RELATIONSHIP BETWEEN THE FORMATION OF INTERMETALLIC COMPOUNDS BY MATRIX MODIFIERS AND ATOMIZATION IN GRAPHITE-FURNACE ATOMIC-ABSORPTION SPECTROMETRY, AND AN OBSERVATION OF THE VAPORIZATION OF INTERMETALLIC COMPOUNDS BY MEANS OF ELECTRON-MICROSCOPY
K. Yasuda et al., RELATIONSHIP BETWEEN THE FORMATION OF INTERMETALLIC COMPOUNDS BY MATRIX MODIFIERS AND ATOMIZATION IN GRAPHITE-FURNACE ATOMIC-ABSORPTION SPECTROMETRY, AND AN OBSERVATION OF THE VAPORIZATION OF INTERMETALLIC COMPOUNDS BY MEANS OF ELECTRON-MICROSCOPY, Analytical sciences, 10(4), 1994, pp. 623-631
It was found, only in cases of palladium-tin and platinum-lead interme
tallic compounds, that the atomization shifted to a higher temperature
region when analytes formed an intermetallic compound with a matrix m
odifier, and its activity decreased. Due to a smaller amount of data f
or the activity coefficient, it is difficult to confirm the above-ment
ioned phenomena. In this study, tin, indium and lead were chosen as an
alytes, and palladium, nickel and manganese were used as matrix modifi
ers. On the other hand, the atomic-vapor temperature was measured by t
he two-absorption-line method. This temperature is close to the temper
ature of the sample loaded into the graphite furnace. Unfortunately, t
he setting temperature, or the measured temperature of the graphite fu
rnace using an optical pyrometer, did not express the true temperature
of the sample. By measuring the temperature of the alloy, we were abl
e to verify that atomization begins in certain phases of the intermeta
llic compounds, and finished in specific phases. The phases of the int
ermetallic compounds were then confirmed by measuring the atomic-vapor
temperature. Since there has not been much data concerning the activi
ty coefficient being measured until now, the pressure of the atomic va
por evaporated from the intermetallic compound was compared with the p
ressure from pure metal at the beginning of atomization. We also estim
ated whether the activity coefficient was bigger or smaller than 1.0.
In other words, the absorbance profiles of the vaporization of the int
ermetallic compound and pure metal were compared with each other; we c
ould thus estimate the activity coefficient. Regarding Sn, the beginni
ng of atomization shifted to a higher temperature region in the order
Mn, Ni and Pd. This order is the same as that of the decreasing activi
ty coefficient. In the cases of In and Pb, the beginning of atomizatio
n of Pd-In and Pd-Pb only shifted to a higher temperature. The other i
ntermetallic compounds did not shift, because the compounds melted in
the early stage of atomization. The atomization of the intermetallic c
ompound, whose activity coefficient is smaller than 1.0, was observed
by means of high-resolution transmission-electron microscopy. As a typ
ical example, an intermetallic compound of Pd and Sn was chosen. Vapor
ization was observed in the phase of Pd3Sn2 beginning with atomization
. The atoms in the top surface were tightened until the temperature of
the intermetalic compound increased and almost reached the melting po
int. After the atoms had vibrated and swayed for a while, about 10 ato
mic layers evaporated explosively. On the other hand, the atoms gradua
lly evaporated from the top surface in pure Sn and Pd metals. These ph
enomena indicate a big difference between the intermetallic compound a
nd the pure metal.