Scanning Electron Microscopy and X-Ray Microanalysis

336

20

CO, CO 2 , and H 2 O that evaporate into the vacuum, causing

substantial mass loss from the interaction volume. At the

highest beam currents, typically 10–100  nA, used with a

focused beam at a static location, it is also possible to cause

highly damaging temperature elevations, which further exac-

erbate mass loss. Indeed, when analyzing this specimen class

it should be assumed that significant mass loss will occur dur-

ing the measurement at each point of the specimen. If all con-

stituents were lost at the same rate, then simply normalizing

the result would compensate for the mass loss that occurs

during the accumulation of the X-ray spectrum. Unfortunately,

the matrix constituents (principally carbon compounds and

water) can be selectively lost, while the heavy elements of

interest in biological microanalysis (e. g., Mg, P, S, K, Ca, Fe,

etc.) remain in the bombarded region of the specimen and

appear to be present at effectively higher concentration than

existed in the original specimen. What is then required of any

analytical procedure for biological and polymeric specimens

is a mechanism to provide a meaningful analysis under these

conditions of a specimen that undergoes continuous change.

Marshall and Hall (1966) and Hall (1968) made the original

suggestion that the X-ray continuum could serve as an inter-

nal standard to monitor specimen changes. This assumption

permitted development of the key procedure for beam-sensi-

tive specimens that is used extensively in the biological com-

munity and that is also applicable in many types of polymer

analysis. This application marks the earliest use of the X-ray

continuum as a tool (rather than simply a hindrance) for

analysis, and that work forms the basis for the development of

the peak-to-local background method applied to challenging

geometric forms such as particles and rough surfaces. The

technique was initially developed for applications in the high

beam current EPMA, but the procedure works well in the

SEM environment.

The Marshall–Hall method (Marshall and Hall 1966)

requires that several key conditions:

1. The specimen must be in the form of a thin section,
   where the condition of “thin” is satisfied when the
   incident beam penetrates with negligible energy loss. For
   an analytical beam energy of 10–30 keV, the energy loss

Albite

E 0 = 15 keV

100 s spectra

1 μm x 1 μm

10 μm x 10 μm

Fixed beam

Photon energy (keV)

Counts

140 000

120 000

100 000

80 000

60 000

40 000

20 000

0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Albite

E 0 = 15 keV

Fixed beam

10 s spectra

1 st (10 s)

5 th (50 s)

10 th (100 s)

Photon energy (keV)

Counts

7 000

6 000

5 000

4 000

3 000

2 000

1 000

0

0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0

Albite_Point_10s_15kV15nA

Albite_Point_100s_15kV15nA

Albite_Point_50s_15kV15nA

Albite_100kX_15kV15nA

Albite_spot_15kV15nA

Albite_10kX_15kV15nA

. Fig. 20.18 Albite (NaAlSi 3 O 8 ); E 0 = 15 keV, 15 nA: (upper) effect of increasing dose on the Na peak; (lower) effect of fixed beam versus scanned
beam on the Na peak

Chapter 20 · Quantitative Analysis: The SEM/EDS Elemental Microanalysis k-ratio Procedure for Bulk Specimens, Step-by-Step