Arxiv:1905.00776V1 [Cond-Mat.Mes-Hall] 2 May 2019

Long-Range Microwave Mediated Interactions Between Electron Spins

F. Borjans,1 X. G. Croot,1 X. Mi,1, ∗ M. J. Gullans,1 and J. R. Petta1, † 1Department of Physics, Princeton University, Princeton, New Jersey 08544, USA (Dated: May 3, 2019) Entangling gates for electron spins in semiconductor quantum dots are generally based on exchange, a short- ranged interaction that requires wavefunction overlap. Coherent spin-photon coupling raises the prospect of using photons as long-distance interconnects for spin qubits. Realizing a key milestone for spin-based quantum information processing, we demonstrate microwave-mediated spin-spin interactions between two electrons that are physically separated by more than 4 mm. Coherent spin-photon coupling is demonstrated for each individual spin using microwave transmission spectroscopy. An enhanced vacuum Rabi splitting is observed when both spins are tuned into resonance with the cavity, indicative of a coherent spin-spin interaction. Our results demonstrate that microwave-frequency photons can be used as a resource to generate long-range two-qubit gates between spatially separated spins.

Nonlocal qubit interactions are a hallmark of advanced and the cavity. Overcoming these challenges, we use a mi- quantum information technologies [1–5]. The ability to trans- crowave frequency photon to coherently couple two electron fer quantum states and generate entanglement over distances spins that are separated by more than 4 mm, opening the door much larger than qubit length scales greatly increases connec- to a Si spin qubit architecture with all-to-all connectivity. De- tivity and is an important step towards maximal parallelism liberately engineered asymmetries in micromagnet design, in and the implementation of two-qubit gates on arbitrary pairs conjunction with three-axis control of the external magnetic of qubits [6]. Qubit coupling schemes based on cavity quan- field, allow us to tune separated spin qubits into resonance tum electrodynamics [2,7,8] also offer the possibility of us- with each other and the cavity. ing high quality factor resonators as quantum memories [3,9]. The device consists of two double quantum dots (DQDs), Furthermore, extending qubit interactions beyond the near- denoted L-DQD and R-DQD, that are fabricated on a Si/SiGe est neighbor is particularly beneficial for spin-based quantum heterostructure and positioned at the antinodes of a half- computing architectures limited by short-range exchange in- wavelength Nb superconducting cavity [Fig.1(a)] with a cen- teractions [10]. ter frequency fc = 6.745 GHz and decay rate κ/2π = Silicon spin qubits can be fabricated in dense arrays and 1.98 MHz. Device 1 is fabricated on a nat-Si quantum well, the device technology has matured rapidly in the past several while Device 2 utilizes an enriched 28Si quantum well with an years, with single qubit gate fidelities that rival superconduct- 800 ppm residual concentration of 29Si. A single electron is ing qubits [11], demonstrations of exchange-based two qubit isolated in each DQD, which is defined by a tri-layer overlap- gates [12–14], and promising initial steps towards quantum in- ping aluminum gate stack [24] operated in accumulation mode formation transfer along qubit arrays [15, 16]. Long-range co- [Fig.1(b)]. The electron interacts with the electric field of the herent coupling between separated semiconductor spin qubits cavity through the electric dipole interaction. Our device de- is an outstanding challenge. Efforts towards nonlocal spin- sign uses a split-gate cavity coupler [labelled CP in Fig.1(b)] spin coupling have been limited to interactions between indi- that is galvanically connected to the center pin of the super- vidual spins and photons in a microwave cavity [17–20]. conducting cavity [25]. The split-gate coupler relaxes the con- One of the key challenges associated with achieving long straint that plunger gates be held at the same potential [18] and distance spin-spin coupling stems from the lack of indepen- allows simultaneous tuning of both DQDs to the one electron dent control of the spin Zeeman energies. In order to co- regime, which is necessary for a demonstration of spin-spin herently couple spin qubits via cavity photons [21, 22], it is coupling. Measurements take place in a dilution refrigerator necessary to bring the qubits into resonance with each other and an in-plane external magnetic field Bext is used to tune (virtual coupling), or into resonance with each other and the the spin transition into resonance with a cavity photon. microwave cavity (resonant coupling). Unlike charge qubits, The small magnetic field generated by the vacuum fluctu- whose energy splittings can easily be electrically tuned [8], ations of the cavity, combined with the weak magnetic mo- spin qubits confined in quantum dots have limited electrical ment of the electron µB ≈ 58 µeV/T, leads to an intrinsic tunability. Their energies are set by a total magnetic field

