# Fileset

[20240125 Elarabi_2024_Jpn._J._Appl._Phys._63_020801.pdf](https://mdr.nims.go.jp/filesets/1ae418be-5465-46e0-b42a-4f01236c8e9e/download)

## Creator

Asem Elarabi, Yoshito Saito, Hidehiro Asai, Ryota Kobayashi, Ken Hayama, Keiichiro Maeda, Shuma Fujita, Yusuke Yoshioka, [Yoshihiko Takano](https://orcid.org/0000-0002-1541-6928), Manabu Tsujimoto, Itsuhiro Kakeya

## Rights

© 2024 The Japan Society of Applied Physics
[In Copyright](http://rightsstatements.org/vocab/InC/1.0/)

## Other metadata

[Polarized terahertz electromagnetic-wave radiation from cuprate superconductor Bi2212 mesa structures](https://mdr.nims.go.jp/datasets/6a312c11-412a-4495-b3dd-2692ebc11277)

## Fulltext

Polarized terahertz electromagnetic-wave radiation from cuprate superconductor Bi2212 mesa structuresJapanese Journal of AppliedPhysics     PROGRESS REVIEWPolarized terahertz electromagnetic-wave radiationfrom cuprate superconductor Bi2212 mesastructuresTo cite this article: Asem Elarabi et al 2024 Jpn. J. Appl. Phys. 63 020801 View the article online for updates and enhancements.You may also likeFabrication of thin-film-typeBi2Sr2CaCu2O8+ intrinsic Josephsonjunctions by pulsed-laser-depositionMasakazu Haruta, Eiji Kume and ShigekiSakai-Engineering and characterization of apackaged high-Tc superconductingterahertz source moduleManabu Tsujimoto, Takuji Doi, GenkiKuwano et al.-C-axis electrical resistivity of PrO1aFaBiS2single crystalsMasanori Nagao, Akira Miura, SatoshiWatauchi et al.-This content was downloaded from IP address 144.213.253.16 on 15/04/2024 at 06:55https://doi.org/10.35848/1347-4065/ad0cdd/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/0953-2048/22/12/125004/article/10.1088/1361-6668/aa67aa/article/10.1088/1361-6668/aa67aa/article/10.1088/1361-6668/aa67aa/article/10.1088/1361-6668/aa67aa/article/10.1088/1361-6668/aa67aa/article/10.7567/JJAP.54.083101/article/10.7567/JJAP.54.083101/article/10.7567/JJAP.54.083101/article/10.7567/JJAP.54.083101https://pagead2.googlesyndication.com/pcs/click?xai=AKAOjstmqYHzpD2210UVSSARazNIWAjBqa8rtWmANqAFN9Rb9GR7T_9_FsHmLaBMI3MRYl3cOYQHZHRf8-IZ6eaTtPL5vWB_OFzNdO7mjscBFc_GjCKTUuecO7Ffj28fbjDRfqWh9HCsHk7pJOteGbhWGjdoJMzWCES6fEoEFrlCjW2h94txjd3oBc7Vgyp4_qIsexRrkX4TRxnfuF4u6BkxxZf17rqPUtjBfgj9FRzRgrYBOmEZR71PIRlW0bIOJED61CQMZiSSQqXepHSzAXjJmOJeDbNJlThW99ht3QylXjf7nY2BXy4rj_xGWjqeG2E2bjAJIJfqwwfhnXgVp0cElKw&sig=Cg0ArKJSzDVNT0oRcYp6&fbs_aeid=%5Bgw_fbsaeid%5D&adurl=https://ecs.confex.com/ecs/prime2024/cfp.cgi%3Futm_source%3DIOP%26utm_medium%3Dbanner%26utm_campaign%3Dprime_abstract_submissionPolarized terahertz electromagnetic-wave radiation from cuprate superconductorBi2212 mesa structuresAsem Elarabi1 , Yoshito Saito2,3, Hidehiro Asai4, Ryota Kobayashi1 , Ken Hayama1, Keiichiro Maeda1, Shuma Fujita1,Yusuke Yoshioka1, Yoshihiko Takano2,3 , Manabu Tsujimoto5 , and Itsuhiro Kakeya1*1Department of Electronic Science and Engineering, Kyoto University, Kyotodaigaku Katsura, Nishikyo, Kyoto 615-8510, Japan2International Center for Materials Nanoarchitectonics (MANA), National Institute for Materials Science, Tsukuba, Ibaraki 305-0047, Japan3Graduate School of Pure and Applied Sciences, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-8577, Japan4Semiconductor Frontier Research Center, National Institute of Advanced Industrial Science and Technology (AIST), Central 2, 1-1-1 Umezono,Tsukuba, Ibaraki 305-8568, Japan5Global Research and Development Center for Business by Quantum-AI Technology (G-QuAT), National Institute of Advanced Industrial Science andTechnology (AIST), Central 2, 1-1-1 Umezono, Tsukuba, Ibaraki 305-8568, Japan*E-mail: kakeya@kuee.kyoto-u.ac.jpReceived July 31, 2023; revised November 2, 2023; accepted November 15, 2023; published online January 4, 2024Polarized terahertz (THz) sources are important components in THz technologies. This paper highlights and discusses recent progress andmeasurement methods in the monolithic generation of polarized THz radiation using intrinsic Josephson junction oscillators made of high-Tcsuperconductors. The polarized radiation is generated from three mesa designs: truncated-edge square, notched cylindrical, and rectangularmesa structures. The polarization control depends on the excitation of two orthogonal TM modes in these mesas, comprising stacked intrinsicJosephson junctions in single crystalline Bi2Sr2CaCu2O8+δ. This method maintains a high output intensity and low axial ratios while avoiding thesignal loss associated with external polarimetric modulators prevalent in the THz frequency range. Moreover, it demonstrates the manipulation ofterahertz wave helicity by adjusting the current injection position, with experiments substantiating the device’s capability to switch between left-handed and right-handed elliptical polarization at designated frequencies. © 2024 The Japan Society of Applied Physics1. IntroductionElectromagnetic (EM) wave sources operating in the terahertz(THz) frequency range have recently attracted increasinginterest.1) Their potential applications span a wide range ofscientific and technological domains, from materials scienceand astronomy to ultra-high-speed communications,2,3) securitysystems,4) and biomedical research.5,6) To meet the diverseneeds of these applications, a variety of EM wave sources havebeen developed. Among these sources, semiconductor-baseddevices such as quantum cascade lasers7) and resonant tun-neling diode-based oscillators8) are often cited as highlypromising candidates. Superconducting devices made fromBi2Sr2CaCu2O8+δ (Bi2212), proved to be a strong candidatefor reliable THz sources.9–14)In the case of highly anisotropic superconductors like suchas Bi2212, it is possible to study the properties of individualintrinsic Josephson junctions (IJJs), revealing a rich array ofsuperconducting phenomena. These include the d-wave nodalsuperconducting gap, the first-order vortex lattice meltingtransition, macroscopic quantum tunneling at low tempera-tures, and EM wave emissions resulting from the synchro-nous oscillations of numerous IJJs.15) Terahertz EM waveemissions from IJJ stacks in Bi2212 single crystals wereinitially demonstrated in 2007 and were found to be the resultof synchronized Josephson plasma waves excited in IJJsaccording to the ac Josephson effect.9) We refer to suchdevices as Josephson plasma emitters (JPEs).11) To makeJPEs practical for telecommunication and imaging applica-tions, their output power must exceed 1 mW, which is thesensitivity limit of state-of-the-art terahertz cameras. Recentadvances in JPE research have focused on improvingfrequency tunability,16–18) operating temperature,19–21) andfabrication techniques.22,23) Although the radiation power ofthe JPEs still lags behind other THz sources, recent studieshave shown promise in improving the output power throughsynchronized emissions from coupled devices. The max-imum output power of a JPE attained by driving three mesastructures on a crystal is 0.61 mW, which is approximately 32times the individual emission power.24)Efforts to improve radiation efficiency and discover newfunctionalities for JPEs are of great significance in the field ofsuperconducting electronics. Efficient HF radiation inJosephson junctions is achieved through phase-locking toan external resonance condition. Various techniques havebeen studied to achieve phase-locking in one and two-dimensional