RECEIVED BY D T l t
WIMI Z<i i^'w
MASTER
Gulf G^n^rai A-l-omic
lncorporat»a
AEC RESEARCH AND
DEVELOPMENT REPORT
GA-85T^
Vol. 2
CESIUM SORPTION IN MATERIALS FOR THERMIONIC CONVERTERS, II
by
M. K. Yates and G. 0 . F i t z p a t r l c k
P r e p a r e d under
C o n t r a c t AT(0i^-3)-l6T
P r o j e c t Agreement No. lU
for the
San F r a n c i s c o O p e r a t i o n s Office
U. S . Atomic Energy Commission
A p r i l 1970
^^^ ^^,, .oC^iENT U. L-KUMMBft
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,
makes any warranty, express or implied, or assumes any legal
liability or responsibility for the accuracy, completeness, or
usefulness of any information, apparatus, 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 necessarily state or reflect those of the United States
Government or any agency thereof.
DISCLAIMER
Portions of this document may be illegible in
electronic image products. Images are produced
from the best available original document.
»
Gulf General Ato m i (
Incorporot.d
P. O. Box 608, San Diego, California 92112
AEC RESEARCH AND
DEVELOPMENT REPORT
GA-851^, Vol. 2
C-93b, Advanced
"concepts f o r Future
Application Conversion Devices
M-3679 ( 5 6 t h Ed.)
CESIUM SORPTION IN MATERIALS FOR THERMIONIC CONVERTERS, I I
by
M. K. Yates and G. 0 . F i t z p a t r i c k
Prepared under
C o n t r a c t AT(Oi<-3)-l67
P r o j e c t Agreement No. lijfor the
San F r a n c i s c o O p e r a t i o n s Office
U. S. Atomic Energy Commission
LEGAL
NOTICE
TWB report was prepared ae an accounl o£ Government sponsored work Netther the UL.II
States, nor the Commission, nor any person acting on behalf of the Commluslon
A. Makes anj warranty or representation, expressed or implied with respect to the f. t
racy, completeness, or usefulness of the Information contained in this report, or that Uit L
of any information, apparatus, method, or process dlaclosed In this report may not Infi Ir
privately owned rights, or
B. AsBumes any liabilities with respect to the USP of, or fur ulamages resulting frun. 1
use of any Information, apparatus, method, or process disclosed in this report
As used in the above, "person acting on behalf of the Commission" Includes an> e.
ployee or contractor of the Commission, or employee of such contractor, to the extent U
such employee or contractor of the Commission, or employee of such contractor prepafi
dlBseminates, or provides access to, any infoi mation pursuant to his employment or con'".
with the Commission, or his employment with such c
Gulf General Atomic P r o j e c t U022
A p r i l 1970
;T IS \-7>!UM?rsD,
^r^TS^'**'^y
SUMMARY
This report covers work done from January 1968 to June 19^9 under AEC
Contract Number AT (OU-3)-l67 on cesium sorption in materials which serve
as high-temperature cesium reservoirs in thermionic converters.
Cesium vapor pressure/sample temperature/sample loading chaiBcteristics were measured for two different types of graphite and one sample of
porous tungsten.
When the data on the graphite samples were compared to
some earlier data, it was found that the isotherais for different types of
graphite were similar; but the tnansition between two cesium graphite
equilibrium states was better defined in some samples than in others. Hysteresis effects were present in most of the graphite samples hut appear
to pose no particular problems to the thermionic converter. The data concerning the cesium loading in the porous tungsten sample were obscured by
soluble cesium compounds present on the inner oven surfaces. It was determined, however, that the primary sorption mechanism was surface adsorption.
Long-term isostere stability tests were run on two graphite samples
and one carbon sample. All samples were operated ~ 800 C at a cesium vapor
pressure of ~ 10 torr. Periodically cesium pressure/sample temperature
curves were taken.
Stability testing of one graphite sample using a radio-
active technique for cesiimi pressure determination was discontinued after
3360 hours because of high background radiation levels. No changes in
either heat of reaction or cesium pressure larger than the experimental
error of 1 20^ were observed.
Stability testing of the carbon sample, also
using the radioactive technique of cesium pressure determination was discontinued after 2000 hours for the same reason. The variations observed in
the carbon's heat of reaction were not larger than the experimental error
of 1 30^. During the initial 1200 hours a 20-35'^ cesium pressure decrease
ii
was observed, an amount corresponding to that expected from observed cesium accumulation in the counting volijme.
The second graphite sample was tested in an apparatus where the cesium vapor pressure was determined by measuring the cesium condensation temperature of an insulator. This sample was tested for 1225 hours, and no
significant change was detected in the sampled temperature-cesium pressure
isostere. The insulator technique proved capable of a cesium condensation
temperature determination within 1 10 C, with relative determinations
accurate to _ 2 C.
