Nature Physics  Letter
Fewsecondlong correlation times in a quantum dot nuclear spin bath probed by frequencycomb nuclear magnetic resonance spectroscopy
 A. M. Waeber^{1}^{, }
 M. Hopkinson^{2}^{, }
 I. Farrer^{3}^{, }
 D. A. Ritchie^{3}^{, }
 J. Nilsson^{3}^{, }
 R. M. Stevenson^{4}^{, }
 A. J. Bennett^{4}^{, }
 A. J. Shields^{4}^{, }
 G. Burkard^{5}^{, }
 A. I. Tartakovskii^{1}^{, }
 M. S. Skolnick^{1}^{, }
 E. A. Chekhovich^{1}^{, }
 Journal name:
 Nature Physics
 Volume:
 12,
 Pages:
 688–693
 Year published:
 DOI:
 doi:10.1038/nphys3686
 Received
 Accepted
 Published online
One of the key challenges in spectroscopy is the inhomogeneous broadening that masks the homogeneous spectral lineshape and the underlying coherent dynamics. Techniques such as fourwave mixing and spectral holeburning are used in optical spectroscopy^{1, 2, 3}, and spinecho^{4} in nuclear magnetic resonance (NMR). However, the highpower pulses used in spinecho and other sequences^{4, 5, 6, 7, 8} often create spurious dynamics^{7, 8} obscuring the subtle spin correlations important for quantum technologies^{5, 6, 9, 10, 11, 12, 13, 14, 15, 16, 17}. Here we develop NMR techniques to probe the correlation times of the fluctuations in a nuclear spin bath of individual quantum dots, using frequencycomb excitation, allowing for the homogeneous NMR lineshapes to be measured without highpower pulses. We find nuclear spin correlation times exceeding one second in selfassembled InGaAs quantum dots—four orders of magnitude longer than in strainfree III–V semiconductors. This observed freezing of the nuclear spin fluctuations suggests ways of designing quantum dot spin qubits with a wellunderstood, highly stable nuclear spin bath.
Subject terms:
At a glance
Figures
Main
Pulsed magnetic resonance is a diverse toolkit with applications in chemistry, biology and physics. In quantum information applications, solid state spin qubits are of great interest and are often described by the socalled central spin model, where the qubit (central spin) is coupled to a fluctuating spin bath (typically interacting nuclear spins). Here microwave and radiofrequency (rf) magnetic resonance pulses are used for the initialization and readout of a qubit^{18}, dynamical decoupling^{5} and dynamic control^{6} of the spin bath.
However, the most important parameter controlling the central spin coherence^{9, 11, 19}—the correlation time τ_{c} of the spin bath fluctuations—is very difficult to measure directly. The value of τ_{c} is determined by the spin exchange (flipflops) of the interacting nuclear bath spins. By contrast, pulsed NMR reveals the spin bath coherence time T_{2}, which characterizes the dynamics of the transverse nuclear magnetization^{7, 8, 20} and is much shorter than τ_{c}. The problem is further exacerbated in selfassembled quantum dots, where quadrupolar effects lead to inhomogeneous NMR broadening exceeding 10 MHz (refs 21,22), so that pulsed NMR requires practically unattainable rf field amplitudes exceeding 1 T.
Here we develop an alternative approach to NMR spectroscopy: we measure noncoherent depolarization of nuclear spins under weak noiselike rf fields. Contrary to intuitive expectation, we show that such measurement can reveal the full homogeneous NMR lineshape describing the coherent spin dynamics. This is achieved by employing rf excitation with a frequencycomb profile (widely used in precision optical metrology^{23}). We then exploit nonresonant nuclear–nuclear interactions: the homogeneous NMR lineshape of one isotope measured with frequencycomb NMR is used as a sensitive noninvasive probe of the correlation times τ_{c} of the nuclear flipflops of the other isotope. Although initial studies^{9, 17, 19} suggested τ_{c} ~ 100 μs for nuclear spins in III–V semiconductors, it was recently recognized^{20, 24, 25} that quadrupolar effects may have a significant impact in selfassembled quantum dots. Here, for the first time, we obtain a quantitative measurement of extremely long τ_{c} ≳ 1 s, revealing strong freezing of the nuclear spin bath due to the inhomogeneous strain—a crucial advantage for quantum information applications of selfassembled quantum dots.
