- Email: [email protected]
Chemical shift in Lα, Lβ1, Lβ3,4, Lβ2,15, Lγ1 and Lγ2,3 emission lines of 47Ag, 48Cd and 50Sn compounds
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
Contents lists available at ScienceDirect
Nuclear Inst. and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
Chemical shift in Lα, Lβ1, Lβ3,4, Lβ2,15, Lγ1 and Lγ2,3 emission lines of 48Cd and 50Sn compounds
47Ag,
MARK
⁎
Harpreet Singh Kaintha, , Ranjit Singha,b, Gurjot Singha, D. Mehtaa a b
Physics Department, Panjab University, Chandigarh 160014, India Department of Radiotherapy, PGIMER Chandigarh 160012, India
A R T I C L E I N F O
A B S T R A C T
Keywords: L emission lines Chemical effect Theoretical methods Bond length Relative line-width CK process Effective charge and WDXRF
Positive and negative shifts in L shell emission lines of 47Ag, 48Cd and 50Sn elements in different chemical compounds were determined from their recorded X-ray emission spectra in high resolution wavelength dispersive X-ray fluorescence (WDXRF) spectrometer. In 47Ag compounds, the measured energy shifts in Lα X-ray emission line were in the ranges from (0.12 to 0.40) eV, Lβ1 (0.27 to 0.36) eV, Lβ3,4 (1.10 to 4.89) eV, Lγ1 (−0.09 to 1.13) eV and Lγ2,3 (−2.08 to 0.59) eV. Likewise, for 48Cd compounds, the estimated shifts in Lα X-ray emission lines were in the range (−0.27 to 0.69) eV, Lβ1 (0.50 to 2.06) eV, Lβ2,15 (0.12 to 0.79), Lβ3,4 (−0.62 to 1.79) eV, Lγ1 (0.10 to 1.35) eV and Lγ2,3 (−0.73 to 1.75) eV, while for 50Sn compounds, the measured shifts in Lα X-ray emission lines were in the range of (0.02 to 1.81) eV, Lβ1 (0.11 to 0.78) eV, Lβ2,15 (0.15 to 1.40), Lβ3,4 (0.17 to 2.01) eV, Lγ1 (0.09 to 1.08) eV and Lγ2,3 (0.17 to 1.40) eV respectively. The effective charges (qP, qS, qL and qB) were calculated by four different theoretical methods (Pauling method, Suchet method, Levine method and Batsonav method) and found to be linear dependent with the chemical shift. Further, the measured chemical shifts were correlated with bond length, relative line-width (FWHM), effective charge, electronegativity, number of ligands and Coster-Kronig (CK) transition processes.
1. Introduction Chemical state analysis becomes more interesting topic for many researchers working in the spectroscopy field. The chemical effects can be studied by measuring either the shift in energies of characteristic Xray photons [1–2], relative intensities of X-ray photons [3] or the formation of satellite lines [4]. X-ray photoelectron spectroscopy (XPS), energy dispersive X-ray fluorescence (EDXRF) spectrometers, photoemission spectroscopy (PES) and X-ray absorption near-edge structures (XANES) are the most common employed techniques used to identify the chemical state analysis in different compounds. All these techniques are non-destructive and bulk-sensitive to the chemical states of ir-radiating atoms, which provide useful insight of electron density, quantum states and atomic and electronic structures in the materials. During the formation of compound, metal atom transforms into a positive ion, results in change in the core binding energy of the electron. This deviation in the energy of a metal atom with respect to the different chemical compounds termed as chemical shift (ΔE), and is defined as Emetal – Ecompound, where Emetal denotes central peak position of pure metal and Ecompound is the central peak position of the relevant compounds. However, having limited detector resolution and poor data
⁎
reliability, these types of spectrometers suffer with lack of stability, electrostatic effects and uncertainty in the emission curves which requires large number of complex corrections [4,6–10]. To avoid these types of problems to some extent, wavelength dispersive X-ray (WDXRF) spectrometer with different ranging of high resolution crystals are used. Earlier many authors [5,11–15] reported the chemical effects in K and L shell emission lines of different low and medium Z elements. Chemical effect combination and their relative theoretical explanation on different chemical compounds, however, have not been established completely. Various factors like CosterKronig (CK) transitions, line-width (FWHM), effective charge and bond length etc are responsible for the cause of chemical shift. In past, PutilaMantyla et al. [16] calculated the line-widths of Lα1 and Lβ1,3,4 emission lines in the atomic range 41 ≤ Z ≤ 50 and showed sudden drop of L1 line-width at 50Sn element due to the closing of L1-L3M4,5 CK process. Likewise CK transitions, effective charge also plays an important role in chemical shift. Many researchers [17–22] studied the effective charge by various methods and techniques to show their impact on the chemical shift. In the present work, our objective is to study the cause of chemical shift in L shell emission lines of different 47Ag, 48Cd and 50Sn
Corresponding author. E-mail address: [email protected] (H. Singh Kainth).
http://dx.doi.org/10.1016/j.nimb.2017.10.027 Received 14 August 2017; Received in revised form 23 October 2017; Accepted 24 October 2017 0168-583X/ © 2017 Elsevier B.V. All rights reserved.
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 1. The schematic diagram of S8 TIGER WDXRF X-ray tube setup (situated at Panjab University Chandigarh).
Table 1 Represents the oxidation state, manufacture’s name and nearest bond distance of 48Cd and 50Sn compounds. Compounds
Oxidation state
Manufacturer
Bond distance (Å)
Ag foil
0
–
AgCl AgBr AgI Cd foil
+1 +1 +1 0
CdCO3 CdB4O7 Cd3(PO4)2 CdCl2 CdI2 CdO CdS Sn foil
+2 +2 +2 +2 +2 +2 +2 0
SnF4 SnF2 SnO SnO2 Sn(CrO4)2 SnCl2
+4 +2 +2 +4 +4 +2
Micromatter, Deer Harbor, WA CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent Micromatter, Deer Harbor, WA CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent Micromatter, Deer Harbor, WA Sigma Aldrich Sigma Aldrich Sigma Aldrich Sigma Aldrich Sigma Aldrich Sigma Aldrich
conditions. The schematic diagram of WDXRF apparatus was shown in Fig. 1. In WDXRF spectrometer, the high performance X-ray tube with narrow beam path and advanced analyzer crystals gave the highest intensity and fast data acquisition. The collimator masks were located between the sample and collimator to cut-off the radiation coming from the edge of the cup aperture. The space between the collimator plates restricted the maximum angle for divergence to the X-rays incident on the analyzing crystals. In the present measurement, LiF (2 0 0) high resolution analyzing crystal with 2d spacing (4.026 Å) is used. The Al filters of different thickness were used to avoid the contribution from bremsstrahlung radiations. The characteristic lines of different energies emitted from the sample were diffracted into different directions by the analyzing crystals. The diffracted characteristic lines were further detected by scintillation counter and proportional counter. For analysis procedure, SPECTRA plus software automatically corrected all matrix effects, measurement of calibration set, mathematical data processing and optimal calibration equations based on the fundamental parameters. During measurement, WDXRF spectrometer was evacuated down to ∼15 Pa and temperature was stabilized at 37 °C. The typical acquisition time for each spectrum was 20 min. Each compound was analyzed at least five times to check the reproducibility. The abscissa axis of a chart was displayed in the 2θ angle with step size of 0.046° and step time of 0.15 s. In the present study, the abscissa axis was converted into the energy (eV) with the use of Bragg’s law given by
47Ag,
2.365 2.528 2.737 — 2.023 2.027 2.125 2.404 2.769 1.813 2.035 – 1.890 1.943 1.771 1.981 2.765 2.355
nλ = 2dsinθ
(1)
where ‘λ’ is the wavelength of the X-ray, ‘d’ is the inter-atomic spacing between two layers of crystals, ‘θ’ is the angle between incident rays and surface normal to the crystal and ‘n’ is the order of diffraction pattern. The information about manufacturer, oxidation state and bond length of central metal atom in the compounds are given in Table 1. All the fine powder samples were spread uniformly on mylar film (∼2 µm thickness). The given powder samples were then placed on the top of sample holder cup and thin mylar film was put on the top of the powder samples. Samples were analyzed in plastic cups using mylar film as base. The standard thin foils were measured as such as received from Micro-matter, Deer Harbor, WA, US.
compounds. The effective charges calculated by different methods and their role are also discussed. 2. Experimental details 2.1. Instrumentation The chemical shifts in L shell of different chemical forms of 47Ag, and 50Sn elements were measured on a high performance WDXRF spectrometer (Model: S8 TIGER, Bruker, Germany). The spectrometer was equipped with an Rh anode X-ray tube (4 kW, 60 kV and 170 mA), automatic collimator and beam changers operated under vacuum
48Cd
85
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 2. Representation of measures spectrum of standard 50Sn foil before and after second order five-point SavitzkyGolay smoothing method.
Fig. 3. The normalised Lα X-ray emission spectra of 48Cd and 50Sn compounds.
86
47Ag,
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 4. The normalised Lβ1 X-ray emission spectra of 48Cd and 50Sn compounds.
47Ag,
Fig. 5. The normalised Lβ3,4 X-ray emission spectra of 47Ag, 48Cd and 50Sn compounds.
87
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 6. The normalised Lβ2,15 X-ray emission spectra of 48Cd and 50Sn compounds.
of L X-ray emission line was normalised by dividing with respect to their median counts of given compounds. During evaluation procedure, two parameters were needed. One was peak position and other was line-width (FWHM) of given X-ray fluorescence spectra. The peak position of L emission lines of 47Ag, 48Cd and 50Sn elements was easily available in Storm and Israel Table [24]. For each spectrum, peak position was determined by non-linear curve fitting method which was already installed in computer graphical software Origin 9. For each run of the calculations of given spectra, the software gave out new values of these parameters with minimum value of chi-square (χ2 ∼ 0.04). The peak positions were calculated at the centre point of 9/10 intensity of emission line of given compounds [25]. The chemical shift in L shell Xray fluorescence spectra of 47Ag, 48Cd and 50Sn compounds were given in Figs. 3–8 respectively.
2.2. Evaluation procedure In this paper, second order five-point Savitzky-Golay smoothing method [23] was applied twice to smooth the experimental data of L Xray emission lines of 47Ag, 48Cd and 50Sn compounds. The experimental data points of standard 50Sn foil spectra for L emission lines before and after smoothing were presented in Fig.2. It should be noted that smoothing method did not affect the data points of the given compounds. In order to find the chemical shift in L X-ray emission lines of different forms of 47Ag, 48Cd and 50Sn elements, the constant background for all spectra was chosen in such a way that least counts were counting in the spectral range. The given spectra were fitted with Voigt function (convolution of Lorentzian with Gaussian function). For conventional spectrometer, Gaussian function was dominated because of high resolution than natural line width, whereas measurement with high resolution spectrometers, Voigt function was used and defined as
y = y0 + A.
