# Fileset

[Tajiri_JSynchrotronRad_32_2025_125-132.pdf](https://mdr.nims.go.jp/filesets/1850b5bf-e4c2-4f1e-9733-4297abba12b2/download)

## Creator

[Hiroo Tajiri](https://orcid.org/0000-0002-8457-5631), [Shinji Kohara](https://orcid.org/0000-0001-9596-2680), [Koji Kimura](https://orcid.org/0000-0001-5485-3672), Sekhar Halubai, Haruto Morimoto, [Naohisa Happo](https://orcid.org/0000-0003-1888-6762), [Jens R. Stellhorn](https://orcid.org/0000-0002-5579-6902), [Yohei Onodera](https://orcid.org/0000-0002-3080-6991), [Xvsheng Qiao](https://orcid.org/0000-0002-6411-1274), [Daisuke Urushihara](https://orcid.org/0000-0002-2967-0337), Peidong Hu, [Toru Wakihara](https://orcid.org/0000-0002-3916-3849), [Toyohiko Kinoshita](https://orcid.org/0000-0002-8030-0796), [Koichi Hayashi](https://orcid.org/0000-0002-8782-4293)

## Rights

[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)

## Other metadata

[Diffractometer for element-specific analysis on local structures using a combination of X-ray fluorescence holography and anomalous X-ray scattering](https://mdr.nims.go.jp/datasets/61140d8f-96ba-49f9-9ed2-c3f411465a6c)

## Fulltext

ok5128 125..132electronic reprintISSN: 1600-5775journals.iucr.org/sDiffractometer for element-specific analysis on localstructures using a combination of X-ray fluorescenceholography and anomalous X-ray scatteringHiroo Tajiri, Shinji Kohara, Koji Kimura, Sekhar Halubai, HarutoMorimoto, Naohisa Happo, Jens R. Stellhorn, Yohei Onodera, XvshengQiao, Daisuke Urushihara, Peidong Hu, Toru Wakihara, ToyohikoKinoshita and Koichi HayashiJ. Synchrotron Rad. (2025). 32, 125–132IUCr JournalsCRYSTALLOGRAPHY JOURNALS ONLINEThis open-access article is distributed under the terms of the Creative Commons Attribution Licencehttps://creativecommons.org/licenses/by/4.0/legalcode, which permits unrestricted use, distribution, andreproduction in any medium, provided the original authors and source are cited.J. Synchrotron Rad. (2025). 32, 125–132 Hiroo Tajiri et al. · Diffractometer for element-specific analysis on local structureshttps://journals.iucr.org/s/https://doi.org/10.1107/S1600577524011366https://creativecommons.org/licenses/by/4.0/legalcodehttps://crossmark.crossref.org/dialog/?doi=10.1107/S1600577524011366&domain=pdf&date_stamp=2025-01-01research papersJ. Synchrotron Rad. (2025). 32, 125–132 https://doi.org/10.1107/S1600577524011366 125ISSN 1600-5775Received 30 August 2024Accepted 21 November 2024Edited by K. Kvashnina, ESRF – The EuropeanSynchrotron, FranceKeywords: X-ray diffractometers; X-rayfluorescence holography; anomalous X-rayscattering; element-specific measurements;carry-in equipment.Supporting information: this article hassupporting information at journals.iucr.org/sPublished under a CC BY 4.0 licenceDiffractometer for element-specific analysis onlocal structures using a combination of X-rayfluorescence holography and anomalous X-rayscatteringHiroo Tajiri,a* Shinji Kohara,b,a Koji Kimura,c,a,b Sekhar Halubai,c,aHaruto Morimoto,c Naohisa Happo,d,a Jens R. Stellhorn,e Yohei Onodera,b,fXvsheng Qiao,g Daisuke Urushihara,h Peidong Hu,i Toru Wakihara,iToyohiko Kinoshitaa and Koichi Hayashic,aaJapan Synchrotron Radiation Research Institute, Hyogo 679-5198, Japan, bCenter for Basic Research on Materials,National Institute for Materials Science, Ibaraki 305-0047, Japan, cDepartment of Physical Science and Engineering,Nagoya Institute of Technology, Nagoya 466-8555, Japan, dGraduate School of Information Sciences, Hiroshima CityUniversity, Hiroshima 731-3194, Japan, eCo-Creation Institute for Advanced Materials, Shimane University, Matsue 690-8504, Japan, fInstitute for Integrated Radiation and Nuclear Science, Kyoto University, Osaka 590-0494, Japan, gState KeyLaboratory of Silicon Materials & School of Materials Science and Engineering, Zhejiang University, Hangzhou 310027,China, hDivision of Advanced Ceramics, Nagoya Institute of Technology, Nagoya 466-8555, Japan, and iInstitute ofEngineering Innovation, School of Engineering, The University of Tokyo, Tokyo 113-8656, Japan. *Correspondencee-mail: tajiri@spring8.or.jpTo tackle disorder in crystals and short- and intermediate-range order inamorphous materials, such as glass, we developed a carry-in diffractometer toutilise X-ray fluorescence holography (XFH) and anomalous X-ray scattering(AXS), facilitating element-specific analyses with atomic resolution using thewavelength tunability of a