ORNL-TM-8298.txt

ORNL /TM-8298
Dist. Category UC-95

    

METALS AND CERAMICS DIVISION ORNL/TM-~-8298

DE83 (04228

THERMAL~-CONVECTION-LOOP STUDY OF THE CORROSION OF Fe-Ni-Cr
ALLOYS BY MOLTEN NaNO3-KNOj

P. F. Tortorelli
J. H. DeVan

Date Published - December 1982

NOTICE: This document contains information of 1
preliminary nature. It is subject to
revision or correction and therefore does
not represent a final report.

/'; Prepared by the
: OAK RIDGE NATIONAL LABORATORY
. Oak Pidge, Tennessee 37830
\ operated by
UNION CARBIDE CORPORATION
for the

Office of Industrial Programs
U.S. DEPARTMENT OF ENERGY
Under Contract No. W-7405-eng-26
o f
o

T Shieiit e

D iy N i -
Ry TEE .

2
o)
-

 
 

ABSTRACT .+« ¢« & o o « &
INTRODUCTION . ¢ ¢ o &
EXPERIMENTAL PROCEDURES
RESULTS « &« ¢ & & & « »
DISCUSSION . « « « « &
SUMMARY . « « ¢ ¢ & « &
ACKNOWLEDGMENTS . . . .
REFERENCES . . . . . .
APPENDIX .« ¢ & o & « &

CONTENTS

N =

25
26
26
29
THERMAL CONVECTIOK LOOP STUDY OF THE CORROSION OF Fe-Ni—Cr
ALLOYS BY MOLTEN NaNO3-KNO3*

 

P. F. Tortorelli and J. H. DeVan
ABSTRACT

We studied the corrosion of Fe-Ni—Cr alloys by draw salt
(60 wt % NaNO;—40 wt % KNO3) with thermal convection loops of
alloy 800 and types 304L and 316 stainless steel. The main
corrosion processes at 600°C and below were the growth of thin
oxide scales and the dissolution of chromium by the salt.
Spallation of oxide layers occurred on type 304 stainless steel
specimens at intermediate temperatures. Results Iindicated rela-
tively low corrosion rates (<13 ym/year in most cases) for tem-—
peratures of 600°C and less. Corrosion of type 316 stainless
steel was greatly accelerated when the maximum loop temperature
was raised to 620°C. Tt therefore appears that 600°C may be the
limiting temperature for use of the above alloys In draw salt.

 

INTRODUCTION

A mixture of 60 wt I NaNO3—40 wt Z KNO3 (draw salt) has been
proposed! for use in solar thermal power systems as both a solar receiver
working fluid and a thernal energy storage medium. Draw salt is attrac-
tive for such applications because of its high sensible heat capacity, its
low reactivity in the event of a leak to air of steam, and the low
operating pressures required for its use. However, the feasibility of
such a system depends partly on the compatibility of the salt with can-
didate structural alloys. The subject experiments were therefore per-
formed in closed-loop systems to study the corrosion of alloy 800 and
types 304 and 316 stainless steel by slowly flowing NaNO3;-KNO; under an
impressed temperature gradient. Previous work? with a nitrate-nitrite
[49 NaNO;-44 KXO3-7 NaNO, (wt 2)] salt revealed unacceptably high corrosion
rates of type 316 stainless steel at 550°C. The study presents comparable

 

data for a pure nitrate salt system.

  
   
 

 

*Research performed under subcontract with Sandila National
Laboratory, Livermore, California, per purchase agieement 92-8568, spon-
sored by the U.S. Department of Energy, Division of Energy Storage Systems.

R :.."‘1,_ Pl 5'?9‘?:_':?75'_1“‘«53;

  
 

 The tests reported here are companion experiments to those conducted
at Sandia National Laboratory.3® The Sandia experiments were matched
to our experiments in terms of loop materlals, specimens, and temperature
conditions. However, the Sandia experiments operated under air, while
ours were sealed, the cover gas being composed of salt decomposition

products.

EXPERIMENTAL PROCEDURES

The as-received draw salt, with a nominal composition of 60 wt %
NaNO3—40 wt Z KNO3* and a melting point of 221°C, was outpassed in a
closed type 304 stainless steel vessel ar 250 to 300°C while simul-
taneously being purged by argon flowing above and through the salt. A
significant amount of water vapor was removed in this way. Subsequently,
this pot of salt was used to fill three thermal convection loops (TCLs) of
the type shown schematically in Fig. 1. Natural circulation of the salt
was induced 1in these closed loops by controlling the loop's temperature
profile to produce convective flow. Each of the three loops was coan-
sfructed of different alloys: types 304L and 316 stainless steel and
alloy 800 (Table 1). The loops were designed to permit the inscrtion and
removal of corrosion coupons and salt samples with minimal disruption of
the salt flow. In this way, changes in salt composition, specimen
weight, and microstructure were monitored as functions of exposure time.
Tubing for the loops ranged from 1.9 to 2.5 cm in outside diameter and
from 0.2 to 0.3 cm in wall thickness. 1Insert coupons were approximately
1.9 x 0.8 X 0,1 cm and were in the as—annealed condition (unpolished). 1In
any given loop, the insert specimens and tubing were of the same nominal
compositions, although, in the case of the type 304L stainless steel 1loop,
specimens of bot: normai- and low-carbon grades were exposed. In each
TCL, the pressure above the salt rapidly increased to above 6.9 X 10% Pa,

gage (10 psig) when the temperature was initially raised above about

 

*Partherm 430, Park Chemical Co.

 
 

ORan. Dw. 58 0N T

‘o™ TWIK BALL '
T 7 vaves T

     
   
      
     
     

H1
N2
4
H3
H4
HS - CLAMSHELL
ssem |7 EATERS 5
— e INSULATION
CORROSION C6
SPECINENS —
HI
¢
Hg ! FLOW
DIRECTION || g
‘?\:\_;/
o \\f
tm P
oo
FREEZE VALVE

| Dump
TANK

Fig. 1. Thermal convection loop used in NaNO3-KNO; studier. The "H"
and "C" numbers denote specimen positions iIn the hot and cold legs,
respectively. Coupons Hl and Cl are located in the cover gas and are not

wet by the salt.

520°C. Analysis of the evolved gas showed that it was principally cxygen.
Subsequently, care was taken to maintain this gas composition over the
loop salt by using only oxygen as a cover gas duriny specimen manipulation
and salt sampling. During most of our study the three TCLs operated with

maximum temperatures and temperature differentials ATs of about 600 and

o "J=-|_.“'"u~, l‘.= - NI A’.‘.‘j,'*‘_ I t h

 
Table 1. Nominal corpnsitions of tested alloys

 

Concentration (wt %)

 

 

Alloy
Fe Ni Cr Mn Mo Si Al Ti ca

Alloy 800 45 32 21 1 0.2 0.5 0.5 0.10
Type 304 68 10 19 2 1 0.08

stainless

steelb
Type 316 67 11 17 2 2 1 0.08

stainless

steel

 

IMaximum cncentrations.