2ln2 wL . 3 2 π 2 wG
e−t
∞
∫−∞
(
ln2 wL 2 ) wG
+(
2.3. Theoretical methods to calculate effective charge
2
4ln2 X − Xc −t )2 wG
To understand the cause of chemical shift more deeply, effective charge (q) plays a very crucial role in the compound formation. In the present study, we adopt four theoretical methods to calculate the effective charge values which are explained one by one as below:
dt (2)
where ‘y0’ was the peak offset value, ‘A’ was the area of Voigt peak profile, ‘w’ was the FWHM of the peak, ‘wL’ was the width of Lorentzian profile, ‘wG’ was the width of Gaussian profile and ‘xc’ was the central peak position of Voigt curve. After Voigt fitting procedure, each spectra
2.3.1. Pauling method According to this method [26,27] effective charge (qP) is
88
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 7. The normalised Lγ1 X-ray emission spectra of 48Cd and 50Sn compounds.
q S = n[1−0.01185(z/r1 + z1/r)]
represented by
qP = nI
(5)
where, ‘n’ denotes oxidation state of cation, ‘z’ and ‘z1’ represent total number of electrons and ‘r’ and ‘r1’ are the ionic radius of both cations and anions respectively.
(3)
where, ‘n’ and ‘I’ symbol denote the valence of absorbing atom (cation) and ionicity of the metal-ligands bond of given compounds in the present work which is given by equation
I= 1−n/cexp[−(x1−x2)2 /4]
47Ag,
2.3.3. Batsanov method It has been seen that the above methods such as Suchet and Pauling method are applicable only for binary compounds. For other complicated compounds, Batsanov et al. [30] developed a new method (so called Batsanov method) for calculating effective charge (qB) on the cation and is defined as
(4)
The symbols ‘x1’ and ‘x2’ represent the electro-negativities of ligands and absorbing atom and ‘c’ denotes the coordination number of absorbing atom. In the present study, the values of electro-negativity have been taken from Pauling scale [28]. By using equations (3) and (4), the effective charges (qP) of 48Cd and 50Sn compounds are calculated.
qB = Z−nc
(6)
where, ‘Z’, ‘n’ and ‘c’ refer to oxidation state, coordination number and covalency of the central metal atom.
2.3.2. Suchet method A simple method given by Suchet [26–27,29] generally depends on the transfer of charge between cations and anions on the binary compounds formation. According to Suchet (1965), the effective charge on central metal atom (cation) is represented by
2.3.4. Levine method According to Levine method [31] when the cation denotes its charge
89
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 8. The normalised Lγ2,3 X-ray emission spectra of 47Ag, 48Cd and 50Sn compounds.
Table 2 Chemical shift in all L X-ray emission lines of Compounds
Lα
Chemical shift (ΔE in eV) Ag foil 0 AgBr 0.40 AgCl 0.28 AgI 0.12
47Ag
Table 3 Chemical shift in all L X-ray emission lines of
compounds.
Lβ1
Lβ3
Lβ4
Lγ1
Lγ2,3
0 0.36 0.47 0.27
0 1.27 2.64 1.10
0 2.92 4.89 1.79
0 0.26 1.13 −0.09
0 0.13 0.59 −2.08
Compounds
Lα
Chemical shift (ΔE in eV) Cd foil 0 Cd3(PO4)2 0.69 CdB4O7 0.42 CdCO3 0.28 CdO −0.27 CdCl2 1.06 CdI2 0.20 CdS 0.02 Sn foil 0 Sn(CrO4)2 1.81 SnO2 1.45 SnO 0.89 SnCl2 0.66 SnF2 0.51 SnF4 0.02
The chemical shift of 47Ag compounds were calculated by taking reference energy given by E. Storm and H.I. Israel theoretical table [24]. The uncertainty in the chemical shift (ΔE) for 47Ag compounds in all L shell emission lines was found to be ∼1–3 × 10−5 eV respectively.
to its neighbouring atom (anion), some part of transferring of that charge remains in the bond screens during compound formation. This bond charge arises from two sources. One is from the contribution of incomplete screening of the ion cores (q) while other comes from the overlap of spherical part of atomic scattering form factor (q0)
48Cd
and
50Sn
compounds.
Lβ1
Lβ2,15
Lβ3
Lβ4
Lγ1
Lγ2,3
0 1.15 0.82 0.80 0.50 2.06 0.87 0.72 0 0.78 0.54 0.21 0.32 0.14 0.11
0 0.38 0.23 0.21 0.12 0.79 0.39 0.36 0 1.40 1.15 1.04 0.66 0.21 0.15
0 0.43 0.20 0.05 −0.05 0.96 −0.34 −0.31 0 1.21 1.04 0.59 0.43 0.21 0.17
0 1.16 0.62 0.27 −0.62 1.79 0.73 0.18 0 2.01 0.71 0.58 1.75 0.46 0.34
0 1.27 1.13 0.71 0.62 1.35 0.35 0.10 0 1.08 0.98 0.87 0.19 0.12 0.09
0 1.75 0.72 0.59 −0.33 1.44 1.10 −0.73 0 1.40 1.09 0.30 0.65 0.18 0.17
The chemical shift of 48Cd and 50Sn compounds were calculated by taking reference energy given by E. Storm and H. I. Israel theoretical table [24]. The uncertainty in the chemical shift (ΔE) for 48Cd and 50Sn compounds for L shell emission lines was found to be ∼2–3 × 10−4 eV respectively.
90
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Table 4 Effective charges (qP, qS, qL, qB) calculated from Pauling method, Suchet method, Levine method and Batsanov method for
47Ag
compounds.
Compounds
Pauling MethodEffective charge (qP)
Suchet MethodEffective charge (qS)
Levine MethodEffective charge (qL)
Batsanov MethodEffective charge (qB)
Ag foil AgCl AgBr AgI
0 0.48 0.87 0.21
0 0.51 0.35 0.11
0 0.56 0.70 0.69
0 0.14 0.10 0.08
Table 5 Effective charges (qP, qS, qL, qB) calculated from Pauling method, Suchet method, Levine method and Batsanov method for
48Cd
compounds.
Compounds
Pauling Method Effective charge (qP)
Suchet Method Effective charge (qS)
Levine Method Effective charge (qL)
Batsanov Method Effective charge (qB)
Cd foil Cd3(PO4)2 CdS CdB4O7 CdCl2 CdCO3 CdI2 CdO
0 (1.92,1.94,1.93,1.95,1.94,1.96,1.95) * 1.18 (1.87,1.89,1.90,1.91,1.88,1.91,1.89) * 1.61 (1.84,1.85,1.86,1.87,1.85,1.86,1.85) * 1.47 1.76
0 (1.20,1.21,1.19,1.22,1.23,1.24, 1.25)* 0.17 (1.16,1.13,1.14,1.17,1.16,1.17, 1.18)* 0.95 (1.12,1.09,1.08,1.10,1.11,1.12, 1.13)* 0.87 0.96
0 (1.79,1.85,1.87,1.84,1.81,1.88,1.85) * −2.38 (1.50,1.58,1.67,1.55,1.58,1.60,1.58) * −1.11 (1.29,1.38,1.52,1.31,1.37,1.39,1.32) * −2.21 0.78
0 (1.16,1.17,1.18,1.18,1.17, 1.18, 1.19)* 0.36 (1.13,1.14,1.15,1.14,1.13,1.14, 1.15)* 0.83 (1.10,1.11,1.13,1.11,1.10,1.11,1.11) * 0.42 1.07
Effective charges (qP, qS, qL and qB) in * are calculated from the experimental curves given by above methods within uncertainty of ± (0.052, 0.073, 0.077 and 0.071) L X-ray emission lines of all 48Cd compounds.
Table 6 Effective charges (qP, qS, qL, qB) calculated from Pauling method, Suchet method, Levine method and Batsanov method for
50Sn
compounds.
Compounds
Pauling MethodEffective charge (qP)
Suchet MethodEffective charge (qS)
Levine MethodEffective charge (qL)
Batsanov MethodEffective charge (qB)
Sn foil SnF4 SnF2 Sn(CrO4)2 SnO2 SnO SnCl2
0 2.56 1.82 (2.71,2.75,2.73,2.72,2.69, 2.68,2.77)* 2.46 1.42 0.61
0 1.71 1.18 (1.80,1.81,1.81,1.82,1.81, 1.81,1.82)* 1.75 0.99 1.02
0 0.15 0.11 (1.01,1.04,1.02,1.05,1.03, 1.02, 1.01)* 0.92 0.07 0.09
0 2.56 1.28 (2.84,2.85,2.86,2.87,2.83, 2.84,2.87)* 1.69 0.84 1.21
Effective charges (qP, qS, qL and qB) in * are calculated from the experimental curves given by above methods within uncertainty of ± (0.06, 0.10, 0.12, 0.14) L X-ray emission lines of all 50Sn compounds.
more sensitive as compared to the inner energy level and it is believed that energy also depends on the charge transfer between cation and anion of the given compound. The total effective atomic charge (qP) on central metal atom is given as
contribution. The value of q0 is given by
q0 =
Nnv f c c
(7)
where, the term ‘N’ represents coordination number of cation, ‘nv’ represents number of valence electron per bond as mentioned in Levine’s method [31] ‘fc’ is the covalency of bond given by
f c = 1−fi
qP = Q−q 0
(9)
The term ‘Q’ refers to formal valence state of central metal atom and (q0) being given in (7).
