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

Journal logoJOURNAL OF
SYNCHROTRON
RADIATION
ISSN: 1600-5775

Simultaneous measurements of solid–liquid interfaces and bulk liquids using soft X-ray absorption spectroscopy

crossmark logo

aPhoton Factory, Institute of Materials Structure Science, High Energy Accelerator Research Organization, 1-1 Oho, Tsukuba, Ibaraki 305-0801, Japan, bDepartment of Chemistry, Keio University, 3-14-1 Hiyoshi, Yokohama 223-8522, Japan, cInstitute for Molecular Science, Myodaiji, Okazaki 444-8585, Japan, and dGraduate Institute for Advanced Studies, SOKENDAI, Myodaiji, Okazaki 444-8585, Japan
*Correspondence e-mail: [email protected]

Edited by S. Bone, Forschungszentrum Jülich, Germany (Received 8 March 2026; accepted 5 May 2026; online 1 June 2026)

Simultaneous measurement methods for soft X-ray absorption spectroscopy (XAS) of solid–liquid interfaces and bulk liquids have been developed. The O K-edge XAS spectra of H2O/Au interfaces and bulk H2O were measured using a transmission-type liquid cell, in which the liquid layer was sandwiched between Au/Cr/Si3N4 and Si3N4 membranes. The XAS spectrum of bulk H2O was obtained using the transmission method by controlling the thickness of the liquid layer, whereas the XAS spectrum of the H2O/Au interface was obtained using the electron-yield method by measuring the drain currents from the Au surface after soft X-ray absorption. The XAS spectra of bulk H2O and the H2O/Au interfaces were differentiated to determine the appropriate measurement conditions for the solid–liquid interfaces interfaces.

1. Introduction

Water molecules at solid–liquid interfaces play important roles in various catalytic, electrochemical and biological reactions (Björneholm et al., 2016View full citation; Shimizu et al., 2018View full citation; Hu et al., 2024View full citation; Qiu et al., 2025View full citation). Recently, several measurement methods have been developed for observing solid–liquid interfaces. The interface between liquid water and a silicon thin film has been measured using X-ray photoelectron spectroscopy (XPS) (Masuda et al., 2013View full citation). The structure of water molecules on electrode surfaces has been investigated using ambient-pressure XPS (Favaro et al., 2016View full citation) and infrared spectroscopy (Ashley & Pons, 1988View full citation; Nakamura et al., 2008View full citation). The adsorption structures of water molecules at solid–liquid interfaces have been studied using sum-frequency generation spectroscopy (Nihonyanagi et al., 2004View full citation; Tabassum et al., 2025View full citation), Raman spectroscopy (Pettinger et al., 1981View full citation; Wang et al., 2021View full citation) and X-ray scattering (Toney et al., 1994View full citation). The alignment of water molecules on biological membranes has been investigated using surface-specific vibrational spectroscopy (Dreier et al., 2018View full citation). The electronic structures of solid–liquid interfaces during various electrocatalytic reactions have been extensively studied using X-ray absorption spectroscopy (XAS) in both the hard and soft X-ray regions (Timoshenko & Cuenya, 2021View full citation).

Soft X-rays below 2 keV include the K-edges of C, N, O and F, and soft X-ray XAS is an element-specific method for investigating the electronic structures of liquid water and organic molecules (Stöhr, 1992View full citation). However, XAS measurements of liquids are difficult to obtain in transmission mode owing to the thickness of the liquid layer, which should be below 1 µm owing to the strong absorption of soft X-rays by air and liquids (Chantler, 2000View full citation). Recently, however, the development of several detection methods has enabled XAS measurements of liquids and solutions (Smith & Saykally, 2017View full citation). Because soft X-ray transmission to liquid samples is challenging, XAS measurements have been performed to detect fluorescent X-rays (Myneni et al., 2002View full citation; Odelius et al., 2006View full citation; Tokushima et al., 2008View full citation) or Auger electrons (Wilson et al., 2001View full citation; Smith et al., 2004View full citation), which are formed as secondary processes after soft X-ray absorption. XAS spectra of liquids have been measured via the fluorescence-yield method using a liquid cell consisting of a Si3N4 membrane, where soft X-rays irradiate the liquids through the membrane and fluorescence X-rays are emitted from the liquids through the same membrane. However, the XAS spectra in the fluorescence yield deviate from the true absorption spectra because fluorescent X-rays are absorbed by liquid samples, a phenomenon known as the self-absorption effect. The XAS spectra of liquids have been measured in electron-yield mode by detecting the Auger electrons emitted by the soft X-ray absorption of liquid microjets under vacuum. Pressure and temperature should be compensated for in microjets under vacuum because these phases differ from those of liquid under atmospheric conditions. The XAS spectra of liquids have been measured by detecting ion currents and exhibited bulk-sensitive electronic structures (Schön, Golnak et al., 2017View full citation; Schön, Xiao et al., 2017View full citation; Ren et al., 2019View full citation). Further evaluation of the XAS spectra in ion-yield mode is necessary because ions in solution dissipate before they reach the detector.