Z"'I‘ype 304L stainless steel 1s of similar composition except for a
maximum carbon concentration of 0.03.

225°C, respectively. The AT was established by first setting the minimum
temperature at the bottom of the cold leg and then increasing the output
of the heaters along the hot leg to achleve a maximum temperature of 600°C
just below the bottom of the hot-leg surge tank (Fig. 1). Typlcal
temperature profiles for each of the three TCLs are shown in Fig. 2. The
nominal salt velocity was 4 mm/s. After 5300 h of coupon exposure, the

AT of the type 316 stainless steel TCL was changed to 150°C, and its
maximum temperature was maintained at 600°C to assess any changes in
behavior due to mass transfer effects (see below). This temperature pro-
file is also shown in Fig. 2.

Before weighing, the loop specimens were cleaned with water after each
exposure to flowing salt. After 1500 and 4500 h of exposure, corners were
clipped from the specimens in each loop for metallographic examination.
Periodically, samples of salt from the respective loops were taken and
analyzed for Fe, Ni, Cr, Mo, NO,, and NO3. Total chromium and chromium(VI)
concentrations in the salt were measured spectrophotometrically,®:’ and
nickel and molybdenum levels were determined by the atomic absorption
spectroscopic graphite furnace technique. The salt samples were analyzed
for iron spectrophotometrically with the o-phenanthroline method.8 The
nitrate concentrations were determined by a modified DeVarda method;? the

nitrite levels were measured volumetrically by titration with cerium(IV)

(ref. 10).

 
ORNL-DWG 84-20917

 

  
 

  
    
   
 

 

 

 

650 T I 1 T T | 1

600 |- 7

550 }— ]
g
W 500 |- 7
= ~°
E o \\
W 450 F>e. . -
= // TYPE 346 LOOP, AT=250°C
- /// ——--—= TYPE 316 LOOP, AT=450°C

400 ' —— TYPE 304 LOOP \

ALLOY 800
350 |—
300 ] | 1 | | 1
0 5 10 15 20 25 30 35 40

DISTANCE FROM BOTTOM OF COLD LEG IN DIRECTION OF FLOW (cm)

Fig. 2. Typical temperature profiles for the three draw salt thermal
convection loops; AT, temperature differential.

RESULTS

As previously discussed, the weights of the loop specimens were
measured as a function of exposure time. Such data are shown in Fig. 3
>for the specimens at the two hottest coupon positions (H3 and H4) in the
alloy 800 TCL. Very little weight change occurred. Similar behavior was
found for specimens in these positions in the type 304L stainless steel
loop, which contained both types 304 and 304L stainless steel at the 600°C
(H3) position (Fig. 4). Equivalent weight change data for the type 316
stainless steel loop are shown in Fig. 5. As with the other two loops,
the weight changes of the H3 (595°C) specimen were small. Fowever, the
weight of the H4 (565°C) specimen steadily decreased during the initial
3000 h of exposure to the salt, after which it became approximately

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6
ORNL- DWG 81-2092% 5
4
T T I I 1 ]
@
e
2 |- * o o
— ® 0
~ o
£ . o
2 ° 0 0
W © o o©
Z o :
d "
T
-2 b ® 600°C - S
o O 570°C - 4
w
= g
-4 | ] ‘
" z | | | 1 1 |
0 1000 2000 3000 4000 5000 6000 TO00 8000 _‘_‘7
EXPOSURE TIME (h) A
Fig. 3. Weight change versus exposure time for alloy 800 exposed to fé(
flowing NaNO3-KNO;. | &
ORNL-DWG B4-20924
4 1 1 T T T I
2+ 4 & — "
<
E & * . S o
& ° *
0 e
5 2 o
- -2 [ — Bt
5 e TYPE 304L, 600°C
W ° o TYPE 304, 600°C
4 & TYPE 304L, 570°C _
8
’ «
-6 i j ] ] i i 1
0 1000 2000 3000 4000 5000 6000 TO00 8000

EXPOSURE TIME (h}

Fig. 4. Weight change versus exposure time for types 304 and 304L
stainless steel exposed to flowing NaNO3~KNO,.

 
ORNL-DWG 81-20518

 

 

 

 

 

 

2 1 1 1 ] T 1 I
AT CHANGED Twuaxs CHANGED TO
NP T0 150C—— l—szox:, AT=150°C
O
Q o
o o °© o .
— 1—2!—-— g
€
o ° e 565-590°C
®
w -4 0 595-620°C ]
Z
o §
5
_6,__ e
e . o
D
l:l Y ®
2 *8[—- ° ° —
® [ ]
-"O'— —y
O
42 b

 

 

 

0 {000 2000 3000 4000 5000 6000 7000 8000 9000 40,000
EXPOSURE TIME (h)

Fig. 5. Welght change versus exposure time for type 316 stainless
steel exposed to flowing NaNO3-KNOj3.

constant. Changing the AT of the type 316 stainless steel loop from 225
to 150°C after 5000 h of coupon exposure had little effect on the welght
changes, but raising the maximum loop temperature to 620°C after 6000 h
resulted in the onset of significant weight-loss rates.

As noted, the welght change data in Figs. 3 through 5 are for only
two of the coupon positions in each of the TCLs. The data for all loop
coupons are listed in the Appendix in Tables A.1 through A.3 (the location
of each specimen is shown in Fig. 1). With the exception of the type 316
stainless steel H4 coupon, the weight changes of all the specimens in the
alloy 800 and type 316 stainless steel loops were small. However,

significant weight gains were measured on certain coupons in the type 304L

 

 
stainless steel TCL. The weight gains were particularly high for several
ccupons in the lower half of the hot leg. These welight changes are
plotted as a function of exposure tire in Fig. 6. The decreases in weight
et longer times appeared to be related to spallation.

Metallographic specimens were obtained by clipping corners from the

loop coupons after 1500 and 4500 h of exposure to NaX0;-KNO3. Polished

S TR o

cross sections of specimens exposed to the draw salt at the H3 and Hé4
positions in each of the loops are shown in Figs. 7 through 9. Each alloy %j
exhibited multilayered corrosion products at the higher temperatures. 1In |
certain cases, but particularly in the case of types 304 and 304L stainless
steel, etching of these metallographic specimens Increased the number of
apparent layer: that could be differentiated on a given specimen.

Examples of this can be seen by comparing the micrographs in Fig. 10 with
the associated ones in Figs. 8 and 9. Micrographs of specimens exposed at
other loop positions are shown in Figs. 11 through 14. Those in Fig. 14

are of the surfaces of the coupons that exhibited the larger weight gains

O - Dk §14- 20919

 

-~
o

 

 

 

 

WEIGHT CHANGE (g/mh)
" o

 

 

 

 

 

EXPOSURE TiME (n)

Fig. 6. Weight change versus exposure time for type 304L stainless
steel exposed to flowing KaNO;-FNOj.

