Tile:
Author(s):
Submitted
to:
Silicon photodiode characterization
from
1
eV to
10
kev
SPE:
Optical Science, Engineering,
and
Instrumentation conference
Jul27
-Aug
1,1997
in
San
Diego,
CA
DISCLAIMER
This report
was
prepared
as
an
account of work sponsored by
an
agency of the United
States Government. Neither the United States Government
nor
any agency thereof, nor
any
of
their employees, make any
warranty,
express or implied, or
assumes
any
legal
liabili-
ty
or responsibility for the
accuracy,
completeness,
or
usefulness
of any information, appa-
ratus,
product,
or
process
disclosed,
or represents that
its
use
would not infringe privately
owned rights. Reference herein
to
any specific commercial product, process, or
service
by
trade name, trademark, manufacturer, or otherwise does not
necessarily
constitute
or
imply its endorsement, recommendation, or favoring by the United States Government
or
any agency thereof. The views and opinions of authors expressed herein
do
not necessar-
ily
state
or
reflect those of
the
United States Government or any agency thereof.
Silicon photodiode characterization from
1
eV to
10
keV
G.
C.
Idzorek
and
R.
J.
Bartlett
Los
Alamos
National Laboratory,
Los
Alamos,
NM
87545
ABSTRACT
Silicon
photodiodes offer a number of advantages over conventional photocathode
type
soft
x-ray detectors in pulsed power
experiments. These include a nominally flat response, insensitivity to
surface
contamination, low voltage biasing
requirements, sensitivity to low energy photons, excellent detector to detector
response
reproducibility, and ability to operate
in
poor vacuum or
gas
backfilled experiments.
Silicon photodiodes available from International Radiation Detectors
(IRD),
Torrance, Californa have
been
characterized
for absolute photon response from
1
eV to
10
keV photon energy, time response, and signal saturation levels. We have
assembled individually filtered photodiodes into an array designated the
XUV-7.
The
XW-7
provides seven photodiodes in
a vacuum
leak
tight, electrically isolated, low noise, high bandwidth, x-ray filtered assembly in a compact package with a
3.7
cm
outside diameter. In addition we have assembled the diodes in other custom configurations
as
detectors for
spectrometers.
Our
calibration measurements show factor of ten deviations from the silicon photodiode theoretical flat response due to
diode sensitivity outside the center 'sensitive
area'.
Detector
response
reproducibility between diodes appears to be better
than
5%.
Time
response
measurements show a
10-90%
rise time of about
0.1
nanoseconds
and a fall time of about
0.5
nanoseconds.
Silicon photodiodes
have
proven to
be
a versatile
and
useful
complement to
our
standard photocathode detectors for
soft
x-
ray measurement and
are
very competitive
with
diamond for a number of applications.
Keywords:
silicon photodiodes, x-ray detectors
1.
INTRODUCTION
Silicon P-I-N diodes
are
excellent for
use
as
x-ray detectors due to their high sensitivity, stability, and essentially flat
response'.
Also
unlike photoemissive
soft
x-ray detectors that emit photoelectrons only from the top few angstroms
of
their
surface the silicon photodiodes behave similarly to an ion chamber with the sensitive region being the internal intrinsic
silicon layer.
Having
internal rather
than
external sensitivity renders the silicon virtually immune to surface contamination.
A
drawback to older silicon photodiodes
has
been
the doped dead layer that blocked low energy x-ray photons. However,
using special doping techniques the International Radiation Detectors
(IRD)
2s3
silicon photodiodes are constructed without
a doped surface dead layer. For
our
pulsed power work we have
used
the
IRD
type
HS-1
diode chips mounted into a
custom housing of
our
design. Despite its small sensitive
area
of
0.05
mm2
centered
on
the
1
mm2
chip we must stand back
several meters from typical pulsed power radiation
sources
to reduce detector signals to usable levels.
On
the Sandia
National Laboratories PBFA-2 pulsed power machine that
can
output
200
TW
our
detectors saturate at a distance
of
20
meters. These extreme operating conditions motivated
us
to characterize the
HS-1
diode saturation, time, and spectral
response.
2.
DIODE CONSTRUCTION
Figure
1.
shows
the P-I-N diode
used
in
OUT
detector system. Diodes are constructed by epitaxially
growing
a
"-type
silicon
(equivalent to intrinsic) layer on top of a heavily doped
pi-
silicon wafer substrate. The cathode contact is constructed
by
heavily ndoping the top of the
z-type
layer and then a passivating Si02 layer is deposited. X-rays must pass through the
thin
Si02 surface passivating layer designed to protect the silicon from atmospheric moisture. Newer designs nitride the
Si02 window to improve radiation hardness but the x-ray transmission characteristics remain almost the same. By using the
minimum thickness Si02 entrance window the detector remains responsive through the vacuum-ultraviolet and very
soft
x-
ray region.
.
,-
cathode contact ring
8000A
window
I
10
micron epitaxial layer
f
anode contact
Figure
1.
HS-1
P-I-N
diode construction
3.
REPRODUCIBILITY
The
HS-1
P-I-N
diodes we
use
are
mass
produced
on
a
4"
or greater diameter
silicon
wafer. Each wafer
can
yield over
5000
diodes,
all
created under identical conditions and therefore should have identical characteristics. Figure
2.
shows
a
visible
light monochromator calibration where we tested
49
diodes and found response variations with
a
standard deviation well
within the
instrumental
measurement error
bars.
The
vacuum
ultraviolet and
soft
x-ray
response
should behave similarly
as
they too
are
sensitive only to the upper layers of the diode. Note that the data shows
a
peak
response
value of
1150
A*cm2
/
MW
which
is
equivalent to
a
response of
2.3
A/W
for
a
0.05
mm2
detector.
This
is about four times higher
than
is
normal
for silicon visible responsivily. We explain
this
apparent anomaly
in
the
'5.
Spectral
Response'
section.
400
L
*
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
eV
Figure
2.
Average
response
and
1-0
error
bars
for
49
HS-1
silicon diodes.
4.
TIME RESPONSE
The diode
response
time
is
determined by the resistances and capacitances present
in
the detector and measuring circuitry
and
the
charge carrier transit times.
In
conventional p-n diodes
if
the bias voltage is not
high
enough to
fully
deplete the
lightly doped p-type or n-type silicon
or
if
the photogenerated charge reduces the effective bias then the undepleted lightly