Waveguide-integrated colour centres in silicon carbide with broadband photonic crystal reflectors for efficient readout

TL;DR

This work addresses efficient readout of spin-active V2 colour centres in 4H-SiC by integrating them into waveguides capped with broadband Dinosaur photonic crystal reflectors. Through simulations, precise nanofabrication, and cryogenic testing, the authors demonstrate a broadband reflectance window of about 64 THz with peak reflectance above 80%, facilitated by a tapered waveguide-PhC interface. Cryogenic PLE reveals comparable spectral stability to bulk emitters at low powers and saturation counts around 103–125 kcps, with a charge-resonance check enabling near-Poissonian readout statistics. Theoretical analysis suggests optical single-shot readout with fidelity exceeding 98% under ideal conditions, highlighting the potential of this platform for scalable quantum information processing in 4H-SiC.

Abstract

Spin-active colour centres in 4H silicon carbide are promising candidates as building blocks for quantum information applications. To increase the photon count rate of the emitters at low temperatures, the colour centres must be integrated into nanophotonic structures and characterised under cryogenic conditions. Here, we design and fabricate waveguide structures attached with an efficient Dinosaur photonic crystal reflector at one side. The devices show broadband reflection over a range of 60 THz with a peak reflectance above 80 %. Additionally, colour centres were integrated into these structures and characterised at cryogenic conditions. The emission was collected by a tapered-waveguide-tapered-fibre interface. Although the spectral stability of the emitters must be further improved for high excitation powers, the saturation intensity in standard PLE measurements is about 104 kcps. The count rate can be further improved to about 125 kcps with a charge-resonance check measurement scheme. To highlight the relevance of our devices, we theoretically show that these count rates enable optical single-shot readout with a fidelity exceeding 98 %.

Figures (4)

• Figure 1: Illustration and simulations of the Dinosaur reflector. (a) PhC reflector with attached waveguide section and a tapered-waveguide-tapered-fibre (TWTF) interface to collect the emitted photons of an integrated V2 colour centre. The interface between the triangular waveguide and the attached Dinosaur reflector comprises a tapered section, involving $i\in\left[0,5\right)$ unit cells with increasing lengths $a_i$ and corrugation amplitudes $A_i$. The waveguide-interfacing half of the first unit cell $i=0$ matches the waveguide width. The electric field intensity of the interference of an incident waveguide mode and its part reflected off the Dinosaur reflector is displayed in the xz-cross section below the reflector surface at a depth that matches one-third of the triangular cross section's height in y-direction. (b) Stacking identical Dinosaur unit cells (see inset) along the z-direction (for geometry parameters see main text) yields an optical band structure with a set of TE-like (blue lines) and TM-like (red lines) Bloch bands separated by TE-like (dashed areas) and complete (blue-shaded areas) bandgaps within a regime where the wave vector $k_\mathrm{z}$ exceeds the light line $\nu = c|\vec{k}|$ (black line). Here, $c$ is the speed of light. The ZPL of a V2 colour centre (yellow line) is reflected by the upper complete bandgap. Inset: Dinosaur unit cell defined by a triangular cross section and a corrugation profile proportional to a cosine function exponentiated by an even integer $e\geq 2$. (c) The reflection (red dots), transmission (light blue triangles), and scatter (dark blue diamonds) spectra of a tapered Dinosaur reflector consisting of 18 unit cells are obtained by scanning the frequency $\nu_\mathrm{P}$ of an incident probe field in FEM simulations. The reflector's spectral operating range (shaded in light red) is defined by a reflectance larger than $50\%$. Within this range, the red line indicates the mean reflectance and the area shaded in dark red represents the corresponding one standard deviation uncertainty. The orange line shows the V2 emission spectrum.
• Figure 2: Characterisation of fabricated reflector structures. (a) SEM image of a fabricated reflector. On the left side, the tapering region is visible. (b) Setup to measure the reflectance of the devices. White light passes a 90:10 beamsplitter and is sent to a tapered optical fibre which is attached to the taper of the reflector. The reflected photons are measured by a spectrometer in the second arm of the beamsplitter. Polarisation controllers are included because the beamspliter and the reflector are polarisation-dependent. (c,d) Measured (blue dots) and simulated (red triangles) reflection spectra of a fabricated waveguide-attached reflector (c) with and (d) without a tapered waveguide-reflector interface. Areas shaded in blue and red represent measured and simulated spectral operating ranges and one standard deviation uncertainties of the respective mean reflectance values, indicated by horizontal lines. d) does not show a shaded region for the experimental data as the reflectance just reaches $50\%$ at maximum values. The orange lines show the V2 emission spectrum.
• Figure 3: ODMR spectrum and PLE measurements of a V2 colour centre integrated into a waveguide attached to a Dinosaur reflector at one end. The emitted photons are collected via a TWTF interface. (a) ODMR spectrum fitted with Voigt function with its peak maximum at $(67.15\pm0.19)\,\text{MHz}$ and a contrast well above $3\%$. (b) PLE stability over 25 scans. Top panel: Jumping PLE lines for an emitter excited at a resonant power of $5\,$nW and a strong repump laser of $40\,\mu$W. Middle panel: PLE scans of the same emitter but at a much lower resonant power of $0.1\,$nW and no repump applied. Bottom panel: PLE of a bulk emitter in the same sample as stability reference. (b) Averaged PLE spectra of three single lines at a high excitation power of $12.5\,$nW close to saturation (blue) and at a lower power of $1.5\,$nW (orange). The fitted linewidths are $(58.2\pm1.3)\,$MHz and $(40.99\pm0.89)\,$MHz for the blue and orange curve, respectively.
• Figure 4: Saturation curve and charge-resonance check measurements. (a) Saturation curve measured by two different methods. Blue dots: measured and rescaled counts in a $100\,\mu$s bin under the condition that the emitter was on resonance in a preceding charge resonance check. A Poissonian distribution was fitted to the data to extract the mean value and the errorbars. The data are fitted with equation \ref{['eq:saturation']} revealing a saturation intensity of $I_s=(124.3\pm7.2)\,$kcps. Green: standard PLE scans. For each power, the brightest three lines were averaged and the maximum value is taken. The data are fitted with equation \ref{['eq:saturation']} resulting in a saturation intensity of $I_s=(103.8\pm4.2)\,$kcps. (b) Pulse sequence: initializing repump pulse using a $730\,$nm laser followed by a $20\,\mu$s long CRC check during which both resonant transitions A$_1$ and A$_2$ are enabled. Subsequently, the resonant readout pulse is applied for $100\,\mu$s. (c) Histogram of recorded events using the measurement pulse scheme shown in (b). The dashed line marks the filtering threshold of $\text{I}_\text{CRC}>5$. (d) Normalised histogram of the readout events conditioned on that during the CRC pulse more than 5 counts were detected. For the photon statistics, a Poissonian distribution was used to fit the average counts measured in the post-selected data. We obtain $\lambda=(10.5\pm3.2)\,$counts per $100\,\mu$s readout with an uncertainty of $\sqrt{\lambda}$, which corresponds to a count rate of $(105\pm32)\,$kcps. (e) Histogram of a simulated optical SSR attempt with a readout fidelity of the bright state of $98.44\%$. The calculations are based on the measured countrate for $P_\text{res}=8\,$nW. A detailed description of the technique is given in the main text.