The experiments were performed on individual neutral selfassembled InGaAs/GaAs quantum dots at magnetic field B_{z} = 8 T. All measurements of the nuclear spin depolarization dynamics employ the pumpdepolarizeprobe protocol shown in Fig. 1a. Here we exploit the hyperfine interaction of the nuclei with the optically excited electron^{17, 21, 22} both to polarize the nuclei (pump pulse) and to measure the nuclear spin polarization in terms of the Overhauser shift ΔE_{hf} in the quantum dot photoluminescence spectrum (probe pulse). The rf magnetic field depolarizing nuclear spins is induced by a small copper coil. (Further experimental details can be found in Methods and Supplementary Note 1.)
All isotopes in the studied dots possess nonzero quadrupolar moments. Here we focus on the spin I = 3/2 nuclei ^{71}Ga and ^{75}As. The straininduced quadrupolar shifts result in an inhomogeneously broadened NMR spectrum^{21, 22}, as shown schematically by the green line in Fig. 1b. The spectrum consists of a central transition (CT) −1/2 ↔ +1/2 and two satellite transition (ST) ±1/2 ↔ ±3/2 peaks. All nuclei in the dot are coupled by dipole–dipole interactions: an exact description of such a system is a very complex manybody quantum mechanical problem. Here we use a simplified semiclassical description where nuclear–nuclear interactions are taken into account by introducing a phenomenological homogeneous broadening of each nuclear spin transition. Thus, the NMR spectrum in Fig. 1b is an inhomogeneous distribution (with width Δν_{inh}) of individual nuclear spin transitions (shown by the red lines) with much smaller homogeneous linewidth Δν_{hom}.
Radiofrequency excitation is considered to be weak when nuclear spin precession is slower than transverse relaxation, resulting in noncoherent depolarization (that is without Rabi oscillations) of the nuclei^{26}. To make such depolarization sensitive to the homogeneous NMR lineshape, we use rf excitation with a frequencycomb spectral profile. As shown in Fig. 1b (black line), the frequency comb has a mode spacing of f_{MS} and a total comb width Δν_{comb} exceeding Δν_{inh}. The key idea of the frequencycomb technique is described in Fig. 1c, d, where two possible cases are shown. If the spacing is small (f_{MS} Δν_{hom}, Fig. 1c) all nuclear transitions are excited by a large number of rf modes. As a result, all nuclear spins are depolarized at the same rate and we expect an exponential decay of the total nuclear spin polarization. In the opposite case of large mode separation (f_{MS} > Δν_{hom}, Fig. 1d), some of the nuclear transitions are out of resonance and are not excited (for example, the one shown by the dashed red line). As a result we expect a sloweddown nonexponential nuclear depolarization. The experiments are performed at different f_{MS}; the f_{MS} for which a slowdown in depolarization is observed gives a measure of the homogeneous linewidth Δν_{hom}.
Experimental demonstration of this technique is shown in Fig. 2a. The Overhauser shift variation ΔE_{hf} of ^{71}Ga is shown as a function of the depolarizing rf pulse duration t_{rf} for different f_{MS}. For small values of f_{MS} = 80 and 435 Hz an exponential depolarization is observed. However, when f_{MS} is increased the depolarization becomes nonexponential and slows down significantly. The detailed dependence ΔE_{hf}(t_{rf}, f_{MS}) is shown as a colourcoded plot in Fig. 2b. The threshold value of f_{MS} (marked with a white arrow) above which the nuclear spin dynamics becomes sensitive to the discrete structure of the frequency comb provides an estimate of Δν_{hom} ~ 450 Hz. Such a small homogeneous linewidth is detected in NMR resonances with inhomogeneous broadening of Δν_{inh} ~ 6 MHz (ref. 21), demonstrating the resolution power of frequencycomb noncoherent spectroscopy.
The information revealed by frequencycomb spectroscopy is not limited to rough estimates. An accurate determination of the linewidth and the shape of the tails of the full homogeneous lineshape is achieved with spin dynamics modelling based on rate equations (see details in Methods and Supplementary Note 2). We use the following twoparameter phenomenological model for the homogeneous lineshape:
where Δν_{hom} is the homogeneous fullwidth at halfmaximum and k is a rolloff parameter that controls the tails (the behaviour of L(ν) at large ν). For k = 1 the lineshape corresponds to Lorentzian, whereas for k ∞ it tends to Gaussian: in this way equation (1) seamlessly describes the two most common lineshapes. Using Δν_{hom} and k as parameters we calculate the model ΔE_{hf}(t_{rf}, f_{MS}) dependence and fit it to the experimental ΔE_{hf}(t_{rf}, f_{MS}) to find a phenomenological description of the homogeneous NMR lineshape in selfassembled quantum dots.
