Quantum Control of the Tin-Vacancy Spin Qubit in Diamond

Group-IV color centers in diamond are a promising light-matter interface for quantum networking devices. The negatively charged tin-vacancy center ($\mathrm{Sn}\mathrm{V}$) is particularly interesting, as its large spin-orbit coupling offers strong protection against phonon dephasing and robust cyclicity of its optical transitions toward spin-photon-entanglement schemes. Here, we demonstrate multiaxis coherent control of the $\mathrm{Sn}\mathrm{V}$ spin qubit via an all-optical stimulated Raman drive between the ground and excited states. We use coherent population trapping and optically driven electronic spin resonance to confirm coherent access to the qubit at 1.7 K and obtain spin Rabi oscillations at a rate of $\mathrm{\Omega }/2\pi =19.0(1)\mathrm{MHz}$. All-optical Ramsey interferometry reveals a spin dephasing time of $T_2^{*}=1.3(3)\mu \mathrm{s}$, and four-pulse dynamical decoupling already extends the spin-coherence time to $T_2=0.30(8)\mathrm{ms}$. Combined with transform-limited photons and integration into photonic nanostructures, our results make the $\mathrm{Sn}\mathrm{V}$ a competitive spin-photon building block for quantum networks.

Popular Summary

The basic building block of a quantum communication network requires a matter-based quantum bit (qubit) for fast control of local information, a long-lived memory qubit for storage, and an efficient interface to light-based qubits. Optically controllable spins embedded within a solid-state medium make natural candidates for such a building block, and atomic-scale defects in diamond known as color centers unify all the requirements in a single piece of hardware. Here, we demonstrate the first full quantum control over a tin-vacancy spin qubit, one of a group of diamond color centers identified to offer the right combination of optical, spin, and nuclear properties.

Leveraging the tin-vacancy center’s excellent optical interface to its spin degrees of freedom, we implement high-fidelity multiaxis qubit operations using lasers. We show that a qubit coherence time of 0.3 ms is readily available, and we verify strong coupling of our qubit to a nuclear isotope, a natural candidate for quantum memory. Further, all this is achieved at a temperature 1.7 K, which offers a feasibility advantage over silicon and germanium alternatives that necessitate millikelvin refrigeration for comparable performance.

Taken with recent work demonstrating near-ideal single photons and nanophotonic integration, our results are a milestone, establishing this qubit as a viable building block for quantum network hardware.

Figure 1

$\mathrm{Sn}\mathrm{V}$ and Raman drive. (a) Lilac panels: energy levels of the $\mathrm{Sn}\mathrm{V}$ split by spin-orbit coupling (left), Zeeman splitting (middle), and hyperfine splitting (right). Blue panel: qubit defined by the $|\downarrow ⟩$ and $|\uparrow ⟩$ electronic levels. An optical lambda scheme is defined by two optical fields ${\omega }_1$ and ${\omega }_2$ detuned by $\mathrm{\Delta }$ from the excited state $|E⟩$. The Raman drive frequency ${\omega }_R={\omega }_1\text{−}{\omega }_2$ is the energy offset between the two fields. (b) $|E⟩$ population plotted in orange circles as a function of ${\omega }_R$ for $\mathrm{\Delta }=0$ and 200 mT magnetic field strength applied at 54.7° relative to the $\mathrm{Sn}\mathrm{V}$ symmetry axis. The pulse sequence consists of an approximately $10\text{−}\mu \mathrm{W}$ green stabilization pulse for $50\mu \mathrm{s}$, $B2$ reset pulse for $30\mu \mathrm{s}$, and Raman drive with $p=40(4)\mathrm{nW}$ for $2\mu \mathrm{s}$. The solid curve is a three-level model describing coherent population trapping. (c) $|\downarrow ⟩$ population plotted in orange circles as a function of ${\omega }_R$ for $\mathrm{\Delta }=600\mathrm{MHz}$ at 204-mT magnetic field strength. After the stabilize and reset pulses, the sequence includes an initialization pulse for $30\mu \mathrm{s}$, a Raman drive with $p=40(4)\mathrm{nW}$ for $1\mu \mathrm{s}$, and a readout pulse on $A1$ for $30\mu \mathrm{s}$. The solid curve is a two-Lorentzian fit resulting in an average linewidth of 900(200) kHz and spin resonances split by 42.6(4) MHz.