JJs arrays, such as by coupling the array to anexternal LCR resonator, or using a common ground plane,and locking them by cavity resonance as explained in thesereferences.25,26) The power emitted by such an array isdirectly proportional to N2, where N is the number ofcontributing junctions. It has been confirmed that, based onits current–voltage (I−V ) characteristics, the layered struc-ture of Bi2212 functions as a series-connected one-dimen-sional array of Josephson junctions, leading to the termIJJs.27–29) By using Bi2212 IJJs, it was theoreticallyforeseen30–32) and experimentally realized9) that high radia-tion power can be achieved through phase-locked synchro-nized Josephson plasma waves in resonance with the cavity.The ability to manipulate radiation properties like direc-tionality and polarization in JPEs is a subject of particularinterest, as it expands the range of potential applications forthese devices. Theoretical models have predicted that emis-sions from rectangular mesas should exhibit linearpolarization:33,34) a prediction that has been corroboratedby experimental studies utilizing radiation patterns.35)Additionally, it has been theoretically shown that polarizationin JPEs can be modified through induction of thermalinhomogeneity on the surface of the device.36,37) Theseworks imply that polarization is an important measureincluding rich variery of physics not only electromagnetismbut also superconductivity and the patch anttena theory020801-1 © 2024 The Japan Society of Applied PhysicsJapanese Journal of Applied Physics 63, 020801 (2024) PROGRESS REVIEWhttps://doi.org/10.35848/1347-4065/ad0cddhttps://crossmark.crossref.org/dialog/?doi=10.35848/1347-4065/ad0cdd&domain=pdf&date_stamp=2024-01-04https://orcid.org/0000-0001-7541-3756https://orcid.org/0000-0001-7541-3756https://orcid.org/0009-0003-7915-3666https://orcid.org/0009-0003-7915-3666https://orcid.org/0000-0002-1541-6928https://orcid.org/0000-0002-1541-6928https://orcid.org/0000-0003-4296-5137https://orcid.org/0000-0003-4296-5137https://orcid.org/0000-0003-4999-2111https://orcid.org/0000-0003-4999-2111mailto:kakeya@kuee.kyoto-u.ac.jphttps://doi.org/10.35848/1347-4065/ad0cdddevelopped in microwave range38) is applicable for con-troling terahertz radiation from JPE. In this paper, we discussa subset of experiments related to this subject that we haveconducted. It is pertinent to note that some of the findings anddiscussions presented in this review have their origins in theauthor’s published thesis and other publications.39–48)2. Polarization in the THz rangePolarization control in the THz region has been a highlytargeted field for the past 30 years. Achieving such controlcould pave the way for numerous practical applications,ranging from telecommunications and imaging to spectro-scopy. Although polarization control of pulsed THz radiationhas been well documented in various studies using methodssuch as liquid crystal cells,49) laser pulse combinations,50,51)dual-color lasers,52,53) and femtosecond pulse modulators,54,55)it is generally more challenging to control polarization incontinuous-wave devices. Conventionally, circularly polarizedradiation has been achieved in laboratories using polarizationconverters or phase-retarders. While this is straightforward atoptical frequencies using birefringence or chiral effects ofanisotropic materials, such as quarter-wave plates, achievingthe requisite phase delay is more challenging at submillimeterwavelengths.56) Various alternative methods have beenexplored, including frequency selective surfaces,57–59)metasurfaces,59–62) waveguides,63) and artificially periodicmetamaterials.64–67)It has been previously demonstrated that the polarizationstate of spin-polarized lasers and LEDs can be tuned by anapplied external magnetic field,68,69) achieving a degree ofcircular polarization (DOCP) of up to 50%. A more practicalapproach for field applications would be to integrate polar-ization control into a single, unified structure, commonlyreferred to as monolithic polarization control. This wouldminimize radiation loss and simplify the overall system. Insupport of this concept, various studies have developedmonolithic circularly polarized terahertz sources. Thesesources typically use quantum cascade lasers (QCLs) as aradiation source and are covered with “fishbone” surfacegratings composed of orthogonally oriented apertureantennas.70,71) A high DOCP (≈ 98%) was observed usingthese types of sources.70) Liang et al. showed that THzradiation could be continuously tuned from linear to circularpolarization (CP) by electronically and monolithically inte-grating in-plane metasurfaces with two phase-locked semi-conductor-based THz QCLs.72) Furthermore, a device with aswitchable polarization state was also demonstrated.73)In the microwave region, polarization control finds appli-cations in diverse fields including telecommunications, radar,and navigation systems. There are two primary methods forachieving polarization control in this frequency range: the useof antennas and waveguides. The antenna approach becomesparticularly relevant when exploring THz radiation generatedby IJJs. The mesa structure in IJJs closely resembles a patchantenna, especially since its radiation is governed by reso-nance conditions.38,74,75) This similarity enables the applica-tion of antenna theory for the design and understanding ofIJJ-based devices to control radiation properties,40,41,76–78) atopic that will be explored more later in this review.3. Polarization of EM waves and Stokes formalismIn the case where the polarization is linear, the electric fieldoscillates in one direction. If the electric field rotates at aconstant rate in the direction of propagation, then thepolarization is considered elliptical or circular, dependingon the rotation properties. The rotation could be rotating tothe right (viewing from the detector to the emitter, termedleft-hand polarization) or rotating to the left (termed Right-Hand Polarization). The pioneering studies by Fresnel andArago showed that an EM wave consists of two orthogonallylinearly polarized electric field components:79–82)E z t E t kz, cos , 1x x x0 w d= - +( ) [ ] ( )E z t E t kz, cos , 2y y y0 w d= - +( ) [ ] ( )where, E0x and E0y represent the peak amplitudes, ωt− kzacts as the electric field propagator, z denotes the direction ofpropagation, t is indicative of time, while δx and δy symbolizethe phases at the x and y axis, both of which are perpendicularto the direction of propagation. If one were to map thetrajectory of this polarized electric field over time and space,the result would be an elliptical shape. This ellipse, known asFig. 1. (Lefft) Polarization ellipse. (Right) Poincaré sphere represents a visualization method to Stokes parameters, pointing to a point on the sphere surface(red dot). The length of the Stokes vector (S) corresponds to the total intensity (Itotal or S0), while the remaining Stokes parameters (S1, S2, S3) are denoted bytheir projections onto the x, y, and z axis, respectively. Reproduced from Ref. 39 with permission of Asem Elarabi.020801-2 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWthe polarization ellipse, is depicted in Fig. 1. The nature ofthis ellipse can be further detailed by:82)E tEE tEE t E tE E2cos sin . 