Scanning Electron Microscopy and X-Ray Microanalysis

386

23

. Figure 23.9a shows examples of additional geometric
shapes created with K411 glass: deep, narrow pits from
diamond scribe impacts, microscopic particles (major
dimen sions < 100  μm), and macroscopic particles (major
dimensions > 500  μm). When random locations are ana-
lyzed on these targets, the range of measured Mg and Fe
concentration values is shown in. Fig. 23.9b, covering an
order of magnitude in both constituents. This huge range
occurs with EDS spectra that were readily measurable
despite the severe departure from the ideal flat specimen
geometry.
These results demonstrate that the SEM microanalyst
must realize that just because an EDS spectrum can be
obtained when the stationary beam is placed on a topo-
graphic feature of interest, the resulting analysis may be sub-
ject to such egregious errors so as to be of little use.
Sometimes analysis locations that are surprisingly close on a
microscopic scale can produce very different results.
. Figure 23.10 shows a fractured fragment of pyrite (stoi-
chiometric FeS 2 ) which has been analyzed at various loca-
tions (conditions: E 0 = 20 keV; DTSA-II calculations with Fe
and CuS as standards, followed by normalization). Despite
the proximity of the analyzed locations, the results vary
greatly. Thus, analysis at location 3 produces a nearly perfect
match to the stoichiometric values with relative deviation
from expected value (RDEV) within ±0.15 %, while analysis
at nearby location 9 (about 25 μm away) suffers relative
accuracy of ±36 %, while at location 7 (about 50 μm away)
the RDEV is ±100 %.

kThe Takeaway

Just because a feature can be observed in an SEM image and

an EDS spectrum can be recorded does not mean that a suc-

cessful and useful quantitative analysis can be performed!

23.4 Useful Indicators of Geometric Factors

Impact on Analysis

There are strong diagnostic indicators that reveal the impact

of geometric factors on analysis:

23.4.1 The Raw Analytical Total

The raw analytical total is the sum of all the constituents

measured (including any constituents such as oxygen calcu-

lated on the basis of assumed stoichiometry of the cations).

For an ideal flat sample measured with the beam energy

selected in the “conventional range” (E 0 = 10 keV to 25 keV)

and following a standards-based–matrix correction factor

protocol, the analytical total typically will fall between 0.98

and 1.02 mass fraction (98–102 wt %), a consequence of the

uncertainties inherent in the measurement process (counting

statistics) and in the calculated matrix correction factors. If

the raw analytical total exceeds this range, it is usually an

indication of a deviation in the measurement conditions

(e.g., beam current drift). If the raw analytical total is below

this range, this may again indicate a deviation in the

11

11.05

11.1

11.15

11.2

11.25

11.3

11.35

11.4

11.45

11.5

11.55

7.8 7.9 8 8.1 8.2 8.3 8.4 8.5 8.6 8.7 8.8 8.9

K411_1-mu_Diamond

Mean compared to

SRM values

Fe 11.41 (11.21%) +1.8% rel

Mg 8.76 (8.85 %) -1.0% rel

Mg (weight percent)

Fe

(weight

perc

en

t)

1 mm diamond

0.1 mm alumina polish

SRM

Certificate

values

K411_Highly polished

. Fig. 23.7 Analysis of SRM470
(K411 glass) with surface rough-
ness produced by abrading with
1-μm diamond grit

Chapter 23 · Analysis of Specimens with Special Geometry: Irregular Bulk Objects and Particles