 
 

 

9
) Fig. 7.- Polished cross sectionsbf’ alkloy 800 _ex‘posed_ to fiowing
‘NaNO3;~KNO3: (a) 1500 h at 600°C, (b) 4500 h at 600°C, - (¢) 1500 h at
. 570°C, and (d) 4500.h1at_570‘c_. L \ - ’

 
 

 

 

Y-173085

 

(b)

{a)

e

 

T

ok

 

(d)

40

316 stainless steel exposed

....
s
uy
N
v

&
%
<o
o
0,
W
4.
-~
b,
o
o .
]
Bes
Wy D
S e O WM
Q Wy
_s.tm
g9

)

t.nnu
OO0
Y O N
WY -
w ”~
D
QO o
o
%) <
.. " e @™
2o
"

Smc
L ) o
=l |
m:..eb
O W™
mr
a.

nwoo
et ©
=32
oy
"~
QW
T

 

"

 
 

 

)

b

(

 

()

Y1801

 

Polighed cfoss'sectioné_of types 304 and 304L stainless

‘Type 304 for 1500 h at 600°C.
{c) Type 304 for 4500 h at 600°C

(2) Type

Fig. 9.

: -KNO;.

steel exposed to flowing NaNO,

L Y P

- -an B

°c.

:'699

(b) Type 304L for 1500k a

&

B
 

  

— [ e e et e Rt T el ek A sty e D

12

(a)

 

{b)

(c)

 

Fig. 10. Cross sections of stainless steel exposed to flowing
' RaR03-KNO3 for 4500 h at approximately 600°C; etched with aqua regia.
. (2) Type 316 stainless steel. (b) Type 304 stainless steel.

{¢) Type 304L stainless steel.

 
 

a) ©

Figo 11.
NaNO3-KNO3 for

 

 

Polished cross sections 6f.a116y 800 exposéd to flowing
4500 h at (a) 520°C, (b) 410°C, end (c) 375°C.
 

 

 

 

o

[ ST

ETE

 

14

Y-177228

 

 

 

Y-177244

 

 

Polished cross sections of type 316 stainless steel exposed

to flowing NaNO3—KNO; for 4500 h at (a) 505°C, (b) 410°C, and (c) 375°C.

 

Fig. 12.

 
 

 

%

Tl h?:".’ e
ik i e A I RS

 

Fig. 13. Polished cross sections of type 304L stainless steel

_ exposed to flowing NaKO3-KNO3 for 4500 h at (a) 410°C and (b) 375°c. . B :: K "
5
 

Y-178312

 

Y-178814

 

Y-178816

Polished cross sections offityfie 304i'stain1ess stéel_

 

 

Fig. 14.
exposed to flowing NaNO,

(¢) 450°C.

-KNO3 for 4500 h at (=) 545°C, (b) 520°C, and

 
   

17

 

reported in Fig. 6. Note that the large welght gains are also reflected
in relatively thick reaction layers and that spallation of the layers
seems imminent. Tables A.4 through A.6 in the Appendix list the total
depth of the corrosion products xr for each coupon as measured from the
respective micrographs.

High-resolution electron microprocbe analysis allowed determination of
the chemical composition of the reaction zone of type 316 stainless steel
after 4500 h of exposure at 595°C. The results indicated three layers of
different compositions, shown schematically in Fig. 15. The entire reac-
tion zone contained oxygen, and the individual layers comprising this zone
differed in their relative ratio of iron to chromfum. The outermost layer
was depleted in chromium, but the underlying one was not. Finally, a thin
nickel-rich layer was observed at the corrosion product zone—base metal
interface.

The salt in each of the TCLs was periodically analyzed to determine
the extent and nature of corrosion product specles; results are shown in
Table 2. (The sampling times in the table represent the number of hours
the salt circulated in a particular loop before analysis and, as such, do
not coincide with the coupon exposure times used 1n reporting the weight
change and microstructural results because of the periods in which the
coupons were out of the respective loops.) The data in the table clearly
show a general increase of chromium in the salt with increasing operating
time in all three loops. The concentrations of the other principal ele-
ments of the structural alloys did not differ appreciably from their
levels in the as-received salt. We are unsure of the reason for the occa-
sional, unusually high molybdenum concentration in the salt. Most prob-
ably the measurements are errcneous because we observed a decrease to a
"normal” concentration in the alloy 800 loop at a later sampling time. 1In
all the loops, the salt's nitrite ccncentration increased rapidly to about
1.9 wt Z but then Increased more slowly.

As described in the experimental procedures section, the pressure
above the loop salt rapidly increased vhen the temperature was raised
above about 520°C. Analyzed samples of evolved gas were mainly oxygen.

These analyses are given in Table 3.
 

 

 

 

 

REACTION OXYGEN
ZONE TRACE
MATRIX
/ IRON CHROMIUM
/] TRACE TRACE
e arsmnd L™
Fig. 15. Results from electron microprobe analysis of type 3154

18

ORN(-DWG 81-20923

 

 

 

 

 

stainless steel exposed to NaNO3-KNO3 for 4500 h at 595°C.

 

 

 

 

 

 

 

 

 

 

 

 

Table 2. Salt composition as a function

of loop operating time

 

 

Concentrat{on in salt? ~,

 

 

 

Loop
Loop operating
material tioe (vt ppm) (vt 1)
(h) crd N1 Ye Ho MO, »,
0 4 Qo 2t <2 0.05 65.5
Alloy 800 2,403 130 <10 1.6 2 1.7 68.1
5,619 137 Qo <1 27 3.2 63.6
7,755 156 <10 3 3.3 64,1
10,1042 237 (218) a
Type 304L 1,89 163 <10 Q 1.7 67.6
stainless 2,715 120 <10 2.3 <2 2.0 66.6
steel 4,249 200 <10 1.6 <2 2.0 67.2
8,809 222 Qo 3 3.9 63.2
9,672 246 Qo 4 4.0 63.4
12,122 322 (310) <
Type 316 1,126 63 ao 14 Q@ 1.9 64.3
stainless 4,034 118 <10 <1 i.8 66.5
steel 5,355 128 1o Qa 1.% 68.5
7,419 120 <10 1.1 <2 1.9 64.3
12,578 163 ao 29 4.3 62.7
14,638°¢ 270
15,313¢ 283 (282) a

 

OMeasurerent uncertainties are 25%.

biusbers 1n parentheses represent concentrations of Cr{VI) in salt.

“Max{mum loop temperature Tgay vas sbout 625°C; all other
measurement times were fcr & T, of about 600°C.

