accuracy of composition measurements (see ).
This allowed the use of the Cliff-Lorimer formula
for calculating the Si/Al weight concentration ratio
based on the simultaneously registered
intensities for Al and Si:
where CSi and CAl are the weight concentrations of
the Si and Al elements in the volume excited by
the electron beam (see Fig. 1), ISi and IAl are the
intensities, (i.e. the integral counts of the respective
Fig. 1. Schematic representation of the beam±specimen±detector
geometry during microanalysis, corresponding to a single particle
(see the text for the discussion concerning the geometrical values).
C. SaÃrbu, B. Delmon / Applied Catalysis A: General 185 (1999) 85±97 87
peaks obtained after subtraction of the background
counts) of the respective X-ray K
lines emitted by the
Si and Al atoms registered simultaneously, whereas
KAlSi is a speciÆc parameter characterising the spectrometer.
Fig. 1 shows the geometry used in our measurements.
We tried to fulÆl two conØicting conditions: to
explore the narrowest specimen surface area and
volume and, on the other hand, to get an acceptable
counting accumulation in a reasonable lapse of time.
The former condition requires a strongly focalised
electron beam, whereas the second one requires a
relatively high counting rate per investigated ``point''.
By working in the STEM-HR mode, using the smallest
condenser aperture which is recommended by the
manufacturer of the electron microscope for getting
the highest attainable STEM image resolution, these
conditions were satisfactorily matched when the accumulation
time was 300 s for all analysed points, except
for two specimens where the time was 600 s per point.
During that time the electron beam was in a Æxed
position under the control of the scanning device. The
particles subjected to analysis, as well as the explored
``points'', were selected in a random manner.
Chemical composition data of comparable samples belonging to the two series of sol±gel silica±aluminas subjected to the microanalytical
nanoscale investigation by EDS X-ray spectroscopy
Sample name () Laboratory prepared samples Industrially prepared samples
Global chemical composition
(as established by laboratory
composition (as indicated
by the producer)
Percent Global Si/Al
weight ratio a
Percent Global Si/Al
SA-95 4.5% Al2O3 18.74 Fig. 1(a) 6.5% Al2O3 12.70 Fig. 1(b)
95.5% SiO2 93.5% SiO2
SA-90 10.8% Al2O3 7.29 Fig. 1(c) 10% Al2O3 7.95 Fig. 1(d)
89.2% SiO2 90% SiO2
SA-85 14.8% Al2O3 5.08 Fig. 2(a) 12% Al2O2 6.48 Fig. 2(b)
85.2% SiO2 88% SiO2
13.3% Al2O3 5.76 Fig. 5(a)
13.3% Al2O3 5.76 Fig. 5(b)
SA-75 26.1% Al2O3 2.501 Fig. 2(c) 24±26% Al2O3 2.65e Fig. 2(d)
73.9% SiO2 76±74% SiO2
SA-50 50.6% Al2O3 0.862 Fig. 3(d)
SA-40 62.2% Al2O3 0.536 Fig. 3(a) 60% Al2O3 0.588 Fig. 3(b)
37.8% SiO2 40% SiO2
SA-30 70.5% Al2O3 0.369 Fig. 3(c)
SA-15 86.8% Al2O3 0.134 Fig. 4(a) 85% Al2O3 0.1558 Fig. 4(b)
13.2% SiO2 15% SiO2
a These values are marked in the graphical representations by (A).
b First delivery.
c Second delivery.
d The pro-ducer did not indicate a precise value for the chemical global composition.
e This value is calculated by considering an average composition of 25% Al2O3á75% SiO2
88 C. SaÃrbu, B. Delmon / Applied Catalysis A: General 185 (1999) 85±97
As shown in Fig. 1, because of the unavoidable drift
effect occurring always in the electron microscope
during long-lasting observation of an object, the electron
beam normally sweeps an area larger than the
electron beam cross-section during each ``point'' analysis.
The extension of this area depends on the beam
diameter and the drift length.We tried to minimise the
drift effect by keeping for some time a larger area of
the observed grid hole under the action of the unfocussed
electron beam, until the drift rate diminished
and some stabilisation was observed, and starting the
Чтобы распечатать файл, скачайте его (в формате Word).
Ссылка на скачивание - внизу страницы.