Finally, the required temperature control accuracy for some sorption
reservoirs is shown to be approximately one-half that required for a liquid
cesium reservoir, and tolerable cesium losses in sorption reservoirs are
given for a cesium vapor pressure change.
iii
CONTENTS
SUMMARY
il
INTRODUCTION
1
PRESSURE-TEMPERATURE-LOADING TESTS
1
GRAPHITE PRESSURE-TEMPERATURE-LOADING TEST
3
HYSTERESIS EFFECTS IN GRAPHITE
11
POROUS TUNGSTEN SAMPLE PRESSURE-TEMPERATURE LOADING TEST
lU
LONG-TERM THERMAL STABILITY TESTING OF CESIUM LOADED SAMPLES
lU
RADIOACTIVE TECHNIQUE
17
INSULATOR TECHNIQUE
19
APPLICATION TO THERMIONICS
27
REFERENCES
37
FIGURES
1.
P l a n view of p r e s s u r e - t e m p e r a t u r e l o a d i n g a p p a r a t u s
2
2.
P r e s s u r e - l o a d i n g i s o t h e r m s sample No. 1 GGA annealed
pyrolytic graphite
5
3.
P r e s s u r e - l o a d i n g isotherm sample No. 2 speer carbon
p r o d u c t s molded g r a p h i t e 3^995
6
h.
P r e s s u r e - l o a d i n g i s o t h e r m s carborundum CARB-I-TEX 700
7
5.
P r e s s u r e - l o a d i n g i s o t h e r m sample No. k CARB-I-TEX 700
8
6.
Sample No. h CARB-I-TEX 700 c o n s t a n t cesium l o a d i n g
i s o s t e r e s carborundum CARB-I-TEX sample No. U
iv
13
7.
8.
Isotherm h y s t e r i s i s s o r p t i o n b r a n c h carborundum
CARB-I-TEX sample No. U
15
Isotherm h y s t e r e s i s on d e s o r p t i o n branch carborundum
CARB-I-TEX 700 sample No. U
l6
Long term s t a b i l i t y t e s t s t a t i o n
l8
10.
Long term s t a b i l i t y t e s t - l i q u i d sample
20
11.
Long term s t a b i l i t y t e s t - carborundum CARB-I-TEX 700
21
12.
Long term s t a b i l i t y t e s t - carborundum CARB-I-TEX 700
22
13.
Long term s t a b i l i t y t e s t - pure carbon FC-50
23
lU.
Long term s t a b i l i t y t e s t - pure carbon FC-50
24
15.
Background cesiimi c o n c e n t r a t i o n p r o f i l e s i n long
9.
term s t a b i l i t y t e s t ovens
25
16.
PT T e s t i n s u l a t o r assembly
26
17.
I n s u l a t o r t e m p e r a t u r e - r e s i s t a n c e curve long term
s t a b i l i t y t e s t - carborundum CARB-I-TEX 700
Cesium c o n d e n s a t i o n t e m p e r a t u r e - sample t e m p e r a t u r e data
carborundum CARB-I-TEX 700
Six sample long term cesium p r e s s u r e - s a m p l e temp e r a t u r e t e s t assembly
18.
19.
20.
T o l e r a b l e cesium l o s s e s a s a f u n c t i o n of sample temperature
28
29
30
33
TABLES
k
1. Physical Characteristics
2.
Sample Impurities (ppm)
10
3. Graphite Sample Characteristics
12
k.
31
Slope of P-T curve at 10 torr for various materials
V
INTRODUCTION
This report is a continuation of work on cesium sorption in materials
for thermionic converters.
The cesium sorption testing of various sam-
ples is discussed, where the primary characterization involves the measurement of cesium pressure/cesi\jm loading isotherms. The design of long-term
thermal test apparatus is also discussed, and results of the tests are given.
Finally some aspects of the incorporation of sorption reservoirs into thermionic converters are considered, namely, tolerable cesium losses and required temperature control for various materials used as sorption reservoirs.
FRESSURE-TEMPERATURE-LOADING TESTS
The design of the apparatus used in most of these tests has been described before.
It has been modified slightly (as shown in Fig. l) in
order to better outgas the sample and to eliminate any possibility that
cesium might collect on the reference side of the capacitance manometer.
Basically, the apparatus consists of a liquid-cesium reservoir (with
~ 2 grams of cesium having a specific Cs
activity of 25O |j.curies/gm), the
manometer, and the sample. The sample's cesium loading is determined by
counting the radioactive cesium present. The cesium vapor pressure is determined by measuring the temperature of the liquid reservoir and using the
P-T curve recommended by Nottingham and Breitwieser as that giving the best
2
fit to experimental vapor pressure data. The capacitance manometer is
calibrated against the liquid reservoir.
The cesium sorption characteristics of two graphite samples and one
porous tungsten sample have been measured.
Table 1 lists the physical
characteristics of these samples as well as those of two other graphite samples that were tested earlier. Table 2 lists the impurities for these
samples.
1
TO Nf
TO VACUUM
SYSTEM
COMPRKSSION
PORTS
SAMPLE
ro
BAKEABLE
VALVES
LEAD SHIELDING
CESIUM RESERVOIR
NEGATIVE PRESSURE
HOUSING
2 IN. Nal CRYSTAL
PRE-AMPLIFIER'
Fig.