The transmission method is effective for obtaining the true absorption spectra of bulk liquids. Recently, XAS spectra of thin layers of bulk liquids with a thickness of less than several micrometres in flat microjets were measured in transmission mode (Ekimova et al., 2015View full citation; Fondell et al., 2017View full citation; Koralek et al., 2018View full citation; Kubin et al., 2018View full citation). Although flat microjets show no radiation damage to liquid samples owing to the windowless system, they cannot be applied to liquid samples with high vapor pressures. A transmission-type liquid cell consisting of two Si3N4 or SiC membranes can be applied to all liquid samples (Yang & Kirz, 1987View full citation; Schreck et al., 2011View full citation; Meibohm et al., 2014View full citation; Sellberg et al., 2014View full citation). XAS spectra of solutions in a wide concentration range were measured by precisely controlling the thickness of the liquid layers (Nagasaka et al., 2020View full citation; Nagasaka & Kosugi, 2021View full citation). Liquid cells can precisely control chemical environments, such as the liquid temperature (Nagasaka et al., 2017View full citation) and electrode potential, using a gold-deposited Si3N4 membrane as a working electrode (Nagasaka et al., 2013View full citation; Nagasaka, Yuzawa et al., 2014View full citation).

XAS spectra acquired in transmission mode are mainly derived from bulk liquids, and the contributions of the solid–liquid interfaces are smaller than those of bulk liquids. The XAS spectra of solid–liquid interfaces have been frequently obtained by detecting Auger electrons, which mostly originate from solid–liquid interfaces owing to the low escape depth of electrons (Velasco-Velez et al., 2014View full citation; Wu et al., 2018View full citation; Kao et al., 2020View full citation; van Spronsen et al., 2021View full citation; Wang et al., 2025View full citation). The XAS spectra of solid–liquid interfaces with induced electrode potentials have been measured using the electron-yield method (Velasco-Velez et al., 2014View full citation; Wu et al., 2018View full citation; Kao et al., 2020View full citation; van Spronsen et al., 2021View full citation). The hydrogen bonds in bulk H2O have been extensively investigated using O K-edge XAS (Smith et al., 2004View full citation; Wernet et al., 2004View full citation; Fransson et al., 2016View full citation; Nagasaka et al., 2017View full citation; Frati et al., 2020View full citation). The pre-edge peak in bulk H2O reflects the electronic structures of the acceptor sites in H2O because it corresponds to the transition of O 1s electrons to 4a1 unoccupied orbitals that are distributed around oxygen atoms. Valesco-Velez et al. measured the O K-edge XAS spectra of bulk H2O in fluorescence-yield mode, and those of the H2O/Au interfaces in electron-yield mode (Velasco-Velez et al., 2014View full citation). The pre-edge peak intensity of the H2O/Au interface is smaller than that of bulk H2O and varies with electrode potentials. By contrast, the pre-edge peak intensities of isolated H2O molecules in aqueous aceto­nitrile solutions are stronger than those of bulk H2O (Nagasaka, 2024View full citation). Wang et al. measured the O K-edge XAS spectra of bulk H2O using the ion yield and those of H2O/Au interfaces using the electron yield, finding that the pre-edge peaks of the H2O/Au interfaces are also weaker than those of bulk H2O (Wang et al., 2025View full citation).

As described above, the XAS spectra of the solid–liquid interfaces were measured using the electron-yield method, and those of bulk liquids were measured using fluorescence or ion yield methods. As shown in Fig. 1[link], this study developed simultaneous XAS measurements of bulk H2O and H2O/Au interfaces by applying the electron-yield method using a transmission-type liquid cell, including an Au/Cr/Si3N4 membrane. Because the XAS spectra in transmission mode reflect the true absorption spectra of bulk H2O, the differences between bulk H2O and H2O/Au interfaces can be accurately determined. The electron-yield XAS spectra were obtained by measuring drain currents from the Au/Cr/Si3N4 membrane to compensate electrons for H2O+ cations, which are formed by emitting Auger electrons after soft X-ray absorption. Because the effective attenuation length of electrons in liquid water is 6 nm at the O K-edge (Suzuki et al., 2014View full citation), most of the electron-yield XAS spectrum is derived from the H2O/Au interfaces. Although H2O+ cations are also formed in bulk H2O by emitting Auger electrons after soft X-ray absorption, there is few amount of electrons from the Au/Cr/Si3N4 membrane due to long distances. Note that a small amount of H2O+ cations in bulk H2O reaches the Au/Cr/Si3N4 membrane owing to the diffusion or exchange of H2O+ cations because the XAS spectra of liquid H2O were obtained by the ion-yield measurements (Wang et al., 2025View full citation). Thus, a small drain current is formed for the H2O+ cations come from bulk H2O. This study explores the appropriate measurement conditions for H2O/Au interfaces using electron-yield XAS measurements.

[Figure 1]
Figure 1
Schematic of simultaneous XAS measurements for H2O/Au interfaces and bulk H2O. The H2O/Au interfaces were measured using the electron-yield method, and bulk H2O was measured using the transmission method. The excitation processes of H2O molecules after soft X-ray absorption at the interface and bulk liquid are also described.

2. Experimental setup

Fig. 2[link](a) shows schematics of the simultaneous XAS measurement system for solid–liquid interfaces and bulk liquids. The experiments were performed at the soft X-ray beamline BL-13A at the Photon Factory, Institute of Materials Structure Science, High Energy Accelerator Research Organization (KEK-PF), Japan (Mase et al., 2010View full citation; Toyoshima et al., 2013View full citation). Details of the transmission-type liquid cells used for the XAS measurements are provided elsewhere (Nagasaka et al., 2018View full citation; Nagasaka et al., 2020View full citation; Nagasaka & Kosugi, 2021View full citation). The liquid cell was set in an atmospheric helium chamber, which was separated by a 100 nm-thick Si3N4 membrane with a window size of 0.2 mm × 0.2 mm (NTT-AT) from the soft X-ray beamline under ultrahigh vacuum. The liquid layer in the liquid cell consisted of two membranes with a window size of 2 mm × 2 mm: a 100 nm-thick Si3N4 membrane and an Au (20 nm)/Cr (5 nm)/Si3N4 (100 nm) membrane. Transmission measurements were performed by detecting the transmitted soft X-rays using a photodiode detector (Opto Diode IRD AXUV 100). The XAS spectrum of bulk H2O was obtained using the Beer–Lambert law, ln(I0/I), where I is the transmission signal of bulk H2O and the membranes, and I0 is the transmission signal of the membranes.