Figure 2

Multiaxis coherent spin-qubit control. (a) $|\downarrow ⟩$ population (orange circles) as a function of the Raman drive duration $T$ with the pulse sequence shown at the top. The Raman drive is applied with $\mathrm{\Delta }/2\pi =1.2\mathrm{GHz}$ and $p=650(70)\mathrm{nW}$. The black curve is a fit to a two-level model under a master-equation formalism (see Supplemental Material [61] Sec. 3). Inset: $\mathrm{\Omega }/2\pi$ as a function of $p/(\mathrm{\Delta }/2\pi )$ with a linear fit to the data (solid curve). Here, $\mathrm{\Delta }/2\pi$ and $p$ are varied from 300 to 1200 MHz and 40 to 650 nW, respectively. (b) Pulse sequence (top) with one $\pi /2$ pulse about $x$ and a second about an axis rotated by an azimuthal angle $\varphi$ from the $x$ axis. The $\pi /2$ pulse duration is determined from Rabi measurements taken with $\mathrm{\Delta }/2\pi =300\mathrm{MHz}$ and $p=260(30)\mathrm{nW}$. Illustrated on the Bloch spheres are trajectories for $\varphi =0$ (left), $\varphi =\pi /4$ (center), and $\varphi =\pi$ (right). The $|\downarrow ⟩$ population (orange circles) is plotted as a function of $\varphi$. The solid curve is a cosine function.

Figure 3

Ramsey interferometry. (a) Ramsey pulse sequence with two $\pi /2$ pulses separated by a delay time $\tau$ (top). The phase of the second pulse is swept according to $\varphi =\tau {\omega }_S$, where ${\omega }_S/2\pi =5\mathrm{MHz}$. The color indicates $|\downarrow ⟩$ population plotted as a function of $\tau$ and the two-photon detuning, $\delta$. Dotted white curves provide a guide to the eye for the expected $|\downarrow ⟩$ population recovery. Fits of the form $a\exp [-(\tau /T_2^{*}{)}^2]\sin ({\omega }_{\text{Ramsey}}\tau +\alpha )$ at each $\delta$ yield an average $T_2^{*}=1.3(3)\mu \mathrm{s}$ and ${\mathrm{\Delta }}_{\mathrm{ac}}/2\pi =3.3(5)\mathrm{MHz}$ (see Supplemental Material [61] Sec. IX). (b) Line cut at $\delta /2\pi =-1\mathrm{MHz}$ indicated by an arrow in (a), where the $|\downarrow ⟩$ population (orange circles) is measured as a function of $\tau$. The black curve corresponds to the fit described in (a), where ${\omega }_{\text{Ramsey}}/2\pi =7.27(3)\mathrm{MHz}$, $T_2^{*}=1.4(3)\mu \mathrm{s}$, $a=0.054(4)$, and $\alpha =0.16(9)$. The purple shaded region is bounded by the envelope $\pm a\exp [-(\tau /T_2^{*}{)}^2]$.

Figure 4

Dynamical decoupling. (a) Decoupling pulse sequence (top) with two implementations: (left orange panel) Hahn echo, (right purple panel) CPMG-2. The phase $\varphi$ of the second $\pi /2$ pulse is variable. In both panels, color indicates the $|\downarrow ⟩$ population shown as a function of the total decoupling time $\tau$ and phase $\varphi$. (b) Visibility $a/b$ obtained from fitting the function $a\cos (\varphi )+b$ to the data shown in (a) at each delay time $\tau$ plotted as a function of $\tau$. Hahn echo data (orange circles) are fitted to the function $v_0\exp [-(\tau /T_2{)}^n]+v_{\infty }$, where $v_0=0.26(1)$, $v_{\infty }=0.013(5)$, $n=3.7(4)$, and $T_2=28.3(6)\mu \mathrm{s}$ (solid orange curve). CPMG-2 data (purple circles) are fitted to the same function with $v_0=0.11(1)$, $n=2.1(9)$, $T_2=0.19(2)\mathrm{ms}$, and $v_{\infty }$ is fixed to 0 (solid purple curve). CPMG-4 data (see Supplemental Material [61] Sec. X) are plotted (pink circles) and fit to the same function with $v_0=0.044(7)$, $n=1.2(6)$, $T_2=0.30(8)\mathrm{ms}$, and $v_{\infty }$ is fixed to 0 (solid pink curve).