The solid line in Fig. 3a shows the bestfit lineshape (Δν_{hom} ≈ 221 Hz and k ≈ 1.67) for the measurement shown in Fig. 2a, b. The dashed and dasheddotted lines in Fig. 3a show for comparison the Lorentzian (k = 1) and Gaussian (k ∞) lineshapes with the same Δν_{hom}. The difference in the lineshape tails is seen clearly in Fig. 3b, where a logarithmic scale is used. The model ΔE_{hf}(t_{rf}, f_{MS}) dependence calculated with the bestfit parameters is shown in Fig. 3c and with lines in Fig. 2a—there is excellent agreement with experiment. By contrast, modelling ΔE_{hf}(t_{rf}, f_{MS}) with Lorentzian (Fig. 3d) and Gaussian (Fig. 3e) lineshapes shows a pronounced deviation from the experiment, demonstrating the excellent sensitivity of the frequencycomb spectroscopy to the tails of the homogeneous spectral lineshape.
We have also performed frequencycomb NMR spectroscopy on ^{75}As nuclei (Fig. 4a). Despite their larger inhomogeneous broadening, Δν_{inh} ~ 18 MHz, the model fitting reveals even smaller Δν_{hom} ≈ 117 Hz and k ≈ 1.78. From the Δν_{hom} derived here we can estimate the nuclear spin coherence times T_{2} ≈ 1/(πΔν_{hom}) ~ 1.4 and 2.7 ms for ^{71}Ga and ^{75}As, respectively. These values are in good agreement with the corresponding spinecho T_{2} ≈ 1.2 and 4.3 ms derived in ref. 20. However, spinecho could be measured only on the central transitions, for which Δν_{inh} is relatively small. Moreover, pulsed NMR does not allow determination of the full homogeneous lineshape, which for dipole–dipole interactions typically has a ‘tophat’like (Gaussian) profile^{27}. And, most importantly, owing to parasitic effects such as ‘instantaneous diffusion’^{7} and spin locking^{8}, pulsed NMR does not reveal the characteristic correlation time τ_{c} of the spin exchange flipflops. In particular, owing to the ‘instantaneous diffusion’, the slow flipflop dynamics become obscured by the much faster and trivial spin dynamics arising from a combined effect of secular spin–spin interactions and strong rf pulses^{7}.
As we now show, the nonGaussian lineshapes can be understood and τ_{c} can be derived using experiments with two frequency combs exciting nuclei of two isotopes (^{75}As and ^{71}Ga). We again excite ^{75}As nuclei with a frequency comb to measure their homogeneous lineshape. The difference is that now we simultaneously apply a second comb exciting the ^{71}Ga spins. Importantly, in this experiment the ^{71}Ga nuclei are first fully depolarized after the optical nuclear spin pumping and before the twocomb excitation (see Supplementary Fig. 3)—in this way the excitation of ^{71}Ga has no direct effect on the measured hyperfine shift ΔE_{hf}. By contrast, it leads to ‘heating’ of the ^{71}Ga spins, which has only an indirect effect on ΔE_{hf} by changing the ^{75}As lineshape via dipolar coupling between ^{71}Ga and ^{75}As spins. The result of the twocomb experiment is shown in Fig. 4b: a clear increase of Δν_{hom} for ^{75}As is observed. From model fitting we find that ^{71}Ga ‘heating’ leads to a three times broader homogeneous linewidth Δν_{hom} ≈ 355 Hz of ^{75}As and its homogeneous lineshape is modified towards Gaussian, observed as an increase in k ≈ 2.32.
To explain this result we note that the NMR lineshape is a statistical distribution of NMR frequency shifts of each nucleus produced by its dipolar interaction with all possible configurations of the neighbouring nuclear spins. However, in a frequencycomb experiment the evolution of the ^{75}As spin is limited in time by its depolarization. If the nuclear spin environment of each ^{75}As nucleus does not go through all possible configurations during the ^{75}As characteristic depolarization time, the frequency shifts are effectively static, and hence are eliminated from the lineshape, as for any other inhomogeneous broadening.
Thus, we conclude that the narrowed, nonGaussian (k ≈ 1.6–1.8) homogeneous NMR lineshape arises from the ‘snapshot’ nature of the frequencycomb measurement, probing the strongly frozen nuclear spin configuration. When the additional ^{71}Ga ‘heating’ excitation is applied it ‘thaws’ the ^{71}Ga spins, detected as broadening of the ^{75}As lineshape (as demonstrated in Fig. 4a, b). If all isotopes were fully ‘thawed’ we would have observed a closetoGaussian (k 1) lineshape. By contrast, in the case of completely frozen flipflops, the lineshape should be close to Lorentzian (k = 1), ultimately limited by the longitudinal nuclear spin relaxation. In a real experiment some spin fluctuations are always present: in particular the spins depolarized by the frequency comb influence one another via homonuclear dipolar interaction, resulting in a homogeneously broadened lineshape with some intermediate k > 1. This interpretation is further supported by a microscopic Bloch equation model of the frequencycomb experiments, as discussed in Supplementary Note 5C.
