Applications of Hard X-ray Photoelectron Spectroscopy

Applications of Hard X-ray Photoelectron Spectroscopy to bulk materials

Jingru Chi

ABSTRACT

X-ray Photoelectron Spectroscopy is a powerful, relevant and non-destructive method for studying atoms, molecules and surfaces [1]. However, investigations are limited to atoms, molecules and surfaces since low energy electrons limit the depth resolution, resulting in three-dimensional bulk state cannot be detected. Hard X-ray Photoelectron Spectroscopy (HAXPES) with high kinetic energy photoelectrons uses excitation by X-ray of 2-15keV and a high energy analyzer which make it possible to measure bulk and determine bulk electronic structure properties of materials [2]. The recoil effects of photoelectrons in valence band states and core levels are the principal studies of HAXPES. In this paper, HAXPES uses undulator synchrotron X-rays at SPring-8. The results of high energy photoelectron spectroscopy of the valence bands and sensitivity of bulk have shown that the measured valence band spectra are indispensable in studying the bulk electronic structure.

Introduction

The Nobel Prize in 1981 was awarded to Kai Siegbahn for developing the method of Electron Spectroscopy for Chemical Analysis (ESCA), now which is presented as X-ray Photoelectron Spectroscopy (XPS). Since then, XPS becomes one of the most useful and non-destructive techniques for analyzing the electronic structure of atoms, molecules and surfaces. In early studies, XPS uses Mo K? (= 17.479keV), Cr K? (= 5.417keV) or Cu K? (= 8.047 keV) hard X-ray sources. Gradually it out of date with the discovery that lower energy soft X-rays such as the Mg K? (= 1253.6eV) and Al K? (= 1486.7eV) sources has the higher energy resolution [3]. Reducing incident energy improved the surface sensitivity of XPS. However, soft X-rays source limits the depth resolution to 5nm for Angle-Resolved X-ray Photoelectron Spectroscopy (ARXPS) and 10nm for inelastic loss analysis [4], so it cannot detect the deeply buried layers without destructive ways like ion sputtering and etching which are time-consuming and hard to control. In the last few years, the developments of new bright synchrotron photon sources, the availability of monochromators with a resolving power about 10⁵, and electron analyzers which can analyze the 10 keV electrons with meV resolution have created the new possibilities for Hard X-ray Photoelectron Spectroscopy (HAXPES) [5], which made it possible to analyze the bulk structure without using destructive methods.

HAXPES has not been well-developed until the appearance of the third generation synchrotrons, which can generate high-brilliance, high-flux X-rays that enables one to perform experiments with the HAXPES in very low photoionization cross-section [1]. As well the developments of electron analyzer with the high kinetic energy range also contributed to improve HAXPES measurements [6]. The superiority of HAXPES is the considerable probing depth owing to the increased electron mean free path [7], made it possible to detect the electronic structure of bulk materials. With the excitation energy of 8 keV, the escape energy can be greater than 90 A [1]. With high kinetic energies of electrons, the coral level and the valence band can be detected in the bulk materials. According to these advantages, HAXPES is one of the best ways to perform sensitive photoemission spectroscopy on correlated systems such as thin films, multilayer systems and devices [1]. Several investigations on bulk materials have been reported. One is the measurement of the valence band of Co2Mn1?xFexSi (x = 0, 0.5, 1) be excited by photons which have energy about 8 keV [8, 9]. This experiment was the proof that the XPS with hard X-rays has better sensitivity on bulk electronic structure than the conventional XPS with soft X-rays.

Fundamentals of HAXPES
Model of photoionization using hard X-rays

When use the conventional theoretical model to describe the photoemission, including the differential cross-section of photoionization for photon energies from 2 keV to 15 keV, the model based on the power series expansion of the electron-photon interaction operator cannot perfectly explain photoemission though just using a limited number of the terms of the expansion [2]. However, a more complex expansion of the electron-photon interaction operator developed by Fujikawa [11, 12] contained all the electric dipole operators and other multipole terms can explain it well. These models indicate that contributions from electric quadrupole and magnetic dipole transitions cannot be ignored anymore when photoelectrons are excited by high energy X-rays and beyond the electric dipole transitions [2].

Systems

Fig. 1. HEARP Lab system [10]

There are just so many kinds of HAXPES, here just introduce the HEARP Lab system. In this system, it needs the monochromatic X-ray source with 5–6 keV photon energies and the high energy electron analyzer with angular resolution capability for the measurements of takeoff angle dependence and X-ray photoelectron diffraction [10]. To meet these requirements, HEARP uses the Cr K? X-ray source, a wide acceptance objective lens, and a high energy version of VG SCIENTA R4000 analyzer [10], as it showed in Fig. 1.

A Cr K? X-ray source is shown in Fig. 2(a). The main body contains the water cooling system and Cr target. At the first, the monochromatic Cr K? X-rays are emitted by a focused electron beam with the maximum acceleration energy of 20keV, then X-rays go through the bent crystal monochromator with a 300mm Rowland circle and focus onto a sample surface [10], as schematically shown in Fig. 2(b). The X-ray spot size ranges from 1.5?m to 200?m by raster scanning of the electron beam [10].

(b)

(a)

Fig. 2. X-ray source [10]. (a) Photograph of UHV compatible flange mounted Cr K? X-ray source [10]. (b) Schematic diagram of X-ray source. The electron beam is focused on a water cooled Cr target [10] and it excite Cr K? X-rays. Then it monochromatized by an elliptically bent Ge crystal with 422 reflection [10] and direct irradiated to the sample surface.

