/ /' / : > RRC-0102

Mf ? i *^t;^

J«S^
n i\ ,4

CESIUM-137
POWER
PROGRAM
Facsimile Price $ //__
M i c r o f i l m Price $

A v a i l a b l e from the

QUARTERLY O f f i c e o f Technical Services

Department- o f Commerce
Washington 25, D. C.

REPORT
m privately owned rights or
LEGAL NOTICE -
Th 6 report was prepared aa an account o£ Govemtns t sponsori-d work ^eilce^ the United
States nor the Comonssioii nor any i>sr3on acting on behalf oi the Comnussior
A Makes anj warranty or representation expressed or rapUed with respect to the a ca
racv completeness or usefulness of the InformaUon contained 1 thifi report or thai the use
of anj Information apparatus method or process disclosed ii tMt report mav not infringe

B Aastimes any liabilities wltJb respect to the use of or for damigss resulting !ro«i th"
use of any information aoparatiis method or process disclosed in thl repo £
As used m the above person acting on behalf of the Commisaton inclu'' s any *.&.
ployec or contractor of the Cotrmission cr emplovee of such conlracjsr >o the exlen that
such employee or contractor of the i^aaimission or employee of such contra tor p spares
dlssemi atea or provides access to any Inforioatlon pursuant U his employment o coptract
with the Commission or his employment with such contractor

PREPARED FOR THE

DIVISION OF JSOTOPE DEVELOPMENT
U. S. ATOMIC ENERGY COMMISSION

ROYAL RESEARCH CORPORATION

A SUBSIDIARY OF R O Y A L INDUSTRIES INC.
 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.
 RRC-0102
Special Distribution

THE CESIUM-137 POWER PROGRAM

This work p e r f o r m e d for

The Division of Isotopes Developraent
United S t a t e s Atomic E n e r g y C o m m i s s i o n
C o n t r a c t No. AT(04-3|-366

3rd QUARTERLY REPORT

by

ROYAL > RESEARCH CORPORATION

a s u b s i d i a r y of ROYAL INDUSTRIES, INC.
Hayward, California

^dyOoiuA if J J5 %.A U (If..

Hampden O, B a n k s , J r ,
Project Director

October 3 1 , 1961
 I i: I I I I I
*• ii«(i«i, CO
o
I I U
f I I ¥ r i l l
f- |t oi
iim^li^li^l
 iii

LEGAL NOTICE

This report is an account of Government sponsored work. Neither the

United States, nor the Commission, nor any person acting on behalf of
the Commission:

A. Makes any warranty or representation, expressed or im-

plied, with respect to the accuracy, completeness, or
usefulness of the information contained in this report, or
that the use of any information, apparatus, method or
process disclosed in this report may not infringe upon
privately owned rights; or

B. Assumes any liabilities with respect to the use of, or

for damages resulting from the use of any information,
apparatus, method, or process disclosed in this report.

As used in the above, "person acting on behalf of the Commission," in-

cludes any employee or contractor of the Commission, or employee of
such contractor, to the extent that such employee or contractor of the
Commission, or employee of such contractor prepares, disseminates,
or provides access to, any information pursuant to his employment or
Contract with the Commission, or his employment with such contractor.

RRC-0102
 SPECIAL DISTRIBUTION LIST FOR CESIUM-137 REPORT

In a c c o r d a n c e with Exhibit I, Page 8 of the c o n t r a c t , the q u a r t e r l y

r e p o r t will be d i s t r i b u t e d to the following:

Division of Isotopes Development

U, S. Atomic Energy C o m m i s s i o n
Washington 2 5 , D. C.

Attention: D r . Paul C. A e b e r s o l d , D i r e c t o r
(25 copies)

Technical S e r v i c e s Division
San F r a n c i s c o Operations Office
B e r k e l e y , California

Attention: Robert W. Hug hey, D i r e c t o r

(2 copies)

Oak Ridge National L a b o r a t o r y

P o s t Office Box X
Oak Ridge, T e n n e s s e e

Attention: M r . John H, Gillette, Superintendent

(2 copies)

Lamont Geological O b s e r v a t o r y
of Columbia U n i v e r s i t y
P a l i s a d e s , New York

Attention: D r . John Ewing, A s s i s t a n t D i r e c t o r

(2 copies)

RRC-0i02
 V

CONTENTS
Page
Title Page i
Frontispiece ii
Legal Notice iii
Distribution L i s t iv
Contents , v
Illustrations vi
Tables viii
Foreword ix
Summary x
I. Subtask 1.2 - C e s i u m Connpound Conversion 1
II. Subtask 1. 3 - Cesium Fuel F o r m Investigation . 5
III. Subtask 1.4 - Long T e r m Solubility Testing , 8
IV. Subtask 2. 1 - Heat T r a n s f e r Analysis 19
V. Subtask 2. 2 - Power Conversion Analysis , . 25
VI. Subtask 2. 3 - G e n e r a t o r Design Analysis 39
VII. Subtask 2 . 4 - G e n e r a t o r Shielding Analysis 47
VIII. Subtask 2. 5 - Conceptual Design of G e n e r a t o r . 54
IX. Subtask 2 . 6 - H a z a r d s Evaluation 56
X. T a s k III - Design and Construction of the Model . , 6l
XI. Subtask 4. 1 - F a b r i c a t i o n Analysis 64
XII. Subtask 4 . 2 - Fuel Element F o r m i n g & Cladding 65
Appendix A - C e s i u m - 1 3 7 E n e r g y Calculations . . . . . . . . 71
Appendix B ~ Calculation of T e m p e r a t u r e Distribution in
Uraniunn Shield 75
Appendix C - Shielding Calculations 86

>

RRC-0102
 vi

^ ILLUSTRATIONS
Figure
No. Page
i Compound Conversion With AG 1X8 R e s i n 3
2 Solubility of Cesium P o l y g l a s s in Aqueous Media 11
3 Solubility of I r r a d i a t e d C e s i u m P o l y g l a s s 14
4 P o l y g l a s s Samples Before and After I r r a d i a t i o n . 15
5 S t a i n l e s s - S t e e l Bellows Envelope for Thermocouple Package 21
6 T h e r m a l Conductivity of Tungsten (Sintered) As a Function
of T e n n p e r a t u r e 24
7 T h e r m a l Conductivity of Depleted Uranium (Theoretical
Density) As a Function of T e m p e r a t u r e 24
8 Westinghouse Miniature T h e r m o e l e c t r i c Module . . , , . . . 26
9 T e s t Setup of Miniature T h e r m o e l e c t r i c Module 27
10 T h e r m o e l e c t r i c Module Evaluation Apparatus 32
11 E n l a r g e m e n t of Royal R e s e a r c h Lead Telluride Therniocouple
Design, With Operating T e m p e r a t u r e Profile 34
12 Reduced-Scale P r i n t e d - C i r c u i t B o a r d Used in Life Testing
of Royal T h e r m o c o u p l e s 35
13 T h e r m o c o u p l e A s s e m b l y in I n e r t A t m o s p h e r e E n c l o s u r e . . . 35
14 Royal T h e r m o c o u p l e s A s s e m b l e d on P r i n t e d Circuit Board . . 36
15 Double-Walled Hastelloy C Fuel Can 41
16 Reduced Scale Aluminum T e s t V e s s e l s 42
17 V e s s e l P r e s s u r e T e s t i n g at Naval E l e c t r o n i c s L a b o r a t o r y ,
San Diego, California , 42
18 Imploded T e s t V e s s e l , Showing C r a c k e d Wall and
F r a g m e n t e d Bottom 43
19 E x t e r i o r and I n t e r i o r Views of Lead Cave Used in
Shielding E x p e r i m e n t s 48
20 Attenuation of C e s i u m - 1 3 7 by Tungsten and Lead . . . . . . . 49
21 Attenuation of Cesiuni-137 by Aluminum 50

t
RRC-0102
 Vll

ILLUSTRATIONS (Cont'd)
Figure
No. Page
22 C o r e T e m p e r a t u r e As a Function of E x t e r n a l Insulation . . . . . 59
23 G e n e r a t o r Model, L e s s DC-DC C o n v e r t e r 62
24 High T e m p e r a t u r e F u r n a c e Used for Cesium Polyglass
Fabrication 66
25 E l e c t r i c a l l y Driven C e s i u m - F u e l Mixer 67
26 C e s i u m P o l y g l a s s Flow Sheet 69

•
RRC-0102
 viii

TABLES

P h y s i c a l C h a r a c t e r i s t i c s of the R e v i s e d Fuel Configuration . . . 6

Solubility Data Sheet , . 9
Testing of P r o d u c t s in Chennical s 16
Reduced Scale T h e r m o e l e c t r i c Module Output 29
Anticipated Output of the 5-Watt Conversion System 30
G e n e r a t o r Shell P r e s s u r e T e s t - Reduced Scale . . . . . . . . . 46

RRC-0102
 FOREWORD

This report has been prepared by Royal > Research Corporation,

Hayward, California, as required under U, S. Atomiic Energy Commission
Contract AT(04-3)-366. It is the third quarterly report for the Cesium-137
Power Program and covers the period June 22 to September 22, 1961,

RRC-0102
 X

SUMMARY

The third q u a r t e r of the Cesiumi-137 Power P r o g r a m was devoted to

completion of a definite fuel form and the investigation of those p a r a m e t e r s
which r e g u l a t e a r e p r o d u c i b l e product m a t e r i a l . Such qualities as color of
the polyglass w e r e analyzed insofar as it identified c o n t a m i n a n t s , incomplete
reaction and m a t e r i a l density. The p r e s e n c e of minute surface bubbles w e r e
found to be attributable to a t r a c e of water in the starting constituents. It
was d i s c o v e r e d that preheating m i n i m i z e d this phenomenon. Final removal
of all r e m a i n i n g bubbles could then be accomplished by mechanical m e a n s .

Fuel quality is p r e d i c a t e d upon safety. Toward this end, a s e r i e s of

l o n g - t e r m solubility m e a s u r e m e n t s was continued. Aqueous m e d i a , including
n a t u r a l sea water and distilled w a t e r , w e r e used, with all physical and cheiti-
ical p a r a m e t e r s maintained at a constant. Initial dissolution did occur because
some surface c e s i u m was found to be in an unreacted s t a t e . Following this
dissolution, the r a t e d e c r e a s e d m a r k e d l y . After 100 days i m m e r s i o n , the
total dissolved c e s i u m was in the low p a r t s - p e r - m i l l i o n region, although this
i s purely a function of testing v e s s e l v o l u m e . Weekly i n c r e a s e is in the p a r t s -
p e r - b i l l i o n r a n g e . F o r this work, t r a c e r c e s i u m was added to stable polyglass
since the solubility is well below conventional m e a s u r e m e n t l e v e l s . Fuel
quality, a s a function of r a d i a t i o n d o s a g e , was investigated and found to have
a stabilizing effect that r e t a r d s dissolution. This may be due to co-poly-
m e r i z a t i o n of the complex b o r o s i l i c a t e s .

The amount of C e s i u m - 1 3 7 r e q u i r e d to produce 120 therwial watts was

r e c a l c u l a t e d , employing the m o r e r e c e n t l y accepted half-life for C e s i u m - 1 3 7 ,
of 26. 7 y e a r s . Individual energy contributions from prinnary beta decay was
redefined, based upon a m o r e r i g o r o u s t r e a t m e n t of the irsethod of calculation.
As a r e s u l t of t h e s e r e v i s i o n s in available e n e r g y , a total of 25,500 c u r i e s
will be needed.

A second and m o r e a c c u r a t e shielding experiment was p e r f o r m e d ,

using a lO-curie C e s i u m - 1 3 7 s o u r c e containing 1.5 per cent of C e s i u m - 1 3 4 ,
This s a m p l e was p r e p a r e d by the Oak Ridge National L a b o r a t o r y , and was
made r e p r e s e n t a t i v e of the 2 5 , 5 0 0 - c u r i e batch accumulated for fueling the
5-watt g e n e r a t o r . Half-value l a y e r s for lead agreed with those values
obtained by ORNL r e s e a r c h e r s . Both half and tenth-value l a y e r s w e r e d e t e r -
mined for tungsten, and showed that over 3 in, of that m a t e r i a l would be
needed to r e d u c e the radiation to a level of 100 m r / h r at 4 ft. from the c e n -
t e r of the g e n e r a t o r . Since this would r e s u l t in an unacceptable o v e r - a l l

RRC-0102
 XI

g e n e r a t o r s i z e , depleted u r a n i u m was s e l e c t e d . Uranium has twice the

attenuation qualities a s lead, t h e r e f o r e , not m o r e than 2 in. would be needed.
P r e c i s e shielding e x p e r i m e n t s a r e in p r o g r e s s .

The Westinghouse effort on power conversion has yielded a package

which is approximately h a l f - s c a l e , with an output of 2,75 watts at 0.99 volts.
T h e r m o c o u p l e cladding was m a d e by m e a n s of a poured i n s u l a t o r ; this has
r e s u l t e d in a high heat l o s s through the package and a r e q u i r e d t h e r m a l
wattage of 140 watts to provide the r e q u i r e d 5 watts e l e c t r i c a l .

Royal R e s e a r c h is continuing its backup p r o g r a m and has instituted

life t e s t s on a typical Royal R e s e a r c h configuration. This includes printed
c i r c u i t r y , b r a z e d " n " and " p " e l e m e n t s and a suitable c e r a m i c cladding.
E l e m e n t s a r e then insulated with felted f i b e r g l a s s . The outer container will
be a right c i r c u l a r bellows type of wall with end plates of a highly (thermal)
conductive alloy. Couple r e s i s t a n c e has been reduced to 0.027 o h m s .

The fuel will be doubly encapsulated in two s e p a r a t e c o n t a i n e r s , each

holding t h r e e e l e m e n t s . This was adopted to p e r m i t use of the ORNL r e m o t e
welding a p p a r a t u s . Final dimensioning of the inner fuel can and cladding have
been c o m p l e t e d . Two f u l l - s c a l e 7075-T6 aluminunn g e n e r a t o r shells have
been o r d e r e d . T h e s e will be used for p r e s s u r e testing and air dropping to
deternnine i n t e g r i t y .

Reduced s c a l e models of the g e n e r a t o r shell were tested at the Naval

E l e c t r o n i c s L a b o r a t o r y for r e s i s t a n c e to d e e p - s e a p r e s s u r e . Excellent
r e s u l t s w e r e obtained, indicating that as little as 1,40-in. wall thickness
would afford ample protection against a 6,000-fathom p r e s s u r e . At this s a m e
installation, u n d e r w a t e r e x p o s u r e t e s t s w e r e initiated to d e t e r m i n e a suitable
coating agent to p r o t e c t the aluminum from s e a w a t e r attack. Four s a m p l e s ,
each of different cladding, w e r e i m m e r s e d in running s e a w a t e r , to be left
for 90 d a y s . Interim inspection will be made by nneans of closed circuit
television provided at the offshore s i t e .

Two acceptable d c - d c c o n v e r t e r s have been defined and a r e being

a s s e m b l e d by E l e c t r o s y s t e m s Co, One is a solid state device of g r e a t e r
than 85 p e r cent efficiency but with a longevity a s s u r e d for approxiniately
10,000 h o u r s . This device i s not position s e n s i t i v e . The second will operate
at from an e s t i m a t e d 85 to 92 p e r cent efficiency, has a component longevity
well in e x c e s s of 30,000 h o u r s and m u s t be o p e r a t e d within ^ 10 deg. of v e r -
t i c a l . Since both devices r e p r e s e n t a b r e a k t h r o u g h in longevity, no decision
will be made at this t i m e a s to which will be used in the g e n e r a t o r package.

RRC-0102
 xii

The basic information which has evolved from the various tasks of this
program has made possible the dimensioning of the main components com-
prising the 5-watt generator. From this information, a full-scale model
was designed and fabricated. Although inoperable, this model was sufficient-
ly detailed to permit inspection of all major parts, as well as indicate their
relative size and position in relation to their function. The model consisted
of three sections: the generator body, the dc-dc converter cavity and the
lid. The model was delivered to the customer in a carrying case, together
with color slides and descriptive literature.

At the convenience of the customer, the site selected for fuel fabrica-
tion and cladding has been changed irom the Vallecitos Atomic Laboratory,
California, to Oak Ridge National Laboratory. Necessary hot cell equipment
required to fabricate the fuel elenients is under design by the contractor and
will be delivered, subject to approval by the fabricating facility.

RRC-0102
 1

I. SUBTASK 1.2 - CESIUM COMPOUND CONVERSION

Research into the methodology of converting starting cesium chloride
(in a dry state) to aqueous cesium carbonate was completed during the third
quarter of the program. Evaluation of contaminants in a typical sample of
starting Cesium-137 fuel material has shown that no appreciable dissimilar
elements are present as radioisotopes. For this reason the prelinninary
scrubbing step, which had been assumed necessary, may be omiitted fro3:n
the process.

Oak Ridge National Laboratory has provided the project with a 10-curie
sample, representative of the 25,500-curie batch being reserved for fueling
the operational generator, A small aliquot of this material was extracted,
diluted and counted, on both the end window Geiger-Mueller counter and the
liquid-well counter. No short-lived constituents were identified other than
the Cesium-134 component, A sample was then assayed with the 256-chanKel
pulse-height analyzer. No gamma peaks attributable to radioactive con-
taminants were seen above the photoelectric region. This sample contained
a reported 1.5 per cent Cesium-134,

Because of the relative purity of the above sample, the anticipated use
of ammoniunn phospho-molybdate (in a column) for decontaminating the fuel
material, has been obviated. In addition to reducing the number of steps
required for the overall cesium polyglass fabrication, the processing time
also has been reduced. Instead of the fuel entering the compound conversion
column diluted to 1. 68 liters and then coming off as the carbonate with a
volume of 13,4 l i t e r s , only 200 ml, of chloride are introduced, and 570 ml,
are collected as the carbonate. This will result in smaller vessels and a
tremendous saving of time, when evaporating the carbonate solution down to
the desired 200 ml. prior to mixing it with the stable borosilicates, Perpass
time through the compound conversion column has been reduced, as well.
However, information gained as a result of considering the presence of a
contaminant is useful for future fuel processing, particularly if the charge of
a future generator is found to be radiocontaminated.

Changes in the physical dimensioning of the generator fuel array, as

well as a reduction of Cesium-137 needed to provide the necessary 120 thermal
watts, has resulted in a change in cesium processing volumes, equipment and
running time. Because of restrictive dimensions encountered in the cladding
welder equipment, a single, doubly-encapsulated Hastelloy C configuration
cannot be used. Instead, two smaller doubly-encapsulated cans must be em-
ployed. In so doing, an even number of equally-filled platinum cups must be
fabricated, instead of seven, as originally planned. For this reason, the
numiber of cups has been reduced to six, and this makes mandatory a greater
Cesium-137 content per cup. This, in turn, has required a reduction in the

RRC-0102
 2

number of i n c r e m e n t a l p r o c e s s c o l u m n s , although resulting in i n c r e a s e d

total volume. T h e r e f o r e , each compound conversion column will contain
1100 m l . of ion exchange r e s i n . The column d i a m e t e r will be 4 . 8 4 c m . and
contain a r e s i n height of 59. 6 cnn. This m a t e r i a l r e p r e s e n t s a 10 per cent
e x c e s s amount of r e s i n (as a safety factor) and is capable of converting l69 g,
of c e s i u m c h l o r i d e . T e s t s have shown that such a colunnn has an efficiency
of from 97 to +99 per cent, with a column retention of <0, 5 per cent after
washing. To p r o c e s s six fuel i n c r e m e n t s , six such colunnns would be r e -
quired. This approach i s s u p e r i o r to r e p e a t e d r e c h a r g i n g of a single ion
exchange column in that it s a v e s t i m e , r e d u c e s the possibility of channeling
and g u a r d s against the p a s s - t h r o u g h of u n r e a c t e d cesium c h l o r i d e . The
columns will be m e d i u m - w a l l e d pyrex with a c o a r s e frit forming the bottom
of each. A 3-in. -long, t a p e r e d stopcock will be incorporated in the column,
positioned below the frit. I n t e r i o r b o r e on the d i s c h a r g e end will be 4 m m ,
in d i a m e t e r ,

Dowex 1X8 anion-exchange r e s i n has been replaced by the m o r e uniform

and m o r e efficient type - - AG 1X8. The l a t t e r is simply a m o r e refined form
of Dowex 1 which has an i n c r e a s e d exchange capacity of 15 per cent. The
AG 1X8 has a r e p o r t e d value of 1.4 m e q / m l . of moist r e s i n . This m a t e r i a l
is provided, a s the c a r b o n a t e , by the BioRad L a b o r a t o r i e s , Richnnond, Calif-
ornia.

