150
ELECTRON SPECTROSCOPY
IN
MATERIALS SCIENCE
sources,
within Siegbahn's Group
at
Uppsala, Sweden,
led to a
com-
mercial
product.
The
launch instrument
was
installed
in
ICI's Wilton
Laboratories,
UK, in
1989.
Graham Beamson
and
David Briggs
at ICI
used this spectrometer
to
record high-resolution spectra
of
over
100
homopolymers. These were subsequently published
as the
High Resolu-
tion
XPS of
Organic Polymers, although
now out of
print
the
spectra
are
available
in
electronic
form
from
SurfaceSpectra
Ltd,
Manchester
(www.surfacespectra.com/xps)
The
influence
of the
prototype
spectrometer,
and the
seminal hand-
book that sprang from
the
polymer
XPS
data acquired
on it,
cannot
be
overemphasized:
the
routine
use of a
high-resolution X-ray source
pro-
vided
a
much greater level
of
information than
had
previously been
obtainable.
The
issue
of
charge compensation using
an
electron flood
gun for the XPS
analysis
of
polymers
had
also been established
at a
higher level
of
precision.
In a
very
short
time
all the
major manufac-
turers
of XPS
systems were
offering
high-resolution
XPS
systems with
performance approaching,
and
eventually surpassing, that
of the
ESCA300.
One
area
of
fertile
development
has
been
in the
design
of
charge compensation systems. Thus, although good quality
XPS can be
carried
out on
polymers using twin anode (non-monochromatic)
sources,
for the
best quality analysis
the use of a
monochromatic source
is
necessary.
As
all
organic polymers contain substantial quantities
of
carbon,
it is
the
chemical
shift
of the
carbon
1s
electrons which predominates
the
interpretation
of XPS
data
from these materials. Figure
5.36
shows
the
C 1s
spectrum
of
poly(methyl methcrylate)
recorded
using
a
monochro-
matic AlKa source.
In
order
to
achieve satisfactory peak
fitting of the
experimental spectrum
it is
necessary
to use
four
singlets.
These
peaks
correspond
to
aliphatic
carbon
at a
binding energy
of 285 eV (a
useful
internal
standard
in
polymer analysis), carboxyl carbon
at a
separation
of
approximately
4.2 eV
from
the
C-C/C-H
peak,
and
ether-like
car-
bon at a
distance
of
about
1.8 eV (as
indicated
on the
structural
formula
of
this polymer, which
is
also included
in
Figure 5.36).
The final
component
at a
separation
of
0.7-0.8
eV is due to a
second-
ary
chemical
shift,
which
is the
effect
of the
carboxyl group
on the
unsubstituted carbon atom
in the
C-COiR
structure. Such secondary
shifts
(also known
as
nearest neighbour
effects)
have only been
reported
in
the
literature relatively recently
and it is
clear that their
identification
results
from
improvements
in
peak-fitting
methods
as
well
as the