Science - USA (2021-12-03)

SCIENCE science.org

As with the Sycamore processor, this
atomic quantum simulator is programma-
ble. The authors ran programs on the quan-
tum processor and monitored the long-
range entanglement properties among
the processor’s 219 atoms. Specifically, the
authors measured how these quantum
correlations establish themselves among
the spins along a meandering path, which
produce data that directly reflect the topo-
logical order of the quantum phase. Like
Satzinger et al., Semeghini et al. demon-
strated how quantum information can be
encoded into the system, by showing that a
bit of encoded information can be read out
again, and established a path for creating a
quantum memory.
The two experiments represent the first
definitive detection of a topologically or-
dered phase with time-reversal symmetry.
Neither experiment was achieved by us-
ing new materials, as is usually the case.
Instead, the achievement was realized
virtually with quantum processors. And
although the processors provide a mecha-
nism to create quantum states entangled at
long range, their most critical contribution
is to provide a way to measure the long-
range entanglement structure that charac-
terizes topological order (see the figure).
Satzinger et al. and Semeghini et al. illus-
trate the potential for quantum computer
technology to serve as a tool for exploring
quantum many-body systems.
To protect quantum information dur-
ing computation, it will also be necessary
to initialize, manipulate, and measure the
quantum information in these codes and to
do so by using circuits that are tolerant to
errors. For practical applications, the error
rates must be reduced even further from
the levels achieved by the two experiments.
And critically, quantum error correction
will require repetitive measurements of the
check operators for detecting errors and to
update the logical information by decod-
ing these measurements. This will require
not only a powerful quantum processor
but also a highly integrated classical pro-
cesser and controller ( 5 ). However, despite
the challenges still in the way of achieving
a practically useful quantum processing
device, the two experiments mark the first
steps toward harnessing topologically or-
dered quantum phases for error correction
in quantum computers. j

REFERENCES AND NOTES

1. K. J. Satzinger et al., Science 374 , 1237 (2021).


2. G. Semeghini et al., Science 374 , 1242 (2021).


3. F. Arute et al., Nature 574 , 505 (2019).


4. A. Kitaev, Ann. Phys. 303 , 2 (2003).


5. S. J. Pauka et al., Nat. Electron. 4 , 64 (2021).

10.1126/science.abl8910

INSTRUMENTS

Room-temperature

mid-infrared detectors

Two independent groups designed

nanoantennas for detecting mid-infrared light

By Reuven Gordon

F

rom astronomy to the surveying of

greenhouse gases, a wide range of sci-

ence and engineering applications rely

on the detection of mid-infrared (mid-

IR) photons. However, because pho-

tons from the mid-IR have less than a

tenth of the energy of the visible range, the

detection is not as easy as taking a photo

with a camera. IR detectors are generally

very sensitive to thermal noise and must be

kept cool, but for applications where a cryo-

genic setup is impractical, other engineer-

ing solutions must be used. On pages 1264

and 1268 of this issue, Chen et al. ( 1 ) and

Xomalis et al. ( 2 ), respectively, demonstrate

two techniques to “up-convert” mid-IR light

into the near-IR band, where it can be de-

tected using standard room-temperature

semiconductor detectors.

A common solution to increase the en-

ergy of IR photons so that they can be

measured efficiently is the use of a pro-

cess known as up-conversion. This process

shifts up the frequency, and therefore the

energy, of a photon so that it can be sensed

using ordinary detectors. This is often

realized by combining the incoming mid-

IR photons with near-IR photons from a

pump laser to produce a new photon with

a higher energy. However, current design

schemes to accomplish this require compli-

cated matching of the propagation phase

of the mid-IR photons from the source and

the near-IR photons from the pump laser.

Chen et al. and Xomalis et al. each pres-

ent an antenna design to sidestep the size

and momentum mismatch issues that

make existing up-conversion schemes inef-

ficient. Both teams achieve high detection

efficiency with four orders of magnitude

less near-IR laser power. Whereas Chen

et al. used a mid-IR slot antenna that

resonantly confines the mid-IR energy to

a narrow slot in a metal film, Xomalis et

al. used a mid-IR disk antenna that uses a

gold film to accomplish the same (see the

figure). Both devices boost the interaction

between near-IR and mid-IR light by using

gold nanospheres in conjunction with mid-

IR antennas to confine the light in the gap

region containing biphenyl-4-thiol mole-

cules. Both schemes combine two antennas

in the same location to interact efficiently

with these molecules without the need for

complicated phase-matching—one mid-IR

antenna and one near-IR one.

As the incoming mid-IR light ap-

proaches the antenna from the underside

and the pump laser from the topside, they

are focused down to an area 1 nm across.

This goes well beyond the diffraction limit

of the microscope objectives used in the

setup. Using IR light with a wavelength of

several microns and a beam width of the

same size, the antennas used by Chen et al.

and Xomalis et al. reduced the wavelength

and beam width of the incoming IR light

down to the nanometer range. Both de-

signs achieved a focusing factor of 10,000

beyond the conventional diffraction limit

to enhance the interaction with the mol-

ecules in the gap. In addition, the focus-

ing factor for the near-IR light was around

1000. The resulting near-IR photon from
      the up-conversion is only slightly larger in
      energy than the incident light photons, so
      it also benefits from this extreme focusing.
      The focused IR light is absorbed by a
      molecule placed inside a nanocavity within
      the antenna, which causes the molecules to
      vibrate. The energy from this vibration is
      then transferred to a photon from a near-IR
      laser in a process called anti-Stokes Raman
      scattering. After passing through a series of
      filters for cleaning up the signal, the photon
      from the anti-Stokes Raman scattering is
      detected using a conventional detector.
      Raman scattering and IR absorption
      require vibrational modes with differ-
      ent symmetries. Raman scattering probes
      vibrations of even symmetry, and IR ab-
      sorption probes odd. This means that the
      molecular vibration needs to be chosen
      to have both even and odd parts for this
      up-conversion scheme to work. With this
      in mind, both studies used the biphenyl-
      4-thiol as an up-converter. This molecule
      not only has a specific type of bond that

Department of Electrical and Computer Engineering,

University of Victoria, Victoria, BC, Canada.

Email: [email protected]

3 DECEMBER 2021 • VOL 374 ISSUE 6572 1201