arXiv:1905.00776v1 [cond-mat.mes-hall] 2 May 2019 spin-photon coupling rate gs/2π ≈ 10100 Hz, which is much ~ tot B , which results from the vector addition of a global ex- too slow to be useful as a quantum technology [26, 27]. In ternal field B~ ext and stray fields from micromagnets B~ M de- our device architecture a large electric dipole coupling rate posited on-chip to facilitate spin-photon coupling [18, 23]. gc/2π ≈ 40 MHz is combined with an artificial spin-orbit in- Small variations in micromagnet fabrication at each qubit lead teraction generated by a micromagnet to achieve spin-photon to variations in B~ tot, making resonant coupling of spins chal- coupling [17, 23, 28]. The resulting spin-cavity coupling rate lenging. While spin-charge hybridization permits a small de- is given by gs ∝ gµB∆Bx, where g ≈ 2 is the electronic gree of frequency tunability, generally it is insufficient to si- g-factor and µB is the Bohr magneton and ∆Bx is the ampli- multaneously tune spin qubits into resonance with each other tude of the effective oscillating transverse field generated by the electrons motion in the cavity field [18, 19, 23]. ∆Bx is maximal at the interdot charge transition, where the level detuning = 0, and gc/2π is largest. Large gradient fields have ∗ Present address: Google Inc., Santa Barbara, California 93117, USA been used as a resource to enable high fidelity single spin ma- † [email protected] nipulation [29, 30] and recent demonstrations of single spin- 2

(a) (b) 1 mm 200 nm (a) R-DQD CP

S1L L-DQD L-DQD D Bext R-DQD SL L 휙 S2 DQD (ẑ) L 0.0 0.5 1.0 (b) A/A0 (c) B1L P1L B2L P2L B3L 0.0 0.5 1.0 0.0 0.5 1.0 6.78 (c) A/A0 A/A0

