ORNL-TM-1854.txt

  
     
       

 

OAK RIDGE NATlONAI. I.ABORATORY
- operated by
UNION CARBlDE CORPORATION
ErfNUCLEAR DIVISION S
ST for the s
U S ATOMIC ENERGY COMMISSION

ORNI. TM—'- -

(LTS

  
   
   
    
  
   

- 'fi-:':H*E"‘MéC"??f -"'." d IR wear;_ | J L

 

- fmfi flflfiddgffi' HAS Gery Rca 7
. V1
B9 INVENTIONS OF PATSHT 1 EweD.

  

i

 

 

. -S_HHTIOE Thls document ,--con!ams lnforrnchon of @ prelammory nufure L

ond was’ prepared “primarily for’ ‘internal wse at the Ook Ridge Nahonut Rty e
" Laboratory. It s ‘subject to revision or correction und therefore &oes
T not represent a hnal ‘repor SR P : ,

 

 

 
 

 

e S e b A e by A 2

1 r b o e iy

g e e

© ot 2 e 5.

s T e i ] 1L

 

o A 4 ke e ey

af

 

LEGAL ROTICE

 

 

. This report was prepared as an account of Gé\;cmmont sponsored work, Noiqhor the United Stof;s,
_ nor the Commission, nor any person oefing on behalf of the Commission:

A. Makes eny warranty or representation, expressed or Implied, with respect to the accufccy,
completeness,. or usefulness of the lnformation contained in this report, or that the vse of
any information, apparatus, molhod or process disclosed in this _teport may not infringe
privct-ly owned rights; or

. B, Assumes any liabilities with respect to ihe use of, or for damcgos nsuhlnq from the use of

any information, apparatus, methed, or process disclosed in this report.
As vused in the cbove, '"person acting on behalf of the Commlulon” includes any employes or -
contractor -of the Commission, or employes of such tontractor, to the extent that such employse

- or contractor of “the Commission, or empleyee of such contractor prepares, disseminates, or

provides access to, any Information pursuant to hu omploymont or contract with the Commission, -

Cor hls lmploymem with such contractor.

 

A P s A R e

 

 

 

 

 
 

e) q-’ »E

*) &Bm* ¥
‘ ‘

L

ORNL~-TM-1854

iy

CFSTI PRICES |

T

rfie. $.3 %% un - és/

Contract No. W-7405-eng-26

METALS AND CERAMICS DIVISION
| , LEGAL NOTICE

. This report was prepared as an sccount of Guverment sponsored work. Nefther the United
i Btates, nor the Commiasion, nor any person acting on behalf of the Commission: ‘ :

+"of any information, apparatus, method, or process disclosed in this report may not infringe
i, privately owned rights; or ! o . .

‘-« B. Assumes any lHabilities with respect to the use of, or for damages resulting from the
; use of any information, ipparatus, method, or process dizclosed in this report.

MATERTALS DEVELOPFIENT FOR MOLTEN-SALT BREEDER REACTORS

H. E. McCoy, Jr., and J. R. Weir, Jr.

 JUNE 1967

" 0AK RIDGE NATIONAL LABORATORY
. Oak Ridge, Tennessee
- operated by
UNION CARBIDE CORPORATION
: : .. for the :
U.S. ATOMIC ENERGY COMMISSION

 
 

b AR LSS

él‘E%-g

 

.

4)) ‘5?

w) fi%fi ¢

131

- CONTENTS

Abstraet e e w
Introduction .. |
Status of the'Developmentsof'HaStelloy.N
General Properties | .
.Physical Properties . . . . . C e e
Mechanical Properties . . . . .
Tensile Properti€s . v v v v v v « o o o o«
Creep Properties . . . . . . . | |

Fatigue Propertles . . . . . . « .« . . . . .

Effects of Irradisation . . « v o v o v v v v o 0 o .

Corrosion by Molten Fluorlde Salts

: Loop Studies . . o v v v v v u e w0 .
MSRE Operatlng'Experience

Resistance to Gaseous Contaminants
Oxidation Resistance_r.'.b. « . Q e e e
Resistance to Nitriding .. .'. e e e e
Compatibillty with Superheated ‘Steam f-.-; .

Fabrication of Hastelloy N Systems and ComponentSy

Raw Material Fabrication

Welding and Brazing of Hastelloy N ... c e e

Joining for Reactor Component Fabrlcatlon

Pressure Vessel and. Piplng S e e e e

Heat Exchangers .~.*.*;~5 e e e a e e

. Dissimilar Metal Joints . . .. . . .
: Remote Joining .. .',_ _ | | _
" Remote - Welding".ffi }-;*. e e e e e e
~ Brazing and Mbchanieel_Joints. C e e e e
Remote Inspectlon f}. e e e e e e e e e e
Status of Development of Graphite'_,;”._. . |
Grade CCB Graphlte _; . e
SETUCEUTE o o o o o e e e e .

Permeability . « « ¢« « ¢ ¢ ¢ v o ¢ o 4 e .

Page

 

VW U W W |

w N NN MNNMNNN N MM OWOWOWOW W W WD R P
B 5 0 o MK L oW n:_éS 5 0 360 6D RN M = 30, NP

 
 

iv - -
<;;§
Mechanical Properties: . « v v v v v o 4 ¢ o v o s s o . 53 «
Isotropic Graphite . « + v v v v o o v o v o . e 54
JOININE v v v e o e o e et e e e e e e e e 56
Graphite-to-Hastelloy'N Joints . . . . ... ... ... 57 |
Graphite-to-Graphite Joints . .:.'. .'.,.'.-. e « os . 60
Compatibility of Graphite with Mblten Salts . . ... ... 60
Rediation Effects on Graphite . . . . . .. ........ 6l

Nondestructive Testing of Tubing oo e e e e e e .'_64
Materials Development Program for Mblten-Salt |

Breeder Reactors . . . « v v v o v o v s s 0 o . Q:, e e e 65
“Hastelloy N PrOGF&Il . o v v v o o o v v ot v v oo n e e 65
Resistance to Irradiation Damagé'-.\. .;.}, . c .. - 65
Corrosion Program . « o & & & o & o & o o o o o o s o o 67
NIEELAINE  « « ¢ v v e e e b e e e e e e e, 69
Joining DevelOPmENt . o v o 4 v v v e aie o e o e s .. 69
Inspection Developmeht e e e e e e e e e e e e e e e T4 7
Materials Development for Chemical Processing | -_ | . i;
Bquipment . . . . ¢ 00 0 0 s e e e s e e e e e 75 -
Graphite Program . . . . . . « ¢« ¢« « « « . e e e e e e 75 Y
Graphite Fabrication and Evaluation - . . . « . . « « ¢ 4 76
Irradiation Behavior . . . . . .. e e e e e e e 77
Graphite Joining . . . . . . . o ¢ o o0 e e e e e 79
Permeability Studles . . . . . . .+ o v o v v o o0 . 79
Corrosion and Compatibility of Graphite O (0
Graphite Inspection . . . . . . . . .+ o o o o o o 80
General Development and Project Assistance . . . . .. ... 8L
ACKNOVIEdGEENtS . o v o v o v e e e e e e e e e .. Bl
Appendix ... . . . ¢ o o & e e e e e e e e e e 83
{

 
 

 

 

 

.'¢§(»»ug*

4 J’j vl -

o

) B

MATERIALS DEVEIOPMENT FOR MOLTEN-SALT BREEDER REACTORS
H. E. McCoy, Jr., and J. R. Weir, Jr.

- ABSTRACT

We have described the materials development program
that we feel necessary to ensure the successful construction -
and operation of a molten-salt breeder reactor, The pro-
posed reactor is a two-region system utilizing a uranium-
bearing fluoride fuel salt and a thorium-bearing fluoride
blanket salt. A third lower melting fluoride salt will be
used as a coolant for transferring the heat from the fuel

and blanket salts to the supercritical steam. The primary

structural materials are graphite and modified Hastelloy N,
The individual fuel cells will be constructed of graphite

tubes. These tubes must withstand neutron doses of the order

of 1072 neutrons/cm and must have very low permeability: to

gases (because of 137Xe entrapment) and fused salts. These

requirements mean that we need a graphite that is slightly

. better than any currently available. We have described in

detail what graphites are available and their respective
properties. A line of action for obtaining improved grades
of graphite is proposed along with a test program for
evaluating these new products.

Modified Hastelloy N will be used in all parts of the
system.except the reactor core. Since standard Hastelloy N
is embrittled at elevated temperatures by neutron irradiation,

‘it has been necessary to modify the composition of the alloy

with a small addition of titanium., The program necessary for
fully developing this modified alloy as an. engineering material
is described in detall ; \ ,

An integral part of the proposed system is a JOlnt between
the tubular graphite fuel channels and the modified Hastelloy N.

‘Brazing alloys have been develoPed specifically for this job
“and a reasonable design for the joint has been made. The

1ntegr1ty of the goint must be demonstrated by engineering

_tests.

. Several areas requlre the development of suitable inspec-
tion techniques. These techniques are further complicated by

the fact that they must be adaptable to remote 1nspection

inside the Ieactor cell

_ Although numerous problems exist which will require Turther
development, none of these appear insolvable. Hence, we feel
that the materials development program can proceed at a rate
consistent with that proposed for the Molten-Salt Breeder Reactor.

 
 

 

14"[ilv"

INTRODUCTION

The proposed molten-salt‘breeder'reactorsllwiiljrequire some advances
in materiais technology. However, the'construction and operatioh of the
_Mbiten-Salt Reactor Experiment have given us iovaluable iasight concerning
what advances are necessary. We feel that we can make these advances on a
time schedule that is consistent with that pr0posed for the Mblten-Salt
. Breeder Reactor A

From & materials standpoint, the reactor 1s easily divided into two
sections: (l) the core and (2) the reactor veseelrand associated piping.
The requirements of the material for use within the core are (1) good
moderation, (2) low neutron absorption, (3) compatibility with the molten
salt~Hastelloy N system, (4) low permea.bility to both salt and fiseion ga.ses,
(5) fabricable into tubular shapes, (6) capable of being jolned to the rest
- of the system, and (7) capable of maintaining all the above. properties after

A

accumulated neutron doses of 1023 neutrons/cm . Graphite 1s the preferred
material for the core, and the Grade CGB graphite used in the MSRE satis-
fies most of the stated requirements. The main additional requirements.

Nh

®

are that we develop the technology for producing tubular shapes of a com-
parable material with slightly improved gas permeability and then demon—
strate that this material will retain its integrity to neutron doses of

1023 neutrons/cm®. Tubes of the desired quality can be produced in the

near future. .Available data on the radiation damage of graphite to doses'
of 2 to 3 X 1022 neutrons/cm? indicate that the material is capable-of the
anticipated doses. , : o

| The rest of the system — the core vessel, heataexchahgers, and piping -
- will see & considerably lower neutron flux. The requiremeats of.a‘material
~ for this application include (1) resistance to corrosion by fluoride salts,
(2) compatibility with'the core material (3) capable of(being fabricated
into complicated shapes by conventional processes such as rolling, forging,
and welding, (4) good mechanical strength and ductility at temperatures over
the expected service range of lOO to 1300°F, and (5) capable of maintaining-

 

¥

1p, R. Kasten, E. S. Bettis, and R. C. Robertson,~Design Studies of i~
1000-Mw(e) Molten-Salt Breeder Reactor, ORNL-3996 (August 1966) A g{

 
 

4 c -

oo,

) (n! .

“reasonable strength and ductility after eXposure to a neutron environment.

Hastelloy N satlsfies most of these requirements This alloy was developed

- at the Oak Ridge National Laboratory specifically for. use in molten-salt

lsystems.' HOWever, our experlences w1th this alloy in the MSRE indicate

~that 1t has at least two disadvsntages: 8 propensity_forxweld cracking

and severe reduction of high-temperature ductility after irradiation. The
- first problem can be solved by using vacuum-melted material, but slight

changes in the alloy comp051tion will be necessary to minimize the radia-
tion damage problem, We have found that a slightly modified alloy con- -
taining O 5wt % Ti has good weldability and reasonable resistance to
radiation damsge - _

 This report presents the present stste of knowledge of these two

'materials, graphite and Hastelloy'N, as they affect an MSBR. The first

section discusses Hastelloy N and the second discusses graphlte The
final section briefly presents the program requlred to extend the develOp-
ment of these materials to_s stage where they may be used in the MSBER
Wnilke considerable development and testing must be accomplished to pro-

vide the aSSured.performance'necessary for a resctor_system, it appears

that adequate materials can'be'obtsined for a molten-salt breeder reactor.
STATUS OF THE DEVELOFMENT OF HASTELLOY N
JGeneral\PrOperties

'3 During the early stages of the Aircraft Nuclear Propulsion Program,

'.msny metals and’ alloys were: screened to determine their re81stance to.

';,f'molten fluorides. Primsrlly on' the. basis of these reSults, ‘the nickel—'

'r,molybdenum system was selected as the most: promising for additlonal study.

A molybdenum concentration range of 15 to 20% was selected since this

Vylelded s1ngle-phase alloys with their inherent metallurgical stability

—:varlous other alloying additiOns were studied in order to 1mprove the

| ';mechanical prOperties and cxidation resistance of the basic binary alloy.

 The Hastelloy N alloy reSulted from this program and was used for the MSRE.

The chemical comp081t10n is 11sted in Table l

Hastelloy N is a nickel-base alloy that is solution strengthened with

| molybdenum and has an optimized chromium content to max1m1ze oxidation

 
 

 

Table 1. Chemical Composition Requirements for the
Nickel-Molybdenum-Chromium Alloy Hastelloy N

 

 

Element S L o Wfi'%(é) -

Nickel T . bal
Molybdemum - 15.00-18.00
Chromwm . 6.008.00

~ Iron [ - 5.00

Carbon o o _tn‘ - 0.04-0.08
Mangenese o SR _-'i;r: ©.L00
Silicon - , - ~ - 1.00
Tungsten o o Lo e 0.50 -
Aluminum + titenium o Q.50”

~ Copper B . S 0.35 -
Cobalt | | o 0.20

~ Phosphorus D — - 0.015
Sulfur - - | - 0.020
Boron . | o . 0.010
Others, total . o S 0.50

 

Single values are maximum percentages unless other-
wise specified

resistance and to minimize corrosion by fluoride salts. The chemistry |
has been controlled to preclude aging embrittlement. The aluminum, -
titanium, and carbon contents are limited to minimize severe fabrication
and corrosion problems, and the boron content is limited to prevent weld
,cracking. Iron is included to allow more choice of starting materials
for melting. The extreme examples of permissible combinations of elements
'allowed by the chemistry specification were studied, and in no case did.
any undesirable brittle phases develop. Carbides of the form Mé305 and
M¢C exist in the alloy and are stable to at least 1800°F.

The MSRE was constructed using conventional practices (comparable to
those used for a stainless steel system) and from material obtained from
commercial vendors. The major materials problem encountered was one of

weld cracking, which was eventually overcome by slight changes in melting

-
 

 

 

 

" L)‘

S )

&) o,

qt) (“;'

'__ '”practice and by strict quality control of the material 'Heats of material

subaect to weld cracklng were 1dentif1ed and thereby eliminated by means

. of a special weld—cracking test that was included as a part of the

| Specifications

Physical Properties

Several phy51cal properties of Hastelloy N are llsted in Table 2.
Specific heat, electrical re51stiv1ty, and thermal conductiv1ty data all

»show inflections with respect to temperature at l200°F This is thought

to be due to an order-disorder reaction, however, no changes in mechanical

v
pr0pert1es are detectable as a result of this reaction. -

Mechanical Properties

The composition and the fabrication procedure for Hastelloy N have been

optimized with resPect to. resistance to salt corrosion, ox1dation resistance,

 and freedom from embrittling aging reactions. The strength of the alloy

is greater than the austenltic stainless steels and comparable with the
stronger alloys of the "Hastelloy type." |
‘Design data for this alloy were established by performing mechanical

prOperty tests on experimental heats of commercial size. Data from this

study were reviewed by the. ASME Boller and Pressure Vessel Code Committee
and Code approval wa.s obtained under Case 1315 for Unfired Pressure Vessel

construction and Case 1345 for Nuclear Vessel constructlon.l The recognized

-~ allowable stresses are tabulated 1n Table 3. The" commercial heats used
for MSRE construction exhibited strengths equal to or greater than the
- experimental heats...

"fTEn31le Properties3

Tensile tests were performed on the experimental heats and Spec1fica—

tlcns were thereby established for the commercial heats. Flgure l is a

 

2R 'W. Swindeman, The Mechanical Propertles of INOR-S ORNL-2780
(Jan. 10, 1961).

2. T. Venard, Ten511e and Creep Properties of INOR- 8 for the Molten-_

'_Salt Reactor Experiment ORNL—TM*1017 (February 1965)

 
Table 2. Ifhyaicnl Properties at Va:bims Temperatures |

 

Flectrical | ' Cosfficient of Thermsl Modulus of

 

 

Temperature Density Resiptivity . Thermal Conductivity Specific Heat Expansion Flasticity
(°F) (a/em®) (1b/1in.?)  (uohm-cm) (v em™t °C”1) (Btu ™ nr™t °F°Y)  (Btu Wt 7Y (°F)-2 . (1b/4n,2)
| _ % 10~8 x 108
75-85 g.93 0.320 120.5 ”
1300 : ' 126.0
‘340 ' 0.098
572 : : : 0.109 .
1000 ‘ _ _ 0 0,115
212-1752 , R , Lo . 7.0
752-1112 ‘ : o , ' C 8.4
1112-1832 ' ‘ Lo ' - 9.9
212-1832 S 8.6
300 0.12 6.94
75 0.14 8.21
825 0.16 9.25
985 0.18 10.40
1165 0.20 - 11,56
1475 - 0.24 13.87 o
55 ‘ 31.5
425 - 29,0
925 . . 27.0 -
1075 -~ 26.3
ns 26.0
1300 ! 2408 -
1475 23.7
1575 2.7
1750 20.7
. 1825 19.1
- 1925 A7

 

P Cs _ o ¢ Vo

O,

 
“C"

13 I

a’)

 

"Table 3. Msx1mum.Allowable Stresses for Hastelloy N Reported
~ Dby ASME Boiler and Pressure Vessel Code

 

Maximum Allowable Stress, psi

 

 

Temperature
(°F) | "~ Material Other Cmso
. : - than Bolting Boltrng
100 S 25,000 ' 10,000
200 | o 24,000 | o 9,300
300 | ' 23,000 8,600
400 o | - 21,000 ' 8,000
500 A - 20,000 7,700
600 o 20,000 7,500
700 EEE 19,000 | 7,200
800 | 18,000 7,000
900 18,000 - - 6,800
1000 - 17,000 | 6,600
1100 ' 13,000 | 6,000
1200 . - - 6,000 ' | 6,000

1300 o 3,500 3,500

 

- ) OfiNL-_-Dfi’G 64-4414R2
o ' TEMPERATURE (°C) - ' |

O 100 200 300 400 SO0 €600 700 800 900 1000
140 1T TT- T T 1771 1T T1 T

 

 

 

 

 

 

 

 

420

    
 

 

 

SCATTER BAND FOR-
EXPERIMENTAL HEATS

 

 

 

100

 

 

 

o
o

 

 

D
Q

 

 

 

 

. WEAT 5075 / -

‘ '4PARALLEL TO R. o.‘ : //

- ®NORMAL TO R.D. - - _ - %
HEAT 5081 - R | 8

e PARALLEL TO R.D. RREEN . e

vNORMAL. TO R.O..