 

 

 

 

 

 

 

 

 

 

 
 

19

Table 3. Analyses of gas initially
evclved from KaKO3~KNJ; in closed
thermal convection loops

 

Concentration (wt I)

 

 

Species
Type 3C4L loop Type 316 loop
Ara <0.1 4.4
H,0 <0.1 1.3
N, + CO 0.7 3.6
0, 99.2 90.3

 

AArgon was the original cover gas.
DISCUSSION

A substantial amount of gaseous oxygen rapidly evolved when the draw
salt in these closed loops was heated to above about 520°C. This behavior

is consistent with the decomposition reaction
XNO3 = XNO, + 1/2 0, , (1)

where X can be sodium or potassfum. Such a reaction was reported
earlier for this salt.! As seen in Table 2, the measured nitrite con—
centrations in the salt increased In accordance with the decomposition
reaction and are in approximate agreement with those calculated with the
equilibrium constant for Eq. (1) (ref. 11) and the assumption that a mass
balance was maintained between nitrate and nitrite. Although metal oxida-
tion reactions were also occurring in the system, as evidenced by the
growth of oxide filas and the buildup of chromium fn the salt, 1t appears
that the above decomposition reaction controls the oxidation potential of
the system. This is consistent with the fact that the oxidation of the
metal is a small effect relative to the amount of salt in the loop and
therefore should not significantly affect salt chemistry.

As already discussed, metal oxidation processes were manifected by

the formation of surface reaction layers (Figs. 7-10) and in increased

 
el .

i ¥ . S
o g 5;.. 1—_:‘2__?‘_
e O

20

levels of chromium in the draw salt, and microprobe examination of
exposed type 316 stainless steel (Fig. 5) Indicated that significant
depletion of chromium from hotter loop surfaces also occurred. Both the
oxidation and depletion reactions are discussed in more detail below.

The weight cha:-es reported in Tables A.l1 through A.3 are not {indica-~
tive of the true corrosjon rates. The measured weight change AW is
actually the difference between the weight of the reacted oxygen AW, and
the sum of the weights of the dissolved elements (principally chromium)

8Wn and of the spalled material AVg:

e i

AW = AW, — (AWp + AWg) . (2)

The weight of the oxygen acquired by a specimen AW, can be roughly calcu-
lated from the thickness of the oxide layer x (see Tables A.4-A.6):

Mg = (64.0/231.8) ppe 0, X T, (3)

vwhere the density of Fe30,, pg, . » was used to approximate that of the Ef
VY 2
composite oxide layer. The total weight of oxidized metal AWp is defined N

as the weight of metal in both the oxide scale and molten salt. This
weight "loss™ is thus the sum of the weight of the dissolved elements AWp
and the weight of the reacted (oxidized) iron and chroamium AWM (o)t

AWp = AWp + AWYy(p) » (4)

where

 

MM(O} = 0.72( Pre 30“ X x + MS) . (5)

When AWp 1s divided by the time of salt exposure, an average approximate
corrosion rate for the particular specimen is obtained. The values of AVr
can be obtained by substituting in Eq. (4) values of AWp and AWy(,)
obtained by use of Eqs. (2) and (5), respectively, and appropriate
measurements of AY and x from Tables A.l1 through A.6. For our case, we
will assume A¥g = O because observation of possible spallation after 4500 h
of exposure was restricted to a very few specimens in one loop (specimens

H6 and H7 in the type 304L stainless steel TCL). With this assumption,

 
 

A N
-,fiz,"-l .¥“‘.

R
1!

 

 

 

!

21 e

approximate values of the total weight of oxidized metal of eagh loop }gi
specimen are listed in Tsbles A.7 through A.12. Measurements of the vfi%
weights of descaled specimens are In progress to determine the accuracy of %%f
the above approach. Corrosion weight changes and rates obtalned ir this éfi%
manner for the alloy 800 and type 316 stainless steel loops (Tables A.7— ffi%

A.10) are consistent with the expectation that oxidation kinetics, sad
therefore total weight loss, should increase with {increasing temperature
(this is not the case for the uncorrected weight changes measured with
the oxide films intact). In contrast, the weight losses calculated for
specimens in the type 304L stainless steel loop after 4500 h of exposure
did not maximize at the highest loop temperature. In this case, relati-
vely large corrosion rates resulted at intermediate temperatures (see
Table A.12). The accelerated oxidation at these intermediate positions
was characterized by the presence of relatively thick and apparently
poorly adherent oxide films (Fig. 14). However, the actual corrosion
rates at these positions are somewhat less than those calculated because
the calculations assumed a 1007 dense oxide film of the same thickness as
that of the observed porous film.

Regardless of poesible inaccuracies in the determinations of AVp,
the weight change results shown in Tables A.7 through A.12 indicate that
the metal in the oxide film constitutes most of the metal loss and that
the total corrosion rates are relatively small. Figure 16 gshows that the
oxide thicknesses on the alloys of this study are about an order of magni-
tude less than those expected for annealed type 304 stainless steel in
steam.12 The total weight losses after 4500 h in draw salt have been con-
verted into corrosion rates in the last column of Tables A.8, A.10, and
A.12 and indicate that, while the “worst case” corresponds to 25 mm/year
(1 mil/year), the corrosion rate is substantially leses in most ceses. The
one potential concern in these tests was the nature of the oxide films on
type 304L stainless steel specimens at intermediate temperatures, where
spalling appeared imminent.

Although the exact neture of the metallographically distinct surface

layers on exposed specimens is not known, it appears that, in generszl, the

 
22

ORNL-DWG B - 209211

 

 

 

 

’Ozt T T T YT
" ]
8 p -1
3 30458
1 2 STEAM’%iS:i:>/,/”fl
w
v
z
\j . -—
5 0F 316 S5 ]
r I ALLOYESB‘"““\L ,,,, — ]
b o :
g st ;:;" -
g N NaNO3~ KNOy B
600*C
. 304 8 304LSS
2} -
10° i i 14:1ui 1 i1l 141
10 2 3 10° 2 3 10

EXPOSURE TiME (n}

Fig. 16. Comparison of oxide thickness versus exposure time for
exposures to NaNO3;-KNO3; and steam.

reaction zone at higher temperatures consists of an iron oxide (probabdly
Fe30,) with an underlying spine! layer such as Fe(Fe,Cr),0,. This would
be consistent with the microprobe data and agrees with the results from a
study of the corrosion of type 304 stainless steel by thermally convective
draw salt.3 However, details revealed by optical metallography indicate a
more complex reaction zome than that just described. For example, a dark-
etching phase was sometimes observed optically at the salt—-allov interface
[Fig. 10{a)], although this phase was not revealed by microprobe analysis.
The other three layers seen in Fig. 10(a) were distinct arecas in the
microprobe analysis (Fig. 15). Another complicating factor is the multi-
tude of layers observed in the reaction zones of types 304 and 304L
stainless steel when their polished cross sections are etched (Fig. 10).
This variety of metallographically distinct structures was not observed

in another study of type 304 stainless steel,? and the reason why so

many layers should develop on these alloys and not on the other materials

 
23

in this study is not obvious from compositional differences. Finally, the
micrographs of the higher temperature alloy 800 coupons indicated some
tendency for internal oxidation at the reaction zone—base metal Interface
[Fig. 7(a)], although longer term tests will be required to establish the
significance of this observation.