6.78 L-DQD R-DQD (GHz) photon photon f

6.74 6.70 휙 = 0.0º 휙 = 0.0º

(GHz) f 6.78 6.70

106 108 110 112 100 102 104 106 (GHz) ext ext B (mT) B (mT) f 6.70 휙 = 2.8º 휙 = 2.8º FIG. 1. Cavity-coupler for spins. (a) Optical micrograph of the 100 104 108 112 100 104 108 112 ext ext superconducting cavity containing two single electron DQDs. The B (mT) B (mT) electron spin in each DQD is coupled to the cavity through a com- bination of electric-dipole and artificial spin-orbit interactions. (b) FIG. 2. Tuning towards spin-spin resonance. (a) The micromagnets False-color scanning electron microscope image of the L-DQD. A are purposely fabricated at an angle relative to the DQD z-axis to double well potential is formed beneath plunger gates P1L and P2L, allow for additional control of the spin resonance conditions. By ro- ext ext and the barrier gate B2L is used to adjust interdot tunnel coupling. tating B relative to the z-axis, and tuning the magnitude of B , Spin-orbit coupling is induced by a Co micromagnet (dashed lines). the energy splitting of both spins can be brought into resonance with ext (c) Cavity transmission A/A0 plotted as a function of the external the cavity mode. (b) A/A0 as a function of B and f. For a field magnetic field Bext for the L-DQD and R-DQD. Vacuum Rabi split- angle φ = 0◦ the R-spin and L-spin resonance conditions are sep- ting, a hallmark of strong coupling, is observed for each spin. arated by Bext = 6 mT. Increasing the field angle to φ = 2.8◦ dramatically reduces the field separation. (c) Theoretical model for A/A0 using a master equation simulation. photon coupling with gs/2π ≈ 10 MHz [18, 19]. We first demonstrate strong coupling of a spin trapped in of the micromagnet. A similar approach was adopted to re- each DQD to a cavity photon. To probe spin-photon coupling, motely couple two ensembles of nitrogen vacancy centers in the cavity transmission A/A0 is plotted as a function of cavity diamond, where the crystal axes of the two diamond samples ext probe frequency f and B in Fig.1(c). The Zeeman split- were rotated relative to one another [31]. We show now that ting increases with the total magnetic field B~ tot = B~ ext+B~ M. this approach is well-suited for coupling two single spins in Field tuning allows us to bring each spin into resonance with silicon via a cavity mode. ext the cavity photon of energy hfc, where h is Plancks constant. Figure2(b) shows A/A0 as a function of f and B with We observe coherent coupling between the spin trapped in the φ = 0◦. The R-spin (L-spin) is in resonance with the cav- L-DQD (L-spin) and the cavity photon, as evidenced by the ity mode at Bext ≈ 103.1 mT (Bext ≈ 109.1 mT). With vacuum Rabi splitting, when Bext = 109.1 mT. Strong spin- φ = 2.8◦, the R-spin resonance condition shifts up to Bext ≈ photon coupling is achieved with gs,L/2π = 10.7 ± 0.1 MHz 104.7 mT and the L-spin resonance condition shifts down to exceeding the cavity decay rate κ/2π = 1.98 MHz and spin Bext ≈ 107.4 mT. To compare the data with theory we model decoherence rate γs,L/2π = 4.7 MHz (AppendixB). Like- the system with an effective Jaynes-Cummings Hamiltonian wise, we observe strong spin-photon coupling for the R-spin [18, 23] ext at B = 103.1 mT with gs,R/2π = 12.0 ± 0.2 MHz ex- γ /2π = 5.3 MHz κ/2π 6 mT X ¯hωj + − ceeding s,R and . The difference H = σz + ¯hω a†a + ¯hg σ a + σ a† , (1) in Bext is equivalent to a 268 MHz difference in the spin res- 2 j c s,j j j j onance frequencies after accounting for the finite susceptibility of the micromagnet (AppendixB), and is largely due to where j = L, R, and ¯hωj is the energy level splitting of the ef- inaccuracies in micromagnet placement. This substantial fre- µ fective j-spin, σj are the Pauli operators in the j-spin Hilbert quency detuning between the spins precludes the observation space and ωc/2π is the cavity frequency. Decoherence in our of resonant spin-spin coupling [18]. system is incorporated by a Lindblad master equation (Ap- To compensate for local differences in the magnetic field, pendixA), using the measured γs,j and κ. We plot the theo- the device was purposely fabricated with the long axis of the retical predictions for the cavity transmission in Fig.2(c) and Co micromagnets tilted ±15 degrees relative to the interdot observe strong agreement with the experimental data. axis of the DQDs [Fig.2(a)]. Because of this intentional We now demonstrate control over the difference between asymmetry, adjusting the angle φ of the in-plane magnetic the spin resonance frequencies by fixing the external mag- field relative to the DQD axis provides an additional degree netic field magnitude Bext and varying the angle of Bext of freedom for simultaneous tuning of both spins into reso- in the plane of the sample. The expected spin resonance nance with the cavity. Qualitatively, the high permeability Co frequencies are plotted in Fig.3(a) as a function of φ with micromagnet concentrates the magnetic field lines, leading to Bext = 106.3 mT (upper panel) and Bext = 110 mT (lower a maximum total field when B~ ext is aligned with the long axis panel), confirming that these two control parameters will 3

(a) (b) 0.0 0.5 1.0 (c) A/A0 7.1 ext B = 106.3 mT Bext = 106.3 mT Bext = 106.3 mT

Z 6.78 (GHz)

f 6.76

6.6

ext

7.1 B = 110 mT (GHz) f

6.72

(GHz) f 6.70 6.6 0 10 3 4 5 6 7 8 3 4 5 6 7 8 휙 (º) 휙 (º) 휙 (º)