We use the ^{75}As lineshape as a probe to measure the dynamics of the ^{71}Ga equilibrium spin bath fluctuations. Rather than performing lengthy measurements of the entire ^{75}As lineshape, we excite its spins with a frequency comb with a fixed f_{MS} = 903 Hz, for which the ^{75}As depolarization dynamics is most sensitive to the ^{71}Ga ‘heating’. We find that the resulting ^{75}As depolarization is well described by a stretched exponential function ∝exp[−(t_{rf}/τ_{As})^{r}] with r ≈ 0.6–0.9 (see Methods); the measured value of τ_{As} is used to characterize the response of ^{75}As to ^{71}Ga ‘heating’. Furthermore, we now use selective ‘heating’ of either the −1/2 ↔ +1/2 CT or the + 1/2 ↔ +3/2 ST of ^{71}Ga. The amplitude β_{1, Ga} of the ‘heating’ frequency comb is varied—the resulting dependences of the ^{75}As depolarization time τ_{As} at different ^{75}As comb amplitudes β_{1, As} are shown in Fig. 4c, d for CT and ST ‘heating’, respectively. For analysis we also express β_{1, Ga} in terms of the rfinduced ^{71}Ga spinflip time τ_{Ga} ∝ β_{1, Ga}^{−2} (top scales in Fig. 4c, d, see details in Methods).
For vanishing ^{71}Ga excitation (β_{1, Ga} < 2 nT Hz^{−1/2}, τ_{Ga} > 20 s) the rfinduced spinflip time τ_{Ga} of ^{71}Ga is much larger than the correlation time τ_{c} of the ^{71}Ga intrinsic spin flipflops (τ_{Ga} > τ_{c}). By contrast, strong ‘heating’ (β_{1, Ga} > 200 nT Hz^{−1/2}, τ_{Ga} < 2 ms) completely ‘thaws’ the ^{71}Ga spins (τ_{Ga} < τ_{c}), broadens the ^{75}As lineshape (via heteronuclear interaction), and is observed as a reduction of τ_{As}. Another clear trend in Fig. 4c, d is that the τ_{As}(β_{1, Ga}) dependence becomes less pronounced at small β_{1, As}. This is because ^{75}As nuclei probe ^{71}Ga spin fluctuations on a timescale of the ^{75}As spin depolarization ∝β_{1, As}^{−2}. At very small β_{1, As} the ^{75}As spin evolution time becomes longer than τ_{c}, hence the ^{75}As lineshape is broadened by intrinsic ^{71}Ga fluctuations, and additional ^{71}Ga heating leads only to a minor reduction in τ_{As}. As we now show, this effect allows τ_{c} to be deduced from the measurements with variable β_{1, As} and β_{1, Ga} shown in Fig. 4c, d.
First, as we show in Supplementary Note 5B, the heatinginduced spin flips (with correlation time τ_{Ga}) and intrinsic dipolar flipflops (with correlation time τ_{c}) of ^{71}Ga have the same effect on ^{75}As. Therefore, τ_{As} depends only on the effective correlation time (τ_{Ga}^{−1} + τ_{c}^{−1})^{−1} of ^{71}Ga fluctuations. Second, as discussed above, ^{75}As homogeneous broadening is determined by fluctuations of ^{71}Ga only during the characteristic ^{75}As depolarization time. Consequently, the ^{75}As homogeneous linewidth, and hence τ_{As}, depend only on the ratio of the ^{71}Ga effective correlation time and the ^{75}As characteristic evolution time (∝β_{1, As}^{−2}). This ratio is proportional to β_{1, As}^{2}(τ_{Ga}^{−1} + τ_{c}^{−1})^{−1} and τ_{As} is its function. Although the analytical expression for this function is not known, it can be written as a Taylor series, from which we find that τ_{As} depends linearly on (τ_{Ga}^{−1} + τ_{c}^{−1})^{−1} in the limit of small β_{1, As} (see details in Supplementary Note 4). Taking into account that (τ_{Ga}^{−1} + τ_{c}^{−1})^{−1} ∈ (0, τ_{c}) and (τ_{Ga}^{−1} + τ_{c}^{−1})^{−1} = τ_{c}/2 when τ_{c} = τ_{Ga}, we find that τ_{c} can be estimated as the value of τ_{Ga} for which τ_{As} is a mean of its minimum and maximum values. From the measurements with the smallest β_{1, As} = 3.4 nT Hz^{−1/2} we estimate in a straightforward way τ_{c, CT} ~ 0.5 s and τ_{c, ST} ~ 1.5 s for CT and ST respectively. A more accurate derivation of τ_{c} is achieved by fitting the entire τ_{As} data set with a stretched exponential function of β_{1, As}^{2}(τ_{Ga}^{−1} + τ_{c}^{−1})^{−1}, as shown by the lines in Fig. 4c, d (see details in Methods and Supplementary Note 4). Fitting yields correlation times τ_{c, CT} ≈ 1.0 ± 0.2 s for CT and τ_{c, ST} ≈ 6.2 ± 2.5 s for ST. Furthermore, in Supplementary Note 5 we reproduce the experimental data of Fig. 4c and d with a numerical model based on Bloch equations. We then successfully verify the fitting procedure used here to derive τ_{c} by applying it to numerical model results with known τ_{c}.