 

 

H
O

 

TENSILE STRENGTH (1000 psi)

 

L o0 b

 

 

 

 

 

 

 

 

 

 

 

 

O 200 400 600 8OO 1000 1200 4400 1600 1800
TEMPERATURE. (°F) '

Fig. 1. Tensile Strength at Various Temperatures of Hastelloy N.

 
 

 

\

summary of the ultimate strengths at temperatures from ambient to 1800°
for both types of material. Similar data on the 0. 2% offset yield strength
are shown in Fig. 2. The values for fracture ductility are presented in
Fig. 3. In all cases, the values for the commercial heats were well within
the band obtained from the experimental heats. The values from both the
.longitudinal and transverse specimens are comparable, show1ng no anisotropy
effects. Metallographic data indicate that the heats with low carbon, and |
consequently large grain size, tend to exhibit the lower strengths,
Tensile tests of notched sPecimenS‘were performed using a'notch o
radius of 0.005 in. The notched-to-unnotched strength ratios varied from
1.07 to 1.38 at test temperatures from ambient to 1500°F o |

, ORNL-DWG 64-4415R2
' , TEMPERAT_URE {°C) _ , '

-0 100 200 300 400 S00 €00 700 800 900 {000

0 7 | | 11 T | I

 

 

60

 

 

 

 

S0

 

 

/SCATTER BAND FOR
/ EXPER!MENTAL HEATS

. é%fi/%

 

 

 

 

Z

 

 

 

20 ' e — 1 4.
HEAT 5075 |
' 4 PARALLEL/TO R.D.

¢ NORMAL TO R.D.

10 I HEAT S081
s PARALLEL TO R,D.
vNORMAL TO R.D.
0 1 I - : . '
‘0 200 400 600 8OO 1000 {200 4400 1600 4800
: TEMPERATURE °F) e

0.2% YIELD STRENGTH (1000 psi)

 

 

 

 

 

 

 

 

 

 

Fig. 2. Yield Strength Values for Hastelloy N.

»

 
 

 

“C”

"

“»

‘-J)

. " ORNL-DWG €4-4416R2
o : : TEMPERATURE ey o
O 100 200 - 300 400 500 600 700 800 900 1000
T 1 I I T 1 ol
: : __SCATTER BAND FOR |
EXPERIMENTAL‘ HEATS

 

 

- 70

 

 

60

 

\\\\\\\‘_‘

N
&

.

i i/%//// / /;

 

 

8.
\\\\\\
i

X

N
N

 

-~

 

 

 

 

\\
&\
\

 

 

o PARALLEL TO R. D
v NORMAL -TO- R.D. "

0 1 ‘[, — 1 '
O 200 400 600 - BOO. {4000 1200 400 600 1800
- TEMPERATURE (°F) -

Z

30 |- - 7
ol e e % \%z‘
| e e . ////

 

 

 

 

 

 

 

 

 

s

o Fig. 3. Totai-Elongatidn-Values for HaStelloy'N.

Creep. Properties3

Creep*rupture tests were performed on sheet and rod sPec1mens in

both air and molten salts Mbst of the testlng was. conflned to the

1100 to 1300°F temperatfire range however, a few teste were_eonducted at
ftemperatures up to 1700°F Sfifimary curree'representing etrees vs minimum
. creep rate for the MSRE heats of Hastelloy'N are shown in Flg 4.  These
- data for heat 5055 are plotted to show the tlme to varlous strains at
~ 1300°F in Fig. 5. The rupture life of Hestelloy N plotted as a function

of stress is shown in Flg 6. The creep propertles in molten- salt

environments were not signlflcantly different from those obtained in air.

 
 

10

{x10°)
80

{4100°F)

40
o

 

ORNL-DWG 64-4434R

:‘_'_; -
v
"
E (1300°F)
w
10 - . {NOR-B
s ' AS RECEIVED-TESTED IN AIR
| L | MATERIAL FROM MSRE PLATE -
6 ' 1 o HEAT 5055
- A 8 HEAT 5075
(1500°F) - G HEAT s081 .
« _ OPEN SYMBOLS ~PARALLEL TO R.D.
- CLOSED SYMBOLS ~TRANSVERSE TO R.D.
. ! \ :

'o-' K w0 s ot ‘0-3

WMINIMUM CREEP RATE (i '- >

—}

1o - ottt

Fig. 4. Effect of Stress on Minimum Creep Rate of Ha.stelloy N

& 88

H
O

STRESS {1000 psi)

N-
o

02% 05 1.0 20 5.0

10
oo .0 0 100

TIME (hr)

- Flg. 5. Time to Reach Various Strains for a Giiren' Stress. Numbers

in parentheses indicate fracture ductilities.

   

ORNL-OWG 64-4438R

704°C
{(1300°F)
HEAT 5055

- AIR

RUPTURE

{000 10,000

(Open symbols — specimens

parallel to rolling direction, closed symbols — 8pecimens perpendicular

to rolling direction.)
 

 

¢ t‘}

b1

&)

ad (.fl

STRESS (1000 psi)

100

50

20

10

ORNL-DWG 64-4418R

593°C (1100°F)
~704°C (1300°F)

816°C_(1500°,F)

o'HEAT 5055
& HEAT 5075
0 HEAT 5081

OPEN SYMBOLS - PARALLEL TO R.D.
CLOSED SYMBOLS-TRANSVERSE TO R.D.

 

o 0 100 1000 10,000

o 'TIME TO RUPTURE (hr)
Fig. 6. Rupture Life of Hastelloy N at Given Stresses and
Temperatures. o

Fatigue Prop_rties

Rotating-beam fatigue tests were conducted on Hastelloy N at 1100 and
1300°F by Battelle MEmorlal Instltute 4 Grain size and frequency were the
major variables studied. - The results are shown in Flgs 7 and 8. )

The thermal-fatigue hehav1or of the alloy was investigated at the
Univer51ty of Alabama Flgure 9 presents a graph of the plastic straln
range Vs cycles to fa:.lure » a.nd the results are seen to obey 8 Coff:m-—type

relation The tests 1nvolveé several me.ximum temperatures, as noted, as

well as rapid cycling and hold time cycling Furthermore, the data were

obtalned from two sPecimen geometries These data show good agreement
with isothermal straln-fatigue data on. this alloy. '

An analys1s of these same tests, based on a plastlc strain energy -

rcrlterlon, 1nd1cates that the total plastlc work for fallure of Hastelloy N

by fatlgue is constant in this temperature range The data have been

l_plotted‘in Fig. 10 as plast;c~strain energy Vs cycles.to failure. It is

‘seen that the data for 130Q'afia’1500°r fit’curveS'having the same slope

 

“R. G. Carlsen, Fatigue Studies of INOR-&, BMI-1354 (January 1959),

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

00 — ' o . ‘ ORNL-LR-DWG 54252 | o'
_ : 100 cpm 3000c¢cpm
80 1115.p. 49 COARSE GRAINED o o
S.P 19 FINE GRAINED .
~ T0
w ‘ ~
g €0 al . Jouuy _
- ._- - h.~ ~~~~4
s o gl il o | T ,
< 50 o-HbNoo— " ‘
@ . f %\\ \D#‘--—.._.__ .
fl' - B B""'-n-.,, h_# i
5 40 V 0 T .
30
20 1 ! - |
10% ~10* 10° | 108 107 10°

. CYCLES TO FAILURE -

Fig. 7. Fatigue Properties of Hastelloy N at 1300°F s Rotatirig
2

ORNL LR-DWG 354253

"

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

g0 I H ” TTTTIT 1T T 1711
HHIL 1 1 1 100cem ‘3000cpm
S.P.19 COARSE GRAINED o o
g0 , S.P. 19 FINE GRAINED . .
. . '
70 o
o g @ ol
= ?\:\ d ~ | |
E 60 ’ Bhiq — e e
3
o
g - T~
< 50 m |_
I‘g ) {4 ) Dot b
Wt
=
‘w40
- 30
) 20 2 |
103 10* 10° 108 107 108

| ' CYCLES TO FAILURE
'Fig. 8. TFatigue Properties of Hastelloy N at 1100°F; Rotating

Beé.m. . S
y ' : ' L : -  ORNL-DWG 64-4006

10~

 

 

@
g
£
g 2
[
& 2
10~
g
g .
2 5
o .
- K
z
4
. -2 -
E, 0 1600°F
E _ bl -
-t .
& 5
-
9
2
1074 : e —
409 2 -8 0t 2 B -'102 2 -85 - 10 2 s - 0% 2 5 10°

 

N,. CYCLES TO FAILURE

Fig. 9. Effect of Stra.in Range on the Thermal Fa‘tigue Life of
Hastelloy N.

s

3 0% P ‘ . ORNL-Owa €4-4007 -
10°

i S - 1300°F

W), PLASTIC. STRAIN ENERGY PER CYCLE lin.~ib/in,)

o , .

10%. o _ :

‘0% 2 5 1w 2 5 102 2 5 10 2 5. 10* 2 5. 108
' - N, , CYCLES TO FAILURE .

 

- Fig. lO Relation o:f‘ Plastic Stra,in Energy Absorbed per Cycle to
the Fatigue ‘Life of Hastelloy N

v
?_

 

 

 
 

14

(approximately'—i) but different 1ntercepts. The'intercept'values (at
N, = 1/2) are in fair agreement with plastic strain energy values derived
from tensfl.e tests at the approPriate temperatures. . | o

Effects of Irradiation.

Although it has been shown thattHastelloy N'has suitebie:properties
for long-time use at high temperatures, a deterioration of high*temperature
properties occurs in a hlgh neutron-radiation field 5 we have found
that Hastelloy'N is susceptible to a type of high—temperature dirradiation
damage that reduces the creep—rupture life and the rupture ductillty
Data from 1n-reactor creep-rupture tests run on a. heat of MSRE material
are_presented in Fig. 1l. The rupture life has been reduced by a factor |
of 10 and the rupture ductility to strains of 1 to 3%. There is,.however,
an inuicatiOn that the radistion effect becomes less as the'strese'levels

are reduced.

 

W. R. Martin and J. R. Weir, Effect of Elevated Temperature Irradia-
tion on the Strength and Ductility of the Nickel-Base Alloy, Hastelloy N,

03NL—TM~1005 (1965)
onm.-nws 65-5779R _

 

70 o

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

29 N
N\~\ )
€0 . éia ,
2199 . 11 |
S Negr2 i) | 1
50 N T
a—— . 9 .
3 25 UNIRRADIATED
o . Y L Il
g 40 _ P66
£ 158
fi [ | \\\ .
& 30 \33‘.5__,2_5..12.5 _ .
LT N L
' ¢ . ’
g, =6x10%m122) N 125
20 Ty <8
'.5‘ "6.... \\i
244>~
0 30
10° 2 5 o 2 5 0 2 s 100 2 s o

' ' "RUPTURE LIFE (hr)
Fig. 11. Creep-Rupture Properties of Hastelloy N at 650°C, Heat 5065.
Numbers indicate rupture strains.

T
 

 

ifl ( )

¥

-x)

e

 

15

. The amount of damage present is a function of the thermal flux and is

| essentially independent of fast flux.  This supports the hypothesis that

the damage is due to the thermal neutrons with the 108 transmutation to
lithium and helium thought to be the most probable reaction. The helium

| __collects in,the grain boundaries and promotes the formation of intergranu-

lar cracks ThlS type of irradiation damage has been found to be quite

"'general for iron- and nickel—base structural alloys.

Helium.may be formed in Hastelloy'N from two sources. - The normal

alloys contain 30 to 100 ppm. B, and, when exposed to thermal neutrons,

the transmutation of the 10B Just discussed is by far the predominant
source of helium A second source is the (n,q) reactions of fast neutrons

with nickel molybdenum, and iron. Thise source of helium predominates

after all the 10B has been transmuted (requiring a thermal dose of

approximately 1021 neutrons/cmz) or in materials Where the 108 content 1s

- extremely low.

In the reference MSBR there is no- stressed metal in the core and the

3Hastelloy'N is shielded from the core by the breeder blanket Most of the

neutrons reaching the metal will originate as delayed neutrons that are
born with t00 low an energy level to produce the fast neutron reactions.\-
Therefore, the best way to reduce the amount of helium in the metal is to

reduce the boron content
. The main source of boron in commercial alloys is the melting crucible,

and we have found that with reasonable care ‘commercial heats can be prepared'

»,;with boron in the range L—5 ppm. However, the proPerties of irradiated
. metal of this low boron content are no better than ‘those of metal contalning

fi750 ppm B 1ndicating that the threshold helium levels necessary for damage

can ‘be. produced in metal containing below 1 ppm- B. Hence, it appears that

ffreducing the boron level alone is not an adequate solution to the prdblem.

We have found that the resistance of Hastelloy N to irradiation

o damage can be improved by slight changes in chemical composition. A

r*reduction in the molybdenum content seems desirable to prevent the forma-

tion of mass1ve MSC in the alloy. Flgure 12 shows the 31mpler micro-

ldd;structure of the a110y containing 12% Mb Further increases in the

 

6. R. Martin and J. R Weir, Jr., Solutions to. the.Problem of

’ High-'.I‘emperature Irradiation Embrittlement ORNL-E[M-1544 (June 1966)

 
 

 

 

     

 

 

 

 

 

 

 

 

 

 

 

 

 

 

r.
L
3

 

 

' L 14% Mo . - | _ w:“ 'Wfs% M;' Lol
 Fig. 12. Effect of Molybdenum Content on the Microstructure of Ni=7% Cr—0.2% Mn—0.05% C Alloys
After Annealing 1 hr at 2150°F. ” ' ‘ . | S

 

 
 

 

) ( : ¥ -

Sy

@y

17

mblybdenum content result'ip_'the' 'formei_tion of the -molybdenumfijich MeC and -
have little effect on st’rength’ ' The addition of small amounts (approx
0.5 wt %) of Ti, Zr and HE reduces the irradiation damage problem signif-

| 1cant1y F:Lgure 13 illustrates the fact that several alloys have been

‘developed with propertles after irradlatlon that are superior to those of
unirradiated starid_ard Hastej_.ley N. V.We are beginning work to optimize the

'c‘:ompos.it.ions and heat treatments forvth_ese alloys. Ve are also initiating
the procurement of 1500 1b commercial melts of some of‘the'm‘ore attractive
compositions. ' | |

The ex-reactor proPerties of the modified Hastelloy N seem very

 attractive.. Strengths are slightly better than standard Ha.stelloy N and

fracture ductilities are about double. The weldability of the titanium-
and hafnium-bearing ’a'lloys appears excellent;-'however, ‘we realize that

welds need to be made under',_higher reetraint and in 1afg'er sections than

ORNL-DWG 67-3523

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Nl(22)
. ‘\\ J
€0 T 1= UNIRRADIATED AIR MELTS ™
-_\\,
50 | e 1IN L Hue0s L) |
- . o8 - 408 Hf'“‘. ‘ M (o570
' §40 : ""'"~\‘ - ”“ N2/ .,” 'i 1
2 1 R _.._\\figs_fl.c)_ \\\ (15 osm L)I l |
o IRRADIATED AIR MELTS{ ~ |
8 . | a0 F | \\ ™ ,49051: U |
52 1 TN {3, m-—L)
20 -+ IRRADIATION couomous ——t ey NS
1 | r=esoc T . |
K- ¥ -a-suo”m
o - L ‘ ‘. - .7._. . : . - : -
C e 0 1000 - ',-_';——r—-TlO.OQO :

RUPTURE LIFE (br)

Fig. 13, Comparison of the Postirradlation Creep PrOperties of

Several Hastelloy N Alloys at 1202°F. The first number indicates the

fracture strain, the second indicates the alloy addition, and the thlrd
indicates the source, C = commercial, L = laboratory

 
 

 

 

have been studied thus far. The zirconium'eddition appears'very detri- ¥
mental to the weldability with extensive weld metal cracking resulting o
Thus we feel that the further development of the 21rcon1um-bearing alloy '
must await the development of a suitable filler metal | '

Corrosion by Molten Fluoride Salts

A unique feature of the molten-salt\type reactors_ie'the_circuiation
of molten salts both as the fuel and the coolant media. The Salts_being
" used are mixtures of metal fluorides. Essentially no experience is |
available from‘indnstry on handling such salt mixtures at the proposed
temperatures,‘tut much information has been produced at ORNL. B o
Studies on molten fluoride mixtures for reactor applieatione began . .
in(the early 1950's and gave-prinafy consideration to the compatibility
of these salt mixtures with stiucturel materials. In the intervening
15 years, an extensive corrosion program has been conducted at,ORNLrwith"
several families of fluoride mixtures using‘both commercial and develop-
mental high-temperature alloys.”™ 15 As a consequence, the corrosion - :

 

7L. S. Richardson et al., Corrosion by Molten Fluorides, ORNL-1491
(Mar. 17, 1953). |

8G. M. Adamson et al., Interlm.Report on Corrosion by'Alkali—Metal
Fluorides: Work to May 1, 1953, ORNL-2337.

°G. M. Adamson et al., Interim Report on Corrosion by Zirconium—
Base Fluorides, ORNL-2338 (Jan, 31, 1961).

10y, B. Cottrell et al., Disassembly and Postoperative Examlnation
of the Aircraft Reactor Experiment, ORNL-1868 (Apr. 2, 1958). ' o

11y, D. Manly et al., Aircraft Reactor. Experiment —-Metallurglcal
Aspects, ORNL-2349 (Dec. 20, 1957) pp. 2-24.

12y, D. Manly et al., Prog. in Nucl. Energy, Ser. IV 2, 1964 (1960).

137, A. Lene et al., pp. 595-604 in Fluid Fuel Reactors, Addison-
~ Wesley, Reading, Pa,, 1958,

~ 14Molten-Salt Reactor Program Status Report ORNL-CF-58-5-3, pp. 112-13
(May 1, 1958). |

153, &, Devan and R. B. Evans III, "Corrosion Behavior of Reactor
Materials in Fluoride Salt Mixtures," pp. 557—79 in Conference on Corro- e
gion of Reactor Materials, June 4-8, 1962, Vol, II, International Atomic "\Ej
Energy Agency, Vienna, 1962. ‘

 

 

 

.

 
 

 

 

1Y C «}

N

&)

19 .

“technology for the molten-salt'reactor'concept’is’now in an advanced stage

of development Furthermore, container materials are. available that have

ghown extremely low corrosion rates in fluoride mixtures at temperatures

_ considerably above the 1000 to 1300 F range proposed for the MSER.

- Unlike more conventional oxidi21ng media, ‘the . products of oxidation
of metals by molten;fluorides tend to be completely,soluble in the cor-

roding media; hence, passivation is preciuded, and corrosion depends

~ directly on the thermodynamic driving force of the corrosion reactions.

Design of a chemically stable system utilizing fluoride salts, therefore,
requires the selectiOnVOf:salt'constituentsrthat7are not'appreciably
reduced by available StruCtural metals and the-development of containers

whose components are in near thermodynamic equllibrlum with the salt

‘medium. -TIn practical applications, the maJor source .of corrosion by

fluoride mixtures has derived from trace,impurities in the melt rather
than the major Salt'constituents TheSe impurities may originate within

»the melt or from oxide films or other contaminants on the metal surface,

| Further details concerning corrosion mechanisms in fluoride melts have been

presented previously, 2
Corrosion data on Hastelloy N in LiF -BeF; - ‘based salts have been
generated in out-of-reactor thermal- and forced convection loops, in-

reactor capsules,,and more_recently_the MSRE,

Loop Studies

The origlnal studies that serve as a base for the MSBR corrOS1on
studies were conducted at ORNL a8 part of the Aircraft Nuclear PrOpulsion
Project 713 Work on this project wag aimed at a 1500°F maximum tempera-

v'ture. Nickel alloys were shcwn to be the most promising for containing
the fluorides however, strength considerations restricted the candidate
-materials to the commercial nickelvchromium group,_with Inconel 600

‘receiving the most attention. These‘alloys’incurred corroSion by selective

oxidation of chromium through trace impurities and by mass transfer of

chromium through a redox reactionrinvolv1ng UF4.' Corrosion‘was in the

form of subsurface vbids'withfithe-depth'prop0rtionalrto the test time

and temperature.