3xxyyx yx y2022020 02d d+ - =( ) ( ) ( ) ( )( )The polarization state can be characterized using the geo-metric parameters of the polarization ellipse. This involvesthe use of the AR, which represents the ratio of the ellipse’smajor axis to its minor axis (a/b), or its inverse, known as theellipticity (b/a). The orientation angle (Ψ) is defined as theangle between the ellipse’s major axis and the x-axis. Theellipticity angle (χ), indicative of the polarization ellipse’s“roundness” or “stretch”, is also depicted in the figure. BothΨ and χ can be mathematically described according to thefollowing equations:38,82)E EE Etan 22 cos, 4x yx y0 00202dY = ( )E EE Esin 22 sin, 5x yx y0 00202cd=+( )ARabARmajoraxisminoraxis1 , 6= = ¥ ( ) whereb E E E E E E122 cos 2 , 7x y x y x y020204040202 d= + + + +( ) ( )anda E E E E E E122 cos 2 . 8x y x y x y020204040202 d= + - + +( ) ( )In order to provide a more comprehensive mathematicaldescription of the polarization state, four parameters aretypically used. These parameters, known as Stokes polariza-tion parameters (SPPs),83) are useful and quantifiable metricscapable of accurately characterizing fully polarized, partiallypolarized, and nonpolarized EM waves. They can be definedmathematically as:S E E , 9x y0 0202= + ( )S E E , 10x y1 0202= - ( )S E E2 cos , 11x y2 0 0 d= ( )S E E2 sin . 12x y3 0 0 d= ( )The first parameter S0 represents the total intensity of theEM wave. The second parameter S1 measures the value of thehorizontal and vertical linear polarization. S2 characterizesthe linear polarization at angles of 45◦ or −45◦. Lastly, S3indicates the rotation direction of the EM wave, specifyingwhether it is right-hand CP (RHCP) or left-hand CP (LHCP).Figure 2 provides a schematic representation of the states ofpolarization of EM waves with respect to their defining SPPs.To gauge the portion of unpolarized EM waves within thetotal measured radiation, an insightful expression known asthe degree of polarization (DOP) can be utilized. It is definedas the ratio of the intensity of polarized waves to the totalmeasured intensity, and can be expressed as follows:S S S SDOP ,0 DOP 1. 131222320= + +( )( ) If DOP= 0, this signifies that the entire measured radiationis unpolarized. On the contrary, if DOP= 1, it denotes that allmeasured radiation is polarized. When 0<DOP< 1, itimplies the presence of partially polarized radiation.An additional method to visualize the polarization state isthrough the use of a Poincaré sphere [Fig. 1]. The SPPs aredepicted as a vector originating from the center point (0,0,0)and extending to the sphere’s surface. Figure 2 furtherillustrates the states of polarization as represented by thePoincaré sphere for different SPPs.4. Experimental details4.1. Samples fabrication and measurement setupsHigh-quality single crystals of Bi2212 are prepared in-labusing the traveling solvent floating zone technique. The mesashape is then fabricated using photolithography and argon ionmilling.Figure 3 shows the electrical wiring block diagram. THzemission is detected using a lock-in amplifier (SanfordResearch System SR850), an optical chopper and a Sibolometer, as shown in Fig. 4. All measurements are madein a nitrogen-purged box. A thermally controlled He-flowcryostat equipped with an optical window was used. For thefrequency spectra measurements, a Martin–Puplett type FTIRspectrometer84) was employed for the truncated-edge square(TES) and rectangular mesa, while a split lamellar mirrorspectrometer was used for the notched-side cylindrical mesatype.85,86)4.2. Polarization measurementsThe polarization of a radiated EM wave is described by thecurve traced by the momentary electric field radiated by theantenna in a plane perpendicular to the radial direction.38)This projection takes on an elliptical shape.The polarization of sources is often characterized by theAR, the tilt of the polarization ellipse (Ψ), and the rotationdirection or helicity. In practice, it is challenging to achieve aradiation source that has a constant polarization state in alldirections. Therefore, the polarization is typically measuredat the point where the source is intended to be used. Acomprehensive characterization of polarization should in-clude not only AR and Ψ but also the direction of rotation(either right-handed or left-handed rotation). It should benoted that some of the results presented in this review wereobtained using the method known as the polarization-patternmethod. Although this method can determine the AR and tiltangle of the polarization ellipse, it does not provide informa-tion on the rotation direction.4.2.1. AR measurement with a wire-grid polarizer. Tomeasure AR and Ψ, the polarization-pattern method calls forthe use of a linearly polarized receiver as a probing device. Insome studies, a combination of a silicon bolometer with awire-grid polarizer (WGP) is utilized for this purpose. Asdepicted in Fig. 5(a), the THz emission source is mountedinside a cryostat. The angle of the WGP is controlled using astage controller and a motor. As the WGP rotates in the beampath [Fig. 5(a)], the change in intensity is recorded in relationto the angle of the polarizer. In cases of linear polarization,020801-3 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWthe angle-dependent intensity measured should resemble theshape of a peanut {Fig. 5(b)[left]}, where β is the rotationangle of the WGP relative to the reference direction. Forelliptical polarization, the central area of the peanut shapewidens {Fig. 5(b)[middle]}. In the context of CP, a minimalto zero change in intensity will be detected at any angle,causing the polarization pattern to appear as a circle{Fig. 5(b)[right]}. From these measurements, the polarizationellipse can be estimated as the tangent to the polarizationpattern, thereby providing the AR and the tilt angle. By fittingthe recorded intensity using a sine function, the AR in dB canbe determined by the following:IIAR 20 log 14maxmin= ( )where Imax and Imin are the maximum and minimumintensities, respectively. It is to be noted that polarizationwith an AR of less than 3 dB can be regarded as CP.87)Fig. 2. Schematic representation of the states of polarization with exclusive contributions of S1 (left), S2 (middle) and S3 (right) with a positive sign (top) anda negative sign (bottom). The orthogonal electric field components Ex (blue line) and Ey (cyan line) together with the phase difference (orange stroke), as wellas the projection of the resulting electric field radiation(black line) on the xy plane (red) are visible. Corresponding Poincaré spheres are presented for eachstate. Reproduced and modified from Ref. 39 with permission of Asem Elarabi.020801-4 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEW4.2.2. Stokes polarimetry with a WGP and a quarter-wave plate. The precise determination of SPPs throughStokes polarimetry is crucial for a comprehensive under-standing of the polarization state. Various techniques havebeen developed to measure these parameters. One conven-tional method employs a rotating polarizer and a quarter-wave plate (QWP), while another uses a fixed polarizer witha rotating QWP.80,82,88,89)The intensity of EM waves passing through a QWP andWGP can be expressed by:82,90)Fig. 3. Block diagram of the electrical setup. Reproduced from Ref. 39with permission of Asem Elarabi.Fig. 4. Sketch for the measurement setup used. Reproduced from Ref. 39with permission of Asem Elarabi.