The p r o c e d u r e which h a s been found m o s t suitable for fuel p r o c e s s i n g

is a s follows. The e n t i r e c h a r g e for one i n c r e m e n t (l69 g. c e s i u m chloride)
is dissolved in 200 ml, of w a t e r . Occasionally, a gentle warming has been
found n e c e s s a r y . This then, is fed into the conversion column from the
d i s s o l v e r . B e c a u s e of dimensional c h a r a c t e r i s t i c s of the column d e s c r i b e d
e a r l i e r in this r e p o r t , the volume would r e s u l t in an 11 c m , head above the
r e s i n . Wash water collected in the d i s s o l v e r is not added at this t i m e .
Instead, the original solution is p a s s e d through the colunan at a r a t e of
8, 5 m l , p e r m i n u t e , until the columin head has been reduced to 1 c m . above
the r e s i n . The wash water is then (introduced at the top of the column) added
to maintain a 1-5 c m . head. F i g . 1 shows a typical elution curve obtained
from a full s c a l e r u n . T r a c e r C e s i u m - 1 3 7 was employed as the elution
i n d i c a t o r . A total of 570 nal. of solution was p a s s e d through - - 200 m l . of
original c e s i u m chloride solution and 370 m l . of wash w a t e r . After elution,
the r e s i n was analyzed for r e s i d u a l c e s i u m . Sensitive counting showed that
only 0 , 0 1 per cent of the cesiuam was r e t a i n e d . This would amount to l e s s
than 15 m g , of r a d i o c e s i u m lost during a hot r u n , a negligible n u m b e r .

RRC-0102
 9

>
8
o x^ry^
o ^-0-
9i
u ^ c
0) o

7
K
/;
/
3 1
•S D
I
/) 169 g r . CsCl in 200 m l .
/ solution passed through
5 column - CS2CO2 eluted
/
C
(D
3 / :>
a /
w

s3 3
/?
'M

0) /
O 2
/
O
«
s
1
/
o
0 6 i2 18 24 3 0 36 42 4 8 54 6 0 66 72 78 84
I x^

Total Elapsed Elution T i m e in Minutes

Fig. 1 - Compound Conversion vVith AG 1X8 Resin

RRC-0102
 4

As indicated by the curve in Fig. 1, the break-point is clean and well

defined. The rise time is as narrow as could be expected from any resin.
It is flattened through the use of a concentrated cesium chloride solution.
Flattening is necessary so that the steep drop-off can be more readily anti-
cipated and better control exercised. While a more concentrated starting
solution would provide a smaller product volume (classical data suggests
a much higher solubility in water),it has been our experience that such is
not possible.

Control of the endpoint during hot cell operations may be achieved by

means of a shielded counter tube having a high saturation point. This could
be positioned perpendicular to the column and would identify the cut-off in
radiocesium passage through the resin.

The carbonate form of AG 1X8, as supplied by the BioRad Laboratory,

will be made available to Oak Ridge National Laboratory for use in their
processing of the cesium fuel.

RRC-0102
 5

n . SUBTASK 1 . 3 - CESIUM FUEL FORM INVESTIGATION

This subtask is concerned not only with cesium polyglass as a product

material of certain chemical and physical characteristics, but with heat
producing capabilities of the final product as well. A suitable hot cell scale-
up of laboratory techniques involved in the manufacture of cesium polyglass,
has also been included. Evolution of specific pieces of apparatus which
becoine identified for fabrication are discussed in Task IV.

Originally, 28,000 curies were identified as being needed to provide

the 120 thermal watts of energy for the cesium generator. As described in
the first quarterly progress report, this was predicated upon calculations
which assumed the useful beta energy as 1/3 the total energy. A more
rigorous examination has recently been made of beta energies available from
Cesium-137 decay chain, resulting in certain revisions. It was shown that
the 514 Kev. beta has a useful energy of 183 Kev. and the 1, 176 Kev. beta
has a useful energy of 283 Kev, based upon relative abundance of 92 and 8
per cent, respectively. These values are derived in Appendix A. As before,
no contribution to thermal energy has been derived for the Cesium-134 addi-
tion to the calculated 137-curie strength. In resolving the generator fuel
requirement, the more recent value of 26.7 y r s . for Cesium-137 half-life
was used. As a consequence, the customer has been advised that 25,500
curies of radioisotope will be sufficient to provide the required 120 thermal
watts.

The reduction in fuel quantity has resulted in a dimensional revision

of the fuel core. Further changes were dictated by the Oak Ridge National
Laboratory requirement that the double encapsulants fit in the existing hot
cell welding enclosure. Therefore, the decision was made to fabricate the
25,500 curies in the form of six increments, instead of the previously
reported seven. While the dixnensions of each platinum fuel cup will be
reduced in height, the 2. 50-in.-dia, has been retained. This is being done
partly to conform to external design commitments already made, to optimize
self-absorption and thereby reduce external shielding, and lastly, to keep the
linear heat path through the fuel cans to a minimum. If a narrower fuel can
and shield were employed, the previously determined thermal insulability
would be invalidated, it being more difficult to force heat through a long,
narrow cylinder. By reducing the fuel load and retaining the original
diameter, essentially we have foreshortened the fuel-configuration height-.and
the shield height, thus resulting in improvement of thermal economy.

The following characteristics, based upon final revision of fuel incre-

ment dimensions, are descriptive of the fuel package:

RRC-0102
 TABLE I

Physical Characteristics of the Revised Fuel Configuration

Pre-Fused Mix Hred Product

1,874 Weight (grams) 1 687

1.53 - .02 Density (grams per cc.) 3,00 + ,02
1,230 Volume (cubic centimeters) 563
26, 100 Activity (curies) 25 500
Diameter (centimeters) 6.35
Increments 6
Increment Height (centimeters) 3,44

These changes were brought about, in part, by a revision upward (by

Oak Ridge National Laboratory) in specific activity of the starting fuel material.
Originally, they had estimated a specific activity of 25 curies per gram. How-
ever, the strength of the 10-curie representative source, as provided for the
shielding experiment, showed an increase of 5,62 per cent, or 26, 7 curies/g.
Further consideration of specific activity of the fuel is unnecessary; it will be
assumed that a mechanical mixing-drying device will not affect product den-
sity, or specific activity. Physical loss of fuel will be kept to a 2 per cent
maximum.

Repeated use of the false fuel-cup wall, as provided by the dental

porcelain, has demonstrated its reliability. It is strong enough to insure
against accidental breakage from handling, yet brittle enough to permit later
reinoval with suitable tools. Its intrinsic non-wetting qualities insure against
fuel-material creep upward on the cup wall. Samples of a fired false wall,
starting powder and complete fabricating instructions have been provided ORNL
representatives for their evaluation. The alternative to this method would be
a full-height platinum cup. However, if the latter method is used, several
problems must be overconne. First of all, molten cesium polyglass has a
tendency to "wet" the inner wall and then creep up it. Furthermore, removal
of the excess platinum can be achieved only through the use of a lateral saw,
if rim irregularities a r e to be avoided.

RRC 0102
 7

The p r e s e n c e of tiny air bubbles in the upper 1/8 in. of fuel i n c r e m e n t

surface has been i n v e s t i g a t e d , in an effort to explain their p r e s e n c e and
seek a method of removing t h e m , if p r a c t i c a l . Bubbling, in g l a s s , has been
a major p r o b l e m a s s o c i a t e d with g l a s s m a n u f a c t u r e . These bubbles a r e
caused by (1) water v a p o r , (2) decomposition of i m p u r i t i e s , and (3) internal
volatilization. T h e r e has been no convenient method found, to d a t e , for the
rapid r e m o v a l of such b u b b l e s , or mixing g l a s s without this side effect o c c u r -
r i n g . In the c a s e of c e s i u m p o l y g l a s s , the second cause can be omitted since
pure constituents a r e used. However, the e x t r e m e l y high hygroscopicity of
c e s i u m c a r b o n a t e m a k e s complete water r e m o v a l virtually i m p o s s i b l e . When
this was a t t e m p t e d , through evacuated heating f u r n a c e s , some cesium c a r -
bonate decomposed to form c e s i u m oxide, which readily volatilizes unless
it can be m a d e to r e a c t with a b o r o s i l i c a t e . Internal volatilization of cesium
oxide does occur but b e c a u s e of its high reactivity with the b o r o s i l i c a t e s and
an e x c e s s of silica p r e s e n t in the c h a r g e , this would be of a m o m e n t a r y
nature.

E x t r e m e c a r e e x e r c i s e d in the p o s t - d r y i n g stage (400 deg. F . ) does

evaporate a l m o s t all of the m o i s t u r e . When fusion o c c u r s , m o i s t u r e bubbles
form and then m i g r a t e upward through the viscous m a t r i x . Surface tension
inhibits total e s c a p e of the l o w e r m o s t o n e s . T h e r e f o r e , after the p r o p e r
fusion p e r i o d , a l / 8 - i n , - d e e p l a y e r usually contains tiny bubbles.

Surface volatilization of c e s i u m oxide has been minimized to where it

accounts for a l e s s than 2 per cent l o s s . However, it can not be omitted
comipletely. E a r l i e r r e s e a r c h e r s , who endeavored to entrain cesium in a
vitreous m a t r i x , w e r e unsuccessful b e c a u s e they sought thin-film s o u r c e s .
No fused head was p r e s e n t to prevent the high l o s s of c e s i u m , as is true with
c e s i u m polyglass i n c r e m e n t s .

As explained in preceding s u b t a s k s , bubbles have been p r e s e n t in

s a m p l e s which w e r e subjected to l o n g - t e r m solubility d e t e r m i n a t i o n s , the
i r r a d i a t e d sample test and all fuel-product evaluation t e s t s . Chemical and
physical p r o p e r t i e s a s s o c i a t e d with c e s i u m polyglass t o l e r a t e the p r e s e n c e
of bubbles or v a c u o l e s , which a r e only 1 mil in d i a m e t e r and approximately
10 m i l s a p a r t .

RRC-0102
 8

>
III. SUBTASK 1.4 - LONG TERM SOLUBILITY TESTING

A p r e c i s e d e t e r m i n a t i o n of the solubility of cesium polyglass in various

aqueous media is p e r h a p s the m o s t imiportant single test being conducted on
the p r o g r a m . T h u s , the destined subsea use of the five-watt g e n e r a t o r , makes
contact with s e a water far m o r e p r o b a b l e than any other credible incident.
C o r r o s i o n of the outer m e t a l l i c components could proceed independently of a
physical accident which might r u p t u r e the g e n e r a t o r . T h e r e f o r e , in addition
to the evolution of a s t r u c t u r a l a r r a y to p r o t e c t the fuel c o r e from i m p a c t ,
fire or v a n d a l i s m , a c h e m i c a l compound was sought which would be inert to
n a t u r a l aqueous m e d i a . C e s i u m polyglass exhibited acceptable r e s i s t a n c e to
both distilled and n a t u r a l sea water during e a r l y t e s t i n g . As the i m m e r s i o n
t i m e extended to m o n t h s , this fact was substantially confirmed. Testing will
continue throughout the duration of the c o n t r a c t , in o r d e r to make possible the
m o s t a c c u r a t e extrapolation of u l t i m a t e dissolving t i m e .

Although solubility i s r e p o r t e d a s p a r t s p e r million of aqueous m e d i u m ,

this t e r m i n o l o g y does not reflect the unusually low solubility of cesium poly-
g l a s s . P h y s i c a l r e s t r i c t i o n s of l a b o r a t o r y space prevented controlled testing
in l a r g e r v e s s e l s ; and the use of g r a n u l a r s a m p l e s prevented m e a s u r e m e n t of
surface a r e a exposed to the testing m e d i u m . This is the t r u e p a r a m e t e r of
m e a s u r e m e n t , since dissolution i s a surface r e a c t i o n . Dissolution has p r o -
ceeded so slowly, that the 13 l i t e r v e s s e l used actually r e p r e s e n t e d an
infinite volume to the polyglass s p e c i m e n s . Although spiked with a relatively
high specific activity of C e s i u m - 1 3 7 , the amount found in i n t e r i m aliquots
has r e p r e s e n t e d the lowest limit of counting efficiency, and r e s u l t s a r e most
probably an o r d e r of magnitude higher than they should b e .

A distilled water e x p e r i m e n t consisting of two independent 13-liter

v e s s e l s , each containing speciinens of c e s i u m p o l y g l a s s , was run for a
period of lOO d a y s . During this t i i n e , the volume was kept constant, t e m p e r a -
t u r e was maintained at 50 deg. C. and the w a t e r constantly agitated, Sainples
w e r e drawn every 5 days and subjected, in duplicate, to end-window beta
counting of mounted evaporated s a m p l e s , and s c i n t i l l a t i o n - c r y s t a l g a m m a
counting of liquid s a m p l e s . R e s u l t s of this e x p e r i m e n t a r e concluded in
Table 2. H e r e , uniformity of i n c r e a s e obviates r a n d o m - t y p e r e s u l t s . During
the l a s t 50 days of testing an a v e r a g e i n c r e a s e of l e s s than 2 ppm. was observed
for both r u n s . A plot of t h e s e r e s u l t s , which includes e a r l i e r testing for c o m -
p a r i s o n , m a y be s e e n in F i g , 2. Although physical testing has been t e r m i n a t e d ,
an acceptable extrapolation is r e a d i l y p o s s i b l e , suggesting a d i s s o l u t i o n - r a t e
of 0.095 g. of c e s i u m polyglass per y e a r . N a t u r a l l y , it is understood that
dissolution studies conducted in distilled w a t e r have only a c a d e m i c value.

t
RRC-0102
 TABLE 2

Solubility Data Sheet

Elapsed Time, A vg.. ppm C
Css in
in
Days Solvent (13 L,) Solvent

50 8.8 Distilled Water

55 8.8
60 9.0
65 9, 1
70 9.2
75 9.2
80 9.7
85 9.7
90 9.8
95 9.8
100 9.9

50 10,0 Distilled Water

55 10, 1
60 10. 1
65 10. 1
70 11.4
75 11,3
80 11.5
85 11.7
90 11,8
96 11,8
100 11.9

50 12. 1 Sea Water

55 12,3
60 12,4
65 12.6
70 12„6
75 12,8
80 12.9
85 12.9
90 12.9
95 13.2
100 13.3
105 13, 1
110 13. 1
115 13,4
120 13,4

RRC-0102
 10

TABLE 2 (Continued)

Solubility Data Sheet

Elapsed Tinne, Avg,. ppm Cs in

Sample Weight Days Solv ent (13 L . ) Solvent

4a 15.2g, 50 12.7 Sea Water

4b 55 13.0
60 13.0
65 13.2
70 13.5
75 13.6
80 13,8
85 13.8
90 13.9
95 13,9
100 14.1
105 14,3
110 14,3
115 14.3
120 14,5

RRC-0102
 I ig. 2 - Solubility of C e s i u m P o l y g l a s s in Aqueous Media

to

Sample IV
14
,—- Sample HI
^
12 Sample II
...^
--"'"
Extrapolat ed
>'
. / .-—- . ^ Sample 1
P 10 K .
H
^ CO
^ • ^
^ ^ —
T) / ^
fs 0 / =;^^i:ll
X
9
1 H >5
o -^"'^
o %
H«a
ro !-«*
H*
o1:5
6 -X—j.
/
/
qolvfnts
Distill 5d Vrate- - [ anjd II
A.

i
/ Sea ^VMer - IIJ an<nv

f
2

C) 10 20 30 40 50 60 70 QO 90 100 II 0 120
Days in Solution
 12

At the conclusion of t e s t i n g , the 13 l i t e r contents of both fresh water

e x p e r i m e n t s w e r e evaporated down to 100 m l , with gentle w a r m i n g , A 3 m l ,
aliquot from each was taken and evaporated to d r y n e s s for counting p u r p o s e s .
R e s u l t s indicated that total c e s i u m dissolved during the testing period was
only 67 per cent of the amount r e p o r t e d at the end of 100 days, Fromi this
then, it nnay be a s s u m e d that all i n t e r i m values r e p o r t e d w e r e also in e r r o r
by a like amount.

Of i n t e r e s t is the method of i n t e r i m aliquoting ennployed on these and

all other solubility m e a s u r e m e n t s a m p l e s . The p r o c e d u r e is as follows:
E v e r y 5 d a y s , a 1-liter portion of the original solution is drawn and evap-
orated down to 100 m l . F r o m t h i s , 2 - m l . aliquots a r e taken for counting
p u r p o s e s . At the end of each a n a l y s i s , all liquid i s collected, r e - d i l u t e d
to 1 liter and r e t u r n e d to the solubility t e s t v e s s e l . In this m a n n e r , no
accumulated c e s i u m is lost from the testing m e d i u m .

The second s e r i e s of solubility studies under evaluation is natural sea

water t e s t i n g . This is p e r h a p s the inost important t e s t , since the g e n e r a t o r ' s
initial application r e q u i r e s deep s e a i m m e r s i o n . T h e r e f o r e , attack by s e a
water is a m o r e probable h a z a r d . The third q u a r t e r period saw a continuation
of t e s t s throughout the 120-day p e r i o d . As in the c a s e of the distilled water
p h a s e , i n t e r i m s a m p l e s w e r e drawn every 5 d a y s . A tabulation of t h e s e data
may be seen in Table 2. Confirmation between the beta and gamma counting
techniques is good, and the gradual i n c r e a s e indicated is uniform. The net
i n c r e a s e in dissolved c e s i u m for the past 50 days r e p r e s e n t s only 1 ppm.
F r o m the 15-g, s a m p l e , this i s only 95 m g , per y e a r of dissolved cesium
p o l y g l a s s . Since c e s i u m r e p r e s e n t s only 50 per cent of it by weight, over
t h r e e c e n t u r i e s would be r e q u i r e d to d i s s o l v e the 15-g. s p e c i m e n s .

A graphic i n t e r p r e t a t i o n of dissolution, a s a function of t i m e , may be

seen in F i g . 2. Weekly buildup in the 13 l i t e r testing solution is of the
o r d e r of p a r t s - p e r - b i l l i o n , a s in the c a s e of the distilled-water e x p e r i m e n t .
It is i n t e r e s t i n g to note that the solubility i n c r e a s e in a s e a water medium is
only 35 per cent g r e a t e r than in distilled w a t e r .

Upon conclusion of distilled water t e s t i n g , the equipment was thoroughly

cleaned, w i p e - t e s t e d for r e s i d u a l radioactivity and p r e p a r e d for the f r e s h
water t e s t s . Although p r e s e n t c o n s i d e r a t i o n s do not include f r e s h - w a t e r
bodies as locations for isotopic powered g e n e r a t o r s , such t e s t s still a r e in
o r d e r since t h e r e i s a possibility of an accidental drop in a lake or r i v e r .
Such impact could r u p t u r e the e x t e r n a l c a s e under special c i r c u m s t a n c e s ,
as when n e a r - s u r f a c e r e e f s a r e in the impact a r e a . T h e r e f o r e , running
long t e r m solubility d e t e r m i n a t i o n s of c e s i u m polyglass in this aqueous
mediuna would be m o s t useful in the h a z a r d s evaluation. Should future appli-
cation extend to g e n e r a t o r placement in f r e s h w a t e r , these studies would b e -
come ever m o r e v a l u a b l e .
RRC-0102
 13

Potable water was obtained, with m i n e r a l content, pH and organic

count a p p r o x i m a t e l y that of l a r g e f r e s h - w a t e r b o d i e s , r a t h e r than that which
is typical of any one local a r e a . It was found during previous testing that
when polyglass is fabricated, t h e r e is a thin surface filim of unreacted cesium
r e m a i n i n g , or deposited upon the product after firing. This cesium dissolves
off the s p e c i m e n quite r a p i d l y , and in doing s o , suggests a g r e a t e r than
actual dissolution r a t e . Examination of the distilled and sea water solubility
r u n s i l l u s t r a t e s this fact. Since a superficial cleaning action is feasible in
the fuel fabricating c e l l , this step was introduced p r i o r to i m m e r s i o n .

After 5 days of i m m e r s i o n , no activity was found in the aliquot s a m p l e s

drawn from the f r e s h - w a t e r t e s t i n g v e s s e l s . T h i s , is in c o m p a r i s o n to the
3 ppm, r e p o r t e d at the end of a s i m i l a r period for both previous t e s t s .
Since the r a t e of dissolution in f r e s h water should be between distilled and
sea w a t e r , the absence of e a r l y dissolving action m u s t be attributed to the
added cleaning s t e p . T h i s should r e s u l t in a curve whose g e n e r a l shape
r e s e m b l e s that of the s e a and distilled water t e s t s , but which will level off
below e i t h e r of t h e i r m a x i m a .

The foregoing solubility t e s t s w e r e conducted for the purpose of d e t e r -

mining the absolute dissolution r a t e of c e s i u m polyglass in various aqueous
m e d i a . Radioactivity was i n t r o d u c e d , in o r d e r to m e a s u r e cesium in amounts
not detectable by conventional m e t h o d s . However, the six fuel i n c r e m e n t s
will contain a total of 25,000 c u r i e s of r a d i o c e s i u m . The intense radiation
field to which the fuel would be constantly subjected could conceivably change
the r a t e of dissolution, after sufficient dosage had been r e c e i v e d . In o r d e r
to r e s o l v e this question, s e v e r a l s a m p l e s of polyglass w e r e p r e p a r e d . They
contained sufficient C e s i u m - 1 3 7 to p e r m i t detection of even a t r a c e of d i s s o -
lution. T h e s e s a m p l e s w e r e then subjected to a p r e l i m i n a r y i m m e r s i o n in
distilled water (at room t e m p e r a t u r e ) for a period of 15 d a y s , as indicated
in Fig, 3 . Dissolution p r o g r e s s e d to 4 . 5 and 3,7 ppm, r e s p e c t i v e l y , for
the two s a m p l e s . The f o r m e r weighed Z.l g. and the l a t t e r , 1, 8g, , size
being kept to a miinimum in o r d e r to a c c o m m o d a t e the s a m p l e s in the isotope
cans provided for i r r a d i a t i o n .