 
 

 

 

Using information gained in corrosion testing of the commercial |
alloys and from_fundamental,interpretations ofthe.corrosionprocess, an .
alloy was developed at ORNL to provide improved coerSion resistance as
- well as-acceptable mechanical:prOperties; The alloy system used as the,-
basis for this'deveIOpment wasrcomposed_ofrnickel with a_prinary_strengthe
ening addition of 15 to 20%-Mb.. Evaluations of other'strengthening addi-

- tions culminated in the selection of an alloy containing 16% Mo, 7% Cr,
and 4% Fe (INOR-8, now Hastelloy N). | . S

'The corrosion testing program for Hastelloy N involved three sequential
phases., In the first phase, the corrosion properties of 13 fluoride salt
mixtures were compared in Hastelloy N thermal-convection loops operated
for 1000 hr. Specific salt mixtures, whose compositions are_listedhin
Table 4, were selected_to_provide an evaluation of (1) the corrosion
properties of beryllium-bearing fuels, (2) the corrosion properties of
beryllium-fluoride mixtures containing large quantities of thorium, and
(3) the corrosion properties of nonfuel-bearing fluoride cooclant salts.
The second phase of testing, which again used thermal-oonvection 1o0ps,
involved more extensive investigations for longer time periods and at
two teaperature levels. The third phase of the'testing was conducted'in
 forced-circulation loops at flow rates and temperature conditions simu- '
lating those of an operating reactor system. These 100ps Operated with
maximum fuel-salt temperatures of 1250 to 1500°F, ma.ximum coolant salt
temperatures of 1050 to 1200°F, and all with approximately 200°F tempera-
- ture drOps. A total of 49 thermal loops and 15 forced-circulation loops
- were operated during these phases of the. program | o |
Essentially no attack or deposition was found with any of the
- fluorides in Hastelloy'N lOOpS in the phase-I studies. The max1mum‘ |
attack found after 8760 hr in the phase-II tests was a 1imited surface 7
‘roughening and pitting to a depth of 1/2 mil Attack in most cases was
accompanied by a thin surface 1ayer. ‘The typical appearance of a hot-
leg surface is shown in Fig. l4 Nb deposit or other evidence of mass
transfer wvag found in any of the cold legs | A -

In the third phase of the _progranm, employing forced—convection lOOpS,
tubular inserts in the heated sections of some of the 100ps provided infor-
mation about the weight losses occurring during the tests. 8Salt samples

.’)

 
 

 

4 C 4

[

Xy

8

_ Salt Mixture . -

e e s

21

Table 4 Compositions of Mblten-Salt Mixtures Tested for
Corrosiveness in Thermal—Convection Loops e

 

- fgcomposition,-moles% -

 

. NeF L KF %P, BeF, R, THR

 

‘*g_Fuelfsnd,Blanket Salts

 

w2 os7 e

2o . 553 . 4007 4

D am T
s s o5 o5
126 s e 1
127 s s 7
128 St 29

e oma
e m e
13m0 16 13

134 e . 35 05 1

‘s s 455051

136:tm m _ t; ti '7Q:i:fl‘:' H. | 107 - . 20 7

Bl +osU e 185 o5 1

_ Coolant Salts

 

 

| "iftaken from the pump bowls provided a semicontinuous indication of ;;'f
"ff;;corrosion-product concentration in‘the circulating systems.11 A summary
':_[_fof the oPeratlng conditions and results of the metallographic examination
_ijfof the forced- 1rculation loops are presented in Table 5 the that o
.'~;;the operating times of these systems range from a minlmnm of 6500 hr to a
rs;i’tmaximum of 20 OOO hr., Nine of the loops were - operated for over 14, OOO hr.

The corrosion rates of Hastelloy N in ‘the pumped 1oops operating with

-'maximum temperatures between 1300 and 1500°F (ref.15) indicate that

 
 

22

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 14 Appearance of Metallographic’Specimen from Hot leg
(1250°F) of Hastelloy N Thermal-Convection Loop. Operating time:
8760 hr. Salt mixture: LiF-BeFp-UF,-ThF, (62—36.5-0.51 mole %).
The small "voids" in the microstructure below the surface layer are
microconstituents in the Hastelloy N which have been darkened and partially
removed by metallographic etching.

corrosion reactions effectively go to completion in the first few thousand
hours of loop Operation. As shown in Table 6, weight losses in 10, 000-
and lS,OOO—hr testsrin pumped loops containing LiF-Bng-UF4 salts showed
no measurable increase'after,the-first_50007hr of 0peration;furthermore,
changes in concentrations of corrosion products'in the circulating salt
leveled off in the same time period. The temperature dependence of the
maximum corrosion rate, as judged from total weight losses of" sPecimens,‘
was less than a factor of 5 over the temperature range investigated |
(Table 5). 30wever, the metallographic appearance of the loop surfaces
was noticeably affected by the test temperature. At the highest test i
temperature (1500°F), Hastelloy-N surfaces were depleted of chromium, as
indicated by the appearance of shallow subsurface voids to a maximum depth
of 4 mils. At 1300 and 1400°F the surfaces exhibited no evidence of
attack or other metallographic changes during the first 5000 hr of 0pera-t

tion, at still longer test times a thin continuous 1ntermetallic layer was
 

 

X) (-" . &X) W

stle-5 Hastelloy N Qperating Conditions of Fbrced Convection Loqps and Results of

 

 

85ce Table 4.

Mstallographic Examinations of Loop Materials
P S R Maxinmum | | ' . ‘ - |
~ Duration | ' Flow » L,
Loop of Test Salt Mixturea - Fluid-Metal em Reynolds Rate Results of.Mstsllographlc
Nunber (hr) - | ~ Interface (°F) Number (gal/min) Examination
- Temperature ' & ' o
o (-
9354-1 14,563 126 . 1300 200 2000 2.5  Heavy surface roughening and
U e B - | plttlng to 1 1/2 mils
 9354-3 19,942 "‘ff847f‘ sfl*~ 1200 1100 3000 2.0 No attack, slight trace of _
' o “'Tfi; j . _;J L | - metallic deposit in cooler coil
9354-4 15,140 . 130 1300 200 3000 2.5 No attack
9354-5 14,503 130 1300 200 /3000 2.5  No attack N
MSRP-6 1 20,000 134 1300 200 . 2300 1.5 --Pitted surface layer to 2 mils '
MSRP-7 20,000 133 1300 200 3100 1.8 Pitted surface layer to 1 mil
MSRP-8 9,633 124 1300 200 4000 2.0 No attack
'MSRP-9 9,687 134 1300 200 2300 1.8 No attack
MSRP<10 20,000 135 1300 = 200 3400 2.0 Pitted surface layer to 1/2 mil
 MSRP-11 20,000 123 1300 200 3200 2.0 ' Pitted surface layer to 1 mil
MSRP-12 14,498 134 1300 200 2300 1.8  No attack
MSRP-13 8,085 - 136 1300 200 3900 2.0 Heavy surface roughening and
T el = ' pittlng
MSRP-14 9,800 - Bu-14 + 0.5 U 1300 200 Pitted surface layer to 1/2 mil
MSRP-15 10,200 Bu-14 + 0.5U - 1400 200 Pitted surface layer to 2/3 mil
MSRP-16. 6,500 Bu-14 + 0.5 U 1500 200 Moderate subsurface void forma-

" tion to 4 mils

 
 

 

24

Table 6. Corrosion Rates of Inserts Located in the Hot Legs of Hastelloy N
Forced—Convection Ioops as & Function of Qperating Temperature '

Loop temperature gradient: (360°F)
Flow rate: = approximately 2.0 gal/min
Reynolds number: approximately 3000

 

Insert b Time Welght Loss Equivalent Ioss

 

Loop Salt
. Temperature per Unit Area in Wall Thickness
Numb M & o
er Maxture” "m0 Tg/en?) T ()
9354-4 130 1300 5,000 1.8 2.0
- ~ | - 10,000 2.1 2.3
o . \ 15,140 1.8 2.0 -
'MSRP-14 Bu-14 1300 2,200 0.7 0.8
- ‘ , 8,460 3.8 4.3
; 10,570 - 5.1 5,8_'
MSRP-15 . Bu-1l4 1400 - 8,770 .11.2c 12.7
T -~ - 10,880 10.0° 11.2°
MSRP-16 Bu-14 - 1500 5,250 9.6c 10.9 .
7,240 9.0 9.1

 

83a1t Compositions:
130 LiF-BeF,-UF; (62-37-1 mole %)
Bu-14 LiF-BeF,-ThF,;-UF,; (67-18.5-14—0.5 mole %).

Same as maximum wall temperature,

Average of two inserts.

faintly discernible. PFigure 15 illustrates the metallographic-appearance
~ of hot-leg surfaces from Hastelloy N pumped loops after various operating
‘times at 1300°F. Chemical analyses of exposed surfaces suggested that the

layer was an intermetallic transformation product of the nickel-molybdenum

system.

- Although most of our work was with Hastelloy N, we looked at the
behavior of several nickel-base alloys with variations of the Hastelloy N
- composition.1® Several alloys were screened by_thermal-convection_loop,
tests at 1500°F in salt 107 (composition given in Table 7). 'The composie
-tion ef these alloys and the depth of corrosion are shown in Fig.-lé.

 

165, H. DeVen, Effect of Alloying Additions on Corrosion Behavior of
Nickel-Molybdenum Alloys in Fused Fluoride Mlxtures, M.S. Thesis, the
Uhlversity of Tennessee, 1960. - '

‘w
7 " . .
¢ 4
i &

e
 

. | s

 

= [ - 0070 - 19459].

 

 

 

 

 

9,700 hr 4,500 hr 20,000 hr
Fig. 15. Effect of Operating Time on the Corrosion of Hastelloy N

 

 

 

’ Forced-Convection Loops Operated with Mixtures of LiF-BeF,-UF;-ThF,
(62—-36.5-0.5-1 mole. %g Maxlmum_salt-mgtal‘interface temperature: 1300°F.
= Loop AT: 200°F. The. differencee in microstructure smong the three
specimens below the surface layer are attributable to differences in
metallographic etching techniques rather than the test conditions.
Table 7. Composition of Fluoride Mixture Used to
Evaluate Experimental Nickel—Mblybdenum Alloys
Component . Mole% - . Welght®
F . 453 24
KF ;;'a;;'41 o 494
'asalt-107jgliQfiiGUS_témperéfure;490°Cr

 

 

 

 

 

 
 

 

DEPTH OF CORROSION {in. x 103)

" ALLOY ADDITION {at%)

26
. UNCLASSIFIED
ORNL—LR—DWG 46944

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

[ @ N
3 N woor- S 7 — />\ \‘ % |
S NN
Nem ADAD D DN ANN Y
vo | 1036|1268 | 1135|1078 | 1400 | 102 | 10.2 [48:77 |14.20| 1042 | 966 | 1005

ol T [es 4.-30 479 sa7|  |ss0 654 |
Fe - .56 | 0.3

al | 140 |61 [208 [ n30| |24 Ler | ~ |306 |55t ._1.55_ 264

n 202| 271 10| fres| |is|

Nb [3.23|087 | 148 | ‘ 291 0.9‘2 1'.57. ) __ 6;39_

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig. 16. Depths of Corrosion Observed for Nickel-Molybdenum K Alloys
with Multiple Alloy Additions Following Exposure to Salt 107. Bars desig-
nating corrosion depths appear directly above the alloy compositions which
they represent. = (Where bars have both positively sloped and negatively.

'sloped cross-hatching, the height of positively sloped cross-hatching

indicates depth of corrosion after 500 hr and combined height of both
types indicates depth after 1000 hr. )

Chromium increases the corrosion rate significantly. With chromium pfesént,
the addition of about 2 at. % Ti does not appear to increase the corrosion
rate. Hence,'we feel that the small titanlum addltlon that we are making

“will not be harmful at 1300°F.

 
 

 

#)

9

27

;;~MSRE 0perating ExperiEHCE ;3{:4ff;v

The MSRE initially underwent a series of zero- and low-power tests.

, sThe chemical analyses of the MSRE fuel and coolant salts showed very little
”-change, indicating a 1ow corrosion rate ‘of the Hastelloy N-and negligible
_Ttransfer of impurities from the graphite to the salt, This was substanti-

ated by the examination of graphite and metal sPecimens after the initial
tests, ' N . » '

The full-power operation of the reactor began in March 1965 The
first group of metal and graphite surveillance Specimens was removed in

July 1966 after eXposure to the reactor core environment for 7612 Mwhr

_ which included a - thermal exposure of 4800 hr at 1200°F The peak thermal

dose was 1.3 X 1020 neutrons/cm and the peak fast dose >l 2 Mev was

3 x 10%° neutrons/cm The concentration of chromiwn in the fuel salt
increased from an initial value of 37 to 48 ppm where 1t remained fairly
steady during the 7612 MWhr run The iron and nickel concentrations
remained constant at values of lO and l ppm, respectively Visual exam-
ination indicated the metal and graphite specimens to be in ‘excellent
condition. Machining marks were quite visible on the graphite gpecimens,
Several of the graphite and metal specimens were given to the Reactor
Chemistry Division for analysis with respect to fission product content 17
The graphite specimens have heen stored and mechanical property tests will

mhe run in the near future.i The Hastelloy N specimens were examined metal-
- ;lographically and subjected to tensile and creep—rupture tests.; Control

specimens for ccmparison purposes were exposed to MSRE—type fuel salt and

- -duplicated the thermal history of the in-reactor specimens

Metallographic examination of the Hastelloy'N showed the presence of

‘1fa surface film at’ points where “the graphite and Hastelloy N vere in
f'"intimate contact This product is shown in Fig. 17 and was present on
_'both the surveillance and the control specimens.r Electron micrOprObe
-p;:fexamination showed that the carbon content of the edge of the speCimen
'7i{ranged from 0. 3 to 1 2 wt % compared with a value of 0 05% for the inte-

'f;rior of the sample Nb evidence ef attack or carburization was found on

 

17y, R. Grimes, Chemical Research and Development for Molten-Salt

 

Breeder Reactors, ORNL- TV-1853 (June 1967).

 
 

Y-78266

=

tra

 

3
T

0.007 INCHE
=" 500X

T

 

 

 

b

Fig. 17. Edge of Hastelloy N Specimen Exposed ‘to Molten Salt for
4800 hr at 1200°F. Surface in intimate contact with graphite. Etchant:
glyceria regia.

metal surfaces that were separated from the graphite 1/16 in. This car-

burization is not a concern for the MSRE since we were aware of the problem

"before the reactor was built and placed sacrificial shims ‘between the

graphite and metal where contact was necessary.

Several of the surveillance and controi SPeeimens were'eraluated
by tensile tests. The total elongation at fracture when deformed at a
strain rate of O. 05 min~?t ig shown as a function of temperature in Fig. 18.
Both heats show some refiuction in ductility at lowrtemperatures in the

irradiated condition with a greater reduction being observed'for_heat 5085.

At temperatures above 932°F, the ductility of the irradiated and the

control material decreased with 1ncrea31ng temperature, w1th the 1rrad1ated
materlal showing a greater loss in ductility. At temperatures above
1202 to 1292°F, the control material exhibited improved ductility, whereas

the ductility of the irrediate&'materialvconfinued to deérease.

 
 

 

29

" ORNL-DWG 67-2452

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

- CONTROL = IRRADIATED
5 | 1 a © 508
o . A e  s063
€0 q ] €=.05 MIN"!
A R 1 l |
- V. T
' : ‘ ' 4 o ,
s0 =14 &% 7
- ; . LY
e |4 A \ A |
- T
: ¢
gqo / ' . : \ . : l/,
b 9’ 9 —ne S —— 3 : "
. % _..*_.‘—. — e s & S g 0 \. \ : ) ,,
. ‘ -. - : ) . .)4
a -x_gy L _
430 1 —
< k3
- N
B} N
20 *
10 =
_ ~
o e
o [ | 1 a4 L . I ) 1 2
o 100 200 300 400 800 600  7T00 ~ 800. 900 1000

‘ TEST TEM P, %c

~ Fig. 18. Comparative Tensile Ductilities ofIMSRE Surveillance -
Specimens and Their Controls et a Strain Rate of 0.05 min-1.

We compared the ductilities of the surveillance sPecimens with those

 for specimens irradiated in. other experiments without salt present

Heat 5081 had ‘been irradiated previously in the ORR.18 The ‘ORR experiment

was run at 1292 F to- a thermal dose of 9 X 1020 neutrons/cm and,the

material was in the as-received condition The MSRE surveillance speci-
mens were run at 1202° F to a thermal dose of 1.3 X 1020 neutrons/cm and

‘wlthe preirradiation anneal was different - However, none of these differ—
'_ences are thought to. be particularly significant and - the results can be
’ compered Figure 19 shows that the postirradiation ductilities of
o heat,5081 after both experiments are very. similar. | '

. Several of the surveillance specimens were evaluated.hy‘creep—rupture_

‘tests. . The results of these tests are shown in Fig.VZO _The specimens

 

18y, R. Martin and J. R. Weir,'"Effect of Elevated-Temperature
Irradiation on Hastelloy'N," Nucl Appl. 1(2), 160-67 (1965).

 
 

30 -

~ ORNL-DWG 67-3531

 

 

 

 

 

 

 

 

 

 

 

 

 

= 10
§ T s T

z ¢=0002 min™!

e ¢ | & THIS STUDY, ¢, =13 510",

%' 08 T=650°C

g 47 & ORNL-TM-100S, ¢, -9-|oz°m

Y r=700°C

2 :

"c‘_, 06 +

o

m -

4 &

5 o4 - v

x

o A

5 \ |
2 3

o2 7

<

& .

400 SO0 600 700 800 800 1000

. TEST TEMPERATURE (*C)

Fig. 19. Comparative Effects of Irradiation in MSRE and OBR on the
Ductility cf Hastelloy N, Heat 5081

onm.-pwe-cufls -

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

o :
\ifl , 1 uoes.@...-zxn’ ORR
N : A;oasp..-um'.mnz :
\ 1 loss 3_@m-.3no'?nsnt '
€0 N
N 1]
80 N UNIRRADIATED ALLOYE
%1, tzg} N
3 {9 | | 'H j L
§ | TROS lus n N
40 L &-H "y O= "
2 AVERAGE CURVE FOR SEVERAL' || ____.3-...__ N
: RADIATED ALLOYS (ORR) Ayl R
¢ 4 a8 '
S NN
"\ {125]
M
20 NS
‘2 N /
10
'Y 0 ) " 00 1000 10,000

RUPTURE TIME (KR)

Fig. 20, Comparison of MSRE Surveillance Specimens with Specimens
Irradiated in the ORR. Numbers in parentheses indicate ductilities.
 

 

»

1)

¥

product absorption.--

from the MSRE had propertieS"rery'similar'tO“those observed for specimens

- irradiated to comparable doses in the ORR.