The observed increases in the chromium levels of the test salt in
each loop imply.the formation of one or more chromium-containing specles
that are soluble in NaK03;-KNO;. Because most of the chromium in the salt
was In the +6 oxidation state (Table 2), Na,Cr,0; and Na,CrO, are probable
species. Changes in the chromium concentration of the draw salt as a
function of exposure time can also yield informatinn about the chromium
depletion process. Assuming that, as discussed below, there is no deposi-
tion of chromium from the salt, the increase in the chromfum concentrations
of the salt Cgr(g) should subscribe to a half-pover time dependence if
diffusion of chromium in the alloy or in the salt is rate controlling. A
linear time dependence would imply surface reactifon control. Recent work3
on the corrosion of type 304 stainless steel in draw salt revealed a time
dependence of t0+* for Ccr(s)+ In agreement with other work 3:° our data
generally indicate a continuing Increase in chromium levels of the salt
with time (Table 2) for the duration of the experiments. The chromium
concentration of the salt in the type 316 stalnless steel TCL appeared to
approach a time-independent value after several thousand hours and then
increased again when the maximum loop temperature was increased. 1In no
case, however, was the time dependence of the chromium concentrations 1in
the salt greater than t%<> for operation at a maximum temperature of
600°C; the best-fit exponent was always between 0.1 and 0.5. This would
thus indicate that a diffusional process involving chromium is the rate-
controlling step. Accordingly, the concentratiors are plotted verszus the
square root of exposure time in Fig. 17.

One possible corrosion effect is the movement of surface at-: e
tes* alloys from hot to cold regions of the loop via convection -

salt phase. This phenomenon of thermal gradient mass transport 2

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
24

ORNC-Dw 8- 20977

 

350 T —

 

Twax  600%C .
—ea— TYPE 316 LOOF
300 1 ~@— TYPE 304L {0P

=== AL LOY BOGC LOCF .

Twax=625% /

.
&

o TYPE 346 LOOP ’
& ALLOY 800 LOCF
s

3
T

 

 

 

:
<

8
I

Cr CONCENYRATION IN SALT (wt ppm)
8
7

 

 

 

1 ¢
1 A

60 a0 100 120

 

s+
!

[e xPOSURE TIME (m]'/2

Fig. 17. Chrowmium concentration in KaNO3;-KNO3 versus the square root
of loop opesrating time.

observed in other molten salt systems (principally hydroxides or fluorides)
subjected to a thermal gradfent.l!3 The transport mechanism can involve
oxidation and dissolution of reactive metals as corrosion products in the
hot zone with subsequent reduction and deposition of the metals in the
cold zone. Our tests indicate that an oxidation reaction operates to
dissolve chromium in the salt; however, the weight change data gave no
evidence that the reaction leads to a net transport of chromium from hot
to cold regions. There was also no evidence of cold-leg deposits in
visual or metallographic examinations of cold-leg specimens. Furthermore,
changes Iin the AT of the type 316 stainless steel loop did not signifi-
cantly affect the measured weight changes (Fig. 5). Thus, our results are
encouraging in that thermal gradient mass transport will not be a signifi-
cant operational problem in draw salt systems up to 600°C.

This study indicates two principal corrosion reactions resulting from
exposure of Fe-Ni-Cr alloys to NaNO3;-KNO3: the solid-state oxidation of

the exposed alloys and the selective dissolution of chromium from the

 
    

25

matrix. As such, our results agree with those reported3:“ for similar
alloys exposed to NaNO3-KNO3 in TCLs that were open to air. Spallation
was not extensive, and intergranular attack was not observed.

The results of this study indicate that alloy 800 and types 304,
304L, and 316 stainless steel have low corrosion rates in thermally con-
vective NaN0O3;-KNO3 at a maximum temperature of 600°C. Under such con-
ditions, the greatest metal loss rate corresponded to 25 jm/year, but in
most cases the rate was less than 13 ym/year. These latter rates are
somewhat smaller than those reported for similar loops operated with the
salt exposed to the atmosphere3»> but are within a factor of 2 to 5.
Significant metal oxidation was observed only at intermediate temperatures
in the type 304L stainless steel TCL after 5000 h of salt exposure (Fig. 6),
but the overall rate did not exceed 25 pm/year. Raising the maximum loop
temperature to 620°C resulted in much greater corrosion rates for type 316
stainless steel (Fig. 5). This rapid acceleration of the corrosion pro-
cess was also observed in other studies of types 304 and 316 stainless
steel and alloy 800 in draw salt loops,3» where at 630°C significant
spalling of the oxide films on type 316 stainless steel was observed.>
Our results therefore confirm that 600°C probably represents an upper tem-

perature service limit for such alloys in NaNO3;-KNO; (ref. 5).
SUMMARY

Alloy 800 and types 304, 304L, and 316 stainless steel were exposed
to thermally convective NaNO3;-KNO3 draw salt at 375 to 600°C for more
than 4500 h. The exposure resulted in the growth of thin oxide films on
all alloys and the dissolution of chromium by the salt. The weight change
data for the alloys indicated that (1) the metal in the oxide film consti-
tuted most of the metal loss; (2) the corrosion rate, in general, increased
with temperature; and (3) although the greatest metal loss corresponded
to a penetration rate of 25 ym/year (1 wil/year), the rate was less than
13 pn/year in most cases. Spallation had a sigrificant effect on metal
loss at intermediate temperatures in the type 304L stainless steel loop.

Metallographic examinations showed no evidence of intergranular attack

 
 

26

or of significant cold-leg deposits. Weight change data further confirmed
the absence of thermal gradient mass transport processes in these draw
salt systems.

Raising the maximum temperature of the type 316 stainless steel loop
from 595 to 620°C dramatically increaszd the corrosion rate. It thus
appeared that 600°C may be the limiting temperature for use of such alloys

in draw salt.

ACKROWLEDMENTS

This study was funded by a subcontract from Sandia National
Laboratory; R. W. Carling of Sandia was the task manager. The authors
acknowledge the following ORNL staff members who contributed to this wo~k:
E. J. Lavwrence, who designed and operated the thermal convection loops;

G. C. Marsh and R. S. Crouse for their metallographic and microprobe work;
D. A. Costanzo for his aid in the chemical analyses; Irene Brogden for

editing; and D. L. LeComte for preparation of final copy.
REFERENCES

1. Conceptual Deeign of Advanced Centwal Receiver Power System,
Pinal Report, DOE/ET/20314~1/2, Martin Marietta Corporation, Denver,

September 1978.
2. J. R. Keiser, J. H. DeVan, and E. J. Lawrence, "Compatibility of

Molten Salts with Type 316 Stainless Steel and Lithium,” J. Nuel. Mater.
85&86(11I, A), 29598 (1979).

3. R. W. Bradshaw, Corrosion of 374 Stainless Steel by Molten
NaN03-KNO3 in a Thermal Convection Loor, SANDB0-8856, Sandia National
Laboratory, Livermore, Calif., December 1980.

4. R. W. Carling et al., Molten Ftrate Salt Technology Development
Status Report, SAND80-8052, Livermore, Calif., March 1981, pp. 33-44.

S. R. W. Bradshaw, Thermal Convection Loop Corrosion Tests of 316
Stainless Steel and IN80O in Molten Nitrmate Salts, SAND81-8210, Sandia
National Laboratory, Livermore, Calif., February 1982.