[Figure 2]
Figure 2
(a) Schematic of the simultaneous XAS measurement method for solid–liquid interfaces and bulk liquids. The back side of the liquid cell is also shown. (b) Cross-sectional view of the liquid cell, where solid–liquid interfaces were measured using the electron-yield method and bulk liquids were measured using the transmission method. A schematic of the precise thickness control of the liquid layer from 20 nm to 40 µm for the XAS measurements in transmission mode is also shown.

The liquid cell was composed of polyether ether ketone (PEEK) resin to provide electrical insulation of the entire system during the electron-yield measurements. The back side of the liquid cell was equipped with terminals for measuring drain currents caused by soft X-ray irradiation and for controlling the liquid temperature. Liquid samples were exchanged using a syringe pump. The liquid cell, consisting of stainless-steel plates, can control the liquid temperature from −5 to 80°C using a temperature-controlled Cu plate with a chiller system (Nagasaka et al., 2020View full citation). By contrast, the liquid cell with PEEK resin was used only at room temperature owing to the resin's low thermal conductivity. The developed liquid cell included an Au terminal to control the liquid temperature. The temperature of the liquid layer was adjusted by controlling the temperature of the Au terminal using a chiller. The liquid temperature could be increased to over 200°C using a sheathed heater for catalytic reactions at high temperatures. Note that the Au terminal was electrically isolated from the heaters for the electron-yield measurements. In the present study, the XAS spectra of the solid–liquid interfaces were measured at room temperature without using a thermal control unit.

Fig. 2[link](b) shows a cross-sectional view of the liquid cell. The liquid layer in the liquid cell was sandwiched between Si3N4 and Au/Cr/Si3N4 membranes with window sizes of 2 mm × 2 mm, which were attached to Si frames with sizes of 10 mm × 10 mm. The liquid layer was prepared by setting Teflon and Au spacers to have a thickness of 100 µm at both sides of the Si frames and was tightly sealed using O-rings outside the Si frames. The Si frames in the central parts were close to each other owing to the pressing of the O-rings. The temperature of the liquid layer was controlled using a chiller or heater system through the Au terminal, which was electrically isolated and measured using a Pt100 resistance temperature detector connected to the Au terminal. The XAS spectra of the electron yield were obtained by measuring the drain currents of the Au/Cr/Si3N4 membrane after soft X-ray absorption through the Au spacer. Electron-yield measurements provided the XAS spectra of the H2O/Au interfaces owing to the low escape depth of Auger electrons. The drain currents were converted to voltages using a programmable preamplifier (NF Corporation CA5351), where the signal-to-noise ratio of the drain currents was improved using an electric filter with slow response time. The voltages were converted into TTL signals using a V–F converter (Tsujicon SN2VF-01), and the TTL signals were collected using a TTL counter (Ortec 974).

The XAS spectra of bulk liquids were obtained via transmission measurements. As shown in the inset of Fig. 2[link](b), the thickness of the liquid layer was precisely controlled from 20 nm to 40 µm by changing the pressure of helium gas around the liquid cell. This method realizes XAS measurements of solutions in wide concentration ranges, where thick liquid layers are prepared for dilute solutions and thin liquid layers are used for condensed solutions. The N K-edge XAS spectra of metal complexes in aqueous solutions can be measured at concentrations of several mM using the water window spectral region (Nagasaka et al., 2024View full citation). The intensities of the transmitted soft X-rays were measured using a photodiode detector. The currents from the detector were converted to voltages using a programmable preamplifier (NF Corporation CA5350), voltages were converted to TTL signals using the V–F converter, and TTL signals were counted using the TTL counter.

3. Application to interfacial water on gold surface

Fig. 3[link] shows the O K-edge XAS spectra of liquid H2O on the Au/Cr/Si3N4 membrane. The spectra of bulk H2O were obtained using the transmission method, and those of the H2O/Au interfaces were simultaneously obtained using the electron-yield method. The side of the Au/Cr/Si3N4 membrane in the liquid cell was irradiated with soft X-rays. The O K-edge XAS spectrum of bulk H2O is consistent with those of previous studies (Smith et al., 2004View full citation; Wernet et al., 2004View full citation; Nagasaka et al., 2017View full citation), where the pre-edge, main-edge and post-edge peaks appear at ∼534.7, ∼537 and ∼540 eV, respectively. The small peak at ∼532 eV is attributed to the oxide layers of the Au/Cr/Si3N4 and Si3N4 membranes. The thickness of the liquid layer was estimated to be 180 nm from the edge jump of the XAS spectrum using the calculated absorption coefficient of liquid H2O (Chantler, 2000View full citation).

[Figure 3]
Figure 3
O K-edge XAS spectra of liquid H2O on the Au/Cr/Si3N4 membrane. The XAS spectra were measured by the irradiation of soft X-rays to the side of the Au/Cr/Si3N4 membrane in the liquid cell, as shown in the inset. (a) O K-edge XAS spectrum of bulk H2O obtained using the transmission method. (b) O K-edge XAS spectrum of the H2O/Au interface obtained using the electron-yield method.