FIG. 3. (a) Expected spin resonance frequencies as a function of φ for Bext = 106.3 mT (top panel) and Bext = 110 mT (bottom panel). Bext allows control over both spin frequencies with respect to the cavity, while φ allows control over the spin frequencies with respect to each other. The frequency of the left (right) spin is plotted in blue (purple). (b) A/A0 as a function of f and φ demonstrates simultaneous tuning of both spins into resonance with the cavity at φ = 5.6◦ and Bext = 106.3 mT. Dashed lines indicate left and right spin transition frequencies. (c) Theoretical prediction for A/A0. bring the two spins into resonance with the cavity and each ting is quantitatively analyzed for Device 1 in Fig.4(c) and other. Based on microwave spectroscopy measurements of Device 2 in Fig.4(d). These devices have slight differences in the spins, we expect resonance to occur around φ = 6◦ and gate geometry, and the micromagnets in Device 2 are canted ext B = 106.3 mT. Figure3(b) maps out the field angle at angles of ±10 degrees. In Device 1, we extract gs,L/R ◦ dependence of the spin resonance frequencies over a range by measuring A/A0 at φ = 5.6 and detuning the right/left ◦ ext φ = 38 with B = 106.3 mT. The resonance frequency spin using . In Device 2, we extract the gs,L/R by measuring of the R-spin monotonically moves to lower f as φ is in- A/A0 at an off-resonant angle, at the now separated resonant creased, while the L-spin shows the opposite dependence. fields. For Device 1, we observe a vacuum Rabi splitting of ◦ With φ = 5.6 both spins are tuned into resonance with the 2gs,L/2π = 21.4 ± 0.2 MHz when L-spin is in resonance with cavity. These results are well captured by the theoretical pre- the cavity, 2gs,R/2π = 24.0 ± 0.4 MHz when R-spin is in diction of transmission through the cavity shown in Fig.3(c). resonance with the cavity, and 2gs,LR/2π = 30.2 ± 0.2 MHz when both spins are in resonance with the cavity. Device 2 The spectrum of the Jaynes-Cummings model for a single shows similar behavior and again exhibits an enhanced vac- spin and a single photon in the cavity is shown in Fig.4(a). uum Rabi splitting 2g /2π = 18.4 ± 0.4 MHz when both With the L-spin tuned into resonance with the cavity, the s,LR spins are tuned into resonance with the cavity, compared to spin and cavity photon hybridize leading to a vacuum Rabi the individual splittings 2g /2π = 13.2 ± 0.2 MHz and splitting of magnitude 2g in the cavity transmission [7]. s,L s,L 2g /2π = 12.4 ± 0.2 MHz. Combined, these two data sets In contrast, when both spins are simultaneously tuned into s,R give strong evidence for microwave assisted spin-spin interac- resonance with the cavity, the excited state spectrum of the tions across a 4 mm length scale that is many orders of magni- Jaynes-Cummings model splits into a sub-radiant state and tude larger than what can be achieved using direct wavefunc- two bright states [2]. For N identical spins the Jaynes- √ tion overlap. Moreover, these measurements show that both Cummings model predicts a N enhancement of the cou- field angle and DQD level detuning can be used to modulate pling rate [7]. In our device geometry, the sub-radiant state √ the strength of the spin-spin interactions. is the spin-triplet |0,T0i = 1/ 2 (|0, ↑, ↓i + |0, ↓, ↑i) because the DQDs are located at opposite ends of the cavity The data in Fig.4(c) are fit using a master equation descrip- 180◦ where the electric fields are out of phase. Here, the tion of the spin-cavity system. We independently measure spin states of the L/R-spin are quantized along their local to- the Zeeman splittings ¯hωL/R and linewidths γs,L/R using mi- tal magnetic field axis. The bright states√ are hybridizations crowave spectroscopy (AppendixB). The cavity linewidth κ, |0,S i = 1/ 2 (|0, ↑, ↓i − |0, ↓, ↑i) between the singlet state 0 as well as a complex Fano factor q, are obtained by fitting |1, ↓, ↓i and the state with a single photon that are separated in the bare cavity response with the spins detuned from reso- energy by twice the collectively enhanced vacuum Rabi cou- nance (AppendixB). The spin-photon coupling rates gs,L/R q 2 2 pling, 2gs,LR = 2 gs,L + gs,R. We now search for evidence for each device are obtained by fitting the data with the spins of cavity-mediated single spin coupling. individually tuned into resonance with the cavity as shown in ext Figs.4(c),(d). From the Jaynes-Cummings model we expect Figure4(b) shows A/A0 as a function of f and B with the bright states to split with the enhanced collective coupling φ = 5.6◦, where both spins are in resonance with the cav- q 2 2 ity. As Bext is increased, both spins are simultaneously tuned rate 2gs,LR = 2 gs,L + gs,R. The extracted splittings agree ext into resonance with the cavity at B = 106.3 mT and we with the theoretical predictions of 2gs,LR/2π = 32.1 MHz for observe an enhancement in the vacuum Rabi splitting relative Device 1 and 2gs,LR/2π = 18.1 MHz for Device 2 to within to the data sets shown in Fig.2(b). The vacuum Rabi split- 6%. 4