The observed τ_{c} ≳ 1 s greatly exceeds typical nuclear dipolar flipflop times in strainfree III–V solids τ_{c} ~ 100 μs (refs 9,10,17). We attribute the extremely long τ_{c} in selfassembled quantum dots to the effect of inhomogeneous nuclear quadrupolar shifts making nuclear spin flipflops energetically forbidden^{20, 24}. This interpretation is corroborated by the observation of τ_{c, ST} > τ_{c, CT}, because quadrupolar broadening of the ST transitions is much greater than that of the CT transitions (ref. 21). Furthermore, the ^{71}Ga spins examined here have the largest gyromagnetic ratio γ and the smallest quadrupolar moment Q, so we expect that all other isotopes in InGaAs have even longer τ_{c}, resulting in the overall τ_{c} ≳ 1 s of the entire quantum dot nuclear spin bath. This implies that in high magnetic fields the spinecho coherence times of the electron and hole spin qubits in selfassembled dots are not limited by the nuclear spin bath up to subsecond regimes^{9, 10, 12, 28, 29}. Therefore, the effort aimed at achieving III–V semiconductor quantum dot spin qubits with long coherence should be focused on understanding other mechanisms of central spin dephasing, such as electronmediated nuclear–nuclear interactions, nuclear spin dynamics induced by electron spin manipulation, or charge fluctuations^{13, 14, 15}.
Because the frequencycomb technique is not limited by artefacts in the spin dynamics hampering pulsed magnetic resonance, it allows detection of very slow spin bath fluctuations. Such sensitivity of the method can be used, for example, to investigate directly the effect of the electron or hole on the spin bath fluctuations in charged quantum dots—arising, for example, from hyperfinemediated nuclear spin interactions. The experiments are well understood within a classical rate equation model and confirmed by a microscopic Bloch equation model, although further advances in frequencycomb spectroscopy can be expected with the development of a full quantum mechanical model. Furthermore, the simple and powerful ideas of frequencycomb NMR spectroscopy can be readily extended beyond quantum dots: as we show in Supplementary Note 3, the only essential requirement is that the longitudinal relaxation time T_{1} should be larger (by about two orders of magnitude) than the transverse relaxation time T_{2}, which is usually the case in solid state spin systems. Finally, our approaches in the use of frequency combs can go beyond NMR, enriching, for example, the techniques in optical spectroscopy.
Methods
Sample structures and experimental techniques.
The experiments were performed on individual neutral selfassembled InGaAs/GaAs quantum dots. The sample was mounted in a heliumbath cryostat (T = 4.2 K) with a magnetic field B_{z} = 8 T applied in the Faraday configuration (along the sample growth and light propagation direction Oz). A radiofrequency (rf) magnetic field B_{rf} perpendicular to B_{z} was induced by a miniature copper coil. Optical excitation was used to induce nuclear spin magnetization exceeding 50%, as well as to probe it by measuring hyperfine shifts in photoluminescence spectroscopy^{21}.
Two sample structures have been studied, both containing a single layer of InGaAs/GaAs quantum dots embedded in a weak planar microcavity with a Qfactor of ~250. In one of the samples the dots emitting at ~945 nm were placed in a p–i–n structure, where application of a large reverse bias during the rf excitation ensured the neutral state of the dots. The results for this sample are shown in Fig. 2. The second sample was a gatefree structure, where most of the dots emitting at ~914 nm are found in a neutral state, although the charging can not be controlled. Excellent agreement between the lineshapes of both ^{71}Ga and ^{75}As in the two structures was found, confirming the reproducibility of the frequencycomb technique. The results for this second sample are shown in Fig. 4.
Rate equation model for homogeneous lineshape derivation.