synchrotron X-ray source. Our diffractometer unifiesXFH and AXS configurations to determine the crystal orientation via diffrac-tometry. In particular, XFH was realised even for a crystal with blurred emissionlines by a standing wave in a hologram, and high-throughput AXS with sufficientcount statistics and energy resolution was achieved using three multi-arraydetectors with crystal analysers. These features increase tractable targets byXFH and AXS, which have novel functionalities.1. IntroductionWavelength tunability, a key advantage of synchrotronradiation, facilitates element-specific analysis of structures inmaterials. Several analytical methods, such as X-ray absorp-tion fine structure, photo-electron spectroscopy, resonantX-ray scattering, and multi-wavelength anomalous diffractionin protein crystallography, capitalise on this advantage. Theprogress in the brilliance of synchrotron radiation (Yabashi &Tanaka, 2017) has accelerated the application of wavelengthtunability for a wide variety of characterisation techniques,even in the field of diffractometry, whereas anomalous termsin X-ray atomic form factor realise element-specific analysis.Meanwhile, the demand for the characterisation of localdisorder or order at the atomic level in materials science isincreasing, assisted by the ongoing rapid progress in nano-technology. This is because such local disorder or orderrepresented by environmental structures around a dopant in asemiconductor or both short- and intermediate-range order inglasses plays a crucial role in the characteristics of materials,even triggering the emergence of new functionalities.In fact, although the means to access such local structuresare limited, not only a well known interstitial or substitutionalelectronic reprintstructure by a dopant but also nanoclusters formed bydopants, slight distortions in relaxor ferroelectrics withnanoscopic heterogeneity (Hayashi & Korecki, 2018), dopant–vacancy pairs (Nagaoka et al., 2023), large displacement of adopant (Yamamoto et al., 2022), a two-dimensional van derWaals material (Eguchi et al., 2024), and even soft matterincluding protein crystals (Ang et al., 2023) have becometractable by X-ray fluorescence holography (XFH).Moreover, while understanding ‘order within disorder’(Salmon, 2002) beyond the nearest-neighbour distance indisordered materials (e.g. liquid and glassy materials) remainschallenging owing to the absence of translational periodicityand complexity in their structures, quantum-beam diffractioncombined with advanced topological analysis has beenaddressing this (Salmon et al., 2005; Kohara & Salmon, 2016).To access these unique structures (i.e. disorder in order andorder within disorder) and increase tractable targets withnovel functionalities, we developed a carry-in diffractometerthat utilises XFH and anomalous X-ray scattering (AXS),facilitating element-specific analyses with atomic resolution.Our diffractometer realises XFH even for a crystal withblurred emission lines by a standing wave in a hologram andhigh-throughput AXS with sufficient count statistics andenergy resolution using three multi-array detectors withcrystal analysers.2. Materials and methods2.1. OverviewThe diffractometer on the BL47XU beamline at SPring-8employed in this study is shown in Fig. 1. The light source ofBL47XU is the SPring-8 standard in-vacuum undulator. Theincident X-rays in the energy range of 6–37.7 keVare availablewith a cryogenically cooled Si 111 double-crystal mono-chromator. A pair of Rh-coated mirrors also can be used toreject higher harmonics and vertically focus the incident beam.The usual 2� axis (hereafter, � axis) in the vertical scatteringplane was designed for AXS measurements using a crystalanalyser to observe a large scattering angle over 120�. Adetector system in the horizontal scattering plane was used inthe inverse mode for XFH. The combination of an open �circle with the � axis was used to determine the crystalorientation of a sample by observing the Bragg reflectionswithout detecting emission lines by a standing wave in ahologram, which is a notable advantage of our diffractometer.In particular, the availability of XFH even to a sample withblurred emission lines is enhanced. All axes mentioned aboveare on the motorised translation stage (X and Z) with onemanual rotational motion (R), which enables the alignment ofthe instrument to the X-ray beam path within a few tens