Development of the Soft X-ray AGM-AGS RIXS Beamline at Taiwan Photon Source
A. Singh, H. Y. Huang, Y. Y. Chu, C. Y. Hua, S. W. Lin, H. S. Fung, H. W. Shiu, J. Chang, J. H. Li, J. Okamoto, C. C. Chiu, C. H. Chang, W. B. Wu, S. Y. Perng, S. C. Chung, K. Y. Kao, S. C. Yeh, H. Y. Chao, J. H. Chen, D. J. Huang, C. T. Chen
DDevelopment of the Soft X-ray AGM-AGS RIXS Beamline at Taiwan Photon Source
A. Singh, H. Y. Huang, Y. Y. Chu, C. Y. Hua, S. W. Lin, H. S. Fung, H. W. Shiu, J.Chang, J. H. Li, J. Okamoto, C. C. Chiu, C. H. Chang, W. B. Wu, S. Y. Perng, S. C.Chung, K. Y. Kao, S. C. Yeh, H. Y. Chao, J. H. Chen, D. J. Huang,
1, 2, ∗ and C. T. Chen National Synchrotron Radiation Research Center, Hsinchu 30076, Taiwan Department of Physics, National Tsing Hua University, Hsinchu 30013, Taiwan (Dated: June 25, 2020)
ABSTRACT
We report on the development of a high-resolution andhighly efficient beamline for soft-X-ray resonant inelas-tic X-ray scattering (RIXS) located at Taiwan PhotonSource. This beamline adopts an optical design thatuses an active grating monochromator (AGM) and anactive grating spectrometer (AGS) to implement the en-ergy compensation principle of grating dispersion. Ac-tive gratings are utilized to diminish defocus, coma andhigher-order aberrations as well as to decrease the slopeerrors caused by thermal deformation and optical polish-ing. The AGS is mounted on a rotatable granite plat-form to enable momentum-resolved RIXS measurementswith scattering angle over a wide range. Several high-precision instruments developed in house for this beam-line are briefly described. The best energy resolutionobtained from this AGM-AGS beamline was 12.4 meVat 530 eV, achieving a resolving power 42,000, whilethe bandwidth of the incident soft X-rays was kept at0.5 eV. To demonstrate the scientific impacts of high-resolution RIXS, we present an example of momentum-resolved RIXS measurements on a high-temperature su-perconducting cuprate, La − x Sr x CuO . The measure-ments reveal the A g apical oxygen phonons in super-conducting cuprates, opening a new opportunity to in-vestigate the coupling between these phonons and chargedensity waves. INTRODUCTION
The energy dispersion of low-energy elementary excita-tions in momentum space reflects the fundamental physi-cal properties of materials. Resonant inelastic X-ray scat-tering (RIXS) is a powerful technique to probe these ex-citations with momentum resolution, and provides directinformation about the dynamics arising from fluctuationsof spin, charge and orbital degrees of freedom [1, 2].The process of X-ray absorption in a material and itssubsequent re-emission of an X-ray of different energy isknown as inelastic X-ray scattering (IXS) [3]. If the en-ergy of the incident photons is tuned to an absorptionresonance in which a core-level electron is excited to anunoccupied state, the subsequent X-ray emission spec- trum depends strongly on the incident photon energy;this process is called resonant IXS, i.e., RIXS [1, 2]. It isalso a scattering process in which the energy and momen-tum of the scattered X-ray conform to conservation rules,thus providing information about the energy and mo-mentum of elementary excitations, such as d - d , charge-transfer, plasmon, magnon and phonon excitations etc.of quantum materials. The resonance effect significantlyenhances the scattering cross section and offers a probeof elementary excitations with elemental and chemicalselectivity. In addition, RIXS is a photon-in and photon-out technique that has been applied to explore matter invarious phases.Despite its unique advantages, RIXS was unpopularamong available spectrometric techniques because of alack of good energy resolution and a very weak signalintensity. In the past decade, RIXS has, however, be-come widely accepted as one of the most powerful toolsto investigate the properties of materials in terms of el-ementary excitations. After extensive developments ofinstrumentation, significant improvements in energy res-olution and measurement efficiency have been achieved inthe regime of soft X-ray energy. For example, the energyresolution of the AXES monochromator and spectrome-ter [4, 5] improved from 500 meV at 530 eV in 1996 to50 meV in 2013 [6–9] on replacing the microchannel platedetector with a charge-coupled device (CCD) detector [8]and switching to a Dragon-type monochromator [10, 11].These improvements allowed researchers to study the d - d excitations of 3 d transition-metal