CERAMICS
AND
CATALYSIS
141
material
could
be
associated with
an
iron contaminant, probably ema-
nating
from
steel pipework
or the
reaction vessel, masking
the
highly
active
palladium atoms
as
well
as the
promoter atoms,
and
greatly
re-
ducing
the
catalytic activity
of the
material.
The
surface analysis
of
naturally occurring minerals, although
straightforward
in
principle, provides many problems
in
practice.
Whilst
the
provision
of an
elemental
surface
analysis
by XPS is
straight1
forward,
extracting
the
required
level
of
chemical information
can be
difficult.
There
are two
problems involved. First, electrostatic charging
often
means
that
the
confidence level
on the
charge-corrected
peak posi-
tion
is
greater than
the
spectral window containing
the
chemical
in-
formation!
Second,
the
peaks
often
have
a
very
poor
photoelectron
cross-section (e.g.,
Si 2p and Al
2p).
The
methods
by
which these prob-
lems
have been overcome
are
closely related
to
each other
and
have been
pioneered
by
Castle's Group
at the
University
of
Surrey
UK.
The
problem
of
sample charging
may be
overcome
by
reporting
the
separation
of two
peaks rather than
the
absolute binding energy
of a
photoelectron transmission.
In
some elements
the
Auger chemical
shift
is
equal
to, and
sometimes greater than,
the XPS
chemical
shift,
and the
potential
exists
for
extracting chemical information
via the
Auger
parameter
(a)
defined
in
Chapter
3, as:
where
E
B
is the
binding energy
of the XPS
peak (e.g.,
1s or 2p) and EK is
the
kinetic energy
of the
attendant Auger peak (e.g., KL
2,3
L
2,3
).
In the
case
of
aluminium
and
silicon
the
KL
2,3
L
2,3
Auger
transition
is not
directly accessible although
the
bremsstrahlung radiation from
a
con-
ventional
X-ray
gun is
able
to
eject
sufficient
Si 1s
electrons
to
produce
a
measurable
SiKL
2,3
L
2,3
peak.
The
silicon Auger parameter calculated
in
this
way is
independent
of
sample charging
but
strongly dependent
on
both molecular
and
crystalline structure.
There still
exists
the
problem
of
poor
photoelectron
cross-section
of
the Al 2p and Si 2p
levels
and the
only
way to
overcome this
is to use a
higher-energy
X-ray anode able
to
excite
1s
core levels
of
these
elements.
Various possibilities exist including
ZrLa
and
TiKa,
but the
ones which represent
the
best combination
of
sensitivity
and
resolution
are
rnonochromated AgLa
and
CrK/3. Using such
a
source
it is
possi-
ble
to
record
1s-KLL
Auger parameters;
a
spectrum obtained with
a