Electron probe microanalysis of Ni-silicides at low voltage: difficulties and possibilities

E Heikinheimo,S Louhenkilpi,X Llovet,S Richter,P.T Pinard

doi:10.1088/1757-899x/109/1/012005

Abstract

Interest in the use of EPMA at low voltage has grown considerably in recent years, mainly because of the availability of electron-beam instruments equipped with field-emission guns. However, EPMA at low voltage is marred by both experimental and analytical problems which may affect the accuracy of quantitative results. In the case of the analysis of transition elements, both the emission and absorption of X-rays are still poorly understood when they originate from electron transitions involving the partially filled 3d-shell. This is the case for the most intense Lα (L3-M5 transition) and Lβ (L2-M4 transition) lines. In this communication, we point out anomalies which appear to afflict the accuracy of EPMA of Ni-silicides using the Ni-Lα X-ray line and we discuss possible solutions.

Highlights

The incorporation of the field-emission (FE) gun into electron-beam instruments has stimulated the use of these instruments at low voltage and has opened up new opportunities for the characterisation of complex materials at a sub-micrometre scale
electron probe microanalysis (EPMA) at low voltage is marred by both experimental and analytical problems which may affect the accuracy of quantitative results
Systematic EPMA measurements on the Ni-Si system have been performed in order to illustrate the difficulties of using the Ni-Lα line for the quantitative analysis of Ni-silicides at low beam energy

Summary

Introduction

The incorporation of the field-emission (FE) gun into electron-beam instruments has stimulated the use of these instruments at low voltage and has opened up new opportunities for the characterisation of complex materials at a sub-micrometre scale. Conventional EPMA measurements using the Ni-Lα and Si-Kα lines, pure Ni and Si metals standards, and the matrix correction procedure XPP, as implemented in the JEOL microprobe software, were performed at 6 keV. 3. Calculations k-ratios for the different X-ray lines and studied compounds were calculated using the Monte Carlo (MC) simulation programmes PENEPMA [13] and MONACO [14], and the analytical model XPP [11], as implemented in the software DTSA-II [15]. Using the libraries and databases within DTSA-II, an in-house programme was developed to derive empirical MACs from the EPMA measurements of the studied materials This programme follows the method described by Pouchou and Pichoir [21] and is implemented in the software package XMAC [2], which was used for comparison purposes. The programme yields the MAC values which give the best fit between the experimental and calculated X-ray intensities

Results and discussion

Findings

Conclusions