oxides with effec-tive energy resolution [12, 13]. The SAXES spectrom-eter at Swiss Light Source [9] enabled measurements onmagnetic excitations in cuprate superconductors [14, 15].Furthermore, beamline ID32 at European SynchrotronRadiation Facility achieved resolution 30 meV at the Cu L -edge [16]. This beamline enables RIXS intensity map-ping as a function of momentum transfer through the ro-tation of the spectrometer in ultra-high vacuum (UHV).A polarimeter has also been installed allowing the polar-ization analysis of scattered photons. Recent years haveseen an enhanced development of new high-resolutionsoft X-ray RIXS instruments, including beamline I21 atDiamond Light Source, beamline SIX at National Syn-chrotron Light Source II [17], the VERITAS soft X-rayRIXS beamline at MAX IV, the soft-X-ray spectrometerPEAXIS at BESSY II [18] and beamline 41A at Taiwan a r X i v : . [ phy s i c s . i n s - d e t ] J un Photon Source (TPS) [19].To meet the two stringent requirements–high resolu-tion and high efficiency–in soft X-ray RIXS experiments,a design concept of an active grating monochromator(AGM) and an active grating spectrometer (AGS) basedon the principle of energy compensation of grating dis-persion was conceived in 2002 and publised in 2004 [20].In the AGM-AGS design, the efficiency of RIXS mea-surements becomes greatly enhanced on increasing thebandwidth of the incident photons, while maintainingthe energy resolution. The energy-compensation princi-ple for RIXS has been successfully tested at Taiwan LightSource (TLS) beamline 05A [21]. Our theoretical simu-lations indicate that a resolving power better than 10 for photon energies from 400 eV to 1000 eV is achievablewith an AGM-AGS beamline, motivating us to build anew soft X-ray RIXS beamline at TPS [19].In this paper, we report on the development of the softX-ray AGM-AGS RIXS beamline located at TPS port41A. This paper is organized as follows. In Sec. 2, weintroduce the RIXS branch of TPS 41A, including thedesign concept, a summary of precision instruments de-veloped in house for this beamline and their performance.In Sec. 3, we discuss commissioning results, includingadjustment of the grating surface profile, resolution op-timization and RIXS measurements on the phonon ex-citations of high-temperature superconducting cuprates(HTSC). A summary and future plan follow in Sec. 4. RIXS OF TPS BEAMLINE 41APhoton source
Beamline 41A at TPS composes two branches, i.e.,high-resolution RIXS and coherent soft X-ray scattering.Both branches share the same monochromator, slits andfront-end focusing optics. The photon source originatesfrom two elliptically polarized undulators (EPU) in tan-dem in a 12-m straight section with a double-minimum β function to enhance the brilliance. Each EPU magnethas length 3.2 m and period 48 mm. The brilliance of theEPU tandem is designed to be greater than 1 × pho-tons s − mrad − mm − per 0.1 % BW in the energy rangefrom 400 eV to 1200 eV; the photon flux of the centralcone exceeds 1 × photons s − . In this energy range,the calculated beam sizes are about 386 µ m and 28-35 µ m at the full width at half maximum (FWHM) in thehorizontal and vertical directions, respectively; the beamdivergences are, respectively, 42 − µ rad and µ radat FWHM, in the horizontal and vertical directions, de-pending on photon energy. Detailed parameters of EPU48 are reported elsewhere [22, 23]. FIG. 1. Optical layout of AGM-AGS RIXS. Abbreviations ofoptical elements are defined in the text. Distances betweenoptical elements are summarized in Tables 1 and 2.FIG. 2. Photographs of the RIXS setup installed at TPSbeamline 41A. The AGS grating and the EMCCD detectorare placed on two separate granite platforms. Through anair-cushion mechanism, the spectrometer can swing over awide scattering angle from 17 ◦ to 163 ◦ about a vertical axisat the sample position. Inset photographs show the water-cooled active grating and slit assemblies. Focusing optics and slits
