beamlines\(\def\hfill{\hskip 5em}\def\hfil{\hskip 3em}\def\eqno#1{\hfil {#1}}\)

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775

A dedicated beamline for wide-energy-range X-ray spectroscopy at SSRF: combining soft and hard X-ray capabilities

crossmark logo

aShanghai Synchrotron Radiation Facility, Shanghai Advanced Research Institute, Shanghai 201000, People's Republic of China, and bUniversity of Chinese Academy of Sciences, Beijing 100100,People's Republic of China
*Correspondence e-mail: [email protected], [email protected], [email protected]

Edited by K. Kvashnina, ESRF – The European Synchrotron, France (Received 10 September 2025; accepted 21 December 2025; online 30 January 2026)

The advancement of renewable energy critically depends on the rational design of catalysts, which necessitates a thorough understanding of the underlying materials. At Shanghai Synchrotron Radiation Facility (SSRF), the Energy Material beamline (E-line) has been established with three independent endstations – soft X-ray, hard X-ray and a combined soft/hard station – dedicated to fundamental studies of photovoltaic and catalytic processes, energy conversion mechanisms and related phenomena. Covering a broad photon energy range from 130 eV to 10000 eV, the combined endstation is specifically designed for in situ photoemission spectroscopy, enabling layer-by-layer analysis of materials and devices. Through techniques such as wide-range hard X-ray photoemission spectroscopy (HAXPES) and X-ray absorption spectroscopy (XAS), this endstation provides comprehensive insights into the chemical and electronic properties of catalysts. This report outlines the layout of the combined soft/hard beamline and the endstation, and evaluates its performance in terms of photon flux, energy resolution and representative applications in model catalysis. In particular, the use of in situ X-ray photoemission spectroscopy/HAXPES is expected to significantly advance the fundamental understanding of functional materials, thereby accelerating the development of efficient, reliable and affordable renewable energy solutions.

1. Introduction

Over the past decades, the continuous rise in fossil fuel consumption has led to a substantial increase in carbon dioxide emissions, contributing significantly to global warming and other adverse environmental impacts. In response, there is a growing demand for clean and renewable energy alongside a greener environment, which has driven research not only into noble-metal catalysts such as Pt and Ir, despite their high cost, but also into transition-metal-based catalysts (e.g. Fe, Co, Ni), which offer a promising alternative due to their cost-effectiveness (Ezhov et al., 2024View full citation; Zeng et al., 2023View full citation; Guo et al., 2024View full citation). Furthermore, the development of high-performance and low-cost catalysts is closely related to elucidation of complex reaction mechanisms in the energy conversion and storage process (Yano & Yachandra, 2009View full citation; Cutsail & DeBeer, 2022View full citation; Rocha et al., 2012View full citation). Fortunately, X-ray spectroscopy offers a powerful approach for determining both atomic structures and electronic structures of materials. Its element-specific ability enhances the probing of individual elements in complex systems without mutual interference (Xu et al., 2025View full citation; Eggart et al., 2021View full citation; Garcia-Diez et al., 2025View full citation). By advancing fundamental insights through techniques such as X-ray absorption spectroscopy (XAS), X-ray emission spectroscopy (XES) and X-ray photoemission spectroscopy (XPS), a solid foundation will be established for developing more efficient, reliable and sustainable energy solutions.

In general, XAS/XES measures provide bulk-sensitive information, with penetration depths on the order of micrometres or more. In contrast, XPS is surface-sensitive, typically probing only the top few nanometres of the investigated material, as determined by the inelastic mean free path (IMFP) of photoelectrons (Abbate et al., 1992View full citation; Frazer et al., 2003View full citation). Importantly, it is widely recognized that most chemical reactions in energy conversion and storage processes occur at surfaces or interfaces, involving the adsorption of reactants, formation of intermediate species and desorption of products (Nørskov et al., 2011View full citation; Ruiz-Lopez et al., 2020View full citation). In practice, conventional soft X-ray XPS using Al/Mg Kα sources or synchrotron radiation below 2000 eV has a probing depth of less than 10 nm. By contrast, hard X-ray photoemission spectroscopy (HAXPES), which employs photons with energies above 2 keV, can probe buried interfaces up to several hundred nanometres deep, as the IMFP of photoelectrons increases with kinetic energy above approximately 100 eV (Risterucci et al., 2014View full citation; Woicik, 2016View full citation; Rueff et al., 2018View full citation; Kalha et al., 2021View full citation; Favaro et al., 2021View full citation). Moreover, XPS has evolved into an accessible method for studying samples in various forms under in situ or operando conditions, such as in gas environments, and under varying temperatures or external electric fields. Given these capabilities, wide-energy-range XPS is now regarded as a versatile approach for revealing the atomic structure of materials under diverse in situ conditions. Consequently, several wide-energy X-ray beamlines have been constructed and are now operational worldwide, including the EMIL beamline at BESSY II, the X SPEC beamline at the KIT light source, the SST beamline at NSLS-II and the I09 beamline at Diamond Light Source (Hendel et al., 2016View full citation; Weinhardt et al., 2021View full citation; Lee & Duncan, 2018View full citation; Weiland et al., 2018View full citation).