1
P l a n View of P r e s s u r e - Temperature Loading Apparatus
PHOTO-TUBE
GRAPHITE PRESSURE-TEMPERATURE-LOADING TEST
The reproducibility of sorption characteristics in nominally identical types of graphite was of interest, as well as comparisons between samples of different types. As shown in Table 1, samples of annealed pyrolytic, molded, and a layered graphite cloth were tested.
The annealed pyrolytic graphite was produced by vapor deposition followed by a high-temperature heat treatment (3000 C).
The result was a dense
graphite that is highly oriented with almost all of its carbon layers parallel to the substrate surface. As a result, its physical properties vary
greatly with the axis chosen, i.e., it is very anisotropic.
The molded graphite was made from particles less than 0.003 inch in
diameter compacted under high pressure.
It is less dense and much more
isotropic than the pyrolytic graphite.
The last sample type tested was CARB-I-TEX, a unique structure made
up of layered graphite cloth impregnated with a graphite binder. We have
found its expansion properties to be anisotropic, expanding much more in a
direction perpendicular to the cloth layer planes than in the other two when
loaded with cesium. This type of graphite is the least dense of those tested,
due primarily to imperfect binder impregnation. Two samples of the same
nominal type, but from different batches were tested.
The presence in the
sample of two forms of graphite, the cloth and the binder, makes it very
difficult to get meaningful x-ray data on layer spacing and crystallite
size.
The pressure-temperature-loading characteristics of the samples are
shown in Figures 2 through 5*
Data have been taken over a cesium pressure
range of 0.1 to 30 torr and sample temperatures from 600 to 1100 C. Loadings up to 1.2 grams cesi\;im/gram graphite have been obtained. Also shown
in Figures 2 through 5 is the 700 C isotherm based on Salzano and Aronson's
3
work on the cesium-graphite lamellar compounds.
3
Table 1
Physical Characteristics
Np BET Area
Sample
No.
Type
Source
Density
gm/cm
/
m2 /cm
X-Ray Data''"
2 / 3
m'^/cm-^
Axis 1
1
Annealed
Pyrolytic
Graphite
Gulf General
Atomic
2.08
0.8
1,7
2
Molded
Graphite
Speer Carbon
Products
1.62
2.28
3.69
1.35
1.48
0.39
0.75
0.56
1.11
V
5
Porous
Tungsten
W-2 w/oB
E l e c t I&'
OpticalSyste ms
11.81
Axis 2
Powder
Lc
1150
6.734
675
6.729
675
6.719
350
6.750
370
6.713
54o
6.724
Avg. p o r e diameter=1.92(i
S
Lc
%
\
Carborundum
Flat
Co.
layered
Graphite
Cloth
bonded with
Graphite,
CAEB-I-TEX
700
(i)
CQ
No meaningful d a t a o b t a i n e d
No meaningful data o b t a i n e d
Pore c r o s ^ s e c t i o n a l d e n s i t y =
3.87 X 10° pores/cm2
, ..... —
_ . ~
"Axis 1 and Axis 2 are perpendicular a x i s , one chosen t o correspond a p p r o x i m a t e l y t o t h e s a m p l e ' s n a t u r a l a x i s
determined by a n i s o t r o p y . The powder t e s t e d was o b t a i n e d from sample s c r a p p i n g s and should r e p r e s e n t a sample
average.
Cf, represents an average l a y e r spacing determined from t h e angle of r e f l e c t i o n of X-ray peak.
Lc r e p r e s e n t s an average c r y s t a l l i t e s i z e , determined from t h e X-ray peaks h a l f - w i d t h . A c r y s t a l l i t e i s a
t h r e e - d i m e n s i o n a l a t o m i c a l l y o r d e r e d s t r u c t u r e bounded by micropores o r h i g h l y a n g l e d carbon l a y e r s .
200
300
400
500
600
700
800
LQADIBG (MG C S / G GRAPHITE)
Fig.
P r e s s u r e - Loading Isotherms Sample No
GGA Annealed P y r o l y t i c G r a p h i t e
900
1000
noo
1200
100.0
10.0
!^
OA
1 .0.
0. 1
100
,
200
300
400
500
600
700
800
900
LOADING.(MG Cs/G GRAPHITE)
Fig. 3
P r e s s u r e - Loading Isotherm Sample No, 2
Speer Carbon P r o d u c t s Molded G r a p h i t e 3^4-995
1000
1 too
TOO!
-J
100
200
300
itOO
500
600
700
800
900
LOADING (MG Cs/G GRAPHITE)
Fig. h
P r e s s u r e - L o a d i n g Isotherms Carborundum CARB-I-TEX 700
1000
1100
00 i l
0001
OOi Xai-I-aaVD ^ 'OM aidmBg uuaqq-osi S u T p B o i - a j n s s a j j
006
008
00^
009
( 3 i l H d V y D 9 / s o 9W) ONIOVOI
OOS
00*7
00£
C, '2xg.
003
001
0
1
- ° ' D SD
S3
Qtjs 3 °''3S3
/
£ J3 «
~o
/1
/
J:
—
D
NOiidyo S
" ' 3S3 ^^oso
y
?