ofmicrometres. The incident X-ray optics in the beamline hutchconsisted of a precision slit, attenuator, vacuum path, and ionchamber as a beam-flux monitor. A sufficiently low back-ground noise level was achieved by installing a radiation shieldbox surrounding the incident X-ray optics, adequate leadscattering shields around the detectors, and a beam stop forthe incident beam, as shown in Fig. 2(a).2.2. X-ray fluorescence holographyXFH is a technique to observe the atomic resolutionhologram produced by interference between fluorescenceX-rays from the source atom and those scattered bysurrounding atoms. The theoretical foundations of XFH wereestablished by Szöke (1986), and the first XFH hologram wasrecorded a decade later by Tegze & Faigel (1996). In thekinematical approximation, the holographic modulation of thefluorescence yield can be approximated as (Len et al., 1994)�ðkÞ ¼ �2reReXifiðk; rÞriexp iðkri þ k � riÞ� �; ð1Þwhere re and ri (|ri| = ri) are the classical electron radius andposition of the i-th scattering atom from the emitter atom ofthe fluorescence X-rays, respectively; and k (|k| = k) and fi arethe wavevector and atomic form factor of the i-th scatterer,respectively. A three-dimensional (3D) atomic image aroundthe emitter atom is reconstructed from the obtained hologramusing the Helmholtz–Kirchhoff integral transformation(Barton, 1988).XFH has two measurements modes, namely, normal andinverse. Observed intensities using these two modes areessentially the same on the basis of the reciprocity theorem inoptics (James, 1962), that is, if a point source of radiation andresearch papers126 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures J. Synchrotron Rad. (2025). 32, 125–132Figure 1Photograph of the diffractometer for XFH and AXS on the BL47XUbeamline at SPring-8. The dimensions of the instrument are 1.8 m(length) � 1.6 m (width) � 2.2 m (maximum height), with a weightof 2.5 t.electronic reprintan observed point are interchanged, the measured intensity atthe new observation point will be the same as that at theprevious point. Multiple-energy measurements using theinverse mode by changing the incident X-ray energy have theadvantage of reducing the ghost images inherited in XFH(Gog et al., 1996). Thus, the inverse mode can produce a betterreconstructed 3D atomic image than the normal mode.Another advantage is the higher count statistics that can beachieved using the inverse mode.Therefore, in our instrument, we applied a measurementsystem in the inverse mode as the standard setup, as shown inFigs. 2(a) and 2(b). Note that the normal mode measurementis also possible using a two-dimensional detector. The XFHdetector system in the inverse mode uses three axes for asample (�, �, and �) and the � axis for the angle control of thedetecting system, composed of an analyser and detector in thehorizontal plane. In the inverse mode, we measured thefluorescence yield by changing the angle of the incident X-raysusing the � and � axes. A brushless motor was applied for the� axis to achieve continuous rotation in the same rotationaldirection, which is preferable to effectively measure a holo-gram and achieve angular precision by XFH. Three transla-tional motions were used to adjust the irradiated area of thesample. For the analyser crystal, we apply a graphite cylind-rical analyser (Hayashi et al., 2001), toroidally bent analyser(Sekioka et al., 2005), and C-shaped analyser (Happo et al.,2018, 2024) for the effective collection of a large solid angle offluorescence X-rays. Typically, an avalanche photodiode or asolid-state detector is used to collect fluorescence X-raysignals.2.3. Anomalous X-ray scatteringAXS is a method for obtaining element-specific informationfrom X-ray total scatterings (XTSs) using the anomalous termin the complex atomic form factor, f, of each element,f ðQ;EÞ ¼ f 0ðQÞ þ f 0ðEÞ þ if 00ðEÞ; ð2ÞwhereQ is defined by the wavelength of the incident X-rays, �,and scattering angle, 2�, as Q = (4 /�)sin�; f 0 and f 00 are thereal and imaginary parts of the energy (E)-dependent term,respectively.In XTS, the pair distribution function, g(r), is expressed bythe total structure factor, S(Q) (Waseda, 2002; Egami &Billinge, 2003):gðrÞ ¼ 1þ 12 2rZ QmaxQminQ½SðQÞ � 1� sinðQrÞ dQ; ð3Þwhere r, ,Qmin, andQmax represent the distance in real space,atomic number density, and minimum and maximum observedQ, respectively. Qmax determines the real space resolution