The RIXS branch includes the front end, focusing mir-rors, slits, AGM grating, sample station, AGS gratingand detector. Figure 1 illustrates its optical layout; Fig-ure 2 shows the beamline photographs. The first opticalelement is a horizontal focusing mirror (HFM) locatedat 25.805 m from the center of the EPU tandem. Afterthat, a vertical plane mirror (VPM) and a vertical focus-ing mirror (VFM) are located at 1.2 m and 2.4 m from theHFM, respectively. The VFM focuses the photon beamonto a vertical slit, i.e., the entrance slit of the AGM,with demagnification 28.2. The entrance-slit assembnlycomprises two water-cooled diamond blades. Through aflexure-based high-precision positioning mechanism, thetranslation and the tilt of the two blades are controlledindependently with four actuators with resolutions 0.002 µ m and 0.02 µ rad, respectively. The opening of the en-trance slit and its center can be varied continuously from0 to 4000 µ m. The monochromator uses an active gratinglocated 4 m after the vertical entrance slit to focus theincident soft X-rays vertically onto the sample located TABLE I. Major parameters of mirrors used at the RIXSbranch of TPS beamline 41A. All mirrors have a Au-coatedsurface. HFM is with a Glidcop substrate; others are with aSi substrate. All values of r , r , and radius are given in unitsof m.optic type r r radius ∗ deviation angle ( ◦ )HFM cylindrical 25.805 3.100 211 3VPM plane ∞ ∞ ∞ active 28.205 1.000 74 3M plane-elliptical 6.200 0.600 42 3M plane-elliptical 0.750 9.500 36 4 ∗ radius at the mirror center HFM, VPM and VFM are also known as M , M and M ,respectively. µ m, can be selected.The exit slit placed 27.5 mm before the sample also setsthe vertical beam size of the incident soft X-rays on thesample. In the horizontal direction, soft X-rays are fo-cused in two steps, first with the HFM which has demag-nification 8.3, followed by a horizontal refocusing mirrormirror M placed 0.6 m before the sample with demagni-fication 10.3. The designed horizontal beam size on thesample is 4.5 µ m at FWHM. The soft X-rays scatteredfrom the sample are focused with another horizontal fo-cusing mirror M which has a collection angle 18 mradto enhance the efficiency of the AGS. The spectrome-ter uses an active grating located 2.5 m after the sampleto focus the scattered soft X-rays vertically onto a two-dimensional (2D) detector located 5.5 m after the AGSgrating. The major parameters of the focusing mirrorsare summarized in Table I. AGMAGS scheme
The design of the monochromator and spectrometeris based on the energy-compensation principle of grat-ing dispersion. Incident X-rays from the entrance slitare diffracted, dispersed and focused onto the samplewith the AGM grating; similarly, scattered X-rays arediffracted, dispersed and focused onto the 2D detectorwith the AGS grating. The entrance and exit arms ofAGM are, respectively, r and r ; those of AGS are r (cid:48) and r (cid:48) , respectively. The AGM-AGS scheme requiresthat the two gratings have an identical groove density n at the grating center and that r (cid:48) = r . As shown in TABLE II. Optical parameters of AGM and AGS gratings.Both gratings have a laminar groove profile with duty ratio2:3 and groove depth between 8 nm and 6 nm with a position-dependent density n ( x ) = n + n x , where x is defined inSection 3.2. AGM AGSSize (L × W × T) ( mm ) 186 × ×
10 186 × × r = 4.0 r (cid:48) = 2.5Exit arm (m) r = 2.5 r (cid:48) = 5.5 n (grooves mm − ) 1200 1200 n (grooves mm − ) 0.80 -0.32 (low-energy)-0.13 (high-energy) Fig. 1, the inelastically scattered X-rays with the sameenergy loss but different incident energies have the samedispersion from the AGS, so implementing the energy-compensation principle of grating dispersion. The AGM-AGS scheme has two important features. (1) The resolu-tion and spectral-weight distribution of RIXS are insen-sitive to the incident bandwidth as long as the selectedbandwidth is smaller than the core-hole lifetime width;(2) The measurement efficiency is proportional to the se-lected bandwidth, because the energy-loss spectrum isthe summation of the inelastic scattering excited by theincident photons within the bandwidth. Therefore, theAGM-AGS scheme can greatly enhance the efficiency ofRIXS measurement while maintaining the energy resolu-tion.
Active gratings
To implement the AGM-AGS scheme for high-resolution RIXS, two high-precision active gratings arerequired. Using a multi-actuator bender, one can obtaina surface profile of grating with a high-degree polynomialto diminish defocus, coma and higher-order aberrationsas well as to decrease the slope errors caused by thermaldeformation and optical polishing. In this RIXS beam-line, each active grating is made of a varied-line-spacingplane grating mounted on a high-precision 25-actuatorbender [24]; its surface slope is monitored with an in-position long-trace profiler (LTP) [25]. Table II lists theoptical parameters of the AGM and AGS gratings.