After establishing n o n - i r r a d i a t e d dissolution r a t e , the s a m p l e s w e r e

subjected to a g a m m a flux of 1, 7 x 10 r / h r for a period of 10 d a y s . This
amounted to a m a s s i v e dose of 4 x 10'^ r r e c e i v e d by each specimen. P o s t -
i r r a d i a t i o n examination showed p e r c e p t i b l e darkening of the p o l y g l a s s , as
i l l u s t r a t e d in F i g , 4 . No powdering was o b s e r v e d on the s u r f a c e , and no
fracturing had taken place during i r r a d i a t i o n . When subjected to sudden j a r -
r i n g , no disintegration o c c u r r e d . In g e n e r a l , no physical change was
detected, other than an optical one.

RRC-0102
 14

Fig. 3 - Solubility of Irradiated Cesium Polyglass

u
C)
OH
m
u
(T3

6 12 18
Immersion Time (Days)

RRC-0102
 15

•»'t*,
'^?ti
•%».

F i g . 4 - P o l y g l a s s Samples Before and After I r r a d i a t i o n

(Samiple size was limited by s m a l l isotope can
shown in photo above)

14-

NON-IRRADIATED IRRADIATED

RRC-0102
 16

The fuel c o r e will be effective only if it dumps internal heat promptly. The
color transfornnation i s not d e t r i m e n t a l to this t h e r m a l t r a n s f e r since the
polyglass will t r a n s m i t heat to the platinum cup by conduction. The heat
r e f e r r e d to in this i n s t a n c e , is p r i m a r i l y b e t a - p a r t i c l e energy conversion
h e a t , as g a m m a - r a y energy u n d e r g o e s conversion in the radiation shield.

When the i r r a d i a t e d specijnens w e r e i m m e r s e d in containers of new

distilled w a t e r , the dissolution r a t e was appreciably l o w e r , as shown in
F i g , 3 . Since both s a m p l e s had been given a p r e l i m i n a r y rinsing before
initial i m m e r s i o n studies began, the lower r a t e observed after i r r a d i a t i o n
m u s t be a t t r i b u t e d to this t r e a t m e n t . One possible explanation is that
r a d i a t i o n - i n d u c e d c r o s s - p o l y m e r i z a t i o n o c c u r r e d in the g l a s s s t r u c t u r e ,
further tying up the cesiumi p r e s e n t and linking it to m o r e durable bonds.
A second c o n s i d e r a t i o n is that a t e m p e r i n g action o c c u r r e d , making the
polyglass s t r u c t u r a l l y h a r d e r , so that it better r e s i s t e d the e r o s i v e action
of the swirling w a t e r . This l a s t effect would add to t r u e dissolution by
causing a t e m p o r a r y suspension of minute p a r t i c l e s , whose high s u r f a c e -
t o - v o l u m e would be m o r e v u l n e r a b l e to attack.

In o r d e r to m o r e c l e a r l y define the chemical p r o p e r t i e s of the new

c e s i u m compound which has evolved a s a r e s u l t of the p r o g r a m , a s e r i e s
of spiked polyglass s a m p l e s w e r e subjected to the action of v a r i o u s common
r e a g e n t s . Table 3 shows the effect of 0. ION solutions of n i t r i c acid,
h y d r o c h l o r i c acid, sulfuric acid and sodium hydroxide. It is interesting to
note that n i t r i c acid proved to be the m o s t c o r r o s i v e , while sodium hydroxide
had the l e a s t effect. This m a y , in p a r t , explain the fact that sea water did
not exhibit a p a r t i c u l a r affinity for c e s i u m a s might be expected, since sea
w a t e r also has a slight alkalinity.

TABLE 3
 17
\
In reviewing the r e l a t i v e action of distilled, fresh and natural sea water
upon c e s i u m complex b o r o s i l i c a t e , it b e c o m e s increasingly obvious that the
Debye-Huckel T h e o r y of "uncommon ion effect" does hot apply. If cesium
l o s s by c o r r o s i v e action of an aqueous medium w e r e the only factor involved,
then the l o s s would be g r e a t e r by many o r d e r s of magnitude. The e x t r e m e l y
low solubility of cesiunn polyglass in any aqueous medium (with or without
s a l t s dissolved in it) suggests that the silica and s i l i c a t e s may be the d e t e r -
mining f a c t o r . Dissolution, then, would be a t w o - s t a g e p r o c e s s . F i r s t the
silica linkage must be s e v e r e d , leaving c e s i u m l e s s protected than b e f o r e .
Next, the cesiuirj b o r o s i l i c a t e may be reduced to a s i m p l e peninsular shape.
T h i s , in t u r n may be f r a c t u r e d by the swirling water action, suspending tiny
fragments of c e s i u m - c o n t a i n i n g m a t e r i a l in the testing m e d i u m . Orderly
i n c r e a s e s in cesiunn content preclude a g r o s s random effect. Since so little
is known of the n a t u r e of g l a s s e s , an i m m e d i a t e solution and explanation is
not p o s s i b l e . However, r e p r o d u c i b l e r e s u l t s substantiate the claim that the
action i s p r e d i c t a b l e and the e x p e r i m e n t a l l y obtained dissolution r a t e is
reliable.

T h e s e t e s t s have all been conducted with s a m p l e s of polyglass containing

surface b u b b l e s . T h e s e bubbles a r e of the o r d e r of magnitude of 1 m i l , in
d i a m e t e r . They occur during p r o c e s s i n g of the polyglass and in no way i n -
fluence solubility of the product. While they can be r e m o v e d , either by
mechanical s t i r r i n g , or extended t i m e in a molten s t a t e , the f o r m e t involves
a difficult naanipulator s t e p , while the l a t t e r will cause a slight i n c r e a s e in
c e s i u m l o s s . Since excellent insolubility has been d e m o n s t r a t e d with s a m p l e s
containing the b u b b l e s , and since they a r e too s m a l l and too localized to
affect heat flow, they cannot be c o n s i d e r e d injurious to the fuel. No m e a s u r -
able density change has ever been detected between s a m p l e s taken from the
pellet surface (containing the bubbles) and those taken from the middle or
b a s e of the pellet. T h e r e f o r e , it is strongly felt that they need not be r e m o v e d ,

A c o r r o s i o n p r o b l e m which m o r e closely r e l a t e s to g e n e r a t o r longevity

r a t h e r than fuel containment, is the r e s i s t i v i t y of the outer s h e l l . F o r the
sake of c o m p a c t n e s s and saving of weight, 7075 T 6 aluminum has been
selected for the p r e s s u r e r e s i s t a n t outer envelope. Since this m a t e r i a l is
subject to attack by sea w a t e r , a m e a n s was sought to provide a thin film
b a r r i e r between the aluminum and the ocean. S e v e r a l t e s t blocks of alloy
w e r e coated with different m a t e r i a l s . Each was subjected to a p r e s s u r e depth
of 3,600 fathoms (9j500 psi) in sea water by nneans of a c o m p r e s s i o n c h a m b e r ,
Phenoline-300 failed, in that water seeped under the film, causing it to lift
away fromi the a l u m i n u m . The t h r e e r e m a i n i n g t e s t blocks which s u c c e s s -
fully withstood the p r e s s u r e t e s t , w e r e then subjected to refluxing boiling sea
water for a period of 24 h r s . Having withstood t h i s , they w e r e then taken down

RRC-0102
 18

to the Naval Electronics Laboratory, San Diego, California, and there put
into running sea water at a recommended depth of 15 ft. for 90 days. Interim
reports from the offshore station indicate, that after 60 days numerous col-
onies of barnacles and other marine life have collected on the surfaces of
each one. This prevents underwater television from examining the surfaces
for failure. At the end of the 90 day period they will be retrieved and examined.

RRC-0102
 19

I¥, SUBTASK 2,1 - HEAT TRANSFER ANALYSIS

Heat transfer experiments perforiBed earlier in the program and reported
in the second quarterly report, have identified the optimal thermal insulants
for a 5-watt device. Based upon the then defined fuel-can dimensions, a
shield height of 9,5 in, was used. Results indicated that, in addition to the
double reflector configuration, an additional 3/4 in, of compacted AA Felted
Fiberglass would be necessary. This combination of insulators would waste
only 21 theriBal watts in maintaining an upper shield temperature of 500 deg.
C. Such loss represents the passage of 0,10 8 thermal watts per square
inch. However, it must be remembered that the experimental configuration
consisted of an insulating envelope completely surrounding the shield.
Therefore, the above number actually represents heat loss per unit area
of insulation. Additional heat was also lost through the thermocouple shield
and leads, as well as the electrical heater leads.

Subsequent increase in fuel-can dimensions, necessitated by restrictions

in the ORNL remiote welding fixture, have caused an increase in shield
dimensions. The final radiation shield is a right cylinder, 7 in. in diameter
by 13 in. in height. As in the experiment, complete thermal insulation must
cover the sides and base. The upper section will be in intimate contact with
the thermoelectric convertor packages. This will have an active area whose
thermocouples will cover 30 %. The remaining 70 % must be thermally
insulated. The combined exposed surfaces represent a sizable increase
over the original consideration. If only 3/4 in. of insulation were employed,
the heat loss would be 46, 6 thermal watts. It nnust be remembered that
the overall generator diameter cannot exceed 13 in. i therefore, unlimited
insulation cannot be used. However, by increasing the fiberglass thickness
to 1.25 in. , it is possible to reduce loss to 25. 6 watts, and still maintain an
acceptable generator outside diameter. An additional 3,8 thermal watts are
saved by subtracting the lead-wire loss present in the experimental apparatus.
This reduces total loss to 21,8 watts, which is approximately the same
value as reported earlier.

There are three principle avenues through which internally-produced

heat may travel. The first, lateral heat l o s s , had been discussed in the
above. The second is through the thermoelectric assembly. Here, heat
loss is a function of couple efficiency and hot and cold junction temperatures.
The last is through the metal envelope which surrounds the power conversion,
system.

RRC-0102
 20

P o w e r , through t h e r m o e l e c t r i c c o n v e r s i o n , can only become accepted

on a w i d e s p r e a d b a s i s if cost i s optimized. Since prototype models must
be hand-fitted, the unit cost is understandably high. This p r a c t i c e m u s t ,
however, be phased out of any volume production. One of the m o s t time
consuming p h a s e s of g e n e r a t o r c o n s t r u c t i o n is assemibly of the conversion
s y s t e m . In an attempt to anticipate production r e q u i r e m e n t s , a p r e -
packaged c o n v e r s i o n s y s t e m h a s been p u r s u e d . Suitably high t h e r m a l l y
conductive m e t a l s w e r e found with little difficulty. Selection of the
v e r t i c a l wall m a t e r i a l for the package p r e s e n t e d a m o r e complex p r o b l e m .
Such p a r a m e t e r s a s flexibility, n o n p o r o s i t y , low t h e r m a l conductivity
and compatibility with m^etal bonding m^ade m o s t m a t e r i a l s unacceptable.
Stainless s t e e l was s e l e c t e d , since a double, 5-mil wall b e s t imet all
r e q u i r e m e n t s . S t a i n l e s s , with an ambient t h e r m a l conductivity of 15, 65
B t u / h r / f i r / d e g , F / f t . , d r o p s s h a r p l y with a r i s e in t e m p e r a t u r e . At
operating conditions, it i s actually m o r e r e s i s t a n t to t h e r m a l flow than
e l e c t r i c a l and m e c h a n i c a l c e r m i i c s , whose ambient conductivity is 2 B t u / h r /
ft / d e g , F / f t . This is t r u e , p a r t l y b e c a u s e a 1/4-in, -thick wall would be
r e q u i r e d to provide s t r u c t u r a l strength and prevent diffusion inward during
vacuum pumipdown. A l / 4 - i n . -thick c e r a m i c wall would have no flexibility,
would bond to m e t a l only with e x t r e m e difficulty and waste m o r e heat
through i t s c r o s s section than s t a i n l e s s s t e e l .

An examination of F i g . 5 shows the c r o s s section of the power

c o n v e r s i o n package. The bellows shape of the v e r t i c a l o u t e r m o s t wall
p e r m i t s intimacy of contact under r e d u c e d p r e s s u r e . This i n s u r e s good
heat flow by minimizing heat backup i m m e d i a t e l y in front of the hot junction
plate and e x t e r n a l to the cold junction s h o e . Since heat l o s s through the
insulation located between couples has a l r e a d y been c o n s i d e r e d , all that
r e m a i n s to be done is to identify the heat l o s s through the s t a i n l e s s - s t e e l
wall.

A calculation of heat l o s s through the s t a i n l e s s s t e e l walls may be miade

in the following m a n n e r :

KCCross section exposed to heat) 0.208 in? or 0,0014 ft?

Thfhot Junction) 900 F ,
Tj,(cold junction) 145 F ,
A i f t e m p e r a t u r e drop) 755 F ,
u ( t h e r m a l conductivity) 15, 65 B t u / h r / f t ^ / d e g . F/ft,

The heat l o s s Q. (heat) m a y then be d e t e r m i n e d :

Q - ( T h - Tc) (u) (K)

RRC-0102
 21

Stainles,s.-Steel
Skin
Cold P l a t e

Printed
Circuit Board

AA F i b e r g l a s s
Insulation

Hot P l a t e

Double-Walled
Stainless-Steel
Bellows

F i g , 5 - Stainless-Steel Bellows Envelope for Thermocouple Package

I
RRC-0102
 22

Q " 16«54 B t u / h r ,
(or) Q = 4 . 8 t h e r m a l w a t t s / l i n e a r foot

The length of the pleated v e r t i c a l wall, when unfolded, is 5 i n c h e s .

T h e r e f o r e , this length factored into the above value (4. 8 t h e r m a l watts)
i n c r e a s e s it t o :
Q = 2 . 4 X 4, 8 - 11. 5 t h e r m a l watts l o s s

If we add this to the heat l o s s sustained through the insulation surrounding

the shield and between c o u p l e s , t h e r e will be a total conductive heat l o s s
of 33. 3 t h e r m a l watts experienced by the p r e s e n t 5-watt g e n e r a t o r design.

It is strongly believed that the above l o s s is p e s s i m i s t i c , since

s e v e r a l c o n t r i b u t o r y factors w e r e purposely omitted. Available heat was
p r e d i c a t e d upon only that which i s derived from Cesium-137. Analysis of
the Cesium-134 contribution indicates that an additional 4 . 7 t h e r m a l watts
will be c r e a t e d by t h i s i s o t o p e . While r e l a t i v e l y s h o r t e r half-life ( 2 . 4 y r s . )
suggests a rapidly d e c r e a s i n g contribution, t h e r e will be heat available,
r e g a r d l e s s of the t i m e factor since the g e n e r a t o r p e r f o r m a n c e is predicated
upon a 3 to 5 y r , operational r e q u i r e m e n t .

The e x p e r i m e n t that was run to d e t e r m i n e l a t e r a l heat l o s s , employed

twin polished r e f l e c t o r s as well a s a goldplated shield s u r f a c e . After
t e r m i n a t i o n of the e x p e r i m e n t , the a p p a r a t u s was dismiantled and it was
seen that the gold had suffered from e x p o s u r e to a i r , r e s u l t i n g in l o s s of
m o s t of i t s reflective s u r f a c e . This o c c u r e d during initial running when a
leak developed and vacuum was lost overnight. F o r this r e a s o n , no change
in heat l o s s was o b s e r v e d . Improved methods of sealing in a s s e m b l y of
the actual g e n e r a t o r , will obviate such a possibility and make for improved
insulability.

Tungsten was originally selected as the radiation shield, h e a t - s i n k

m a t e r i a l . This was p r e d i c a t e d upon the feasibility of obtaining a m a s s i v e
p i e c e , which would have a uniform density, and be of sufficient size for
fabricating the shield. Subsequent testing of v a r i o u s s a m p l e s of m a t e r i a l
indicated that 86 % of t h e o r e t i c a l density i s the b e s t that is p r e s e n t l y
available. Shielding e x p e r i m e n t s ( d i s c u s s e d l a t e r in this report) indicated
that over 3 in. of c r o s s section would be r e q u i r e d to attenuate the radiation
down to an acceptably safe v a l u e . T h i s , in t u r n , would i n c r e a s e the o u t e r -
m o s t g e n e r a t o r d i m e n s i o n s beyond 13 in. dia. F o r this r e a s o n tungsten is
totally u n a c c e p t a b l e .

RRC-0102
 23

Depleted u r a n i u m is a high-density i:Tjetal whose attenuation is almost

twice that of lead and over 1. 5 t i m e s that of tungsten. A r i g o r o u s i n v e s t -
igation h a s shown that, with u r a n i u m , the original dimensions apply.
However, in making the change, it was n e c e s s a r y to r e - e v a l u a t e the heat
t r a n s f e r p r o p e r t i e s of a u r a n i u m shield. Sintered tungsten has a high
(89 B t u / h r / f t ^ / d e g . F/ft) t h e r m a l conductivity. At 49 deg. F . , however,
this value d e c r e a s e s to 75 at operational t e m p e r a t u r e as indicated in F i g . 6,
U r a n i u m , Fig. 7, exhibits the opposite effect. At ambient t e m p e r a t u r e ,
its conductivity i s 15 B t u / h r / f t ^ / d e g . F/ft w h e r e a s , at operational t e m p e r -
a t u r e , (900 F . ) this value is 20, Since the operational t e m p e r a t u r e
r e f l e c t s a t h e r m a l conductivity lower than tungsten v a l u e s , a r i g o r o u s
t r e a t m e n t was given the problem of " h o t - s p o t " formation throughout the
uranium shield. As may be seen in Appendix B , solution ©f the problem
is i n t r i c a t e . C e r t a i n c o n s e r v a t i v e a s s u m p t i o n s w e r e made in this c a l c u -
lation. They tend to make the r e s u l t s reflect a t e m p e r a t u r e differential
c o n s i d e r a b l y higher than a c t u a l . Despite t h i s , the g r e a t e s t difference
found was only 140 d e g r e e s F , between top and bottom of the shield.
B e c a u s e of t h e s e r e s u l t s employment of a depleted uraniun) shield (in cast
form) is not only a dimensionally economical nneans of shielding but c o m -
pletely safe from the h a z a r d s a s p e c t a s w e l l . It must be r e m e m b e r e d that
the p r e s e n t design c r i t e r i o n i n v e r t s the shield, so that the threaded cap is
at the b a s e , r a t h e r than at the top of the heat flow p a t t e r n . By so doing,
the a i r gap between t h r e a d s does not impede heat t r a n s f e r .

Additional protection will be given the uranium by plating it with a

heavy l a y e r of nickel on all exposed s u r f a c e s . By so doing, the heat t r a n s -
fer is actually i m p r o v e d , while obviating t h e possibility of pyrophoric con-
sumption in the event of an a c c i d e n t .

RRC-0102
 24

100

90

" - -
u 80
3 »
T3 ^ - ^
70
"•O

60

50
0 2()0 4C)0 6()0 8C)0 10()0 1200 1400 1600 1800 2000 2200 24^0 26S
T e m p e r a t u r e in D e g r e e s F a h r e n h e i t

F i g . 6 - T h e r m a l Conduct ivity of T u n g i e n (Sintered) As

a Function of T e m p e r a t u r e , *

30

•43 f ^ 20
Q
3 *

10
£•2

0 100 200 300 400 500 600 700 800 900 1000 llOO 1200 1300
T e m p e r a t u r e in D e g r e e s F a h r e n h e i t
F i g . 7 - T h e r m a l Conductivity of Depleted Uranium (Theoretical Density)
A s a F u n c t i o n of T e m p e r a t u r e ,
*Data Supplied by Stauffer Metals Co,

RRC-0102
 25

V. SUBTASK 2. 2 - POWER CONVERSION ANALYSIS

As indicated in earlier reports, a dual effort on the power conversion
system for the cesium generator is being expended by both Royal Research
Corporation and the Westinghouse Electric Corporation, That system which
most closely meets contractual requirements as to element cladding, couple
cladding and dimensional compatibility, while optimizing the available heat,
will be used. Data resulting from RRC and Westinghouse efforts are
presented individually in this subtask, for the purpose of delineation and
comparison.

*A. WESTINGHOUSE EFFORT

Westinghouse design and fabrication of the rmoelectric couples for

the Cesium-137 Power Generator were described in the second Quarterly
Report, Since then, thermoelectric couples have been assembled and
enclosed in a module for endurance evaluation. It is our feeling that suffi-
cient data points will be obtained by testing a miniaturized thermoelectric
module. From these data, calculations will be made to determine the
specifications necessary to produce 5 watts power output at 3 volts load.