In this context, a wide-energy-range beamline, referred to as the Energy Material beamline (E-line), has also been proposed and constructed at the Shanghai Synchrotron Radiation Facility (SSRF), as introduced earlier (Chen et al., 2018View full citation). The E-line consists of three branches: soft X-ray, hard X-ray and a combined soft/hard X-ray branch, which successfully entered commissioning in 2024 (Mei et al., 2024View full citation). SSRF is a third-generation light source comprising a 150 MeV linear accelerator and a 3.5 GeV electron storage ring, operating in top-up mode with a beam current of 200 mA. Like other wide-energy-range beamlines, the E-line incorporates two independent canted undulators to simultaneously generate soft and hard X-rays. A distinctive feature, however, lies in the soft/hard combined branch, which is specifically optimized for wide-range XPS under in situ conditions. Accordingly, this article provides a detailed account of the performance of the combined soft/hard branch and presents initial experimental results obtained at its wide-energy-range endstation.

2. Description of the combined soft/hard beamline and its performance

To deliver soft, tender and hard X-rays to the respective branches of the E-line, two independent undulators, an elliptical polarized undulator (EPU) and an in-vacuum undulator (IVU), are installed along the same straight section with a canted angle of 2 mrad, as shown in Fig. 1[link]. The EPU covers the soft X-ray range from 130 eV to 1500 eV for the soft X-ray branch, which is equipped with resonant inelastic X-ray scattering (RIXS) and resonant elastic X-ray scattering (REXS). The IVU supplies tender and hard X-rays from 1500 eV up to 18 keV for the hard X-ray branch, which supports high-energy-resolution fluorescence detection X-ray absorption near-edge spectroscopy (HERFD-XANES) and resonant X-ray emission spectroscopy (RXES) (Chen et al., 2018View full citation; Mei et al., 2024View full citation). Key parameters of the EPU and IVU are summarized in Table 1[link]; further details can be found in previous reports (Chen et al., 2018View full citation; Mei et al., 2024View full citation) and in Fig. S1 of the supporting information. As illustrated in Fig. 1[link], the soft X-ray beam (red line) is monochromated using a water-cooled variable-line-spacing plane-grating monochromator (PGM-1). PGM-1 employs blazed gratings with central line densities of 300, 800 and 1200 lines mm−1, similar to the in-focus variable-line-spacing PGM design reported earlier (Cai et al., 2019View full citation). On the hard X-ray branch, most of the IVU beam (2500–18000 eV, blue lines) is monochromated by a double-crystal monochromator (DCM) using Si(111) and Si(311) crystals. The tender-energy portion (1500–2500 eV) is monochromated by a second plane-grating monochromator (PGM-2) with fixed groove densities of 800 and 1200 lines mm−1. Due to the extremely high heat load, PGM-2 is cooled with liquid nitro­gen. PGM-2 and the DCM are arranged in series along the hard X-ray branch. Either monochromator can be moved into or out of the beam path as needed, ensuring that only one is in the beam at a time.

Table 1
Undulator parameters

  No. of periods Period length (mm) Length in total (m) Maxium k value Peak magnetic field (T) Energy range (eV)
EPU 30 60 1.8 5.3 0.907 130–1500
IVU 65 24 1.56 1.7 0.96 1500–18000
[Figure 1]
Figure 1
Diagram of the combined soft/hard beamline at SSRF and the corresponding photographs of PGMs, where the interweave of PGM-1 and PGM-2 is highlighted.

To merge the soft and hard X-ray beams together, each beam is first deflected by its respective mirror: the soft-branch mirror S2 and the hard-branch mirror H2, as shown in Fig. 1[link]. The soft X-ray beam is then refocused both horizontally and vertically by the toroidal mirror S3 into the combined endstation. Meanwhile, the hard X-ray beam is pre-focused horizontally by the toroidal mirror H1 onto the H-slit while remaining parallel in the vertical direction. The cylindrical mirror H2 subsequently deflects the hard beam into the combined branch and focuses it vertically onto the V-slit. Finally, as indicated in Fig. 1[link], a bent toroidal mirror H3 refocuses the hard X-ray beam into the combined endstation. The soft and hard beams are thus brought to a common focal point with a crossing angle of 2°, forming a new branch designated as the combined soft/hard beamline. Although the IVU can supply photons with energies beyond 10000 eV, the combined endstation is specifically configured for XPS studies. Consequently, the photon energy (hv) range is restricted to 130–10000 eV, due to the upper kinetic energy limit of the electron spectrometer and the very low photoionization cross section at higher photon energies.