ofthe analysis. S(Q) is experimentally extracted from the XTSintensities, I(Q), by normalising with f 2 of the constitutingelements, while S(Q) consists of the Faber–Ziman partialstructure factors, Sij(Q), of the i-th and j-th elements (Faber &Ziman, 1965).The intensity difference, �I, is derived from normalisedscattering intensities measured with two different X-rayenergies, Enear and Efar:�I ¼ AðQ;EfarÞ � AðQ;EnearÞ� �þ BðQ;EfarÞ � BðQ;EnearÞ� ��SðQÞ; ð4Þwhere A = h f 2i � h fi2 and B = h fi2 are described usingchemical averages of the atomic form factors, hfi, and squaredaverages, h f 2i. Enear and Efar are close to the absorption edgeof a specific element, which are typically at an energy below30 eV and 300 eV from the edge, respectively. As only f 0changes drastically, while f 00 remains almost constant in thisenergy range, this condition yields a simple interpretation ofthe obtained AXS spectrum. The differential structure factor,�S(Q), is given as a linear combination of Sij(Q), i.e.�S(Q) =PiPj wijðQ;Efar;EnearÞSijðQÞ, with weighting factors, wij,according to fi and fj (Waseda, 2002; Egami & Billinge, 2003).We can define �g(r) by substituting S(Q) into �S(Q) inequation (3). �S(Q) strongly enhances the related partialstructure factors of the i-th element and suppresses those ofthe others (Kohara et al., 2013), facilitating the element-specific analysis.research papersJ. Synchrotron Rad. (2025). 32, 125–132 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures 127Figure 2(a) Experimental setup for XFH and AXS in the horizontal and verticalscattering planes, respectively. Measurement geometries for (b) XFH and(c) AXS. The X and Z directions are the same as those in Fig. 1.electronic reprintThree multi-array detector systems with three crystalanalysers arranged every 30� can accelerate high-throughputmeasurements, as shown in Fig. 2(a). Each AXS detectorsystem comprises slits for incoming and outgoing X-raystowards a crystal analyser, where the double-slit configurationwas applied for the outgoing beam, i.e. in front of a detector toreduce background noise [Fig. 2(c)]. The analyser crystalswere set under vacuum in cylindrical chambers with two X-raywindows made of polyimide films to avoid analyser contam-ination. The analyser crystal and detector were set to the �–2�configuration to measure symmetric reflections from theanalyser. The analyser axis was motorised with high precision(0.001� per pulse), which is an order of magnitude finer thanthose for the other axes. The angle of the 2� arm can beadjusted to detect the scattered signal with near- and far-energy configurations without readjusting the arm angle byselecting adequate slit sizes. Currently, we have two detectoroptions, namely, plastic scintillation counters and avalanchephotodiodes, depending on the energy of the incident X-rays.According to the absorption corrections in the three detectors,i.e. asymmetric transmission geometry (Egami & Billinge,2003; Rowles & Buckley, 2017), we recommend the trans-mission geometry for samples as a standard for a simplerinterpretation of absorption corrections while offering bothreflection and transmission geometries.2.4. Sample preparationA natural zeolite (Junnar, Pune District, Maharashtra,India) purchased from N’s Mineral Co. Ltd was used as thescolecite crystal in this study. For the measurements, thecrystal was cut and polished to approximately 3 mm� 8 mm�1 mm. A crystal structure analysis of the sample wasperformed using a laboratory X-ray source with an Mo targetbefore the synchrotron XFH experiments to clarify the crystalquality. The sharp X-ray diffraction (XRD) peaks revealedthat the sample is a fine single crystal of the space group Cc(monoclinic), with the structural parameters of a = 6.52520(10) Å, b = 18.9769 (3) Å, c = 9.7779(2) Å, and � = 108.8570(6)�, resulting in an R factor of 0.0240. Detailed structuralinformation of the scolecite crystal is provided in thesupporting information.The Ag2O–ZnO–B2O3 systems, which have garneredincreasing attention in various fields, including catalysis, whereAg quantum clusters are formed in borate glass (Zheng et al.,2023), were selected for AXS. The 15Ag2O–15ZnO–70B2O3glass was prepared using the melt-quenching method.Mixtures of raw materials (ZnO, Ag2O, and HBO2) accordingto the target composition were melted in a closed aluminacrucible at 1300�C for 90 min, followed by pouring into a brasstemplate for quenching.3. Results3.1. 