0% found this document useful (0 votes)
8 views5 pages

Supporting Information NL

The document discusses the methodology for achieving higher-order exceptional points in terahertz metasurfaces, detailing numerical simulations, sample fabrication, and terahertz frequency-domain spectroscopy measurements. It outlines the processes involved in creating multilayer SRR structures and the mathematical framework for understanding the characteristic polynomial related to exceptional points. Additionally, it presents experimental results that validate the theoretical predictions regarding the behavior of the system near these exceptional points.

Uploaded by

fuyueyue1122
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
8 views5 pages

Supporting Information NL

The document discusses the methodology for achieving higher-order exceptional points in terahertz metasurfaces, detailing numerical simulations, sample fabrication, and terahertz frequency-domain spectroscopy measurements. It outlines the processes involved in creating multilayer SRR structures and the mathematical framework for understanding the characteristic polynomial related to exceptional points. Additionally, it presents experimental results that validate the theoretical predictions regarding the behavior of the system near these exceptional points.

Uploaded by

fuyueyue1122
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 5

Supporting Information for Achieving

Higher-order Exceptional Points in Terahertz


Metasurface

Peng Fu,†,‡,# Pai Peng,¶,# Wenze Lan,†,§,# Shuo Du,† Jiawei Shao,¶ Leyong

Hu,† Chensheng Li,† Yuyang Zhang,‡ Zi-Lan Deng,∥ Baoli Liu,∗,†,‡ Luyi

Yang,∗,¶,⊥ and Changzhi Gu∗,†,‡

†Beijing National Laboratory for Condensed Matter Physics, Institute of Physics, Chinese
Academy of Sciences, Beijing 100190, China
‡School of Physical Sciences, CAS Key Laboratory of Vacuum Physics, University of
Chinese Academy of Sciences, Beijing 100190, China
¶State Key Laboratory of Low Dimensional Quantum Physics and Department of Physics,
Tsinghua University, Beijing, 100084, China
§Institute of Condensed Matter Physics, Technische Universität Darmstadt, 64289
Darmstadt, Germany
∥Institute of Photonics Technology, College of Physics &Optoelectronic Engineering, Jinan
University, Guangzhou 510632, China
⊥Frontier Science Center for Quantum Information, Beijing 100084, China
#Contributed equally to this work

E-mail: [email protected]; [email protected]; [email protected]

1
Numerical Simulation

The COMSOL multiphysics eigen solver calculates the complex frequencies of the tri-SRR
metasurfaces. The silicon substrate is represented as a lossless dielectric with permittivity of
ϵ = 11.56, whereas the metal is modeled as a PEC. Periodic boundary conditions are applied
in the x- and y-directions and PMLs are imposed in the z-direction.

Sample Fabrication

A double-sided polished silicon wafer, 2 mm thick and of high purity (001), was selected as
the substrate due to its high resistivity (>10 kΩ cm), which ensured optimal THz trans-
mission. Before microfabrication, the substrate underwent ultrasonic cleaning in acetone,
isopropanol, and ultrapure water, followed by drying with nitrogen gas. The fabrication
process comprised five steps: First, a 1 µm thick layer of UV photoresist (SPR955-CM
0.7) was spin-coated onto the substrate at 4000 rpm, then pre-baked at 90C for 90 seconds
to harden the resist. Second, the first layer of SRR array structures was patterned using
standard photolithography (DWL66+). Then, a 200 nm gold layer was deposited via elec-
tron beam evaporation (EBE, FU12PEB). The unwanted material was removed through a
lift-off process in acetone using ultrasonic assistance. A 3 µm thick SU8-2002 layer was spin-
coated at 1000 rpm, pre-baked at 95C for 2 minutes, and exposed to UV light (MA6) with
180 mJ/cm2 energy. The sample was hard-baked at 300C for 30 minutes to form a robust
spacer. Finally, the second layer of SRR arrays was patterned using the same procedure,
with precise alignment. This sequential process ensured precise fabrication of the multilayer
SRR structures.

Terahertz Frequency-Domain Spectroscopy Measurements:

The measurements were conducted utilizing a commercially available terahertz spectrome-


ter(TeraFlash Pro by TOPTICA Photonics). This THz time-domain spectroscopy apparatus

2
operates based on a fiber laser featuring a central wavelength of 1560 nm and an 80 fs pulse
duration, with a repetition rate of 100 MHz. The majority of the pulses serve to excite an
InGaAs photoconductive emitter, thereby generating linearly polarized THz pulses. A minor
portion of the beam traverses a delay stage before being directed onto another InGaAs pho-
toconductive receiver. The resulting spectral range is 0.1-3THz. To mitigate the detrimental
effects of water vapor absorption on THz radiation, the THz beam path is purged with dry
nitrogen gas.

The characteristic polynomial of EP3

An EP3 is closely related to third-order algebraic multiplicity in the roots of characteristic


polynomial P (ω) of the system, which is given by:

P (ω) = det(H 2 − ω 2 I)

This means that P (ω) has a triple root at a specific frequency ω EP 3 . To describe this
mathematically, we expand P (ω) around ω EP 3 using a Taylor series as follows:

1 (1) EP 3 1
P (ω) = P (ω EP 3 ) + P (ω )(ω − ω EP 3 ) + P (2) (ω EP 3 )(ω − ω EP 3 )2 + ...
1! 2!

Since P (ω EP 3 ) (because ω EP 3 is a root), and for a third-order root, both the first and
second derivatives must vanish at ω EP 3 , we have the following conditions:

P (ω EP 3 ) = 0, P (1) (ω EP 3 ) = 0, P (2) (ω EP 3 ) = 0

These conditions ensure that ω EP 3 is a triple root of P (ω).


According to the Taylor expansion, the third derivative P (3) (ω EP 3 ) must be nonzero to
guarantee that the root at ω EP 3 is a third-order root. Therefore, the expansion takes the

3
form:
1 (3)
P (ω) ≈ P (ω − ω EP 3 )3
3!

This indicates that near ω EP 3 , the characteristic polynomial behaves as a cubic function,
and the behavior of the system near this frequency is primarily determined by the third
derivative of P (ω).

a b

Re Re
[ ] [ ]
] Re[ ] Re[

c d
]

]
Re[

Re[

Im[ ] Im[ ]

Figure 1: Theory results of the real and imaginary parts of the eigenmode in the intricate
complex coupling coefficient spaces. (a, b) The construction of the eigenmode space when
the coupling coefficient κ2 is a real number. (c, d) The construction of the eigenmode space
when the coupling coefficient κ2 is an imaginary number.

4
Figure 2: Proof of the coupling relationship. (a, b) The evolution of the eigenvalue of the
two resonators SRR1 and SRR3 with b3 , When b3 increases from 40 µm to 60 µm, the
resonance frequency of the SRR3 decreases from 0.64 THz to 0.53 THz (blue dotted line in
Fig. S2a), and the eigenvalues after coupling (fixing d2 = ±12 µm) are nearly unchanged.
(c) Transmission spectrum as b3 increase. (d, e) SRR1 and SRR2 are in the same layer with
symmetry in the y-direction, as d1 increases change the eigenfrequencies’ behavior, proving
that they exhibit weak coupling. (f) The transmission spectra always present one peak and
redshift as the coupling distance d1 increases

Figure 3: Compares the fitted transmission spectrum with the experimental data.

You might also like