Fig. 3[link](b) shows the O K-edge XAS spectrum of the H2O/Au interface, obtained using the electron-yield method. The pre-edge peak at the H2O/Au interface merges with the shoulder of the main-edge peak owing to the higher energy shift of the pre-edge peak compared with that of bulk H2O. These spectral shapes are consistent with previous XAS measurements obtained using the electron-yield method (Velasco-Velez et al., 2014View full citation; Wang et al., 2025View full citation). The pre-edge peaks in the O K-edge XAS spectra of liquid methanol (Nagasaka, Mochizuki et al., 2014View full citation) and liquid ethanol (Nagasaka et al., 2022View full citation) also appear at the shoulder of the main-edge peak because the electronic structures of the OH groups are changed by the methyl and ethyl groups. The pre-edge peaks of the H2O/Au interfaces reflect the electronic structures of H2O molecules interacting with Au surfaces. The oxide layers of the Au/Cr/Si3N4 membrane give rise to a sharp peak at ∼530.7 eV. These results confirm that the electronic structures of the solid–liquid interfaces are clearly obtained from the XAS spectra using the electron-yield method.

Fig. 4[link] shows the O K-edge XAS spectrum of the H2O/Au interface for a thick liquid layer, obtained using the electron-yield method. The side of the Au/Cr/Si3N4 membrane was irradiated with soft X-rays. Although soft X-rays could not be transmitted to the thick liquid layer under the present conditions, the membrane-related oxide peak and pre-edge peak at the H2O/Au interfaces were obtained in the electron-yield measurements. In other words, XAS measurements of the solid–liquid interfaces are possible by irradiating the side of the Au/Cr/Si3N4 membrane with soft X-rays. Note that the electron-yield XAS spectra of H2O/Au interfaces would have the same spectral profiles at different thickness of the liquid layer for soft X-rays irradiating the Au/Cr/Si3N4 side. However, the profile of the electron-yield XAS spectrum shown in Fig. 4[link] is slightly different from that shown in Fig. 3[link](b). This discrepancy is caused by deforming of the Au/Cr/Si3N4 membrane by varying the thickness of the liquid layer, which would change the soft X-ray irradiation positions.

[Figure 4]
Figure 4
O K-edge XAS spectrum of the H2O/Au interface for a thick liquid layer obtained via the electron-yield method, where soft X-rays irradiated to the side of the Au/Cr/Si3N4 membrane.

4. Discussion on appropriate measurement conditions of interfaces

For obtaining appropriate measurement conditions of bulk H2O and the H2O/Au interfaces, the O K-edge XAS spectra of liquid H2O in transmission mode and the electron yield were measured at different thicknesses of the liquid layer with soft X-ray irradiation to the side of the Si3N4 membrane. Figs. 5[link](a) and 5(b) show the O K-edge XAS spectra of liquid H2O with thicknesses of 80 and 270 nm, respectively, measured using the transmission method. The transmission spectra include the contributions of both bulk liquids and solid–liquid interfaces and exhibit a pre-edge peak at ∼534.7 eV and a peak at ∼532 eV derived from the oxide layers of the Au/Cr/Si3N4 and Si3N4 membranes. The XAS spectrum at 270 nm thickness is mainly derived from bulk liquids because of the thick layers of bulk H2O. In the transmission spectrum at 80 nm thickness, the intensity of the pre-edge peak relative to the main-edge peak is lower than that in the XAS spectrum of the thick liquid layer of bulk H2O, indicating that the contribution of the H2O/Au interface increases in the XAS spectra of the thin liquid layer. However, the contribution of bulk H2O still dominates the XAS spectrum of the thin liquid layer in transmission mode, as indicated by the distinct pre-edge peak of bulk H2O. Note that the spectral contributions of bulk H2O and the H2O/Au interfaces can be separated from the XAS spectra of liquid H2O at different thicknesses, where the ratio of bulk H2O and H2O/Au interfaces varies with different thickness of the liquid layers. In a recent infrared spectroscopy study, the spectrum of vicinal H2O on hydro­gels has been separated from that of bulk H2O by varying the contribution of vicinal H2O and bulk H2O (Maeda et al., 2022View full citation).

[Figure 5]
Figure 5
O K-edge XAS spectra of bulk H2O and H2O/Au interfaces at varying thicknesses of the liquid layer, measured by irradiating soft X-rays to the side of the Si3N4 membrane in the liquid cell (inset). The thicknesses of the liquid layers were (a) 80 nm and (b) 270 nm. (c) Thick liquid layer, where the soft X-rays were strongly absorbed by bulk H2O.

Fig. 5[link] shows the electron-yield XAS spectra of liquid H2O at different thicknesses of the liquid layers with soft X-ray irradiation to the side of the Si3N4 membrane. The electron-yield spectrum at 80 nm thickness is mainly derived from the H2O/Au interface, which shows shoulder structures of the pre-edge peak to the main-edge peak, similar to the electron-yield spectrum shown in Fig. 3[link](b). Note that the membrane-related oxide peak at 532 eV shows a broader profile compared with that shown in Fig. 3[link](b). The soft X-ray irradiation positions of the membranes in the electron-yield measurements at the Si3N4 side were different from those at the Au/Cr/Si3N4 side, and the amount of oxide layers in the irradiation positions would be different from each other. The electron-yield XAS spectrum at 270 nm thickness also shows shoulder structures of the pre-edge peak to the main-edge peak, which is derived from the H2O/Au interface. Fig. 5[link](c) shows the O K-edge XAS spectrum of H2O/Au interfaces with a thick liquid layer. Because soft X-rays were mostly absorbed by bulk H2O before they reached the Au/Cr/Si3N4 membrane, the membrane-related oxide peak exhibits low intensity. The intensity of the pre-edge peak also increases compared with the main-edge peak, indicating that the ratio of bulk H2O increases at the thicker liquid layers.