The miniature thermoelectric module, with power terminals, is

shown in Fig. 8. This module contains 15 thermoelectric couples connect-
ed electrically in series and thermally in parallel to form the thermoelectric
ladder.

The thermoelectric couples a r e sandwiched between but electrically

insulated from, the hot and cold sides. The electrical insulation used
maintained its dielectric properties at the high, 482°C. hot side tempera-
ture while insuring reasonably good thermal conductance, A potting
nnaterial having low thermal conductivity, was also incorporated in the
module. This material will be described later.

The module hot and cold sides were machined from a stress-free
18-8, 300 series steel.

Special welding was employed to join a thin 18-8 tubular steel shell
to the hot and cold sides, to insure a leak-free assembly.

The module temperature was raised to 100° C. with an inert gas

flowing through the module. This final precautionary step was used
to minimize oxidaticn of the thermoelectric materials. After sufficient •
time, a positive pressure of the inert gas was sealed in the thermoelectric
module.

RRC-0102
o

t\>

F i g . 8 - Westinghouse Miniature T h e r m o e l e c t r i c Module

50
O

o
IS)

Fig. 9 - T e s t Setup of Miniature T h e r m o e l e c t r i c Module

 28

Perfornnance

The evaluation of the m i n i a t u r e t h e r m o e l e c t r i c module shown under

t e s t in F i g . 9, was to be conducted in two s e p a r a t e t e s t s . The first t e s t
was conducted only on a t e m p e r a t u r e gradient c r i t e r i o n , i . e . , the t h e r m a l
l o s s e s w e r e not c o n s i d e r e d .

At the m a x i m u m A T of 433° C. , the t h e r m o e l e c t r i c nnodule power

output is 2. 75 watts at 0.99 v o l t s .

Tabulated r e s u l t s a r e listed in Table 4, Column 1.

The ^ factor in Column 1 is a r a t i o of conductor r e s i s t a n c e plus

contact r e s i s t a n c e to bulk t h e r m o e l e c t r i c m a t e r i a l r e s i s t a n c e . Special
designing and a s s e m b l y techniques would reduce p further; but a r a n g e
between 0 - 10% for 0 is m o s t a c c e p t a b l e .

Column 2 in Table 4 c o m p a r e s the m e a s u r e d data with values c a l c u -

lated from t h e o r e t i c a l c u r v e s . The c u r v e s will be published by Westinghouse
and a r e p r o p r i e t a r y informsation.

Table 5 is a full-powered extrapolation, based on r e s u l t s obtained with

the m i n i a t u r i z e d module evaluation, for a 3-volt load voltage and 5 watt power
output. The thermioelectric module will be 3 - 3 / 4 in. s q u a r e and enclose 35
t h e r m o c o u p l e s ( 1 / 4 X 1/4 x 1-1/8 i n , ) .

The second t e s t has been designed to minimize t h e r m a l l o s s e s and

r e q u i r e s a special testing a p p a r a t u s . The m e c h a n i s m s involved in heat
t r a n s f e r and the techniques for mieasurement a r e complex. It is planned to
conduct the t e s t in the n e a r future.

RRC-0102
 TABLE 4

Reduced Scale T h e r m o e l e c t r i c Module Output

Column 1 Column 2

Hot wall t e m p e r a t u r e (T^)° C. 482 482

Hot Junction t e m p e r a t u r e ("^ C ) 472 (estimated) 472
Cold wall t e m p e r a t u r e (T^) ° C . 49 49
Cold junction t e m p e r a t u r e ( ° C . ) 73 73
Couple voltage (open circuit) 0. 132 volt/couple 0. 132 volt/couple
Module voltage (open c i r c u i t ) 1.98 volt 1.98 volt
Module voltage (roiatched load) 0.99 volt 0,99 volt
Couple power output 0, 183 watt/couple 0, 183 watt/couple
Module power output 2. 75 watt 2. 75 watt
Couple r e s i s t a n c e (27° C ) ,0074 o h m / c o u p l e .0074 ohm/couple
Module r e s i s t a n c e (482° C.) 0,361 ohm 0.336 ohm
0,0745 0
*
26. 3 watt
Q^ (Conducted through couple)
10. 15 watt
Qp (Heat a b s o r b e d at hot junction)
« tion) 1. 36 watt
Qj (Joule heat r e t u r n e d to hot junction) 7, 75%
Couple efficiency

RRC-0102
 30

TABLE 5

Anticipated Output of the 5-Watt Conversion System

Hot wall temperature (T^^) ° C. 482

Cold wall temperature |T^) ° C. 49
Temperature difference (AT) ° C. 433
Module voltage (open circuit) 4, 62 volts
Module voltage ( 1 , 74 ^-v load) 3,00 volts
Module resistance (482 ° C ) 0.936 ohms
Power output (1. 74 _A_ load) 5, 18 watts
Q^ (Conducted through couples) 54. 7 watts
Q (Absorbed at hot junction) 14,9 watts
«
Q. (Joule heat returned to hot junction) 1. 39 watts
Couple efficiency 7.6 %

Potting Thermal Insulation

Considerable effort was applied to the development of several potting

thermg-l insulators by Westinghouse. These thermal insulations are charac-
terized by four p a r a m e t e r s ,

1, The thermal conductivity of the thermal insulator has to be con-

siderably less than that of the thermoelectric materials.

2. The thermal insulator has to support the thermoelectric pellets

during shock and vibration loadings.

3. The thermal insulator must retain a high electrical resistance

(10 ohm) at the hot junction temperature with negligible
chemical reactivity.

4, It must have handling properties which permit its incorporation

within the thermoelectric module simply and economically.

RRC-0102
 31

Satisfying the f i r s t r e q u i r e m e n t of the above was the most difficult. It

was n e c e s s a r y to c o n s t r u c t a t h e r m a l conductivity m e a s u r i n g a p p a r a t u s for
testing the insulating m a t e r i a l s .

After conducting t e s t s on a number of m a t e r i a l s and m i x t u r e s , the

final potting compound was formulated by mixing exploded silica and a
smooth, working r e f r a c t o r y m a t e r i a l . T h i s potting t h e r m a l insulator best
fitted t h e outlined r e q u i r e m e n t s .

T h e r m a l Conductivity

The a p p a r a t u s used for the t h e r m a l conductivity m e a s u r e m e n t s was a

modification of the 8-in. guarded hot plate method approved by the National
B u r e a u of S t a n d a r d s ( A . S . T . M . Designation: C 177-45).

F i g . 10 shows such an a p p a r a t u s , consisting essentially of a c i r c u l a r

h e a t e r (2), guard h e a t e r s (1) and (6), a section of the insulating m a t e r i a l
to be t e s t e d (3) and the cooling plate (4). The heater (2), established heat
flow through the insulating m a t e r i a l . The guard h e a t e r s , (1) and (6) and
h e a t e r (2), a r e kept at the s a m e t e m p e r a t u r e by adjustments of the input to
the guard h e a t e r s . This p r o c e d u r e p r e v e n t s heat flow from h e a t e r (2) to
h e a t e r (1) and (6), r e s u l t i n g in all the heat flow from h e a t e r (2) being p e r p e n -
d i c u l a r to i t s s u r f a c e .

The s p a c e between the insulating m a t e r i a l to be t e s t e d , h e a t e r s (2) and

the guard h e a t e r s , is filled with alumina insulating g r a i n .

As indicated in F i g . 10, t e m p e r a t u r e s a r e m e a s u r e d at v a r i o u s points

in the insulating nnaterial and h e a t e r section by m e a n s of c h r o m e - a l u m e l
t h e r m o c o u p l e s . The differential t h e r m o c o u p l e method was employed between
the e l e c t r i c a l powered h e a t e r (2) and guard h e a t e r s (1) and (6). By then
m e a s u r i n g energy input to the h e a t e r (2), insulating m a t e r i a l t h i c k n e s s , and
t e m p e r a t u r e gradient a c r o s s the insulating m a t e r i a l , the t h e r m a l conduc-
tivity may be d e t e r m i n e d by the standard equation:

K =Q ^ L (Btu)
A XT hr71F~F.

RRC-0102
 32

Fig. 10 - Thermoelectric Module Evaluation Apparatus

RRC-0102
 33

B. ROYAL > RESEARCH CORPORATION E F F O R T

Royal > R e s e a r c h C o r p o r a t i o n ' s effort on the power conversion s y s t e m ,

which is designed to produce the r e q u i r e d e l e c t r i c a l power through conversion
of t h e r m a l energy of the isotopic fuel s o u r c e , is continuing with favorable
r e s u l t s . Design p a r a m e t e r s have been defined and attention i s now focused
on the e l e m e n t s t h e m s e l v e s . To a s s u r e maximum efficiency of individual
t h e r m o c o u p l e s , p r o b l e m s such a s heat flow through the couple, internal r e -
s i s t a n c e , contact r e s i s t a n c e and c o m p a c t n e s s m u s t be solved. Research
was initiated, by defining a thermocouple design completely different from
any known to d a t e . In our d e s i g n , c o n s i d e r a b l e emiphasis was placed on
couple r e s i s t i v i t y to t h e r m a l and vibrational shock.

A c o m p r e h e n s i v e evaluation was next made of v a r i o u s e l e c t r i c a l

insulator m a t e r i a l s . The c r i t e r i o n for acceptability was high e l e c t r i c a l
r e s i s t a n c e with good t h e r m a l conductance. Many kinds of m a t e r i a l s w e r e
t e s t e d by applying them to t h e r m o c o u p l e s which w e r e then placed under t e s t .
Only those mtaterials which survived extended p e r i o d s of subjection to t e s t
conditions w e r e selected for u s e in the t h e r m o e l e c t r i c package. As shown
in F i g , 11, the cold e n d - c a p design has m a d e possible the a s s e m b l y of
couples having e x t r e m e l y close t o l e r a n c e s . M a t e r i a l s having s u p e r i o r bond-
ing c h a r a c t e r i s t i c s a r e employed in the cold junction a s s e m b l y .

The need for a nrjethod of joining the completed thernnocouples in a

s e r i e s a r r a y , which would be capable of reproducing the r e q u i r e d output
voltage, was r e s o l v e d by m e a n s of the special printed circuit board shown
in Fig, 14. Capable of withstanding t e m p e r a t u r e s over 300*^ F . , the
printed c i r c u i t board p e r m i t s l e s s than ,002 ohms combined contact r e s i s -
t a n c e , per t h e r m o c o u p l e . F a b r i c a t i o n and testing of an a r r a y designed to
a c c o m m o d a t e 6 c o u p l e s , F i g . 12, showed that the internal r e s i s t a n c e per
couple d e c r e a s e d , when an i n e r t - a t m i o s p h e r e e n c l o s u r e was used to fabricate
and a s s e m b l e our P and N e l e m e n t s , F i g . 13,

E a r l y couples of R o y a l ' s design r e s u l t e d in an a v e r a g e r e s i s t a n c e of

. 027 o h m s . At the p r e s e n t t i m e extensive work is being undertaken to lower
couple r e s i s t a n c e to the a l t e r n a t e value of 0,0055 o h m s , the m a t e r i a l r e s i s -
tance of lead t e l l u r i d e . F o r a completed thermocouple to be efficient and
d u r a b l e , e v e r y p r o b l e m a s s o c i a t e d with construction m u s t be defined.
Toward this end, each detail i s under study, with the goal that of obtaining
the highest efficiency for all of the t h e r m o c o u p l e s .

RRC-0102
 34
Temperature Electrical
Profile Insulation

P -Type Element
G opper Adapter

Circuit Contact
P oints

N-Type Element
Copper Adapter

N-Type Lead
Telluride Eleroient

P-Type Lead
Telluride Element

M txxco Iron
Hot Junction Shoe
Electrical Insulation

Heat Sink
Heat Source

Fig. 11 - Enlargement of Royal Research Lead Telluride Thermocouple

Design, With Operating Temperature Profile

RRC-0102
 '-'''%ir''-«
^r-

Fig. 12 - Reduced-Scale Printed-Circuit Board Used

in Life Testing of Royal Thermocouples

Fig, 13 - Thermocouple Assembly in Inert Atmosphere Enclo

sure

RRC-0102
 CIRCUIT BOARD

LEAD TELLURIDE THERMO-

ELECTRIC ELEMENT/N"IN
E W / ^ " INSIDE

COPPER COLD
S H O E / P " SIDE

5»
O
I COPPER COLD
o SHOEfN" SIDE
»—»
o

SILVER- PLATED
PRINTED CIRCUIT
COUPLE HOLD-DOWN
ASSEMBLY

Fig. 14 - Royal Thermocouples Assembled on Printed Circuit Board

 37

To insure that a completed element will be free of any oxidation during

fabrication and asseitibly, all thernaoelements are handled within an inert
gas enclosure. Here, an argon atmosphere protects the miaterials until
fabrication and assembly are complete. Actual testing is now under way to
identify maximum operating temperatures of the hot and cold junctions and
so determine the most efficient temperature differential. In the thermo-
couple tester (described in the 2nd Quarterly Report), temperature-indicating
thermocouples were placed on the hot Junction, iron shoe of the couple,
copper cold junction cap and the cold junction plate. Individual temperatures
were recorded, which later proved very valuable in determining the insula-
tion best suited for both the hot Junction iron shoe and cold Junction copper
cap. Anticipated operating conditions are simulated within the tester, in
order to realistically analyze behavior of the thermocouples,

A study also has been initiated to identify those methods which aid in
assembling the completed package. At this writing, the method decided upon
is as follows: After successfully completing acceptance tests, each thernso-
couple will be attached to the printed circuit board. When all 46 therino-
couples have been attached, the assembly will be tested to determine the
power output from the system. After satisfying testing requirements, the
system shall be enclosed in an hermetically sealed envelope. A stainless
steel bellows comprising the walls of the envelope shall be bonded to the cold
junction plate. Copper was chosen because it can be brazed to the bellows.
This container is then inverted and an electrical insulator placed on the
interior surface of the copper plate. Next, the printed circuit board, con-
taining the thermocouples, will be inserted in the container so as to make
physical contact with the cold junction. A pre-punched, circular piece of
AA Fiberglas shall be inserted between the thermocouple pattern. At this
time, an Armco iron plate, with location holes matching the position of each
thermocouple, will be placed over the insulator. "Xhe electrical insulator
will then be attached to individual iron hot shoes. Locating holes are threaded,
to permit individual adjustment of the thermocouples so that intimate contact
and pressure necessary for both the cold and hot junctions, can be maintained.
After all thermocouples have been adjusted, a thin, stainless steel plate is
placed over the hot Junction plate. This di skis then welded to the stainless
steel bellows, completing assembly of the thermoelectric package. A valve
located on the cold junction copper plate will permit the package to be pumped
down, under vacuum. After such time as is needed to outgas the systenn, a
partial atmosphere of reducing gas and argon will be introduced into the
package. Thermocouple performance is optimized by such an atmosphere.
Finally, the package will be sealed off and evaluation initiated in preparation
for its use in the Cesium-137 generator.

RRC-0102
 38

Royal T h e r m o c o u p l e P e r f o r m a n c e Under T e s t

A typical t e m p e r a t u r e profile encountered in our thermocouple t e s t s

may be seen in Fig. 11. T h e s e couples have an o p e n - c i r c u i t voltage of
0,218 volts when the shoe t e m p e r a t u r e is 925*-* F . and the cold junction
t e m p e r a t u r e is 175° F . Each t h e r m o c o u p l e is designed to dissipate 2.07
watts of heat, with 1.97 watts of it going through the lead telluride and 0. 10
watts through the c e r a m i c s .

T h e o r e t i c a l r e s i s t a n c e of a typical couple at room t e m p e r a t u r e would

be 5.53 m i l l i o h m s , 33 m i l l i o h m s at operating input t e m p e r a t u r e of the couple.
It i s anticipated that t h e s e r e s i s t a n c e s actually will be achieved in future fab-
r i c a t i o n r e s e a r c h . To d a t e , t h e s e couples have a room t e m p e r a t u r e r e s i s -
tance of 19 to 25 m i l l i o h m s ; r e s i s t a n c e i n c r e a s e s to about 100 milliohms per
couple at operating t e m p e r a t u r e . At this t i m e , it has been established that
this i n c r e a s e in r e s i s t a n c e is not due to junction r e s i s t a n c e but r a t h e r to
m i c r o - f r a c t u r e s found in the lead t e l l u r i d e m a t e r i a l . Work i s p r o g r e s s i n g
on developing an annealing p r o c e s s to eliminate e x c e s s i v e r e s i s t a n c e .

A design objective of our theriwoelectric p r o g r a m is to produce 0. 128

watts p e r t h e r m o c o u p l e , with an efficiency of 6,2 per cent at the designed
o p e r a t i n g t e m p e r a t u r e . T h u s , the output of the thermocouple package, which
contains 46 c o u p l e s , will be 5.89 w a t t s . T h i s power then, will produce 5,0
watts of e l e c t r i c i t y , b a s e d on an 85 per cent efficient dc-dc c o n v e r t e r . Open-
c i r c u i t voltage of the t h e r m o c o u p l e package will be 10 v o l t s , d . c . To d a t e ,
e x p e r i m e n t a l data on t h e s e couples indicate a maximum power when the i n t e r -
nal load is 1.25 t i m e s the e x t e r n a l load. T h e r e f o r e , maximum power will
be produced when the t h e r m o e l e c t r i c power package is operating at 4 . 4 v o l t s .
This figure will i n c r e a s e when the power i s optimized a s a function of the
d c - d c c o n v e r t e r efficiency and the efficiency of the t h e r m o e l e c t r i c s y s t e m ,
although this p r i m a r y voltage to the d c - d c c o n v e r t e r will r e m a i n within the
r a n g e of 4 . 4 to 5,0 v o l t s .

RRC-0102
 39

VI. SUBTASK 2.3- GENERATOR DESIGN ANALYSIS

A b a s i c overall design was p r e s e n t e d and d e s c r i b e d in the Second

Q u a r t e r l y R e p o r t . This has been subject to s e v e r a l i n t e r n a l c h a n g e s , due
p r i m a r i l y , to development of m o r e detailed information regarding fuel
fabrication, a r r i v a l at a b e t t e r understanding of shielding r e q u i r e m e n t s and
the r e s u l t s of e x p e r i m e n t a l data on outer p r e s s u r e shell integrity.

The f u e l - i n c r e m e n t configuration has been changed from a single,

doubly-encapsulated stack of seven i n c r e m e n t s , to two stacks of t h r e e each.
This was n e c e s s i t a t e d b e c a u s e of l i m i t a t i o n s of the Oak Ridge National
L a b o r a t o r y h o t - c e l l welding e n c l o s u r e . B e c a u s e of t h i s , the fuel configura-
tion had to be shortened and i n c r e a s e d in d i a m e t e r , or fabricated as two
s e p a r a t e u n i t s . The f o r m e r a l t e r n a t i v e was u n d e s i r a b l e , since it would have
invalidated all p r e v i o u s work done on l a t e r a l dimensioning of the g e n e r a t o r .
F u r t h e r , it would have r e s u l t e d in a design whose outer d i a m e t e r would have
exceeded 13 in. Since the r e q u i r e m e n t for the device specifically defined
the t o l e r a b l e l i m i t s of d i a m e t e r , the fuel i n c r e m e n t d i a m e t e r s w e r e held at
2.50 in. The only a l t e r n a t i v e was in going from a seven i n c r e m e n t fuel
load, to an even n u m b e r e d one and dividing the i n c r e m e n t s into two b a t c h e s .
Six e l e m e n t s would r e d u c e the fabrication time previously r e q u i r e d for
seven, and would not m a t e r i a l l y affect the fabrication technique or r e q u i r e
a g r o s s e n l a r g e m e n t of p r o c e s s equipment. The final design of the fuel
elenaent a r r a y is shown in F i g . 15. All dimensional t h i c k n e s s e s and s p a c e r s
a r e r e q u i r e m e n t s identified by ORNL and t r a n s m i t t e d to the c o n t r a c t o r . The
l a t e r a l walls will be ground to s i z e , b a s e p l a t e s welded in place and the r e -
sulting c o n t a i n e r s d e l i v e r e d for fuel loading. Additional h a r d w a r e to be
include4 will be fitted caps and i n t e r i o r can s p a c e r s of varying t h i c k n e s s e s .
The l a t t e r a r e for filling any a i r voids which may develop fromi i n c r e m e n t
stacking. Since the exact height of each i n c r e m e n t is not known, v e r t i c a l
can dimensions can not be anticipated. In addition to two cans and lids of
each s i z e , one e x t r a set of cans will be d e l i v e r e d , to p e r m i t orientation
welding and f a m i l i a r i z a t i o n p r i o r to actual fuel e n c l o s u r e .