Optical tables and AGS rotation platform
The implementation of high-resolution RIXS requiresmechanical adjustments and supports with great preci-sion and stability for all mirrors and gratings. To ful-fill these requirements, a high-precision, highly rigid andhigh-load optical table was designed and constructed forHFM, VPM, VFM, AGM and AGS. This optical tablecan support a UHV chamber for an optical element upto 1000 kg in weight with adjustments in all six degreesof freedom, i.e., three translations and three rotations.The resolution and repeatability of translation adjust-ments are 0.01 µ m and 0.05 µ m, respectively; those ofrotation adjustments are 0.02 µ rad and 0.1 µ rad, respec-tively. With comparable resolution and repeatability, anall-flexure-made high-precision optical table with adjust-ments in five degrees of freedom was designed and con-structed for M , M and the 2D detector.To facilitate the variation of scattering angle, hori-zontal refocusing mirror M , the AGS grating and the2D detector are mounted on a rotational platform com-posed of two movable air-cushioned granite blocks witha high-precision quickly detachable connecting bridge inbetween, as shown in Fig. 2. This design can min-imize the ground micro-vibrational effects during dataacquisition. M and the AGS grating are placed on oneblock, the 2D detector on the other. During a rotationof the platform, the two blocks are connected with thebridge; the air gap between the granite block and thegranite floor is kept at 30-50 µ m. Coupled with a uniqueUHV-compatible scattering chamber without a differen-tial pumping, a wide scattering angle, 146 ◦ , i.e., from 17 ◦ to 163 ◦ , can be achieved. Sample manipulation
The sample manipulator enables linear motions alongthree orthogonal directions with resolution 1 µ m. Thesemotions are driven with a translational stage using UHVstepper motors and optical encoders. Figure 3 shows aphotograph of this manipulator, which is mounted ontop of an in-vacuum single-axis goniometer. The sam-ple holder made of oxygen-free high-conductivity cop-per is connected to a liquid-He cryostat through copperbraids. The sample holder is thermally isolated from thelinear stage through a Vespel rod. The sample can betransferred in vacuum through a load-lock chamber andcan be cooled to 20 K. In addition, the sample holderis electrically insulated from the cryostat, which enablesus to perform measurements of X-ray absorption spectra(XAS) in the total-electron-yield mode. To minimize theradiation-induced damage and saturation of the detector,two shutters have been installed, one before mirror M and the other before the 2D detector. FIG. 3. Photograph of the sample manipulator inside theRIXS chamber. The sample xyz stage was developed usingUHV compatible stepper motors and optical encoders. Thesample position can be aligned with the incident X-ray beamwith a resolution 1 µ m. Two photodiodes are installed tomeasure the incident photon flux and fluorescence.FIG. 4. RIXS 2D image detector. A photograph of the de-tector mounted on a granite floor is shown in the left. Detailsof a customized EMCCD sensor located inside the chambermarked with a red circle are revealed in the right. Soft X-ray 2D detector
The scattering cross section of RIXS is typically smalland its signal is weak. To achieve a detection scheme nearphoton counting, a detector of high sensitivity and lownoise is required because only a few hundred electron-holepairs can be generated with a silicon-based detector forsoft X-rays. Figure 4 shows a photograph of a custom-made electron-multiplication CCD (EMCCD) detectorwithout anti-reflection coating at grazing angle 10 ◦ . Thepixel size of the EMCCD is 13.5 µ m; the grazing inci-dence gives an effective pixel size 2.3 µ m. The EMCCDis back-illuminated to avoid absorption of photons by the FIG. 5. Screenshot of the user interface, which is composed offour panels and one CLI. Panel 1 shows the measured RIXS2D image. Panel 2 displays the experimental parameters inreal time. Panels 3 and 4 show the command history and the1D spectra for RIXS or XAS, respectively. integrated circuits on the front side of the wafer. The de-tector head is cooled to − ◦ C. Through the electron-multiplication mechanism, the signal-to-noise ratio is en-hanced. In addition, two four-jaw apertures are installedbefore both gratings to block stray light.