3D atomic imaging by XFHFigure 3(a) shows the typical hologram pattern from a Gesingle crystal of the Ge K� line projected onto the [001]direction measured at 14.5 keVof the incident X-ray energy inthe inverse mode. Higher harmonics were rejected using X-raymirrors. We used a cylindrical graphite crystal analyser(Hayashi et al., 2001) to collect the fluorescence X-rays. With50 mm � 50 mm incident X-rays, it took 3 h to complete onehologram. We reconstructed the 3D atomic images using eightholograms measured at the X-ray energies of 11.5–15.0 keVatincremental steps of 0.5 keV. Each hologram was extractedfrom the measured data using symmetry operations based onthe crystal orientation, as determined from the emission linescaused by a standing wave.Figure 3(b) shows the X–Y-reconstructed image of thesection at z = 2.8 Å from the Ge atom as the source point,where the X and Y axes correspond to the h100i and h010idirections, respectively. The reconstructed image from theholograms shows the surrounding Ge atoms at atomic reso-lution, as expected, towards the central Ge atom as the sourceof fluorescence X-rays. We used the 3D-AIR-IMAGE soft-ware for the data analysis and visualisation of the atoms(Matsushita, 2015; Matsushita et al., 2018).Besides, we present the XFH results of the scolecite as anexample that has a hologram with blurred emission linescaused by a standing wave, posing difficulties in refining thecrystal orientation of the sample. Symmetry operations on theobserved hologram based on the crystal orientation areresearch papers128 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures J. Synchrotron Rad. (2025). 32, 125–132Figure 3(a) Hologram pattern of XFH from a Ge single crystal under 14.5 keVincident X-rays, with intensity variation in the range of �0.3%. Clearemission lines by a standing wave can be observed. (b) Reconstructed Geatomic image from the hologram. The open circles denote the expectedpositions of Ge atoms.electronic reprintrequired to produce a hologram covering a wide solid anglebecause observing the complete hologram over the entiresolid angle is experimentally limited, even in inverse mode.Therefore, knowing the fine crystal orientation at an angle ofless than 0.1� is essential to obtain a precise reconstructedimage. Determining the crystal orientation by measuring theBragg reflections through single-crystal XRD can address thisdifficulty. Our instrument unifying the XFH and AXSconfigurations enables us to employ this approach.Figure 4(a) shows the hologram from the scolecite crystal ofthe Ca K� line projected onto the [001] direction with 9.2 keVincident X-rays in the inverse mode. Eight holograms with abeam size of less than 50 mm in the X-ray energy range of 9.2–13.2 keV measured in steps of 0.5 keV were used for theanalysis. Figure 4(b) shows the rocking curve (RC) of the 001Bragg reflection from the scolecite. From the angular set (�, �,�, and �) obtained from the fine Bragg peak, we successfullyobtained the crystal orientation of the scolecite crystal andexecuted symmetry operations on the hologram. The angularresolution was sufficient for determining the crystal orienta-tion though the Bragg peak split in the � scan.As expected from the crystal structure of the scoleciteshown in Fig. 5(a), atomic images corresponding to thesurrounding Ca atoms are clearly observed, as indicated by thegreen open circles in the reconstructed image, as shown inFig. 5(c). These results are in good agreement with oursimulation results [Fig. 5(b)], confirming the ability of oursystem to visualise local atoms with a hologram, even withblurred emission lines by a standing wave.3.2. Element-specific analysis by AXSThe RC of the LiF analyser crystal (OKEN Co. Ltd) isshown in Fig. 6(a) for the 002 Bragg reflection with an incidentX-ray energy of 20 keV. The full width at half-maximum(FWHM), ��, of the RC was 0.0059� (ca 13 eV in �E),corresponding to less than 0.1% �E/E, which is sufficient toresolve elastic, Compton, resonant Raman scattering, andfluorescence observed in X-ray scattering (Fischer et al., 2006).The energy resolution is defined by �E/E = �� cot�B, where�B denotes the Bragg angle. The observed energy spectrum ofthe 15Ag2O–15ZnO–70B2O3 glass at an X-ray energy(25.484 keV) close to the Ag absorption edge obtained byangular scanning of the analyser is shown in Fig. 6(b). Asexpected, the Compton component increased and graduallyshifted from the elastic components in terms of energyaccording to the scattering angle. The resonant Ramancomponents are independent of the scattering angle. Theseresults highlight the suitability of the LiF crystal for reducingresearch papersJ. Synchrotron Rad. (2025). 