2 2 (a) 2 gs,L+ gs,R (c) (d) 2g Device 1 Device 2 s,L 〉 |1, ↓ 〉 |0, ↑〉 |1, ↓ ↓ 〉 |0, S0 0.6 2gs,R/2p 2gs,R/2p = 24 MHz 1.2 = 12.4 MHz

|0, ↓〉 |0, ↓ ↓ 〉 R-spin R-spin 0.0 0.5 1.0 (b) A/A p p 0 0.4 2gs,L/2 2gs,L/2 = 21.4 MHz 0.8 = 13.2 MHz

6.78

/ A 6.76 A L-spin L-spin

0.2 2gs,LR/2p 0.4 2gs,LR/2p

(GHz) = 30.2 MHz =

f 18.4 MHz 6.72 both both spins spins 6.70 휙 = 5.6º 0.0 0.0 102 106 110 112 6.70 6.74 6.78 6.66 6.68 6.70 6.72 Bext (mT) f (GHz) f (GHz)

FIG. 4. Cavity-mediated spin-spin coupling. (a) Tuning the L-spin into resonance with the cavity results in vacuum Rabi splitting with q 2 2 magnitude 2gs,L. A vacuum Rabi splitting of magnitude 2gs,LR = 2 gs,L + gs,R is expected when both spins are tuned into resonance with ext ◦ the cavity. (b) A/A0 as a function of f and B with φ = 5.6 indicates an enhanced vacuum Rabi splitting when the L-spin and R-spin are tuned into resonance with the cavity. (c) A/A0 as a function of f for the R-spin in resonance (upper curve), L-spin in resonance (middle curve), and both spins in resonance (bottom curve). The enhancement of the vacuum Rabi splitting with both spins on resonance with the cavity is indicative of spin-spin coupling via the cavity mode. (d) Cavity-assisted spin-spin coupling is also observed in a second device with a different gate pattern. Dashed lines in (c,d) are ﬁts to a master equation simulation. Insets in (c,d) are scanning electron microscope images of the devices, with 200 nm scale bars.