The most 1mportant question to be answered concerning these data is

how - they apply to the operation of the MSRE

The surveillance specimens were exposed to a thermal dose of

1.3 x 1020 neutrons/cm ~ (The MSRE vessel will reach this dose after
about 150,000 MWhr of operation.) ‘This burned out about 30% of the 10B
~and produced a helium content of about 107 5 atom fractlon 1n both heats.

The high- temperature tensile and creep-rupture properties are exactly what

- we would expect for this dose Our work indleates a saturation in the
| degree of radiation damage at a helium atom fraction of about 10'5 and we

feel that the properties of the material will not deteriorate further

| The.lowetemperature_ductility reduotionmwas not expected. It is thought

tO'be_a‘result'of graineboundary‘precipitates forming“due to the long
thermal exposure. Irradlation playa some role in this process that is yet
undefined The low-temperature properties are not "brittle“ by any stan-
dards, but will be monitored closely when future sets of surveillance
speC1mens are removed. L -' ' o . |

The MSRE has since operated for a total of approximately 32,500 Mwhr.

- The chromium content of the salt has been somewhat higher during this run
at approximately 60 ppm we removed a group of surveillanoe speoimens |
,(graphite and Hastelloy N) for study about May 15, 1967. The Hastelloy N
_removed was. of the modified type, With small additions of titanlum ‘and-
izirconium The graphite and metal specimens will be evaluated with respect

;:to corr031on, metallographic changes, mechanical properties, and fission

.Reeistancefto-GaseOustontaminants o

VOxidation Resistance'h .

The ox1dation resistance of nickel-molybdenum alloys depends on the

.l_rservice temperature, the temperature cycle, the molybdenum content - and
“'dthe chromium content, . The ox1dat10n rate of the binary nickel-molybdenum
--alloy passes through a max1mum for the alloy containing 15% Mo, and the
'-scale formed by the ox1dat10n is- N1M004 and NiO. Upon thermal cycling

 
 

 

32

from above 1400°F to below 660°F;_the'N1M004'undergoes a phase transfor-
mation that causes the protective scale on the oxidized metal to spall.
Subsequent temperature cycles then result in an accelerated oxidation
rate. Similarly, the oxidation rete of nickel-molybdemum alloys contein-
ing chromium passes through a maximum as the chromium content of the
alloys is increased from 2 to 6% Cr. ‘- Alloys containlng more ‘than 6% Cr
are insensitive to thermal cycling and to the molybdenum content because
B the oxide scale is predomlnantly stable Cr203 An abrupt ‘decrease (by a
factor of about 40) in the oxldation rate at 1800°F is dbserved when the
chromium content is increased from 5.9 to 6.2%. | - |

The oxidation resistance of Hastelloy N is excellént#‘and'centinuous"
~ operation at temperatures up to 1800°F is feasible. Intermittent use at
- temperatures as high as 1900°F could be tblerefed.'-For'temperetures;up'__
to 1200°F, theVOxidatiOn'rete is not measurable; it'is*essentielly zero
after 1000 br of exposure in static air; as well as in nitrogen containing
small quantities of air (thefMSRE'cell environment). It is estimated
that oxidation of 0.001 to 0.002 in. would occur in 100,000 hr ef opera-
tion at 1200°F. The effect of temperature on the oxidation rate of ‘the
alloy is shown in.Fig 21. '

Resistance to Nitriding

 

Hastelloy N does not-react-noticeably with nitrogen at temperatures
up to 2200°F in the absence of irradiation. 'waerer, in a neutron
environment, the nitrogen is dissociated'and same shalloW‘surface-reac-“
tion (a few mils) is encountered. Although only limited data are avail-
able, there appears to be a considerable variation in the depth of such |
layers with various lots of material. While such layers are not thick
- enough in themselves to affect the strength of the proposed cemponents, -
they are brittle and crack eas1ly, 50 that they could possibly act as

stress risers.

 

Compatibility'with Superheated Steam
In parts of the system Hastelloy N will be exposed to'snpercritical"'

steam having & maximum temperature of 1050°F and.a pressure of 3800 psi.

Some heat exchange equipment will have coolant salt on one side and steam

 
 

 

 

1)

| (WElG;HT‘. GAI‘~)2 ' (mg/cmé)z‘:_

33

, ' T S ~ORNL-DWG 64-2855R

 

 

 

4.5

~ INOR-8

 

4.0

 

 

35

 

v
o

 

N
K

 

»
o.

 

 

1.5

 

1.0

 

1800°F o —T"*

 

 

0.5

 

 

 

 

g

 

L

 

 

 

 

 

 

ol et VT Y
%0 s 10 10 200 250 300 30 400

 

" Fig. 21. Effect of Temperature on Oxidation of Hastelloy N.

 
 

 

34

on the other side. Because.of the higher pressure of the'steam, a failure
in the Hastelloy'N7wou1d resultrin:steam.being forced into the coolant-
 salt circuit. Thus, we cannot adopt the "cheap material and. frequent
maintenance" policy of the power industry. The stressacorrosion’type of
failure in boiling water and in superheated steam systems'has_been cbserved
for the austenitic stainlese steels in several nuclear systems;l?’zo_ The
resistance of high-nickel alloys'to this type of attack has also been
established and it seems reasonable that our system should utilize such an
alloy. 19, 20

, Hastelloy N has been included in steam-metal compatibility programs
 at BNWL, 21522 gn3 some of the results of these studies are sumarized in
Table 8. Hastelloy N exhibited much better resistance to corrosion by
steam than type 304L stainless steel. It also compared very favorably
with the other nickel-base alloys that were included in these evaluation
programs (e.g., Hastelloy X, Inconel 600, and Incoloy 800). Although the
depth of attack varied slightly with steam pressure and oxygen content, a
.rather pessimistic value for the corr081on rate is about 0. l mil attack
in 2400 hr. Assuming a linear rate of attack, which again 1s on the pessi-
mistic slde, the depth of attack in 20 years would be about 7 mils.
Assuming the more realistic parabolic behavior, the expected depth of
attack would be only 0.8 mil. Either of these values is quite acceptable |
and we feel that Hastelloy N is a very attractive material for use in the

steam circuit

 

%W, 1. Pearl, G. -G. Gaul, and G. P. Wozadlo, "Fuel Cladding Corrosion
Testing in Simulated Superheater Reactor Environments: - Phase II. Iocal-
ized Corrosion of Stainless Steels and High Nickel Alloys," Nucl Sci. Eng.
19 274 (1964) '

200, N. Spalarls et al Materials for NUclear Superheater Applications,
GEAP-3875 (1962).

21p, Claudson and R. E. Westerman, An Evaluation of the Corrosion
Resistance of Several High Temperature Alloys for Nuclear Applications,
BRWL-155 (November 1965).

S 22p, T, Claudson and H. J. Pessl Evaluation of Iron- and Nickel-Base
‘Alloys for Medium and High Temperature Reactor Applications, Part II,
BNWL-154 (Nbvember 1965).

 
 

 

 

Table 8.

 Summery of Data on the Corrosion of Hastelloy N in Steam

 

Test Results

 

. Welight Gain (mg/cm?)

Oxide Penetration (mil)

 

Efivironment?éhdfTesfilCon&ifiions

'.?-'Héstelldy_NI 

. Type 304L
Stainlegs_Steel_

 

] H&Stélldy'N - Type 304L

 

Superheated steam, cos e
1022°F, 3000 psi, 1062 hr :

Deoxygenated.steam,‘;‘-. L
932°F, BOOO'psi, 2400 hr
932°F 5000 psi, 2400 hr =~ =
(< 50 ‘ppb), 1022°F, 3000 p51,-'
2400 hr - o .
Oxygenated steam

uto4mm,maw mmpfl
2400 hr

Helium plus 15 torrs water vapor,f -
1500°F 300 hr, ~ 1 atm pressure

0.57

0.109

2.20

0.37

 

. Stainless Steel{f

 
 

36

Fabrication of Hastelloy N Systems and Components

The fact that Hastelloy N is a'material_from which a complicated
reactor system may be constructed has been demonstrated by the'successfnl
construction and operation of the MSRE. Evidence of its general recog-
nition by industry is demonstrated by its acceptance and.approval by the
ASME Boiler and Pressure Vessel Code, |

" Raw Material Fabrication

Hastelloy N has proven to be readily fabricable into the many raw
material forms (pipe, tube, plate,'bar, castings,etc.)‘required in the
fabrication of a compdex engineering system such as a nuclearreactor.
Fabrication of the various rew material items required for the MSRE was
by normal commercial sonrces Forming techniques'deveIOPed'for austen- .
itic stainless steels can usually be used for Hastelloy N with a small
increase in temperature.

Melting and casting is carried out by using conventional practices
for nickel and its alloys. Ingots for the MSRE were prepared by both
air- and vacuun induction or consumable arc melting Individual.melts up
to 10,000 1b and a total quantity of almost 200,000 1b were produced

| Castings have been made in molds of water-cooled copper, graphite, rammed ,

magnesia, cast iron, and sand. While castings are considerably weaker
than wrought metal at room temperature, they are slightly stronger at
MSBR.operating temperatures. It is easier to control the chemical anal-
ysis by vacuum melting and the metal has better mechanical properties
and fabricability. Vacuum melting has become & common commercial'practice
during the last few years, and this melting practice will probably be used
in.preference to air melting in the future for nuclear grade materials

In making the many different forms and sizes of material required
during the developnent and construction of the MSRE, Hastelloy N was fab-
ricated using the normal technigues for nickel-base alloys. The initial

fabrication or'breaking down of the ingots was accomplished by forging.
or extrusion. Temperatures varied from 1825 to 2250°F. - Secondary fabri-

_cation was accomplished hot and—coid; Acceptable techniques included
rolling, swaging, tube reducing, and drawing. The alloy work_hardensr

w
 

 

 

37

| when‘fabricated cold, but reductions in area of up to'50% are possible
7between anneals.. Temperatures of 2100 to 2200°F are used for hot rolling
“with reductions of about -10% per pass. L
~_ The standard finished annealing temperature for this alloy is 2150°F
~All material for the MSRE was annealed after forming or working. After
lfinal fabrication, all material was. stress relieved at 1600°F. For any
‘heat treatment the'cooling;rate to 600°F was limited to 400°F per hour
per inch of thickness. The slow cooling'helped to impart dimensional
stability and improved the ductility in the 1400 to 1800° F range.
A necessary adgunct of fabrication is an inspection to demonstrate

the acceptance of the product - Products fabricated from Hastelloy N
may be inspected With the same nondestructive testing procedures used

for other high-nickel alloys or austenitic stainless steels Tubing and
pipe received for the MSRE'were inspected using 1iquid penetrants, eddy

'currents, and ultrasonic techniques.
Weldingland Brazing of Hastelloy N

Extensive experienceuith-flastelloy N has shown-that it exhibits
relatively good weldability. mhe MSREtcontains literaily'hundreds of
satisfactory tungsten-arc welds in varied section sizes. The welds were
- made using procedures developed at ORNL for nuclear quality applications
The subsequent inSpection of these welds showed that they met the very
tight 'ORNL requirements ' '

| Hastelloy N has also been shown to be readily brazeable. Good
:_wetting and flow have been demonstrated using commerCially available
brazing alloys and conventional brazing techniques.; o

During the course of the investigation of the weldability of
: Hastelloy N} welding prdblems were encountered _ Hot cracking of the type.
’occasionally encountered in welding high nickel alloys was sometimes
iobserved in heavily restrained JOints both in experimental heats and in

';'later commercial heats. In some of these unacceptable heats, the cause -
7of cracking was established as high boron content, however,'certain lower

boron heats were stil1l crack-sensitive for unknown reasons Although the

exact cause of the trouble;was never,established,.indications‘were that

 
 

38

it was associated with the proprietary melting practice used by the manu-
factnrer;. The impurity specifications were relaxed slightly, the melting
practice was revised, and weldable heats were produced. ' :

Although these changes upgraded the quality of the commercial pro-
ducts they did not assure the complete elimination of the propensity
toward cracking. A crack test was conceived and used &s the basis for
segregating unsatisfactory heats.”” Since the effects'of processing vari-
ables and impurity levels on welding were not determined' the‘weldabilr'
ity of a particular heat had to be determined by a test weld. Conse-—
quently, metallurgical studies aimed at understanding the basic reasons
for cracking and-eventually reaching'a more permanent solution were ini-
tiated. Cracking was shown to be essociated-with'the segregetionof
allofing elements in the heat-affected zone of Hastelloy N weldments. 23
Figure 22 shows the microstructural changes that can occur'ée a result
of welding; microprobe analyses of these samples showed that the brittle
eutectic-type structure had a markedly different composition from that
of the matrix. ' - - _

To date, the gas tungsten-arc welding process is the only technique
that haS'been used for the construction of reactors'of this naterial |
The MSRE experience has proven the applicability of this process " The
weld filler metal used for Joining Hsstelloy N has been of the same
basic composition as the wrought product.

The room-temperature mechanical properties of Hestelloy N'weldments.
have been extensively investigated. In the course of qualifying welding
procedures for this alloy in accordance with the ASME Boiler and Pressure
Vessel Code, numerous bend and tenS1le tests were conducted The results
readily satisfied the Code requirements However, as was expected the
ductilities of the welds were not as good as those obtained for the base :
metal. Elevated-temperature studies of welds24,25 showed that_theychadr

 

23R. G. Gilliland, “Microstructures of INOR-8 Weldments, Metalsk |
and Ceramics Div. Ann. Progr. Rept. June 30, 1965, ORNL-3870, pp . 24648,

2R, @ Gilliland end J. T. Venard, Welding J. (N Y.) 45(3), 1035-105
(March 1966).

- 2%sR Program.Semiann. Progr Rept Aug 31, 1965, ORNL-3872,
PP- 94—101

 

 

 
 

 

 

 

 

 

 

 

[ 4l

39

-

 

1N00X -

V00D INCHEY
iro

Tew

]
i

 

000X

QUUSD INLHED

1

 

 

 

[ PP g
. Fig. 22. Mlcrostructural Changes that Occurred as a’ Result of

~ Welding. (a) Structure of unaffected base metal showing stringer-type -
phase; (v) eutectic-ty'pe structure in heat-affected zone. = Electrolytically

etched in HaPO,. ' '

 
 

 

 

 

40

lower duotiiity and creep-rupture_strength than the base metal. However,
stress relieving at 1600°F improved the properties to where they were
comparable with those of the base metal. |

~ The room- and elevated-temperature shear strengths of Hastelloy N
joints brazed with Au-18% Ni filler metal nave also been determined.26
Joints tested at room temperature hed-antaverage shear strength of
73, 000 psi, while those tested at 1300°F possessed an average strength
of 18,000 psi. Diffusion and mlcrohardness studies of aged brazed joints
'showed that they possessed excellent microstructural stablllty.

Joining for Reector Component Fabrication

Pressure Vessel and Plping

As mentioned prev1ously, only‘weldable heats of Hsstelloy N were
used for the MSRE application. The tungsten-arc welding,process, with
| filler metal additions, was used throughbut,randltheifabrication was
carried out without‘undoeodifficultj A weidifig was ‘done in accor-
dance with ORNL-approved ‘specifications by qualifled operators The
flnished product readlly met the strict ORNL requirements for nuclear

systems.

Heat Exchangers: | | 7

The production of reliable‘tsbe-to-header joints for.heat exchange
applications_in molten-selt systems-is msndatory.' One means of assuring
this reli&biiity, welding and. bhck-brazing; was successfully developed
and tested exten51ve1y under severe operating eondltlons for the
Aircraft Nuclear Propulsion Project. 27 The technique was further
improved and adapted for use in the fabrication of the primary heat

 

26E, C. Hise, F. W. Cooke, and R. G. Donnelly, "Remote Fabrication
“of Brazed Structural Joints in Radioactive Piping," Paper 63-WA-53 of ‘the
Winter Annual Meeting, Philadelphia, Pa., November 17—22 1963 of the
American Society of Mechanical Englneers

27G. M. Slaughter and P. Patriarca, Welding and Brazing of High-
Temperature Radiators and Heat Exchangers, ORNL-TM-147 (February 1962).

 

 
 

 

 

 

 

 

 

 

 

¥

X

 

 

 

41

exchanger for the. MSRE 28 F:Lgure 23 shows the comple'bed tube bundle

in its handling flxtu_re. Figure 24 is a sketch of the omnt 'I‘o'date,
several thousand Jo:p.nts have been fabricated by the welded and back-brazed
technlque without a 51ngle reported failure.

 

28R. @. Donnelly and G. M‘ Slaughter, We.ldi.ng J. (N'.'Y..' ) 43(2),
118-24 (Pebruary 1964). = . T =

 

 

 

 

 

 

 

Fig. 23. Completed Tube Bundle in Fixture.

 
 

 

42

ORNL-LR-DWG 65682R3

—TUBE

 
 

 
 

TREPAN GROOVE WITH
/ BRAZING ALLOY RING

BRAZE SIDE

_THREE ‘
/FEEDER HO
7

WELD SIDE ' P S \TREPAN
| (a) BEFORE WELD!NG AND BRAZING

i
N Y

N

 

]4%&%%&2%%%&%4»%%&&%2»%&

R\

27
NS :
B 7SS S A ST TSI ST LS LTS TIS SLS S LSS LIS,

N

 

[
e

|
,.:Z::_'A.\. SR - ‘ L
L AT PTII RIS SIS TS S IS S S S S LS LSS SIS, ‘

 

g@mnmm&&%%%fi%%m&hfi%n%«-

 

(&) AFTER WELDING AND BRAZING

Fig. -24 Sketch Showing Joint Design Used for Welding and Back-
Brazing MSRE Heat Exchanger ‘

 

'Dissimiiar Metal Joints

 The incorporation of a variety of camponents in a reactor system
freqfiently requires the use of dissimilar metal weidS"'Wélding proce-
dures have been developed for joining Hastelloy N to several different

materlals, 1nclud1ng Inconel 600, austenitic stainless steels, ‘and nickel.
 

r

§

 

43
Remote Joining

‘Molten- salt reactors have: the characteristic that the fuel circulates
through much of the external piping system \ During operatlon, residual
activity is 1mparted to the walls, and during drainlng some fuel always
remains either in the crevices or by sticking to the wall. The resulting
radioact1v1ty requires that remote maintenance procedures be used, and

they include remote Joining

Remote Welding

While'many'remote'joining methods'have been-developed for the nuclear:
industry,they'were_alwayS”for'special-applicatiOns such as replacing the
seal weld on a vessel, —or'making-a closure weld for'a'capsule or a fuel
element, %° Commerc1al equlpment is not generally available for such
aIgflications.. The nearest approaches to developed remote maintenance
systems for a large reactor were one designed and partially developed by
Westinghouse Electric Corporation, Atomic Power Departmentjrfor use with
the proposed Pennsylvania Advanced Reactor,30 and one at Atomics

International for use with'aflarge sodium graphite reactor.3! However,

these projects were terminated_before'the developed techniques could be

demonstrated. ':Includediwere-teChniqnes for plugging heat exchanger tubes
and for butt welding of pressure Piping. |

The practicality of remote operations has been- improved by the '
recent rapid advances in the automatic welding field aimed at the nuclear
and aerospace 1ndustr1es, an excellent example of an automated machlne is
the one developed for welding the HFIR fuel elements. The only function

“of the operator is to pos1tion the torch and o press the start button;

'both of these operations could also be done automatically 1f de51rable.

 

29G M. Slaughter, Pp. 1?3—86 4in Welding and Brazing Techniques for

"Nuclear Reactor Components, Rowman and Littlefield, New York, 1964.