32, 125–132 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures 129Figure 5(a) Crystal structure of scolecite drawn using VESTA (Momma & Izumi,2011). Reconstructed images of scolecite obtained by (b) simulation and(c) experiment. Ca atoms were clearly visualised from the experiment,indicated as green open circles.Figure 4(a) Hologram pattern of XFH from a scolecite single crystal, withextremely blurred emission lines. (b) RC of the 001 Bragg reflection fromthe scolecite. The inset shows � scan results for the angle refinement.electronic reprintthe unwelcome background, including Compton and resonantRaman scatterings, to an AXS spectrum, while maintaining anacceptable energy window for elastic scattering.Using the LiF analyser crystal configurations, we measuredthe XTS of silica glass with three detector arrays at an X-rayenergy of 20 keV. The transmission geometry was applied to a10 mm � 10 mm � 1 mm silica glass plate. Figure 7(a)represents the XTS from the silica sample as the standardsample with a measurement time of 1 h. The scattering dataobtained by the three detectors with the analysers depictedgood agreement. For this data connection, we used an over-lapping Q region to calibrate the data. Notably, sufficientcount statistics required for AXS were achieved because thetotal count at the first sharp diffraction peak reached over 3.5� 106 (ffiffiffinp=n is less than 0.1%). Therefore, our instrumentachieved high-throughput measurements three times fasterthan conventional instruments with a single detector.The systematic error owing to the acceptance angleinstability of the analyser crystal in a wide-range scan,resulting from the instrument precision and misalignment, wasreduced considerably by limiting the scan range from 0� to 30�in each system. S(Q) was derived using corrections for self-absorption, normalisation by the flux of incident X-rays, andcombining the datasets from the three detector systems. Asshown in Fig. 7(b), the measured S(Q) using our instrumentand the reference S(Q) observed by high-energy (HE) XTS(Ohara et al., 2021) are in excellent agreement with sufficientstatistics, even at Q over 10 Å�1.Finally, we present the typical AXS data obtained for the15Ag2O–15ZnO–70B2O3 glass. Figure 8(a) presents �S(Q),the differential structure factor between the two S(Q)measured with X-rays near to (�30 eV) and far from(�300 eV) Ag K absorption edge (25.514 keV), together withS(Q). �S(Q) possesses Ag-specific structural informationowing to the difference in the anomalous term (f 0) of the Agform factor. The total g(r) and �g(r) were derived from theFourier transform of S(Q) and �S(Q), respectively, with Qmax= 17.3 Å�1, as shown in Fig. 8(b). Bond lengths of the B2O3–Ag2O glass extracted by neutron scattering (Ushida et al.,2001) are also indicated by dashed lines, together with those ofthe MoO3–ZnO–B2O3 glass confirmed by neutron scatteringand reverse Monte Carlo modelling (Fabian et al., 2016).Evidently, �g(r) distinctly enhanced the peak correspondingto the Ag—O correlation by the element-specific feature,whereas total g(r) has a less visible peak of Ag—O togetherwith those of B—O and Zn—O bond lengths.4. DiscussionIn Section 3, we presented typical XFH and AXS data usingour instrument. The unification of the XFH and AXSconfigurations into one instrument was beneficial becauseboth methods offered element-specific analyses using wave-length tunability of synchrotron X-rays and required photon-hungry experiments that need count statistics finer than 0.1%,showing similarities. XFH could obtain deeper insights intothe structures of local disorders at the atomic level in a crystal,e.g. defects, substitutions, and distortions. Meanwhile, AXSprovided information on ordering beyond the nearest-neigh-bour distance in disordered materials, for example, both short-and intermediate-range order in glass, representative ofdisordered materials. In other words, our diffractometer offersdual-experimental approaches to reveal local structures inresearch papers130 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures J. Synchrotron Rad. (2025). 