(8)
The term ‘fi’ refers to the Phillips ionicity. However, in X-ray emission spectra, we have to deal with two types of energy levels. One is inner core energy level (less affected by chemical state of emitted atom), while other is outermost energy level (strongly affected by chemical environment around cation). The outer energy level is much
3. Results and discussion The chemical shift (ΔE) measured for L X-ray emission lines in different chemical forms of 47Ag, 48Cd and 50Sn elements are listed in 91
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 9. The effective charge (qP) versus chemical shift (ΔE) for
48Cd
compounds calculated by Pauling method.
electron towards the parent nucleus. This twin behaviour of ligands with respect to the central atom affects the change in peak position of all compounds. The effective charge is quite useful for better understanding the energy difference and chemical bond separation between central metal and its neighbouring atoms in different chemical forms. Earlier workers [26,33–34] developed various types of relations (linear, polynomial, exponential, quadratic and cubic) on the basis of least-squares regression analysis in different transition and non-transition compounds. Using various analytic techniques like X-ray photoemission spectroscopy (XPS) [35–36], bent crystal X-ray spectrograph (BCXAS) [37] and dispersive extended X-ray absorption fine structure (DEXAFS) [26], many researchers reported the linear dependency behaviour between chemical shift and effective charge in different binary and ternary compounds. Leonhardt and Meisel [38] proposed a free-ion model to calculate the small chemical shift in Kα emission lines of third period elements and show the dependency on the effective charge. Ishii et al. [39], examined both positive and negative effective charges on 14Si atom in Mg2Si and CaSi metal silicides using DV Xα method and correlate with its chemical shift. The calculated values of effective charges by different theoretical methods are given in Tables 4–6 respectively. In Table 4, the theoretical calculations of effective charges (qS and qB) calculated with Suchet and Batsanov method show similar sequence
Tables 2 and 3 respectively. From Table 2, it is seen that the AgCl compound show large chemical shift than AgBr and AgI compounds (AgCl > AgBr > AgI) for all L emission lines except Lα emission peak. From Table 3, it is observed that for 48Cd and 50Sn compounds, chloride compounds show large chemical shift in all L shell emission lines. For example, for 48Cd compounds (CdCl2 > CdI2 > CdS) and for 50Sn compounds (SnCl2 > SnF2 > SnF4). From Table 3, it is also observed that, chemical shift in all L shell emission lines gradually increases as the number of oxygen (ligands) atoms attached to the 48Cd atom increases. Similar behaviour of chemical shift has also been found in the compounds of 50Sn. In the past, Baydas et al. [25] and Kaur et al. [32] also studied the chemical effects in Kα and Kβ1,3 X-ray emission lines of different 19K, 20Ca and 27Co compounds and found that chemical shift is proportional to the number of ligands attached to the central metal atom. In the present work, Tables 2 and 3 clearly show both positive and negative shifts in 47Ag, 48Cd and 50Sn compounds. The positive sign of ΔE indicates the increase in L X-ray energy emitted from the given compounds. L X-ray energy is the difference in L shell and outer shell binding energies. It looks like the ligands attached to the central atom pull away the outermost valence electrons from the parent nucleus. On the other hand, negative sign of ΔE indicates the decrease in L X-ray energy emitted from the compounds. It means that ligands attached to the central atom are exerted to push the outermost valence shell
92
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 10. The effective charge (qS) versus chemical shift (ΔE) for
48Cd
compounds calculated by Suchet method.
shell vacancy is transferred to higher shell vacancy of the same shell. The process of CK transitions is radiationless and involves some special properties. (a) High intensity (b) Low transition energy (c) Energy dependence of emission rates. CK transitions energies are defined as the binding energy differences between two subshell in same shell. For 47 ≤ Z ≤ 48, the L1-L3M4,5 CK process is energetically allowed, but for Z = 50, the L1-L3M4,5 CK process is energetically forbidden. In general, the initial states of Ag+, Cd2+ and Sn4+ central metal atoms are likewise to each other and come out similar to the ground state electronic configuration of 46Pd element i.e., [Kr] 4d10. If L1-L3M4,5 CK process occurs, 2 s−1 electron vacancy is created and makes initial state of X-ray fluorescent peak to 2p−1 3d−1 4d10 (see Fig. 13) and helps Ag+, Cd2+ atoms to attract electron with its neighbouring ligands. Also, for 47Ag, 48Cd and 50Sn elements, the L1-L3N4,5 and L2-L3N4,5 CK processes are energetically allowed [40]. In present measurement, it is difficult to analyze the Lβ2,15 peak for all 47Ag compounds because the counts of Lβ2,15 peak is too much less to identify chemical shift so we neglect Lβ2,15 peak in 47Ag case. When L1-L3N4,5 CK process happens, again 2 s−1 electron vacancy is formed and makes initial state configuration 2p−1 4d9 and for L2-L3N4,5 CK processes, 2p−1 electron vacancy is created and make the initial state configuration to 2p−1 4d9 (see
in halogen compounds (AgCl > AgBr > AgI), while using Pauling method and Levine method, the effective charges (qP and qL) show different trend (AgBr > AgCl > AgI and AgBr > AgI > AgCl), while from Tables 5 and 6, it is seen that for 48Cd and 50Sn compounds, effective charge increases with decreasing electronegativity. It should be noted that the above four methods are valid only for binary compounds. To calculate the effective charge values in non-binary compounds like Cd3(PO4)2, CdB4O7, CdCO3, CdO and Sn(CrO4)2 compounds, we develop a linear relation by plotting graph between chemical shift and effective charges in different L shell X-ray emission lines which are listed in Figs. 9–12 respectively. These figures show the linear dependency behaviour of 48Cd compounds in different L shell emission lines. During the fitting procedure we observed that the estimated uncertainty in Pauling method is less than other methods. Similarly, the effective charge values for 50Sn compounds in L shell emission lines are also calculated by same procedure. The value of uncertainty obtained for 48Cd and 50Sn compounds are given in Tables 6 and 7 respectively. Thus, it clearly shows that Pauling method is the most suitable method for linearly dependency behaviour in chemical shift and effective charge in given compounds. Coster-Kronig (CK) is a special kind of Auger process in which inner
93
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 11. The effective charge (qL) versus chemical shift (ΔE) for
Fig. 13). Likewise, for Lγ1 emission peak formed in 47 ≤ Z ≤ 50, the L1L2N4 CK process is energetically allowed [40] and makes the initial state configuration 2p−1 4d9 from other subshell (see Fig. 13). The remaining Lβ1, Lβ3,4 and Lγ2,3 X-ray fluorescent peaks in 47 ≤ Z ≤ 50, the values of L1-L2M4, L1-L1M2,3 and L1-L1N2,3 CK transition are offset and dominant one. On the other side, in Sn2+ central atoms, the initial state electronic configuration is 4d10 5 s2. Due to the presence of outermost 5 s2 state, the CK transition process is quite different from the initial states configuration of the Ag+, Cd2+ and Sn4+ central atoms. Similarly, from above criteria, the initial state configuration in Lα emission line during the CK transition process becomes 2p−1 3d−1 5 s2. On the other hand, in Lβ2,15 and Lγ2,3 emission peaks, the initial state configuration from lower to higher subshell in presence of CK processes lead to 2p−1 4d−1 5 s2. In the present work, we would like to highlight the correlation between effective charges (qP, qS, qL and qB) and CK transition processes. It is believed that both these factors play a vital role in the chemical shifts of all relevant compounds. For small chemical shifts seen in L X-ray emission lines of all compounds, we observe that the effective charge and CK transition affects the flow of valence electron between cation and anion. This flow of valence electron disturbed the central peak position of the given compound which results in chemical shift.
48Cd
compounds calculated by Levine method.
To study the chemical effect in L X-ray emission lines more deeply, line-width (FWHM) and bond length are also considered as a major role. The relative line-width (ΓR in eV) and bond length (Å) for all samples are shown in Tables 1 and 7 respectively. Similar to the chemical shift (see Tables 2 and 3), the measured value of ΓR also follow same order except in Lα (L3-M4,5) line of 47Ag compounds (AgCl and AgI), Lβ2,15 (L3-N4,5) lines of 48Cd (CdO) and 50Sn (Sn(CrO4)2, SnO2 and SnCl2) compounds. This abrupt small change may occurs due to the presence of L1-L3M4,5, L1-L3N4,5 and L1-L2N4 CK processes. By using the values of ΔE and ΓR, we generated a calibration curve to show linear dependency between them. The linear relation between ΓR and ΔE for 47Ag, 48Cd and 50Sn compounds in Lβ1 X-ray emission lines are shown in Fig. 14. On the other hand, it is observed that for halogen compounds, bond length decreases with increasing electronegativity and found proportional to the complex oxides of 48Cd and 50Sn compounds. Thus it is clearly seen from the above discussion that both line-width and bond length depend on the chemical shift. 4. Conclusion The chemical shift in Lα, Lβ1, Lβ3,4, Lβ2,15, Lγ1 and Lγ2,3 X-ray emission lines of 47Ag, 48Cd and 50Sn compounds have been measured
94
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 12. The effective charge (qB) versus chemical shift (ΔE) for
Table 7 Relative FWHM (ΓR in eV) calculated for L emission lines of pounds.
47Ag, 48Cd
and
50Sn
Lα
Lβ1
Lβ2,15
Lβ3
Lβ4
Lγ1
Lγ2,3
Ag foil AgBr AgCl AgI Cd foil Cd3(PO4)2 CdB4O7 CdCO3 CdO CdCl2 CdI2 CdS Sn foil Sn(CrO4)2 SnO2 SnO SnCl2 SnF2 SnF4
1 1.04 1.03 1.03 1 1.06 1.05 1.04 1.03 1.10 1.03 1.02 1 1.38 1.28 1.16 1.03 1.01 0.98
1 1.04 1.05 1.03 1 1.11 1.10 1.06 1.04 1.38 1.20 1.07 1 1.47 1.01 0.95 0.99 0.96 0.97
– – – – 1 1.37 1.33 1.29 1.65 1.36 1.27 1.23 1 1.03 1.03 1.00 0.98 1.00 0.95
1 1.03 1.04 0.93 1 1.10 1.02 0.99 0.98 1.16 1.03 1.03 1 1.40 1.22 1.14 1.03 0.97 0.96
1 0.95 1.00 0.76 1 1.15 1.10 1.09 1.05 1.28 1.16 1.04 1 1.13 0.99 0.95 1.10 1.08 1.04
1 1.25 1.17 1.02 1 1.32 1.16 1.11 1.09 1.33 1.13 1.03 1 0.92 0.90 0.77 0.85 0.83 0.80
1 1.10 1.13 1.02 1 1.33 1.27 1.05 1.02 1.00 1.04 0.93 1 1.16 1.05 0.93 1.12 1.04 1.03
compounds calculated by Batsanov method.
with high resolution advanced S8 Tiger WDXRF spectrometer. Effective charge, bond length, line-width (FWHM) and Coster-Kronig (CK) transition processes are the responsible factors for the cause of chemical shift. From the present results, the following conclusions are obtained for the profile change of L shell emission spectra of these compounds:
com-
Compounds
48Cd
1. Halogen compounds are more sensitive for chemical states in different 47Ag, 48Cd and 50Sn compounds. 2. The chemical shift, effective charge, bond length and line-width are increased with increasing with number of oxygen atoms attached to the central metal atoms. 3. For halogen compounds, bond length is found inversely proportional to the electronegativity. It means for higher electronegativity, shorter is the bond length between two nearest neighbour atoms. In complex oxides of 48Cd and 50Sn compounds, bond length increases with increasing chemical shift. 4. Chemical shifts and effective charges show linear dependency behaviour with each other in all L X-ray emission lines for 47Ag, 48Cd and 50Sn compounds. 5. Chemical shift show both positive and negative shifts in L shell emission lines. This happens due to pull and push of outermost valence electrons from the ligands and central metal atoms. The present study also gives better view about various factors affecting
The uncertainty in the FWHM for L shell emission lines for all compounds is ∼1–1.5 × 10−5 eV respectively.