30g, H. Siedler et al., Pennsylvania Advanced Reactor —-Reference

feDe31gn Two, Layout and Maintenance, Part I, WCAP-llO4 Vol. 4, (March 1959).

.31, Newcomb, Calandria Remote Maintenance Tool- Development
NAAPSR-11202 (Apr 15, 1966}.

 
 

b

Brazing and Mechanical Joints

 

 Studies on remote Joining,hsve also included brszing and'meohsnical

joining techniques. Standard ring joint flanges were used at high pres-
sures but lower temperatures in the Homogeneous Reactor Expériment.; |
Freeze flanges are used to join the major pieces of equipment:in the'
MSRE. Similar developments have been made for other reactors and for
fuel reprocessing plants. in all cases the jolnts were developed for.
use in smaller pipes than are being proposed for the MSBR. The principal

deterrent to the use of mechanical joints is the dlfficulty in repeatedly
| assuring a seal, particularly in high-temperature cyclic service.
| - Equipment was developed and tested for remotely cutting and bra21ng

the drain line and the salt piping in the drain tank cell of the.MSRE.
The piping is standard 1 1/2-in. sched-40 pipe. These joints were .
cafefuiiy inspected _snd fiere determined to be of high qus.lity. '

Remote Inspection

 

A.necessary accompaniment of remote welding must be a remote inspec-__‘

tion. Less development has been done on inspection than on weldlng
Sufficient work has been done at ORNL to show that such operations are |
feasible and the required techniques should be possible with a_limited B
development program. | |

Ultrasonic techniques along with remote positloning equipment were
developed and used for the remote measurement of the wall thickness of
the Homogeneous Reactor Test core vessel. 3? The precision'of messurement

vas approximately 1%.

~ Ultrasonic and radiographlc technlques were developed but not trled |

for the remote inspection of the EGCR through-tube weldments.Bsflir

 

32R, W. McClung and K. V. Cook, Develomment of Ultrasonic Techniques

 

for the Remote Measurement of the HRT Core Vessel Wall Thickness,
ORNL-TM-lOB (Mar. 15, 1962).

33R. W. McClung and K. V. Cook, Feasibillty Studles for the Non-
destructlve Testlng of the EECR Through—Thbe Weldment, ORNL TM,46
(NOV. 14, 1961). ] | S |

 

 

 

§

oy
 

 

 

[ 1]

45

Another class of remote inspection devices has been developed for

use in hot cells. Radiographic techniques have been shown to be appli- |

~cable to the evaluation of radioactive materials,in high radiation )

backgrounds. This includes both television®4 and film techniques.’’
STATUS OF. Dm_aomm oF VGRAPHITE

In the proposed MSBR, graphite is used in the core as & pipe mate-

‘rial for separating the fuel and blanket salts and also for moderator

blocks around the tubes The former is the more critical use. The
graphite in the tubes must have low permeability to salt and must be free
from flaws that would permit appreC1able seepage of- salt between the

 fuel and blanket streams. A,value for the permeability to gases has not

been specified yet but a very low one is de51rable to minimize the rate

of diffusion of xenon- into thefgraphite and thereby minimize the loss of
neutrons to +>%Xe. - Although high strength is desirable, 1t need not be

‘higher than has already been obtained readily Techniques sre required
- for joining graphite to graphite and graphite to metals. Finally the

graphite in the MSEBR will be irradiated to a dose as high as

2 % 1032 neutrons/cm (E >'0 18 Mev) per year and the tubes must perform -

satisfactorily at this irradiation level for at least three and poss1b1y:
five years to have g practical power reactor

Graphites have been used for many years as. the moderator in a

,.variety of power and pdutonium production reactors where the require-
‘ments are less. severe and the radiation levels are much lower - The type
?'CGB graphite used - in the. MSRE approaches the requirements of the MSER -

- but the MSRE 1s not. a two—fluid $ystem and is less. demanding Loosely

| :assembled bars were used small cracks were not serious faults, and the

"._radiation levels were almost two orders of magnitude lower ‘than those 1n‘
"the MSER. Some of the newer isotropic graphites promise greater stabilityr

_1-under 1rradiation than does the anisotropicntype CGB. Improvement of

 

o 34J Wallen and R .W.vMeClung, An Ultrasonic, High Resolution, X—Ray
Imaging System, 0RNL-2671 (May 1959)

35R. W. McClung, Mater Evaluation 23 (1), 4145 (January 1965).

 

 
 

thdse’materials to satisfy other requirements imposed by the usecof'molten- o
salt fuels could result in a better graphite for use.in the breeder reac-
tors. Extrapolat1on of existing data indicates that the life will probably

be adequate
Grade CGB Craphite

_ Grade CGB graphite was designed for use in molten-salt reactors and
was first made in comnercial quantities for the MSRE. dIfi is'basicelly x5
a petroleum needle coke that is bonded withdceei—tar pitch, extruded to
rough shape, and heated to 5072°F. Improved dimehsional”stabilifyfunder =
irradistion is ensured by not using amorphous carbon ma%erialsrsuch-ase
carbon black in the base stock.- High density and low permeation are
achieved fhroughfmultiple'impregnatiens'and'heat treatments. All compo-
nents are fired to 5072°F or higher. The product is essentially a well-
graphitized material and.is'highly"anisotropic.w Its properties are
tabulated in Table 9. LT e Tl

Grade CGB graphite represents significant progress in the develop- “,
ment of a low-permeability radiation-resistant graphite. Ekperiments '
have demonstrated that the multiple 1mpregnations required to dbtain the
low permesbility and small open-pore size do not appreciably alter the |
dimensional stability compared to conventional needle-coke graphites.

In manufacturing the material for the MSRE, - experimental equipment
and processes were used on a commercial scale for -the first time by the
- Carbon Products Division of Union Carbide Corporation to‘pfoduCe'bars
2.50 in. square and 72 in. long which were machined to the shapes -
required for the MSRE. The bars were the 1argest that could be produced
by the then-current technology to meet the requirements of. low permeabil—-
ity to salt and gases end resistance to't&diatibn'effects. Assembly of
the MSRE core from bars of the graphite is shown in Fig. 25

- Structure

The MSRE graphite has an accessible void sPace of only 4. O% of: 1ts

bulk volume. The pore entrance diameters to the accessible voids are

L QO
b g g~ . —

o

 

 

47

T&ble 9. PropertiéS'of'MSRECo:e'Graphite;Grade.CGB
Phjsiéalrpropértiés | -
: : 8 /.m3 : : _
Bulk density, g/em® R el
Range - | o L — L 1.82-1.89 -
Average : - 1.86
.-Porosity,b'% | | o - c
Accessible
Inaccessible
Total . o | .
Thermal conductivity, ‘Btu £EL hr'17(°F)’1.-
‘With grain . : _
- At 9Q°F. 112
At 2200°F - L
Normal to grain [ | S
At 90°F . SRR 63
At 1200°F L o 34
| Temperature coefficlent of expan31on,c (‘M- |
With grain = ; o
At 68°F 0.27 x 10~6

b3
sJNJCD

Normal to grain o S . | ,
| At 68°F oL o 1.6 x 10-6
Specific heat;d ‘Btu 11 (oF)-l T D e
At O°F R o 0,14
- At 200°F '.,' _171;_ -;‘7 o 0,22
At 600°F L 70,33
CAt 1000°F - o - 0.39
: At 1200°F . e | - - 0.42
Matrix coefficient of permeability to helium 3 x1074
~ at 70°F, cm /sec :1' S e .
"Salt absorption at 150 psig, vol % o 5 ;iff 0,20
Mbchanlcal strength at 68°F ' e T |
- Tensile strength, psi
With grain o ,;a,,-;;-'; L el e
"Range xf{ TS o 15006200
Average Q*irinjgj;f‘ o S 11900
NOrmal togrein © .
R Range  43'f¢a”5iTiifff' S '1 :,1 L ;fl;mf 5{{1100fi4500
-~ Average o 1_;:_¥'_‘  afj‘_‘ S 7-14OQ
Flexural strength, psi LT Ll
With grain ,'__-,=;*;]gr* B S
-~ Range . o . -~ 3000-5000
Average o o 0 4300

 
 

S48 : o

 

 

 

‘Table 9. (continued) - | ¥
Mechanical strengthb at 68°F
Flexural strength, psi.
- Normal to grain | | | IR
Range - | - .2200-3650
- Average o 7 . 3400
Modulus of elasticity, psi T
With grain | . -+ 3,2 X 108
Normal to grain - ' : ' 1.0 x 106
Compressive stren'gth,dpsi' ConT T 8600
Chemical purityf - | | L
Ash, wt % - - | | - 0,0041
Boron, wt % , o 0.00009 .
Venadium, wt % . - - - 0.0005
- Sulfur, wt % | | | - 0.0013
Oxygen, cm? of CO per 100 cm of graphite 9.0
Irradiation data® e IR
[ Exposure: 1.0 x 102° neutrons/cm , (E>2.9 Mev) .
Shrinkage, % (350-475°C)] | | - | r
Ao . | | 0,10 |
aMeasurements made by ORNL. ’
blesed on measurements made by the Carbon Products Division and
ORNL..
“Measurements made by the Carbon Products Division. |
d'Representsti\are data from the Carbon Products Division. 7
®Based on measurements made by the Carbon Products Division on
pilot-production MSRE graphite. _ -
f1'.}zatta. from the Carbon Products Division
small, over 95% of the entrances are < 0.4 u in diameter.3_6 'I.’he molten |
salts do not wet the graphite, and calculations indicate. that pressures
in excess of 300 psig would be required to force salt into these acces-
sible voids. The microstructures of a conventional nuclear graphite and
the graphite in the MSRE are compared in Fig 26. '
24, H. Cook, MSRP Semiann. Progr. Rept. July 31, 1964 ORNL-3708 . —

 
49

Photo 70797

)

 

"

$t

:
i

 

 

L - Fig. 25. Anisotropic; I
| | -~ for MSRE Core. o

 

   

 

ngh-density and small-pore structures are des;rable characteristics

 

; - - of unclad graphite for molten-salt reactors, HOwever, for the MSRE bars,

 

-this particular structure was perfected to such a degree that in the
’final fabrication operations_gases could not escape rapldly enough

 

through the natural pores. in the’ graphite.' Consequently, the matrix'of
thefgrephitéferacked'as*thefgaseS'fOrCed their way out. However,
0.50-1in. -thick sections (heevieet wall for MSBR fueletubeS)~have been

made successfully without cracks.

 

 
 

50

“PHOTC 67670

 

asoT ceB

1.68 g/cm3 * "f 186gmm3
247 % ACCESSIBLE VOID VOL. 4.0 % ACCESSIBLE VOID VOL.
3.7 % INACCESSIBLE VOID VOL.  14.0 % INACCESSIBLE VOID VOL.
25.4% TOTAL VOID VOL. = _"1 e.o' o TO_TAL VOID VOL.

Flg 26 The Mlcrostructures of a Conventional Nuclear Graphlte,
Grade AGOT, and of the MSRE Graphlte, Grade CGB

Mechanical tests indicate that"the-—'prbc'essing-cracks, in the graphite
resist propagation.37 Since it is probable that cfacks'ih'gréfihite will
£fi11 with salt, tests were made to determine if repeated melting and
freezing of the salt would propagate the cracks. _Imfiregfiatédépecimens
were cycled from 390 to 1830°F at various rates of temperature rise.
Specimens and cracks were not detectably changed by the thermal qycles,38’39
The maximum temperature in the tests considerably exceeded the 1350°F

expected in the MSRE or the MSEHR.

 

37c. R. Kennedy, GCRP Semiann. Progr Rept. Mar. 31, 1963, ORNL-3445,
. 221,

38y, H. Cook, MSRP Semiann. Progr. . R@t July 31, 1963, ORNL-3529
p. 76.

39y, H. Cook, MSRP Semiann. Progr. Rept July 31, 1964 ORNL-B'?OS
p. 382.

L1

 
 

 

 

»

"

51

 Permeability

'Some'of the more”stringent specifications on graphite for MSER will

_ be on permeability. 4O Values will be specified for perrieability to both:

molten salts and to helium The screening test for salt permeability is

to expose evacuated specimens for 100 hr to molten salt at 1300°F under

8 pressure of 150 ps1g.4 The MSRE de51gn spec1f1cations reqnired that

f-less than O. 5% of the bulk’ volume of the graphite should be filled with
salt and that the salt should not penetrate the graphite matrix Speci~
‘mens of grade CGB had an average of less than 0. 2% of their bulk volume

permeated by salt in these screening tests, and . the salt distribution

-~ . was limited to shallow penetrations at the surfaces and along cracks
" which intersected ‘the exterior surfaces. The salt that entered the

cracks was confined to the cracks and a8 shown in Fig 27, did not

.. permeate the matrix. The permeation characteristics of the graphite as

 determined by test should be representative since irradiation does not

appear to change the wetting characteristics of the salt and the screening

pressures are three times the maximum design preBSure for the MSRE core.
The limit on gas permeability ‘has played a major. role in establishing

the fabrication sequence for the graphite Low permeability to gases is

desirable to prevent gaseous fission products from entering the graphite.

We would like a graphite for the 'MSER with a helium permeability of

10"7 2/sec. While it has been possible to obtain graphites meeting the

o salt-permeability requirements, 1t has not been possible to reduce the

- gas . permeability to the lO 7 level | Crack-free CGB graphite in the MSRE

f;pexhibited average helium permeability values of 3 X 10'4 cmz/sec As
"jpjpreviously discussed this graphite was fabricated using special proce—_.

f;fdures aimed at low porOSity.. The graphite tubes for MSER'will have much
_"ihthinner cross sections and should be less permeable. ‘At present, it
pf'fiwould appear. that values of O s cmz/sec would ‘be about 2ll that cen be
,fgdproduced by present technology Experimental and other spec1a1 graphite |
.'“_.ypipes have been produced w1th permeabilities less than 10'5 cmz/sec

 

40Hielium is a convenient medium to use for quality control to
establish permeabllity 11mits to gaseous fission products.

 
 

 

 

 

    

 

  

 

  

   

 

 

   

 

 

 

 

 

520 | B L

 

 

 

 

 

' Fig. 27,1_Radibgréphs of Thin'Seétiofis Cut from an Unimpregnated
Specimen and Three Salt-Impregnated Specimens. (a) Control (no salt = - .
present); (b), (c), and (d) salt-impregnated specimens; white phase is . . o

 

salt. _ , 7 - o | ) = o -

 

   

 

i}

 

 

 

 

 

 
 

 

 

"

”

53

(refs. 41, 42). The 1atter~contained carbon black in the base stock which
would significantly decrease itsdimensional stahility-under radiation.
The former had amorphous‘carhon impregnantS‘ but'this'appears to cause
only slight decrease in its. dimens1onal stability under irradiation. %3
While & homogeneous graphite is desirable, one w1th a surface seal
must also be considered Such 8 seal may be & penetrating, integral
skin of graphite or a thin coating of 92Mo or Nb The metallic coatings
would be applied hyrdeposition from the metal chlorideggas at elevated
temperatures. The surface COating_would_be quite thin (approx 0.0001 in.)
but would be applied under.conditions where the metal sould be linked into

the pore structure so that‘good.adherencefWOuid be. ensured,

Mechanical Properties

 

The tensile and flexural strength.measurements on grade CGB graphite
for the MSRE averaged71900 and 4300,p31, reSpectively,_despite the pres-
ence of cracks in the graphitert There'fiere no values less than the
minima of 1500 and 3000 psi spec1fied for ‘the ten511e and flexural
strengths, respectively. These strength ‘values were spe01f1ed primarily

to assure a good quality of . graphite.‘ The stresses expected in the

'reactor are very low in the absence of irradiation However, under fast-

neutron 1rrad1ation, the differences in shrlnkage rates across the 1/2 -in.

wall of a pipe caused by flux gradients can produce high stresses.

The tensile strength of ‘the MSRE graphite is. shown in Flg 28 The

 band with the single cross-hatching indicates the stress strain behav1or

: of the material However, some of the material contained cracks and |
lfallure occurred at lower stresses and strains as indicated by the solid

__;points ‘The crack- free material represented by the solid p01nts, failed

'd:at higher stresses Spec1mens that demonstrated definite effects of

| cracks (the ‘black- points) had a minimum strength of 1510 pe1 and an average

 

4*K Worth Techniques and Procedures for Evaluating Low-Permeability

_Craphite Properties for Reactor Application, GA~3559 . 7 (March 1, 1963).

421, W. Grahsm and M.S.T. Price, "Special Graphite for the Dragon

-Reactor Core," Atompraxis 11 549-54 (September—October 1965).

43G. B. Engle et al 3 Irradlation of Graphite Impregnated with’
Fu.rfuryl Alcohol, GA-6670, p. 2 @ct 5, 1965).

 

 
 

 

_54

ORNL-DWG $3-J354R

© NORMAL FRACTURE
® STAR-STEP FRACTURE
WITH A RISER 5 ¥4 in.

 

0 0.0% ) 45 0.20 028 0%
' STRAIN (%) S

Fig. 28. Tensile Properties of Grade CGB Graphite Parallel with -
Grain (Axis of Extrusion); Bar 45, 2 X 2 in.

strength of 2940 psi. Specimens that were not appreciably affected

by cracks (the open circles) yielded an average strength of 5440 psi and

g modulus of elasticity of 3.2 X 106 psi, corresponding to an unusually |
strong graphite. These results indicate the variation of properties.within'
a single bar (No 45); however, there was algo some yariation between
various bars. The double cross-hatched portion indicates the overall

range of prOperties obtained by sampling various bars of the MSRE graphite
Isotropio cfapme |

There has been recent progress in the development of isotropic '
grades of graphite that offer greater dimensional stability under 1rra~g
| diation and potentially can be used as a ‘base for the 1mpregnations to
Obtain the necessary low permeability. Development of this material
has occurred in the last few years and.much of the data are proprietary
and have not been released for publication. However, properties of
some available grades are listed in Tatleilo :

 
 

. Table 10,

 

 

 

 

 

 

‘D1men31onal 1nstabi11ty at 700°C

2 x1072

. 1/2 to 1/3 that of AGOT

| Properties of Advanced Reactor‘GradeS'of G-raphitea
o v _ - Gradet
- Property : — ' —
| o - H-207-85 H-315-A ‘H-319 .
-Density, g/cm3 o 1.80 - 1.85 1,80
Bend strength, psi j\.761“ - >6000 4500 ~>4000
;Mbdulus of elastlclty, psi S o 1.351. 65 X 106 R
;Thermal conductiv1ty,.\ | ‘i“fiffi; 22 22 22
Btu ft'l hr~t °pt (1832°F) a o |
- | / n
}Thermal expansion,(l L - 5.2-5.7 4.8-5.8 3.7%.4 \
108/°¢ (22-700°C) | o |
 Electrical resist1v1ty, 8.9 9.4 9.9
~ ohm-cm X lO E
'Isotropy factor, CTE”/CTEL 1.11 1.20 l;l8
Permesbility to helium, c /sec‘_ g x 1072

 

Adapted from G. B. Engle, The Effeet of Fast Neutron Irradiatlon from 495°C to 1035°C on Reactor

 Grephite, GA-6888 (February 11,

1966) .

All propertles measured at room temperature unless stated otherW1se.

Based on very prelimlnary results on H-207-85 and type B isotropic data to 2 X 1022 neutrons/cm

 
 

 

56

The isotropic graphite is made in various shapes with conventional
carbonraad'graphite molding and extruding equipment. Bulk densities can
range from < 1,40 to 1.90 g/cm3. The strengths of various grades tend
to exceed those of grades of conventional, anisotropic graphite.