(a)(b)Fig. 5. (a) Sketch of the polarization (AR) measurement setup used. Reproduced from Ref. 40 with permission of Americal Physical Society.(b) Polarization patterns, The dashed curve represents the polarization ellipse. Reproduced from Ref. 39 with permission of Asem Elarabi.Fig. 6. Sketch of the SPPs measurement setup. Reproduced from Ref. 39with permission of Asem Elarabi.020801-5 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEW15IS S S S,12cos 2 cos sin 2 sin cos 2 ,0 1 2 3q fq f q f q= + + + ( )( )( )Where θ is the angle between the transmission axis of thepolarizer and the vertical axis, f is the phase retardation(fQWP= π/2). In this method, the parameters can be acquiredby measuring four intensities, in terms of f and θ, as follows:I S S0, 012, 160 1q f= = = +( ) [ ] ( )I S S4, 012, 170 2⎛⎝⎞⎠qpf= = = +[ ] ( )I S S2, 012, 180 1⎛⎝⎞⎠qpf= = = -[ ] ( )I S S4,212. 190 3⎛⎝⎞⎠qpfp= = = +[ ] ( )Upon solving for S0, S1, S2, S3 we can determine the Stokesparameters as follows:S I I0, 02, 0 , 200 ⎛⎝⎞⎠q f qpf= = = + = =( ) ( )S I I0, 02, 0 , 211 ⎛⎝⎞⎠q f qpf= = = - = =( ) ( )S II I24, 00, 02, 0 , 222 ⎛⎝⎞⎠⎛⎝⎞⎠qpfq f qpf= = =- = = - = =( ) ( )S II I24,20, 02, 0 . 233 ⎛⎝⎞⎠⎛⎝⎞⎠qpfpq f qpf= = =- = = - = =( ) ( )This conventional technique has certain limitations. It requiresmeasurements with and without QWP in the beam path(f= 0), which may not be practical for realistic measurementswithin a nitrogen-purged box and could result in variable totalintensity. Moreover, the accuracy may be compromised as onlyfour intensity points are measured. An improved method,adopted in some of our experiments, provides multiplebenefits, such as automation and stable intensity. This en-hanced technique involves a rotating QWP and a fixed WGP,is illustrated in Fig. 6 and detailed in Refs. 82, 88, 90.In this improved method, Muller matrices offer a simpleand effective means to describe the output state (Sout) of EMwaves passing through (Sin) an optical element:S SM . 24out in= ( )The Stokes vectors of the EM wave out of the QWP andfixed WGP using the Muller formalism are expressed by:82)S SM M , 25WGP QWP b¢ = ( ) ( )where β is the angle between the transmission axis of theQWP and the vertical axis. By solving the matrices, we canfind the outcome of S0 b¢( ), which is also known as the totalbeam intensity, as follows:I S S SS S12cos 2cos 2 sin 2 sin 2 . 260 0 122 3b b bb b b= ¢ = ++ +( ) ( ) ( ( )( ) ( ) ( )) ( )This equation can be used to find the Stokes parametersthrough a multiparameter fitting for the detected intensity.82,88)This can be accomplished either during measurements withdata acquisition software, or post-measurements using fittingtools.The DOP is calculated by Eq. (13) The QWP used in ourexperiment is not conventional (typically made of transparentmaterials like quartz). Instead, it employs parallel, puncturedmetallic plates to create a phase delay between the horizontal andvertical components of the passing electric field [Fig. 7 91,92)].This type of QWP benefits from a wide bandwidth and anadjustable target frequency. However, some asymmetry in itstransmittance properties were found during testing.5. Polarizations of rectangular mesas5.1. Long rectangular mesasThe first JPE was demonstrated on a long rectangular mesawith a high aspect ratio, measuring 300 μm on the long sideand 60–100 μm on the short side.9) It was pointed out that theemitted EM wave was polarized in the direction of the c-axisof the IJJ stacking. This conclusion was drawn based on thetransmission intensity ratio of the oscillating EM wave whenpassed through a parallel plate filter.Here we first discuss the consistency of AR measurementsusing only WGP and SPP measurements with the addition ofQWP for radiation from long rectangular mesas. We thencompare polarizations between mesas on the same singlecrystal with very close superconducting properties, and thencompare polarizations between mesas on different crystals todiscuss the effects of superconducting properties and devicegeometry on the modifications.Figures 8(a), 8(b) show the WGP angular (θ) dependenceof the detected intensity of mesa B1 at a bias voltage of1.05 V and the QWP angular (β) dependence of the SPPmeasurement system, respectively. It should be noted that thedevice names presented in this manuscript are identical tothose of the original papers if they are published. Whencomparing the ARs obtained listed in Table I, the deviation isabout 10 %, indicating a close match. On the other hand, thedifference with respect to the specified variable range of 180degrees is still about 10 %, although the orientation angle Ψseems to show a relatively large difference. Next, wecompare with the polarization in mesa B2 as shown inFigs. 8(c), 8(d). Bias voltages for B1 and B2 are fixed to theconditions that maximize the radiation intensity. The dataobtained from both the AR measurement and the SPPmeasurement agree very well; although the AR measurementof B2 shows a fairly large AR, the AR is greatly affected bythe intensity at the node, so it is difficult to say that this is anessential difference in radiation characteristics due to theambiguity in the origin of the intensity. From the above, itcan be seen that the mesas on the same crystal radiate withalmost identical polarization characteristics. On the otherhand, the polarization of mesa B1 on another crystal is shownin Figs. 8(e), 8(f).020801-6 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEW(a) (b)(c) (d)(e) (f)Fig. 8. Polar plots of AR (a), (c), (e) and SPP (b), (d), (f) measurements in mesas B1 (a), (b), B2 (c,d), and C1 (e), (f). Measured data are plotted as blacksymbols and fitted curves are drawn in red. Partially reproduced from Ref. 46 with permission of Americal Physical Society.Fig. 7. (a) Photo of the WGP. (b) Photo of QWP. (c) Photo of the plate perforations. Reproduced from Ref. 39 with permission of Asem Elarabi.020801-7 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWNext, we discuss the device structure and polarization.Polarization measurements of mesas B1 and B2 show that themajor axis of polarization, i.e. the major axis of the radiatedelectric field, is close to the short side direction of the mesastructure. This can be interpreted as the current flowing onthe device surface in the short-edge direction being the mainsource of radiation, which can be explained by the existingstanding waves in the short-edge direction alone. However,the results of C1 suggest that the major component originatesfrom the current flowing in the long-sided direction of themesa. In this case, it implies excitation of Josephson plasmastanding waves with finite wavenumber in the y-direction. Inboth cases, the principal axis of the polarization ellipse istilted more than 20 degrees from the x- or y-axis, indicatingthat the oscillating current has components in the x and ydirections. Although not with any accuracy that can beconsidered significant, the observation of the tilted polariza-tion ellipse, including the fact that it is observed, requires acorrection to the discussion of the radiation frequencyassuming a resonance mode, which has been done inprevious studies. The numerical calculations of the radiatedelectric field are important for a deeper understanding ofthese phenomena. In the next section, we discuss polarizationmeasurements and numerical calculations to reproduce themfor a rectangular