The objective lens, as it shown in Fig. 3(a), is set in front of the analyzer. The angle acceptance of the lens is about ±7°, when it is combined with the VG SCIENTA R4000 10 KV hemispherical analyzer, the angle acceptance can achieve to ±35° [10]. The magnification factor is 5, as well the magnification factor of analyzer input lens is 5, so the total magnification factor is 25 [10], and the working distance is 11mm from the aperture [10].

(a)

Fig. 3

Performance

The performance of HEARP system is evaluated by measuring Au 3d5/2 photoelectrons emitted from an Au strip. The total energy resolution is 0.53 eV [13]. The result of the experiment showed the acceptance angle is ±35° and a resolution less than 0.5° [10]. When it is provided with the objective lens, the acquisition time for the Au 3d spectrum excited by the same Cr K? source is 16 min, which is seven times better than without the objective lens [10].

Table 1 shows the differences in HEARP Lab system and HXPES at BL47XU beamline system.

Table 1

Comparison of HEARP Lab system with HXPES at BL47XU beamline system. They all have the same analyzer with the same pass energy of 200 eV. In the Lab system, it uses entrance slit of curved 0.8mm, beamline system uses curved 0.8mm. The X-ray excitation power is 45W (15 kW, 3.0 mA) [10].

Applications

References

[1] Siham Ouardi, Gerhard H. Fecher, Claudia Felser, ‘Bulk electronic structure studied by hard X-ray photoelectron spectroscopy of the valence band: The case of intermetallic compounds,’ Journal of Electron Spectroscopy and Related Phenomena, 190 (2013) 249–267.

[2] Laszlo Kover, ‘X-ray photoelectron spectroscopy using hard X-rays,’ Journal of Electron Spectroscopy and Related Phenomena, 178–179 (2010) 241–257.

[3] M. Taguchi, Y.Takata, A.Chainani, ‘Hard X-ray photoelectron spectroscopy: A few recent applications,’ Journal of Electron Spectroscopy and Related Phenomena 190 (2013) 242–248.

[4] P. Risterucci, O. Renault, E. Martinez, B. Detlefs, V. Delaye, J. Zegenhagen, C. Gaumer, G. Grenet, and S. Tougaard, ‘Probing deeper by hard x-ray photoelectron spectroscopy,’ Applied Physics Letters 104 (2014) 051608.

[5] Ronny Knut, Rebecka Lindblad, Mihaela Gorgoi, Hakan Rensmo, Olof Karis, ‘High energy photoelectron spectroscopy in basic and applied science: Bulk and interface electronic structure’, Journal of Electron Spectroscopy and Related Phenomena 190 (2013) 278–288.

[6] S. Ueda, ‘Application of hard X-ray photoelectron spectroscopy to electronic structure measurements for various functional materials,’ Journal of Electron Spectroscopy and Related Phenomena, 190 (2013) 235–241.

[7] C. Dallera, L. Duo, L. Braicovich, G. Panaccione, G. Paolicelli, B. Cowie, J. Zegen- hagen, ‘Results and perspectives in hard X-ray photoemission spectroscopy (HAXPES) from solids’, Appl. Phys. Lett. 85 (2004) 4532.

[8] B. Balke, G.H. Fecher, H.C. Kandpal, C. Felser, K. Kobayashi, E. Ikenaga, J.-J. Kim, S. Ueda, ‘Properties of the quaternary half-metal-type Heusler alloy Co2Mn1?xFexSi,’ Phys. Rev. B 74 (2006) 104405.

[9] G.H. Fecher, B. Balke, S. Ouardi, C. Felser, G. Schonhense, E. Ikenaga, J.J. Kim, S. Ueda, K. Kobayashi J., ‘High energy, high resolution photoelectron spectroscopy of Co2Mn1?xFexSi,’ Phys. D: Appl. Phys. 40 (2007) 1576.

[10] Keisuke Kobayashi, Masaaki Kobata, Hideo Iwai, ‘Development of a laboratory system hard X-ray photoelectron spectroscopy and its applications,’ Journal of Electron Spectroscopy and Related Phenomena, 190 (2013) 210–221.

[11] Takashi Fujikawa, Rie Suzuki,Hiroko Arai, Hiroshi Shinotsuka,Laszlo Kover, ‘Nondipole effects in photoemission angular distribution excited by high-energy X-rays’, J. Electron Spectrosc.

Relat. Phenom. 159 (2007) 14.

[12] Rie Suzuki,Hiroko Arai,Hiroshi Shinotsuka,Takashi Fujikawa, ‘Theory of High-Energy Photoemission’, e-J. Surf. Sci. Nanotechnol. 3 (2005) 373.

[13] Kobata M,Pis I,Iwai H,Yamazui H,Takahashi H,Suzuki M,Matsuda H,Daimon H,Kobayashi K., ‘Development of the hard-X-ray angle resolved X-ray photoemission spectrometer for laboratory use’, Anal Sci.26 (2010) 227-32.

[12]

. In Heusler thin films of Co2MnSi and Fig. 1 shows the thin films covered with MgO, SiOx and the protective layers with AlOx or Ru, the thickness of the MgO and SiOx layers are from 1 nm to 20 nm. When it is covered with AlOx layer,