The switching of the X-ray beam between the soft/hard branch and the combined branch can be achieved by linearly translating the S1 mirror or by shifting the cylindrical mirror H2. Notably, to effectively resolve the spatial conflict between the two PGMs, PGM-2 is integrated such that it is partially enclosed by the vacuum tube of PGM-1, as shown in the inset of Fig. 1[link]. This unique design allows PGM-1, PGM-2 and the DCM to operate simultaneously without interference. With the combined branch configured as shown in Fig. 1[link], a wide-range X-ray beam is delivered to the combined endstation [Fig. 2[link](a)], and the resolution power of the combined beamline is first verified by measuring the broadening of the Ar 2p3/2 absorption spectrum at 244 eV (the L3 edge), and the rocking curve of the DCM at 5000 eV. In the hard X-ray regime, the energy resolution is primarily determined by the Darwin width of the given Bragg reflections in the DCM. According to the Bragg diffraction principle, the relative bandwidth Δλ/λ (or ΔE/E) is exactly given by Δθ × cot θ. θ is the Bragg angle at the chosen photon energy, and Δθ can be characterized by the rocking curve – a high-resolution X-ray diffraction technique that measures the variation in diffracted intensity as the crystal is gradually tilted (`rocked') relative to the incident beam, thereby providing a quantitative assessment of the resolution power. In addition, the choice of 244 eV and 5000 eV to test the resolution power is based on ray-tracing results. After subtracting the pristine Lorentzian broadening of the Ar 2p inner-shell ionization (the natural lifetime of the core-excited state) from the overall full width at half-maximum (FWHM) of the Ar peak (Carroll et al., 2001View full citation), the resolution at 244 eV is then revealed to be 1.8 × 10−4 (ΔE/E) in Fig. 2[link](b). Similarly, the rocking curve of the DCM at 5000 eV (the Bragg angle θ is 23.3°) gives the FWHM of 82.0 µrad in Fig. 2[link](c), which corresponds to the energy resolution [ΔE/E is roughly equal to FWHM × cot (θ)] of 1.9 × 10−4 at 5000 eV. Secondly, photon flux has been checked at the endstation via measurement of the photocurrent of the calibrated photodiode (AXUV-100G); this reveals that fluxes of the soft X-ray beam and hard X-ray beam are 3.3 × 1012 photons s−1 at 300 mA and 244 eV (0.1% bandwidth) and 3.0 × 1012 photons s−1 at 300 mA and 5.0 keV (0.1% bandwidth), respectively. With a photon flux of the magnitude of 1012 photons s−1, we can reliably conduct both in situ HAXPES and XAS measurements not only for light elements, but also for 3d transition metals and heavy elements in the combined endstation. In addition, the beam spot size was measured at 244 eV and 5000 eV by scanning a sharp-edged Au foil horizontally and vertically (step size 1 µm) and recording the corresponding photocurrent (Fig. S2). The spot profile was obtained by fitting the first derivative of the photocurrent curve with a Gaussian function. Consequently, beam sizes are revealed to be: 97.2 µm × 42.1 µm at 244.0 eV, 54.1 µm × 39.5 µm at 5.0 keV, as shown in Figs. 3[link](a)–3[link](d).

[Figure 2]
Figure 2
(a) Photograph of the combination of the soft X-ray (red) and hard X-ray (blue) into the combined endstation. (b) The resolution power (ΔE/E) of the combined beamline at 244 eV and (c) 5000 eV.
[Figure 3]
Figure 3
Measurement of beam spot sizes (Gaussian fitting of the first derivation) at horizontal (a) and vertical directions (b) at 244 eV, and at horizontal (d) and vertical directions (d) at 5000 eV.

3. The design of the endstation

The layout of the endstation is illustrated in Fig. 4[link](a). As noted previously, this endstation in the combined branch is designed for wide-energy in situ XPS and in situ XAS studies. Here, in situ XPS refers specifically to near-ambient-pressure XPS (NAPXPS), with the current focus on elucidating solid–gas and solid–solid interfaces. Accordingly, the endstation comprises a fast-entry load-lock (FELL), a transfer chamber (UFO), an ultrahigh-vacuum (UHV) preparation chamber, an operando preparation chamber (up to 20 bar and 800°C) and an analysis chamber, as illustrated in the top-view schematic in Fig. 4[link](b). Samples in various forms of powders, thin films or crystal pieces can be mounted on flag-type sample holders and introduced via the load-lock. They may then be transferred to the UHV preparation chamber for further processing before being moved into the analysis chamber [Fig. 4[link](c)] via the UFO transfer system. Available processing options include in situ cleaning by Ar+ sputtering, thermal annealing and molecular or thin-film deposition using an electron-beam evaporator or a Knudsen-cell evaporator. Alternatively, samples can be further treated under operando conditions in the high-pressure chamber [Fig. 4[link](d)], such as reduction or oxidation reactions in various gas atmospheres at pressures up to 20 bar and temperatures up to 800°C. The analysis chamber is equipped with a HIPP-2 electron analyzer (ScientaOmicron), a four-axis manipulator and an automated gas-mixing system [Fig. 4[link](e)] consisting of five mass-flow-controlled gas lines (O2, CO2, CO, H2 and CH4). The HIPP-2 spectrometer can precisely detect electrons with kinetic energies from 5 eV to 10000 eV, matching well the photon energy range of the combined branch. Gasses are introduced into the entire analysis chamber via backfilling at pressures up to 100 mbar. The entrance cone of the analyzer can be selected with diameters of 800 µm or 300 µm (or smaller), depending on the required pressure limit. To protect the beamline's UHV, a 100 nm-thick Si3N4 window is installed on a CF100 flange in front of the analysis chamber [highlighted in Fig. 4[link](b)]. Thanks to the three-stage differential pumping system, the vacuum around the microchannel plate (MCP) of the analyzer remains below 10−9 mbar, ensuring the stable operation of the MCP (Cai et al., 2019View full citation). For in situ heating under gas atmospheres, the laser heating system on the manipulator allows sample temperatures up to 800°C at 50 mbar. Conventional radiative heating is also available on a separate sample stage. Additionally, samples can be cooled with liquid nitro­gen. During XPS measurements, the sample is typically grounded but can also be floated or biased relative to ground, for example, to enable total-electron-yield (TEY) detection for XAS measurements.