Although not shown in the full scale m o d e l , the shield has been modified.
Originally specified to be f a b r i c a t e d from s i n t e r e d tungsten, it will now be
made of c a s t , depleted u r a n i u m . Additionally, the s c r e w cap originally shown
at the t o p , will become the b a s e . This enhances heat flow in the d e s i r e d d i r e c
tion, upward, by removing the a i r gap which must exist between cap and body
of the shield. F i n a l l y , both ends will be t a p e r e d to c o n s e r v e on weight and
i m p r o v e insulability of the component. Since uranium is to be u s e d , all inner

RRC-0102
 40

and outer s u r f a c e s will be heavily nickel plated to obviate combustion in the

event of a d i s a s t e r . The e x t e r n a l , nickel surface will be polished and gold
plated, to s e r v e a s a r e f l e c t o r and i m p r o v e r a d i a n t - h e a t insulation.

The shield cap will be provided with t h r e e indented slots to p e r m i t

easy a s s e m b l y . A t a p e r e d pin slot and hole will be provided so that final
c l o s u r e m a y be made by m e a n s of a pin hamimered h o m e . It is interesting
to note that the t h r e e holding slots m a y a l s o be used to rigidly position the
shield in the inverted position, preventing eventual sag into the felted fiber-
g l a s s . Should this o c c u r , intimacy of contact between shield and t h e r m o -
e l e c t r i c c o n v e r t e r would be l o s t , causing a power d r o p .

No change has been made in the t h e r m a l insulation m a t e r i a l s ,

AA Felted F i b e r g l a s will be used to insulate against conductive heat l o s s .
This m a t e r i a l will be comtpacted to a density of 7 pounds/ft , which the m a n u -
f a c t u r e r quotes a s having a t h e r m a l conductivity of l e s s than 0 , 0 1 B t u / h r / £ t ^ / f t
in a vacuum of l e s s than 100 m i c r o n s . The dual, gold plated r a d i a n t - h e a t r e -
flectors will be used a s d e s c r i b e d in the Second Q u a r t e r l y R e p o r t . Braided
f i b e r g l a s s s p a c e r s will be used a s standoffs between shield and s p a c e r s .

One significant fact was brought to light after the l a s t t h e r m a l insulation

e x p e r i m e n t was concluded. When the a p p a r a t u s was d i s a s s e m b l e d , it was seen
that the gold plating had all but d i s a p p e a r e d . I m p r o p e r determination of a i r
tightness of the outer shell p r i o r to t e s t i n g , r e s u l t e d in development of an a i r
leak during evaluation. Since this o c c u r r e d during the night, considerable
oxygen was admitted to the i n t e r i o r before r e m e d i a l m e a s u r e s could be taken.
At the high operating t e m p e r a t u r e , to which the shield must be continually
subjected, exclusion of a i r is i m p e r a t i v e . M e a s u r e s have been taken to
prevent the admittance of air during g e n e r a t o r pump-down. Although AA Felted
F i b e r g l a s will also be used to t h e r m a l l y shield the d c - d c c o n v e r t e r (in the upper
c o m p a r t m e n t ) , the higher efficiency made p o s s i b l e with a p a r t i a l vacuum is not
n e c e s s a r y h e r e , and t h e r e f o r e no pump-down will be needed. The c o n v e r t e r
itself will be sealed within a potted package which will exclude oxygen.

Final dimensioning of the o u t e r m o s t 7075-T6 alunninum shell will be made

after full scale testing to verify e x p e r i m e n t a l r e s u l t s obtained. F o r these t e s t s ,
r e d u c e d - s c a l e c o n t a i n e r s w e r e fabricated a s shown in F i g . 16, Wall thickness
was p r e d i c a t e d on the Lamont p r e s s u r e c u r v e s , which though limited to 2024
a l u m i n u m , w e r e the only available data p r e d i c a t e d upon actual testing r a t h e r
than on calculated r e s i s t i v i t y to p r e s s u r e . Although the inside d i a m e t e r of the
t e s t v e s s e l s was 3 in, , the r e s u l t s w e r e believed to be extrapolatable to anti-
cipated d i m e n s i o n s of the g e n e r a t o r .

RRC-0102
 41

To be Welded
'wzizuzmz by O. R. N. L,

Cesium Typical Spacer

Polyglass

Platinunn
Cups Outer
Hastelloy C
Can

Inner
Hastelloy C
Can

F i g . 15 - Double-Walled Hastelloy C Fuel Can

RRC-0102
 42

Fig, 16 - Reduced Scale Aluminum Test Vessels

Fig. 17 - Vessel P r e s s u r e Testing at Naval Electronics

Laboratory, San Diego, California

Official U, S. Navy Photograph

RRC-0102
 43

Fig, 18 - Imploded T e s t V e s s e l , Showing C r a c k e d Wall, Above,

and Fragmented Bottom, Below

RRC-0102
 44

T h e s e t e s t v e s s e l s w e r e rounded at the b a s e s and provided with a single

" O " ring seal at the lid, A r a i s e d a r e a on the lid was machined, drilled and
taped to provide a t h r e a d e d hole for the lifting lug. In this way, the lid thick-
n e s s was not weakened by p e r f o r a t i o n . In threading the lid to the v e s s e l body,
a gap of 0.030 inches was left, to d e t e r m i n e if the two p a r t s would p r e s s u r e -
weld under p r e s s u r e .

P r e s s u r e testing was p e r f o r m e d at the Battery Whistler T e s t A r e a ,

Naval E l e c t r o n i c s L a b o r a t o r y , San Diego, California, H e r e , a naval rifle
has been f o r e s h o r t e n e d , the tube plugged and inner bore tapped for sea water
a d m i s s i o n and a t w o - s t a g e pumip connected to provide p r e s s u r e . B e c a u s e
implosion could damage b r e e c h p a r t s , the inner cavity of the first test v e s s e l
was filled with wood and then wrapped in b u r l a p . F i g . 1 7 shows i n t e r i m
rennoval of the v e s s e l after a typical r u n .

Runs w e r e a c c o m p l i s h e d by a l t e r n a t e l y p r e s s u r e testing and machining

m e t a l from the outside of bottom and side w a l l s . After four such r u n s , the
wall and b a s e t h i c k n e s s had been reduced to a point considerably below Lamont
r e c o m m e n d a t i o n s . The t e s t s w e r e then t e m p o r a r i l y concluded with no r u p t u r e
occurring.

F o r the second set of r u n s , the r e s e r v e t e s t v e s s e l was used. H e r e ,

the s t a r t i n g wall and b a s e t h i c k n e s s was equal to final dimensions on the first
e x p e r i m e n t . The p u r p o s e of this t e s t s e r i e s was to d e t e r m i n e the r u p t u r e
point in wall t h i c k n e s s a n d / o r b a s e t h i c k n e s s . This t e s t v e s s e l differed from
the f i r s t , in that the inner b a s e was rounded, so that no right angle was
p r e s e n t e d to the o u t s i d e . Table 6 shows the r e s u l t s of the second s e r i e s of
t e s t s . After each p r e s s u r e t e s t , f i r s t at 9,500 p s i . and then at 15,900 p s i , ,
the v e s s e l was r e m o v e d and machined down. The foregoing p r e s s u r e s
r e p r e s e n t sea p r e s s u r e at depths of 3,600 and 6,000 fathoms, r e s p e c t i v e l y .

During machining, an effort was miade to maintain the original r a t i o

of wall to base t h i c k n e s s , so that ultimate collapse could be attributed to a
final configuration, r a t h e r than to a weak spot in the c r o s s section t h i c k n e s s .
However, despite all p r e c a u t i o n s , r u p t u r e did o c c u r , as indicated by t e s t
No. 7. The a p p e a r a n c e of the v e s s e l and fragmented b a s e may be seen in
F i g . 18 . Analysis of the r u p t u r e point indicates that a weak a r e a developed
c i r c u m f e r e n t i a l l y along a line opposite the i n t e r i o r round on the bottonn.
During machining, an outside r a d i u s was r e c u t on the c o r n e r formed by the
sides and b a s e , to maintain a curved s u r f a c e . However, in so doing a portion
of the bottom roundness was r e d u c e d to a point that it did not exceed the v e r -
tical wall in t h i c k n e s s , as it should h a v e . This may be seen by the concavity
of the b r e a k surface w h e r e the b a s e imploded and also by the s y m m e t r i c a l
c u r v a t u r e on the o u t e r m o s t f r a g m e n t s as shown in the photo of the r e a s s e m b l e d
base.

RRC-0102
 45

TABLE 6

G e n e r a t o r Shell P r e s s u r e T e s t - Reduced Scale

( P r e s s u r e T e s t i n g , 5 Min. at P r e s s u r e )
7075-T6 Aluminum - 81,000 p s i . m a x , tensile strength
72,000 p s i , m a x . yield strength

Test Side Bottom Pressure

No. Thickness Thickness

1 Wall 0.58 Bottom 0.875 OK OK

O . D . 4 . 15 Length 4 . 2 8

2 Wall 0. 559 Bottom 0.750 OK OK

O.D. 4.088 Length 4. 155

3 Wall 0.528 Bottom 0.625 OK OK

O.D. 4.026 Length 4 . 0 3

4 Wall 0.497 Bottom 0.550 OK OK

O . D . 3.964 Length 3.915

5 Wall . 4 6 6 Bottom 0.550 OK OK

O . D . 3.902 Length 3.915

6 Wall . 3 8 8 Bottom 0.550 OK OK

O . D . 3.777 Length 3.915

7 Wall .326 Bottom 0.460 OK A t 15,850 p s i ,

O . D . 3.652 Length 3.79 B ottom Collapsed

The e x t r e m e t h i n n e s s of b a s e and wall at the time of f r a c t u r e verifies

the value of the e x p e r i m e n t . According to t h e s e data, a full scale p r e s s u r e
shell of approximately 1,4 in. s i d e - w a l l t h i c k n e s s and 2 , 2 in. b a s e thickness
would withstand i m m e r s i o n to a full 6,000 fathoms. This is a considerable
reduction over initial e s t i m a t e s of 2 . 5 and 3 . 5 in. thickness for sides and
base, respectively.

Repeated cyclings of p r e s s u r e , during the n u m e r o u s t e s t s performed

on both t e s t v e s s e l s produced no leaks through the single "O" r i n g . (This is
highly d e s i r a b l e since r e s e a r c h e r s at Lamont have r e p o r t e d that placement
at depth and subsequent r e t r i e v a l of i n s t r u m e n t p a c k a g e s , caused the con-

RRC-0102
 46

t a i n e r s to " b r e a t h e " e . g . , the extreinely high p r e s s u r e caused slight con-

t r a c t i o n of the e n t i r e c o n t a i n e r . T h i s , in t u r n , r e s u l t e d in l e a k a g e . ) When
the first t e s t v e s s e l was examined, no p r e s s u r e welding was found to have
o c c u r r e d , despite r e p e a t e d c o m p r e s s i o n s . A galling effect was identified,
which made reopening e x t r e m e l y difficult. This phenomenon will be r e -
evaluated during full-scale p r e s s u r e t e s t i n g .

The s u c c e s s of r e d u c e d - s c a l e testing provided wall thickness a p p r o x i -

mations which will be applied to two full-scale p r e s s u r e s h e l l s . In anticipa-
tion of t h e s e t e s t s , two forgings have been o r d e r e d from the Harvey Aluminum
Company, Oakland, California. T h e s e will consist of b a s e s and caps only,
since the upper portion of the configuration will be s t r u c t u r a l l y s t r o n g e r by
design. Starting as forgings, the p i e c e s will be s t r e s s relieved and machined
to a p p r o x i m a t e d i m e n s i o n s . After receiving t h e m , final machining and thread
cutting will be p e r f o r m e d . Inside d i a m e t e r s will be predicated upon a 2-in, -
thick radiation shield since t h i s t h i c k n e s s a p p e a r s adequate when using depleted
u r a n i u m as shield m a t e r i a l .

RRC-0102
 47

VII, SUBTASK 2,4 - GENERATOR SHIELDING ANALYSIS

Particular emphasis has been placed upon the shielding of gamma radia-
tion in the proposed cesium generator. This is important from the standpoint
of safety and efficiency. Concerning safety, this means that the nuclear core
nnust have sufficient shielding to reduce to a tolerable level the dosage received
by persons working in close proxinnity to it. As to efficiency, adequate shield-
ing ineans effective utilization of gamma-ray conversion to heat. Cesium is
a useful power source only if both beta and gamma decay radiation can be con-
verted to heat. Since the shield is also the heat accumulator, all attenuated
radiation is made available as heat to the thernrsoelectric converter.

While both foregoing situations suggest the generous use of shielding

material, a third parameter influences the careful avoidance of excessive
shielding. The cesium generator is predicated upon a subsea application. As
such, its diameter may not exceed 13 in. (for the Lamont application) and its
overall weight must be kept to a practical minimum. The latter feature is
necessitated by ultimate plans for retrieval of the generator from great ocean
depths.

Initial consideration of the radiation shield design suggested the use of

pressed and sintered tungsten. This was partly due to its high structural
strength, good thermal conductivity and high density. The most promising
material available (several manufacturers of malleable tungsten have been
contacted) has a quoted 92 per cent theoretical density. Preliminary shielding
experiments using lead were made and later, a tungsten equivalent thickness
calculated, based on a 100 per cent tungsten density. A one-curie source was
used for this determination. However, it was felt that results reported
earlier in the program, although carefully obtained, were subject to inaccura-
cies introduced by extrapolation from one curie to 28,000 curies. For this
reason, a larger source was procured from Oak Ridge National Laboratory.
To improve reliability, this source was aliquoted from the same massive
batch of radiocesium held in reserve for fueling the 5-watt operational generator.
This precaution was important since the amount of Cesium-134 present can
determine the amount of shielding required if it exceeds 1 per cent of the fuel
source. The assay on the 10-curie source indicated 1,5 per centcontent of the
shorter-lived isotope (2.4 y r . ) of cesium. Therefore, this fraction was used
in the shielding experiment.

In order to provide a safe working area, as well as to exclude scatter

radiation from the meter, the 10-curie source was shielded within a 2~in, -thick
lead cave, as shown in Fig, 19. The lead glass windows and castle manipulators
permitted safe handling of the source and attenuator disks. A heavy-walled

RRC-0102
 48

.^•P
<^ r - J

•; fe,.

.^S
F i g . 19 - E x t e r i o r and I n t e r i o r Views of Lead Cave
Used In Shielding E x p e r i m e n t s

RRCȤ102
 49

2000

1» ^ a t e r i a l
L 2 in.
T tiick i 10 i n >
Slick

^ 100« b t h e o r e t i c a . d e n s ity l e a d 0, 2 4 i n . 0. 79 in,

1000 \ •^ n
' \ ^ \. 8 6 % b v wfr . . 9 2 7i th<^OT" e t i r a l
\ \ ~..r4«-*.»-v-* ft "^O in 0
\ o FC*^^ rit-y-ttr CXgSt'&ti u. V,
^ . \
u \^ \
^ .
"^
N ^ \

\
\

ci i n t e r e
V
r
?ungst e n C a St Le< ^d
U
100
\ \
\~~~" _ _ ^
03
V, \
\ \
oi ^ ^ ^
0) \
to
^
o \
p
0) \
> \
>
•
to \
O \

\
10
- ^ ^ ^ \
Wieter \ ) V

E TTO' ^ X
r 1 1 1 i !
I • i o s o r i >er
\ ff
X nLie KUti i» » i l l i ut^acsj
\
0 .25 .50 .75 1.0 1.25 1.50

F i g . 20 - Attenuation of C e s i u m - 137 (Incl. 1, 5% C s - 134) by Tungsten and Lead

RRC-0i02
 Tes t A - 3.500 in. Ti mgste 1 + 0 . 250 in , 0« D »
Tes t B - 1. 138 in. Ti mgste 1 + 0 , 250 in . s.s.
200* I

" " \
100 v. \ ,
XV
N \
\
\
X
\ ,
\

) \.
\
Test i
\ T est B

\
10-
\
\
\
N\
\

\
\

X
1 Inches of Al uminu m Abs o r b e r
0 ][ 2 3 I\

F i g . 21 ~ Attenuation of C e s i u m - 1 3 7 (Containing 1. ^o Cs-134) by Aluminum

RRC-0102
 51

inner cave was so designed a s to p e r m i t only a c i r c u l a r beam of radiation to

e m e r g e from the configuration. The d i a m e t e r of the beam was controlled by
an i r i s l a r g e enough to expose the e n t i r e liquid contents of the s o u r c e bottle,
permitting an unshielded reading to be t a k e n . Approximately 3 in. outward
from t h i s c i r c u l a r orifice was a second orifice 0.5 in. g r e a t e r in d i a m e t e r .
Into this c i r c u l a r tunnel w e r e placed the attenuator disks used to d e t e r m i n e
the half and tenth-value l a y e r s for the C e s i u m - 1 3 7 , C e s i u m - 1 3 4 s o u r c e .
The enlarged d i a m e t e r of the second, continuing a p e r a t u r e , provided a
stepped radiation seal against s c a t t e r r a d i a t i o n . A calibrated ionization
c h a m b e r m e t e r was positioned at a d i s t a n c e of 4 ft. from s o u r c e center of
the e x p e r i m e n t . In this way, the m e t e r was protected from s t r a y radiation
which m a y have existed within the confines of the o u t e r m o s t shield. With the
beam hole plugged i n s i d e , the m e t e r r e g i s t e r e d no radiation field, verifying
the containment g e o m e t r y of the e x p e r i m e n t .

The p u r p o s e of this e x p e r i m e n t was threefold. We wished to e x p e r i -

mentally deterinine the combined, unshielded g a m m a radiation from an aqueous
s o u r c e of r a d i o c e s i u m , to find the half and tenth-value l a y e r of lead and a l s o ,
to obtain like values for s i n t e r e d tungsten. The unshielded dose at 4 ft, was
m e a s u r e d and found to be 2.00 r / h r . V a r i o u s t h i c k n e s s e s , first of lead and
then s i n t e r e d tungsten, w e r e then used to obtain attenuation c u r v e s shown in
Fig, 20, F r o m the lead c u r v e , a half-value thickness of 0 , 2 4 in. and a tenth
value of 0,79 in. w e r e obtained. S i m i l a r t h i c k n e s s e s obtained from the tung-
sten c u r v e w e r e found to be 0.20 in, (half-thickness) and 0,67 in, (tenth-
t h i c k n e s s ) . As indicated by the tungsten c u r v e , the slope continued on c o u r s e
up to the point that random r e s u l t s overshadow the values obtained. This is
unfortunate, since the c u r v e s did not reflect a resolution of C e s i u m - 1 3 4 g a m m a s ,
after attenuation of the softer C e s i u m - 1 3 7 g a m m a s . During the conduct of this
e x p e r i m e n t , duplicate values w e r e obtained for all points during attenuator
stacking and subsequent individual r e m o v a l of a t t e n u a t o r s .

Since it is difficult to a c c u r a t e l y anticipate the shielding r e q u i r e d to

r e d u c e the g a m m a r a y energy from a high kilocurie s o u r c e , s e v e r a l safety
factors w e r e introduced into our c a l c u l a t i o n s . The contribution to attenuation
made by the s t e e l r e f l e c t o r s surrounding the shield was omitted as well a s
that of the Hastelloy C c a n s . In a s i m i l a r m a n n e r , attenuation by the ( o u t e r -
most) thick aluminum p r e s s u r e shell, will also be left out of the p r e p a r a t i o n
of final shielding d i m e n s i o n s . Although t h e s e contributions to overall shield-
ing a r e valid, they will not be included in shield r e q u i r e m e n t s . However, it
is important to know how valid they a r e in reducing r a d i a t i o n , in o r d e r to
define them a s safety f a c t o r s . F o r this r e a s o n , a third shielding experiment
was r u n . P e r m a n e n t a t t e n u a t o r s (tungsten and steel) w e r e first positioned
to i n t e r c e p t the b e a m . Then varying known t h i c k n e s s e s of aluminum w e r e
added. R e s u l t s of this e x p e r i m e n t a r e indicated by the c u r v e s shown in Fig. 2 1 .

RRC-0102
 52

Here too, it can be noted that no break in the plot is seen, indicating that only
the Cesium-137 fraction was evaluated. Values for the half and tenth thick-
nesses of lead obtained in the foregoing experiments were subsequently veri-
fied by similar work performed at Oak Ridge National Laboratory. For their
work a 20-curie source was employed, obtained from the same kilocurie batch
as the 10-curie source used in the experiments described above.
Temporarily overlooking the inaccuracy of experimental results, a
shield thickness was determined for tungsten. For this calculation, co33amer-
cially available tungsten of 86 per cent purity (tungsten content) and of
92 per cent theoretical density, was considered. After extrapolating the un-
shielded dose to 25,500 curies, it was found that at 4 ft. some 5, 100 r / h r .
must be attenuated down to 100 m r / h r .

Using the experimentally obtained half and tenth values for sintered
tungsten, a thickness in excess of 3 in. would be required. Actually, the
amount needed would be a net value obtained after factoring-in the Cesium-134
contribution minus self-absorption and source geometry, Never-the-less,
3 in. of thickness is completely untenable, since it would result in a generator
of decreased thermal efficiency due to increased surface. Of greater con-
sequence is the fact that the resulting 5-watt device would have a diameter of
nearly 24 in. Original requirements have made mandatory a 13-in, maximum
diameter.