The observation of enhanced vacuum Rabi splitting when Appendix A: Master Equation Theory the separated spins are simultaneously on resonance with the cavity is evidence of long-range spin-spin coupling. The non- Since the transverse magnetic field ∆Bx ≈ 30 mT is large local interaction of two spins marks an important milestone compared to the charge-cavity coupling rate gc, we can use for all-to-all qubit connectivity and scalability in silicon-based the effective Hamiltonian described in Equation1. We intro- quantum circuits. In the near term, this demonstration paves duce dephasing in our system by coupling the spin-cavity sys- the way for the implementation of modular qubit architectures tem to external baths. Integrating out these baths in a Born- in silicon, wherein nearest-neighbor coupled registers of spin Markov approximation leads to a Lindblad master equation qubits [24] can be interfaced with other sparsely distributed for the spin-cavity dynamics registers via microwave photons. With further improvements in cavity quality factors and spin-photon coupling rates, time- dρ i X = − [H, ρ]+ γd D σz + γr D σ− ρ+κD [α] ρ domain demonstrations of cavity-assisted spin-spin coupling dt ¯h s,j j s,j j should be within experimental reach. j (A1) † 1 † † where D[A] = AρA − 2 (A Aρ + ρA A) is the Lindblad d r super-operator, γs,j is the spin-dephasing rate, γs,j is the re- laxation rate of the spins, and κ is the total cavity decay rate. d r The spin linewidth (FWHM) is given by 2γs,j = 2γs,j + γs,j. We decompose κ = κ1 + κ2 + κin into the contributions from ACKNOWLEDGMENTS the decay rates into the two transmission lines κ1,2 and an intrinsic cavity decay rate κin. In this experiment, DQDs are probed via the cavity trans- We thank Lisa Edge, Joe Kerckhoff, Thaddeus Ladd, and mission. The transmission can be computed from the master Emily Pritchett of HRL Laboratories, LLC for providing the equation evolution using input-output theory [32], which re- 28 Si heterostructure used in these experiments, device sim- lates the output field operator an,out of each transmission line ulation support, and technical comments on the manuscript. to the input field operator an,in through the relation We acknowledge helpful discussions with Monica Benito and √ Guido Burkard. an,out(t) = κna(t) − an,in(t) (A2) 5 where a(t) is found from the operator evolution under the work [18]. Substituting the extracted qubit frequency ωj into Lindblad master√ equation with the addition of a drive term Eq. A4, we generate the theory plots in Fig.2(c). H = P κ a†a + h.c.. In the limit of weak To generate the fit in Fig.3(c), the angular dependence of drive √ n n n,in driving κn han,ini κ (here operator expectation values the resonance frequencies of the two DQDs is extracted us- of the input-output fields are taken with respect to the density ing microwave spectroscopy in the dispersive regime. We can matrix of the bath, which is treated as unentangled with the then shift the dependence in frequency based on the external system in the Born-Markov approximation), we can neglect magnetic field and the known magnetic susceptibility to over- the population of the excited states of the Jaynes-Cummings lay with the data. model with more than one excitation, leaving the reduced The fits to the line cuts shown in Figs.4(c),(d) are modeled spin-photon Hilbert space with Equation A4, using the input parameters in TableI. We leave gs as a free parameter in nonlinear regression fits and |0, ↓, ↓i , |0, ↑, ↓i , |0, ↓, ↑i , |1, ↓, ↓i (A3) fix the remaining parameters extracted from separate measurements. The errors on gs given in the main text are determined where the first quantum number labels the number of photons from these fits. We assume the gradient field between the two in the cavity and the left/right spin labels refer to the L/R-spin dots is ∆Bx = 30 mT based on previous experiments [18]. direction quantized along the axis of the local magnetic field. We define ±θ as the rotation angle of the micromagnet long axis relative to the interdot axis of the left (right) DQDs, and When gs,jTr[ρa] γs,j, we can further neglect populations outside of the ground state |0, ↓, ↓i and only need to account tc,L/R is the interdot tunnel coupling of the left (right) DQDs. for off-diagonal coherences in the master equation evolution. To fit the data in Figs.4(c),(d) with both spins on resonance, In this limit, the complex transmission amplitude takes the we use an effective model where we extract the splitting of approximate form the two bright states using a single collective coupling gs,LR and a single effective decay rate γs,LR. These fits are in good √ √ agreement with the predicted curves (not shown) for the two- ha i κ Tr[ρa] −i( κ κ + ∆ /q) A = 2,out = 2 = 1 2 0 spin system based on equation A4 with the parameters from g2 ha1,ini ha1,ini κ P s,j −∆0 − i − i TableI. 2 j γs,j +iδj (A4) where ∆0 = ωd − ωc is the detuning of the cavity drive ωd from the cavity frequency, δj = ωj − ωd is the detuning of TABLE I. Physical parameters of the two devices. the j-spin from the drive frequency, and we have introduced a Parameter Device 1 Device 2 complex parameter q to account for Fano interference effects gs,L/2π 10.7 MHz 6.6 MHz in the bare cavity transmission [33]. For simplicity, we ne- gs,R/2π 12.0 MHz 6.2 MHz glect all higher order corrections in gs,j/q to the cavity trans- gs,LR/2π 15.1 MHz 9.2 MHz mission. γs,L/2π 4.7 MHz 3.0 MHz γs,R/2π 5.3 MHz 3.3 MHz γs,LR/2π 5.0 MHz 4.0 MHz 2tc,L/h 8.8 GHz 9.1 GHz Appendix B: Theory Fits 2tc,R/h 8.6 GHz 9.2 GHz κ/2π 2.0 MHz 3.46 MHz q 33.3e−0.3iπ 20e−0.25iπ From the spin-photon transitions in Fig.1(c), we fit the ∆Bx 30 mT 30 mT qubit frequency and extract the finite magnetic susceptibil- θ 15◦ 10◦ ◦ ity χ of the micromagnet, finding χ ≈ 0.6 at φ = 0 over χ 0.6 0.6 small ranges of magnetic field, in agreement with previous