[Figure 4]
Figure 4
(a) Scheme of the combined NAPXPS endstation, which consists of the fast-entry load-lock, the UFO chamber, preparation chambers and the analysis chamber with the laser heating and (b) the top view photograph of the NAPXPS endstation. (c) The configuration of the sample stage (laser heating from back) and electron analyzer during the ambient-pressure measurement with the inset showing the HIPP-2 electron analyzer integrated in the analysis chamber. (d) The high-pressure preparation chamber with gas up to 20 bar and the sample heated to 800°C. (e) The gas mixture system for the setup of the solid–gas environment, where five gas lines are introduced and controlled by the automatic control system.

Regarding the geometric configuration between the soft/hard X-rays and the electron analyzer, the solution is not straightforward in the present NAPXPS station. As is known, the best geometry for XPS investigations considering the dipole approximation is to keep the magic angle between the electric vector of incoming polarized X-rays and the photoelectron momentum at about 54.7° (Shard, 2020View full citation). However, when it comes to the tender and hard X-ray region, the non-dipole transitions including the dipole–quadrupole contribution, become significant and have to be taken into consideration (Tanuma et al., 2013View full citation; Guillemin et al., 2006View full citation; Isomura et al., 2014View full citation). Of course, the non-dipole effect might also be witnessed in the soft X-ray region in some special cases (Hemmers et al., 2004View full citation). Roughly, it is preferable to keep the angle between the electric field vector of incident X-rays and the electron analyzer close to 90° in the hard X-ray range. This overall approximation could be described as follows:

Mathematical equation

where σ is the total photoionization cross section, β is a dipole asymmetry parameter, and θ is the angle between the incident electric field vector direction of the linearly polarized X-ray and the photoelectron analyzer. This part is the so-called dipole approximation. δ and γ are angular parameters considering electric dipole–quadrupole and electric dipole–magnetic dipole contributions, respectively, while φ is the angle between the photon direction and the plane passing through the photoelectron direction. As seen, the magic angle (54.7°) is valid with the pre-factor of the β parameter in the cross section [Mathematical equation] as the major contribution. Taking account of all these factors, the maximum photoelectron intensity at the photon energy of 8000 eV is obtained with θ at about 90° for s, p states, and differs by 10° or 20° for d and f states, taking Pt as an example (Isomura et al., 2014View full citation). Notably, the photoelectron intensity at θ of 70° is still about 80% of the maximum intensity at 8000 eV. Meanwhile, in the soft X-ray region, utilizing a geometry larger than the magic angle is also acceptable with the loss of a certain part of the photoelectron intensity, depending on the polarization of X-rays (Shard, 2020View full citation). Consequently, based on these findings and considerations, the angle between the analyzer and hard X-ray is finally set to about 70°, and 68° for the soft X-ray since the angle between the hard X-ray and soft X-ray is about 2°. In most cases, the photon energy will not exceed 8 keV, due to the extremely low photoionization cross section at higher photon energies.

Based on the configuration of the endstation, the available photon energy range of the combined branch was determined by measuring the valence band and the Au L3 absorption edge of a gold foil, respectively, as shown in Fig. 5[link]. Fitting the valence band spectrum with a standard Fermi function [Fig. 5[link](a)] indicates that the lower limit of the photon energy reaches 128.7 eV, slightly below the nominal design value of 130 eV. The observed broadening of the Fermi edge is about 0.06 eV, resulting from the convolution of the intrinsic bandwidth of the incident X-rays, the energy resolution of the electron spectrometer and the thermal smearing of the electron distribution at room temperature. The upper energy limit was established from the Au L3 absorption edge, recorded by scanning the incident photon energy while using the photocurrent from the sample holder as the I0 reference. In typical XAS measurements, I0 is obtained from an Au mesh placed upstream of the endstation; however, this configuration could not be used for the Au L3 absorption measurement. As shown in Fig. 5[link](b), the acquired absorption spectrum agrees well with the NIST reference, clearly resolving the absorption edge, confirming that the upper photon energy extends to at least 12 keV. Although the DCM in the combined branch can deliver photons above 12 keV, the photoionization cross section drops significantly at higher energies (Weiland et al., 2016View full citation). Therefore, photons exceeding 10 keV are not currently utilized for research at this endstation.