The electron-yield XAS spectra of liquid H2O at soft X-ray irradiation to the Au/Cr/Si3N4 side are mostly derived from the H2O/Au interface because the effective attenuation length of electrons in liquid H2O is 6 nm (Suzuki et al., 2014View full citation). In the electron-yield spectra at soft X-ray irradiation to the Si3N4 side, on the other hand, the spectral contributions of the H2O/Au interface and bulk H2O would be changed because soft X-rays were absorbed by bulk H2O before they reached the Au/Cr/Si3N4 membrane. Thus, the influence of the different thicknesses of the liquid layers on the electron-yield XAS spectra was discussed using the simulated soft X-ray transmission of liquid H2O. Fig. 6[link](a) shows the simulated soft X-ray transmission of liquid H2O at different thicknesses of the liquid layer, obtained from the calculated soft X-ray absorption coefficients of liquid H2O (Chantler, 2000View full citation). Soft X-rays are not strongly absorbed by liquid H2O before the O K-edge at 529 eV, and the transmission of soft X-rays is 74.5% at 3000 nm thickness. On the other hand, soft X-rays are strongly absorbed by liquid H2O after the O K-edge at 548 eV, and the transmission of soft X-rays is 0.6% at 3000 nm thickness. This means that the transmission spectra of liquid H2O need thin liquid layers below 1000 nm.

[Figure 6]
Figure 6
(a) Simulated soft X-ray transmission of liquid H2O at 529 and 548 eV as a function of the thickness of the liquid layer. (b) Ratio of bulk H2O spectra at 529 and 548 eV as a function of the thickness of the liquid layer, estimated from the simulated soft X-ray transmission.

When the soft X-ray intensities at the Au/Cr/Si3N4 membrane decrease with strong absorption of soft X-rays by bulk H2O, the ratio of bulk H2O spectra caused by the diffusion or exchange of H2O+ cations becomes enhanced, as shown in Fig. 1[link]. Thus, the ratio of the H2O/Au interface would be reduced by decreasing the soft X-ray intensities at the Au/Cr/Si3N4 membrane, whereas that of bulk H2O would not be changed with the different thickness of the liquid layer. Fig. 6[link](b) shows the simulated ratio of bulk H2O in the electron-yield XAS spectra for different thicknesses of the liquid layer, estimated from the simulated soft X-ray transmission. The ratio of bulk H2O was assumed to be 0.1% in the initial condition (0 nm). When the soft X-ray intensity at the Au/Cr/Si3N4 membrane is reduced by increasing the thickness of the liquid layer, the ratio of bulk H2O spectra increases. Because soft X-ray transmission at 529 eV is not strongly changed at 3000 nm thickness, the ratio of bulk H2O spectra is not changed from 0.1%. The ratio of bulk H2O at 548 eV is not different below 1000 nm thickness. This means that the electron-yield XAS spectra mostly come from the H2O/Au interface when the thickness of the liquid layer is below 1000 nm, even if soft X-rays irradiate to the side of the Si3N4 membrane. Therefore, the electron-yield spectra shown in Figs. 5[link](a) and 5(b) mostly come from the H2O/Au interfaces. The ratio of bulk H2O spectra at 548 eV increases at the thicker liquid layers and reaches 14.2% at 3000 nm thickness. Therefore, the electron-yield spectrum at the thick liquid layer includes the contribution of bulk H2O. Note that the electron-yield XAS spectra at the thick liquid layer would show deformed spectral shapes because soft X-ray absorption of liquid H2O at 548 eV is largely different from that at 529 eV. These results suggest that the irradiation of soft X-rays from the Au/Cr/Si3N4 membrane side is important for clear observation of the H2O/Au interface.

5. Conclusion

In this study, simultaneous XAS measurement methods for solid–liquid interfaces and bulk liquids were developed using the electron-yield and transmission methods. The liquid layer in the transmission-type liquid cell consisted of Au/Cr/Si3N4 and Si3N4 membranes, and the thickness of the liquid layer was precisely controlled from 20 nm to 40 µm (Nagasaka et al., 2020View full citation; Nagasaka & Kosugi, 2021View full citation). The O K-edge XAS spectra of bulk H2O at varying liquid-layer thicknesses were obtained using the transmission method. The XAS spectra of the H2O/Au interfaces were obtained using the electron-yield method, which measures the drain current from the Au surface after soft X-ray absorption. When soft X-rays irradiate the side of the Au/Cr/Si3N4 membrane in the liquid cell, the XAS spectrum of the H2O/Au interface acquired in electron-yield mode exhibits a peak derived from the oxide layer of the membrane, as well as a pre-edge peak of H2O molecules that merges with the shoulder of the main-edge peak, which is consistent with previous measurements of H2O/Au interfaces (Velasco-Velez et al., 2014View full citation; Wang et al., 2025View full citation). Pre-edge peaks of the H2O/Au interfaces in the electron-yield XAS spectra were also obtained with soft X-ray irradiation to the Si3N4 side when the thickness of the liquid layer was below 1000 nm. The XAS spectra at the thick liquid layer would include the contribution of bulk H2O owing to the strong absorption of soft X-rays by liquid H2O and would show deformed spectral shapes because the soft X-ray absorption coefficient before the O K-edge is largely different from that after the O K-edge. These results suggest that the XAS spectra of the H2O/Au interfaces and the oxide layer of the membrane should be measured using the electron-yield method with soft X-ray irradiation of the Au/Cr/Si3N4 membrane side.