Instrumentation control and user interface
The beamline instrumentation control for data acc-quistion is implemented on three levels: (1) the intrin-sic commands of commercial devices, (2) the algorithmand software for a group control of commercial devices,(3) the user-level commands and interface. Levels 2 and 3are developed in house using FORTRAN and Python lan-guages, respectively; the communications between themare based on EPICS tools. Figure 5 shows the screenof the user interface, which is composed of four pan-els and a user-oriented command-line input (CLI). Thefirst panel displays the RIXS 2D image recorded by theEMCCD. The second panel displays the real-time sta-tus of measurement parameters, including the positionsand orientations of optical elements, monochromator andspectrometer energies, sample rotation angle ( θ ), scatter-ing angle (2 θ ), photon flux, sample temperature, shutterstatus etc. The third panel shows the command history.The fourth panel presents a graph of RIXS spectra, XAS, θ -2 θ angular scan etc. COMMISSIONING RESULTSBeam size
To achieve high-resolution RIXS measurements, asmall soft X-ray source is required. The knife-edge-scanmethod was used to measure the beam profile and to
FIG. 6. Measurements of soft X-ray beam profile. (a) & (b)Measurements of beam profile at the entrance slits for verti-cal and horizontal directions, respectively. (c) The horizontalbeam profile of soft X-rays at the sample position. The beamprofile was obtained on detecting the beam intensity with aphotodiode after a blade moving along the transverse direc-tion. The beam intensity from the photodiode as a functionof the blade position is plotted with black circles along witha green curve which shows its Fourier filtered function; thebeam profile after the differentiation of the photodiode signalis plotted with a blue curve for its Fourier filtered function.The measured beam sizes of (a), (b) and (c) are 1.97, 44.38and 3.23 µ m at FWHM, respectively. obtain the beam size at the slit positions. Figures 6(a)and 6(b) show the measured beam-intensity profiles andtheir derivatives as a function of the knife-edge positionof the vertical and horizontal entrance slits, respectively.The derivatives of the measured profile reveal that thebeam sizes at the vertical and horizontal entrance slitsare 1.97 µ m and 44.38 µ m at FWHM, respectively, insatisfactory agreement with the designed values. Figure6(c) plots the horizontal beam profile at the sample posi-tion; the focused beam size is 3.23 µ m at FWHM in thehorizontal direction. Adjustment of the grating surface profile
To operate the AGM-AGS optical system, the surfaceprofile of each of the two active gratings must be adjustedto meet its target profile for a given incident photon en-ergy. The target profile y t (x) is expressed as a polynomialfunction, y t ( x ) = (cid:88) k =0 c k x k , (1)in which x is the position along the longitudinal direc-tion of the grating and c k is the k th coefficient of thepolynomial. Our theoretical simulations indicate that athird-degree polynomial profile is sufficient to achieve anenergy-resolving power 20,000, whereas a fourth-degreepolynomial is necessary for 100,000. To facilitate theadjustment of the profile of the grating surface, an in-position LTP system was used to measure the slope ofthis profile. With the measured slope function, the tar-get slope function and the set of actuator-response func-tions as inputs, one can use an iterative algorithm todeduce a set of actuator incremental values to adjust thesurface profile to match the target profile. The detailsof the adjustment are reported elsewhere [24]. Figure7 plots typical LTP measurement results, including themeasured slope function and its target slope polynomialfunction, the difference between them and the measuredsurface profile of the grating. The slope differnce be-tween the measured and target functions was minimizedto a root-mean-square (rms) value less than 0.25 µ rad forboth AGM and AGS gratings. Because of the limited pre-cision and accuracy of the in-position LTP instrument,the obtained slope difference is larger than the slope er-ror 0.1 µ rad rms of the polished surface provided by themanufacturer. Resolution optimization