Although the density of uranium is not appreciably different from tung-

sten (W - 19.3 g/cc; U - 18.48 g/cc), its shielding properties are superior.
Moreover, uranium can be procured and machined in cast form of theoretical
density; x«achinable tungsten is only 86 per cent pure since it has lower Z
adulterants added to permit machining. Since tungsten must be pressed and
sintered, theoretical density can never be reached, the best being about 92
per cent. Although uranium in finely divided form is pyrophoric, in a massive
shape combustion is quite unlikely. Electroplating of nickel on all outer sur-
faces renders it noncombustible and provides a resistance to corrosion which
is far superior to tungsten. Depleted uranium is plentiful and has been used
before to provide high density ball swivels for shielded cave manipulators.
Since it has no other use, fabrication of shields would provide an excellent
outlet for the stockpiles which have accumulated over the past 15 y r s .

Before final consideration could be given the use of depleted uranium,

it was necessary to determine the attenuation factors of it. Shielding experi-
ments were performed at the Oak Ridge National Laboratory in a similar
manner to those performied at the Royal Research Laboratory, The same
20-curie source was used here, as was employed for the lead shielding expe-
riment described earlier. Although data has not been completely resolved.

RRC-0102
 53

the contract was provided with data showing half and tenth values to be 0. 11
and 0.37 in, , respectively. ORNL r e s e a r c h e r s reported that they were unable
to resolve the shielding plot into a two-component curve. Because of omiission
of this iiwportant factor, common to all experimental work, an analysis was
made of all contributing gamma rays from both radioisotopes of cesium. These
data, shown in Appendix C, include a consideration of buildup factors, internal
conversion coefficients, self-absorption and primary radiation. From this it
is seen that 2, 1 in, of depleted uranium (cast) would be adequate. The dose
rate at 4 ft. from the center of the generator, in a plane perpendicular to, and
bisecting the generator axis, would be 104 m r / h r . The effects of the 1,5 per
cent Cesium-134, a finite volume of fuel inaterial and dose buildup, are in-
cluded.

While the uranium alone would reduce radiation to a safe level, addition-
al contributions to shielding by the Hastelloy C cans, reflector cups and alum-
inum p r e s s u r e shell will tnake a further contribution. They represent a naar-
gin of safety and a r e not included in shielding calculations.

A final consideration is Cesium-134 content. The 1.5 per cent used

throughout the analysis was predicated upon an August use date, whereas the
generator will not be delivered until next spring. Decay of this 2, 3-yr, isotope
will reduce its curie strength to 83 per cent of the August value. This will
result in a reduction of shielding requirements.

RRC-0102
 54

V m . SUBTASK 2.5- CONCEPTUAL DESIGN O F GENERATOR

The b a s i c p a r a m e t e r s of a 5-watt, cesium-fueled t h e r m o e l e c t r i c

g e n e r a t o r have a l r e a d y been defined, based upon a d e e p - s e a application at an
intended operational depth of 3,600 fathoms. Containment to 6,000 fathoms,
c o m p a c t n e s s and e x t r e m e s t r u c t u r a l strength a r e emphasized. In addition to
the operational r e q u i r e m e n t s , comiplete containment of the nuclear fuel load
has dictated c e r t a i n d e p a r t u r e s from original design.

The fuel element configuration has been changed from seven i n c r e m e n t s

to an a r r a y of six. This was n e c e s s i t a t e d by limitations of the ORNL r e m o t e
welding e n c l o s u r e , which cannot accomimiodate the Hastelloy C can d e s c r i b e d
in the l a s t r e p o r t . Instead, two s e p a r a t e c a n s will be fabricated, each to con-
tain t h r e e fuel i n c r e m e n t s of equal s i z e . After c l o s u r e by welding, each can
will be sealed within a second one of the s a m e design and construction, thus
effecting double encapsulation of fuel i n c r e m e n t s . F i g . 15 shows this
modification.

Although r e v i s i o n of the shielding r e q u i r e m e n t s has r e s u l t e d in a change

from tungsten to depleted u r a n i u m m a t e r i a l for its construction, the general
configuration r e m a i n s the s a m e . This will consist of a hollow cylinder with
threaded c a p . After c l o s u r e , the two p a r t s will be locked by m e a n s of a taper
pin. Top and bottom will be beveled to r e d u c e weight and exposed surface.

One important design change involves repositioning of the shield within

the s h e l l , which will have a beneficial effect on t h e r m a l efficiency. This has
been accomplished by r e v e r s i n g the shield e n d s , so a s to locate the threaded
cap at the bottom, instead of at the t o p . Thus t r a n s p o s e d , the heat path will
consist of a continuous wall of u r a n i u m , froms heat source to power conversion
s y s t e m . The bottom location of the c a p - t h r e a d interface a i r gap will not
m a t e r i a l l y impede heat flow. Fuel cans will be loaded in such a manner as to
be in upright position when the g e n e r a t o r is completely a s s e m b l e d .

The change in fuel cladding, to conform to the Oak Ridge National

L a b o r a t o r y specifications of wall t h i c k n e s s and t o l e r a n c e s , has made n e c e s s a r y
an i n c r e a s e in dimiensions of all components e x t e r n a l to it. The inside d i a m e t e r
of the shield has been i n c r e a s e d to 2, 88 in, , n e c e s s i t a t i n g an outside d i a m e t e r
of 7. 07 in. V e r t i c a l d i m e n s i o n s of the shield have been i n c r e a s e d to 8. 81 in.
cavity height, 13.01 in. for o v e r a l l height. Initial t h e r m a l insulation r e q u i r e -
m e n t s w e r e p r e d i c a t e d upon a s m a l l e r shield. However, by increasing it,
additional felted f i b e r g l a s s t h i c k n e s s i s needed. This has been i n c r e a s e d from
3/4 in, c o m p a c t e d , to 1.25 in. , comipacted. S i m i l a r l y , the 7075-T6 aluminum
p r e s s u r e shell now m u s t be i n c r e a s e d to a 1 0 - i n . inside d i a m e t e r , with a

RRC-0102
 55

1.5~in. wall. The overall generator diameter has been held to 13 in. as first
promised the customer.

The thermioelectric conversion package has been reduced in size to opti-

mize the thermocouple array and reduce heat loss through the stainless steel
envelope. The dc-dc converter package is almost complete. Final closure
will be miade by joining the package to the top lid of the pressure shell.

RRC-0102
 56

IX. SUBTASK 2.6- HAZARDS EVALUATION

R e s e a r c h relating to the h a z a r d s evaluation comprises two a r e a s . The

first c o n s i s t s of t e s t s affecting the s t r u c t u r a l design a n d / o r efficiency of the
t h e r m o e l e c t r i c g e n e r a t o r a s an adequate m e a n s of containing the fuel load.
The second a r e a is r e s t r i c t e d to t e s t s of the type that identify the l i m i t s of
the final d e s i g n ' s endurance to n a t u r a l and accidental situations to which a
g e n e r a t o r could be subjected. Since safety is of p r i m a r y concern in the evo-
lution of a portable device employing high kilocurie quantities of C e s i u m - 1 3 7 ,
a connprehensive r e c a p i t u l a t i o n of all informsation pertaining to this subject
will be p r e p a r e d a s a s e p a r a t e r e p o r t . I n t e r i m work is d e s c r i b e d under the
subtask m o s t a p p r o p r i a t e to it, or where r e s u l t s affect the continuing effort.
F o r that r e a s o n data under this heading which is d e s c r i b e d e l s e w h e r e in the
r e p o r t , will not be r e i t e r a t e d h e r e .

The original scope of work of the h a z a r d s evaluation was devoted to

identification of a safe fuel compound, design of a s t r u c t u r a l l y - s o u n d package
and attenuation of radiation through it. P h y s i c a l t e s t i n g , in relation to credible
i n c i d e n t s , was to be r e s t r i c t e d to t h e r m a l insulation by b u r i a l . Recently, an
addition was made to the o r i g i n a l scope of w o r k . Experimentation was out-
lined to t e s t under r e a l i s t i c conditions the following p a r a m e t e r s of safety:

1. Biological Uptake Studies of Cesium P o l y g l a s s .

2. Radiation Effects on C e s i u m P o l y g l a s s .

3. P r e s s u r e Testing the G e n e r a t o r Shell,

4. G e n e r a t o r S u r f a c e - C o a t Evaluation.

5. G e n e r a t o r Shell Impact Evaluation.

1. Biological Uptake S t u d i e s .

The c o n t r a c t o r will f a b r i c a t e c e s i u m polyglass s o u r c e s in s i z e , quantity

and dose r a t e r e q u i r e d by t h e investigating l a b o r a t o r y . This m a t e r i a l will be
used to d e t e r m i n e the uptake of C e s i u m - 1 3 7 in nnarine plants and aniroials.
This work has not yet been initiated,

2. Radiation Effects Study.

D e s c r i b e d in subtask 1.4, this e x p e r i m e n t was concerned with the change

in c e s i u m p o l y g l a s s , after i r r a d i a t i o n by m a s s i v e g a m m a r a y dosage. Testing
showed that while optical change o c c u r e d , other physical p r o p e r t i e s w e r e

RRC-0102
 57

r e t a i n e d . Although solubility t e s t i n g i s i n c o m p l e t e , r e s u l t s indicate that (if

anything) the dissolution r a t e was d e c r e a s e d by the radiation effect. I r r a d i a -
tion was p e r f o r m e d by the High Level G a m m a Facility of Vallecitos Atonnic
Laboratory.

3. G e n e r a t o r Shell P r e s s u r e T e s t s .

Since little was known of how the shape and thickness of a 7075-T6
aluxninunrs v e s s e l would affect i t s r e s i s t a n c e to water p r e s s u r e at 9,500 to
15,900 p s i , , s p e c i m e n s w e r e subjected to t e s t s which would d e t e r m i n e this
f a c t o r . Reduced scale hollow c y l i n d e r s , whose lids w e r e sealed with a
single " O " r i n g , w e r e employed for initial work. P r e s s u r e welding of the
t h r e a d e d s u r f a c e s was sought, a s a m e a n s of p e r m a n e n t c l o s u r e against
t a m p e r i n g , A r a t i o of inside d i a m e t e r to wall and b a s e t h i c k n e s s was e s t a b -
l i s h e d . While p r e s s u r e welding did not o c c u r , galling made lid r e m o v a l ex-
t r e m e l y difficult.

The successful completion of r e d u c e d - s c a l e imodels led to the design

and testing of a full s c a l e g e n e r a t o r s h e l l . After being subjected to the water
p r e s s u r e anticipated at 6,000 fathoms (15,900 psi) for 3.5 h r s , , the specimen
showed no deforirjation of c r a c k s . The single " O " ring seal prevented any
water s e e p a g e , and the lid galled so tightly to the c a s e that d i s a s s e m b l y by a
t e a m of men was i m p o s s i b l e . All testing was p e r f o r m e d at the high p r e s s u r e
t e s t facility. Naval E l e c t r o n i c s L a b o r a t o r y , San Diego. All testing and data
gathering was p e r f o r m e d by authorized naval p e r s o n n e l .

4. G e n e r a t o r Shell Surface Coating.

S e v e r a l potentially a c c e p t a b l e noaterials w e r e given p r e l i m i n a r y t e s t s ,

f i r s t refluxing in s e a water for 24 h r s , , and then subjection to 10,000 p s i ,
w a t e r p r e s s u r e . Of the coatings t e s t e d , t h r e e r e s i s t e d attack satisfactorily:
h a r d anodized s u r f a c e with c h r o m a t e c o v e r , K e l - F and B i s o n i t e . Aluminum
slugs with surface coatings of t h e s e t h r e e m a t e r i a l s w e r e placed in shallow,
running sea w a t e r for a period of 90 d a y s . I n t e r i m r e p o r t s have been r e c e i v e d
from the Oceanographic T o w e r , Naval E l e c t r o n i c s L a b o r a t o r y , San Diego,
w h e r e the slugs a r e under t e s t . T h e s e r e p o r t s , which a r e based upon o b s e r v a -
tions through a s u b m a r i n e TV s c a n n e r , indicate that all specim^ens a r e coated
with colonies of m a r i n e plants and c r u s t a c e a n s but that c o r r o s i v e attack on the
coatings is not p e r c e p t i b l e . At the conslusion of t e s t s , the specimens will be
r e t r i e v e d and examined.

\
RRC-0102
 58

I
5. G e n e r a t o r Shell Impact Evaluation.
The p r e s e n t design of the 6-watt g e n e r a t o r r e p r e s e n t s the safe miniimum
of s t r u c t u r a l strength capable of withstanding a submerged depth of 6,000
fathoms. This is in keeping with the basic objectives of producing a g e n e r a t o r
of m i n i m a l size and weight. Its d i a m e t e r has been held to 13 i n c h e s , the
allowable m a x i m u m , while the weight is well under 500 l b s . Internal shield-
ing r e d u c e s the dose to a p p r o x i m a t e l y 100 m r / h r at 4 ft. It is obvious then,
that a c r e d i b l e incident, such as an a i r accident involving free fall to t e r m i n a l
velocity, cannot be r e a d i l y i n t e r p r e t e d by c a l c u l a t i o n s ,

A second f u l l - s c a l e aluminunn p r e s s u r e shell identical to the one used

for p r e s s u r e testing is being f a b r i c a t e d . When completed, it will be a i r
dropped, after which s t r u c t u r a l containment and the impact b u r i a l p a t t e r n will
be evaluated.

T e s t s p e r f o r m e d by original c o n t r a c t u a l a g r e e m e n t covered the p o s s i -

bility of a t h e r m a l e x c u r s i o n o c c u r r i n g as a r e s u l t of accidental b u r i a l of a
g e n e r a t o r in sand, or envelopment by b u r l a p or c a n v a s . The f o r m e r situa-
tion could obtain only a s a r e s u l t of impact after air d r o p , while the l a t t e r
(highly probable) could occur during s t o r a g e or t r a n s p o r t a t i o n of the device.
To be efficient, a t h e r m o e l e c t r i c g e n e r a t o r naust p a s s t h e r m a l energy rapidly
through its conversion s y s t e m . To do so, the g e n e r a t o r m u s t be in an en-
vironment of infinite heat capacity, whose ambient t e m p e r a t u r e i s lower than
that of the d e v i c e . All conceivable operating conditions would fuUfil this
r e q u i r e m e n t . Carefully planned s t o r a g e or t r a n s p o r t a t i o n methods would take
this into c o n s i d e r a t i o n and provide ventilation even in locations of high
t e m p e r a t u r e . However, accidental covering by burlap or canvas would cause
an insulation effect which might produce a c r i t i c a l t e m p e r a t u r e r i s e at the
outer skin and then move inward to the fuel c o r e . It was the purpose of these
t e s t s to d e t e r m i n e - - b y actual duplication of conditions - - the effect of a c c i -
dental insulation upon the g e n e r a t o r c o r e .

For the b u r l a p e x p e r i m e n t , the same configuration was used as r e p o r t e d

in the Second Q u a r t e r l y R e p o r t ( c o r e and shield mockup containing a 700-watt
c a r t r i d g e h e a t e r with output c o n t r o l ) . This was protected by double gold-
plated radiant heat r e f l e c t o r s and compacted, felted f i b e r g l a s s . The a s s e m b l y
was housed in an evacuated aluminum c y l i n d e r . Thermocouple p r o b e s w e r e
positioned i n t e r n a l l y , to indicate t e m p e r a t u r e s throughout the configuration.
This then, was wrapped in seven t h i c k n e s s e s of b u r l a p and covered by a c a n -
vas t a r p a u l i n .

RRC-0i02
 59

800

700
O

o
P
600
u
3
a
u
a
500
s
H
0)
u
o
O
400

300
0 10 20 30 40 50 60 70 80 90 100 110 120

F i g . 22 - C o r e T e m p e r a t u r e As a Function of E x t e r n a l Insulation

RRC-0102
 60

When the c o r e h e a t e r was e n e r g i z e d to an output of 120 t h e r m a l watts

(anticipated output of the Cesium~137 fuel c o r e ) , the core t e m p e r a t u r e r o s e
steadily, reaching a m a x i m u m of 820 deg, C. after 42 h r s . , as indicated in
F i g . 22. Since the melting r a n g e of c e s i u m polyglass is above 900 deg, C. ,
a modest safety m a r g i n i s provided. More important is the fact that the
e x t e r n a l surface of the t e s t v e s s e l was only 1/5 the anticipated surface of the
5-watt g e n e r a t o r . G r e a t e r surface would d i s s i p a t e heat m o r e r e a d i l y , even
in an insulated e n v i r o n m e n t . N a t u r a l l y , prolonged insulation at the t e m p e r a -
t u r e indicated above would r e s u l t in total d e s t r u c t i o n of the t h e r m o e l e c t r i c
conversion s y s t e m , but this is not a c r i t e r i o n of the h a z a r d s evaluation.

The second t h e r m a l insulation e x p e r i m e n t employed the same device

( d e s c r i b e d above) in a s a n d - b u r i a l situation, to d e t e r m i n e core t e m p e r a t u r e
a s a function of t i m e . To r e p r o d u c e n a t u r a l conditions, bags of sand w e r e
stacked around the a p p a r a t u s , after placing it upon s e v e r a l moist sandbags.
Loose sand was then packed between the b a g s , to fill existing air g a p s . C o r e
t e m p e r a t u r e again r o s e s t e a d i l y , r e a c h i n g a m a x i m u m of 625*^ C. after 90 h r s ,
of e l a p s e d t i m e . Bags of sand w e r e employed instead of bulk sand, to provide
an additional insulant in the b u r l a p , itself. This would simulate any organic
m a t e r i a l which might be p r e s e n t under actual t e r r a i n conditions. Absence of
the b u r l a p would, of c o u r s e , r e d u c e the t e m p e r a t u r e peak and provide a
g r e a t e r miargin of safety a g a i n s t fusion of the p o l y g l a s s .

E a r l i e r e x p e r i m e n t s have shown t h a t , if exposed to a i r , a g e n e r a t o r will

not e x p e r i e n c e a h a z a r d o u s t e m p e r a t u r e e x c u r s i o n , even when in a s t i l l ,
ambient a t m o s p h e r e maintained at 90 deg, F .

RRC-0102
 61

X. TASK m - DESIGN AND CONSTRUCTION OF THE MODEL

When the conceptual design of the generator had been formulated, the
design parameters were incorporated into a model. As per contractual agree-
ment, this model is full scale and in sufficient detail to illustrate all component
p a r t s , although not necessarily to final dimension. The model, with suitable
carrying case which can also be used as a portable pedestal, has been delivered
to the customer.

Designed in three separate increments, the model may be assembled in

such a inanner as to give an exploded view, using clear plastic spacers to
separate the increments. Thus, the original contractual requirements of con-
struction of a thermoelectric generator could be illustrated, apart from the
dc-dc power converter package. Threaded increments permitted integration
of parts to illustrate present plans for a 5~watt device capable of a 12-volt
output.

The basic generator, which includes the complete power generating system
and lid with power-outlet connector, is shown in Fig, 23. Interior components
are illustrated in relief by means of a pie-section cutaway. Six individual fuel
increments, with cutaway platinum cups, are innermost. External to these are
shown the two, double-walled Hastelloy C fuel cans. The radiation shield has
been quarter-sectioned, with the threaded lid in full round to illustrate dimen-
sional characteristics. A cutaway of the power conversion system (simplified)
is positioned directly above the shield, with several conceptual thermocouples
in full round and thermal insulation located between. The envelope surrounding
the power converter, also is simplified and is shown as a right cylinder, in-
stead of the bellows wall which actually will be used in the operational gener-
ator. The two gold-plated reflector cups which surround the radiation shield
may be seen as lines in the vertical and as full diameter disks at the base.
External to thetn is shown the conductive insulation. A thick pressure shell
surrounds the components, which being pie~sectioned to 120 degrees exposes
the generator interior. The threaded p r e s s u r e shell lid is provided with two,
concentric "O" rings, so that when joined to the shell proper, the full size of
the device is seen.

The fully integrated generator, including the dc-dc converter section, is

shown on page ii. While the latter section is not pie-sectioned, a view of the
integrated system is provided, when the plastic spacers are in position. Cool-
ing ribs may be seen on the circumference of the upper section-. When in the
exploded view, a thermal insulation layer also may be seen surrounding the
dc-dc converter package. Since this converter is being purchased outright and
not developed within the program, interior components have not been pie-sectioned.

RRC-0102
 62

F i g , 23 - G e n e r a t o r Model, L e s s DC-DC C o n v e r t e r A c c e s s o r y

RRC-0102
 63

A more basic reason for omitting a pie-sectioned view is because two applicable
but jnechamcally different converters are available and final selection cannot be
made until completion of prolonged life testing of both devices.

On the top of the lid may be seen two lifting rings. They a r e to be used
for lowering the generator into position on the ocean bottom. The male sec-
tion of a through-shell electrical outlet plug is positioned at the center of the
lid.

Since the purpose of the model is to illustrate the basic principle of this
device, components have been simplified to reduce cost, and exaggerated
where necessary, to permit visual identification from a distance. Color has
been used to delineate between different metals and alloys to be used in the
actual device. The exterior surface of the generator was painted to suggest a
protective coating over the aluminum shell. The nuclear fuel core is illumi-
nated by means of a dry-cell battery system. The switch is hidden at the upper
left corner of the generator cutaway.