[Figure 5]
Figure 5
(a) Valence band spectroscopy of an Au foil with the Fermi level fitted. (b) The Au L3 absorption spectrum by scanning the incident photon energy and measuring the photoelectric current of the reference sample.

4. In situ investigation at the combined soft/hard XPS and XAS spectroscopies endstation

The wide-energy-range XPS capability of the E-line was validated through HAXPES measurements. First, a silicon wafer was cleaned following a standard semiconductor cleaning procedure (Kern, 1990View full citation) and transferred into the analysis chamber for UHV measurements. The corresponding XPS spectra are shown in Fig. 6[link], acquired at selected photon energies of 500 eV, 900 eV, 1100 eV, 3000 eV and 5000 eV. At first glance, a paired component is resolved at the binding energy of 103.2 eV for Si 2p3/2 and 103.9 eV for Si 2p1/2 with the broadening peak (FWHM around 1.5 eV) (Moulder et al., 1992View full citation), which is assigned to the SiO2 layer formed on top of the silicon substrate, as the wafer was not flash-annealed in UHV and such an oxidation layer is commonly observed. While the pristine Si state is discovered at 99.2 eV (Si 2p3/2) and 99.9 eV (Si 2p1/2), another doublet appears at 99.5 eV and 100.2 eV, which becomes more distinct at 1100 eV photon energy. This behavior suggests that these paired peaks arise from interface Si atoms, whose chemical state differs slightly from that of bulk atoms (Yamashita et al., 2001View full citation). The hypothesis is further supported by the complete disappearance of these paired peaks at higher photon energies (above 3000 eV). Meanwhile, as expected, the intensity of the oxidation state gradually declines with increasing photon energy and is no longer detectable at 5000 eV, owing to the substantially increased probing depth which enhances sensitivity to bulk layers. Nevertheless, when the photon energy reaches 5000 eV, the intrinsic broadening from X-rays becomes significant (E/ΔE = 6250 at 5000 eV), which affects the overall resolution of the Si 2p spectrum, and the shoulder in the Si 2p spectroscopy becomes less visible accordingly.

[Figure 6]
Figure 6
Si 2p XPS spectra of the SiO2/Si interface at varying photon energies from 500 eV to 5000 eV.

Importantly, to verify the near-ambient condition of XPS in the combined endstation, the solid–gas environment is created by delicately leaking, for instance, H2 into the analysis chamber via the leak valve and the automatic gas mixture system, where the pressure can reach 100 mbar with high precision (±1 mbar). In practice, there are two different ways to build the solid–gas environment for in situ studies. The first one is called backfilling, namely, filling the whole analysis chamber with the reactant gas. The other way is to utilize a small-volume reactor cell in front of the electron analyzer, and the reactant gas is filled into the cell only. While backfilling is user friendly, the upper pressure limit of backfilling is much lower than for the reactant cell design; however, it is not always feasible to use the reactant cell design. To start with, a clean Ag(111) crystal was prepared via cycles of Ar+ sputtering (1000 eV) and post-annealing (700 K), and the cleanness of Ag(111) was checked by XPS with no oxygen or carbon contaminations (Hu et al., 2021View full citation). In situ exploration of the Ag 3d core level with hv = 1200 eV (flux 1.1 × 1012 photons −1 at 300 mA and 0.1% bandwidth) in an atmosphere of 10 mbar and 50 mbar H2 is presented in Fig. 7[link]. The robustness of Ag 3d XPS to the external H2 gas could be identified although the intensity of Ag 3d spectrum is reduced considerably in the gas environment (it roughly dropped by 75% at 50 mbar). Moreover, the FWHM of Ag 3d peaks is well preserved under in situ conditions, while binding energies of both Ag 3d5/2 and 3d3/2 components remain unchanged under the H2 atmosphere. In addition, the electron loss feature at the higher binding energy side of spin–orbit components, closely related to the metallic Ag state, gets weaker with increasing pressure of H2. These in situ experiments clearly demonstrate the reliability and feasibility of NAPXPS at the endstation and, in fact, up to 100 mbar could be achieved with the current configuration; even higher pressure is also possible after improvement of the Si3N4 window, since it is far away from the sample (around 1.5 m) currently. In practice, the backfilling concept is widely adopted at other NAPXPS beamlines worldwide, for instance, at the EMIL beamline in BESSY II (Follath et al., 2013View full citation), where its operation is user friendly. Nevertheless, investigation of the solid–liquid interface will also be possible in the near future with upgrades of the manipulator and chamber (Axnanda et al., 2015View full citation; Zbynek Novotny et al., 2020View full citation), for instance, utilizing a three-electrode setup or the dip–pull approach, while the current research is focused on the solid–gas interface.

[Figure 7]
Figure 7
Ambient-pressure XPS investigations of Ag 3d core level in the H2 atmosphere with hv = 1200 eV.

