. Top: simulated effective detunings of the pump and SH resonances (the gray dashed lines denote \({\delta }_{s}^{{\prime} }=0\)); Middle: simulated pump transmission and generated SH power; Bottom: simulated and theoretically predicted frequency offsets as functions of the effective detuning. e,f, Schematics of dynamic doubly resonant conditions for the leading (e) and trailing (f) cases. During pump wavelength tuning from stage I to IV or i to iv, the pump (λ_pump) stays thermally locked to the blue side of its resonance (λ_p), while the SH resonance (λ_s) varies its relative position from the generated SH wavelength (λ_SH). The leading (trailing) case corresponds to the scenario where the red-shift rate of the SH resonance is two times larger (smaller) than that of the pump resonance when tuning the pump into its resonance. The intensity of the background color indicates the power of the generated SH. At stage V and v, the pump exists the resonance.