95
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 13. The schematic diagram showing (a) L1-L3M4, (b) L1-L3M5, (c) L2-L3N4, (d) L2-L3N5 and (e) L1-L2N4 CK transition processes.
Acknowledgements
the cause of chemical shift in given compounds. 6. Out of these four methods, we found that Pauling method is the most suitable method to calculate the effective charge in given compounds.
Financial support from the Department of Science and Technology (DST), New Delhi, under FIST and the University Grant Commission, New Delhi under the Centre of Advanced Study (CAS) Funds (for H.S. Kainth) is duly acknowledged. 96
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 14. The graph between relative FWHM (ΓR) versus chemical shift (eV) in 47Ag, 48Cd and 50Sn compounds in Lβ1 X-ray emission lines.
193 (2002) 877–882. [14] Y. Mokuno, Y. Horino, T. Tadic, M. Terasawa, A. Kinomura, A. Chayahara, N. Tsubouchi, WDX-PIXE analysis of low energy X-rays using a microbeam, Nucl. Instr. Meth. Phys. B 150 (1999) 109–113. [15] S. Nishibu, S. Yonezawa, M. Takashima, Chemical state analysis of fluorine in various compounds using X-ray fluorescence spectra, J. Fluor. Chem. 126 (7) (2005) 1048–1053. [16] P. Putila-Mantyla, M. Ohno, G. Graeffe, L X-ray linewidths of the elements Nb to Sb, J. Phys. B: At. Mol. Phys. 17 (1984) 1735. [17] Ö. Söğüt, Variation of enhancement effect of Coster-Kronig transition of L3 x-rays of Ba, La and Ce compounds, J. Quant. Spec. Rad. Transf. 97 (2004) 388–394. [18] J.-H. Xu, Y.-N. Xu, Electronic structures and charge transfer of nickel silicides, Solid State Commun. 55 (1985) 891–893. [19] Z. Liu, S. Sugata, K. Yuge, M. Nagasona, K. Tanaka, J. Kawai, Correlation between chemical shift of Si Kα lines and the effective charge on the Si atom and its application in the Fe-Si binary system, Phys. Rev. B 69 (3) (2004) 035106. [20] O. Bisi, L. Braicovich, C. Carbone, I. Lindau, G.L. Olcese, A. Palenzona, Chemical bond and electronic states in calcium silicides: Theory and comparison with synchrotron-radiation photoemission, Phys. Rev. B 40 (15) (1989) 10194. [21] J.J. Jia, T.T. Callcott, A. Asfaw, J.A. Carlislie, L.J. Terminello, D.L. Ederer, F.J. Himpsel, R.C.C. Perera, Occupied electronic states of CaSi 2 and CaSi: Soft-Xray-fluorescence spectroscopy, Phys. Rev. B 52 (7) (1995) 4904. [22] J. Kawai, C. Suzuki, H. Adachi, T. Konishi, Y. Gohshi, Charge-transfer effect on the linewidth of Fe Kα X-ray fluorescence spectra, Phys. Rev. B 50 (16) (1994) 11347. [23] A. Savitzky, M.J. Golay, Smoothing and differentiation of data by simplified least squares procedures, Anal. Chem. 36 (8) (1964) 1627–1639. [24] L. Storm, H.I. Israel, Photon cross sections from 1 keV to 100 MeV for elements Z = 1 to Z = 100, At. Data Nucl. Data Tables 7 (6) (1970) 565–681. [25] E. Baydaş, E. Öz, Chemical effects in the Kα and Kβ1,3 of X-ray emission spectra of Fe, J. Elect. Spectr. Rel. Phen. 185 (1) (2012) 27–31. [26] H.S. Kainth, R. Singh, Chemical shifts of L3 x-ray absorption edges on lead and thallium compounds by DEXAFS using synchrotron radiation source, Nucl. Instr. Meth. Phys. B 407 (2017) 197–202. [27] D. Joseph, S. Basu, S.N. Jha, D. Bhattacharyya, Chemical shifts of K x-ray absorption edges on copper in different compounds by x-ray absorption spectroscopy (XAS)
References [1] T. Konishi, K. Nishihagi, K. Taniguchi, Gearless two-crystal X-ray spectrometer for chemical state analysis, Rev. Sci. Instr. 62 (11) (1991) 2588–2592. [2] H. Hayashi, N. Kanai, N. Kawamura, Y.H. Matsuda, K. Kunga, S. Nakatsuji, T. Yamashita, S. Ohara, Chemical effects of high-resolution Yb Lγ4 emission spectra: a possible probe for chemical analysis, X-Ray Spectrom. 42 (6) (2013) 450–455. [3] L. Mandić, S. Fazinić, M. Jakšić, Chemical effects on the Kβ″ and Kβ2,5 X-ray lines of titanium and its compounds, Phys. Rev. A. 80 (4) (2009) 042519. [4] V.R. Murty, K. Parthasaradhi, M.V. Murti, K x-ray satellite spectrum of Ca excited by photons, X-Ray Spectrom. 27 (1) (1998) 23–25. [5] E. Öz, E. Baydaş, Y. Şahin, Chemical shifts of Kα and Kβ1,3 x-ray emission spectra for oxygen compounds of Ti, Cr, Fe Co, Cu with WDXRF, J. Radioanal. Nucl. Chem. 279 (2) (2009) 529–537. [6] S. Porikli, Chemical shift and intensity ratio values of Dyspersium, Holmium and Erbium L x-ray emission lines, Rad. Phys. Chem. 81 (2) (2012) 113–117. [7] O. Sogut, E. Büyükkasap, H. Erdoğan, Chemical-effect variation of Kβ/Kα x-ray intensity ratios in 3d elements, Rad. Phys. Chem. 64 (5) (2002) 343–348. [8] Ö. Sögüt, S. Seven, E. Baydas, E. Büyükkasap, A. Küçükönder, Chemical effects on Kβ/Kα x-ray intensity ratios of Mo, Ag, Cd, Ba, La, Ce compounds and total mass attenuation coefficients of Fe and Cu, Spectrochim. Acta B 56 (8) (2001) 1367–1374. [9] N. Kallithrakas-Kontos, X-ray chemical shift determination by energy dispersive detection, Spectrochim. Acta B 51 (13) (1996) 655–1659. [10] K. Kaur, R. Mittal, Relative K X-ray energy shifts in Fe, Cu and Zn metals and their compounds with EDXRF set-up, Radiat. Phys. Chem. 83 (2013) 28–33. [11] E. Baydaş, E. Öz, Effects of chemical combination on X-ray Kα and Kβ1,3 emission spectra of Co, Phys. Script. 81 (1) (2010) 015302. [12] M. Büyükyıldız, E. Baydaş, M. Kurudirek, E. Öz, Chemical shifts observed in Lα xray emission lines of elements in the range of 26 ≤ Z ≤ 30 in their halogen compounds, J. Radioanal. Nucl. Chem. 292 (2) (2012) 467–472. [13] K. Kawatsura, N. Takeshima, T. Imaoku, K. Takahiro, Y. Mokuno, Y. Horino, T. Sekioka, M. Terasawa, High-resolution X-ray spectroscopy for copper and copper oxides and a new WDX system using an ion microbeam, Nucl. Instr. Meth. Phys. B
97
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
[35] V.K. Kondawar, C. Mande, Chemical shifts of the K absorption discontinuity of cobalt in some binary and ternary intermetallic systems, J. Solid State Phys. 9 (7) (1976) 1351. [36] B.J. Lindberg, K. Hamrin, G. Johansson, U. Gelius, A. Fahlman, C. Nordling, K. Siegbahn, Molecular spectroscopy by means of ESCA II. Sulfur compounds. Correlation of electron binding energy with structure, Phys. Script. 1 (5–6) (1970) 286. [37] A.R. Chetal, P. Mahto, P.R. Sarode, Chemical-shift of the X-ray K-absorption edge of Co in some compounds, complexes and minerals, J. Phys. Chem. Solids 49 (3) (1988) 279–283. [38] G. Leonhardt, A. Meisel, Determination of effective atomic charges from the chemical shifts of X-ray emission lines, J. Chem. Phys. 52 (12) (1970) 6189–6198. [39] H. Ishii, S. Matsuo, P. Karimov, K. Tanaka, J. Kawai, Effective charge on silicon atom in the metal silicides Mg 2 Si and CaSi, Phys. Rev. B 71 (20) (2005) 205202. [40] M.H. Chen, B. Crasemann, K.N. Huang, M. Aoyagi, H. Mark, Theoretical L-shell Coster-Kronig energies 11 ≤ Z ≤ 103, At. Data Nucl. Data Tables 19 (2) (1977) 97–151.
with Synchrotron radiation, Nucl. Instr. Meth. Phys. B 274 (2012) 126–128. [28] L. Pauling, The Nature of the Chemical Bond, Third Edition, Cornell University Press, Ithaca, New York, 1960. [29] M.N. Ghatikar, B.D. Padalia, R.M. Nayak, Chemical shifts and effective charges in ternary and complex systems, J. Phys. C: Solid State Physi. 10 (20) (1977) 4173. [30] S.S. Batsanov, I.A. Ovsyannikova, Chemical Bonds in Semiconductors and Thermodynamics N. N. Sirota (New York: Consultants Bureau) 65 (1968). [31] B. Levine, d-Electron effects on bond susceptibilities and ionicities, Phys. Rev. B 7 (6) (1973) 2591. [32] K. Kaur, R. Mittal, Measurement of shift in K X-ray peak energies of potassium and calcium in different compounds using EDXRF, Eur. Phys. J.-App. Phys. 68 (3) (2014). [33] M.M. Ballal, C. Mande, Chemical shifts of the copper and cobalt K absorption discontinuities in some ternary compounds, J. Phys. Chem. Solids. 38 (12) (1977) 1383–1389. [34] A. Goldmann, J. Tejeda, N.J. Shevchik, M. Cardona, Density of valence states of CuCl, CuBr, CuI, and AgI, Phys. Rev. B 10 (10) (1974) 4388.