 
 

27

6. E. B. Sandell, Coloritmetric Determiration of Traces of Metals,
3d ed., Interscience, New York, 1959, p. 392.

7. J. Applebaum and J. Marshall, "Silver(II) Oxide Oxidation and
Spectrophotometric Determination of Chromium(III),” 4Anal. Chim. Acta 35,
409 (1966).

8. A. E. Harvey, J. A. Smart, and E. S. Arnis, "Simultaneous
Spectrophotometric Determination of Iron{(II) and Total Iron with
1,10-Phenanthroline,” Anal, Chem. 27, 2627 (1955).

9. W. W. Scott, Standanrd Methods of Chewical talysis, 6th ed.,

Van Nostrand, New York, 1962, p. 748.

10. I. M. Kolthoff and E. B. Sandell, Textbook of Quantitative
Inorganic Analysis, 3d ed., MacMillan, New York, 1952, p. 581.

11. D. A. Nissen, The Chemistry of the 3inary %ak0;-KNOy Syetem,
SAND81-8007, Sandia National Laboratory, Livermore, Calif., June 1981.

12. J. C. Griess and W. A. Maxwell, The Long-Term Oxidation of
Selected Alloye in Superheated Steam at 482 and 538°C, ORNL-5771, July 198l.

13. J. H. DeVan and R. B. Evans III, "Corrosion Behavior of Reactor
Materials in Fluoride Salt Mixtures,” pp. 557-80 in Corrosion of Reactor
Materials, Int. Atomic Energy Agency, Vienna, 1962.

 
 

 

 

 

 

 

 

 

 
P e g Ry B gt

 

Appendix
CORROSION DATA
 

 

 

 

 

 

 

 

 
Table A.l. Weight changes for alloy 800 specimens exposed to draw salt
in an alloy 800 thermal convection loop

 

Weight change (g/m2) for total exposure in h

 

 

Speacimen Tem?sggture

378 1500 2002 3010 4498 4998 6146 6650
H1 580 0.0 +0.3 +0.9 +1.4 +2.6 +3.0 +3.4 4+3.9
H2 595 +0.3 +1.7 +2.0 +2.5 +3.7 +3.7 +4.7 45,7
H3 600 +0.3 +1.1 +0.9 +1.4 +2.0 +2.5 +3.0 +3.5
H4 570 0.6 -0, 0,3 +0.3 +0.3 +0.8 +1.3 +1.8
Hs 545 -0.6 ~0.6 ~1.1 ~0.3 0.0 0.0 +1.0 +1.5
H6 520 0.0 0.0 —0.6 0,3 0,3 =0, 0.2  +0.7 o
R7 495 0.6 0.6 -1.1 0.9 0.6 -0.6 ~0.6 ~0.6
H8 470 0.0 +0.3 -0.6 0.0 0.0 0.0 0.0 0.0
c3 485 -0.6 ~0.8 -1.1 -0.6 0.0 ~1.4 0.4 +0.1
ch4 455 -0.6 -0,.3 —~0.6 -2 .0 -1.4 ~0.9 -0.4 -0.9
C5 4135 -0.3 ~0.3 -~0.6 -0,3 0.0 ~1.0 -0.5 0.0
cé6 410 0.6 -0.3 ~0.6 ~0.3 -0.3 -0.3 ~0.3 4+0.2
c7 390 ~0.6 -0.3 0.6 -0.3 0.0 ~0.5 0.0 0.0
Cc8 375 ~0.6 ~0.6 -0.9 -0.3 0.3 ~0.3 ~0.8 0.3

 

 
Table A.2. Weight changes for types 304 and 3041 stainless steel
specimens exposed to draw salt in a type 3041, stainless
steel thermal convection loop

 

Weight change (g/m?) for total exposure in h

 

 

S Temperature
pecimen (°C)
500 1500 2121 3006 4488 4988 5489 6279 6804

H1 580 0.0 +0.3 +0.3 +0.3 +0.6 +0.6 -1.3 -1.8 —0.9
H2 595 -1.7 -0.3 -2.0 -2.0 +0.9 +2.6 +3.5 +2.2 +3.1
H3 600 -1.1 0.0 +0.9 +0.9 +0.6 —0.2 - 0.0 -1.3 -0:9
H3Aa 600 -2 .R -0.8 -0, 3 0.0 +0,9 +0.1 40,1 -1.5 -1 .%
H4 570 +0.8 +2.3 +2.0 +1.7 +2.0 +0.7 ~0.2 ~1.5 —5.1
H5 545 +0.8  +1.1  +11.4 +8.8 +0.6 +1.1  +0.6 0.8 -1.3 e
Hé6 520 0.0 +0.6 +16.8 +14.5 415.4 416,99 +11.0 +10.5 +11.0
H7 490 0.0 +0.8 +4.8 +4.6 +4.8 +4,8 +3.5 +3.5 +3.5
HA 465 +0.3 +2.0 +2.6 +2.8 +3.4 +2.% +2.5 +1.6 +1.6
c3 485 +0.6 +1.7 +4.0 +4.6 +5.7 +4.7 +5.2 +5.7 +6.7
ch 455 40,3 +2.0 +2.8 +3.7 +5.7 +4.2 +6.7 +6.7 +642
ChAa 455 0.0 +1.7 +3.1 +3.7 +6,6 +3.2 +2.7 +3.2 +3.2
C5 435 0.0 +1.1 +2.0 +3.4 +4.0 +3.6 +3.2 +3.2 +3.2
cé6 410 +0.3 +0.8 +1.4 +2.6 +4.0 +4.0 +3.1 +2.2 4+2.2
c7 390 4+0.6 +1.7 +1.7 +2.3 +2.8 +2.4 +2.0 +1.1 +1.5
c8 375 +0.3 +2.8 +2.8 +3.1 +3.7 +2.8 +2.8 +2.4 +1.9

 

aSpecimens H3A and C4A are type 904 stainless steel; remainder are type 304L.