Beyond studies on single crystals, metal-supported metal thin films with engineered interfacial structures have also been examined for model catalysis research. As an example, the Ru/Cu(111) interface was prepared and investigated using XAS in a CO2 environment. Approximately, 1/3 monolayer (ML) Ru was deposited on the Cu(111) substrate via an e-beam evaporator [experimental details are given by Zhang et al. (2023View full citation)]. XAS spectra were collected in the TEY mode by measuring the electron current from the sample. First, the Ru L3-edge absorption spectrum was acquired from the clean Ru/Cu(111) surface. As shown in Fig. 8[link], the absorption peak appears around 2840.1 eV, in agreement with earlier reports (Pedersen et al., 2018View full citation). Subsequently, 1 mbar CO2 was introduced to the analysis chamber while recording XAS in situ. Notably, the TEY signal remained stable under a gas atmosphere with a reasonable signal-to-noise ratio (top panel of Fig. 8[link]). Apparently, a new feature emerges as a shoulder at approximately 2842.5 eV on the higher-energy side of the main peak (orange curve in Fig. 8[link]). This additional shoulder can be attributed to oxidized Ru species (Van Kuiken et al., 2013View full citation), indicating that ruthenium undergoes oxidation upon exposure to CO2, and the oxidation state likely results from the activation and reduction of CO2 at the Ru/Cu(111) interface. The activation of CO2 on the Ru/Cu(111) surface at room temperature aligns with previous observations (Zhang et al., 2021View full citation), in which copper-supported metal atoms exhibit pronounced activity towards CO2 reduction. Most importantly, these experimental observations clearly confirm the reliability of TEY-based XAS measurements performed under a gas atmosphere in the combined branch of the E-line.

[Figure 8]
Figure 8
Investigation of the Ru L3-edge absorption via the TEY mode at UHV and 1 mbar CO2.

5. Conclusions

A wide-energy-range beamline (E-line) comprising soft, hard and combined branches has been constructed at SSRF and user operation began in 2024. By merging soft and hard X-rays into a dedicated combined branch, the beamline offers a specialized platform for wide-energy X-ray spectroscopy, enabling layer-resolved investigation of advanced materials under in situ conditions. As demonstrated, the combined endstation can deliver photons from 130 eV to beyond 10 keV via independent insertion devices (IVU and EPU), thereby allowing a comprehensive characterization of model catalysts using near-ambient-pressure XPS (NAPXPS), hard X-ray photoelectron spectroscopy (HAXPES) and X-ray absorption spectroscopy (XAS). This report details the performance of the combined branch and endstation in terms of photon flux, accessible energy range, beam size and representative applications in model catalysis. Equipped with multiple in situ spectroscopic techniques, including NAPXPS, HAXPES and TEY–XAS, the combined endstation is poised to substantially advance the fundamental understanding of advanced materials and accelerate progress in renewable energy technologies.

Supporting information


Footnotes

These authors contributed equally to this work.

Acknowledgements

Fruitful discussions with Professor Dr Zheng Jiang, Professor Dr Yifan Ye, Dr Hui Zhang and Dr Yong Han are greatly appreciated. We would also like to thank Professor Dr Zhi Liu, Dr Michael Hävecker, Professor Dr T.-L. Lee and Professor Dr Yitao Cui for their assistance during the construction of the combined soft/hard beamline at SSRF.

Conflict of interest

There are no conflicts of interest.

Data availability

The data supporting the results reported in the article can be accessed from within the article and upon request.

Funding information

This work is financially supported by the Strategic Priority Research Program of the Chinese Academy of Science (grant No. XDB1180000), and the National Key Research and Development Program of China (2021YFA1600800).

