
5
Control of collective excitations. The determin-
istic preparation of collective states is essential in or-
der to pave the way for their applications in quantum-
information processing. To this end, we coherently ex-
cite both QDs in order to control the initial collective
state on the Bloch sphere (FIG. 3a). In the magnetic
Zeeman field, the coupled QD transitions are orthogo-
nally (circularly) polarized [34], yet they are efficiently
coupled by the optical mode of the PCW. The phase be-
tween the two driving fields Ω2and Ω3eiθ is adjusted via
the polarization of the excitation laser. Using a single
laser, we implement θ≈ −π/2, which prepares an initial
state with the single excitation close to |egi+i|geifor
∆23 = 0. With detuning, the state either evolves towards
the super- (for ∆23 <0) or the subradiant (for ∆23 >0)
state. This results in a striking difference in the radia-
tion at a short time, followed by out-of-phase coherent
oscillations between the two components (FIG. 3b,c.
The experimental demonstration of this behavior
(FIG. 3e) is accurately described by the theory (FIG. 3d).
We observe a pronounced asymmetry around zero detun-
ing, where for positive detunings, the emission dynamics
is effectively delayed. This stems from the selective pop-
ulation of the subradiant state resulting in a lower emis-
sion at early times. Subsequently, the emission intensity
increases as the coherent evolution increases the popula-
tion of the superradiant state. The reverse behavior is
found for ∆23 <0. This is a result of the out-of-phase
oscillation of the population of the super- and subradiant
components (FIG. 3f,g). As opposed to the case where a
single QD was excited (FIG. 2), we observe here multiple
coherent oscillations. This is due to the state |egi+i|gei,
starting to coherently evolve on the Bloch sphere directly
after excitation. This is in contrast to the case where |egi
is prepared, and the coherent evolution sets in once the
state has partially decayed to |si. We track the maxi-
mum emission intensity (dashed line) for ∆23 >0 (and
minimum for ∆23 <0) in the plots of FIG. 3e. It is conse-
quently found that the collective light emission intensity
can be controlled via the QD detuning. By simultaneous
driving, we populate the doubly excited state |eeicompo-
nent, which reaches 0.07 for ≈π/3 excitation pulses that
are used. We note that the doubly excited state |eei,
however, does not contribute to the measured asymme-
try between positive and negative detunings.
Concluding remarks. Our observation of super- and
subradiant emission dynamics using pairs of QDs embed-
ded in PCWs and separated by a distance much larger
than the wavelength is facilitated by the PCW offer-
ing broadband spectral operation and long-range photon-
mediated coherent interaction between QDs. Our work
can constitute a foundational step towards multi-emitter
applications of technological importance, e.g., for real-
izing quantum transduction between microwave qubits
and the optical domain [35] or for quantum memories
with exponential improvement in photon storage fidelity
[31]. The ability to manipulate the super- and subradiant
state dynamics by controlling the detuning and pumping
conditions will lead to a whole new range of opportuni-
ties when implementing a coherent spin inside the QD
[34,36]. For instance, advanced photonic cluster states
may be generated deterministically [37], providing a uni-
versal resource for measurement-based photonic quantum
computing. To this end, waveguide-mediated dissipa-
tive coupling can be exploited to realize spin-spin en-
tanglement between distant quantum emitters [38]. An-
other direction will exploit photon scattering from cou-
pled QDs to realize efficient Bell-state measurements [39]
or photon-photon quantum gates [40]. Taking a broader
perspective, the ability to deterministically couple mul-
tiple quantum emitters opens a new arena of studying
non-equilibrium quantum many-body physics of strongly
correlated light and matter, which could be used in quan-
tum simulations of strongly correlated condensed matter
systems [41].
Acknowledgments: The authors thank Xiao-Liu
Chu for fruitful discussions at the beginning of the
project.
Funding: We gratefully acknowledge financial sup-
port from Danmarks Grundforskningsfond (DNRF 139,
Hy-Q Center for Hybrid Quantum Networks). B.S. ac-
knowledges financial support from Deutsche Forschungs-
gemeinschaft (DFG, German Research Foundation),
Grant No. 449674892. O.S. acknowledges funding from
the European Union’s Horizon 2020 research and innova-
tion programme under the Marie Sk lodowska-Curie grant
agreement No. 801199. S.S., A.L. and A.D.W. acknowl-
edge financial support of the German-French University
DFH/UFA within the CDFA-05-06 as well as BMBF
QR.X project 16 KISQ 009.
Author contributions: A.T., V.A., and C.J.v.D.
carried out the measurements. Y.W. and L.M. designed
and fabricated the sample. O.S., B.S. and A.S devel-
oped the theoretical model, A.T., V.A and C.J.v.D. an-
alyzed the data and prepared the figures. S.S., A.D.W.,
and A.L. carried out the growth and design of the wafer.
A.T., V.A., C.J.v.D., and P.L. wrote the manuscript with
input from all the authors. P.L. supervised the project.
Competing interests: P.L. is founder of the com-
pany Sparrow Quantum that commercializes single-
photon sources.
Data and materials availability: The data pre-
sented in the main text and Supplementary Materials
for this publication are openly available from [42].
[1] R. H. Dicke, Physical Review 93, 99 (1954).
[2] M. Gross and S. Haroche, Physics Reports 93, 301
(1982).
[3] R. G. DeVoe and R. G. Brewer, Physical Review Letters