mesa of low aspect ratio.5.2. Short rectangular or square mesasA wide range of continuous frequency tuning has beenreported for rectangular mesas with low aspect ratio (some-times disk-shaped) with near-square planar geometry sincethe early stages of research.35,43) However, identifyingresonance modes has proven challenging, leading to exten-sive discussions regarding the physics of inter-mode transi-tions. Conversely, in the research on radiation devices fromwhisker crystals,44) which advanced due to the simplicity ofdevice fabrication, radiation was obtained from mesa struc-tures with relatively low aspect ratios. However, the distinctchemical composition of whisker crystals means that knownvalues of the refractive index cannot be used, rendering theestimation of the mode challenging. To address this, wecompared the radiated EM field calculations with polarizationmeasurements using CST Studio Suite, a three-dimensionalEM field simulation software.Figure 9(a) shows the AR measurement and polarizationellipse for sample 1 of Ref. 45. To reproduce this polarizationellipse, a model is built consisting of a PEC antenna patchand a dielectric with a relative permittivity of òr= 26. Here,the feeding point of the terahertz current was taken near thecorner of the patch. Figure 9(c) shows the polarization ellipsefor each probe set up in the calculated space. The polarizationobserved in the experiment can be considered to be averagedover space. At 729 GHz, near the antenna’s resonant fre-quency (730 GHz), the polarization has a very large AR andis slightly tilted from horizontal. As we move away from theresonant frequency, we see a bulge in the polarization ellipse,and the AR becomes smaller while Ψ becomes larger. It canalso be clearly seen that the AR varies with the location of theprobe. Spatially averaged results show that 735 GHz bestreproduces the experiment. This means that to identify thespatial distribution of polarization is the key information toestimate the standing wave mode, and thus we propose this asa new research method.6. Polarization of synchronously driven mesasIn the past, attempts have been made to increase the radiationintensity of JPEs by synchronizing the oscillations acrossmultiple mesa devices, aiming to achieve a higher outputpower compared to operating them individually.93) Thepinnacle of these efforts is marked by a recorded radiationoutput of 0.6 mW when three mesas were biasedsimultaneously.24) This increment in output is hypothesizedto stem from the coherent radiation of each mesa. However,no experiments have conclusively demonstrated phase syn-chronization between the mesas: only a nonlinear enhance-ment in radiation intensity and a retraction of the radiationfrequency have been observed.Drawing from the insights on polarization measurementsdelineated in the previous section, we propose a novelapproach to analyze the phenomenon of mutual synchroniza-tion. By measuring the polarization of EM waves emittedfrom multiple mesas, whether biased singly or simulta-neously, we aim to dive deeper into the underlying mechanicsof this synchronization.6.1. Two mesas synchronizationFigure 10 presents the results when mesas B1 and B2,described in Sect. 5.1 are connected in parallel and biased.The electric field radiated when the mesa is individuallybiased is expressed as EB1 and EB2, while the electric field inthe parallel bias configuration, denoted EB1∥B2 is expressed asa linear coupling with single bias radiation as the basis. Inother words, the complex numbers α and β given byE E E , 27B B B B1 2 1 2a b= + ( )facilitating a mathematical description of the two-mesasinteraction. SPPs for the parallel connection of B1 and B2are also shown in Table I. The derived values from Table Iare ∣β/α∣= 0.98 and arg 68.7b a =( ) degrees. To evaluatethe validity of the coefficients obtained, we compared thephase difference of the coupled standing waves estimatedfrom arg b a( ) and the distance between the mesas. As aresult, a phase difference dependent on the distance betweenthe mesas was found by comparing the individual bias andthe parallel connection bias.46) Furthermore, the analysis ofthe coupling coefficients for another combination of mesadevices B and E, both in parallel and in series connection, atmultiple frequencies of synchronous oscillation with aspecific basis, showed a systematic change in the declinationwith the synchronous frequency.47) The results also show thatthe difficulty of synchronization changes between parallelTable I. Summary of obtained polarization parameters by AR (upper rows)and SPP (lower rows) measurements. Here, normalized SPPsS S S i 1, 2, 3i i 0= =˜ ( ) are listed.Mesa B1 B2 C1 B1∥B2AR (dB) 8.26 13.6 8.99 5.85Ψ (deg) 10.7 11.8 −69.3 −3.22S1̃ 0.298 0.395 −0.60 0.48S2˜ 0.384 0.478 −0.60 0.01S3˜ −0.142 −0.179 −0.13 −0.15DOP 0.50 0.65 0.86 0.50AR (dB) 8.35 8.50 11.3 8.06Ψ (deg) 25.6 25.2 −67.7 0.51020801-8 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWand series connections due to the margin of synchronizationconditions and differences in individual radiation character-istics.The proposed analysis methodology for synchronousradiation phenomena, when applied to relatively diversemesa structures, promises a pathway towards controllingsynchronous oscillation across numerous diverse mesa de-vices. This progression is anticipated to lead to the develop-ment of superconducting devices with enhanced radiationintensity and diversified characteristics, paving the way forreal-world applications like ultra-high-speed communicationdevices.6.2. Three mesas synchronizationFigure 10(a) shows the result of radiation with mesas B, C,and E connected in parallel.47) The maximum radiationstrength is slightly weaker than the sum of the maximumradiation strengths of the two mesas, but significantly higherthan any of the two mesas in parallel [Fig. 10(b)]. Theradiation spectrum at voltages close to the maximumintensity [Fig. 10(c)] is unimodal, indicating that the threemesas oscillate synchronously. The relation between theradiation frequency and the average mesa voltage is shown inFig. 10(d). The Stokes polarization measurements of the BCEparallel connection are shown in [Fig. 10(e)], and theestimated Stokes parameters of the BCE parallel connectionare detailed in Table II. Here, adjustments have been madefor the intensity reduction due to the tilt of the polarizationanalyzer. The values to be obtained are the linear combina-tion coefficients α, β, and γ. These coefficients use theoscillating electric field in a BCE stand-alone operation as thebasis to reproduce the Stokes parameters in BCE parallelconnection by their linear combination.7. Circularly polarized radiationCircularly polarized terahertz waves are useful for inter-mobile communications and in circular dichroism spectro-scopy to identify chemical substances. Circularly polarizedEM waves have been conventionally obtained in the micro-wave region by supplying high-frequency current to aspecially designed patch antenna and in the optical regionby inserting a QWP. In this experiment, we applied the patchantenna concept to demonstrate a monolithic circularlypolarized terahertz source using a superconductor.7.1. TES and notched cylindrical mesasFigure 11(a) shows the JPE mesa device S1 that radiatescircularly polarized terahertz waves with the highest DOCPof 99.7%. The transmission