The present study investigated H2O/Au interfaces using Au/Cr/Si3N4 membranes. XAS measurements of other interfaces will be applicable by forming thin films on the Au surfaces with the formation of SrTiO3 or CoOOH (Wang et al., 2025View full citation), which have been used for catalytic reactions in solutions. Transmission measurements are possible when the catalytic thin films are within several tens of nanometres. Electron-yield measurements are possible when drain currents flow to thin films. Recently, operando XAS measurements of electrochemical reactions have been performed using Au/Cr/Si3N4 membranes as working electrodes (Nagasaka et al., 2013View full citation; Nagasaka, Yuzawa et al., 2014View full citation). Nickel borate electrocatalysts on Au surfaces have been investigated using O K-edge XAS (Yoshida et al., 2015View full citation). O K-edge XAS measurements also observed a high-valent iron-oxo species, involved in catalytic methane oxidation by the μ-nitrido-bridged iron phthalocyanine dimer deposited on a carbon surface (Yamada et al., 2025View full citation). The electron-yield XAS measurements will be applicable to investigate the interfaces of electrocatalysts during reactions. For biological reactions at solid–liquid interfaces, XAS measurements of membrane proteins can be performed using lipid bilayers including proteins (Goh et al., 2023View full citation). The simultaneous XAS measurement methods for solid–liquid interfaces and bulk liquids are applicable for investigating the mechanisms of various catalytic, electrochemical and biological reactions involving solid–liquid interfaces.

Acknowledgements

The experiments were performed under the approval of the Photon Factory Program Advisory Committee (Proposal No. 2025S2-003). The authors thank Editage (https://www.editage.jp) for English language editing.

Conflict of interest

The authors declare no conflicts of interest.

Data availability

The data underlying this study are openly available in Figshare at https://doi.org/10.6084/m9.figshare.31330666.

Funding information

This work was funded by JSPS KAKENHI (Grant Nos. JP25K03396, JP24K21042 and JP25K01843).