In addition to the model and carrying case, the customer was also pro-
vided with descriptive literature to aid in making oral presentations. Color
slides of the model were prepared in 35 mm and lantern-slide sizes.

RRC-0102
 64

XI. SUBTASK 4. 1 - FABRICATION ANALYSIS

The conceptual design shown in the second quarterly progress report
is currently under re-evaluation, based upon certain changes which have since
been made. Machining details and assembly techniques are being formulated
in accordance with fabrication requirements. Data, derived by testing and
calculations in previous subtasks, is being reviewed. Various suppliers of
raw materials and certain fabricated parts have been contacted in an effort to
synchronize delivery of necessary items, which must be procured before
going ahead with dependent p a r t s .

An AEC permit has been requested, to allow possession by Royal

Research of the 25,500-curie fuel load, as well as the heavy-walled depleted-
uranium shield. Auxiliary equipment is being constructed, including a heavy-
duty vacuum pumping system. This will be used to evacuate the fiberglass
insulation and thereby enhance heat economy within the generator.

Coordination has been established with the Lament Geological Observatory,

so that the electrical leads and physical shape of the generator will meet the
requirements of all associated seismographic equipment.

Initially, this subtask is concerned with identifying any necessary devia-

tions to the conceptual design which may become necessary as the result of
final testing, or changes required in certain generator components. During
generator assembly, surveillance will be nnaintained to anticipate needed
modifications and thereby save time by obviating rework effort.

RRC-0102
 65

XIL SUBTASK 4. 2 - FUEL ELEMENT FORMING & GLADDING

A change in p r o g r a m scope has r e s u l t e d in redefinition of the facility

r e s p o n s i b l e for fuel p r o c e s s i n g . The Vallecitos Atomic L a b o r a t o r y , General
E l e c t r i c Company, will not p r o c e s s the 25,500 c u r i e s of C e s i u m - 1 3 7 , Instead,
this work will be p e r f o r m e d at Oak Ridge National L a b o r a t o r y . The project
has been notified that Royal R e s e a r c h is to provide the following i t e m s for fuel
p r o c e s s i n g and encapsulation:

1. Seven platinunn fuel c u p s ,

2. T h r e e complete double-walled fuel cans and c a p s ,

3. Sufficient ion exchange r e s i n to p r o c e s s fuel.

4. A suitable fusion f u r n a c e , with b l u e p r i n t s .

5. A suitable s l u r r y m i x e r , with b l u e p r i n t s ,

6. Complete fuel p r o c e s s i n g data.

Oak Ridge r e p r e s e n t a t i v e s will r e s e r v e the right to utilize either a

c e r a i n i c - w a l l e d , platinunn fuel cup, or a full platinum cup. If the l a t t e r is
employed, they will provide a cutting wheel to r e m o v e e x c e s s platinum wall
after fusion. The cups will be spun and will have 0, 004-in. -thick walls and
0 , 0 0 5 - i n . -thick b a s e s . Although only six a r e r e q u i r e d , one e x t r a will be fur-
nished as a p r e c a u t i o n a r y m e a s u r e .

T h r e e s e t s of double-walled Hastelloy C fuel cans will be fabricated to

the r e q u i r e d d i m e n s i o n s , with the b a s e s sealed by welding. All t h i c k n e s s e s
and t o l e r a n c e s will be to specifications supplied by ORNL. In addition,
Hastelloy C s p a c e r s of t h i c k n e s s e s varying from 0,032 to 0, 125 in. thickness
will be provided. Shaping will be achieved through the use of s e a m l e s s tubing.
If unavailable, rolled sheet may be substituted for tubing, provided it is
acconnpanied by X r a y certification. Machine cutting will not be employed;
i n s t e a d , high speed grinding will be used to fabricate cup p a r t s to dimension.
All welds will be s t r e s s r e l i e v e d . Final c l o s u r e will be performed at ORNL,
after fuel loading.

Royal R e s e a r c h has developed a high t e m p e r a t u r e furnace for f a b r i c a -

ting the c e s i u m polyglass fuel, as .shown in Fig, 24. The lower section may
be swung away from the rigid top and loaded with a fuel cup. The fuel charge
can then be added as shown in photo A. Next, the lower section is r e t u r n e d
to a position d i r e c t l y under the top, photo B , and elevated tight against it by

RRC-0102
JO
O
I
o
»-•
o

F i g , 24 - High T e m p e r a t u r e F u r n a c e Used for Cesiunn P o l y g l a s s Fabrication

a-
 67

^^U^^f^^-

Fig, 25 - Electrically Driven Cesium-Fuel Mixer

RRC-0102
 68

m e a n s of a r e m o t e l y operated s c r e w d r i v e , a s in photo C. A s t a i n l e s s s t e e l
off-gas vent that p r o j e c t s from the furnace top, c a r r i e s away m o i s t u r e and
fusion g a s e s . F u r n a c e t e m p e r a t u r e is controlled by means of a r h e o s t a t
mounted e x t e r n a l to the hot c e l l . Heat i s provided by special n i c h r o m e h e a t -
ing e l e m e n t s wound around a c e r a m i c c y l i n d e r . T h e r m a l insulation is p r o -
vided at the sides and b a s e of the furnace. This type of device has been used
by Royal r e s e a r c h e r s to fabricate c e s i u m polyglass t e s t specimens and has
proven r e l i a b l e and efficient.

A second piece of h a r d w a r e designed by the contracting l a b o r a t o r y and

under developmient, is an e l e c t r i c a l l y driven a u g e r . Its purpose is to mix an
aqueous solution of C e s i u m - 1 3 7 c a r b o n a t e with stable complex b o r o s i l i c a t e s .
Water is r e m o v e d by heating during the mixing p r o c e s s . As the s l u r r y goes
first to a p a s t e and then to a dry m i x , the revolving s c r e w action b r e a k s up
lumps and finally r e d u c e s the contents to a d r y , finely divided and intimately
mixed p r o d u c t . Since this is a complex m e c h a n i s m , considerable development
t i m e is needed to perfect it. Components which c o m p r i s e the basic machine
a r e shown in F i g . 25, Power is derived from a variable speed, d. c, Bodine
miotor. This p r o p e l s a s p i r a l a u g e r m e m b e r housed within a hopper. When
the hopper is inclined, the s l u r r y is forced up the r a m p in the i n t e r i o r of the
hopper, whereupon it r e t u r n s to the r e a r by the pushing action of the auger on
the s l u r r y . Heat i s provided to the hopper walls by externally-mounted heating
e l e m e n t s . C o n s i d e r a b l e flexibility is provided, permitting the angle of eleva-
tion, or d e p r e s s i o n to be changed, the a u g e r speed v a r i e d , and its d i r e c t i o n
r e v e r s e d . The hopper may be tilted for dumping and the e n t i r e a p p a r a t u s
rotated on its b a s e for convenient charging with the dry and liquid constituents.

P r o b l e m a r e a s include the shape of the hopper, the position of the auger

and redundancy of motion. In addition, the s e a r c h for a suitable bearing
miaterial has been e x t r e m e l y difficult. Such p a r a m e t e r s as r e s i s t a n c e to r a d i a -
tion, heat and the a b r a s i v e action of the b o r o s i l i c a t e s have damaged all b e a r -
i n g s . An offset chain drive proved to be i m p r a c t i c a l , when dry product caked
in the chain and caused b r e a k s in the chain l i n k s .

S e v e r a l highly-successful r u n s have been made with the e x p e r i m e n t a l

a u g e r - h o p p e r yielding a finely divided, dry powder of uniform p a r t i c l e s i z e ,
after only 2 h o u r s running t i m e . The l o s s of product on the walls was l e s s
than 3 p e r cent. T h i s would not become an additive l o s s , for each of the six
f u e l - p r o c e s s i n g r u n s since subsequent addition of aqueous c e s i u m carbonate
would r e d i s s o l v e any dry r e s i d u e . This would be further aided by the a b r a s i v e
action of the b o r o s i l i c a t e p a r t i c l e s in the s l u r r y .

RRC-0102
 R e s i n Dimensions
CsCl d. ~ 4, 84 cna.
Dissolver h, = 50,40 c m .

169 g. Blender Fusion Cup

in Dryer Cup T r imming
^ -^ -^
200 m l . and
In: 165 g. CS2CO3 (Aq) d, - 6. 35 cm,
165 g. B o r o s i l i c a t e s h. ~ 3. 56 cm. Transfer
Out: S a m e to
T e m p : 70° C. (Under Construction) Generator
T i m e : 30 m i n .
Heating Cycle Cooling Cycle
2 h r s , @ 600° C, To 650° C, in five, lOO-deg.
?3 4 h r s , @ 1150° C. steps of 30 m i n . duration
aI for e a c h . Total elapsed
t i m e , 2. 5 h r s .
o
o Volume in: 200 rril. CsCl
+265 m l . wash w a t e r
Volume out: 465 m l ,
Elution t i m e : 105 m i n .

Fig, 26 - Cesium Polyglass Flow Sheet

 70

Because of the connplexity of the auger-hopper approach to mixing and

drying, alternative methods are also under evaluation. One particularly
promising approach involves rotating knife blades within an externally heated
hemispherical bowl. Constant rotation of the specially shaped blades con-
tinually blends the slurry, then breaks up the lumps and finally, reduces par-
ticle size by erosive action of the particles ttemselves. Water is driven off by
a heater which surrounds the bowl. Although somewhat slower than the auger
method, basic simplicity of the concept suggests easier reproducibility.

The revised fuel-process schedule appears in Fig, 26 and includes such

improvements as the elimination of a cesiuni cleanup step and reduction of
eluent from the carbonate conversion column. This, in turn, results in the
elimination of a liquid-volume reduction vessel since the carbonate eluent may
be fed directly to the borosilicate mixer. This is shown as a broken, lined box
since the apparatus has not been selected as yet. Changing from a seven incre-
ment system to one containing six incre ments of fuel has caused an increase in
processing reagents. Despite this, the overall process requires less time
than that reported earlier in the program.

RRC-0102
 71

APPENDIX A

Cesium-137 Energy Calculations

The numiber of c u r i e s of Cs r e q u i r e d to produce 120 watts of

t h e r m a l energy depends on the a v e r a g e energy r e l e a s e p e r d i s i n t e g r a -
tion and the p e r c e n t a g e of this energy deposited within the radiation
shield. Calculation of the l a t t e r p r e s e n t s no p r o b l e m . All the b e t a s
will be e a s i l y stopped within the shield and since it i s designed to
provide many tenth-value l a y e r s for the 662 kev g a m m a r a y s , it is
safe to a s s u m e that e s s e n t i a l l y all the g a m m a energy will be available
t h e r m a l l y . It should be noted that although the 662 kev gammias a r e
p a r t i a l l y c o n v e r t e d , the r e s u l t a n t e l e c t r o n s a r e m o n o e n e r g e t i c , with
no accompanying n e u t r i n o , as in t r u e beta d e c a y . T h u s , assumption
of complete availability of the g a m m a energy r e m a i n s valid, r e g a r d -
l e s s of the actual c o n v e r s i o n coefficient.

Calculation of the a v e r a g e e n e r g y p e r disintegration is not s t r a i g h t -

forward b e c a u s e of the continuous s p e c t r u m c h a r a c t e r i s t i c of beta decay.
Since the energy c a r r i e d away by the neutrino in each disintegration i s
l o s t , an a v e r a g e b e t a energy m u s t be deternnined« F o r t h i s , the shape
of the s p e c t r u m and the end-point (maxinnum) energy m u s t be known.
If t h e r e i s m o r e than one b r a n c h , a s in the c a s e of C^^''', the branching
r a t i o s also a r e r e q u i r e d . F r o m a s e a r c h of the l i t e r a t u r e ^ the best
values for end-point e n e r g i e s *1 and branching r a t i o s ^ and ^ w e r e
d e t e r m i n e d . T h e s e a r e shown on the decay s c h e m e below*

137 1176 Kev

5 14 Kev
92%
137m 662 Kev
Ba

I T

137
\1/ Ba

RRC-0102
 72

Data on a v e r a g e beta energiffifound in the l i t e r a t u r e a r e s p a r s e ,

with the following r u l e - o f - t h u m b frequently quoted;*'* ^ ^ E = 1/3 '^xnax'
That t h i s is a r a t h e r r i s k y a s s u m p t i o n , i s amply d e m o n s t r a t e d by
M a r i n e l l i et, a l . " , whose compilation shows a substantial s c a t t e r of
values about the figure of 1/3 for the well-behaved "allowed" t r a n s i t i o n s .

Neglecting s c r e e n i n g effects, , we have for any beta spectruttj;

w h e r e /y{^)(^^l ' the n u m b e r of b e t a s with m o m e n t a between V and 7? + c£ H

V' -^^ : r e l a t i v i s t i c m o m e n t u m in units of mc
^ mC ^
lA/ r tJJlLE^ :. total energy in units of m c ^
E - kinetic energy
\Ao = -• °—^ ^ — - total e n e r g y c o r r e s p o n d i n g to E = EQ= '^nA'X,
and IAJ = T^ -f / i s the r e l a t i o n betweenil/and /?
p (i)f^)- r e l a t i v i s t i c n u c l e a r coulomb factor
/7^(H/o'*H^lf^==statistical weight factor
C (V^ TJQ 'Z) = shape c o r r e c t i o n factor depending on the type of
transition
The a v e r a g e b e t a e n e r g y , E , i s then defined a s : „

The factorF('£^jis r a t h e r complicated and includes a P function

of complex a r g u m e n t . F o r t u n a t e l y , this function has been tabulated
*^ in the ioxTt4i;:])q^F(2;q)/(fi(Z), for 2 = I to 100 and a sizeable range of
Y[ . (The value of Z to be u s e d in the t a b l e s is that of the daughter
n u c l e u s , Z s 5 6 in this c a s e . ) The above i n t e g r a l s thus simplify
somewhat to: K

and n u m e r i c a l integration can r e a d i l y be c a r r i e d out. F o r the c a s e of

f-(i,Tl)-'C(J},l):,^2)= I , which is a fair approximation for allowed t r a n s i t i o n s
in low-Z n u c l e i , the i n t e g r a t i o n s , though somewhat t e d i o u s , can be
c a r r i e d out d i r e c t l y .

RRC-0102
 73

Cesium-137 E n e r g y Calculations

The shape f a c t o r , C , for allowed transitions» is unity. F o r the

514 Kev beta of Cs^^l a unique f i r s t - f o r b i d d e n t r a n s i t i o n , (.\- (Wf^-lA/J^'^
(W'^-l) *2 and 9 and for the 1176 Kev b e t a , a second-forbidden t r a n s -
ition, CzT-'C^/s-^i^+ft a3(^'^-/^ *2 and 9 . R e f e r r i n g to the shape factor
for t h e 514 Kev b e t a , we note that it a u g m e n t s both the high and low
energy ends of the s p e c t r u m r e l a t i v e to an allowed transitions while
the shape factor for the 1176 Kev beta enhances mainly the low energy
contribution of the s p e c t r u m .

Using NBS t a b l e s *8 ^ ^-^^ the above shape f a c t o r s , the a v e r a g e beta

e n e r g i e s for both b r a n c h e s w e r e d e t e r m i n e d by n u m e r i c a l integration
and the r e s u l t s tabulated below.

Em ax % E E/E„^ax.
514 Kev 92% 183 Kev ,356
1176 Kev 8% 283 Kev .240

The only value for E/Ep^g_^ found in the l i t e r a t u r e was a value of

0,367 (*192/523) for the 514 Kev b r a n c h that a p p e a r s unreferenced in
an ORNL compilation of fission product p r o p e r t i e s *10^ As was
suggested by the c o m m e n t on the 1176 Kev shape factor j, the calculated
E/E evidences a substantial energy d e p r e s s i o n .
max.

Summary

The r e s u l t s of the a v e r a g e b e t a energy calculations and the a s s u m -

ption of e s s e n t i a l l y complete deposition of the unconverted 662 Kev
g a m m a s within the radiation shield indicate a value of (0. 92) (183) +
(0,08) (283) + (0.92) (662) = 800 Kev for the a v e r a g e available energy
per disintegration.

Since P (watts) * 5.94x10" x M e v / d i s i n t e g r a t i o n x C u r i e s , we find

that 120 watts of t h e r m a l energy r e q u i r e s 25,300 c u r i e s of C s ^ ' ' .

RRC-0102
 74

^References

1. Y. Yoshizawa. Nucl. P h y s . 5, 122 (1958).

2. L . M . L a n g e r a n d R . J . D . Moffat, P h y s . R e v . 82, 635(1951).

3. T . K a t o h , M, Nozawa^ Y. Yoshi^awa^ and Y, Koh, J , P h y s . Soc.

(Japan) 12, 738 (1957),

4. I, Kaplan, " N u c l e a r P h y s i c s " Addison-Wesley Publishing Co. , I n c . ,

Reading, M a s s , (1955) p . 301.

5. S, G l a s s t o n e , " P r i n c i p l e s of N u c l e a r R e a c t o r Engineering"
D, Van N o s t r a n d Co, , I n c . , P r i n c e t o n , New J e r s e y (1955) p . 68,

6. L . D . M a r i n e l l i , R. F , Brinkerhoffj a n d G . J . H i n e , R e v s . Mod.
P h y s . 19, 25 (1947),

7. K, Siegbahn, ed. , " B e t a and G a m m a - R a y Spectroscopy-,"

Inter science P u b l i s h e r s , I n c . , New York (1955) p . 282.

8. National B u r e a u of S t a n d a r d s , " T a b l e s for the Analysis of Beta

Spectra,"NBS Applied M a t h e m a t i c s S e r i e s - 1 3 j U . S . Government
P r i n t i n g Office, Washington, D . C , (1952).

9. R . A . R i c c i , F h y s i c a XXIII. 693 (1957).

235
10. J . O , B l o m e k e , ed, " N u c l e a r P r o p e r t i e s of U Fission P r o d u c t s , "
ORNL 1783 (1959).

RRC-0102
 75

APPENDIX B

Calculation of T e m p e r a t u r e Distribution in Uranium Shield

Summary

The possibility of "hot s p o t s " developing in the uranium shield of the

Royal R e s e a r c h 5-watt C e s i u m - 1 3 7 t h e r m o e l e c t r i c g e n e r a t o r has been inves-
tigated a s r e p o r t e d in this note.

The a s s u m p t i o n s n e c e s s a r y to provide a mathematically t r a c t a b l e

solution a r e d i s c u s s e d and justified. Sample calculations show a maximum
possible t e m p e r a t u r e differential of 140° F , from the cooled top surface of
the uranium shield to the hottest point in its i n t e r i o r .

Introduction

The uraniuna shield of the Royal R e s e a r c h 5-watt Cesium-137 g e n e r a t o r

is shown in the sketch below. The beta and g a m m a decay energy of the
8. 8 - i n , - h i g h fuel cylinder containing 25,300 c u r i e s of C e s i u m - 1 3 7 , produces
120 watts of t h e r m a l power within this shield. The division of energy between
the beta and g a m m a components is calculated as follows: Of the . 800 Mev
r e l e a s e d , on the a v e r a g e , p e r C e s i u m - 1 3 7 disintegration (see section on
C e s i u m - 137 energy c a l c u l a t i o n s ) , (0, 92) (, 183) + (0, 08) (.283) = . 19 1 Mev is
apportioned the betas a r i s i n g from t r u e beta decay. About 10% of the , 662
Mev gamnaas a r e i n t e r n a l l y converted
and e m e r g e a s monoenergetic c o n v e r -
sion e l e c t r o n s , bringing the total
a v e r a g e beta energy per d i s i n t e g r a t i o n
to .252 Mev, The r e m a i n d e r , or , 5 4 8
Mev per d i s i n t e g r a t i o n , is the g a m m a
component. Since the r a n g e of the
betas is s m a l l c o m p a r e d to the fuel
cylinder d i a m e t e r and the major p o r -
tion of the g a m m a r a y s will escape
the fuel c y l i n d e r , we will a s s u m e the
following behavior for the spatial
deposition of the n u c l e a r radiation:

RRC-0102
 76

1. All the beta e n e r g y (32% of the total) i s deposited within the fuel
cylinder,

2. All the g a m m a energy (68% of the total) is deposited exponentially in a

r a d i a l d i r e c t i o n in the u r a n i u m shield.

We can now w r i t e the equation governing the t e m p e r a t u r e distribution

within the u r a n i u m shield, which i s simply the time-independent, non-
homiogeneous heat conduction equation:

w h e r e S is the s o u r c e t e r m a r i s i n g from the exponential deposition of g a m m a

energy within the shield.

Assumptions

1. The n a t u r e of the s o u r c e t e r m and the solution of the p a r t i a l differential

equation a r e c o n s i d e r a b l y simplified if we extend the cesium fuel cylinder
the e n t i r e height of the u r a n i u m shield without changing the cesium loading
per unit length or the fuel cylinder d i a m e t e r . This m o s t certainly p r o -
duces a w o r s e situation than encountered in the actual g e n e r a t o r since we
a r e now generating (120) ( 1 3 / 8 , 8 ) " 177 watts of power, instead of 120.
This g e o m e t r y is shown in t h e sketch below.