We measured the linewidth of elastic scattering froma W/B C multilayer (ML) to tune the energy resolu-tion. At the beginning, the grating surface profile wasadjusted using the 25 actuators to match the initial tar-get profile from the theoretical simulations of the de-signed beamline. Because of the alignment imperfectionsof the beamline, various sets of polynomial coefficients,i.e., [c , c , c ], were tested as the target profile to op-timize the energy resolution. Figure 8 shows the elasticscattering of 530 eV soft X-rays from the ML near a re-flection condition with entrance slit 4 µ m and exit slit100 µ m, which corresponds to bandwidth 0.5 eV of inci-dent soft X-rays. The best achieved spectral resolutionwas 12.4 meV, which is defined as the minimum separa-tion between two spectral lines when they can be resolvedthrough a criterion that the core-to-wing ratio is smaller FIG. 7. Slope measurements of an active grating with anin-position LTP. (a) Measured grating slope and its targetpolynomial slope function y t (cid:48) ( x ) = c x +2 c x +3 c x +4 c x ,in which c = 1 . × − , c = 9 . × − , c = − . × − and c = 5 . × − . The function y t (cid:48) ( x ) is vertically offsetfor clarity. (b) Difference between measured and target slopefunctions. The rms of this difference is 0 . µ rad. (c) Profilefunction of the grating surface obtained from the integrationof the measured slope, with the height at the grating centerdefined as zero. than the ratio 0.9272 of two identical Gaussian functionsseparated by their FWHM. Photon flux on the sample
As RIXS is a photon-demanding technique, a largephoton flux on the sample is essential to achieve high-resolution measurements. The AGM-AGS scheme allowsus to increase the bandwidth of the incident photons onthe sample while maintaining the energy resolution. Toverify this condition, we measured the energy resolutionof the elastic scattering of 530 eV soft X-rays with exit-
FIG. 8. Elastic scattering spectra of a W/B C multilayermeasured in a near reflection condition at 530 eV. The mea-sured scattering spectrum, which has a FWHM of 14.8 meV,is plotted with black circles along with a green curve whichshows its Fourier filtered function. The spectrum plottedalong with a blue curve for its Fourier filtered function isa duplicate of this measured spectrum. When the separationbetween these two spectra is 12.4 meV, the core-to-wing ra-tio in their summation plotted together with a red curve forits Fourier filtered function is 0.926. The measured spectrumwas obtained without applying any pixel elimination with anintensity threshold on the 2D image recorded in 1 sec. slit opening set at 50, 100 and 200 in units of µ m. Themeasured photon flux increased linearly from 3.6 × photons s − to 1.3 × photons s − with increased exit-slit opening. The measured energy resolution remainednearly unchanged while the exit-slit opening increased.These data are consistent with our previously results [21]and demonstrate that the AGM-AGS scheme is also ap-plicable to high-resolution soft X-ray RIXS. Phonon excitations of superconducting cuprates
The dressing of electrons by phonons in a mate-rial plays an important role in their novel electronicproperties; in particular, the coupling between elec-trons and phonons in HTSC is pivotal yet remains un-der debate. RIXS has the potential to measure thestrength of the coupling of electrons to lattice excitations.It has been shown theoretically that high-resolutionRIXS provides direct, element-specific and momentum-resolved information about the electron-phonon coupling[26, 27]. Here we present O K -edge RIXS measurementson La . Sr . CuO (LSCO) to reveal its momentum-resolved phonon excitations.La − x Sr x CuO is a doped Mott insulator [28] com-posed of CuO conducting layers with relatively weakcoupling between the