References

Return to citationAshley, K. & Pons, S. (1988). Chem. Rev. 88, 673–695.  CrossRef CAS Google Scholar
Return to citationBjörneholm, O., Hansen, M. H., Hodgson, A., Liu, L.-M., Limmer, D. T., Michaelides, A., Pedevilla, P., Rossmeisl, J., Shen, H., Tocci, G., Tyrode, E., Walz, M.-M., Werner, J. & Bluhm, H. (2016). Chem. Rev. 116, 7698–7726.  PubMed Google Scholar
Return to citationChantler, C. T. (2000). J. Phys. Chem. Ref. Data 29, 597–1056.  Web of Science CrossRef CAS Google Scholar
Return to citationDreier, L. B., Nagata, Y., Lutz, H., Gonella, G., Hunger, J., Backus, E. H. G. & Bonn, M. (2018). Sci. Adv. 4, eaap7415.  CrossRef PubMed Google Scholar
Return to citationEkimova, M., Quevedo, W., Faubel, M., Wernet, P. & Nibbering, E. T. J. (2015). Struct. Dyn. 2, 054301.  Web of Science CrossRef PubMed Google Scholar
Return to citationFavaro, M., Jeong, B., Ross, P. N., Yano, J., Hussain, Z., Liu, Z. & Crumlin, E. J. (2016). Nat. Commun. 7, 12695.  Web of Science CrossRef PubMed Google Scholar
Return to citationFondell, M., Eckert, S., Jay, R. M., Weniger, C., Quevedo, W., Niskanen, J., Kennedy, B., Sorgenfrei, F., Schick, D., Giangrisostomi, E., Ovsyannikov, R., Adamczyk, K., Huse, N., Wernet, P., Mitzner, R. & Föhlisch, A. (2017). Struct. Dyn. 4, 054902.  Web of Science CrossRef PubMed Google Scholar
Return to citationFransson, T., Harada, Y., Kosugi, N., Besley, N. A., Winter, B., Rehr, J. J., Pettersson, L. G. M. & Nilsson, A. (2016). Chem. Rev. 116, 7551–7569.  CrossRef CAS PubMed Google Scholar
Return to citationFrati, F., Hunault, M. O. J. Y. & de Groot, F. M. F. (2020). Chem. Rev. 120, 4056–4110.  Web of Science CrossRef CAS PubMed Google Scholar
Return to citationGoh, M. W. S., Tozawa, Y. & Tero, R. (2023). Membranes 13, 854.  CrossRef PubMed Google Scholar
Return to citationHu, L.-Y., Huang, X.-X., Zhou, R.-Y., Wang, Y.-H., Yang, Z.-L. & Li, J.-F. (2024). J. Catal. 430, 115355.  CrossRef Google Scholar
Return to citationKao, L., Feng, X., Ha, Y., Yang, F., Liu, Y.-S., Hahn, N. T., MacDougall, J., Chao, W., Yang, W., Zavadil, K. R. & Guo, J.-H. (2020). Surf. Sci. 702, 121720.  CrossRef Google Scholar
Return to citationKoralek, J. D., Kim, J. B., Brůža, P., Curry, C. B., Chen, Z., Bechtel, H. A., Cordones, A. A., Sperling, P., Toleikis, S., Kern, J. F., Moeller, S. P., Glenzer, S. H. & DePonte, D. P. (2018). Nat. Commun. 9, 1353.  Web of Science CrossRef PubMed Google Scholar
Return to citationKubin, M., Guo, M. Y., Ekimova, M., Baker, M. L., Kroll, T., Källman, E., Kern, J., Yachandra, V. K., Yano, J., Nibbering, E. T. J., Lundberg, M. & Wernet, P. (2018). Inorg. Chem. 57, 5449–5462.  CrossRef CAS PubMed Google Scholar
Return to citationMaeda, S., Chikami, S., Latag, G. V., Song, S., Iwakiri, N. & Hayashi, T. (2022). Molecules 27, 2130.  CrossRef PubMed Google Scholar
Return to citationMase, K., Toyoshima, A., Kikuchi, T., Tanaka, H., Amemiya, K. & Ito, K. (2010). AIP Conf. Proc. 1234, 709–712.  CrossRef CAS Google Scholar
Return to citationMasuda, T., Yoshikawa, H., Noguchi, H., Kawasaki, T., Kobata, M., Kobayashi, K. & Uosaki, K. (2013). Appl. Phys. Lett. 103, 111605.  Web of Science CrossRef Google Scholar
Return to citationMeibohm, J., Schreck, S. & Wernet, P. (2014). Rev. Sci. Instrum. 85, 103102.  Web of Science CrossRef PubMed Google Scholar
Return to citationMyneni, S., Luo, Y., Näslund, L. Å., Cavalleri, M., Ojamäe, L., Ogasawara, H., Pelmenschikov, A., Wernet, P., Väterlein, P., Heske, C., Hussain, Z., Pettersson, L. G. M. & Nilsson, A. (2002). J. Phys. Condens. Matter 14, L213–L219.  CrossRef CAS Google Scholar
Return to citationNagasaka, M. (2024). J. Phys. Chem. Lett. 15, 5165–5170.  CrossRef CAS PubMed Google Scholar
Return to citationNagasaka, M., Bouvier, M., Yuzawa, H. & Kosugi, N. (2022). J. Phys. Chem. B 126, 4948–4955.  CrossRef CAS PubMed Google Scholar
Return to citationNagasaka, M. & Kosugi, N. (2021). Chem. Lett. 50, 956–964.  CrossRef CAS Google Scholar
Return to citationNagasaka, M., Mochizuki, K., Leloup, V. & Kosugi, N. (2014). J. Phys. Chem. B 118, 4388–4396.  Web of Science CrossRef CAS PubMed Google Scholar
Return to citationNagasaka, M., Tsuru, S. & Yamada, Y. (2024). Phys. Chem. Chem. Phys. 26, 23636–23645.  CrossRef CAS PubMed Google Scholar
Return to citationNagasaka, M., Yuzawa, H., Horigome, T., Hitchcock, A. P. & Kosugi, N. (2013). J. Phys. Chem. C 117, 16343–16348.  CrossRef CAS Google Scholar
Return to citationNagasaka, M., Yuzawa, H., Horigome, T. & Kosugi, N. (2014). Rev. Sci. Instrum. 85, 104105.  Web of Science CrossRef PubMed Google Scholar
Return to citationNagasaka, M., Yuzawa, H., Horigome, T. & Kosugi, N. (2018). J. Electron Spectrosc. Relat. Phenom. 224, 93–99.  CrossRef CAS Google Scholar
Return to citationNagasaka, M., Yuzawa, H. & Kosugi, N. (2017). J. Phys. Chem. B 121, 10957–10964.  CrossRef CAS PubMed Google Scholar
Return to citationNagasaka, M., Yuzawa, H. & Kosugi, N. (2020). Anal. Sci. 36, 95–99.  CrossRef CAS PubMed Google Scholar
Return to citationNakamura, M., Kato, H. & Hoshi, N. (2008). J. Phys. Chem. C 112, 9458–9463.  CrossRef CAS Google Scholar