[1] L.-M. Duan, M. D. Lukin, J. I. Cirac, and P. Zoller, “Long- crowave cavity memories,” Nat. Phys. 14, 705 (2018). distance quantum communication with atomic ensembles and [6] J. Preskill, “Reliable quantum computers,” Proc. R. Soc. Lon- linear optics,” Nature (London) 414, 413 (2001). don, Ser. A 454, 385 (1998). [2] J. Majer, J. M. Chow, J. M. Gambetta, J. Koch, B. R. John- [7] A. Wallraff, D. I. Schuster, A. Blais, L. Frunzio, R.-S. Huang, son, J. A. Schreier, L. Frunzio, D. I. Schuster, A. A. Houck, J. Majer, S. Kumar, S. M. Girvin, and R. J. Schoelkopf, “Strong A. Wallraff, A. Blais, M. H. Devoret, S. M. Girvin, and coupling of a single photon to a superconducting qubit using R. J. Schoelkopf, “Coupling superconducting qubits via a cavity circuit quantum electrodynamics,” Nature (London) 431, 162 bus,” Nature (London) 449, 443 (2007). (2004). [3] M. A. Sillanp, J. I. Park, and R. W. Simmonds, “Coherent quan- [8] D. J. van Woerkom, P. Scarlino, J. H. Ungerer, C. Mller, J. V. tum state storage and transfer between two phase qubits via a Koski, A. J. Landig, C. Reichl, W. Wegscheider, T. Ihn, K. En- resonant cavity,” Nature (London) 449, 438 (2007). sslin, and A. Wallraff, “Microwave Photon-Mediated Interac- [4] C. Monroe and J. Kim, “Scaling the Ion Trap Quantum Proces- tions between Semiconductor Qubits,” Phys. Rev. X 8, 041018 sor,” Science 339, 1164 (2013). (2018). [5] C. J. Axline, L. D. Burkhart, W. Pfaff, M. Zhang, K. Chou, [9] W. Pfaff, C. J. Axline, L. D. Burkhart, U. Vool, P. Reinhold, P. Campagne-Ibarcq, P. Reinhold, L. Frunzio, S. M. Girvin, L. Frunzio, L. Jiang, M. H. Devoret, and R. J. Schoelkopf, L. Jiang, M. H. Devoret, and R. J. Schoelkopf, “On-demand “Controlled release of multiphoton quantum states from a mi- quantum state transfer and entanglement between remote microwave cavity memory,” Nat. Phys. 13, 882 (2017). 6