layers. Its mother compound La CuO is an insulating antiferromagnet, in which Cuion has an electronic configuration 3 d with a hole of sym-metry x − y . The system is a charge-transfer insulatorbecause strong correlation effects split the conductionband into the upper and lower Hubbard bands. Thesecorrelation effects manifest themselves in O K -edge X-ray absorption in which an electron is excited from the1 s core level to the 2 p band. The polarization-dependentXAS measurements revealed that the doped holes aredistributed mainly throughout the O 2 p x,y orbital in theCuO plaquette and are hybridized with the Cu 3 d x − y orbital to form a spin singlet 3 d L termed a Zhang-Ricesinglet (ZRS), in which L denotes a ligand hole [29, 30].In the XAS of La CuO , there exists a prepeak at theabsorption edge arising from the O 2 p band hybridizedwith the upper Hubbard band (UHB) of Cu 3 d . In ahole-doped cuprate, a lower-energy XAS feature resultingfrom ZRS emerges and grows linearly with hole concen-tration in the under-doped regime as plotted in Fig. 9(a)[31]. That is, hole doping manifests itself in the spectral-weight transfer from UHB to ZRS as a consequence ofelectron correlations. The existence of ZRS enables mea-surements of phonon excitations using O K -edge RIXS.Figure 9(b) shows the RIXS spectrum of LSCO measuredwith incident X-rays set at 528.4 eV, i.e., the ZRS feature.The inset of Fig. 9(b) illustrates the scattering geometry.The in-plane momentum transfer q (cid:107) was along direction( π , 0) with q (cid:107) = 0.12 πa ; the out-of-plane momentumtransfer was q ⊥ = 1.0 πc , in which a and c denote the lat-tice parameters. The RIXS spectrum in Fig. 9(b) showsan intense elastic peak and pronounced phonon excita-tions. We fitted the measured RIXS spectrum with fourpeaks, i.e. an elastic-scattering peak and three phononpeaks, using Voigt functions. The observed three phononexcitations originate from the half-breathing mode (75meV), the A g apical oxygen mode (45 meV) and thevibrational mode involved with La/Sr (20 meV). Thesehigh-resolution RIXS measurements open a new oppor-tunity to investigate the coupling between phonons andcharge-density waves in HTSC. SUMMARY AND FUTURE PLAN
We have designed, constructed and commissioned ahigh-resolution and highly efficient RIXS beamline basedon the energy-compensation principle of grating disper-sion. The achieved resolving power was 42,000 at photonenergy 530 eV with the bandwidth of the incident soft X-rays set at 0.5 eV. With this high resolution, we observedthree phonon excitations of LSCO. A new in-vacuumLTP instrument with a high precision of 0.005 µ rad rmshas been developed recently. It will be installed in thebeamline to greatly improve the accuracy of the mea-sured slope function of the grating surface, aiming toreach ultra-high resolving power 100,000. We also plan to FIG. 9. (a) O K -edge XAS of LSCO at 23 K measured withthe fluorescence yield mode with X-rays of σ polarization. (b)RIXS spectrum of LSCO with incident energy tuned to ZRSwith q (cid:107) = 0.12 πa and q ⊥ = 1.0 πc . The duration of expo-sure to record the RIXS data shown in the black circles was2 hours. The red curve is the summation of the fitted compo-nents and a linear background function. Other color curvesare fitted Voigt functions for elastic scattering and phonon ex-citations. The inset illustrates the scattering geometry with q = k in − k out , in which k in & k out are incident and scatteredwave vectors, respectively. a and c are crystallographic axesof LSCO. The scattering angle 2 θ was 150 ◦ . install a polarimeter for the polarization analysis of scat-tered soft X-rays and to develop a high-spatial-resolutionand highly efficient soft X-ray 2D detector. ACKNOWLEDGEMENTS