(: oOOZ)
o
N0lida0S3 a
In
0 J
1k-Pn .^<
/
- ^ S '
^
//
/ / '-^
/
/
V
1
/
/rf'
Y^
/
1
II
^iri
/
^
A
1
^
p.
J/
3-009
^
/
[//
^
^
/
^
L^u
/
P
/
A
^
J
DoOO/
3o008
y
J
/A^
/ /
^
/
/
/
/
/f
/ /
J
/]
-f
V/
V
'/
/
/
3o000l
/
3o00ll
Evidence of the transitions between the various cesium-graphite compounds appeared in all of the samples. In particular, the beginning of the
transition between CsCp. and CsC, ^ was well defined in all samples and
occurred at nearly identical sample temperatures and cesium vapor pressures.
However, the constant pressure regime where the CsCpi and CsC,-^ compounds
in equilibrium was narrow and poorly defined in the CARB-I-TEX samples and
very wide and sharply delineated in the annealed pyrolytic sample. The
molded graphite sample fell between these two extremes in this respect.
Of the other transitions, only the one between CsCp. and CsCo/- was well defined and that only in the annealed pyrolytic and molded graphite samples.
In general, cesium loadings were less than that expected if complete formation of a particular compound had taken place. The two CARB-I-TEX samples
differed in cesium loading by 25^.
In an attempt to explain the difference in the graphite isotherms,
various physical characteristics of the samples were measured. These
characteristics (Tables 1 and 2) include density, BET surface area, and
purity. Also included are the average layer spacing and crystallite size
determined for two perpendicular axes and for a sample average.
One mea-
sured parameter that seems to correlate with "compound definition" in the
sample's cesium sorption characteristics is sample density.
Higher densi-
ties correlate well with more distinctly defined compound states. Larger
crystallite size also correlated with better plateau definition.
In order to study the cesium sorption characteristics of graphite
further, a group of samples was prepared by G. Engle of Gulf General Atomic
Incorporated that covered a very wide range of possible physical properties.
Three groups of samples were prepared, each by a separate technique, and
the samples in each group heat treated at a variety of temperatures.
The first group consisted of five low-temperature pyrolytic carbon
samples deposited on graphite disks at lil-50 C from propane in a fluid bed.
Their carbon structure is turbostratic in nature, and their crystallite
orientation is nearly Isotropic. They have a density near I.67 g/cm , Heat
treatment to 3100 C improves the crystallinity of the carbon structure only
slowly.
q
on
Oti
01
I
01 >
01
T
>
Tl
OT >
TS
OT
^ •0
>
OS
^ •0
Ofl
i
OS >
T >
^•0
>
I >
no
A
OS >
OS >
OT
8
I
I
I
8
I
BO
>
^•0 >
§w
9^:
s
^•0
I
^•0 >
a
8
s
I >
I >
rv
^T
£
s
o
H
T
•OM
aiduiBS
(indd) saillHTlcIWI aidWVS
S siqej,
The second group consisted of five samples of massive pyrolytic carbon, a commercial product probably deposited on a graphite substrate from
methane at 2000°C to 2200°C. Their density is ~ 2.2 g/cm . Their carbon
layer planes are highly aligned parallel to the substrate, but still are of
a turbostratic nature in the as-deposited state. With heat treatment they
quickly become more crystalline.
The third group consisted of seven samples of artificial graphite
made from calcined petroleum coke commercially produced and outgassed at
1300 C, and bonded at GGA with carbon from the pyrolysis of a coal-tar pitch.
The final product is quite porous, with a density of ~ 1.7 g/cm , Further
heat treatment greatly increases its crystallinity.
Table 3 lists the samples, their heat treatment, and x-ray parameters.
No cesium sorption tests have been run on these samples.
HYSTERESIS EFFECTS IN GRAPHITE
Hysteresis effects were observed in almost all of the graphite isotherms.
The first item of interest was whether a transition would occur
from one branch of the curve to the other if the cesium loading was held
constant and a pressure-temperature curve taken.
In order to study this
effect, the capacitance manometer was used with one sample (No. h) to take
pressure-temperature data. The data were taken with the valve to the liquid
reservoir closed. The results are shown in Fig. 6.
Two curves taken at
the same loading are of particular interest; one was taken on an adsorption
branch, one on a desorption branch after unloading from 30 torr. Both
curves indicated that a transition from one hysteresis branch to the other
would not occur when changing sample temperature if the cesium loading remained constant.
Because of the volume of the measuring system, actual
loadings dropped by approximately 1^ when going from 0 to 30 torr.