References

Return to citationAbbate, M., Goedkoop, J. B., de Groot, F. M. F., Grioni, M., Fuggle, J. C., Hofmann, S., Petersen, H. & Sacchi, M. (1992). Surf. Interface Anal. 18, 65–69.  CrossRef CAS Web of Science Google Scholar
Return to citationAxnanda, S., Crumlin, E. J., Mao, B., Rani, S., Chang, R., Karlsson, P. G., Edwards, M. O. M., Lundqvist, M., Moberg, R., Ross, P., Hussain, Z. & Liu, Z. (2015). Sci. Rep. 5, 9788.  Web of Science CrossRef PubMed Google Scholar
Return to citationCai, J., Dong, Q., Han, Y., Mao, B., Zhang, H., Karlsson, P. G., Åhlund, J., Tai, R., Yu, Y. & Liu, Z. (2019). Nucl. Sci. Tech. 30, 81.  Web of Science CrossRef Google Scholar
Return to citationCarroll, T. X., Bozek, J. D., Kukk, E., Myrseth, V., Saethre, L. J. & Thomas, T. D. (2001). J. Electron Spectrosc. Relat. Phenom. 120, 67–76.  CrossRef CAS Google Scholar
Return to citationChen, Z. H., Sun, F. F., Zou, Y., Song, F., Zhang, S., Jiang, Z., Wang, Y. & Tai, R. Z. (2018). Nucl. Sci. Tech. 29, 26.  CrossRef Google Scholar
Return to citationCutsail, G. E. III & DeBeer, S. (2022). ACS Catal. 12, 5864–5886.  Google Scholar
Return to citationEggart, D., Zimina, A., Cavusoglu, G., Casapu, M., Doronkin, D. E., Lomachenko, K. A. & Grunwaldt, J.-D. (2021). Rev. Sci. Instrum. 92, 023106.  Web of Science CrossRef PubMed Google Scholar
Return to citationEzhov, R., Olg, M., Lyu, X., Leshchev, D., Stavitski, E., Serov, A. & Pushkar, Y. (2024). ACS Appl. Energy Mater. 7, 604–613.  CrossRef CAS Google Scholar
Return to citationFavaro, M., Clark, P. C. J., Sear, M. J., Johansson, M., Maehl, S., van de Krol, R. & Starr, D. E. (2021). Surf. Sci. 713, 121903.  Web of Science CrossRef Google Scholar
Return to citationFollath, R., Hävecker, M., Reichardt, G., Lips, K., Bahrdt, J., Schäfers, F. & Schmid, P. (2013). J. Phys. Conf. Ser. 425, 2120003.  Google Scholar
Return to citationFrazer, B. H., Gilbert, B., Sonderegger, B. R. & De Stasio, G. (2003). Surf. Sci. 537, 161–167.  Web of Science CrossRef CAS Google Scholar
Return to citationGarcia-Diez, R., Frisch, J., van der Merwe, M., Wibowo, R. E., Gorgoi, M., Kataev, E., Jimenez, C. E., Arce, M. D., Smith, W., Quevedo-Garzon, W., Wilks, R. G., Wallacher, D., Reinschlüssel, L. J., Tok, G. C., Gasteiger, H. A. & Bär, M. (2025). J. Synchrotron Rad. 32, 634–648.  CrossRef CAS IUCr Journals Google Scholar
Return to citationGuillemin, R., Hemmers, O., Lindle, D. W. & Manson, S. T. (2006). Radiat. Phys. Chem. 75, 2258–2274.  CrossRef CAS Google Scholar
Return to citationGuo, J. Y., Haghshenas, Y., Jiao, Y., Kumar, P., Yakobson, B. I., Roy, A., Jiao, Y., Regenauer–Lieb, K., Nguyen, D. & Xia, Z. H. (2024). Adv. Mater. 36, 2407102.  CrossRef Google Scholar
Return to citationHemmers, O., Guillemin, R. & Lindle, D. W. (2004). Radiat. Phys. Chem. 70, 123–147.  Web of Science CrossRef CAS Google Scholar
Return to citationHendel, S., Schäfers, F., Hävecker, M., Reichardt, G., Scheer, M., Bahrdt, J. & Lips, K. (2016). AIP Conf. Proc. 1741, 030038.  Google Scholar
Return to citationHu, J. P., Liang, Z. F., Shen, K. C., Xie, L., Zhang, H., Huang, C. Q., Huang, Y. B., Huang, H., Tang, J. X., Jiang, Z., Yu, M. & Song, F. (2021). Nano Res. 14, 4704–4713.  CrossRef CAS Google Scholar
Return to citationIsomura, N., Kitazumi, K., Kataoka, K., Takahashi, N., Kimoto, Y. & Dohmae, K. (2014). J. Electron Spectrosc. Relat. Phenom. 195, 62–65.  Web of Science CrossRef CAS Google Scholar
Return to citationKalha, C., Fernando, N. K., Bhatt, P., Johansson, F. O. L., Lindblad, A., Rensmo, H., Medina, L. Z., Lindblad, R., Siol, S., Jeurgens, L. P. H., Cancellieri, C., Rossnagel, K., Medjanik, K., Schönhense, G., Simon, M., Gray, A. X., Nemšák, S., Lömker, P., Schlueter, C. & Regoutz, A. (2021). J. Phys. Condens. Matter 33, 233001.  Web of Science CrossRef Google Scholar
Return to citationKern, W. (1990). J. Electrochem. Soc. 137, 1887–1892.  CrossRef CAS Web of Science Google Scholar
Return to citationLee, T.-L. & Duncan, D. A. (2018). Synchrotron Radiat. News. 31(4), 16–22.  CrossRef Google Scholar
Return to citationMei, B. B., Wang, L. X., Gu, S. Q., Su, X. Z., Zhang, S., Wei, Y., Ma, J. Y., Jiang, Z. & Song, F. (2024). Nucl. Sci. Tech. 35, 156.  CrossRef Google Scholar
Return to citationMoulder, J. F., Stickle, W. F., Sobol, P. E. & Bomben, K. D. (1992). Handbook of X-ray Photoelectron Spectroscopy. Perkin-Elmer Corporation.  Google Scholar