 
Table A.3. Weight changes for type 316 stainless steel specimens exposed
to draw salt in a type 316 stainless steel thermal convection loop

 

a ®
Temperature® (°C) Weight change (g/m?) for total exposure in h

 

Specimen

 

 

;gS°C f§0°c 499 1500 2001 3010 4501 5102 6108 7133 7658 8660
Al 580 580 0.0 +1.1 +0.6 +0.3 +0.6 0,3 +0.1 +0.5 +0.5 +0.5
H2 590 580 +0.3 ~4.0 ~2.8 -2.3 -2.0 2.5 -2.0 ~0.6 -0.1 ~0.1
H3 595 595 —1.4 -0.6 ~1.4 -1.1 -0.9 -1.4 -0.9 -1.8 ~6.6 —8.0 w
R4 : 565 575 +0.3 -3.1 -6.6 8.6 8.6 -8.1 ~7.6 -3.7 8.0 =745 w
HS 335 555 +0.8 +0.8 0.0 +0.3 +0.6 +1.1 +1.6 +5.7 +3.3 +5.2
H6 505 535 +0.6 +0.6 +0.3 +0.6 +0.6 +1.1 +1.1 +3.5 +2.5 +4.9
H7 475 - 520 0.0 +0.3 +0.3 +0.6 +0.3 +0.8 +1.3 +1.8 +1.8 +4.2
H8 445 500 0.0 0.0 ~0.3 0.0 -0.3 +1.3 +1.8 +3.4 +3.4 +5.0
c3 485 530 +0.8 +0.6 +0.6 +1.1 +1.4 +2.8 +2.3 +7.0 +5.1 +8.8
Ch 455 505 +0.3 +0.8 -1.1 -1.7 -1.1 ~1.6 —2.5 +0.8 ~1.1 +0.3
c3 435 493 +0.8 +0.8 +0.8 +1.4 +1.7 +1.7 +2.7 +3.2 +3.2 +5.1
Cé6 410 480 +0.3 +2.8 +2.8 +3.1 +3.4 +3.4 +3.4 +6.2 +5.7 +7.6
c? 390 465 +0.3 +0.3 +0.6 +0.8 +0.6 +0.1 +0.1 +2.9 +2.4 +4.8
c8 375 450 -1.1 —1.1 -1.1 —1.1 -1.1 -1.1 ~0.6 +1.3 +0.8 +3.2

 

AT (temperature differential) changed from 250 to 150°C after 5102 h; Tpax (maximum temperature)
changed from 595 to 620°C after 6108 h with corresponding changes in all specimen temperatures.

  
34

Table A.4. Average corrosion product
thickness for alloy 800 specimens
exposed to flowing drav salt

S

(Average of ten measurements)

 

Thickness x (jm)

 

 

Specimen Tem;(afé‘é;ture _— i:
1500 h 4500 h i
E1 580 3.7 5.1 fi?
H2 595 3.7 6.2 '§j
H3 600 5.4 6.8 3
B4 570 2.8 5.3
HS 545 2.3 3.6
H6 520 1.0 1.2 s
H7 495 0.4 1.1 féi§
H8 470 0.3 1.3
c3 485 1.8 2.5
c4 455 1.6 2.7 "
c5 435 0.9 1.7 ,;;'
c6 410 0.4 2.7 .
c7 390 0.4 1.2
c8 375 0.5 2.3
‘; 

 
35

Table A.5. Average corrosion product
thickness for types 304 and 304L
stainless steel exposed to
flowing draw salt

(Average of ten measurements)

 

Thickness x (jm)

 

Specimen Tem?sz§ture —_—

1500 h 4500 h
H1 580 4.1
H2 595 2.6 6.8
H3 600 3.6 5.6
H3AZ 600 3.6 5.0
H4 570 1.9 6.0
H5 545 1.2 11.9
H6 520 0.4 22.8
H7 490 1.1 3.5
H8 465 1.4 1.3
Cc3 485 2.9 5.1
c4 455 2.0 3.1
C4AC 455 1.7 2.4
C5 435 1.4 3.1
c6 410 0.7 0.6
c7 390 <0.4 1.4
c8 375 1.1 0.7

 

IType 304 stainless steel; others are
type 304L.

 
36

Table A.6. Average corrosion product
thickness for type 316 stainless
steel exposed to flowing
draw salt

(Average of ten measurements)

i K X2V gy D

i

 

Thickness x (jm)

 

Specimen Tem?sé§ture
1500 h 4500 h
H1 580 | 0.5
H2 595 1.9 4.4
H3 595 4.0 7.7
H4 565 0.2 8.2
H5 535 1.7 5.7
H6 505 0.6 2.8
H7 475 0.2 0.0
H8 445 0.0 0.0
C3 485 2.2 7.4
C4 455 1.7 4.4
C5 435 0.9 1.3
Cé 410 6.1 4.0
c? 390 6.3 0.2

c8 375 0.0 0.0

 

L 27 L gt et 8
37

Table A.7. Weight change results for alloy 800 exposed
to flowing NaNO3;-KNO3 for 1500 h

 

 

 

b Weight change® (g/m?)
Specimen (;C) (3m)

AW, AW ?P AWyt o) AW
H1 580 3.7 5.4 +0.3 S{& 13.8 18.9
H2 595 3.7 5.4 +1.7 3:7 13.8 17.5
H3 600 5.4 7.9 +1.1 6.8 20.2 27.0
H4 570 2.8 4.1 -0.3 k.4 10.5 14.9
H5 545 2.3 3.3 -0.6 4.0 8.6 12.6
H6 520 1.0 1.5 0.0 1.5 3.7 5.2
H7 495 0.4 0.6 0.6 1.2 1.5 2.7
H8 470 0.3 0.4 +0.3 0.2 1.1 1.3
Cl 470 —2.8
C2 475 +1.1
C3 485 1.8 2.6 —0.8 3.4 6.8 10.2
Cé4 455 1.6 2.3 -0.3 2.6 6.0 8.6
C5 435 0.9 1.3 0.3 1.6 3.4 5.0
Cé6 410 0.4 0.6 -0.3 0.9 1.5 2.4
Cc7 390 0.4 0.6 ~-0.3 0.9 1.5 2.4
cs8 375 0.5 0.7 0.6 1.3 1.9 3.2

 

aT, temperature.
bx, thickness.

CAW,, welght of reacted oxygen; AW, measured ...ght change;
A¥p, total weight of dissolved elements; Aky(,), weight of reacted
iron and chromfum; and AWp, total weight of oxidized metal.

 

 
Table A.8.

Weight change results for alloy 800 exposed to
flowing NaNO3-KNO3 for 4500 h

 

Weight change® (g/m2)

 

 

Specimen (Ez)
AW BWp( o) AWp (m/year) (mil/year)
H1 580 +2.6 19.1 23.9 5 0.2
H3 600 +2.0 25.5 33.4 8 0.3
H4 570 -0.3 19.8 27.3 8 0.3
HS 545 0.0 13.5 18.7 5 0.2
H6 520 -0.3 4.5 6.5 3 0.1
H7 495 -0.6 4,1 6.3 3 0.1
H8 470 0.0 4.9 6.8 3 0.1
Cc1 470 —2.0 12.0 18.6 5 0.2
c2 475 +1.4 B.2 10.0 3 0.1
Cc3 485 0.0 9.4 13.0 3 0.1
C4 455 -~1.4 10.1 15.4 3 0.1
C5 435 0.0 6.3 8.8 K} 0.1
cé6 410 -0.3 10.1 14.3 3 0.1
c7 390 0.0 4.5 6.2 3 0.1
c8 375 -0.3 8.6 12.3 3 0.1

 

 

AL WY e v
e s

ar, temperature.

bx, thickness.

dCR, corrosion rate.

CAW,, weight of reacted oxygen; AW, measured weight change; AWp, total weight of
dissolved elements; AWy(,), welght of reacted iron and chromium; and AWp, total weight of
oxidized metal.