Return to citationNihonyanagi, S., Ye, S., Uosaki, K., Dreesen, L., Humbert, C., Thiry, P. & Peremans, A. (2004). Surf. Sci. 573, 11–16.  CrossRef CAS Google Scholar
Return to citationOdelius, M., Cavalleri, M., Nilsson, A. & Pettersson, L. G. M. (2006). Phys. Rev. B 73, 024205.  CrossRef Google Scholar
Return to citationPettinger, B., Philpott, M. R. & Gordon, J. G. II (1981). Surf. Sci. 105, 469–474.  CrossRef CAS Google Scholar
Return to citationQiu, Y.-R., Peng, W., Zhang, R., Wang, Y.-H., Guo, W. & Li, J.-F. (2025). Natl. Sci. Rev. 12, nwaf284.  CrossRef PubMed Google Scholar
Return to citationRen, J., Achilleos, D. S., Golnak, R., Yuzawa, H., Xiao, J., Nagasaka, M., Reisner, E. & Petit, T. (2019). J. Phys. Chem. Lett. 10, 3843–3848.  CrossRef CAS PubMed Google Scholar
Return to citationSchön, D., Golnak, R., Tesch, M. F., Winter, B., Velasco-Velez, J.-J., Aziz, E. F. & Xiao, J. (2017). J. Phys. Chem. Lett. 8, 5136–5140.  PubMed Google Scholar
Return to citationSchön, D., Xiao, J., Golnak, R., Tesch, M. F., Winter, B., Velasco-Velez, J. J. & Aziz, E. F. (2017). J. Phys. Chem. Lett. 8, 2087–2092.  PubMed Google Scholar
Return to citationSchreck, S., Gavrila, G., Weniger, C. & Wernet, P. (2011). Rev. Sci. Instrum. 82, 103101.  Web of Science CrossRef PubMed Google Scholar
Return to citationSellberg, J. A., Kaya, S., Segtnan, V. H., Chen, C., Tyliszczak, T., Ogasawara, H., Nordlund, D., Pettersson, L. G. M. & Nilsson, A. (2014). J. Chem. Phys. 141, 034507.  Web of Science CrossRef PubMed Google Scholar
Return to citationShimizu, T. K., Maier, S., Verdaguer, A., Velasco-Velez, J.-J. & Salmeron, M. (2018). Prog. Surf. Sci. 93, 87–107.  CrossRef CAS Google Scholar
Return to citationSmith, J. D., Cappa, C. D., Wilson, K. R., Messer, B. M., Cohen, R. C. & Saykally, R. J. (2004). Science 306, 851–853.  CrossRef PubMed CAS Google Scholar
Return to citationSmith, J. W. & Saykally, R. J. (2017). Chem. Rev. 117, 13909–13934.  Web of Science CrossRef CAS PubMed Google Scholar
Return to citationStöhr, J. (1992). NEXAFS Spectroscopy. Berlin: Springer.  Google Scholar
Return to citationSuzuki, Y., Nishizawa, K., Kurahashi, N. & Suzuki, T. (2014). Phys. Rev. E 90, 010302.  CrossRef Google Scholar
Return to citationTabassum, T., Das, B., Park, C., Ghosh, S., Naujoks, D., Piontek, S. M., Ludwig, A., Marx, D. & Petersen, P. B. (2025). Phys. Rev. Lett. 135, 116201.  CrossRef PubMed Google Scholar
Return to citationTimoshenko, J. & Roldan Cuenya, B. (2021). Chem. Rev. 121, 882–961.  Web of Science CrossRef CAS PubMed Google Scholar
Return to citationTokushima, T., Harada, Y., Takahashi, O., Senba, Y., Ohashi, H., Pettersson, L. G. M., Nilsson, A. & Shin, S. (2008). Chem. Phys. Lett. 460, 387–400.  Web of Science CrossRef CAS Google Scholar
Return to citationToney, M. F., Howard, J. N., Richer, J., Borges, G. L., Gordon, J. G., Melroy, O. R., Wiesler, D. G., Yee, D. & Sorensen, L. B. (1994). Nature 368, 444–446.  CrossRef CAS Google Scholar
Return to citationToyoshima, A., Kikuchi, T., Tanaka, H., Mase, K., Amemiya, K. & Ozawa, K. (2013). J. Phys. Conf. Ser. 425, 152019.  CrossRef Google Scholar
Return to citationvan Spronsen, M. A., Zhao, X., Jaugstetter, M., Escudero, C., Duchoň, T., Hunt, A., Waluyo, I., Yang, P., Tschulik, K. & Salmeron, M. B. (2021). J. Phys. Chem. Lett. 12, 10212–10217.  CrossRef CAS PubMed Google Scholar
Return to citationVelasco-Velez, J.-J., Wu, C. H., Pascal, T. A., Wan, L. F., Guo, J., Prendergast, D. & Salmeron, M. (2014). Science 346, 831–834.  CAS PubMed Google Scholar
Return to citationWang, Y.-H., Zheng, S., Yang, W.-M., Zhou, R.-Y., He, Q.-F., Radjenovic, P., Dong, J.-C., Li, S., Zheng, J., Yang, Z.-L., Attard, G., Pan, F., Tian, Z.-Q. & Li, J.-F. (2021). Nature 600, 81–85.  CrossRef CAS PubMed Google Scholar
Return to citationWang, Z., Toyoshima, R., Enomoto, K., Mase, K. & Kondoh, H. (2025). J. Phys. Chem. C 129, 2570–2581.  CrossRef CAS Google Scholar
Return to citationWernet, P., Nordlund, D., Bergmann, U., Cavalleri, M., Odelius, M., Ogasawara, H., Näslund, L. Å., Hirsch, T. K., Ojamäe, L., Glatzel, P., Pettersson, L. G. M. & Nilsson, A. (2004). Science 304, 995–999.  CrossRef PubMed CAS Google Scholar
Return to citationWilson, K. R., Rude, B. S., Catalano, T., Schaller, R. D., Tobin, J. G., Co, D. T. & Saykally, R. J. (2001). J. Phys. Chem. B 105, 3346–3349.  CrossRef CAS Google Scholar
Return to citationWu, C. H., Pascal, T. A., Baskin, A., Wang, H., Fang, H.-T., Liu, Y.-S., Lu, Y.-H., Guo, J., Prendergast, D. & Salmeron, M. B. (2018). J. Am. Chem. Soc. 140, 16237–16244.  CrossRef CAS PubMed Google Scholar
Return to citationYamada, Y., Toyoda, Y., Nagasaka, M., Nakatani, N., Koitaya, T., Kuzume, A. & Tanaka, K. (2025). ChemistryEurope 3, e202500277.  CrossRef Google Scholar
Return to citationYang, B. X. & Kirz, J. (1987). Phys. Rev. B 36, 1361–1364.  CrossRef CAS Google Scholar
Return to citationYoshida, M., Mitsutomi, Y., Mineo, T., Nagasaka, M., Yuzawa, H., Kosugi, N. & Kondoh, H. (2015). J. Phys. Chem. C 119, 19279–19286.  CrossRef CAS 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