Past work has not been directed toward producing isotropic graphite
with the low permeabilitles required for molten-salt breeder reactors.
Pipes of grade H-315-A (Table 10) have been made that are approximately
4.5 in. in diameter with O. 50-in.-wall thickness which would correspond
to the large MSBR pipes. These pipes had an average bulk den81ty of
1.83 g/cm3 and radlally they had a permeability to helium of 2 X 10‘3
to 1 X 10”2 ecm?/sec. The lower value refers to the unmachined pipe and
the higher value to pieces machined from the pipe. walls. Pore spectrum
measurements indicated that approximately 50% of the pore volume had
pore entrance diameters betweeh 0.5 and 3.5 p and less than'lfi% of the
pore volume had pore entrance dismeters > 3.5 p. This pore spectrum  '
suggests that this material, although it was not designed for it, might
be suitable base stock to impregnate to preduce 1cw-permeabi1ityrgraphite.
of ceurse, in fabricating graphite for MSBR use one woald want to opti-
mize the base stock for impregnation which would mean that the base

stock would be fabricated with pore entrances less than 1 p in diameter.
| In the following program, the isotropic graphite will be considered
as the reference material. However, until its development has proceeded
to the point that meeting the MSBR requirements is reasonsbly assured,
it will be necessary to continue the development of the anisotropic
needle-coke graphite as backup material. Most of the data obtained to
date have been with needle-coke graphite; however, much of this
technology is transferable to the 1sotropic graphite.

Joining

The MSBR is unique among reactors in that it uses graphite as an

engineering material and the present conceptual design calls for graphite-

to-graphite and graphite-to-Hastelloy N ,joints.44 These JOints,are oni

 

. 44P, R. Kasten et al., Summary of Molten-Salt Breeder Reactor De51gn'
Studies, ORNL-TM-1467 (March 24, 1966).
ambesine a1 2o gt o g ey £k
.

 

 

461-69 (1962).

g pp 101-04.

27

ithe core pieces but in a region where the neutron flux dose is reduced
\con81derably With all the tons of graphite that have been used in
| ',nuclear reactors, most of the work on graphite—to-metal Joining seems to

 have been done in small programs at ORNL" 46 and Atomics International

Graphite-to-Hastelloy N JOints o

Only 8 few brazing alloys are suitable for brazing graphite for
molten-salt.serv1ce,.since, in addition to,wetting abillty, the braze
material mustfbecorrOSion:resistantiandstable under irradiation. To
promote.bonding,;somechemicaiflreactiontwith the graphite is desirable;
therefore, most proposed:graphite brazes contain some "active" metal.
However, titaniumfand'Zirconium ;;the‘tw0'u8ually_pr0p65ed — may not be

suitable as far as corrosion resistance to molten fluoride salts is con-

. cerned. Similar‘statements can'he made about alloys containing large

quantities of chromium Thus,_brazing development in support of molten-
salt concepts has ‘been cOncerned'with developing an alloy that not only
will braze graphite but also,will ‘be corrosion resistant, melt_high

~ enough to be strong in nse‘butdlow enough to be brazed with conventional

vacuum or inert-stmosphere equipment, andjbe stablefunder irradiation.
'The‘low.cOefficientfof;thermal'expansion_and lowrtenSile strength
of the graphite may make it difficult to join it directly to-the'

‘Hastelloy N or otherlstructural materials. - The problem is further com-

. aplicated by -the- respective shrinkage and expan81on expected in the
graphite as 1t is exposed to the fast neutrons

‘ Two of several approaches are receiving the greatest study at this
'timé; One is to braze ‘the Hastelloy N to the graphite in such a way

‘_that the Hastelloy N applies ‘a compressive 1oading to the graphlte as

the Joint is cooled from the brazing temperature This is ‘desirable

'dfld_51nce graphite is stronger and more ductile in compression than in
_f,_tens1on.; The other approach is to 301n the graphite to materials that
"have coefficients of thermal expans1on approaching or equal to that of

 

43R, G. Donnelly and G M Slaughter, Welding J. (N Y, ) 41(5),

46p. B, Briggs, MSRP Semiann Progr. Rept Feb 28, 1966, ORNL-3936,

 

 
 

58

the graphite. These transition materials_would, in\turn,”be joined to
the Hastelloy N by brazing. Therefore, refractory metals, such as
molybdefium and tungsten, which have thermal expansion coefficients more
closely matched to graphite as well as much higher strengths, have been
selected as prcmising transition materials.
The joint utilizing the refractory metal transition piece ‘has
.received considerable attention. The ductile and corrosion-resistant
alloy, 82%;Au—18% Ni, was chosen:aé.a-starting pcint for mclfen-salt 1
applications. To this base meterial has been added the corrosion- ‘c,- _
resistant, strong carbide formers, molybdenum and tantalum. As a result,
‘alloys in both the Au-Ni-Mo and Au-Ni-Ta systems were developed4’ which,
fulfilléd;the above requirements wifh the'possibleexcepticn of hucléar
stability in a high neutron fluxi(gold_will_transmute to mercury:in high_ ,
flux fields). The most promising alloy contains 35 Au-35 Ni-30 Mo .
(wt %) and has been designated as ANM-16. Graphite-to-graphite and
| graphite-to-molybdenum joints have been made with ANM-16 by induction
braiing in an inert atmosphére and vacuum brazing in.a muffle furnace.
A brazing temperature of 2282 to 2372°F was used. Several componeuts
have been brazed with this alloy, the largest of which have been o
1.50-in. OD X 0.31-in.-wall graphite pipes with mclybdenum.end caps and
sleeves. dJoints of this type as well as T-joints have -been examined
metallographically and found to be sound in all cases.
If brazing conditions are controlled closely, the joints appear
sound in all respects. However, if the assembly is maintained at brazing:

 

temperature longer than necessary or if too high a brazingftemperature
is used, the alloy in the fillet spreads onto the surface of ‘the graphite
in a very thin layer. Upon cooling from the brazing temperature, this
.thifi leyer will crack. This behavior is aggravated when the surface of ..
the graphlte is smooth and free of porosity, such as with the desired _
| MSER graphite. Joints made with high-density CEY graphite pipe exhibited
this behavior, whereas it had not beeh noticed when brazing the.vefy
| pofous grades sfich as AGOT. 'MetallOgraphic examination,'hcqcver,wchqqs_”__
that the cracks do not extend into the body of the £illet,

e et s ikt i

 
 

[

chromium will not cause a corr051on problem

59

A short section-of'graphite pipe with a brazedggraphite-to-metal

‘end closure successfully"contained molten saltsfiforVSOO hr at 1292°F and

a pressure of 150 psig. The graphite pipe was machined to
1, 25~ 1n OD X 0. 75-1n. ID frOm a bar of MSRE graphite ~ Molybdenum end

~ caps were brazed to the ends of the pipe using braze metal ANM-16. No"
leak occurred during the 500 hr test period and posttest radiographlc

rexamlnation 1nd1cated that the salt was contained as planned Metallo-
graphic examination revealed no penetratlon of salt into the graphite
and no detectable corrosion of the braze material

Recently, an alloy w1thout gold has been developed to overcome the
poss1ble problems of transmutation This alloy has & composition of

’35% Ni—60% Efirfi% Cr and has exhibited brazing characteristics very much

~ like the 35% Au-35% Ni—BO% Mo alloy The chromium content was kept low

in the hope that acceptable corros1on resistance would be obtalned. The
brazing alloy wet both materials and appeared to form a joint as good as
that obtained with ANM-16.

Joints of graphite to molybdenum have been brazed w1th this alloy
and survived a 1300°F corrosion test (LiF-Bng-ZrF4-TthfUF4) of
10,000 hr with no attack.,.There wae_a'thin palladium-rich layer_on the
surface of'the_brazing'alloy;IEA.2Q,QOO-hr corrosion test has accumulated

10,700 hr to date. Chromium, the cerbide former, may be tied up in the

form of a corrosion-resistant carblde, Cr3C,, to such an extent that the -
47

To date, no mechanical property tests have been run on brazed

) graphlte 301nts at ORNL but graphite—to—molybdenum specimens ‘brazed W1th

the 35% Ni-60% Pd—fi% Cr alloy ‘have successfully withstood ten thermal

_;'cycles fram 1300 F to ambient., Metallographic examination revealed no
_cracks after this test o

 

47W, H. Cook, MEtallurgy de Semlann Progr. Rept April 10 1956,

| _-1‘GRNL-2080 pp. 44 and 46.

 
 

 

60 | -

R

Graphite-to-Graphite'Joints

| Techniques are commercially available for Joining graphite fs'ff'
itself.48 Most techniques are based on the use of furfural alcohol and
graphite powder The mechanism of sealing is by polymerization of the
- aleohol at temperatures to 1600°F The Joint ‘must be held under pressure
during the curing to prevent gross porosity The por051ty arises frOm 5
the gas that must escape during curing. It may be minimized by close
Joint fitups and slow heating rates The polymerization would be followed
by a graphitizing treatment at about 4892°F waever, it may be more
desirable to use & metallic brazing alloy to obtain a Joint with lower
permeability and higher strength Small graphite-to-graphite Joints have

been made at ORNL using the Au-Ni-Mb and Au-Ni-Ta brazing alloys, however,
| such joints were most effective when graphite that was much more porous
than the CGB grade was used Several other alloys are being investigated
that wet the graphite and are resistant to corrosion by fused salts.

Compatibility of Graphite with Molten Salts

Since the MSBR uses grephite in direct contact with the molten-salt -
fuel and the breeder blanket, good compatibility between graphite and
Hastelloy N in these molten fluorides is needed. Tests performed have
indicated that excellent compatibility exists. The tendency for
Hastelloy N to be carburizediwas investigated'in static pots containing
LiF-BeF,-UF; (67-32-1 mole %) and a graphite at 1300°F for times ranging
from 2000 to 12 000 hr. No carburization was metallographically detected
on specimens from.any of the above tests.49 Tensile properties were'

" determined on specimens included in each test. A comparison of the
tensile strengths and elongation values of the tested specimens with
those of the control specimens indicated that no slgnificant change

| occurred in the test specimens. _

 

48R, E. nghtingale, Nuclear Graphite, Pp. 62-65, Academic Press,
New York, 1962.

“%H. G. MacPherson, MSRP Quar. Progr. Rept. July 31, 1960, ORNL-3014 o/

 

p. 70.

 
 

 

b

™

'5gthe graphite under such conditions.g,_nf"

 

el
'For the'purpOSeioftevaluating thelconpatibility of graphite and'-

convection loop was operated for 8850 hr The 1oop operated at a max1-

‘ir;mum temperature of" 1300 F and circulated a fluoride mixture of ,
;-:LiF-Bng-UF4. On completion of the test, components from the loop were
7 nchecked metallographically and chemically, and specimens were checked
: for dimensional changes and'weight changes. The tests indicated that

(1) there was no corrosion or erosion of the graphite by the flowing
salt; (2) there.was.veryfilittle permeation of the graphite hy the salt
and the permeation'thatfloccurred]was uniform,throughout'the‘graphite

,rodspc(BJ the'various Hastellov’flrloop componentsfexposedlto"the salt
were not carburized; (4) the Hastelloy‘N components:expOSed‘to'the'salt

and graphite were negligibly attacked, and (5) with ‘the possible excep-
tion of oxygen contamination, the salt appeared to have undergone no

chemical changes as a result of exposure to the graphite test ‘specimens.

In-reactor tests 50 confirmed this good compatibility

- Radiation;Effects'on Graphite -

B Although graphite has heen used successfully in reactors for over
20 years, it is only within the past 10 years that the effects of irra-

idiation have been studied inten51vely These studles have been ‘responsi-
: Vble for 51gn1ficant improvements in- both the quality of the graphite and
{the development of reactor concepts that fully utilize the unique prop-
-erties of graphite ' The molten-salt reactor systems are very good :
*; exampdes of reactors that fully utillze the advantageous properties
of graphite. These systems will however, subject. the graphite to much
"Cff'higher neutron doses [for 8’ lO-year life, 2 % 1023 neutrons/cm
,_j'(>-o 18 Mev)] than have been or are being obtained in any other reactor
___l;fljsystem. These Large doses give rise to questions of growth strength |
”nzifliand dimen51ona1 stability that will determine the serv1ceable life of

 

5°w R. Grimes," Chemical Research and Development for MOlten-Salt
Breeder Reactors, ORNL-TM-1853 (June 1967) ' -

 
 

 

62

The thin-wall tubular désign used in the MSER is one of the better
configurations in reducing the differential growth problem to within the
capabilities of the'graphité. The tubular shape also has the advantages '
of good fabricability and‘reliable nondestructuve testing'techniquéé to
ensure maximum integrity initially. Extrapolation of irradiation damage
data for doses to 2 X 1022 neutrons/cm?® leads to the conclusion that
graphite should have an adequate life in the MSER. waever, we cannot
conclude this with certainty until'we have exposed graphlte to the doses"
anticipated for the MSBR.

Although theére is no experimental evidence beyond _

2 x 1022 neutrons/cm s, several factors lead to optimism. about ‘the abillty
of graphite to .sustain much greater‘damage. ‘The first factor is the_
recognition that in the 1292°F temperature range_the:cqntracturalvdimen_
sional changes for the first 10?2 neutrons/cm? are due to the repair of
the damage caused by cooling the graphite from its graphitization temper-
ature. Doses above lOzz_fieu.trons/cni2 produce an expansion in the “c"
direction and a contraction in the "a" direction. 1In effect, the
Stresses produced-normal to the basal planes will be compressivé rather
than tensile. This, of course, is very much preferred in that the
crystal can sustain this type of loading more readily without cracking.
Even so, measurements obtained from-irradiated pyrolytic graphites
indicate that a deformation rate of about 1% shear strain per

1021 neutrons/cem? must be absorbed. Thus, for a 5-year life, the graph-
ite tubes operating in a flux region of 6 X 1014 neutrons cm™? sec™

absorb an internal deformation of about 100% shear strain. There is evi-

must

dence, howeVer, that graphite can sustain this type of internal déformation
without creating micrbcracks.‘ Very fine crystallite pyrocarbons have

been irradiated under conditions prbdubing crystallite shearing'rates

of 60% per 10?2 neutrons/cm® (ref. 51) and have demonstrated the ability
to absorb 160% shear strain without observable internal cracking or loss

of integrity. Thus, 1t is difficult to predict whether the anisotropic

 

51The dimensional changes in graphite are conventionally expressed
as strain per dose. This is referred to as a strain rate although the
variation is with respect to dose rather than time.
 

 

H

»

a3

63

erystal growth Will-producemicrocracksin-the,structure; andlif.the |
| microcracks are formed, whetherrthey_Willlbe more;detrimental to the

structure than the initial microcraoks produced by cooling from'the
graphltlzation temperature | | o

As prev1ously mentloned, the relatively thln-wall tube des1gn
minimizes stresses due to the dlfferential growth, ‘The magnltude of the
maxlmum stress produced can be fairly well calculated with the main

_uncertalnty being the flux difference across the ‘tube wall, ~ Using a

conservative estimate of 2.4.% 10724 cmz/neutron as the growth rate
and a 10% flux drop across the wall a differentlal growth rate of
2.4 X 10725 cma/neutron is Obtalned The restralnt is internal; there-

| fore, about half of the differentlal growth is restrained to produce a

strain rate. Thus, & strain rate of 1.2 x 10723 cmz/neutron is obtained
for comparison to an estimated creep rate coefficient of
4 x 10727 (psi)™t The stress is s1mply and dlrectly calculated to be: -

Thus, the equllibrlum stress level is very low, probably'much below that
required for feilure even at hlgh radistion dose levels. ' |

-Failure,rhowever, could result from inability of the graphite to

-ebsorb the creep deformation even though the stress level is much less
~than the fracture stress. For 1ifetimes of 1 X 1023, 2 X 10%%, and

4 x 1023 neu.trons/cm2 (correspondlng to 5-, 10-, and 20-year llfetlmes),'”

_the strain to ‘be absorbed would be 1.2, 2.4, ‘and 4. 8%, resPectively
,The cons1derat10n of a straln 11mit for failure is. realistlc, ‘however,
“the strain limit for fracture has not been establlshed It has'been |
"__demonstrated that graphlte can absorb strains in excess of 2% in =
. 1022 neutrons/cm? under stresses in excess of 3000 psi. W1thout fracture.
rfThere is also some ev1dence that -growth rates of isotroplc graphltes
Cwill diminish after a dose of 1022 neutron/cm - Thus, the- graphlte o
"'will not be forced to absorb the quantltles of straln calculated |

 
 

 

64

Although there is no direct evidence that"- graphite can, sustain 1.2%
‘strain in 5 years of very high dose and’ low stress, the- indirectfev1-'

dence does indicate that a failure is improbable
Nondestructive Testingfof Tubing

Thererhas'not been extensive development of nondestructive tests
for graphite at ORNL and other AEC installations52 53 or by industry
However, past efforts at ORNL aimed at specific tests have produced
confidence that several NDT methods are applicable. These programs
have produced background knowledge that will aid in subsequent
development programs. '

. Several. low-voltage radiographic techniques have,been developed for

use on 1ow-density materials.f?4 . These innovations. have produced superior-

" achievements in sensitivity and have been applied to numerous. graphite
shapes. Sufficient sensitivity has been obtained to make it possible to
permit examination of details as small as l u in carbon-coated fuel
particles. 55,56

The eddy-current method has been shown to be applicable to graphite
| inSpection. One of the earlier applications of_this‘technique_was for
inspecting the graphite support sleeves for the EGCR fuel assembliesv57__
Because of the local porosity in.the,graphite, the sensitivityiinithis

 

52R. W. Wallouch, "Adaptation of Radiographic Principles to the
- Quality Control of Graphite," Research and Development on Advanced
Graphite Materials, Vol. IV, WADD Tech. Rept. 61-72 (October 1961).

53G, R. Tulley, Jr., and B. F. Disselhorst, The Pore Structure of
Graphite as It Affects and Is Affected by Impregnation Processes,
GA-3194, pp. 40, 47-48 (April 30, 1963).

54R. W. MbClung, "Techniques for Low-Vbltage Radiography,"
Nondestructive Testing 20(4), 248-53 (1962). "

3°R. W. McClung, "Studies in Contact Microradiography," Mater. Res.
Std. 4(2), 66-69 (1964). DR |

56R. W. McClung, E. S. Bomar, and R.- 3. Gray, "Evaluating Coated
Particles of Nuclear Fuel," Metal.frogr 86(1), 90-93 (1964).

 S7R. W. MCClung,'"DeveIOpment of Nondestructive Tests for the EGCR
Fuel Assembly,” Nondestructive Testing 19(5), 352-58 (1961).