98
Contents lists available at ScienceDirect
Nuclear Inst. and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
Chemical shift in Lα, Lβ1, Lβ3,4, Lβ2,15, Lγ1 and Lγ2,3 emission lines of 48Cd and 50Sn compounds
47Ag,
MARK
⁎
Harpreet Singh Kaintha, , Ranjit Singha,b, Gurjot Singha, D. Mehtaa a b
Physics Department, Panjab University, Chandigarh 160014, India Department of Radiotherapy, PGIMER Chandigarh 160012, India
A R T I C L E I N F O
A B S T R A C T
Keywords: L emission lines Chemical effect Theoretical methods Bond length Relative line-width CK process Effective charge and WDXRF
Positive and negative shifts in L shell emission lines of 47Ag, 48Cd and 50Sn elements in different chemical compounds were determined from their recorded X-ray emission spectra in high resolution wavelength dispersive X-ray fluorescence (WDXRF) spectrometer. In 47Ag compounds, the measured energy shifts in Lα X-ray emission line were in the ranges from (0.12 to 0.40) eV, Lβ1 (0.27 to 0.36) eV, Lβ3,4 (1.10 to 4.89) eV, Lγ1 (−0.09 to 1.13) eV and Lγ2,3 (−2.08 to 0.59) eV. Likewise, for 48Cd compounds, the estimated shifts in Lα X-ray emission lines were in the range (−0.27 to 0.69) eV, Lβ1 (0.50 to 2.06) eV, Lβ2,15 (0.12 to 0.79), Lβ3,4 (−0.62 to 1.79) eV, Lγ1 (0.10 to 1.35) eV and Lγ2,3 (−0.73 to 1.75) eV, while for 50Sn compounds, the measured shifts in Lα X-ray emission lines were in the range of (0.02 to 1.81) eV, Lβ1 (0.11 to 0.78) eV, Lβ2,15 (0.15 to 1.40), Lβ3,4 (0.17 to 2.01) eV, Lγ1 (0.09 to 1.08) eV and Lγ2,3 (0.17 to 1.40) eV respectively. The effective charges (qP, qS, qL and qB) were calculated by four different theoretical methods (Pauling method, Suchet method, Levine method and Batsonav method) and found to be linear dependent with the chemical shift. Further, the measured chemical shifts were correlated with bond length, relative line-width (FWHM), effective charge, electronegativity, number of ligands and Coster-Kronig (CK) transition processes.
1. Introduction Chemical state analysis becomes more interesting topic for many researchers working in the spectroscopy field. The chemical effects can be studied by measuring either the shift in energies of characteristic Xray photons [1–2], relative intensities of X-ray photons [3] or the formation of satellite lines [4]. X-ray photoelectron spectroscopy (XPS), energy dispersive X-ray fluorescence (EDXRF) spectrometers, photoemission spectroscopy (PES) and X-ray absorption near-edge structures (XANES) are the most common employed techniques used to identify the chemical state analysis in different compounds. All these techniques are non-destructive and bulk-sensitive to the chemical states of ir-radiating atoms, which provide useful insight of electron density, quantum states and atomic and electronic structures in the materials. During the formation of compound, metal atom transforms into a positive ion, results in change in the core binding energy of the electron. This deviation in the energy of a metal atom with respect to the different chemical compounds termed as chemical shift (ΔE), and is defined as Emetal – Ecompound, where Emetal denotes central peak position of pure metal and Ecompound is the central peak position of the relevant compounds. However, having limited detector resolution and poor data
⁎
reliability, these types of spectrometers suffer with lack of stability, electrostatic effects and uncertainty in the emission curves which requires large number of complex corrections [4,6–10]. To avoid these types of problems to some extent, wavelength dispersive X-ray (WDXRF) spectrometer with different ranging of high resolution crystals are used. Earlier many authors [5,11–15] reported the chemical effects in K and L shell emission lines of different low and medium Z elements. Chemical effect combination and their relative theoretical explanation on different chemical compounds, however, have not been established completely. Various factors like CosterKronig (CK) transitions, line-width (FWHM), effective charge and bond length etc are responsible for the cause of chemical shift. In past, PutilaMantyla et al. [16] calculated the line-widths of Lα1 and Lβ1,3,4 emission lines in the atomic range 41 ≤ Z ≤ 50 and showed sudden drop of L1 line-width at 50Sn element due to the closing of L1-L3M4,5 CK process. Likewise CK transitions, effective charge also plays an important role in chemical shift. Many researchers [17–22] studied the effective charge by various methods and techniques to show their impact on the chemical shift. In the present work, our objective is to study the cause of chemical shift in L shell emission lines of different 47Ag, 48Cd and 50Sn
Corresponding author. E-mail address: [email protected] (H. Singh Kainth).
http://dx.doi.org/10.1016/j.nimb.2017.10.027 Received 14 August 2017; Received in revised form 23 October 2017; Accepted 24 October 2017 0168-583X/ © 2017 Elsevier B.V. All rights reserved.
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 1. The schematic diagram of S8 TIGER WDXRF X-ray tube setup (situated at Panjab University Chandigarh).
Table 1 Represents the oxidation state, manufacture’s name and nearest bond distance of 48Cd and 50Sn compounds. Compounds
Oxidation state
Manufacturer
Bond distance (Å)
Ag foil
0
–
AgCl AgBr AgI Cd foil
+1 +1 +1 0
CdCO3 CdB4O7 Cd3(PO4)2 CdCl2 CdI2 CdO CdS Sn foil
+2 +2 +2 +2 +2 +2 +2 0
SnF4 SnF2 SnO SnO2 Sn(CrO4)2 SnCl2
+4 +2 +2 +4 +4 +2
Micromatter, Deer Harbor, WA CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent Micromatter, Deer Harbor, WA CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent CDH Lab. Reagent Micromatter, Deer Harbor, WA Sigma Aldrich Sigma Aldrich Sigma Aldrich Sigma Aldrich Sigma Aldrich Sigma Aldrich
conditions. The schematic diagram of WDXRF apparatus was shown in Fig. 1. In WDXRF spectrometer, the high performance X-ray tube with narrow beam path and advanced analyzer crystals gave the highest intensity and fast data acquisition. The collimator masks were located between the sample and collimator to cut-off the radiation coming from the edge of the cup aperture. The space between the collimator plates restricted the maximum angle for divergence to the X-rays incident on the analyzing crystals. In the present measurement, LiF (2 0 0) high resolution analyzing crystal with 2d spacing (4.026 Å) is used. The Al filters of different thickness were used to avoid the contribution from bremsstrahlung radiations. The characteristic lines of different energies emitted from the sample were diffracted into different directions by the analyzing crystals. The diffracted characteristic lines were further detected by scintillation counter and proportional counter. For analysis procedure, SPECTRA plus software automatically corrected all matrix effects, measurement of calibration set, mathematical data processing and optimal calibration equations based on the fundamental parameters. During measurement, WDXRF spectrometer was evacuated down to ∼15 Pa and temperature was stabilized at 37 °C. The typical acquisition time for each spectrum was 20 min. Each compound was analyzed at least five times to check the reproducibility. The abscissa axis of a chart was displayed in the 2θ angle with step size of 0.046° and step time of 0.15 s. In the present study, the abscissa axis was converted into the energy (eV) with the use of Bragg’s law given by
47Ag,
2.365 2.528 2.737 — 2.023 2.027 2.125 2.404 2.769 1.813 2.035 – 1.890 1.943 1.771 1.981 2.765 2.355
nλ = 2dsinθ
(1)
where ‘λ’ is the wavelength of the X-ray, ‘d’ is the inter-atomic spacing between two layers of crystals, ‘θ’ is the angle between incident rays and surface normal to the crystal and ‘n’ is the order of diffraction pattern. The information about manufacturer, oxidation state and bond length of central metal atom in the compounds are given in Table 1. All the fine powder samples were spread uniformly on mylar film (∼2 µm thickness). The given powder samples were then placed on the top of sample holder cup and thin mylar film was put on the top of the powder samples. Samples were analyzed in plastic cups using mylar film as base. The standard thin foils were measured as such as received from Micro-matter, Deer Harbor, WA, US.
compounds. The effective charges calculated by different methods and their role are also discussed. 2. Experimental details 2.1. Instrumentation The chemical shifts in L shell of different chemical forms of 47Ag, and 50Sn elements were measured on a high performance WDXRF spectrometer (Model: S8 TIGER, Bruker, Germany). The spectrometer was equipped with an Rh anode X-ray tube (4 kW, 60 kV and 170 mA), automatic collimator and beam changers operated under vacuum
48Cd
85
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 2. Representation of measures spectrum of standard 50Sn foil before and after second order five-point SavitzkyGolay smoothing method.
Fig. 3. The normalised Lα X-ray emission spectra of 48Cd and 50Sn compounds.
86
47Ag,
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 4. The normalised Lβ1 X-ray emission spectra of 48Cd and 50Sn compounds.
47Ag,
Fig. 5. The normalised Lβ3,4 X-ray emission spectra of 47Ag, 48Cd and 50Sn compounds.
87
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 6. The normalised Lβ2,15 X-ray emission spectra of 48Cd and 50Sn compounds.
of L X-ray emission line was normalised by dividing with respect to their median counts of given compounds. During evaluation procedure, two parameters were needed. One was peak position and other was line-width (FWHM) of given X-ray fluorescence spectra. The peak position of L emission lines of 47Ag, 48Cd and 50Sn elements was easily available in Storm and Israel Table [24]. For each spectrum, peak position was determined by non-linear curve fitting method which was already installed in computer graphical software Origin 9. For each run of the calculations of given spectra, the software gave out new values of these parameters with minimum value of chi-square (χ2 ∼ 0.04). The peak positions were calculated at the centre point of 9/10 intensity of emission line of given compounds [25]. The chemical shift in L shell Xray fluorescence spectra of 47Ag, 48Cd and 50Sn compounds were given in Figs. 3–8 respectively.
2.2. Evaluation procedure In this paper, second order five-point Savitzky-Golay smoothing method [23] was applied twice to smooth the experimental data of L Xray emission lines of 47Ag, 48Cd and 50Sn compounds. The experimental data points of standard 50Sn foil spectra for L emission lines before and after smoothing were presented in Fig.2. It should be noted that smoothing method did not affect the data points of the given compounds. In order to find the chemical shift in L X-ray emission lines of different forms of 47Ag, 48Cd and 50Sn elements, the constant background for all spectra was chosen in such a way that least counts were counting in the spectral range. The given spectra were fitted with Voigt function (convolution of Lorentzian with Gaussian function). For conventional spectrometer, Gaussian function was dominated because of high resolution than natural line width, whereas measurement with high resolution spectrometers, Voigt function was used and defined as
y = y0 + A.