32, 125–132Figure 7(a) XTS of silica glass using the three-detector array system with 20 keVX-rays in 1 h (accumulation time of 20 s in each point). Instrumentbackgrounds (BGs) are plotted together. (b) Total structure factor, S(Q),of silica glass derived from the data in (a). For comparison, HEXTS dataare displayed upward by 1 for clarity.Figure 6(a) RC of the 002 reflection from an LiF analyser crystal with 20 keVX-rays, of which the angular width corresponds to the energy resolution(�E) of approximately 13 eV. (b) Energy spectrum using the LiFanalyserin a coarse resolution mode with different scattering angles (40�–70�).Spectra for higher angles are displayed upward for clarity. Elastic andCompton components are distinctly resolved in addition to resonantRaman scattering.electronic reprintboth material states (i.e. crystal and amorphous phases), whichplay pivotal roles in the emergence of functionalities. This willlead to unified understandings of the functionalities acrossthese contrasting states from the structural viewpoints.In Section 3.1, we demonstrated that our system can imagelocal atoms by XFH using scolecite as an example, which hadblurred emission lines by a standing wave in a hologram.Owing to another degree of freedom (an open � circle) andthe � axis used in AXS, our instrument can act as the four-axisdiffractometer that enables determining a crystal orientationrequired for symmetry operations on the observed hologram.This is the great advantage for XFH measurements using ourinstrument, which unifies XFH and AXS configurations,increasing tractable targets.In Section 3.2, we discussed the results of using an LiFcrystal with 002 reflection as the standard analyser of ca 0.1%bandwidth for accurate subtraction of inelastic and re-emis-sion components to extract the elastic scattering in AXS. Insuch a narrow-bandwidth analyser, fine angular alignment ofthe crystal to the scattering plane was required to obtainreliable data. As three detectors with analysers were arrangedevery 30� in our system, each scan range was limited to 0–30�narrower than in a conventional one-detector system, redu-cing the systematic errors caused by the acceptance angles ofthe analyser crystals. In addition to the high throughput, this isanother advantage of the multi-array detector systems. Ananalyser crystal adequate for the experimental demand can beselected. Graphite, which exhibits wider FWHMs in RC thanLiF, is a possible solution for a dilute system that requireshigher flux for count statistics.The large working distance design results in the availabilityof a microbeam option using refractive lenses (typically in theX-ray energy range of 6–25 keV) for microscopic measure-ments. The goniometer for a sample that includes three axes isalso interchangeable with, for example, a sample preparationinstrument for in situ observation.5. ConclusionsWe developed a carry-in diffractometer in SPring-8 thatutilises both XFH and AXS, realising element-specificanalyses with atomic resolution using the wavelengthtunability of the synchrotron X-ray source. The combinationof XFH and AXS configurations facilitates the determinationof crystal orientation via diffractometry. This feature enablesthe application of XFH even for crystals with blurred emissionlines caused by a standing wave in a hologram. Moreover, thethree multi-array detector systems with three crystal analysers,which offer sufficient energy resolutions to resolve elastic,Compton, resonant Raman scattering, and fluorescence,realise high-throughput measurements with sufficient countstatistics required for AXS. These features enable us to tackledisorder in crystals and short- and intermediate-range order inamorphous materials such as glass, and increase tractabletargets by XFH and AXS, which have novel functionalities.6. Related literatureThe following references, not cited in the main body of thepaper, have been cited in the supporting information: Fälth &Hansen (1979); Palatinus & Chapuis (2007); Petřı́ček et al.(2014).Funding informationThe synchrotron experiments were performed with theapproval of JASRI (proposal Nos. 2018A1054, 2018B1068,2020A1458, 2021B1231, 2022A1345, 2022A1359, 2022A2002,2022B1017, 2022B1397, 2022B1400, 2022B2018, 2023A1346,2023A1353, 2023A2302, 2023B2309, 2024A2323). This workwas supported in part by the Japan Society for the Promotionof Science Grant-in-Aid for Transformative Research Areas(A) ‘Hyper-Ordered Structures Science’ (grant Nos.20H05878, 20H05880, 20H05881).research papersJ. Synchrotron Rad. (2025). 