intensity dependence on thelinear polarizer’s angle is shown in the polar coordinate plot[Fig. 11(b)]. Although the DOCP is strongly dependent onthe applied bias voltage or current, it appears relatively stablenear the bias voltage that provides the maximum DOCP. Themaximum DOCP of the device, or the minimum ARmin, isrelated to the ratio of the length of the truncated edge a1 tothe remaining edge length a2.Figure 11(a) illustrates the relationship between the ARand the radiation frequency for a disk mesa device C2, withparts of the outer edges truncated. This behavior is thought tobe attributed to the trapezoidal cross section of the mesa(a) (b)(c) (d) (e)(f) (g) (h)Fig. 9. (a) AR measurement in whisker samples. (b) Calculated anttena model shown by a screenshot of CST studio suite. (c)–(h) Simulated polarizationellipses at 729 and 741 GHz. Virtual probes located at the center (c), (f), right (d), (g), and top (e), (h) with the view shown in (b).020801-9 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWstructure, which induces an entrainment effect betweenJosephson junctions with differing natural frequencies.AR measurements were carried out on four truncated-edgemesas and two notched cylindrical mesas. The results aresummarized in Table III. Chirality is defined as the positionof the perturbed part relative to the current-applying electrode asviewed from the device center. For sample S1 shown inFig. 11(a), the chirality is right, and the opposite is left. Highdegrees of CP less than AR= 1 dB are achieved for elements ofboth chiralities, with the truncated-edge mesa exhibiting theclosest approximation to CP around a2/a1≈ 4. We also find thata larger thickness t yields higher maximum radiation intensities,which is consistent with the known fact that the intensity isroughly proportional to the square of the stack number.7.2. Helicity switchingCircularly polarized radiation arises from the resolution of thedegeneracy between RHCP and LHCP, which initially existin a state of linear polarization. Resolution of this degeneracyis facilitated by two primary factors: the chirality of theperturbed region and the position of current injection withinthe device. These factors influence the local critical current,leading to a decrease in its value.(a)(b)(c)(d)(e)(f)Fig. 10. (a) Current-voltage characteristics of mesa B (red), C (blue), E (green), and all of them connected in parallel (gray). Maximum radiation points werefound in open circles for respective colors. The current for the gray plot is divided by three. (b) Comparison of maximum radiation intensities. (c) Fouriertransform spectra when mesas B, C, and E are biased in parallel. (d) Radiation frequencies versus applied voltages. (e) Polar plot of SPP measurements for twovoltage values of the BCE parallel connection. Asymmetric loss of QWP47) is considered for the fitting curves. (f) Polarization ellipese for fitting curves in (e).Table II. Summary of obtained polarization parameters of three mesasradiation. S i 1, 2, 3i =˜ ( ) denotes a normalized Stokes parameters, as shownin Table I. Data in mesas B, C and E are selected to be as close to the biascondition of the parallel operation as possible.Mesa B C E B&C&EBias voltage (V) 0.94 0.92 0.94 0.95 1.00S0 0.99 0.34 0.56 0.31 0.14S1̃ 0.43 0.94 0.59 0.92 0.89S2˜ 0.62 −0.048 0.22 0.19 0.45S3˜ −0.66 -0.35 0.78 −0.33 −0.037DOP 0.79 0.98 0.77 0.61 0.40AR (dB) 4.26 7.47 3.22 7.64 17.3Ψ (deg) 27.6 −1.48 10.3 5.7 13.3020801-10 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWThe reduction in the local critical current is instrumental inachieving the desired helicity switch, allowing for thetransition between RHCP and LHCP. To demonstrate thishelicity switch within a single device, we fabricated a devicecapable of altering its chirality. By doing so, we can switchthe helicity of the radiated EM wave. We further investigatedthe radiated terahertz wave emanating from the device toobserve the effects of helicity switching.SPP measurement results of the device shown in Fig. 12(a)when current is injected from electrode L are presented inFig. 12(b). The fitting line to obtain the Stokes parameter isshown as a solid line. The radiation frequency was found to be564 GHz and the normalized Stokes vector wasS 0.073, 0.867, 0.491= -˜ ( ). This means that left-handedelliptical polarization was detected. When a bias is applied toelectrode R, the helicity switch is successfully achieved at(a)(b)(c)Fig. 11. (a) TES mesa device S1 for CP. The white areas represent electrodes for bias application and measurement. Reproduced from Ref. 40 withpermission of Americal Physical Society. (b) Plot in polar coordinates showing the variation of transmission intensity with the polarizer angle under conditionsof optimal CP. Reproduced from Ref. 40 with permission of Americal Physical Society. (c) Radiation frequency dependence of the AR in a disk mesa withtrimmed cylindrical mesas. The plots marked with inverted triangle symbols represent experimental data, while the red curves are numerical results based onthe patch antenna model. The interpolated plots depict the measured radiation frequency dependence on bias voltage. Reproduced from Ref. 42 with permissionof AIP Publishing.Table III. Summary of geometrical and polarization characteristics of CPmesas. a1 and a2 are truncated and untruncated length of the truncated-edgesquares. r1 is the depth of the notch and r2 is the rudius of whole disk in thenotched disks. S and ΔS are areas of unperturbed and perturbed parts. t is thethickness of the mesa. Tb and Tc are measured bath temperature and thesuperconducting transition temperature. Pmax and ARmin are measuredmaximum intensity and minimum AR.S1 S2 S3 S4 C1 C2Chirality Right Right Left Left Left Lefta1, r1 (μm) 16 20 13 18 6 11.2a2, r2 (μm) 70 76 69 68 42 46a2/a1, r2/r1 4.38 3.8 5.3 3.77 5.3 4.1ΔS/S (%) 3.46 4.34 2.51 4.38 3.46 3.37t (μm) 2.25 1.9 2.4 2.4 1.9 2.3Tb (K) 21 22 40 21 30 16Tc (K) 84 83 82 84 78 85Pmax (nW) 176.5 23.5 123.5 88.3 15.3 70ARmin (dB) 0.2 0.49 4.6 0.67 0.8 2.4020801-11 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEW570 GHz with S 0.197, 0.867, 0.456= - -˜ ( ), which meansthat the right-handed elliptical polarization is detected.Figure 12(c) shows the radiation frequency dependence of theradiation intensity and Stokes parameters calculated using thesine-Gordon equation for an idealized mesa geometry as shownin the inset. The calculated S 03 >˜ obtained around themeasured radiation frequency of 564 GHz is consistent withthe experimental results. However, the values of S1̃ and S2˜ arequite different from the experiment. This may be due to the factthat not only the actual planar shape of the mesa differs,especially with respect to the angle of the cut, but also that themesa has a larger cross-sectional area toward the bottom nearthe substrate, which requires a 3D model for accurate descrip-tion. A paper is currently being written on the results of detailedexperimental data analysis and numerical simulation refinement.8. SummaryThis manuscript highlights recent progress in exploringpolarized terahertz radiations generated by IJJ emitters. Therelization of CP is led by two distinct designs: a TES mesastructure and a notched cylindrical structure, demonstrating theutility of antenna theory and EM simulations in achieving thedesired polarization. The methodology, based on the excitationof IJJs within a mesa cavity, effectively addresses the insertionloss challenges associated with external polarimetric modula-tors in the terahertz frequency range. A brief discussion ofpolarization characterization methods is provided, with exam-inations that include long and short rectangular mesas, whiskerdevices, and multiple synchronously driven mesas. Althoughthe helicity of the emitted EM wave is a crucial feature ofcircularly polarized waves, its comprehensive understandingremains not fully understood despite recent strides. Theongoing research in circularly polarized terahertz radiation ispoised to drive advancements in practical realms like mobilecommunications and circular dichroism spectroscopy, high-lighting the practical significance of the findings.AcknowledgmentsThis work was supported by the Japan Society for thePromotion of Science (JSPS) KAKENHI (Grant Nos.23K17747, 20H02606, 15KK0204), JSPS—Centre nationalde la recherche scientifique (CNRS) Bilateral Program (GrantNo. 