We acknowledge the NSRRC staff for technical sup-port; particularly we thank D. J. Wang for useful discus-sions. We also acknowledge L. Y. Chang for his help dur-ing the commissioning stage. Thanks are due to The Min-istry of Science and Technology of Taiwan under grantNo. 103-2112-M-213-008-MY3 and MOST 108-2923-M-213-001 for providing partial support to this work. ∗ Corresponding author: [email protected][1] A. Kotani and S. Shin, Rev. Mod. Phys. , 203 (2001).[2] L. J. P. Ament, M. van Veenendaal, T. P. Devereaux,J. P. Hill, and J. van den Brink, Rev. Mod. Phys. ,705 (2011).[3] W. Sch¨ulke, Nucl. Instrum. Meth. A , 338 (1989).[4] C. Dallera, E. Puppin, G. Trezzi, N. Incorvaia, A. Fasana,L. Braicovich, N. B. Brookes, and J. B. Goedkoop, J.Synchrotron Radiat. , 231 (1996).[5] G. Ghiringhelli, A. Tagliaferri, L. Braicovich, and N. B.Brookes, Rev. Sci. Instrum. , 1610 (1998).[6] G. Ghiringhelli, L. Braicovich, T. Schmitt, and V. Stro-cov, Synchrotron Radiat. News , 16 (2012).[7] T. Schmitt, V. N. Strocov, K. J. Zhou, J. Schlappa,C. Monney, U. Flechsig, and L. Patthey, J. ElectronSpectros. Relat. Phenomena , 38 (2013).[8] M. E. Dinardo, A. Piazzalunga, L. Braicovich,V. Bisogni, C. Dallera, K. Giarda, M. Marcon, A. Tagli-aferri, and G. Ghiringhelli, Nucl. Instrum. Meth. A ,176 (2007).[9] G. Ghiringhelli, A. Piazzalunga, C. Dallera, G. Trezzi,L. Braicovich, T. Schmitt, V. N. Strocov, R. Betemps,L. Patthey, X. Wang, et al. , Rev. Sci. Instrum. , 113108(2006).[10] C. T. Chen, Nucl. Instrum. Meth. A , 595 (1987).[11] C. T. Chen and F. Sette, Rev. Sci. Instrum. , 1616(1989).[12] G. Ghiringhelli, N. B. Brookes, E. Annese, H. Berger,C. Dallera, M. Grioni, L. Perfetti, A. Tagliaferri, andL. Braicovich, Phys. Rev. Lett. , 117406 (2004).[13] G. Ghiringhelli, M. Matsubara, C. Dallera, F. Fra-cassi, A. Tagliaferri, N. B. Brookes, A. Kotani, andL. Braicovich, Phys. Rev. B , 035111 (2006).[14] L. Braicovich, J. Van den Brink, V. Bisogni, M. M. Sala,L. J. P. Ament, N. B. Brookes, G. M. De Luca, M. Sal-luzzo, T. Schmitt, V. N. Strocov, et al. , Phys. Rev. Lett. , 077002 (2010).[15] M. Le Tacon, G. Ghiringhelli, J. Chaloupka, M. M. Sala,V. Hinkov, M. W. Haverkort, M. Minola, M. Bakr, K. J.Zhou, S. Blanco-Canosa, et al. , Nat. Phys. , 725 (2011).[16] N. B. Brookes, F. Yakhou-Harris, K. Kummer, A. Fon-dacaro, J. C. Cezar, D. Betto, E. Velez-Fort, A. Amorese,G. Ghiringhelli, L. Braicovich, et al. , Nucl. Instrum.Meth. A , 175 (2018).[17] I. Jarrige, V. Bisogni, Y. Zhu, W. Leonhardt, and J. Dvo-rak, Synchrotron Radiat. News , 7 (2018).[18] C. Schulz, K. Lieutenant, J. Xiao, T. Hofmann, D. Wong,and K. Habicht, J. Synchrotron Radiat. , 238 (2020).[19] D. J. Huang and C. T. Chen, Synchrotron Radiat. News , 3 (2018).[20] H. S. Fung, C. T. Chen, L. J. Huang, C. H. Chang, S. C.Chung, D. J. Wang, T. C. Tseng, and K. L. Tsang, AIPConf. Proc. , 655 (2004).[21] C. H. Lai, H. S. Fung, W. B. Wu, H. Y. Huang, H. W. Fu,S. W. Lin, S. W. Huang, C. C. Chiu, D. J. Wang, L. J.Huang, et al. , J. Synchrotron Radiat. , 325 (2014).[22] C.-H. Chang, C. S. Hwang, C. H. Chang, H. H. Chen,M. H. Huang, F.-Y. Lin, J.-C. Huang, and C. M. Wu,Int. Part. Accel. Conf. , 3183 (2010).[23] T. Y. Chung, C. H. Chang, J. C. Jan, C. S. Hwang, M. S.Chiu, C. C. Kuo, C. H. Chang, F. H. Tseng, J. C. Huang, Y. C. Liu, et al. , Int. Part. Accel. Conf. , 4054 (2016).[24] K. Y. Kao, C. Y. Hua, H. S. Fung, S. W. Lin, H. Y. Chao,S. C. Yeh, S. C. Chung, D. J. Huang, and C. T. Chen,AIP Conf. Proc. , 060005 (2019).[25] S. W. Lin, D. J. Wang, H. W. Fu, H. M. Tsai, C. Y. Hua,C. Y. Kuo, M. Y. Hsu, K. Y. Kao, G. C. Yin, H. S. Fung, et al. , Rev. Sci. Instrum. , 021716 (2019).[26] L. J. P. Ament, M. van Veenendaal, and J. van denBrink, EPL , 27008 (2011).[27] T. P. Devereaux, A. M. Shvaika, K. Wu, K. Wohlfeld,C. J. Jia, Y. Wang, B. Moritz, L. Chaix, W.-S. Lee, Z.-X. Shen, et al. , Phys. Rev. X , 041019 (2016). [28] M. Imada, A. Fujimori, and Y. Tokura, Rev. Mod. Phys. , 1039 (1998).[29] F. C. Zhang and T. M. Rice, Phys. Rev. B , 3759(1988).[30] C. T. Chen, L. H. Tjeng, J. Kwo, H. L. Kao, P. Rudolf,F. Sette, and R. M. Fleming, Phys. Rev. Lett. , 2543(1992).[31] C. T. Chen, F. Sette, Y. Ma, M. S. Hybertsen, E. B.Stechel, W. M. C. Foulkes, M. Schulter, S.-W. Cheong,A. S. Cooper, L. W. Rupp Jr, et al. , Phys. Rev. Lett.66