2. If we neglect axial conduction along the fuel cylinder, then the heat m u s t
flow through the annulus a < r < l b at the top, a s m a l l e r a r e a and t h e r e f o r e
r e q u i r i n g a l a r g e r tennperature gradient than in the actual c a s e .

3. If we a s s u m e the bottom and

sides of the u r a n i u m shield to
be perfectly insulated, we r e -
quire all the heat to flow through
the annulus a ^ r ^ ^ b , which again
requires a larger temperature
gradient.

4. The top of the cylinder a ^ r < i b ,

z - h, i s a s s u m e d to be at a
constant t e m p e r a t u r e . T h i s is
plausible since the r e s i s t a n c e to >A
heat flow a c r o s s the interface at
z = h will be g r e a t e r than the

RRC-0102
 77

r e s i s t a n c e to heat flow in the r a d i a l direction at z - h. In any event,

the t e m p e r a t u r e d i s t r i b u t i o n in the a r e a of g r e a t e s t i n t e r e s t , near the
bottom of the u r a n i u m shield, will be fairly insensitive to any d e p a r t u r e
from this a s s u m e d boundary condition.

Statement of P r o b l e m

The above a s s u m p t i o n s allow us to completely define the problem m a t h -

e m a t i c a l l y . The differential equation i s :

and the boundary conditions a r e : ^

d T fa^z) = -Q^ ^2) aT (X o) --0 (4)

TJT ' r r / I , / 7 j =75 f a constant ) (5)

bn
. 2,
where Q r = g a m m a energy flux ( B T U / h r in )

Q A ~ heat flux into u r a n i u m shield from betas deposited in fuel

cylinder ( B T U / h r in^)

/A. ~ l i n e a r energy a b s o r p t i o n coefficient for g a m m a radiation (in"*)

h = t h e r m a l conductivity ( B T U / h r in ° F , )

h ' height of cylinder (in.)

O. = inner r a d i u s of cylinder (in.)

•&• ~ outer r a d i u s of cylinder (in.)

The above p r o b l e m is a m e n a b l e to solution by a s e p a r a t i o n - o f - v a r i a b l e s

technique which p r o c e e d s , briefly, as follows:

(The m a i n p r o b l e m is f i r s t broken up into two s u b - p r o b l e m s . )

RRC-0102
 78

Sub-problenn 1

Differential equation:

b T>(r,,2) j _ _ JlOt.2) Qi, (n,z) ^ f. (5)

with boundary conditions:

^,^^^•^^=0 (7) 411^^=0 (9)

p i ^ ^ = 0 (8) ];(/t^^;) =75 (10)
^ /t

Sub-problem 2

Differential equation:

with boundary conditions:

|M-,Z^ -^/^ ,12) 4^^-^"^--0 (14)

^(^^^) .0 (13) 72r'/?,i9;.-o (15)

The sum of the solutions to (6) and(11) with their r e s p e c t i v e boundary

conditions, is then the solution to (1) with boundary conditions (2-5). That i s :

rryz^Z)-T,C/t,ZJ-FTa (/I rd) (16)

Now the solution to (6) is obviously

T, fji^z) - T3 (17)
The solution to (11) is obtained by f i r s t applying the separation of v a r i -
ables technique to the homogeneous p a r t of (11). The c h a r a c t e r i s t i c functions
which r e p r e s e n t the z dependence of the product solution and which satisfy
homogeneous boundary conditions (14) and (15) are?

RRG-0102
 79

Since t h e s e functions form a complete orthogonal s e t , we a r e permitted to e x -

pand any piecewise differ^ntiable function in t e r m s of t h e s e orthogonal c h a r a c -
t e r i s t i c functions. It should be pointed out that the original problenn could have
been b r o k e n up in a different fashion, such that the c h a r a c t e r i s t i c functions
r e p r e s e n t i n g the r a d i a l dependence with a p p r o p r i a t e homogeneous boundary con-
ditions w e r e orthogonal, and then the expansions would be c a r r i e d out using
t h e s e c h a r a c t e r i s t i c functions i n s t e a d . F o r the c a s e at hand, we let

2 (18)
A/t '"^ n'O
77

and
oo

In i- I
r^(^.z) - ^ in (/I) Or?
h ^
(19)
n -o

The S ( r ) a r e r e a d i l y shown to have t h e form

(20)

and the r e s u l t of substituting |18) and (19) into (11) is

oo r-

£
n-o
n
dtp(^}
S.K
~{
(An+ I
ffU/i}^5jn) Z h ^
(21)

Now t h e functions t (r) appearing in (19) can be found by solving the

s e c o n d - o r d e r , non-homogeneous (ordinary) differential equation involving
t h e s e functions, which a p p e a r s within the b r a c k e t s in (21), The solution is
obtained by t h e v a r i a t i o n of p a r a m e t e r s technique which y i e l d s :

UCM)^[P,^-^ U^(ni'\Ko(^^ J nU ~P,r.iVr.(nl\IJ'-f>L^I^)

h
(22)
w h e r e the A^ and B a r e constants which a r e d e t e r m i n e d by the two remaining
boundary conditions (12) and (13)_,the IQ and KQ a r e z e r o - o r d e r modified B e s s e l
functions of the f i r s t and second kind, r e s p e c t i v e l y . The n a t u r e of the u^(r)
and the V|5^(r) will be d i s c u s s e d s h o r t l y .

RRC-0102
 80

Thus the solution to our original problem, from (16), (17), (19), and
(22) is: ^^
rr/i,2)^ro^^tt'^^^ + ^n(n)]l<o(-^i^?^j+[Bn+Vn(/ljllJ2^f/7jl
ti-o "*• "" ~ J
X ooo-l^ I 2 (23)
This solution v^ould be amenable to straightforward calculation, using
available tables and modified Bessel functions, were it not for the nature of
the u^(r) and V|i(r). The latter are of the form

Un(n)'-£„j''l,(^tjr^,)g--(^'''=^)^^, (24)

and
n
^.(n)^-f^/^ /<o(^{j^')er^^'''-'^^ JA.^ (25)

where C and the E^ a r e constants. It appears that (24) and (25) cannot be
integrated directly and numerical procedures would be required. It turns out
that, if the problem had been broken up such that the orthogonal functions in
which we expanded were the r-dependent set mentioned earlier, there would
again arise integrals involving the product of Bessel functions and exponentials.

SinDplified-Solution Technique

All the foregoing assumptions and solution techniques are equally appli-
cable if we unfold the cylinder into a slab of the same thickness and height.
The fact that the cross-sectional area perpendicular to the cylinder axis is now
reduced, because of the unfolding, makes the resistance to heat flow in both
the r and z directions even greater and therefore requires an even larger
temperature gradient in both directions. With this modification the problem
becom^es easily manageable since the solution now involves only hyperbolic
and trigonometric functions and ordinary exponentials. Most important, how-
ever, is that each assumption and departure from the actual generator con-
figuration has been shown to require a larger temperature gradient than will
be met in reality and thus will make any so-called hot spots appear even hotter.

RRC-0102
 81

Equation (1) now becomes

with the boundary conditions corresponding to (2)-(5)

d T (o, g) -_^/3 (27) ^ T (x,o; . 0 (29)

a7YZ_,Z) = o (28) T (-x^ h) ^Ts (30)

ax
The nomenclature is the same as before, with the exception of the
independent variable r , which is now replaced by x| and the interval a ^ r ^ b
now corresponds to 0 < x ^ L , The solution procedure follows identically
that which was outlined previously. The main problem is again split up into
two sub-problems.

Sub-problem 1

Differential equation

^ Ti(x,Z) yTi(x,Z)_r^ (31)

^X2 ^" T z ^ "^

with boundary conditions

ir^(0^2) ^__r) (32) a 1 i(x,o) ^ Q (34)
;^x ^ ^^^ az
^T> a . z ; - o (33) T^(x,h)-T^ (35)
ax
Sub-problem 2

Differential equation

(36)

RRC-0102
 82

with boundary conditions

(39)
(37)

(38)
bX (40)

Where J (">( > Z ^ = J^ (y^ Z j + T^ ( X, 2) (41)

Again we find that J"^ f X ^ 2 ) = 7 ^ (42)
F o r the solution of (36), we use the s a m e device of expanding in t e r m s
of the orthogonal c h a r a c t e r i s t i c functions

T)-- O^l.Z^. .
2 ^
obtained from a s e p a r a t i o n - o f - v a r i a b l e s solution of the homogeneous p a r t of
(36) and satisfying boundary conditions (39) and (40).

Thus we let
(43)
T) -O

and M Qr e""-^^" '^-

^^ 5 ^ ()() cw 2J211J1 Z (44)
A h =o Z h

The S„(x) a r e r e a d i l y shown to be

n -A-x
5m(x) -~J:^^J:itlL--^ (45)

We now substitute (43), (44), and (45) into (36), with the r e s u l t :
'n^ -jix
?«o
nn^o
(46)

RRC-0102
 83

The tjj^(x) of (43) are then found by solving the second-order, non-
homogeneous (ordinary) differential equation bracketed in (46), with the result:

2 h Z h
where
2 %>c (-/; n
^'^ Ah{{l22tl A^^^/iiZirtlJk (48)

and the Aj^ and Bj^ a r e constants determined by substituting (47) back into (43)
and satisfying the two remaining boundary conditions (37) and (38),

From (41), (42), (43), and (47) we obtain the final solution:
•oo _

TiK. z) - Ts + ^lA^ cxoA Iznll I {L--X) + dr.^y^^^^ ^ii±LI(t-V-foCr,e ~^^

5 /, ^ (49)
where

h (Zr> i-f T]\.M^ Z^i-I IL (48)

•J L h
Z h

Urrf (50)
2 /?

/n

(5i)
A L
h 2 h

RRC-0102
 84

Calculations

Values of the required parameters are as follows:

^ = 13 in.

Z. = 1.88 in.

A ~ 1.66 BTU/hr. in. ° F , (for T = 900 ° F.)

yU = 6.2 in. "1

Q^ = 3.47 BTU/hr. in. ^

6?/3= 1.65 BTU/hr. in. 2

The first five of each of the coefficients appearing in (49), i . e . , (48),

(50), and (51) have been calculated and are listed below.

Ao = + 143.0 BO =+19.2 x 10-5 ( ^ ^ r -0.431

Ai = -4.93 B j = - 2 , 14 X 10"5 c/j = +0. 144

A2 = + 0.932 B2 = +0.777 X 10"5 c/2-~0-0S7

A3 = - 0.285 B3 = - 0 , 3 9 6 x 10-5 ^ ^ - +0.0622

A4 = + 0, 107 B4 = +0.245 X 10"^ cX^4 = -0.0494

By inspection of (49) it is readily seen that the maximum temperature ,
as might be expected, is to be found along z = O, the bottom surface of the
generator. Two points at the bottom of the shield were chosen for a teiMpera-
ture calculation, T ( 0 , 0 ) next to the fuel cylinder, and T(L,0) at the outer
wall of the uranium shield.

Using the first five t e r m s of (49), the results a r e :

T ( 0 , 0 ) = Tg + 141° F .

T(L,0) = Tg + 139° F .

The sum of the first five t e r m s is estimated to be well within 1 per cent
of the infinite series solution for O^x^L, and z = O,

RRC-0102
 85

Thus the maximum possible temperature in the uranium shield wfl.1 be

about 140° F . above the temperature Tg of the cold face. From the most
pessimistic bias of all of our assumptions, it is reasonable to expect the
maxinnum temperature differential in the actual generator to be less than
100° F. And finally, it should be mentioned that since the thermal conduc-
tivity of uranium actually increases with temperature, in contrast to most
metals, this will tend to flatten the temperature gradient even more.

RRC-0102
 86

APPENDIX C

Shielding Calculations

Summiar>

The dose r a t e 4 ft. from the c e n t e r of the cesium g e n e r a t o r on a plane

p e r p e n d i c u l a r to and bisecting the g e n e r a t o r axis is calculated to be 105
m r / h r . The effects of the 1.5% C e s i u m - 1 3 4 contamination, finite volume of
fuel s o u r c e with attendant s e l f - a b s o r p t i o n and g e o m e t r i c a l c o n s i d e r a t i o n s ,
plus dose build-up in the shield, have been included. The 1/8 in. Hastelloy C
cylinder surrounding the fuel,thernnal insulation, and aluminum p r e s s u r e
shell ( e x t e r i o r to the u r a n i u m shield), w e r e neglected and should provide
additional attenuation.

Nuclear Data

The l a t e s t available data on the e n e r g i e s , conversion coefficients and

i n t e n s i t i e s of g a m m a r a y s a r i s i n g from decay of C e s i u m - 1 3 7 and C e s i u m - 1 3 4
a r e shown below, together with the a p p r o p r i a t e r e f e r e n c e .

Cs^^"^ - 25,500 c u r i e s

Gammia Total
Energy Conversion Unconverted
(Mev) * 1 Coefficient ^ G a m m a s / Disintegration

.662 . 114 .826

Cs 134 383 c u r i e s

Gannma Total
Energies Conversion Unconverted
*3
(Mev) *^ Coefficient ^ Gaitimas / Disintegration

.474 negligible .013

.563 . 10
.569 H . 14
.605 tf .98
.796 I! .80
.801 II ,10

RRC-0102
 87

Cs 134 383 c u r i e s (Cont'd)

Gamma Total
Energies Conversion Unconverted
(Mev) «3 Coefficient '^^ G a m m a s / D i s i n t e g r a t i o n *3

.960 negligible .006

1.040 .015
1. 170 .022
1,370 .033
1.570 .0012

Shielding Calculations

B a s e d on the actual g e n e r a t o r configuration, a calculation was made of

the dose r a t e at 4 ft, from the c e n t e r of the cylinder on a plane perpendicular
t o , and bisecting its a x i s . The v a r i o u s a s p e c t s of the calculation a r e d i s -
cussed below.

1. Fuel C y l i n d e r .

The fuel cylinder was c o n s i d e r e d to be a uniform, cylindrical volume

s o u r c e 2 . 5 in. in d i a m e t e r and 8.75 in, in height. Since Compton scattering
p r e d o m i n a t e s over photoelectric and pair production for the g a m m a energy
r a n g e s of i n t e r e s t h e r e , it is safe to consider the chemically complex cesiunn
fuel as a p u r e Compton s c a t t e r e r . All irsaterials, except hydrogen, contain
roughly the s a m e n u m b e r of e l e c t r o n s p e r grann . The linear absorption
coefficient of the fuel (for the p u r p o s e of self-shielding calculations) i s , to a
good a p p r o x i m a t i o n , only a function of the m a t e r i a l density and p r i m a r y
g a m m a e n e r g y . The l i n e a r a b s o r p t i o n coefficients for the fuel w e r e calcu-
lated, using the data of White ^.

2. Uranium Shield.

The 2. 1-in, -thick cast u r a n i u m shield was a s s u m e d to have a density

of 19.0 g . / c m ' . The m a s s attenuation coefficients used w e r e those of
*7
Grodstein

3. Calculation of Uncollided E n e r g y Flux.

Calculation of the uncollided energy flux at a point e x t e r i o r to a finite

c y l i n d r i c a l - v o l u m e s o u r c e s u r r o u n d e d by an annular shield has been c a r r i e d
out for a wide r a n g e of p r o b l e m s and the r e s u l t s plotted graphically * ° .

RRC-0102
 88

F o r the details and c a l c u l a t i o n s , the r e a d e r is r e f e r r e d to r e f e r e n c e ° .

The calculated values of 1 ( M e v / c m ^ - s e c , ) a r e shown in the table below.
F o r the 1.5 per cent C e s i u m - 1 3 4 contaminant, only those g a m m a s with
e n e r g i e s in e x c e s s of the .662 Mev C e s i u m - 1 3 7 g a m m a w e r e considered
and the , 796 Mev and . 801 Mev g a m m a s w e r e taken a s an equal number of
. 800 Mev g a m m a s .

4. Conversion of Uncollided E n e r g y Flux to Dose R a t e ,

Although the r a t i o of uncollided flux {Mev/cm^ sec) to dose r a t e | r / h r )

is usually a s s u m e d constant over the g a m m a energy range of i n t e r e s t h e r e ,
the actual values ''° w e r e u s e d , producing the r e s u l t s as shown in the t a b l e .

5. Dose Build-Up F a c t o r .

The method of m o m e n t s has had c o n s i d e r a b l e s u c c e s s in describing

the p e n e t r a t i o n of gamnna r a y s "*. Calculations of differential energy s p e c -
t r a and build-up f a c t o r s have been c a r r i e d out for many m a t e r i a l s f i n d ,
uranium) and for a wide r a n g e of p r i m a r y e n e r g i e s and depths of penetration
^". The only s o u r c e g e o m e t r i e s , however, a r e point isotropic and plane
m o n o d i r e c t i o n a l . Considering the d i s t a n c e of detector from s o u r c e (4 ft.)
in c o m p a r i s o n to the fuel cylinder height (8. 75 i n . ) , it i s evident that m o s t
of the g a m m a s reaching the d e t e c t o r have t r a v e r s e d e s s e n t i a l l y the s a m e
t h i c k n e s s of shield. T h u s , the approximation of point i s o t r o p i c geometry
for the p u r p o s e of build-up factor calculation introduces little e r r o r . A l s o ,
the build-up f a c t o r s , although calculated for an infinite m e d i u m , a r e within
a few per cent of those applicable to a finite medium for the number of m e a n -
free paths under c o n s i d e r a t i o n h e r e **. An effective value of m e a n - f r e e
p a t h s , which t a k e s into account the finiteness of the fuel s o u r c e , was c a l c u -
lated in the c o u r s e of computing the uncollided energy flux. Since this con-
tribution is only a few p e r cent of the total number of m e a n - f r e e p a t h s , the
build-up f a c t o r s w e r e calculated by considering the penetration to be c h a r a c -
t e r i s t i c of the s a m e total n u m b e r of m e a n - f r e e paths in u r a n i u m . For p u r -
p o s e s of interpolation, the build-up factor data a r e p r e s e n t e d in m o r e con-
venient form by Moteff '^°»

RRC-0102
 89

Uncollided Flux and Dose Rate for G a m m a E n e r g i e s

Uncollided Uncollided
Gamma energy flux flux dose Dose r a t e Dose
Energy at 4 ft f^ Mev/cm'^-sec r a t e D^ build-up Rate D
(Mev) ( M e v / c m - s e c ) (r/hr) (mr/hr) factor B^ (mr/hr)
.662 1,02 X 10^ 4 . 9 4 X 10^ 20.6 2.4 49.5
,800 2. 13 X 10^ 5,02 X 105 4.2 2.5 10.6
.960 9.92 X 10^ 5. 17 X 10^ . 192 2.6 ,5
1.040 5.03 X 10^ 5,24 X 105 .96 2.6 2.5
1. 170 1.94 X 10^ 5.36 X 10^ 3.62 2.6 9.4
1.370 6,68 X 10^ 5.54 X 105 12.05 2.5 30.2
1.570 4 , 4 9 X 102 5.72 X 10^ .785 2.5 2.0

TOTAL 104.7 m r / h r

T h u s , the calculated dose r a t e at a point 4 ft, from the center of the

g e n e r a t o r , is a p p r o x i m a t e l y 105 m r / h r . Additional attenuation will be p r o -
vided by the 1/8 in. Hastelloy C cylinder surrounding the cesium fuel and the
t h e r m a l insulation and aluminum p r e s s u r e shell e x t e r i o r to the uranium shield,
all of which w e r e neglected in t h i s calculation.

'^References

1. Nuclear D a t a s h e e t s , National R e s e a r c h Council, Washington, D. C.

2. R. A, R i c c i , P h y s i c a XXIII, 693, (1957).

3. R. K. G i r g i s and R. Van L i e s h o u t , Nuclear P h y s i c s , _i2_, 672 (1959).

4. B . T . P r i c e , C, C. Horton and K. T. Spinney, "Radiation Shielding,"

P e r g a m o n P r e s s , New York (1957).

6. G. R. White, National B u r e a u of S t a n d a r d s Report - 1003 (1952).

6. H. Etherington ( e d , ) , " N u c l e a r Engineering Handbook," McGraw-Hill

Book Company, Inc. , New York (1958),

RRC-0102
 90

G. W. G r o d s t e i n , " X - R a y Attenuation Coefficients F r o m 10 Kev to

100 M e v , " National B u r e a u of S t a n d a r d s C i r c u l a r 542, Washington, D. C
(1957),

J, Moteff, " M i s c e l l a n e o u s Data for Shielding C a l c u l a t i o n s , " APEX 176

(1954).

H. Goldstein, " T h e Attenuation of G a m m a Rays and Neutrons in R e a c t o r

S h i e l d s , " U. S. Gov't, P r i n t i n g Office, Washington, D. C. (1957).

H, Goldstein and J . E . Wilkins, J r . , "Calculations of the P e n e t r a t i o n

of G a m m a R a y s , " NYO-3075 (1954).

RRC-0102