120 192 908), and the Murata Science foundation.ORCID iDsAsem Elarabi https://orcid.org/0000-0001-7541-3756Ryota Kobayashi https://orcid.org/0009-0003-7915-3666Yoshihiko Takano https://orcid.org/0000-0002-1541-6928Manabu Tsujimoto https://orcid.org/0000-0003-4296-5137Itsuhiro Kakeya https://orcid.org/0000-0003-4999-21111) M. Tonouchi, Nat. Photonics 1, 97 (2007).2) B. Ferguson and X.-C. Zhang, Nat. Mater. 1, 26 (2002).3) S. S. Dhillon et al., J. Phys. D. Appl. Phys. 50, 043001 (2017).4) J. F. Federici, B. Schulkin, F. Huang, D. Gary, R. Barat, F. Oliveira, andD. Zimdars, Semicond Sci. Technol. 20, S266 (2005).5) K. Kawase, Y. Ogawa, Y. Watanabe, and H. Inoue, Opt. Express 11, 2549(2003).(a)(b)(c)Fig. 12. (a) Polar plot of SPP measurement when current is applied to the electrode “L” of the device shown in (b), which is referred to as DL. (c) Expectedradiation power (upper) and SPPs (lower) as functions of frequency obtained by solving sine-Gordon equation in a model corresponding to DL.020801-12 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWhttps://orcid.org/0000-0001-7541-3756https://orcid.org/0000-0001-7541-3756https://orcid.org/0000-0001-7541-3756https://orcid.org/0009-0003-7915-3666https://orcid.org/0009-0003-7915-3666https://orcid.org/0009-0003-7915-3666https://orcid.org/0000-0002-1541-6928https://orcid.org/0000-0002-1541-6928https://orcid.org/0000-0002-1541-6928https://orcid.org/0000-0003-4296-5137https://orcid.org/0000-0003-4296-5137https://orcid.org/0000-0003-4296-5137https://orcid.org/0000-0003-4999-2111https://orcid.org/0000-0003-4999-2111https://orcid.org/0000-0003-4999-2111https://doi.org/10.1038/nphoton.2007.3https://doi.org/10.1038/nmat708https://doi.org/10.1088/1361-6463/50/4/043001https://doi.org/10.1088/0268-1242/20/7/018https://doi.org/10.1364/OE.11.002549https://doi.org/10.1364/OE.11.0025496) X. C. Zhang, A. Shkurinov, and Y. Zhang, Nat. Photonics 11, 16 (2017).7) B. S. Williams, Nat. Photonics 1, 517 (2007).8) M. Asada and S. Suzuki, J. Infrared Millim. Terahertz Waves. 37, 1185(2016).9) L. Ozyuzer et al., Science 318, 1291 (2007).10) U. Welp, K. Kadowaki, and R. Kleiner, Nat. Photonics 7, 702 (2013).11) I. Kakeya and H. Wang, Supercond. Sci. Technol. 29, 073001 (2016).12) T. Kashiwagi et al., Supercond. Sci. Technol. 30, 074008 (2017).13) R. Kleiner and H. Wang, J. Appl. Phys. 126, 171101 (2019).14) K. Delfanazari, R. A. Klemm, H. J. Joyce, D. A. Ritchie, and K. Kadowaki,Proc. IEEE 108, 721 (2020).15) I. Kakeya, IEICE Trans. Electron. E106.C, 2022SEI0004 (2023).16) T. Kashiwagi et al., Appl. Phys. Lett. 106, 092601 (2015).17) H. Sun et al., Appl. Sci. (Switzerland) 13, 3469 (2023).18) K. Kihlstrom et al., Phys. Rev. Appl. 19, 034055 (2023).19) L. Y. Hao et al., Phys. Rev. Appl. 3, 024006 (2015).20) H. Minami et al., J. Phys.: Condens. Matter 28, 025701 (2016).21) H. Sun et al., Phys. Rev. Appl. 10, 024041 (2018).22) T. Imai, T. Kashiwagi, S. Nakagawa, M. Nakayama, J. Kim, G. Kuwano,M. Tsujimoto, H. Minami, and K. Kadowaki, Jpn. J. Appl. Phys. 60, 126501(2021).23) T. Kashiwagi et al., IEICE Trans. Electron. E106.C, 281 (2023).24) T. M. Benseman et al., Appl. Phys. Lett. 103, 022602 (2013).25) D. Tilley, Phys. Lett. A 33, 205 (1970).26) K. Wiesenfeld, S. P. Benz, and P. A. A. Booi, J. Appl. Phys. 76, 3835(1994).27) R. Kleiner, F. Steinmeyer, G. Kunkel, and P. Müller, Phys. Rev. Lett. 68,2394 (1992).28) R. Kleiner and P. Müller, Phys. Rev. B 49, 1327 (1994).29) K. Schlenga et al., Phys. Rev. B 57, 14518 (1998).30) M. Tachiki, T. Koyama, and S. Takahashi, Phys. Rev. B 50, 7065 (1994).31) T. Koyama and M. Tachiki, Solid State Commun. 96, 367 (1995).32) M. Tachiki, M. Iizuka, K. Minami, S. Tejima, and H. Nakamura, Phys. Rev.B 71, 134515 (2005).33) S.-Z. Lin and X. Hu, Phys. Rev. B 82, 020504 (2010).34) R. A. Klemm and K. Kadowaki, J. Phys.: Condens. Matter 22, 375701(2010).35) M. Tsujimoto, K. Yamaki, K. Deguchi, T. Yamamoto, T. Kashiwagi,H. Minami, M. Tachiki, K. Kadowaki, and R. A. Klemm, Phys. Rev. Lett.105, 037005 (2010).36) H. Asai and S. Kawabata, Appl. Phys. Lett. 110, 132601 (2017).37) H. Asai and S. Kawabata, IEEE Trans. Appl. Supercond. 26, 1800804(2016).38) C. Balanis, Antenna Theory: Analysis and Design (Wiley, New York, 2015).39) A. Elarabi, “Polarization behavior of high-Tc superconducting terahertzemitters,” PhD Thesis Kyoto University (2018).40) A. Elarabi, Y. Yoshioka, M. Tsujimoto, and I. Kakeya, Phys. Rev. Appl. 8,064034 (2017).41) A. Elarabi, Y. Yoshioka, M. Tsujimoto, Y. Nakagawa, and I. Kakeya, Phys.Procedia 81, 133 (2016).42) A. Elarabi, Y. Yoshioka, M. Tsujimoto, and I. Kakeya, Appl. Phys. Lett.113, 052601 (2018).43) M. Tsujimoto et al., Phys. Rev. Lett. 108, 107006 (2012).44) Y. Saito et al., Appl. Phys. Express 14 (2021).45) Y. Saito, I. Kakeya, and Y. Takano, Appl. Phys. Lett. 121, 212601 (2022).46) M. Tsujimoto et al., Phys. Rev. Appl. 13, 051001 (2020).47) R. Kobayashi, K. Hayama, S. Fujita, M. Tsujimoto, and I. Kakeya, Phys.Rev. Appl. 17, 054043 (2022).48) K. Maeda, S. Fujita, A. Elarabi, M. Tsujimoto, and I. Kakeya, 2019 XLIVInt. Conf. Infrared, Millimeter, Terahertz Waves (IEEE), 2019, p. 1.49) M. Sato et al., Nat. Photonics 7, 724 (2013).50) R. Takeda, N. Kida, M. Sotome, Y. Matsui, and H. Okamoto, Phys. Rev. A89, 033832 (2014).51) N. Amer, W. C. Hurlbut, B. J. Norton, Y.-S. S. Lee, and T. B. Norris, Appl.Phys. Lett. 87, 1 (2005).52) J. Dai, N. Karpowicz, and X.-C. Zhang, Phys. Rev. Lett. 103, 023001(2009).53) Y. You, T. Oh, and K. Kim, Opt. Lett. 38, 1034 (2013).54) R. Shimano, H. Nishimura, and T. Sato, Jpn. J. Appl. Phys. 44, L676 (2005).55) Y. Chen et al., Appl. Phys. Lett. 93, 3 (2008).56) S.-Y. Wang, W. Liu, and W. Geyi, Appl. Phys. B 124, 126 (2018).57) C. Zhang et al., IEEE Trans. Antennas Propag. 65, 385 (2017).58) B. Yan, M. Jansen, and C. Felser, Nat. Phys. 9, 709 (2013).59) B. Lin, J.-L. Wu, X.-Y. Da, W. Li, and J.-J. Ma, Appl. Phys. A 123, 43(2017).60) H. Chen et al., Appl. Phys. A 122, 463 (2016).61) G. Zhou et al., Sci. Rep. 6, 38925 (2016).62) Z. Li, W. Liu, H. Cheng, S. Chen, and J. Tian, Sci. Rep. 5, 18106 (2016).63) K. Saito, T. Tanabe, and Y. Oyama, Opt. Express 20, 26082 (2012).64) F. Li et al., Appl. Phys. B 124, 28 (2018).65) X. Ma et al., Adv. Opt. Mater. 