Optical Switching of $χ^{(2)}$ in Diamond Photonics

Abstract

Diamond's unique physical properties make it a versatile materials for a wide range of nonlinear and quantum photonic technologies. However, unlocking diamond's full potential as a nonlinear photonic material with non-zero second-order susceptibility $χ^{(2)}\neq0$ requires symmetry breaking. In this work, we use a nanoscale cavity to demonstrate second-harmonic generation (SHG) in diamond, and demonstrate, for the first time, that the magnitude of the diamond's effective $χ^{(2)}$ strongly depends on the electronic configuration of defects in the diamond crystal, such as nitrogen vacancy centres. The modification of $χ^{(2)}$ arises from photoionisation from the negative to neutral charge state, and is manifested by quenching of SHG upon green illumination. Toggling the green illumination allows for optical switching of the device's $χ^{(2)}$. Optical control of $χ^{(2)}$ by defect engineering opens the door for second-order nonlinear processes in diamond.

Figures (4)

• Figure 1: (a) A schematic of the fibre coupled diamond microdisk system used to demonstrate optically modulated SHG. A proposed mechanism for the observed behaviour of $\chi^{(2)}$ is shown on the right. An electric field ($E_\text{DC}$) created by negatively charged NV centres induces an effective $\chi_{\text{eff}}^{(2)}\neq0$. Photoionisation by a green wavelength ($\lambda_\text{vis}$) field combined with the IR field modifies $\chi^{(2)}_{\text{eff}}$, which leads to quenching of the SHG intensity. (b) Normalised fibre-taper transmission spectrum of the pump cavity mode, showing that the cavity is unaffected by the green illumination. (c) The red spectrum shows that SHG is observed when the IR laser is resonant with the cavity mode in (b). In blue, the SHG signal is quenched when the green laser is on. The inset shows the mechanism of SHG, where two IR photons at frequency $\omega$ combine and upconverts to a photon at frequency $2\omega$.
• Figure 2: Deterministic optical switching of $\chi^{(2)}_{\text{eff}}$ by microscope excitation of the microdisk with a green laser ($\lambda_\text{vis} = 532\,\text{nm}$). (a) Dependence of SHG strength on IR laser wavelength as it is tuned across the cavity resonance in the absence (burgundy) and presence (blue) of the visible wavelength excitation field from the microscope. A quenching of $\sim80\,\%$ is observed when the microdisk is excited with green light. Data in blue has been corrected for background PL from NV centres. (b) Toggling of $\chi^{(2)}_{\text{eff}}$ by performing successive SHG measurement with (blue) and without (burgundy) green excitation. The degree of quenching remains constant over the duration of 40 hours. Both datasets have been normalised to the mean value of the measurements with the green laser off. The dashed black lines and the shaded regions corresponds to the mean value and the standard deviation for the respective dataset.
• Figure 3: Nonlinear IR assisted charge-state conversion. (a) Fibre-taper collected NV centre PL spectrum generated by green illumination of the microdisk. Coupling a strong IR field into the microdisk suppresses the PL (in blue). (b) Quenching of SHG (pink, left axis) and suppression of NV centre PL (blue, right axis) for varying green laser power. A monotonic decrease is observed for both the PL suppression and the SHG quenching. The strong correlation between quenching of SHG and PL suppression suggests a similar underlying mechanism. The SHG data has been corrected for background PL by assuming that all PL fed cavity modes are suppressed to the same degree by the IR laser. (c) Position of the relevant energy levels of NV$^0$ and NV$^-$ drawn to scale. The burgundy and green arrows represent IR ($\lambda_{\textrm{IR}}$) and green photons ($\lambda_{\textrm{vis}}$), respectively. (left) Under $\lambda_{\textrm{IR}}$ excitation, higher-order photon ($>2$) processes are required to excite NV$^-$ and population therefore remains in NV$^-$. (right) Green photons promote population from the $^3A_2$ ground to the $^3E$ excited state of NV$^-$. From the $^3E$ state, two IR photons can pump the NV centre into the dark $^4A_2$ state of NV$^0$ (black dashed arrow).
• Figure 4: (a) Dependency of $\mathcal{R}_{\text{SHG}}$ on visible excitation wavelength, $\lambda_{\text{vis,c}}$. A monotonic increase in $\mathcal{R}_{\text{SHG}}$ is observed with increasing excitation wavelength. The green dashed line indicates the ionisation threshold for N$_{\text{s}}$, while the orange and red dashed lines represents the positions of the zero-phonon lines for NV$^{0}$ and NV$^{-}$, respectively. In Region 1, the laser is sufficiently energetic to excite N$_{\text{s}}$ alongside both the charge-states of the NV centre, resulting in large quenching of the SHG intensity (small $\mathcal{R}_{\text{SHG}}$). A plateau in the quenching is observed in Region 2 and 3, marking the regions where N$_{\text{s}}$ and NV$^0$ can no longer be optically excited, respectively. In Region 3, photoionisation of NV$^-$ is possible, but recombination from NV$^0$ is prohibited. For wavelengths longer than $637\,\textrm{nm}$, the visible laser no longer have sufficient energy to excite NV$^-$, and no quenching is observed (Region 4). The grey shaded region indicates the experimental uncertainty set by the standard deviation in measurements with no visible excitation. Horizontal grey lines are guides to the eye. (b) The relative position of the relevant energy levels for N$_\text{s}$ and both charge-states of the NV centre within the bandgap of the diamond. The sub-panels indicate which photoionisation processes are allowed in the different regions in (a).