Let us consider an ensemble of spin I = 1/2 nuclei with gyromagnetic ratio γ and inhomogeneously broadened distribution of nuclear resonant frequencies ν_{nuc}. We assume that each nucleus has a homogeneous absorption lineshape L(ν), with normalization ∫ _{−∞}^{+∞}L(ν)dν = 1. A small amplitude (nonsaturating) rf field will result in depolarization, which can be described by a differential equation for population probabilities p_{±1/2} of the nuclear spin levels I_{z} = ±1/2
For frequencycomb excitation the decay rate is the sum of the decay rates caused by each rf mode with magnetic field amplitude B_{1}, and can be written as:
where the summation goes over all modes with frequencies ν_{j} = ν_{1} + jf_{MS} (ν_{1} is the frequency of the first spectral mode).
The change in the Overhauser shift E_{hf} produced by each nucleus is proportional to p_{+1/2} −p_{−1/2} and, according to equation (2) has an exponential time dependence ∝exp(−W(ν_{nuc})t). The quantum dot contains a large number of nuclear spins with randomly distributed absorption frequencies. Therefore, to obtain the dynamics of the total Overhauser shift we need to average over ν_{nuc}. Because the spectrum of the rf excitation is periodic and Δν_{inh} f_{MS}, the averaging can be done over one period f_{MS}. Furthermore, because the total width of the rf frequency comb Δν_{comb} is much larger than f_{MS} and Δν_{hom}, the summation in equation (3) can be extended to ±∞. Thus, the following expression is obtained for the time dependence ΔE_{hf}(t, f_{MS}), describing the dynamics of the rfinduced nuclear spin depolarization:
Equation (4) describes the dependence ΔE_{hf}(t, f_{MS}) directly measurable in experiments such as shown in Fig. 2b. ΔE_{hf}(t ∞) is the total optically induced Overhauser shift of the studied isotope and is also measurable, whereas f_{MS} and B_{1} are parameters that are controlled in the experiment. We note that in the limit of small mode spacing, f_{MS} 0, the infinite sum in equation (4) tends to the integral ∫_{−∞}^{+∞}L(ν)dν = 1 and the Overhauser shift decay is exponential (as observed experimentally) with a characteristic time
Equation (4) is a Fredholm integral equation of the first kind on the homogeneous lineshape function L(ν). This is an illconditioned problem: as a result finding the lineshape requires some constraints to be placed on L(ν). Our approach is to use the model lineshape of equation (1). After substituting L(ν) from equation (1), the righthand side of equation (4) becomes a function of the parameters Δν_{hom} and k, which we then find by leastsquares fitting of equation (4) to the experimental dependence ΔE_{hf}(t, f_{MS}).
This model is readily extended to the case of I > 1/2 nuclei. Equation (2) becomes a tridiagonal system of differential equations, and the solution (equation (3)) contains a sum of multiple exponents under the integral. These modifications are straightforward but tedious, and can be found in Supplementary Note 2.
Derivation of the nuclear spin bath correlation times.
Accurate lineshape modelling is crucial in revealing the ^{75}As homogeneous broadening arising from ^{71}Ga ‘heating’ excitation (as demonstrated in Fig. 4a, b). However, because a measurement of the full ΔE_{hf}(t, f_{MS}) dependence is time consuming, the experiments with variable comb amplitudes β_{1, Ga} and β_{1, As} (Fig. 4c, d) were conducted at a fixed f_{MS} = 903 Hz well in excess of the ^{75}As homogeneous linewidth Δν_{hom} ≈ 117 Hz. To extract the ^{75}As depolarization time τ_{As} we fit the ^{75}As depolarization dynamics ΔE_{hf}(t) with the following formula: ΔE_{hf}(t_{rf}) = ΔE_{hf}(t_{rf} ∞) × (1 − exp[−(t_{rf}/τ_{As})^{r}]), using r as a common fitting parameter and τ_{As} independent for measurements with different τ_{Ga}. We find r ≈ 0.6–0.9, depending on β_{1, As}, and the dependence of τ_{As} on β_{1, Ga} obtained from the fit is shown in Fig. 4c, d by the symbols with error bars corresponding to 95% confidence intervals.