In order to further investigate the hysteresis characteristics of
the pressure-temperature curves, small sorption-desorption loops were traversed around one operating point, ~ 800 c/lO torr. The results of the
11
Table 3
Graphite Sample Characteristics
Heat Treatment
QQTnT%"l o
T^trr^o
_^
oa,mpj.e xype
—^
Temp
Time
(hr)
(°c)
X-Ray Parameters
Co
0
L^
c
Massive P y r o l y t i c C!arbon
2100
1
6.850
225
Massive P y r o l y t i c Carbon
2650
1
213
Massive P y r o l y t i c Carbon
2900
1
6.715
6.726
Massive P y r o l y t i c Carbon
3000
1
6.726
1350
Massive P y r o l y t i c Carbon
3100
1
6.726
505
Low Temperature
Carbon
Low Temperature
Carbon
Low Temperature
Carbon
Low Temperature
Carbon
Low Temperature
Carbon
Pyrolytic
2100
1
6.910
49
Pyrolytic
2650
1
6.861
80
Pyrolytic
2900
1
6.826
112
Pyrolytic
3000
1
6.816
119
Pyrolytic
3100
1
6.726
505
Graphite
2100
1
6.850
290
A r t i f i c i a l Graphite
2200
1
6.826
350
Artificial
Graphite
2300
1
6.784
46o
Artificial
Graphite
2650
1
6.736
810
Artificial
Graphite
2900
1
6.726
1000
A r t i f i c i a l Graphite
3000
1
6.724
1150
Artificial
3100
1
6.720
1350
Artificial
The
Lc
Graphite
of this sample is aaomolous and is being remeasured.
12
465
(1.0
563 MG Cs/G
CC
CC
o
CC
•=)
CO
to
UJ
CC
CL
ADSORPTION
BRANCH
.0
CC
o
<
DESORPTION
BRANCH
UJ
320 MG/G
0.1
I
0.8
0.9
1.0
1.1
1.2
1000/T C K ) (PRELIMINARY LOADING CALIBRATION)
1
0.7
Fig. 6
_L
Sample No. h CARB-I-TEX 700 Constant
Cesium Loading I s o s t e r e s Carborundum
CARB-I-TEX Sample Wo. k
13
1.3
data taken are shown in Figs. 7 and 8.
When starting from a sorption point
and unloading the sample, the initial point could always be reproduced.
However, the initial point could not be returned to when starting from a
point on the desorption branch and unloading the sample. From these data
it appears that it is desirable for a reservoir in a thermionic converter
to be operating on the sorption branch of the curve to prevent changes in
its equilibrium cesium pressure characteristics due to reservoir unloading.
POROUS TUNGSTEN SAMPLE PRESSURE-TEMPERATURE LOADING TEST
During testing of the W-2w/oB sample, very long equilibrium times
(over a week) and low cesium loadings (2 mg maximum) were found. Postoperational analysis indicated that over 90^ of the cesium present during the
highest observed loadings existed in the form of soluble cesium compounds
present on the oven inner surfaces.
Only 0.12 mg of cesium were found in
the W-2B sample itself.
2
3
Based on the samples calculated surface area of 0.8 m /cm and a
o°2
cesium cross sectional area of 2oA , this corresponds to a cesium coverage
of 2/3 monolayer. This result agrees with cesium sorption seen on previously
tested porous tungsten samples and indicates that the primary cesium sorption mechanism was surface adsorption.
LONG-TERM THERMAL STABILITY TESTING OF CESIUM LOADED SAMPLES
Long-term testing was intended to demonstrate the stability of the
cesium-substrate system in a realistic temperature environment. Two slightly
different approaches were used, varying only in their methods of cesium
vapor pressure measurement.
In the first approach, pressure was measured
by determination of the quantity of radioactive cesium vapor present in a
calibrated volume. In the second approach, pressure was measured by determining the condensation temperature of an insulator exposed to the cesium vapor over the sample.
Ik
0.0 —
> ^ '
9.0 —
y^
CC
CC
V
/
8.0 —
y^
7.0 —
o
CC
6.0
Z
y
(/)
CO
UJ
CL
o
5.0
CL
<
>
y
=5
to
LU
/
k.O
^
' >
/
^
O DESORPTION
^
D SORPTION
V
o
3.0
1
260
1
270
1
280
1
1
290
300
1
310
LOADING (MG Cs/G GR-PRELIMINARY CALIBRATION)
Fig. 7
Isotherm H y s t e r i s i s Sorption
Branch Carborundum CARB-I-TEX
Sample No. h
15
320
10 0
19
-
'1.0 -
y/^^<^^^^
•3.0 ^
y
a:
X^^^
7.0
CC
o
CC
Z)
y^
6.0
to
to
LU
CC
CL
CC
5.0
o
/
CL
'7
y^
X
<
>
to
y
Z
<
'"^ ^ " " ^ C O M M O N DESORPTION
BASELINE
^
i+.O
LU
O
D
O
^5
SORPTION
DESORPTION
3.0
1
1
1
1
1
1
270
280
290
300
310
320
LOADING (MG Cs/G GR-PRELIMINARY
Fig. 8
CALIBRATION)
Isotherm H y s t e r e s i s on Desorption
Branch Carborundum CARB-I-TEX
700 Sample No. h
l6
330
RADIOACTIVE TECHNIQUE
The d e s i g n of t h e t e s t v e h i c l e used w i t h t h e r a d i o a c t i v e t e c h n i q u e i s
shown i n F i g . 9*
The sample was h e l d i n a l/k
t u b e by crimps a t t h e t o p and "bottom.
i n . - d i a m e t e r , 0.030 i n . w a l l
On t h e t o p i s a 10 cm
volume.