Return to citationNørskov, J. K., Abild-Pedersen, F., Studt, F. & Bligaard, T. (2011). Proc. Natl Acad. Sci. USA 108, 937–943.  PubMed Google Scholar
Return to citationNovotny, Z., Aegerter, D., Comini, N., Tobler, B., Artiglia, L., Maier, U., Moehl, T., Fabbri, E., Huthwelker, T., Schmidt, T. J., Ammann, M., van Bokhoven, J. A., Raabe, J. & Osterwalder, J. (2020). Rev. Sci. Instrum. 91, 023103.  Web of Science CrossRef PubMed Google Scholar
Return to citationPedersen, A. F., Escudero-Escribano, M., Sebok, B., Bodin, A., Paoli, E., Frydendal, R., Friebel, D., Stephens, I. E. L., Rossmeisl, J., Chorkendorff, I. & Nilsson, A. (2018). J. Phys. Chem. B 122, 878–887.  CrossRef CAS PubMed Google Scholar
Return to citationRisterucci, P., Renault, O., Martinez, E., Detlefs, B., Delaye, V., Zegenhagen, J., Gaumer, C., Grenet, G. & Tougaard, S. (2014). Appl. Phys. Lett. 104, 051608.  Web of Science CrossRef Google Scholar
Return to citationRocha, T. C. R., Oestereich, A., Demidov, D. V., Hävecker, M., Zafeiratos, S., Weinberg, G., Bukhtiyarov, V. I., Knop-Gericke, A. & Schlögl, R. (2012). Phys. Chem. Chem. Phys. 14, 4554–4564.  CrossRef CAS PubMed Google Scholar
Return to citationRueff, J. P., Rault, J. E., Ablett, J. M., Utsumi, Y. & Céolin, D. (2018). Synchrotron Radiat. News. 31(4), 4–9.  CrossRef Google Scholar
Return to citationRuiz-Lopez, M. F., Francisco, J. S., Martins-Costa, M. T. C. & Anglada, J. M. (2020). Nat. Rev. Chem. 4, 459–475.  CAS PubMed Google Scholar
Return to citationShard, A. G. (2020). J. Vac. Sci. Technol. A 38, 041201.  CrossRef Google Scholar
Return to citationTanuma, S., Yoshikawa, H., Shinotsuka, H. & Ueda, R. (2013). J. Electron Spectrosc. Relat. Phenom. 190, 127–136.  CrossRef CAS Google Scholar
Return to citationVan Kuiken, B. E., Valiev, M., Daifuku, S. L., Bannan, C., Strader, M. L., Cho, H., Huse, N., Schoenlein, R. W., Govind, N. & Khalil, M. (2013). J. Phys. Chem. A 117, 4444–4454.  CrossRef CAS PubMed Google Scholar
Return to citationWeiland, C., Jaye, C., Quackenbush, N. F., Gann, E., Fu, Z., Kirkland, J. P., Karlin, B. A., Ravel, B., Woicik, J. C. & Fischer, D. A. (2018). Synchrotron Radiat. News. 31(4), 23–28.  CrossRef Google Scholar
Return to citationWeiland, C., Rumaiz, A. K., Pianetta, P. & Woicik, J. C. (2016). J. Vac. Sci. Technol. A 34, 030801.  CrossRef Google Scholar
Return to citationWeinhardt, L., Steininger, R., Kreikemeyer-Lorenzo, D., Mangold, S., Hauschild, D., Batchelor, D., Spangenberg, T. & Heske, C. (2021). J. Synchrotron Rad. 28, 609–617.  Web of Science CrossRef IUCr Journals Google Scholar
Return to citationWoicik, J. (2016). Hard X-ray Photoelectron Spectroscopy (HAXPES). Cham: Springer.  Google Scholar
Return to citationXu, Z., Mao, J., Mei, B., Gu, S., Ma, J., Sun, F., Li, J., Wang, L., Chen, W., Song, F. & Jiang, Z. (2025). Sci. China Chem. 68, 2044–2050.  CrossRef CAS Google Scholar
Return to citationYamashita, Y., Nagao, M., Machida, S., Hamaguchi, K., Yasui, F., Mukai, K. & Yoshinobu, J. (2001). J. Electron Spectrosc. Relat. Phenom. 114–116, 389–393.  CrossRef CAS Google Scholar
Return to citationYano, J. & Yachandra, V. K. (2009). Photosynth. Res. 102, 241–254.  Web of Science CrossRef PubMed CAS Google Scholar
Return to citationZeng, Y., Li, C., Li, B., Liang, J., Zachman, M. J., Cullen, D. A., Hermann, R. P., Alp, E. E., Lavina, B., Karakalos, S., Lucero, M., Zhang, B., Wang, M., Feng, Z., Wang, G. J., Xie, J., Myers, D. J., Dodelet, J. P. & Wu, G. (2023). Nat. Catal. 6, 1215–1227.  CrossRef CAS Google Scholar
Return to citationZhang, H., Liang, Z. F., Huang, C. Q., Xie, L., Wang, H. B., Hu, J. P., Jiang, Z. & Song, F. (2021). J. Catal. 410, 1–9.  CrossRef Google Scholar
Return to citationZhang, H., Xie, L., Liang, Z. F., Huang, C. Q., Wang, H. B., Hu, J. P., Jiang, Z. & Song, F. (2023). Nucl. Sci. Tech. 34, 108.  CrossRef Google Scholar

This is an open-access article distributed under the terms of the Creative Commons Attribution (CC-BY) Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original authors and source are cited.

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775
Follow J. Synchrotron Rad.
Sign up for e-alerts
Follow J. Synchrotron Rad. on Twitter
Follow us on facebook
Sign up for RSS feeds