The spacing of the ^{71}Ga ‘heating’ frequency comb is kept at a small value, f_{MS} = 150 Hz, ensuring uniform excitation of all nuclear spin transitions. The amplitude of the ‘heating’ comb is defined as , where B_{1} is magnetic field amplitude of each mode in the comb (further details can be found in Supplementary Note 1). To determine the correlation times we express β_{1, Ga} in terms of the rfinduced spinflip time τ_{Ga}. The τ_{Ga} is defined as the exponential time of the ^{71}Ga depolarization induced by the ‘heating’ comb, and is derived from an additional calibration measurement. The values of β_{1, Ga} shown in Fig. 4c are calculated using equation (5) as , where γ_{Ga} is the ^{71}Ga gyromagnetic ratio and τ_{Ga} is experimentally measured. The additional factor of 4 in the denominator is due to the matrix element of the CT of spin I = 3/2. For experiments on ST the β_{1, Ga} values shown in Fig. 4d are obtained with the same formulae, but multiplied by .
The effect of ^{71}Ga heating on ^{75}As (probed with τ_{As}) is controlled by the ratio of the ^{71}Ga effective correlation time (τ_{Ga}^{−1} + τ_{c}^{−1})^{−1} and the characteristic time of ^{75}As rfinduced depolarization 2/(γ_{As}^{2}β_{1, As}^{2}) given by equation (5). The experimental data of Fig. 4c, d are fitted with the function τ_{As} = τ_{As, min}(Δτ_{As} + (1 − Δτ_{As})exp[−(γ_{As}^{2}β_{1, As}^{2}(τ_{Ga}^{−1} + τ_{c}^{−1})^{−1}/(2θ))^{s}]), where θ ≈ 0.4, s ≈ 0.6 are dimensionless parameters describing the modified (stretched) exponential function, Δτ_{As} ≈ 2 is the ratio of τ_{As} without and with strong ^{71}Ga heating, τ_{As, min} is τ_{As} in the limit of large β_{1, Ga}, dependent on β_{1, As}, and τ_{c} is the correlation time of the ^{71}Ga spin fluctuations. Further details of the analysis used to derive τ_{c} from experimental data can be found in Supplementary Note 4. Moreover, we have verified this fitting procedure by applying it to numerically simulated spin dynamics with known τ_{c}, as described in Supplementary Note 5D.
References
 Borri, P. et al. Ultralong dephasing time in InGaAs quantum dots. Phys. Rev. Lett. 87, 157401 (2001).
 Volker, S. Holeburning spectroscopy. Annu. Rev. Phys. Chem. 40, 499–530 (1989).
 Yang, L. et al. Twocolour spin noise spectroscopy and fluctuation correlations reveal homogeneous linewidths within quantumdot ensembles. Nature Commun. 5, 5949 (2014).
 Hahn, E. L. Spin echoes. Phys. Rev. 80, 580–594 (1950).
 Biercuk, M. J. et al. Optimized dynamical decoupling in a model quantum memory. Nature 458, 996–1000 (2009).
 BarGill, N. et al. Suppression of spinbath dynamics for improved coherence of multispinqubit systems. Nature Commun. 3, 858 (2012).
 Tyryshkin, A. M. et al. Electron spin coherence exceeding seconds in highpurity silicon. Nature Mater. 11, 143–147 (2012).
 Li, D., Dementyev, A. E., Dong, Y., Ramos, R. G. & Barrett, S. E. Generating unexpected spin echoes in dipolar solids with π pulses. Phys. Rev. Lett. 98, 190401 (2007).
 Merkulov, I. A., Efros, A. L. & Rosen, M. Electron spin relaxation by nuclei in semiconductor quantum dots. Phys. Rev. B 65, 205309 (2002).
 de Sousa, R. & Das Sarma, S. Theory of nuclearinduced spectral diffusion: spin decoherence of phosphorus donors in Si and GaAs quantum dots. Phys. Rev. B 68, 115322 (2003).
 Yao, W., Liu, R.B. & Sham, L. J. Theory of electron spin decoherence by interacting nuclear spins in a quantum dot. Phys. Rev. B 74, 195301 (2006).
 Bluhm, H. et al. Dephasing time of GaAs electronspin qubits coupled to a nuclear bath exceeding 200 μs. Nature Phys. 7, 109–113 (2011).
 Press, D. et al. Ultrafast optical spin echo in a single quantum dot. Nature Photon. 4, 367–370 (2010).
 De Greve, K. et al. Ultrafast coherent control and suppressed nuclear feedback of a single quantum dot hole qubit. Nature Phys. 7, 872–878 (2011).
 Greilich, A., Carter, S. G., Kim, D., Bracker, A. S. & Gammon, D. Optical control of one and two hole spins in interacting quantum dots. Nature Photon. 5, 702–708 (2011).
 Hansom, J. et al. Environmentassisted quantum control of a solidstate spin via coherent dark states. Nature Phys. 10, 1745–2473 (2014).
 Urbaszek, B. et al. Nuclear spin physics in quantum dots: an optical investigation. Rev. Mod. Phys. 85, 79–133 (2013).