The
sample h o l d e r was made e n t i r e l y of niobium w i t h two weld j o i n t s and one
p i n c h off ( o r c o l d w e l d ) .
A f t e r p l a c i n g t h e sample i n s i d e t h e niobium t u b e , t h e assembly was
mounted on t h e p r e s s u r e - t e m p e r a t u r e l o a d i n g a p p a r a t u s and t h e sample o u t gassed.
Then s u f f i c i e n t P-T-L d a t a were t a k e n t o i n s u r e t h a t t h e sample's
s o r p t i o n c h a r a c t e r i s t i c s were s t a b l e .
F i n a l l y t h e sample was e q u i l i b r a t e d
a t ~ 800 C and 10 t o r r cesium p r e s s u r e and s e p a r a t e d from t h e l i q u i d r e s e r v o i r by c l o s i n g t h e v a l v e .
A f t e r c o o l i n g , t h e samples were pinched
i n s t r u m e n t e d , and mounted i n a s e p a r a t e vacuum system.
off,
Pressure-temperature
c u r v e s were t a k e n by counting t h e r a d i o a c t i v e cesium i n t h e volume a t t h e
t o p of t h e sample h o l d e r .
The sample t e m p e r a t u r e s were h e l d c o n s t a n t and
t h e p r e s s u r e - t e m p e r a t u r e curve p e r i o d i c a l l y remeasured.
Three t e s t s were run u s i n g t h i s t e c h n i q u e :
A l i q u i d cesium sample
which served a s a c o n t r o l ; a sample of CARB-I-TEX 700; and a sample of Pure
Carbon FC-50.
The l i q u i d r e s e r v o i r t e s t was t e r m i n a t e d a f t e r 3175 hours
of o p e r a t i o n .
The l i q u i d cesium had been h e l d a t a t e m p e r a t u r e of a p p r o x i -
m a t e l y 366 c (10 t o r r ) and t h e volume a t approximately 500 C during t h e
test.
The r e s u l t s of t h e t e s t a r e shown i n F i g . 10.
No d e t e c t a b l e change
( 1 12^) was observed i n t h e s l o p e of t h e volume count r a t e / c e s i u m p r e s s u r e
l i n e during t h e t e s t .
The CARB-I-TEX 700 sample t e s t was t e r m i n a t e d a f t e r 8226 hours a t
t e m p e r a t u r e , b u t d a t a on t h e s a m p l e ' s s o r p t i o n c h a r a c t e r i s t i c s were taken
o n l y i n t h e f i r s t 336o hours of o p e r a t i o n .
The g r a p h i t e p r e s s u r e - t e m p e r a -
t u r e curves a r e completely c h a r a c t e r i z e d by t h e s a m p l e ' s h e a t of r e a c t i o n ,
and t h e cesium p r e s s u r e a t a g i v e n sample t e m p e r a t u r e .
17
These p a r a m e t e r s
10 CM^ VOLUME
PREAMPLIFIER
PHOTOTUBE
NSTRUMENTATION
i2 IN. Nal):
"CRYSTAL '
LEAD SHOT
oo
•VACUUM CHAMBER
SAMPLE•LONG TERM TEST ASSEMBLY
PINCH-OFF-
D
•TO VACUUM SYSTEM
Fig. 9
Long-Term S t a b i l i t y Test Station
are plotted in Figs. 11 and 12. No detectable changes were noted in either
parameter, both of which are considered accurate to t ^.Qfijo.
The Pure Carbon sample was operated at temperature for a total time
of ij-900 hours, but sorption data were taken only in the first 1200 hours of
operation. The results of this test are shown in Figs. 13 and ik.
No
change was detected in the carbon heat of reaction, which is considered
accurate to + 15-35^. A drop in cesium pressure was observed which correlates well with that expected due to observed cesium losses to the counting
volume.
The carbon-cesium isotherms, reported in Ref. 1, were used to de-
termine the pressure drop expected with the lower cesium loadings.
The termination of testing of these three samples was due to an in3
crease in the radiation background observed in the 10cm volume above the
sample. At the conclusion of the testing, the amount of cesium in the volume was determined by counting techniques. Thirteen, 17 and l6 mg of
cesium were found in the volumes over the carbon, graphite and liquid samples, respectively.
Scans were made across the carbon and liquid volumes
and are shown in Fig. 15-
The residual cesium found correlates well with
the area of inner oven surface exposed to cesium. The volume over the graphite sample was cleaned with water, removing all but 0.05 mg of the cesium
present, thus ruling out the possibility of cesium diffusion into the niobium as a source of cesium buildup. The most likely cause of buildup is
a cesium reaction with oxides or other impurities on the inner volume surfaces.
INSULATOR TECHNIQUE
The design of the sample holder for the insulator technique of cesium
U 5
^
pressure determination ' is shown in Fig. 16. It is fabricated of niobium
and has a Lucalox insulator Cu-Ni brazed in on one end. The sample is
placed in the other end. After outgassing and cesium loading, the sample
holder is sealed by pinching off the niobium tube.