 

"

 
 

 

 

65

-test_waS-limited to detection of discontinuities of approximately 10%
~of the l-in. thickness. |
More recent ORNL eddy-current work has included develOpment of the
phase-sensitive instrument which overcomes some of the disadvantages of ,
conventional equipment. This‘device:has been used to measure thicknesseS-
ofigraphite58-and to detect.flaws in the unfueled graphite shells of
fuel spheres58'Such as are proposed for advanced gas-cooled reactors.
Other testing with graphite has included studies of infrared o
methods59-to detect laminations or unbonded areas in the fueled spheres

- and ultrasonic methods to detect flaws and to measure elastic properties. -

_MATERIALS_DEVELOPMENT PROGRAM FOR MOLTEN-SALT BREEDER REACTORS

In the prev1ous sections of this report, we have shown that we have

& strong technical background in working with the: two prlmary structural

terials in the MSER, graphite and Hastelloy N However, we feel that
certain advancements in technology must be made with reSpect to both
materials to ensure the successful Operation of the MSBR We will
briefly outline the areas of concern and indicate the work necessary for |
develOping suitable technology in these areas. The cost estimate for the
materlals development for the MBBR is appended - | -

7 Hastelloy N Program

Resistance to Irradiation Damage |

 

f The major problem area with Hastelloy N requirlng additional develOp— '
7 :ment before it may be: used in the MSBR 18 that of 1mproving its resistance
fito neutron irradiation The present alloy is susceptible to a type of
._high—temperature radiation damage that reduces the creep-rupture life and
zthe rupture ductility Solving this problem w1ll be a maJOr consideratiOn

 

e 580, V. Dodd, "Applications of o Phase-Sensitive Eddy Current
,Instrument " Mater. Evalustion 22(6), 260-62 and 272 (1964).

59a, V. Dodd, GCR Semiann. Progr Rept ‘Sept. 30 1963 ORNL 3523
PP. 318—24 and GCR Semiann. Progr. Bept Mar. 31, 1964 ORNL-36l9
pp. 75-77 L R . {

 
 

66

.in establishing the schedule for the MSBR. The problem is complicated
by the long lead time required to obtain in-reactor mechanical property
data and by the fact that the solution appears to lie in a change in .
eomposition This'means'that'once a modified radiation—resistant'alloy
is developed, it will be necessary to determine if the changes in composi-
tion have affected any other properties. Another complication is that it
has been shown that the radiation damage is sensitive to fabrication
practice, so to be truly representative, material used will have to be
_gtaken‘from large commercial heatsr' Obviously, a compromise*is required
in which small laboratory heats will be used initially for screening with
the results being confirmed with material from the large heats. The
fabrication practices for the large and. small heats will be kept as nearly
alike 8s possible
Our work has shown that titanium, zirconium, and hafnium are effec-
tive additions that will reduce the radiation damage of Hastelloy N. We
have not established the exact mechanism re5ponsible forrthe ifiprofiement,
but feel that it is associated with the reactivity of these elements with
boron and other impurities in Hastelloy N. Con
During the first year, screening-type tests are being run. | The
major goal is to determine which of the additives appears ‘to be- the most
effective and to begin obtaining an indication of the optimum‘composition
Machined specimens are being irradiated in capsules at elevated tempera-
-tures; on removal they are used to determine the tensile and creep-rupture
properties of the material. During this period, most testing will be on
laboratory-size vacuum-melts and on small (100 lb) commercial melts.
These irradiations are being conducted in the ORNL Research Reactor and
,the Engineering Test Reactor. _ .
, A limited number of ccmpositions will be evaluated by being inserted
into the core of the MSRE. These samples will be added as the surveil-
lance specimens are removed for testing. These samples Wlll be tested
to obtain postirradiation'mechanical properties and studied metallograph-
ically to determine whether corrosion has occurred P
- The ‘second year will be spent in determining the optimum comp031tion,
heat treatment, and fabrication practice for the most promising additives.
| Since these composition changes may adversely affect other prcperties,

T

a
 

 

[ &

o

67

alloys based on &t least two different alloy additions will be carried
through this step. The majority of the testing Wlll still be postirradia—

tion tensile and creep-rupture. The results will, however, be confirmed.

by running a few'in reactorrstress-rupture tests. Both laboratory and
-'“small melts commercialLy fabricatedfiwill be tested with a shift gradually

being made to the latter type | _ : & - ,
, Iarge commercial heats (1500 lb) of a few compositions will be tested
during the second and ‘third years.' This material will also be available

“ for -other testing programs., These ingots will show if. scaling-up in size

has any adverse effects and also if materials from different vendors are

_Vcomparable When the. composition is firmly established we shall procure

& few full-scale commercial heats (10,000 lb) ,

Specifications must be issued during the first year of testing; they
will be continually upgraded by incorporating new data as they become
available. The speCifications for materials for & full—51ze mockup and
for the MSER will be issued during the third year. As both the mockup

and reactor'material'areireceived they will be irradiation tested.

Corrosion Program

Molten Salt. — Since the reference design of the MSBR primary coolant
circult incdrporates,the same,basic‘fuel salts and construction materials
as the MSRE, an extensive-corrosion program will not be required. The

large volume of data'generated during the development and operation of

.rthe MSRE is directly applicable to the MSBR and has demonstrated that an

acceptable system has been developed | _
Corrosion testing for the primary system will be mainly in the area

'of proof testing As compos1tional adjustments are made to the Hastelloy N

or to the graphite, their effects on oorrosion will have to. be determined

| :The active metals being proposed as a solution to the radiation-
'e,embrittlement problem may inorease the corr031on rate, although our pre-‘
. vious experience indicates that the effect will be small.: Alloys of
_modified comp081tion will be evaluated initially in thermal-oonvection
| loops. As The final composition 1s established more firmly, the corrosion
~ behavior Will be better establlshed by pump loops.

 
 

68 . ' = L | 7' '

Sufficient information is ‘available to indicate that corrosion prob-,- : ..
lems are not to be expected in the blanket salt system; however, because |
‘of fewer tests, such a conclusion.is not as well substantiated as that
for the primary system. It will be necessary to operate thermal-convection
and pump loops with the actual pr0posed salt composition~as'proof'tests ,
As with the primary system, any'major changes in Hastelloy N or graphite
will be checked with these salts. ' o

Another area of corrosion testing is anticipated in association w1th
the development of a low-melting coolant salt for the MSBR. Studies of .
both fluorcborate and stannous fluoride Systemsrare currently under way;_
for this application. Neither salt'system has been subjected to;evaluaef
tions in prior corrosion studies, and corrosion data will cbviously be
required in the overall assessment of their properties relative to the
Additional corrosion studies are being planned in support of goals
of longer range than the MSER. In particular, the improvements in purity ‘
and chemical stablility of flnoride systems have brought us to a stage

where it appears reasonable to consider the use of austenitic stainless

4

steels as salt-containment materials. The transition fromta nickel- to -
iron-base alloy offers important economic advantages in larger sized
molten-salt reactor plants, and there is a strong incentive for eXaminingi
the utility of iron-base systems in the coolant-salt region alone, L
Accordingly, we plan to investigate the behavior of stalnless steels in
the presence of the LiF-BeF, salt system both as a function of salt-

processing techniques and exposure temperatures,

Steam. Although Hastelloy N looks very attractive for use in the
steam circult we shall run some proof tests to demonstrate the compat-
ibility of Hastelloy N and supercritical steam Engineering experiments |
- are planned that involve steam and coolant salt separated by Hastelloy N Gb
These tests should yvield useful heat transfer data as well as prov1de

metal corrosion data

 

60Dunlap Scott Components and Systems Development for Molten-Salt | P
Breeder Reactors, ORNL-TM-1855 @une 1967). - o/

 

 
 

»

69

. Witriding

To mlnimize any exp1081ve hazard a nitrogen blanket is prOposed for

'use around the reactor vessel Preliminary studles have shown that nitrogen

- is dissociated by the nuclear env1ronment and msy react w1th the Hastelloy N.

The rate of reaction varies from ‘heat to heat of the alloy.-i

‘Specimens of Hastelloy N will be exposed to NH3, in the absence of
irradistion, and the effects on mechanical properties determined. :Samples
from various heats of msterial will be exposed to reveal the rate-
controlling element in the alloy. If any deleterious effects are found,

_additional samples will Dbe: exposed in in-reactor experiments - If nitriding

is indeed a problem, the cell atmosphere w1ll ‘be changed to another gas,

such as srgon.

Joining,@evelopment

| While a detalled examination of the joining problems for theiMSBR
‘cannot be msde until designs have progressed beyond the conceptual phase,
a tentative evaluation has been- conducted to reveal the ma jor problem
areas. The fabrication problems: fall into two categories: (1) the
original conStruction'snd»(2)fmsintenence of‘the system. The'letter-
will be by far the more: dlfficult of the two since, in most cases, it
must be maintained remotely. ' o R
Welding will be encountered in the reactor care, reactor vessel fab-

,-rication, heat exchangers for: the fuel and blenket systems, and in the

assembly of the piping system.s Fortunstely, the large nuMber of Joining

;ftasks may be combined 1nto s few general types, at- least for the initial

development Thls combining of pmoblems w1ll greatly reduce the effort

: required for the first 2 years The general prdblems are-?rf"
,Ll} Joining graphite to—itself ' '

2. joining graphite to Hsstelloy N, e
3;:fgeneral welding develoPment of Hastelloy N,

| 4. 'Joining Hastelloy N tubes to Hsstelloy N headers, r,ifr L o
5. }development of remote welding techniques for a. variety of Hastelloy N

JOlnts from 4. 1n. to 6 Tt in diameter,'

6, remote capplng or plugging of tubes.

 
 

 

The proposed programs for. solving the graphite Joining prdblems are’ o

included in the graphite section of this report The_programs for the
‘other problems will be presented in the following section,

. General Welding Development of Hastelloy N. — The welding of

Hastelloy N has been developed to the point that it was possible to con-

~ struct the MSRE using a wide variety of welded joints. . However, before

a MSER is eonstructed -additional welding development is desirable,  This

general welding development will be aimed primarily at a better under— \
_standing of the weld-cracklng problem which -has been encountered with
Hastelloy N but is also common to all high-nickel -alloys.

Weld-cracking problems have: sporadically occurred, in commercial _
heats of Hastelloy N manufactured to the same specification and bygthe
samé procedure. While we were able to work around this problem by using

a weld-cracking test to select the material, the cause was not established.
To avoid this problem, welding and metallurgical studies aimed at under-

standing the baeic reasons -for cracking will be conducted. . Previeus-,
indications that cracking'was-associated with segregation-of~alloying
elements need to be expanded to . a widersvariety ofecompoaitions. These
studies will rely heavily on the use of the electron microprobe to =
delineate the segregation. Welds will be made in material prepared-fbrj‘
the radiation damage program and also in 5pecial‘small'heats specifically
for welding studies. The various compositions will be welded at several
heat inpdts with restraint applied to the pieces. Heats showiag tenden-
cles to crack will be studied on the "Gleeble"61’which will be programmed
to magnify any effects resulting from heating.

As work progresses .on .the development of,a=radiationrresistant alloy,-

it will be necessary to conduct weldability studies on the likely candi-
dates.r Zirconium, one of the attractive alloying additions, is knowfi-to
be detrimental to welding. Attempts will therefore be made to;finda set
of welding parameters that will be adaptable to the zirdoniumecontaihing ‘

 

617rade name for a machine designed to simulate the ty?e of heating
obtained in the heat-affected zone of a weld. The specimen can be .
 fractured after heating to determine the resultant ductility

 
 

»

n

alloys. It will also be necessary to make welds that can be fabricated

into 1rradiation samples to demonstrate adequate radiation res1stance of

the weldments

- Currently, the filler metals used for welding Hastelloy N are of the '

same ‘composition as the base material. Results in the two previous

| programs will be examined forlindications-of ways to improve the welds
,hy changing the composition-pf'the filler'metals. Possible improvements
will be sought by.eliminating impurities'known to be harmful or by the

addition of other elements'to alter the form of the harmful impurity.
Preliminary tests have indicated that the addition of either niobium or
tungsten to the filler metal.will 1mprove weldability ThlS lead will
be pursued and the composition optimized

- Joining Development for Components.'— As the engineering components'

get larger and the reactor power increases, it w1ll be necessary to use

heav1er sections An’ construction. With such sections, there w1ll be an
econonic incentlve to use welding processes capable of high depos1tion
rates. All previous welding of Hastelloy N has been with the tungsten-
arc process which is reliable but slow. Processes of potential 1nterest
for the higher depositiOn rates are gas metal-arc, submerged-arc and
plasma welding. These processes by their very natures utilize high heat

inputs and, therefore, studies of their heat-affected zones will be

‘required Welding parameters will be established for any process which

in the prelimunary testing, looks promising These processes will also

be studied as to their suitability for use in remote welding
- Procedures for making tube to-header Joints for small tubes (O 50 in.)

m;were developed and used successfully for the MSRE heat exchanger These
1Joints were welded and then baek-brazed For the MSBR, a- large number |
'.of such Joints, many with much larger tubes, will be required The ma jor
 welding development effort in- this area therefore will be to adapt the
\'present techniques to. the new geometries and to automate them. While
'f manual welding is. possible, the large number of sueh welds make automation,

appear imperative | Because of the larger size components required and

 
 

o

limitetionS'in sizes of trazing or ennealing’furnaces, it msy be necessary
_tothendle the bundles in subsections whichiare‘subsequently welded
together. | .

To reduce the costs and to simplify the designs, the'elimination of
back-brazing will be explored. Conventional mechanical-expanding tech-
niques; such as rolling or plug drawing, will be investigated and tested
‘for reiiability. Use of high energy forming for expanding and bonding _
would appeer to be & promising technique. We have the equipcnent at ORNL
for these processes and will investigate them further.

-Remote Joining. -A.meaor progrem will be required to develop remote
welding and brazing techniques. It will be necessary to first develop
techniques for the specific geometry end then to gdapt them to remote
operation. The development of the positioning end guiding fixtures will
be a major‘part of the program. These phases of the program will be part
'of the maintenance equipment development 62 Several different types of
remote welds will be required and they will be sufflciently different to
require separate development o

. Although the welds differ in 'size and geometry, the programs for
their development will attempt to answer 51milar questions. The general

 

questions ares

1. Which welding technique will be most reliable?

2. What are the necessary welding parameters, such as voltage,
current, wire feed rate, and torch speed, and over vhat limits may they
vary? This will include sequences for start, Operation, and stop.

3. What is the recommended Joint design, and how much misalignment
of the pieces can be permitted? . o o

4, How pure an inert atmDSphere will be required both inside and

 

outside the pipe?
5. What are the effects of small amounts of fluoride contsminetion?,‘

 

62p, Blumberg, Maintenance Development for Molten-Salt Breeder
Reactors, ORNL-TM-1859 (June 1967) . _

 
 

 

 

Bhr

 

o3

6. Wnat is the permissible misalignment of the welding torch?

'7? How will the pipes and the torch be positioned?

The major problem with the Hastelloy N- to-Hastelloy‘N butt JOints
in tubes will be the limited space around each pipe; this is-approx1mately
1 in. Present plans are to develop a s1ngle procedure for the initial
assembly and for remote repairs Several_weld joint designs appear
possible and will be Investigated. .A-concurrent'effort_will be conducted
on remote brazing for joining these tubes. This techniqne, which was
developed for the MSRE, makes use of a conical- -type joint in which one
member must either twist or slide as the braze metal melts .8

‘A single development effort should suffice for all -the large
(approx 12-in.-diam) remote pipe joints. Many of'these Joints are
containment members, so a high degree of reliability will be reqnired

This makes remote welding much more attractive than mechanical-goining=

. techniques,- While these pipes are of large diameter,-they_dolnot have
heavy walls, 'However,:the'additiOn\of fillerumetalrwillibe necessary.

After completion of exploratory studies, a selection of the applicable
welding process will be made. . Welds similar to these were developed by
Westinghouse for the Pennsylvania Advanced Reactor,64 and their procedures
will serve as a starting point A complication with these welds will be
the tight 1imit;onrpipe,lengths required to minimize fuel inventory. The
positioning and aligning-present difficulties.since‘short;;large- -

diameter pipes are very rigid,— An'insert may be.requiredato gnide,the

g _pipes together.

- If one tube of a heat exchanger develOps & 1eak, it w1ll have to

. be plugged or the whole unit will have to be replaced.; It Wlll be neces-
lsary to plug the ends of the failed tube through a small access port in

. each.end of.the exchanger Such plugging procedures are standard on -

 

63, C. Hise, F. W. caské; and R G. Donnelly, "Remote Fabrication

 of Brazed Structural Joints in Radioactive Piping," Paper 63-WA-53 of
 the Winter Annual Meeting, Philadelphia, Pa., November 17—22 1963
~ of. the American Society of ‘Mechanical Engineers SR

 64F. Y, Seidler, Pennsylvania Advanced Reactor —-Reference Design Two,
Layout and Maintenance, Part I, WCAP-1104, Vol. 4 {March 1959]),

 

 
 

 

%

commercial heat exchangers, but are not for remote applicationsrl'The
positioning equipment for such & job would be similar to that used to
cut small pieces from the HRT core tank. ¢ | : L

| For such an application, a plug must be developed which can be
inserted‘into the tube and then fused to the header. A major problem
will be-that'snch a weld is by nature highly restrained and is, therefdre,
subject to cracking. This may require trepanning the head before the
plug is inserted and welded. | | - -

Inspection Development

While a complete evaluation of testing and inSpection problems cannot
be made until a firm design is- available, some are apparent from the con-'r_

ceptual design. The ones requiring most attention are those that will have

to be made remotely. Inspection development for the Hastelloy N tubing.
and pipe will consist of adapting available techniques to the necessary
configurations and to the required sensitivity levels. The techniqpesu
developed for inspecting the MSRE heat exchanger should also be adaptable
" to the MSER heat exchangers. ‘Techniques to be used would include
penetrants, radiography, and ultrasonics. A

Remote inspection will be required for (1) Hastelloy N tube Joints
at the fuel header, (2) large butt-weld pipe joints to disconnect “the
main piping from both the reactor and the heat exchangers, and (3) plugs

in the heat exchanger tubes. ‘Until develo@ment of the joining techniques

has progressed further, it is impractical to speculate on inspection |
. procedures. Several testing techniques will be evaluated

In developing the nondestructive testing techniques, three phases
‘must be completed (not necessarily in sequence). They are (1) demonstra-
tion of feasibility, (2) determination of test sensitivity and establish-

ment of reference standards, and (3) development of detailed techniques
and eqpipment

 

65p, P. Holz, Description of Manipulator System,'Heliarc Underwater
Cutting Torch, and Procedure for Cutting the HRE-2 Core, 0RNL-TM—175
-~ {(Nov. 5, 1962). R _

_,"Zfl,

ks

'
'

L
PN
 

 

O

75.

For the remotely inspected joints, each of the steps must first be

_ performed in'a'coldrlaborétory (but with cognizance of the need to

progress to remote operation) and then the necessary mechanlcal devices
must be,developed for the remote,performance. It may be possible to
devise accessories which can~beiattached»to the manlpulators used to

fabricate the joints.

Materials Development for Chemical Processing Equipment

 

The fuel salt will have to be continually reprocessed to remove

fission products. The present concept of this processing calls for dis--

tilling the salt at a temperature of about 1800°F. The strength require-

ments are quite modeet,;but-the'material will be exposed to salt on one

side and to some gaseous atmosphere on the other side. The easiest gas-

eous atmosphere to obtain is the cell environment which is N, + ~2% Oy .

This would require that the material be fairly resistant to oxidation and

nitriding.--Theioxidation resistance is improved by increasing the chro-

mium content of the alloy, bfit:this.in turn inereases the corrosion rate

on the salt-side; We feel that several possibilities exist for materials

for constructing this vessel. - - |

1. Hastelloy N with an inert cover gas,

2. a molybdenum-base alloyflwith an inert cover gas, |

3. a duplex system with arnickel-molybdenum alloy, exposed to the salt
and an oxldatlon-resistant materlal such as Haynes alloy No. 25

_exposed to the: cell env1ronment

-'4.! a vessel made of an oxidatlon-resistant alloy with a graphlte liner.