 Cai, J., Retzker, A., Jelezko, F. & Plenio, M. B. A largescale quantum simulator on a diamond surface at room temperature. Nature Phys. 9, 168–173 (2013).
 de Sousa, R. & Das Sarma, S. Electron spin coherence in semiconductors: considerations for a spinbased solidstate quantum computer architecture. Phys. Rev. B 67, 033301 (2003).
 Chekhovich, E. A., Hopkinson, M., Skolnick, M. S. & Tartakovskii, A. I. Suppression of nuclear spin bath fluctuations in selfassembled quantum dots induced by inhomogeneous strain. Nature Commun. 6, 6348 (2015).
 Chekhovich, E. A. et al. Structural analysis of strained quantum dots using nuclear magnetic resonance. Nature Nanotech. 7, 646–650 (2012).
 Munsch, M. et al. Manipulation of the nuclear spin ensemble in a quantum dot with chirped magnetic resonance pulses. Nature Nanotech. 9, 671–675 (2014).
 Udem, T., Holzwarth, R. & Hansch, T. W. Optical frequency metrology. Nature 416, 233–237 (2002).
 Dzhioev, R. I. & Korenev, V. L. Stabilization of the electronnuclear spin orientation in quantum dots by the nuclear quadrupole interaction. Phys. Rev. Lett. 99, 037401 (2007).
 Latta, C., Srivastava, A. & Imamoglu, A. Hyperfine interactiondominated dynamics of nuclear spins in selfassembled InGaAs quantum dots. Phys. Rev. Lett. 107, 167401 (2011).
 Madhu, P. & Kumar, A. Direct cartesianspace solutions of generalized Bloch equations in the rotating frame. J. Magn. Reson. Ser. A 114, 201–211 (1995).
 Van Vleck, J. H. The dipolar broadening of magnetic resonance lines in crystals. Phys. Rev. 74, 1168–1183 (1948).
 Khaetskii, A. V., Loss, D. & Glazman, L. Electron spin decoherence in quantum dots due to interaction with nuclei. Phys. Rev. Lett. 88, 186802 (2002).
 Bechtold, A. et al. Threestage decoherence dynamics of an electron spin qubit in an optically active quantum dot. Nature Phys 11, 1005–1008 (2015).
Acknowledgements
The authors are grateful to K. V. Kavokin for useful discussions. This work has been supported by the EPSRC Programme Grant EP/J007544/1, ITN S^{3}NANO. E.A.C. was supported by a University of Sheffield ViceChancellor’s Fellowship and a Royal Society University Research Fellowship. I.F. and D.A.R. were supported by EPSRC. Computational resources were provided in part by the University of Sheffield HPC cluster ‘Iceberg’.
Author information
Affiliations

Department of Physics and Astronomy, University of Sheffield, Sheffield S3 7RH, UK
 A. M. Waeber,
 A. I. Tartakovskii,
 M. S. Skolnick &
 E. A. Chekhovich

Department of Electronic and Electrical Engineering, University of Sheffield, Sheffield S1 3JD, UK
 M. Hopkinson

Cavendish Laboratory, University of Cambridge, Cambridge CB3 0HE, UK
 I. Farrer,
 D. A. Ritchie &
 J. Nilsson

Toshiba Research Europe Limited, Cambridge Research Laboratory, Cambridge CB4 0GZ, UK
 R. M. Stevenson,
 A. J. Bennett &
 A. J. Shields

Department of Physics, University of Konstanz, D78457 Konstanz, Germany
 G. Burkard
Contributions
M.H., I.F., D.A.R., J.N., R.M.S., A.J.B. and A.J.S. developed and grew the samples. A.M.W. and E.A.C. conceived and designed the experiments and analysed the data. A.M.W. performed the experiments. E.A.C. performed the numerical modelling. E.A.C., A.M.W., M.S.S., A.I.T., G.B. and A.J.B. wrote the manuscript with input from all authors. E.A.C. coordinated the project.
Competing financial interests
The authors declare no competing financial interests.
Author details
A. M. Waeber
Search for this author in:
M. Hopkinson
Search for this author in:
I. Farrer
Search for this author in:
D. A. Ritchie
Search for this author in:
J. Nilsson
Search for this author in:
R. M. Stevenson
Search for this author in:
A. J. Bennett
Search for this author in:
A. J. Shields
Search for this author in:
G. Burkard
Search for this author in:
A. I. Tartakovskii
Search for this author in:
M. S. Skolnick
Search for this author in:
E. A. Chekhovich
Search for this author in:
Supplementary information
PDF files
 Supplementary information (10,728,803 KB)
Supplementary information