19
CO
to
Q.
1.5
-
CO
UJ
I—
<
Q:
1 .OO-
3
O
ro
o
o
%
0.5
CO
UJ
>
I—
<
UJ
1000
2000
ELAPSED TIME (HRS)
Fig. 10
Long-Term Stability Test - Liquid Sample
3000
oo
o
s
ro
EH
2000
ELAPSED TIME (HRS)
F i g . 11
Long-Term S t a b i l i t y T e s t - Carborundum CARB-I-TEX 700
ro
ro
1000
2000
3000
ELAPSED TIME (HRS)
F i g . 12
Long-Term S t a b i l i t y T e s t - Carborundum CARB-I-TEX 700
o
o
o
ON
II
u
1.2
1.0
-""5-
CQ
CO
.8
0)
•H
-P
cd
iH
(U
K
I
Measured P r e s s u r e a t T sample = 9^0°
O
C a l c u l a t e d P r e s s u r e A f t e r Cesium Loss
t o Volume (Based of Carbon Data Ref.1 )
,2
_L
0
200
UOO
600
800
1000
1200
Elapsed Time (Hours)
F i g . 13
Long-Term S t a b i l i t y Test - Pure Carbon FC-50
1^4-00
IDOO
o
a
o
•r-l
-P
o
«]
ro
O
-p(d
200
llOO
600
800
1000
1200
lUOO
Elapsed Time (Hours)
F i g . lU
Long-Tenn S t a b i l i t y Test - Pure Carbon FC-50
1600
Carbon Long Term S t a b i l i t y Test Oven
Total Cesium Inventory = 12.7 mg
Liquid Long Term Stability Test Oven
Total Cesium Inventory = lo.O mg
Slit
Width
0)
o
o
ro
•H
-P
a
H
K
I
I
B-?
Fig. 15
^
h
^
^
Background Cesium Concentration Profiles in Long-Term S t a b i l i t y Test Ovens
Pinch-Off
J " ^
ro
Niobium Foil
Sample in
Nb foil
.Niobium
Fig. 16 PT Test Insulator Assembly
.Lucalox
Insulator
Pla sma
Spray
During this reporting period one sample^ CARB-I-TEX 700, was tested
using this technique.
It was tested in the same apparatus used for the
other long-term stability tests (Fig. 9)- The sample's Isostere was determined by measuring the insulator resistance-temperature curve at a fixed
sample temperature. The ascending branch of the hysteresis curve has proved
to be most reproducible, and the insulator temperature giving a resistance
of 1000 Q on the ascending branch has been used as the cesium condensation
temperature. A typical insulator resistance-temperature curve is shown in
Fig. 17. The hysteresis in the insulator resistance - temperature curve
combined with temperature profiles in the sample and insulator areas limits
the absolute determination of the cesium dewpoint to _ 10 C. The relative
accuracies of pressure determination correspond to liquid cesium temperature errors of t 2°C. Data taken on the CARB-I-TEX sample initially and
after 1225 hours of operation at a temperature of 775 C indicated that no
significant change in the sample Isostere had occurred (Fig. l8).
Future tests are planned in a test assembly designed to hold six of
these sample holders (Fig. 19)' This assembly was designed to be used
either in the laboratory or in the GGA TRIGA Mark III reactor. A common
heat sink is used on each end, each with redundant heaters. The lower heat
sink is designed to operate around 800 C and will contain all of the samples.
The upper heat sink will operate ~ 350 C and will provide a uniform
thermal environment for the insulator seals.
APPLICATION TO THERMIONICS
Two questions which have often come up during discussions of the applications of a sorption reservoir to a thermionic converter have been:
What accuracy of reservoir temperature control is required for optimum
cesium pressure operation, and what are the tolerable cesium losses for reservoir materials?
Table h lists the slope of the sample temperature/cesium vapor pressure curves for various materials (taken from Ref. l) at a typical operating
27
T i:,amyle
Or.
- 70l'-'(
,•'1
103
Condensation
Temperature Point
(U
o
a
a)
+J
to
•H
CO
0)
K
fH
O
-P
10^
3
10
_L
21+0
260
280
300
T i n s u l a t o r (oc)
F i g . 17
I n s u l a t o r Temperature - R e s i s t a n c e
Curve Long-Term S t a b i l i t y Test Carborundum CARB-I-TEX 700
28
320-
31+0
Initial Data
—
10
D
Fit to Initial Data
1225 Hour Data
u
-p
U EH
o sd
EH O
-p
0) CO
& w
in d
to (u
CQ
Tj
PM
O
•r-l - H
CD CQ
(D V
1.000
1.100
1.200
1.300
10^/T sample (°K)
Fig. l8
Cesium Condensation Temperature - Sample
Temperature Data Carborundum CARB-I-TEX 700
29
INSULATOR HEAT SINK
SAMPLE HEAT SIKK
Fig. 19
Six Sample Long-Term Cesium PressureSample Temperature Test Assembly
30
TABLE h
Slope of P-T
Curve at 10 torr
For Various Materials
Material
Cesium
Loading
mg/Cs/g
Graphite
Tungsten (1)
Oj-,
10 torr 0^
Slope of P-T
curve at 10 torr
°C/torr
366
639
h.l
350
1031
130U
11.0
U50
866
1139
8.0
550
758
1031
6.k
650
686
959
6.0
150
1053
1326
11.0
250
960
1233
8.7
350
9U2
1215
9.6
U75
795
1068
8.3
~ 1
856
1129
11.1
Liquid Cesium
Pure Carbon FC-50
Temperature at
31
point, 10 torr. The values range from ~ 5 C/torr for a liquid reservoir
to ~ 11 c/torr for carbon loaded to 350 mg Cs/g, graphite loaded to 150
mg Cs/g or a porous tungsten reservoir loaded to ~ 1 mg Cs/g. In both
carbon and graphite the cesixmi pressure becomes more sensitive to temperature as the cesium loading increases.