- These possibllities and. others will have to be explored by screening

| ftests to determlne re51stance to. ox1dat10n, nitriding, and salt corrosion.
”_The ease of fabricability and cost of the candidate materials will also -
',be considered in making a flnal choice.

‘5,1;Gféfifiite_Program;'«f

The graphlte that we need for this reactor requires that some advances
be made in present technology Therefore, our first concern is to procure

material that meets our75pe01f1catlons for study. The ma jor 1tems to be

 
 

 

76

evaluated for this material are (1) determining its behavior at radiation
dose levels as high as possible, (2) joining of graphite-to-itself_and'to
structural materials, (3) fabrication development to yield a low gas
permeability, (4) evaluation and charaCterization of the~m0dified'or'
improved graphite, (5) determining its compatlbllity when used with
Hastelloy'N in a system circulating fused salts at elevated temperatures,
and (6) fabrication of the graphite into engineering systems and evaluating -
its performance. Because of its greater potential for reducing radiation
damage effects, the major effort will be on the development of isotropic
graphite, However, this material is so new and so little is known about
it, development of anisotropic graphite such as the needle-coke'types

will also be continued. The proposed development program is- discussed

in the f01lowing sections. - '

Graphite Fabrication and Evaluation

- The grade CGB graphite used in the MSRE was produced in experimental i

equipment and its properties still leave much to be desired for a MSBR. -
No graphite from any source is now available that w1ll meet all of the e
MSBR requirements, Since there is no present industriel need for low= .
permeability graphite, we are not likely to obtain much allied develop-
ment help from industry. An adequate supply of the highest'quality.
graphite will be purchased to supply material for use in test rigs, to
establish reasonable specifications, and for evaluation. Continuing
orders will be placed as improvements, or potential improvements, are
made in the quality. The intent is to progress in order size from the
laboratory to the pilot plant and then to a.production;size order. Until
msteriel is actually produced in production equipment and is teSted' -
dodbt will exist as to how representative the smaller. batches really are,
It is expected that grade CGB or a similar grsphite will be used in the
first tests.
The various problems to be faced will include:
1. As soon as p0381ble, consult with msnufacturers end.prepare a
specification for small lots (1aboratory quantltles) of graphite of as - e
near MSBR quelity as it is pOSS1ble to obtaln immedietely | ' ' kaj

LG

 

il
 

7

R. Place small orders for both 1sotropic and anisotrqpic grades of
graphite in the form of block and pipe.-f o -

3.  Measure pertinent physical and.mechanlcal properties of
representative graphites. - - -
_ 4, In about 1 year, prepare a Specification for a pllOt-Slze batch
- of the most prom1S1ng type, or types, of graphite | It 1s not expected _
that any significant new radiation-damage 1nformation w1ll be available
at this time. e _ | L

5. If suitable graphite cannot be purchased from out31de sources,

an in-house pilot-production facility will have to be established

. Irradiation Behavior

In laylng out any graphite development program, an enigma rapidly
‘becomes apparent One of the major unknowns is the effect of massive |
neutron doses. To Obtain the desired doses requires irradiation periods
of two to three years, and we,would like to firmly establish in three
years that graphite will have an acceptable 1ife in the MSBR._ The problem
therefore 1is that, if during the development period wmajor changes are made
in the graphite, the graphite being irradiated will no longer be repre-
sentative and the results will therefore be suspect. The most satisfactory.
solution appears to be starting irradiation tests as soon as possible with
samples fabricated from the highest quality material availahle and to fol-
low these with other tests of improved material The purpose of the work
would be to determine first the dimenszonal stability of the graphite for

__.the temperatures and high-radiation doses that would have to be sustained
~in the MSBR plus the effects of irradiation on the. graphite properties,
isuch as creep ductility, mechanical.pmoperties, accessible voids, and

"'-permeability. - The. inclusion of ‘both anisotropic and isotropic graphite
is warranted since the anisotropic has more technological development
~behind it, but the isotrOpic graphite which has been under development for

-approximately three years shows more potential These experiments should

,-Ebe performed on graphlte, graphite~to-graph1te Joints, and graphite-to-

' Hastelloy N joints. | o

 
 

 

78

To demonstrate the ability of graphite to sucCessfully ret&in_its |
integrity after exposure to dose levels of 1 >< 1023 neutrons/cm? (5-year
life), we plan'fo irradiate gfephite to as near this dose as possible.
This will require the irradiations to be performed in reactors that have

fast fluxes of 1015 neutrons cm=? sec™l or greater to obtain the data in

a reasonable time. At present,-there are only tworfast reactors, EBR-1I
and Dbunfeay, and one thermal reactor, HFIR, that approach the'fast-flux

requirements.' Approximate values of the time required to reach a dose

‘of 1 x 1023 neutrons/cm , cost per year, and facility size are listed in,

 

 

Table 11.
Table 11. Comparison of Reactors
EBR-TT Dounreay I-IFIR -
Iflme to 1l X 1023 neutrons/cm year 6 4B 'i 2 IR
Cost per year, $ . ~ 20,000 300, 000 15,000
Facility size, in.  0.75-diem 0,75-diam  0.50-diem
TEWOEEER S s X 20

 

a2’70-operating power days per year

It is, therefore, proposed to use HFIR to demonstrate the ebility 7
of the graphite to retain its integrity after an exposure of -
1 x 102> neutrons/cm®. Due to the size limitation of HFIR, the irradi-
ation will be restricted to simply prepost type of-testing. To |
demonstrate the ability of the MSBR graphite to absorb creep"Strein,
restrained growth type experiments will be performed in EBR-II. By
enforcing & creep rate equal to the growth rate, the creep strain
aceumulation.will be forced into the material ten times faster than.
under MSBR conditions. Although en equivalent neutron exposure will
not be’achieved,'the enforced strain will be-greater than expected in
the MSER, - I

. O

 
 

 

 

 

79

Graphite Joining

Since,the graphite-to-Hastelloy N joint is a_very important part of
the system, we will carry}parallel efforts on at least two different types
of joint;-fBraze'joints.ofnseveral designs will be evaluated using the
35%‘N1—60%'Pd%5% Cr allcy\and'pure copper as brazing,materials.; These. -
joints will be designed so that graphite is initially in compression
where it is strongest and has the ability to undergo small smounts of
plastic strain. This will.minimiZe the'tensile stress that will develop
‘in the graphite as it shrinks due to irradiation.. A mechanical type of
joint will also be evaluated as a second preference. “These joints will
be made u31ng the 5-1n.i0D'Xel/2-in,-wall.graphite pipes that are,presently
in the MSBR design. These joints will be subjected to thermal cycling, -
evaluated'for'corrosion'resistafice'to the'fuel-and'blanket,salts, and
irradiated to determine integrity under service conditions, - o

'The’graphite;to-graphite joint'will be studied in'detail | Based on
this study the choice will be made between a graphitized Jjoint and a |
brazed joint. : | o

Permeability Studies
| As'"imprcred“'grades”dfigraphite are received from producers, it
will be necessary to determine their permeability for molten fluorides,
helium,:and'fissianegases.z'AS'was pointed out in the‘previcUS'diScussion,
| obtaining graphite with the very 1OW'permeabillty of 1 x 10~7 '2/sec
“will be very difficult ‘When the data on xenon stripping and’ the designs
" of the MSBR become firmer, realietic Values for gas permeability will
 have to be established This program will include: s
1. »studies of permeability to gases and.mclten salt of"- the variousfl o
ie-grades of graphite | "I,"” o ' EREE R
'2a-iinvest1gation of the prcperties of graphite which affect the perme-
- rability and how it may be minimized "_' o
33.; measurement of the effects of various coatings on. the permeahility
_ _ _,of gra,phlte, I e o .fi R S
4., determination of the permeability of graphite joxnts
5. determinationrof the effects of fluorides and solid flssicn_products

on the graphite.

 
 

 

80

Corrosion and Compatibility of Graphite

Some additional data will be obtalned on the compatibility of graphite
with Hastelloy N in systems c1rculat1ng high-temperature fused salts. The
testing must also include the brazed 301nts. Initial tests will be simple
capsule tests containing both materials. Then small—scale grephlte and
graphite-Hastelloy N thermal-convection loops will be fabrlcated and
operated with fused salts to determlne the compatibllity and corrosion '
resistance of these materials. The final stage will,be_testing in—the .
engineering loops to evaluate-full-scale.conponents.'.Theseflarge'loops :
are e part of the component developnent progranm, 'The functionrof'this

. task will be to thoroughly examine the components after exposure in the

loops. | | o S
Date on effects of 1rradiation on corrosion and compatibllity Will

be obtained from in-reactor tests in the Chemical Research and Development
Program.®® Various types of graphite are being ineluded in the Surveil-
lance Program with subsequent evaluation of fission—product retention

and changes in mechanical and physical properties.

Graphite InsPection

The necessary nondestructive testing program for the graphite tubes
proposed to be used for MSBR would cover severel test methods It is
7anticipated.that,eddy-current techniqnes would be used for dimensional
kgaging.and both lowsvoltage.rsdiography.and eddy-current“techniques would
be used to detect discontinuities, If laminations are present and unde-
sirable, it may be necessary to use nltresonics or infrared techniqnes;
to detect them, For geometries more complex than concentrlc tubes, the
seme methods would likely be pursued. However, there would be more
requirement for dimensional gaging and the flafiafinding techniques WOuld |
be more difficult to develop. The development would proceediin_the_;;

following steps (not necessarily in sequence):

 

é6W. R. Grimes, Chemical Research and Development for Molten-Salt
Breeder Reactors, ORNL-TM-1853 (June 1967).

 

 
 

L ad

8

1, Vdetermination of feasibility-of application of the test method to

~ the. conflguratlon and grade of graphite,

2. _determlnatlon of characterlstlc flaws and relative test sen31t1v1ty

- attalnable

- 3, establishment of reference,standards,_
b, development of detailed'techniques_(perhaps including equipment).

General Development and Proaect A851stance

- Many details of the f1nal design will depend upon the properties of

the two primary structural materials -~ graphlte and modified Hastelloy N.
. As we progress with the development of these materials, we will keep the

designers informed, so that this information can be 1ncorporated in the

_de51gn. Several of the planned engineering tests‘w1ll,also require
| assistanCe. We shall be called upon for design assistance and fabrica-
ftion technology. The evaluation of many of the test results will be made

by'materials"personnel. ‘After working with the vendors to obtain the
desired products and With.thepdesigners to finalize the MSBR design, we
will take an active part in writing thedSpecifications, procuring the
materials, and assisting;in}the'actual oonstruction.

‘Several parts of the proposed molten-salt systems could probably
be made'of'cheaper materials;~such as the-austenitic stainless steels.
We do not have adequate corrosion data on these materials, but tests

will be initiated to obtaln this informatlon on a second prlorlty basis

e'We w1ll crltlcally assess the properties of these materials for thiS'

o program and 1ntroduce them where they appear advantageous.

- f hnmomsmrmNTs |

The authors are indebted to G M .Adamson, Jr,, and A Taboada for_

'r*assembllng the flrst draft of thls report, 0RNL-CF-66-7-42 Several

' persons contributed informatlon for the various sections-' J H. DeVan,

G. M. Slaughter, R. G Donnelly, D A Canonico, E A. Franco~Ferre1ra

'R, W. McClung; W. H. COOk andC R Kennedy

 
 

 

- APPENDIX

2

 

 

 

 

 

 

 
 

'
'
'
¢

 

 

 

 
 

 

~ Cost Estimate for Materials Development for the MSBR"

 

 

Tasks and Type Costs FY-6¢  FY-69  FY-70  FY-71L  FY-72  FY-73  FY-74
| Manpowver Costf' |
Hastelloy N i | o L | B
‘Irradlatlon testlng " j ' . 250 250 200 200 100 100

<8

Joining . e 120 120 1200 120 90 60
Corrosion and compatibllity o 120 120 . 120 90 60 30
Nitridlng o | - 20 L RN o
Inspectlon develqpment -  . 60 60 60 - - 30 30 ,

subtotal _‘f“j) fi3‘¢‘"f" .;' 570 550 600 440 280 190
Graphite  _ _ "A:f* "_;;j f - | L S
Irradiation testing R 100 100 100 1100 . 35 35
Joining S :,; o 75 75 60 45 30 30
Permeability oo 60 60 - 30 15 15
Procurement and characterization 120 120 90 60 30

System development ‘ - 60 a0 60 60 30

. Corrosion and compatibility 30 30 30 30 S
‘Inspection develqpment | - ‘ 45 60 60 30 30
 Subtotal 490 535 430 340 170 65

General o 'f' R o ) | | | | |
MEtallurglcal serv1ce e 60 90 90 90 30
Outside service . = 30 60 - 60 30 | |
Supervision, secretary, ete. ~60. 60 60 60 30 30 30
Project assistance =~ 60 90 9% 90 - 60 60 60
Subtotal 210 300 300 270 120 90 - 90

Total 1270 1385 1330 1050 570 345 90
 

Cost Estimate (continued)

 

| Tasks and Type Costs - FY-68  FY-69 FY-70 FY-71  FY-72 - FY-73  FY-74

 

Unusual Cost

'Hastelloy N

 

 

 

Reactor . | 50 50 50 40 40 |
Hot cells 60 60 60 60 60 - 30
Experiment and specimen f&brication 110 110 140 100 s - o
Materials 85 100 100 50 30 - | ‘
Analytical chemistry _' | 40 30 30 30 10 5 :
Subtotal 345 350 380 280 190 35
Graphite o o B S B -
Reactor o 100 100 50 100 100 100 &
Hot cells - ‘ ' 10 20 20 20 10 10 |
Experiment and specimen fabrication 60 90 100 100 40 :
Materials 80 120 100 10 10 10
_ Analytical chemistry 20 20 20 10 5
Subcontracts 15 35 25 o - |
 Subtotal . 285 405 315 240 165 120
Equipment | . L 100 50 50 - 50 10
. Totel 730 805 45 570 365 155
| Surmary of Costs ) | |
Manpower total 1270 1385 = 1330 . 1050 570 345 90
| | Uhusua1 £ota1 o o | | ,"730_ - 805 ‘745 i'f" 57Q . 365 155 :"  ‘
| Grend total 2000 2190 2075 . 1620 935 500 90

gAll‘costs in thousands of dollars.

 
e g

oy S
ot 4

 

 

k.

1-2.

413,

14,
15.

16.
17.

18.
19.
20,
21.

22.
23,
24 .

25,

26.

27.
28.
29,

30,

31.
32.
33.
34,
35.

36.

37.
38.
39,
40,
41.
42.
43,
- 44,

46,
47 ..

- 48,
49,

50,

51.

- 52,

53.
54.
55.

>mpmuwgwaMMmzmupmwggpomwgwwwgmpzmaowflwnw

 

- INTERNAL DISTRIBUTION

Central Research Library -

ORNL — Y-12 Technical Library
- Document Reference Section

Laboratory Records: Department
Ieboratory Records, ORNL R C
ORNL Patent Office

Adams

Adamson , Jr.

.—Affel

Alexander

Apple ‘

Baes.

Baker

Ball =

Barthold

Bauman '

Beall

‘Bender -

'S, Bettis -

F. Blankenship

-E. Blanco

0. Blomeke

Blumberg

G. Bohlmann

J. Borkowski

E. Boyd

. Braunstein

A. Bredig '
B, Briggs - oo
R. Bronstein = -l
D. Brunton

A. Canonico -

» - - - - *
. - - -

mmpqzfiwpmgfi

-

T .

. Cantor

. L. Carter

. I. Cathers

. M, Chandler

. L. Compere

. H. Cock -

. T. Corbin

. L. Crovley

. L. Culler
. E. Cunningham -~
. M Dale
, G. Davis

. J. Ditto

« S. Dworkin

69.
70.

56.
57.

58.

59.
60.
61.

62.

63.
64.

- 65.
66.
67‘.

68.

71,

72.

73,
 Yler
75.

76.

7719,

80,

- 81.

82.
83.

84,
85,
86,

- 87.

89,
.90,
91.
92,
© 93,
94
. 95,

96.

97.
98.
99.

- 100.

101.
102.
103,

spawmznwwmnuwwwb;mmoym?pdhuf

Q”“”F?”P““?mfififigs?fi

ORNIL~TM-1854

.- R. Engel

. P. Epler

.. E. Ferguson
M. Ferris
P. Fraas

. A, Friedman
H Frye, Jr.

. H. Gabbard

. B, Gallsher

. BE. Goeller

R. Grimes

. G. Grindell

. H. Guymon

. A. Hannaford

. H, Harley

.. G. Harman

. S. Harrill

.- N. Haubenreich

. A, Heddleson

. G. Herndon

. R, Hightower

. R, Hi1l1

. ‘W. Hoffman

. W. Horton

. L. Hudson
- Inouye

. H., Jordan
"R. Kasten
J. Kedl
T. Kelley

cJ. Kelly

. R. Kennedy

. W. Kerlin

. T. Kerr

. S. Kirslis

;- T. Krakoviak

. W. Krewson

. BE. Lamb
-A. Iane.

. B. Lindauer
P. Litman -
-I. Lundin

. N. Iyon

. G. MacPherson

. E. MacPherson

. D. Martin

 
 

 

&8 - |

104. C. E. Mathews . 184. - C. E. Sessions - -
105. €. L. Matthews 185. J. H. Shaffer |
~ 106. R. W. McClung o | -186. G. M. Slaughter
107-111. H. E. MeCoy, Jr. o - 187. -A. N, Smith
112. H. F. McDuffie o ' '188. F. J. Smith
113, C. K. McGlothlan - 189. G. P. Smith
114. C. J. McHargue =~ = - 190, O. L. Smith
115. L. E, McNeese - . 191. P. G. Smith
- 116. A. S. Meyer ~  192. W. F. Spencer
117. R. L. Moore . .. 193. I. Spiewek
- 118, J. P, Nichols . . - 194, R. C. Steffy
- 119. E. L. Nicholson IR '195. H. H. Stone
~ 120, L. C. Oskes o o 196. J. R. Tallackson
- 121. P. Patrierca , © 197, ‘E. H. Taylor
122. A. M. Perry o - 198. R, E. Thoma
- 123. H. B. Piper I 199. J. S. Watson
124. B. E. Prince - ... . 200. €. F. Weaver
125. J. L. Redford . .- | 201. B. H. Webster -
126, M. Richardson o o 202. A. M. VWeinberg
127. R. C. Robinson ~ o 203—207 J. R, Weir, Jr.
128. H. C. Roller ' | 208. W. J. Werner
'129-178. ‘M. W. Rosenthal 209, K. W. West
179. H. C. Savage ‘ ' | - 210, M. E. Whatley :
180, W. F. Schaffer 211, J. C. White | T
181, C. E. Schilling - 212. L. V, Wilson ' : -
182. Dunlap Scott. - ) 213, G. Young - .
‘H. C. Young '

_183 H. E. Seagren B 214,
EXTERNAL DISTRIBUTION

215-216. D. F. Cope, RDT, SSR, AEC, Osk Ridge National Leboratory

 

217. A. Giambusso, AEC, Washington

218. J. L. Gregg, Bard Hall, Cornell University

219. W. J. larkin, AEC, Oak Ridge Operations
220-234. T. W. McIntosh, AEC, Washington -

235. R. M. Roth, AEC Oak Ridge Operations
236—237. M. Shaw, AEC Washington -

238. J. M Simmons, AEC, . Washington

239. W

. L. Smalley, AEC, Oak Ridge Operations
- 240. D. K. Stevens, AEC, Washington |
- 241, R. F. Sweek, AEC, Washingbon '
242-256. Division of Technical Information Extension

. O