2ln2 wL . 3 2 π 2 wG
e−t
∞
∫−∞
(
ln2 wL 2 ) wG
+(
2.3. Theoretical methods to calculate effective charge
2
4ln2 X − Xc −t )2 wG
To understand the cause of chemical shift more deeply, effective charge (q) plays a very crucial role in the compound formation. In the present study, we adopt four theoretical methods to calculate the effective charge values which are explained one by one as below:
dt (2)
where ‘y0’ was the peak offset value, ‘A’ was the area of Voigt peak profile, ‘w’ was the FWHM of the peak, ‘wL’ was the width of Lorentzian profile, ‘wG’ was the width of Gaussian profile and ‘xc’ was the central peak position of Voigt curve. After Voigt fitting procedure, each spectra
2.3.1. Pauling method According to this method [26,27] effective charge (qP) is
88
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 7. The normalised Lγ1 X-ray emission spectra of 48Cd and 50Sn compounds.
q S = n[1−0.01185(z/r1 + z1/r)]
represented by
qP = nI
(5)
where, ‘n’ denotes oxidation state of cation, ‘z’ and ‘z1’ represent total number of electrons and ‘r’ and ‘r1’ are the ionic radius of both cations and anions respectively.
(3)
where, ‘n’ and ‘I’ symbol denote the valence of absorbing atom (cation) and ionicity of the metal-ligands bond of given compounds in the present work which is given by equation
I= 1−n/cexp[−(x1−x2)2 /4]
47Ag,
2.3.3. Batsanov method It has been seen that the above methods such as Suchet and Pauling method are applicable only for binary compounds. For other complicated compounds, Batsanov et al. [30] developed a new method (so called Batsanov method) for calculating effective charge (qB) on the cation and is defined as
(4)
The symbols ‘x1’ and ‘x2’ represent the electro-negativities of ligands and absorbing atom and ‘c’ denotes the coordination number of absorbing atom. In the present study, the values of electro-negativity have been taken from Pauling scale [28]. By using equations (3) and (4), the effective charges (qP) of 48Cd and 50Sn compounds are calculated.
qB = Z−nc
(6)
where, ‘Z’, ‘n’ and ‘c’ refer to oxidation state, coordination number and covalency of the central metal atom.
2.3.2. Suchet method A simple method given by Suchet [26–27,29] generally depends on the transfer of charge between cations and anions on the binary compounds formation. According to Suchet (1965), the effective charge on central metal atom (cation) is represented by
2.3.4. Levine method According to Levine method [31] when the cation denotes its charge
89
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 8. The normalised Lγ2,3 X-ray emission spectra of 47Ag, 48Cd and 50Sn compounds.
Table 2 Chemical shift in all L X-ray emission lines of Compounds
Lα
Chemical shift (ΔE in eV) Ag foil 0 AgBr 0.40 AgCl 0.28 AgI 0.12
47Ag
Table 3 Chemical shift in all L X-ray emission lines of
compounds.
Lβ1
Lβ3
Lβ4
Lγ1
Lγ2,3
0 0.36 0.47 0.27
0 1.27 2.64 1.10
0 2.92 4.89 1.79
0 0.26 1.13 −0.09
0 0.13 0.59 −2.08
Compounds
Lα
Chemical shift (ΔE in eV) Cd foil 0 Cd3(PO4)2 0.69 CdB4O7 0.42 CdCO3 0.28 CdO −0.27 CdCl2 1.06 CdI2 0.20 CdS 0.02 Sn foil 0 Sn(CrO4)2 1.81 SnO2 1.45 SnO 0.89 SnCl2 0.66 SnF2 0.51 SnF4 0.02
The chemical shift of 47Ag compounds were calculated by taking reference energy given by E. Storm and H.I. Israel theoretical table [24]. The uncertainty in the chemical shift (ΔE) for 47Ag compounds in all L shell emission lines was found to be ∼1–3 × 10−5 eV respectively.
to its neighbouring atom (anion), some part of transferring of that charge remains in the bond screens during compound formation. This bond charge arises from two sources. One is from the contribution of incomplete screening of the ion cores (q) while other comes from the overlap of spherical part of atomic scattering form factor (q0)
48Cd
and
50Sn
compounds.
Lβ1
Lβ2,15
Lβ3
Lβ4
Lγ1
Lγ2,3
0 1.15 0.82 0.80 0.50 2.06 0.87 0.72 0 0.78 0.54 0.21 0.32 0.14 0.11
0 0.38 0.23 0.21 0.12 0.79 0.39 0.36 0 1.40 1.15 1.04 0.66 0.21 0.15
0 0.43 0.20 0.05 −0.05 0.96 −0.34 −0.31 0 1.21 1.04 0.59 0.43 0.21 0.17
0 1.16 0.62 0.27 −0.62 1.79 0.73 0.18 0 2.01 0.71 0.58 1.75 0.46 0.34
0 1.27 1.13 0.71 0.62 1.35 0.35 0.10 0 1.08 0.98 0.87 0.19 0.12 0.09
0 1.75 0.72 0.59 −0.33 1.44 1.10 −0.73 0 1.40 1.09 0.30 0.65 0.18 0.17
The chemical shift of 48Cd and 50Sn compounds were calculated by taking reference energy given by E. Storm and H. I. Israel theoretical table [24]. The uncertainty in the chemical shift (ΔE) for 48Cd and 50Sn compounds for L shell emission lines was found to be ∼2–3 × 10−4 eV respectively.
90
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Table 4 Effective charges (qP, qS, qL, qB) calculated from Pauling method, Suchet method, Levine method and Batsanov method for
47Ag
compounds.
Compounds
Pauling MethodEffective charge (qP)
Suchet MethodEffective charge (qS)
Levine MethodEffective charge (qL)
Batsanov MethodEffective charge (qB)
Ag foil AgCl AgBr AgI
0 0.48 0.87 0.21
0 0.51 0.35 0.11
0 0.56 0.70 0.69
0 0.14 0.10 0.08
Table 5 Effective charges (qP, qS, qL, qB) calculated from Pauling method, Suchet method, Levine method and Batsanov method for
48Cd
compounds.
Compounds
Pauling Method Effective charge (qP)
Suchet Method Effective charge (qS)
Levine Method Effective charge (qL)
Batsanov Method Effective charge (qB)
Cd foil Cd3(PO4)2 CdS CdB4O7 CdCl2 CdCO3 CdI2 CdO
0 (1.92,1.94,1.93,1.95,1.94,1.96,1.95) * 1.18 (1.87,1.89,1.90,1.91,1.88,1.91,1.89) * 1.61 (1.84,1.85,1.86,1.87,1.85,1.86,1.85) * 1.47 1.76
0 (1.20,1.21,1.19,1.22,1.23,1.24, 1.25)* 0.17 (1.16,1.13,1.14,1.17,1.16,1.17, 1.18)* 0.95 (1.12,1.09,1.08,1.10,1.11,1.12, 1.13)* 0.87 0.96
0 (1.79,1.85,1.87,1.84,1.81,1.88,1.85) * −2.38 (1.50,1.58,1.67,1.55,1.58,1.60,1.58) * −1.11 (1.29,1.38,1.52,1.31,1.37,1.39,1.32) * −2.21 0.78
0 (1.16,1.17,1.18,1.18,1.17, 1.18, 1.19)* 0.36 (1.13,1.14,1.15,1.14,1.13,1.14, 1.15)* 0.83 (1.10,1.11,1.13,1.11,1.10,1.11,1.11) * 0.42 1.07
Effective charges (qP, qS, qL and qB) in * are calculated from the experimental curves given by above methods within uncertainty of ± (0.052, 0.073, 0.077 and 0.071) L X-ray emission lines of all 48Cd compounds.
Table 6 Effective charges (qP, qS, qL, qB) calculated from Pauling method, Suchet method, Levine method and Batsanov method for
50Sn
compounds.
Compounds
Pauling MethodEffective charge (qP)
Suchet MethodEffective charge (qS)
Levine MethodEffective charge (qL)
Batsanov MethodEffective charge (qB)
Sn foil SnF4 SnF2 Sn(CrO4)2 SnO2 SnO SnCl2
0 2.56 1.82 (2.71,2.75,2.73,2.72,2.69, 2.68,2.77)* 2.46 1.42 0.61
0 1.71 1.18 (1.80,1.81,1.81,1.82,1.81, 1.81,1.82)* 1.75 0.99 1.02
0 0.15 0.11 (1.01,1.04,1.02,1.05,1.03, 1.02, 1.01)* 0.92 0.07 0.09
0 2.56 1.28 (2.84,2.85,2.86,2.87,2.83, 2.84,2.87)* 1.69 0.84 1.21
Effective charges (qP, qS, qL and qB) in * are calculated from the experimental curves given by above methods within uncertainty of ± (0.06, 0.10, 0.12, 0.14) L X-ray emission lines of all 50Sn compounds.
more sensitive as compared to the inner energy level and it is believed that energy also depends on the charge transfer between cation and anion of the given compound. The total effective atomic charge (qP) on central metal atom is given as
contribution. The value of q0 is given by
q0 =
Nnv f c c
(7)
where, the term ‘N’ represents coordination number of cation, ‘nv’ represents number of valence electron per bond as mentioned in Levine’s method [31] ‘fc’ is the covalency of bond given by
f c = 1−fi
qP = Q−q 0
(9)
The term ‘Q’ refers to formal valence state of central metal atom and (q0) being given in (7).