[10] J. R. Petta, A. C. Johnson, J. M. Taylor, E. A. Laird, A. Ya- [22] M. Benito, J. R. Petta, and G. Burkard, “Optimized cavity- coby, M. D. Lukin, C. M. Marcus, M. P. Hanson, and A. C. mediated dispersive two-qubit gates between spin qubits,” Gossard, “Coherent Manipulation of Coupled Electron Spins in arXiv:1902.07649 (2019). Semiconductor Quantum Dots,” Science 309, 2180 (2005). [23] M. Benito, X. Mi, J. M. Taylor, J. R. Petta, and G. Burkard, [11] J. Yoneda, K. Takeda, T. Otsuka, T. Nakajima, M. R. Delbecq, “Input-output theory for spin-photon coupling in Si double G. Allison, T. Honda, T. Kodera, S. Oda, Y. Hoshi, N. Usami, quantum dots,” Phys. Rev. B 96, 235434 (2017). K. M. Itoh, and S. Tarucha, “A quantum-dot spin qubit with co- [24] D. M. Zajac, T. M. Hazard, X. Mi, E. Nielsen, and J. R. herence limited by charge noise and fidelity higher than 99.9%,” Petta, “Scalable Gate Architecture for a One-Dimensional Ar- Nat. Nanotechnol. 13, 102 (2018). ray of Semiconductor Spin Qubits,” Phys. Rev. Appl. 6, 054013 [12] M. Veldhorst, C. H. Yang, J. C. C. Hwang, W. Huang, J. P. De- (2016). hollain, J. T. Muhonen, S. Simmons, A. Laucht, F. E. Hudson, [25] J. R. Petta, F. Borjans, X. Mi, S. Putz, X. Croot, M. G. Borselli, K. M. Itoh, A. Morello, and A. S. Dzurak, “A two-qubit logic L. F. Edge, J. Kerckhoff, T.D. Ladd, and E. J. Pritchett, (in gate in silicon,” Nature (London) 526, 410 (2015). preparation). [13] D. M. Zajac, A. J. Sigillito, M. Russ, F. Borjans, J. M. Taylor, [26] D. I. Schuster, A. P. Sears, E. Ginossar, L. DiCarlo, L. Frun- G. Burkard, and J. R. Petta, “Resonantly driven CNOT gate for zio, J. J. L. Morton, H. Wu, G. A. D. Briggs, B. B. Buckley, electron spins,” Science 359, 439 (2018). D. D. Awschalom, and R. J. Schoelkopf, “High-Cooperativity [14] T. F. Watson, S. G. J. Philips, E. Kawakami, D. R. Ward, P. Scar- Coupling of Electron-Spin Ensembles to Superconducting Cav- lino, M. Veldhorst, D. E. Savage, M. G. Lagally, Mark Friesen, ities,” Phys. Rev. Lett. 105, 140501 (2010). S. N. Coppersmith, M. A. Eriksson, and L. M. K. Vandersypen, [27] Y. Kubo, F. R. Ong, P. Bertet, D. Vion, V. Jacques, D. Zheng, “A programmable two-qubit quantum processor in silicon,” Na- A. Drau, J.-F. Roch, A. Auffeves, F. Jelezko, J. Wrachtrup, ture (London) 555, 633 (2018). M. F. Barthe, P. Bergonzo, and D. Esteve, “Strong Coupling of [15] T. Fujita, T. A. Baart, C. Reichl, W. Wegscheider, and L. M. K. a Spin Ensemble to a Superconducting Resonator,” Phys. Rev. Vandersypen, “Coherent shuttle of electron-spin states,” npj Lett. 105, 140502 (2010). Quantum Inf. 3, 22 (2017). [28] A. Cottet and T. Kontos, “Spin Quantum Bit with Ferromag- [16] A. R. Mills, D. M. Zajac, M. J. Gullans, F. J. Schupp, T. M. netic Contacts for Circuit QED,” Phys. Rev. Lett. 105, 160502 Hazard, and J. R. Petta, “Shuttling a single charge across a one- (2010). dimensional array of silicon quantum dots,” Nature Commun. [29] M. Pioro-Ladrire, T. Obata, Y. Tokura, Y.-S. Shin, T. Kubo, 10, 1063 (2019). K. Yoshida, T. Taniyama, and S. Tarucha, “Electrically driven [17] J. J. Viennot, M. C. Dartiailh, A. Cottet, and T. Kontos, “Co- single-electron spin resonance in a slanting Zeeman field,” Nat. herent coupling of a single spin to microwave cavity photons,” Phys. 4, 776 (2008). Science 349, 408 (2015). [30] J. Yoneda, T. Otsuka, T. Takakura, M. Pioro-Ladrire, R. Brun- [18] X. Mi, M. Benito, S. Putz, D. M. Zajac, J. M. Taylor, Guido ner, H. Lu, T. Nakajima, T. Obata, A. Noiri, C. J. Palmstrm, Burkard, and J. R. Petta, “A coherent spinphoton interface in A. C. Gossard, and S. Tarucha, “Robust micromagnet design silicon,” Nature (London) 555, 599 (2018). for fast electrical manipulations of single spins in quantum [19] N. Samkharadze, G. Zheng, N. Kalhor, D. Brousse, A. Sam- dots,” Appl. Phys. Express 8, 084401 (2015). mak, U. C. Mendes, A. Blais, G. Scappucci, and L. M. K. [31] T. Astner, S. Nevlacsil, N. Peterschofsky, A. Angerer, S. Rotter, Vandersypen, “Strong spin-photon coupling in silicon,” Science S. Putz, J. Schmiedmayer, and J. Majer, “Coherent Coupling 359, 1123 (2018). of Remote Spin Ensembles via a Cavity Bus,” Phys. Rev. Lett. [20] A. J. Landig, J. V. Koski, P. Scarlino, U. C. Mendes, A. Blais, 118, 140502 (2017). C. Reichl, W. Wegscheider, A. Wallraff, K. Ensslin, and [32] P. Meystre and M. Sargent, Elements of Quantum Optics, 4th T. Ihn, “Coherent spinphoton coupling using a resonant ex- ed. (Springer-Verlag, Berlin Heidelberg, 2007). change qubit,” Nature (London) 560, 179 (2018). [33] U. Fano, “Effects of Configuration Interaction on Intensities [21] A. Warren, E. Barnes, and S. E. Economou, “Long-distance and Phase Shifts,” Phys. Rev. 124, 1866 (1961). entangling gates between quantum dot spins mediated by a superconducting resonator,” arXiv:1902.05704 (2019).