32, 125–132 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures 131Figure 8(a) Total structure factor, S(Q), and differential structure factor, �S(Q),obtained from XTS and AXS, respectively, of the 15Ag2O–15ZnO–70B2O3 glass. (b) Total g(r) and �g(r) derived from the data in (a). Thetotal g(r) shows peaks corresponding to B—O, Zn—O, Ag—O, andcation—cation correlations, whereas �g(r) highly enhanced the Ag—Obond length, depicting element-specific features. Dashed lines areprovided as guides.electronic reprintReferencesAng, A. K. R., Umena, Y., Sato-Tomita, A., Shibayama, N., Happo,N., Marumi, R., Yamamoto, Y., Kimura, K., Kawamura, N., Takano,Y., Matsushita, T., Sasaki, Y. C., Shen, J.-R. & Hayashi, K. (2023). J.Synchrotron Rad. 30, 368–378.Barton, J. J. (1988). Phys. Rev. Lett. 61, 1356–1359.Egami, T. & Billinge, S. (2003). Underneath the Bragg Peaks: Struc-tural Analysis of Complex Materials. Oxford: Elsevier.Eguchi, R., Ikeda, M., Yamamoto, Y., Goto, H., Happo, N., Kimura,K., Hayashi, K. & Kubozono, Y. (2024). Inorg. Chem. 63, 947–953.Faber, T. & Ziman, J. (1965). Philos. Mag. 11, 153–173.Fabian, M., Svab, E. & Krezhov, K. (2016). J. Non-Cryst. Solids, 433,6–13.Fälth, L. & Hansen, S. (1979). Acta Cryst. B35, 1877–1880.Fischer, H. E., Barnes, A. C. & Salmon, P. S. (2006). Rep. Prog. Phys.69, 233–299.Gog, T., Len, P., Materlik, G., Bahr, D., Fadley, C. & Sanchez-Hanke,C. (1996). Phys. Rev. Lett. 76, 3132–3135.Happo, N., Hada, T., Kubota, A., Ebisu, Y., Hosokawa, S., Kimura, K.,Tajiri, H., Matsushita, T. & Hayashi, K. (2018). Jpn. J. Appl. Phys.57, 058006.Happo, N., Kubota, A., Yang, X., Eguchi, R., Goto, H., Ikeda, M.,Kimura, K., Takabayashi, Y., Stellhorn, J. R., Hayakawa, S.,Hayashi, K. & Kubozono, Y. (2024). Chem. Mater. 36, 4135–4143.Hayashi, K. & Korecki, P. (2018). J. Phys. Soc. Jpn, 87, 061003.Hayashi, K., Miyake, M., Tobioka, T., Awakura, Y., Suzuki, M. &Hayakawa, S. (2001). Nucl. Instrum. Methods Phys. Res. A, 467–468, 1241–1244.James, R. W. (1962). The Optical Principles of the Diffraction ofX-rays. Woodbridge: Ox Bow Press.Kohara, S. & Salmon, P. S. (2016). Adv. Phys. 1, 640–660.Kohara, S., Tajiri, H., Song, C., Ohara, K., Temleitner, L., Sugimito,K., Fujiwara, A., Pusztai, L., Usuki, T., Hosokawa, S., Benino, Y.,Kitamura, N. & Fukumi, K. (2013). J. Phys. Conf. Ser. 502, 012014.Len, P. M., Thevuthasan, S., Fadley, C. S., Kaduwela, A. P. & VanHove, M. A. (1994). Phys. Rev. B, 50, 11275–11278.Matsushita, T. (2015). 3D-AIR-IMAGE, https://sites.google.com/hyperordered.org/3d-air-image.Matsushita, T., Muro, T., Matsui, F., Happo, N., Hosokawa, S.,Ohoyama, K., Sato-Tomita, A., Sasaki, Y. C. & Hayashi, K. (2018).J. Phys. Soc. Jpn, 87, 061002.Momma, K. & Izumi, F. (2011). J. Appl. Cryst. 44, 1272–1276.Nagaoka, A., Kimura, K., Ang, A. K. R., Takabayashi, Y., Yoshino,K., Sun, Q., Dou, B., Wei, S.-H., Hayashi, K. & Nishioka, K. (2023).J. Am. Chem. Soc. 145, 9191–9197.Ohara, K., Onodera, Y., Murakami, M. & Kohara, S. (2021). J. Phys.Condens. Matter, 33, 383001.Palatinus, L. & Chapuis, G. (2007). J. Appl. Cryst. 40, 786–790.Petřı́ček, V., Dušek, M. & Palatinus, L. (2014). Z. Kristallogr. 229,345–352.Rowles, M. R. & Buckley, C. E. (2017). J. Appl. Cryst. 50, 240–251.Salmon, P. (2002). Nat. Mater. 1, 87–88.Salmon, P., Martin, R., Mason, P. E. & Cuello, G. J. (2005). Nature,435, 75–78.Sekioka, T., Hayashi, K., Matsubara, E., Takahashi, Y., Hayashi, T.,Terasawa, M., Mitamura, T., Iwase, A. & Michikami, O. (2005). J.Synchrotron Rad. 12, 530–533.Szöke, A. (1986). AIP Conf. Proc. 147, 361–367.Tegze, M. & Faigel, G. (1996). Nature, 380, 49–51.Ushida, H., Iwadate, Y., Hattori, T., Nishiyama, S., Fukushima, K.,Misawa, M. & Fukunaga, T. (2001). J. Alloys Compd. 327, 121–126.Waseda, Y. (2002). Anomalous X-ray Scattering for Materials Char-acterization: Atomic-Scale Structure Determination. Berlin:Springer.Yabashi, M. & Tanaka, H. (2017). Nat. Photon. 11, 12–14.Yamamoto, Y., Kawamura, K., Sugimoto, H., Gadelmawla, A.,Kimura, K., Happo, N., Tajiri, H., Webber, K. G., Kakimoto, K. &Hayashi, K. (2022). Appl. Phys. Lett. 120, 052905.Zheng, W., Li, R., Wang, C., Qiao, X., Qian, G. & Fan, X. (2023). J.Non-Cryst. Solids, 599, 121910.research papers132 Hiroo Tajiri et al. � Diffractometer for element-specific analysis on local structures J. Synchrotron Rad. (2025). 32, 125–132electronic reprint