(8)
The term ‘fi’ refers to the Phillips ionicity. However, in X-ray emission spectra, we have to deal with two types of energy levels. One is inner core energy level (less affected by chemical state of emitted atom), while other is outermost energy level (strongly affected by chemical environment around cation). The outer energy level is much
3. Results and discussion The chemical shift (ΔE) measured for L X-ray emission lines in different chemical forms of 47Ag, 48Cd and 50Sn elements are listed in 91
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 9. The effective charge (qP) versus chemical shift (ΔE) for
48Cd
compounds calculated by Pauling method.
electron towards the parent nucleus. This twin behaviour of ligands with respect to the central atom affects the change in peak position of all compounds. The effective charge is quite useful for better understanding the energy difference and chemical bond separation between central metal and its neighbouring atoms in different chemical forms. Earlier workers [26,33–34] developed various types of relations (linear, polynomial, exponential, quadratic and cubic) on the basis of least-squares regression analysis in different transition and non-transition compounds. Using various analytic techniques like X-ray photoemission spectroscopy (XPS) [35–36], bent crystal X-ray spectrograph (BCXAS) [37] and dispersive extended X-ray absorption fine structure (DEXAFS) [26], many researchers reported the linear dependency behaviour between chemical shift and effective charge in different binary and ternary compounds. Leonhardt and Meisel [38] proposed a free-ion model to calculate the small chemical shift in Kα emission lines of third period elements and show the dependency on the effective charge. Ishii et al. [39], examined both positive and negative effective charges on 14Si atom in Mg2Si and CaSi metal silicides using DV Xα method and correlate with its chemical shift. The calculated values of effective charges by different theoretical methods are given in Tables 4–6 respectively. In Table 4, the theoretical calculations of effective charges (qS and qB) calculated with Suchet and Batsanov method show similar sequence
Tables 2 and 3 respectively. From Table 2, it is seen that the AgCl compound show large chemical shift than AgBr and AgI compounds (AgCl > AgBr > AgI) for all L emission lines except Lα emission peak. From Table 3, it is observed that for 48Cd and 50Sn compounds, chloride compounds show large chemical shift in all L shell emission lines. For example, for 48Cd compounds (CdCl2 > CdI2 > CdS) and for 50Sn compounds (SnCl2 > SnF2 > SnF4). From Table 3, it is also observed that, chemical shift in all L shell emission lines gradually increases as the number of oxygen (ligands) atoms attached to the 48Cd atom increases. Similar behaviour of chemical shift has also been found in the compounds of 50Sn. In the past, Baydas et al. [25] and Kaur et al. [32] also studied the chemical effects in Kα and Kβ1,3 X-ray emission lines of different 19K, 20Ca and 27Co compounds and found that chemical shift is proportional to the number of ligands attached to the central metal atom. In the present work, Tables 2 and 3 clearly show both positive and negative shifts in 47Ag, 48Cd and 50Sn compounds. The positive sign of ΔE indicates the increase in L X-ray energy emitted from the given compounds. L X-ray energy is the difference in L shell and outer shell binding energies. It looks like the ligands attached to the central atom pull away the outermost valence electrons from the parent nucleus. On the other hand, negative sign of ΔE indicates the decrease in L X-ray energy emitted from the compounds. It means that ligands attached to the central atom are exerted to push the outermost valence shell
92
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 10. The effective charge (qS) versus chemical shift (ΔE) for
48Cd
compounds calculated by Suchet method.
shell vacancy is transferred to higher shell vacancy of the same shell. The process of CK transitions is radiationless and involves some special properties. (a) High intensity (b) Low transition energy (c) Energy dependence of emission rates. CK transitions energies are defined as the binding energy differences between two subshell in same shell. For 47 ≤ Z ≤ 48, the L1-L3M4,5 CK process is energetically allowed, but for Z = 50, the L1-L3M4,5 CK process is energetically forbidden. In general, the initial states of Ag+, Cd2+ and Sn4+ central metal atoms are likewise to each other and come out similar to the ground state electronic configuration of 46Pd element i.e., [Kr] 4d10. If L1-L3M4,5 CK process occurs, 2 s−1 electron vacancy is created and makes initial state of X-ray fluorescent peak to 2p−1 3d−1 4d10 (see Fig. 13) and helps Ag+, Cd2+ atoms to attract electron with its neighbouring ligands. Also, for 47Ag, 48Cd and 50Sn elements, the L1-L3N4,5 and L2-L3N4,5 CK processes are energetically allowed [40]. In present measurement, it is difficult to analyze the Lβ2,15 peak for all 47Ag compounds because the counts of Lβ2,15 peak is too much less to identify chemical shift so we neglect Lβ2,15 peak in 47Ag case. When L1-L3N4,5 CK process happens, again 2 s−1 electron vacancy is formed and makes initial state configuration 2p−1 4d9 and for L2-L3N4,5 CK processes, 2p−1 electron vacancy is created and make the initial state configuration to 2p−1 4d9 (see
in halogen compounds (AgCl > AgBr > AgI), while using Pauling method and Levine method, the effective charges (qP and qL) show different trend (AgBr > AgCl > AgI and AgBr > AgI > AgCl), while from Tables 5 and 6, it is seen that for 48Cd and 50Sn compounds, effective charge increases with decreasing electronegativity. It should be noted that the above four methods are valid only for binary compounds. To calculate the effective charge values in non-binary compounds like Cd3(PO4)2, CdB4O7, CdCO3, CdO and Sn(CrO4)2 compounds, we develop a linear relation by plotting graph between chemical shift and effective charges in different L shell X-ray emission lines which are listed in Figs. 9–12 respectively. These figures show the linear dependency behaviour of 48Cd compounds in different L shell emission lines. During the fitting procedure we observed that the estimated uncertainty in Pauling method is less than other methods. Similarly, the effective charge values for 50Sn compounds in L shell emission lines are also calculated by same procedure. The value of uncertainty obtained for 48Cd and 50Sn compounds are given in Tables 6 and 7 respectively. Thus, it clearly shows that Pauling method is the most suitable method for linearly dependency behaviour in chemical shift and effective charge in given compounds. Coster-Kronig (CK) is a special kind of Auger process in which inner
93
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 11. The effective charge (qL) versus chemical shift (ΔE) for
Fig. 13). Likewise, for Lγ1 emission peak formed in 47 ≤ Z ≤ 50, the L1L2N4 CK process is energetically allowed [40] and makes the initial state configuration 2p−1 4d9 from other subshell (see Fig. 13). The remaining Lβ1, Lβ3,4 and Lγ2,3 X-ray fluorescent peaks in 47 ≤ Z ≤ 50, the values of L1-L2M4, L1-L1M2,3 and L1-L1N2,3 CK transition are offset and dominant one. On the other side, in Sn2+ central atoms, the initial state electronic configuration is 4d10 5 s2. Due to the presence of outermost 5 s2 state, the CK transition process is quite different from the initial states configuration of the Ag+, Cd2+ and Sn4+ central atoms. Similarly, from above criteria, the initial state configuration in Lα emission line during the CK transition process becomes 2p−1 3d−1 5 s2. On the other hand, in Lβ2,15 and Lγ2,3 emission peaks, the initial state configuration from lower to higher subshell in presence of CK processes lead to 2p−1 4d−1 5 s2. In the present work, we would like to highlight the correlation between effective charges (qP, qS, qL and qB) and CK transition processes. It is believed that both these factors play a vital role in the chemical shifts of all relevant compounds. For small chemical shifts seen in L X-ray emission lines of all compounds, we observe that the effective charge and CK transition affects the flow of valence electron between cation and anion. This flow of valence electron disturbed the central peak position of the given compound which results in chemical shift.
48Cd
compounds calculated by Levine method.
To study the chemical effect in L X-ray emission lines more deeply, line-width (FWHM) and bond length are also considered as a major role. The relative line-width (ΓR in eV) and bond length (Å) for all samples are shown in Tables 1 and 7 respectively. Similar to the chemical shift (see Tables 2 and 3), the measured value of ΓR also follow same order except in Lα (L3-M4,5) line of 47Ag compounds (AgCl and AgI), Lβ2,15 (L3-N4,5) lines of 48Cd (CdO) and 50Sn (Sn(CrO4)2, SnO2 and SnCl2) compounds. This abrupt small change may occurs due to the presence of L1-L3M4,5, L1-L3N4,5 and L1-L2N4 CK processes. By using the values of ΔE and ΓR, we generated a calibration curve to show linear dependency between them. The linear relation between ΓR and ΔE for 47Ag, 48Cd and 50Sn compounds in Lβ1 X-ray emission lines are shown in Fig. 14. On the other hand, it is observed that for halogen compounds, bond length decreases with increasing electronegativity and found proportional to the complex oxides of 48Cd and 50Sn compounds. Thus it is clearly seen from the above discussion that both line-width and bond length depend on the chemical shift. 4. Conclusion The chemical shift in Lα, Lβ1, Lβ3,4, Lβ2,15, Lγ1 and Lγ2,3 X-ray emission lines of 47Ag, 48Cd and 50Sn compounds have been measured
94
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 12. The effective charge (qB) versus chemical shift (ΔE) for
Table 7 Relative FWHM (ΓR in eV) calculated for L emission lines of pounds.
47Ag, 48Cd
and
50Sn
Lα
Lβ1
Lβ2,15
Lβ3
Lβ4
Lγ1
Lγ2,3
Ag foil AgBr AgCl AgI Cd foil Cd3(PO4)2 CdB4O7 CdCO3 CdO CdCl2 CdI2 CdS Sn foil Sn(CrO4)2 SnO2 SnO SnCl2 SnF2 SnF4
1 1.04 1.03 1.03 1 1.06 1.05 1.04 1.03 1.10 1.03 1.02 1 1.38 1.28 1.16 1.03 1.01 0.98
1 1.04 1.05 1.03 1 1.11 1.10 1.06 1.04 1.38 1.20 1.07 1 1.47 1.01 0.95 0.99 0.96 0.97
– – – – 1 1.37 1.33 1.29 1.65 1.36 1.27 1.23 1 1.03 1.03 1.00 0.98 1.00 0.95
1 1.03 1.04 0.93 1 1.10 1.02 0.99 0.98 1.16 1.03 1.03 1 1.40 1.22 1.14 1.03 0.97 0.96
1 0.95 1.00 0.76 1 1.15 1.10 1.09 1.05 1.28 1.16 1.04 1 1.13 0.99 0.95 1.10 1.08 1.04
1 1.25 1.17 1.02 1 1.32 1.16 1.11 1.09 1.33 1.13 1.03 1 0.92 0.90 0.77 0.85 0.83 0.80
1 1.10 1.13 1.02 1 1.33 1.27 1.05 1.02 1.00 1.04 0.93 1 1.16 1.05 0.93 1.12 1.04 1.03
compounds calculated by Batsanov method.
with high resolution advanced S8 Tiger WDXRF spectrometer. Effective charge, bond length, line-width (FWHM) and Coster-Kronig (CK) transition processes are the responsible factors for the cause of chemical shift. From the present results, the following conclusions are obtained for the profile change of L shell emission spectra of these compounds:
com-
Compounds
48Cd
1. Halogen compounds are more sensitive for chemical states in different 47Ag, 48Cd and 50Sn compounds. 2. The chemical shift, effective charge, bond length and line-width are increased with increasing with number of oxygen atoms attached to the central metal atoms. 3. For halogen compounds, bond length is found inversely proportional to the electronegativity. It means for higher electronegativity, shorter is the bond length between two nearest neighbour atoms. In complex oxides of 48Cd and 50Sn compounds, bond length increases with increasing chemical shift. 4. Chemical shifts and effective charges show linear dependency behaviour with each other in all L X-ray emission lines for 47Ag, 48Cd and 50Sn compounds. 5. Chemical shift show both positive and negative shifts in L shell emission lines. This happens due to pull and push of outermost valence electrons from the ligands and central metal atoms. The present study also gives better view about various factors affecting
The uncertainty in the FWHM for L shell emission lines for all compounds is ∼1–1.5 × 10−5 eV respectively.
