Background radiation in inelastic X-ray scattering and X-ray emission spectroscopy. A study for Johann-type spectrometers
Introduction
With the advent of high-brilliance tunable X-ray sources, as third generation synchrotrons, many spectroscopic techniques have benefited from the possibility of performing high energy resolution experiments along with a high photon flux. Among them, inelastic X-ray scattering spectroscopy (IXS) is nowadays a well-established technique for investigating core and valence electron excitations on a variety of systems [1]. In particular, unique field of IXS using hard X-rays is the study of bulk electronic structure and excitations of systems under extreme conditions [[2], [3], [4]]. In a like manner, X-ray emission spectroscopy has emerged as a powerful element-specific technique to probe the electronic structure of the emitting atom and also its dependence on the chemical environment, having been mainly applied to 3d transition metal complexes [[5], [6], [7]]. Furthermore, in the field of the fundamental atomic physics, high resolution XES measurements can provide valuable information to support newly developed theoretical approaches to calculate atomic structure and to simulate satellite spectra following shake processes [[8], [9], [10]]. Such IXS and XES studies are strongly demanding for high energy resolution. This requirement is usually achieved by means of a Johann-type spectrometer using nearly backdiffracting crystals for the energy analysis of the scattered or fluorescence X-ray photons, according to the kind of experiment.
A precise data processing in X-ray spectroscopy is generally a challenging task. Besides usual energy-dependent corrections, as absorption of the incident and outgoing beam in the sample, scale factors, reflectivity of the analyser crystal, etc., which are required to be applied depending on the type of experiment one is concerned with, the subtraction of spurious spectral components may demand attention. Among them, the background radiation owing to spurious radiation and the contribution from multiple scattering processes occurring in the sample are the most commonly taken into account. An incorrect modelling of these spectral components may lead to misleading results, in particular in experiments intended to investigate weak spectral features in IXS or XES spectra. Several works have been devoted to propose procedures for data analysis and for experiment planning, particularly in the field of IXS [[11], [12], [13]]. The influence of multiple scattering on the IXS spectrum was also investigated using a Monte Carlo code [14]. That study demonstrated that under conditions realised in experiments, the double scattering contribution to the valence electron part of the spectrum is at most and that no additional structure into the IXS spectra of simple metals is introduced by multiple scattering effects.
Usually, in a first approximation, the spectral distribution of the background radiation is assumed to be constant or, in some cases, simulated by a linear function. However, a detailed analysis of the different spectral components of an XES spectrum revealed a non-linear behaviour of the intensity distribution of the background radiation under particular experimental conditions, among them non-evacuated beam paths [15]. While in a few cases X-ray spectrometers have been enclosed in vacuum chambers [[16], [17], [18]], most of the IXS and XES spectrometers are operated in air. For those experiments requiring a reduction of the intensity of background radiation to enhance the signal-to-background ratio, He-filled bags are usually introduced into the beam paths within the spectrometer.
It is the aim of this work to investigate the background owing to parasitic scattering originated by radiation scattered from the beam paths in Johann-type spectrometers. An analytical calculation formalism to simulate the energy distribution of the background radiation intensity in IXS and XES experiments is presented. Additional structures introduced into the IXS/XES measured spectra by the background radiation and, more generally, the validity of approximating the background by a constant or a lineal function will be discussed. The dependence of the background radiation on several experimental and geometrical parameters will be studied.
Section snippets
Calculation formalism
In this section, a general calculation formalism to compute the spectral distribution of background radiation in IXS and XES experiments is presented. It is considered that the experiments are performed using Johann-type spectrometers in Rowland geometry. The background radiation is assumed to be originated by scattering of X-ray photons in the non-evacuated beam paths of the spectrometer. A schematic of a Johann-typespectrometer is shown in Fig. 1. Briefly, the principle of operation is as
General considerations and calculation details
In IXS experiments performed in inverse geometry, which is the usual mode for energy analysis close to the backdiffraction condition, the energy-loss spectrum is measured by varying the incident photon energy while keeping the analysed energy fixed. Hence, sample, analyser and detector are kept fixed on the Rowland circle of the spectrometer. The energy transfer ranges typically from a few tens of eV to several hundreds of eV. Eq. (1) requires, as input quantity, the IXS spectrum of the
General considerations and calculation details
In XES experiments, the X-rayspectrometer of Fig. 1 is operated in scanning mode. The analysis of the X-ray emission spectrum in a narrow energy interval, usually several tens of eV, is performed by scanning the analyser in synchrony with the detector. The excitation energy of the incident X-ray beam is kept fixed.
The X-ray emission spectrum in Eqs. (1), (2) can be described by a simple analytical model, which consists of a sum of Lorentzian functions [32] representing diagram and
Conclusions
A calculation formalism to evaluate the background radiation in IXS and XES experiments performed with Johann-type spectrometers was presented. The calculation model takes into account the background radiation originated by scattering processes along the beam paths in the spectrometer. Different conditions realised in usual experiments were investigated. In IXS as well as in XES experiments, the background radiation is mainly originated in the sample-analyser beam path. In this beam path, the
Acknowledgements
This work was supported by SeCyT–UNC (Grant number 30720150101889CB) and LNLS (Brazilian Synchrotron Light Laboratory), CNPEM/MCTI. O.A.P. and L.M.B. acknowledge CONICET for graduate fellowships.
References (42)
- et al.
High resolution core hole X-ray spectroscopy in transition metal complexes-electronic and structural information
Coord. Chem. Rev.
(2005) - et al.
High resolution study of K and K1,3 X-ray emission lines from Mn-compounds
Chem. Phys.
(2010) - et al.
Multiple scattering in inelastic X-ray scattering spectroscopy: A Monte-Carlo study
Nucl. Instrum. Methods Phys. Res. B
(1997) - et al.
Performance of a spherically bent crystal spectrometer for inelastic scattering experiments with synchrotron X-radiation
Nucl. Instrum. Methods Phys.
(1984) - et al.
Performance of a versatile instrumentation for inelastic X-ray scattering spectroscopy (IXSS) with synchrotron radiation
Nucl. Instrum. Methods Phys. Res. Sect. A
(1992) - et al.
Efficient implementation of core-excitation Bethe-Salpeter equation calculations
Comput. Phys. Comm.
(2015) - et al.
Contribution of the [1s3d] shake process to K spectra in 3d elements
Radiat. Phys. Chem.
(2006) - et al.
K-MM radiative Auger effect in solid Ca, Ti and Cr targets after ionization with 0.7 –1.5 MeV protons
Nucl. Instrum. Methods Phys. Res., Sect. B
(1992) K X-ray production cross sections, K/K ratios, and radiative Auger ratios for protons impacting low-Z elements
Nucl. Instrum. Methods Phys. Res., Sect. A
(1999)Relativistic hartree-slater values for k and l x-ray emission rates
Atom. Data Nucl. Data Tables
(1974)