The question of cesium losses to converter components, residual gases,
or leakage becomes very Important when one contemplates placing material
with a cesium reserve of 10's or 100's of milligrams inside the converter
envelope.
One would prefer a sorption reservoir material which could
tolerate the loss of a large fraction of its reserve cesium before the
vapor pressure dropped much below the operating value.
Using data reported in Ref. 1, the tolerable cesium losses in a variety of materials were compared.
Choosing a point at 10 torr, the amounts
of cesium that could be lost before the vapor pressure dropped by 20^ were
calculated. Thenssults are plotted in Fig. 20 as milligrams cesium loss
3
per cm as a function of sample temperature.
The carbon and graphite samples can tolerate the highest loss, 10 to
•3
900 mg/cm . On the other hand, the tungsten and alumina samples can lose
only 0.02 to 0.2 mg/cm
before the pressure drops 20^. The losses toler-
able in the W-10 Ta sample range down from 12 mg/cm
depending upon sample
temperature.
Tolerable loading losses are not too sensitive to operating temperature except in the two graphite samples, where peaks occur when the transition from one two-phase equilibrium to another occurs. These transitions are sharply defined in the annealed pyrolytlc graphite sample and
less well defined in the CARB-I-TEX 700 sample.
Two methods of loading the reservoir materials with cesium are available. These are:
1. Provision of a cesium excess, the desired loading being achieved
by equilibrating the reservoir material in cesium vapor at the
desired reservoir temperature and cesium pressure.
32
1000
2;
100
rc
PYROLYTIC
GRAPHITE
CARB-I-TEX 700
GRAPHITE
O
(J
<
o
PURE CARBON
FC-50
00
to
a:
a.
en
LLJ
•RANGE OF W-lOTa DATA
n.
o
<
.0 —
o
<
>
CD"
-TUNGSTEN (1 M'^/G SURFACE AREA)
CO
to
O
t/)
LU
tJ
AI203
ALSIMAG S'tS
_L
0.01
500
600
Fig. 20
700
800
SAMPLE TEMPERATURE
900
("C)
1000
Tolerable Cesium Losses as a Function
of Sample Loading Temperature
33
I 100
2.
Provision of the precise amounts of cesium and reservoir material
necessary to achieve the desired loading.
The first method has the advantage of piXDvidlng operation with one
point on the reservoirs cesium pressure-reservoir temperature curve well
known, regardless of variations in reservoir sorption characteristics. It
also does not require an actual knowledge of reservoir material or cesium
quantities used.
However, if operation at a cesium loading allowing the
largest tolerable cesium loss is desired with a graphite reservoir, the
first technique becomes much less attractive since the allowable sample
temperature errors during the loading of a pyrolitic graphite reservoir
are only i 10 C (Fig. 20). Outside of that band the tolerable cesium losses
quickly reach their lowest values. Thus for graphite samples, a loading
using the second method would insure operation at the highest available
tolerable loss levels.
3i^
REFERENCES
Yates, M. K., Fitzpatrick, G. 0., "Cesium Sorption in Materials for
Thermionic Converters, Vo.l I", USAEC Report GA-857^, Gulf General Atomic
Incorporated, 1968.
Breitwieser, R., Nottingham, W. B., "Theoretical Background for Thermionic Conversion Including Space-Charge Theory, Schottky Theory and
the Isothermal Diode Sheath Theory," National Aeronautics and Space
Administration Report, NASA TN D-332U, March I966.
Aronson, S., Salzano, F. J., "Thermodynamic Properties of the CesiumGraphite Lamellar Compounds," J. Chem. Phys., Vol. k3, No. 1, July 1965?
Plij-9.
Levlne, Jules D., "Electrical Conductivity Caused by Adsorbed Cesium
on Insulator Surfaces," Thermionic Conversion Specialist Conference
Proceedings, Cleveland, Ohio, October I96U, PI.
Devin, B., Lesueur, R., Setton, R., "Vapor Pressure of Cesium Above
Graphite Lamellar Compounds," Thermionic Conversion Specialist Conference, Palo Alto, California, October I967, P277.
35