95
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 13. The schematic diagram showing (a) L1-L3M4, (b) L1-L3M5, (c) L2-L3N4, (d) L2-L3N5 and (e) L1-L2N4 CK transition processes.
Acknowledgements
the cause of chemical shift in given compounds. 6. Out of these four methods, we found that Pauling method is the most suitable method to calculate the effective charge in given compounds.
Financial support from the Department of Science and Technology (DST), New Delhi, under FIST and the University Grant Commission, New Delhi under the Centre of Advanced Study (CAS) Funds (for H.S. Kainth) is duly acknowledged. 96
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
Fig. 14. The graph between relative FWHM (ΓR) versus chemical shift (eV) in 47Ag, 48Cd and 50Sn compounds in Lβ1 X-ray emission lines.
193 (2002) 877–882. [14] Y. Mokuno, Y. Horino, T. Tadic, M. Terasawa, A. Kinomura, A. Chayahara, N. Tsubouchi, WDX-PIXE analysis of low energy X-rays using a microbeam, Nucl. Instr. Meth. Phys. B 150 (1999) 109–113. [15] S. Nishibu, S. Yonezawa, M. Takashima, Chemical state analysis of fluorine in various compounds using X-ray fluorescence spectra, J. Fluor. Chem. 126 (7) (2005) 1048–1053. [16] P. Putila-Mantyla, M. Ohno, G. Graeffe, L X-ray linewidths of the elements Nb to Sb, J. Phys. B: At. Mol. Phys. 17 (1984) 1735. [17] Ö. Söğüt, Variation of enhancement effect of Coster-Kronig transition of L3 x-rays of Ba, La and Ce compounds, J. Quant. Spec. Rad. Transf. 97 (2004) 388–394. [18] J.-H. Xu, Y.-N. Xu, Electronic structures and charge transfer of nickel silicides, Solid State Commun. 55 (1985) 891–893. [19] Z. Liu, S. Sugata, K. Yuge, M. Nagasona, K. Tanaka, J. Kawai, Correlation between chemical shift of Si Kα lines and the effective charge on the Si atom and its application in the Fe-Si binary system, Phys. Rev. B 69 (3) (2004) 035106. [20] O. Bisi, L. Braicovich, C. Carbone, I. Lindau, G.L. Olcese, A. Palenzona, Chemical bond and electronic states in calcium silicides: Theory and comparison with synchrotron-radiation photoemission, Phys. Rev. B 40 (15) (1989) 10194. [21] J.J. Jia, T.T. Callcott, A. Asfaw, J.A. Carlislie, L.J. Terminello, D.L. Ederer, F.J. Himpsel, R.C.C. Perera, Occupied electronic states of CaSi 2 and CaSi: Soft-Xray-fluorescence spectroscopy, Phys. Rev. B 52 (7) (1995) 4904. [22] J. Kawai, C. Suzuki, H. Adachi, T. Konishi, Y. Gohshi, Charge-transfer effect on the linewidth of Fe Kα X-ray fluorescence spectra, Phys. Rev. B 50 (16) (1994) 11347. [23] A. Savitzky, M.J. Golay, Smoothing and differentiation of data by simplified least squares procedures, Anal. Chem. 36 (8) (1964) 1627–1639. [24] L. Storm, H.I. Israel, Photon cross sections from 1 keV to 100 MeV for elements Z = 1 to Z = 100, At. Data Nucl. Data Tables 7 (6) (1970) 565–681. [25] E. Baydaş, E. Öz, Chemical effects in the Kα and Kβ1,3 of X-ray emission spectra of Fe, J. Elect. Spectr. Rel. Phen. 185 (1) (2012) 27–31. [26] H.S. Kainth, R. Singh, Chemical shifts of L3 x-ray absorption edges on lead and thallium compounds by DEXAFS using synchrotron radiation source, Nucl. Instr. Meth. Phys. B 407 (2017) 197–202. [27] D. Joseph, S. Basu, S.N. Jha, D. Bhattacharyya, Chemical shifts of K x-ray absorption edges on copper in different compounds by x-ray absorption spectroscopy (XAS)
References [1] T. Konishi, K. Nishihagi, K. Taniguchi, Gearless two-crystal X-ray spectrometer for chemical state analysis, Rev. Sci. Instr. 62 (11) (1991) 2588–2592. [2] H. Hayashi, N. Kanai, N. Kawamura, Y.H. Matsuda, K. Kunga, S. Nakatsuji, T. Yamashita, S. Ohara, Chemical effects of high-resolution Yb Lγ4 emission spectra: a possible probe for chemical analysis, X-Ray Spectrom. 42 (6) (2013) 450–455. [3] L. Mandić, S. Fazinić, M. Jakšić, Chemical effects on the Kβ″ and Kβ2,5 X-ray lines of titanium and its compounds, Phys. Rev. A. 80 (4) (2009) 042519. [4] V.R. Murty, K. Parthasaradhi, M.V. Murti, K x-ray satellite spectrum of Ca excited by photons, X-Ray Spectrom. 27 (1) (1998) 23–25. [5] E. Öz, E. Baydaş, Y. Şahin, Chemical shifts of Kα and Kβ1,3 x-ray emission spectra for oxygen compounds of Ti, Cr, Fe Co, Cu with WDXRF, J. Radioanal. Nucl. Chem. 279 (2) (2009) 529–537. [6] S. Porikli, Chemical shift and intensity ratio values of Dyspersium, Holmium and Erbium L x-ray emission lines, Rad. Phys. Chem. 81 (2) (2012) 113–117. [7] O. Sogut, E. Büyükkasap, H. Erdoğan, Chemical-effect variation of Kβ/Kα x-ray intensity ratios in 3d elements, Rad. Phys. Chem. 64 (5) (2002) 343–348. [8] Ö. Sögüt, S. Seven, E. Baydas, E. Büyükkasap, A. Küçükönder, Chemical effects on Kβ/Kα x-ray intensity ratios of Mo, Ag, Cd, Ba, La, Ce compounds and total mass attenuation coefficients of Fe and Cu, Spectrochim. Acta B 56 (8) (2001) 1367–1374. [9] N. Kallithrakas-Kontos, X-ray chemical shift determination by energy dispersive detection, Spectrochim. Acta B 51 (13) (1996) 655–1659. [10] K. Kaur, R. Mittal, Relative K X-ray energy shifts in Fe, Cu and Zn metals and their compounds with EDXRF set-up, Radiat. Phys. Chem. 83 (2013) 28–33. [11] E. Baydaş, E. Öz, Effects of chemical combination on X-ray Kα and Kβ1,3 emission spectra of Co, Phys. Script. 81 (1) (2010) 015302. [12] M. Büyükyıldız, E. Baydaş, M. Kurudirek, E. Öz, Chemical shifts observed in Lα xray emission lines of elements in the range of 26 ≤ Z ≤ 30 in their halogen compounds, J. Radioanal. Nucl. Chem. 292 (2) (2012) 467–472. [13] K. Kawatsura, N. Takeshima, T. Imaoku, K. Takahiro, Y. Mokuno, Y. Horino, T. Sekioka, M. Terasawa, High-resolution X-ray spectroscopy for copper and copper oxides and a new WDX system using an ion microbeam, Nucl. Instr. Meth. Phys. B
97
Nuclear Inst. and Methods in Physics Research B 414 (2018) 84–98
H. Singh Kainth et al.
[35] V.K. Kondawar, C. Mande, Chemical shifts of the K absorption discontinuity of cobalt in some binary and ternary intermetallic systems, J. Solid State Phys. 9 (7) (1976) 1351. [36] B.J. Lindberg, K. Hamrin, G. Johansson, U. Gelius, A. Fahlman, C. Nordling, K. Siegbahn, Molecular spectroscopy by means of ESCA II. Sulfur compounds. Correlation of electron binding energy with structure, Phys. Script. 1 (5–6) (1970) 286. [37] A.R. Chetal, P. Mahto, P.R. Sarode, Chemical-shift of the X-ray K-absorption edge of Co in some compounds, complexes and minerals, J. Phys. Chem. Solids 49 (3) (1988) 279–283. [38] G. Leonhardt, A. Meisel, Determination of effective atomic charges from the chemical shifts of X-ray emission lines, J. Chem. Phys. 52 (12) (1970) 6189–6198. [39] H. Ishii, S. Matsuo, P. Karimov, K. Tanaka, J. Kawai, Effective charge on silicon atom in the metal silicides Mg 2 Si and CaSi, Phys. Rev. B 71 (20) (2005) 205202. [40] M.H. Chen, B. Crasemann, K.N. Huang, M. Aoyagi, H. Mark, Theoretical L-shell Coster-Kronig energies 11 ≤ Z ≤ 103, At. Data Nucl. Data Tables 19 (2) (1977) 97–151.
with Synchrotron radiation, Nucl. Instr. Meth. Phys. B 274 (2012) 126–128. [28] L. Pauling, The Nature of the Chemical Bond, Third Edition, Cornell University Press, Ithaca, New York, 1960. [29] M.N. Ghatikar, B.D. Padalia, R.M. Nayak, Chemical shifts and effective charges in ternary and complex systems, J. Phys. C: Solid State Physi. 10 (20) (1977) 4173. [30] S.S. Batsanov, I.A. Ovsyannikova, Chemical Bonds in Semiconductors and Thermodynamics N. N. Sirota (New York: Consultants Bureau) 65 (1968). [31] B. Levine, d-Electron effects on bond susceptibilities and ionicities, Phys. Rev. B 7 (6) (1973) 2591. [32] K. Kaur, R. Mittal, Measurement of shift in K X-ray peak energies of potassium and calcium in different compounds using EDXRF, Eur. Phys. J.-App. Phys. 68 (3) (2014). [33] M.M. Ballal, C. Mande, Chemical shifts of the copper and cobalt K absorption discontinuities in some ternary compounds, J. Phys. Chem. Solids. 38 (12) (1977) 1383–1389. [34] A. Goldmann, J. Tejeda, N.J. Shevchik, M. Cardona, Density of valence states of CuCl, CuBr, CuI, and AgI, Phys. Rev. B 10 (10) (1974) 4388.
98