2, 945 (2014).66) Y. Tamayama, K. Yasui, T. Nakanishi, and M. Kitano, Appl. Phys. Lett.105, 021110 (2014).67) L. Wu et al., Appl. Phys. A 116, 643 (2014).68) X. Jiang et al., Phys. Rev. Lett. 94, 056601 (2005).69) M. Holub, J. Shin, D. Saha, and P. Bhattacharya, Phys. Rev. Lett. 98,146603 (2007).70) P. Rauter et al., Proc. Natl. Acad. Sci. 111, E5623 (2014).71) N. Yu et al., Appl. Phys. Lett. 151101, 1 (2009).72) G. Liang et al., ACS Photonics 4, 517 (2017).73) L. Xu et al., Optica 4, 468 (2017).74) M. Tsujimoto et al., Supercond. Sci. Technol. 28, 105015 (2015).75) R. A. Klemm et al., IOP Conf. Ser.: Mater. Sci. Eng. 279, 012017 (2017).76) G. Kuwano et al., J. Appl. Phys. 129 (2021).77) M. Tsujimoto et al., Opt. Express 29, 16980 (2021).78) H. Minami, Y. Ono, Y. Saito, T. Yuhara, R. Kikuchi, T. Kashiwagi,M. Tsujimoto, and K. Kadowaki, Proc. XXIX Int. Conf. Low Temp. Phys.(Journal of the Physical Society of Japan), 2023, 10.7566/JPSCP.38.011047.79) W. A. Shurcliff, Polarized Light: Production and Use (Harvard UniversityPress, Cambridge, MA, 1962).80) E. Collett, “Polarized light: fundamentals and applications,” OpticalEngineering (Marcel Dekker, New York, 1993).81) C. Brosseau, Fundamentals of Polarized Light: A Statistical OpticsApproach (Wiley, New York, 1998).82) A. Molitor, “Investigations on Stokes polarization parameters of Vertical-Cavity Surface- Emitting Lasers : spatially, spectrally and time-resolved,”PhD Thesis Technische Universitat Darmstadt (2015).83) G. G. Stokes, “On the composition and resolution of streams of polarizedlight from different sources,” in Mathematical and Physical Papers(Cambridge Univ. Press, Cambridge, U.K, 2009) 1st ed., Vol. 3, p. 233.84) D. H. Martin and E. Puplett, Infrared Phys. 10, 105 (1970).85) R. L. Henry and D. B. Tanner, Infrared Phys. 9, 59 (1969).86) O. Ferhanoglu, H. R. Seren, S. Luettjohann, and H. Urey, Opt. Express 17,21289 (2009).87) S. Gao, Q. Luo, and F. Zhu, Circularly Polarized Antennas (Wiley,Chichester, UK, 2014), 10.1002/9781118790526.88) H. G. Berry, G. Gabrielse, and A. E. Livingston, Appl. Opt. 16, 3200 (1977).89) B. Schaefer, E. Collett, R. Smyth, D. Barrett, and B. Fraher, Am. J. Phys. 75,163 (2007).90) B. Schaefer, E. Collett, R. Smyth, D. Barrett, and B. Fraher, Am. J. Phys. 75,163 (2007).91) N. Mukai et al., Int. Conf. on Infrared, Millimeter, and Terahertz Waves,IRMMW-THz, 2014, Vol. 39, p. 146.92) M. Nagai et al., Opt. Express 23, 4641 (2015).93) N. Orita, H. Minami, T. Koike, T. Yamamoto, and K. Kadowaki, Phys. C:Supercond. Appl. 470, S786 (2010).020801-13 © 2024 The Japan Society of Applied PhysicsJpn. J. Appl. Phys. 63, 020801 (2024) PROGRESS REVIEWhttps://doi.org/10.1038/nphoton.2016.249https://doi.org/10.1038/nphoton.2007.166https://doi.org/10.1007/s10762-016-0321-6https://doi.org/10.1007/s10762-016-0321-6https://doi.org/10.1126/science.1149802https://doi.org/10.1038/nphoton.2013.216https://doi.org/10.1088/0953-2048/29/7/073001https://doi.org/10.1088/1361-6668/aa6dd7https://doi.org/10.1063/1.5116660https://doi.org/1IRMMW-THz.2/JPROC.2019.2958810https://doi.org/10.1587/transele.2022SEI0004https://doi.org/10.1063/1.4914083https://doi.org/10.3390/app13063469https://doi.org/10.1103/PhysRevApplied.19.034055https://doi.org/10.1103/PhysRevApplied.3.024006https://doi.org/10.1088/0953-8984/28/2/025701https://doi.org/10.1103/PhysRevApplied.10.024041https://doi.org/10.35848/1347-4065/ac3024https://doi.org/10.35848/1347-4065/ac3024https://doi.org/10.1587/transele.2022SEI0001https://doi.org/10.1063/1.4813536https://doi.org/10.1016/0375-9601(70)90735-8https://doi.org/10.1063/1.357387https://doi.org/10.1063/1.357387https://doi.org/10.1103/PhysRevLett.68.2394https://doi.org/10.1103/PhysRevLett.68.2394https://doi.org/10.1103/PhysRevB.49.1327https://doi.org/10.1103/PhysRevB.57.14518https://doi.org/10.1103/PhysRevB.50.7065https://doi.org/10.1016/0038-1098(95)00474-2https://doi.org/10.1103/PhysRevB.71.134515https://doi.org/10.1103/PhysRevB.71.134515https://doi.org/10.1103/PhysRevB.82.020504https://doi.org/10.1088/0953-8984/22/37/375701https://doi.org/10.1088/0953-8984/22/37/375701https://doi.org/10.1103/PhysRevLett.105.037005https://doi.org/10.1103/PhysRevLett.105.037005https://doi.org/10.1063/1.4979317https://doi.org/1IRMMW-THz.2/TASC.2016.2515853https://doi.org/1IRMMW-THz.2/TASC.2016.2515853https://doi.org/10.1103/PhysRevApplied.8.064034https://doi.org/10.1103/PhysRevApplied.8.064034https://doi.org/10.1016/j.phpro.2016.04.022https://doi.org/10.1016/j.phpro.2016.04.022https://doi.org/10.1063/1.5040159https://doi.org/10.1063/1.5040159https://doi.org/10.1103/PhysRevLett.108.107006https://doi.org/10.35848/1882-0786/abe166https://doi.org/10.1063/5.0123290https://doi.org/10.1103/PhysRevApplied.13.051001https://doi.org/10.1103/PhysRevApplied.17.054043https://doi.org/10.1103/PhysRevApplied.17.054043https://doi.org/1IRMMW-THz.2/IRMMW-THz.2019.8874509https://doi.org/1IRMMW-THz.2/IRMMW-THz.2019.8874509https://doi.org/10.1038/nphoton.2013.213https://doi.org/10.1103/PhysRevA.89.033832https://doi.org/10.1103/PhysRevA.89.033832https://doi.org/10.1063/1.2138351https://doi.org/10.1063/1.2138351https://doi.org/10.1103/PhysRevLett.103.023001https://doi.org/10.1103/PhysRevLett.103.023001https://doi.org/10.1364/OL.38.001034https://doi.org/10.1143/JJAP.44.L676https://doi.org/10.1063/1.3046781https://doi.org/10.1007/s00340-018-6997-7https://doi.org/1IRMMW-THz.2/TAP.2016.2623653https://doi.org/10.1038/nphys2762https://doi.org/10.1007/s00339-016-0673-8https://doi.org/10.1007/s00339-016-0673-8https://doi.org/10.1007/s00339-016-9799-yhttps://doi.org/10.1038/srep38925https://doi.org/10.1038/srep18106https://doi.org/10.1364/OE.20.026082https://doi.org/10.1007/s00340-018-6900-6https://doi.org/10.1002/adom.201400212https://doi.org/10.1063/1.4890623https://doi.org/10.1063/1.4890623https://doi.org/10.1007/s00339-014-8252-3https://doi.org/10.1103/PhysRevLett.94.056601https://doi.org/10.1103/PhysRevLett.98.146603https://doi.org/10.1103/PhysRevLett.98.146603https://doi.org/10.1073/pnas.1421991112https://doi.org/10.1063/1.3093476https://doi.org/10.1021/acsphotonics.6b00703https://doi.org/10.1364/OPTICA.4.000468https://doi.org/10.1088/0953-2048/28/10/105015https://doi.org/10.1088/1757-899X/279/1/012017https://doi.org/10.1063/5.0054018https://doi.org/10.1364/OE.420417https://doi.org/10.7566/JPSCP.38.011047https://doi.org/10.7566/JPSCP.38.011047https://doi.org/10.1016/0020-0891(70)90006-0https://doi.org/10.1016/0020-0891(69)90012-8https://doi.org/10.1364/OE.17.021289https://doi.org/10.1364/OE.17.021289https://doi.org/10.1002/9781118790526https://doi.org/10.1364/AO.16.003200https://doi.org/10.1119/1.2386162https://doi.org/10.1119/1.2386162https://doi.org/10.1119/1.2386162https://doi.org/10.1119/1.2386162https://doi.org/10.1364/OE.23.004641https://doi.org/10.1016/j.physc.2010.01.051https://doi.org/10.1016/j.physc.2010.01.051 1. Introduction 2. Polarization in the THz range 3. Polarization of EM waves and Stokes formalism 4. Experimental details 4.1. Samples fabrication and measurement setups 4.2. Polarization measurements 4.2.1. AR measurement with a wire-grid polarizer 4.2.2. Stokes polarimetry with a WGP and a quarter-wave plate 5. Polarizations of rectangular mesas 5.1. Long rectangular mesas 5.2. Short rectangular or square mesas 6. Polarization of synchronously driven mesas 6.1. Two mesas synchronization 6.2. Three mesas synchronization 7. Circularly polarized radiation 7.1. TES and notched cylindrical mesas 7.2. Helicity switching 8. Summary Acknowledgments A10