 

 

gt
39

 

 

 

 

 

Table A.9. Weight change results for type 316 stainless
steel exposed to flowing NaNO;-KNO; for 1500 h

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

ra b Weight change® (g/m?)
Specimen (°C) (3m)
AW, AW Wi AWM(p) bW
H1 580 +1.1
H2 595 1.9 2.8 -4.0 6.8 7.1 13.9
H3 600 4.0 5.8 ~0.6 6.4 15.0 21.4
H& 565 0.2 0.3 —3.1 3.4 0.8 4.1
H5 535 1.7 2.5 +0.8 1.7 6.4 8.C
H6 505 0.6 0.9 +0.6 0.3 2.2 2.5
H7 475 0.2 0.3 +0.3 0.0 0.7 0.7
H8 445 0.0 0.0 0.0 0.0 0.0 0.0
Cl 470 6.0
c2 475 0.6
C3 485 2.2 3.2 +.6 2.6 8.2 10.8
C4 455 1.7 2.5 +0.8 —1.7 6.3 8.0 5
C5 435 0.9 1.3 +0.8 0.5 3.4 3.9 :
cé 410 6.1 8.9 +2.8 6.1 22.8 28.9
c7 390 0.3 0.4 +0.3 0.1 1.2 1.3
C8 375 0.0 0.0 -1.1 1.1 0.0 1.1

 

 

 

 

ar. temperature.

 

 

b:z:, thickness.

 

 

CAW,, weight of reacted oxygen; AV, measured weight change;
AVp, total weight of dissolved elements; AWy(o), welght of
reacted iron and chromium; and AWp, total weigh® of oxfdized
metal.

 

 

 

 

 

dNegative number indicates physical impossibility. N

 

 

 

AT PRTe, - 1 WA
A TS e o

 
Table A.10. Weight change results for type 316 stainless steel
exposed to flowing NaNQO3;~KNO3 for 4500 h

 

 

 

 

 

 

ra b Weight change® (g/m?) CRE
Specimen (°C) (1) ‘
- AW, AW Awa AWyM(0) AWmp (um/year) (mil/year)
H1 580 0.5 0.7 +0.6 0.1 1.9 2.0 <3 <0.1
H2 595 4.4 6.4 -2.0 8.4 16.5 24.9 5 0.2
H3 600 7.7 11.2 -0.9 12.1 28.8 40.9 10 0.4
H4 565 8.2 11.9 -8.6 20.5 30.7 51.2 13 0.3
H5 535 5.7 8.3 +0.6 7.7 21.3 29.0 8 0.3
H6 505 2.8 4.1 +0.6 3.5 10.5 14.0 3 0.1
1 ¥4 475 0.0 0.0 +0.3 -0.3 0.0 0,3
H8 445 0.0 0.0 ~0.3 0.3 0.0 0.3 & <0.1
I
Cl 470 0.0 0.0 -0.3 0.3 0.0 0.3 {3 <0.1 @
C2 475 1.9 2.8 -0.3 3.1 7.1 10,2 3 0.1
C3 485 7.4 10.8 +1.4 9.4 27.7 37.1 10 0.4
C4 455 4.4 6.4 -1.1 7.5 16,5 24.0 5 0.2
C5 435 1.3 1.9 +1.7 0.2 4,9 5.1 <3 <0.1
Cé 410 4.0 5.8 +3.4 2.4 15.0 17.4 5 0.2
c7 390 0.2 0.3 +0.6 -~0.3 0.7 0.4 <3 <0.1
c8 375 0.0 0.0 -1.1 1.1 0.0 1.1 <3 <0.1
ar, temperature.
bx, thickness.
"AW,, weight of reacted oxygen; AV, measured weight change; AWp, total weight of
dissolved elements; AWy(,), welight of reacted iron and chromium; and AWVp, total weight of
oxidized metal.
dNegative numter indicates physical impossibility.
€CR, corrosion rate.
B R - * pA T L. A A L T s e i

 
41

 

 

 

Tabie A.11. Weight change results for types 304 and 304L
stainless steel exposed to flowing NaNO;-KNO3 for 1500 h
ra & Weight changeC (g/m?)
Specimen (°C) (am)

AWo AW s Mo AW
H1 580
H2 595 2.6 3.8 -0.3 4.1 9.7 13.8
H3 600 3.6 5.2 0.0 5.2 13.5 18.7
H3A 600 3.6 5.2 —0.8 - 6.0 13.5 19.5
B4 570 1.9 2.8 +2.3 0.5 7.1 7.6
HS5 545 1.2 1.7 +1.1 0.6 4.5 5.1
H6 520 0.4 0.6 +0.6 0.0 1.5 1.5
H7 490 1.1 1.6 +0.8 0.8 4.1 4.9
H8 465 1.4 2.0 +2.0 0.0 5.2 5.2
Ccl 470 +0.8
c2 475 +2.3
c3 485 2.9 4,2 +1.7 2.5 10.9 13.4
Cé4 455 2.0 2.9 +2.0 0.9 7.5 8.4
C4A 455 1.7 2.5 +1.7 0.8 6.4 7.1
c5 435 1.4 2.0 +1.1 0.9 5.2 6.2
Ccé6 410 0.7 1.0 +0.8 0.2 2.6 2.8
C7 390 <0.4 0.6 +1.7 -1.1 1.5 0.4
c8 375 1.1 1.6 +2.8 -1.2 4.1 2.9

 

AW p,

ar, temperature.
bx, thickness.

CAW,, welght of reacted oxygen; AV, measured weight change;
total weight of dissolved elements; AWys,), weight of

reacted iron and chromium; and AWp, total weight of oxidized

metal

dNegative number indicates physical impossibility.

 

 
Table A.12. Weight change results for types 304 and 304L stainless steel
exposed to flowing NaNO;-KNO; for 4500 h

 

 

 

 

ra - Weight change® (g/m?) CR€
Specimen (°c) (ym)
AW, AW AWpd AWy o) AWp (um/year) (mil/year)

H1 580 4.1 6.0 +6.0 5.4 15.4 20.7 5 0.2
H2 595 6.8 9.9 +0.9 9.0 25.5 34.5 8 0.3
H3 600 5.6 8.2 +0.6 7.6 21.0 28.5 8 0.3
H3A 600 5.0 7.3 +0.9 6.4 18.7 25.1 5 0.2
H4 570 6.0 8.7 +2.0 6.7 22.5 29.2 8 0.3
H5 545 11.9 17.3 +0.6 6.7 44,6 61.3 15 0.6
Hb 520 22.8 33.2  +15.4 17.8 85.4 103.2 25 1.0
H7 490 3.5 5.1 +4.8 0.3 13.1 13.4 3 0.1
HB 465 1.3 1.9 +3.4 -1.5 4.9 3.4 <3 <0.1
Cl 470 +2.3

c2 475 +2.0

3 485 5.1 7.4 +5.7 1.7 19.1 20.8 5 0.2
C4 455 3.1 4.5 +5.7 -1.2 11.6 10.4 3 0.1
C4A 455 2.4 3.5 +4.6 -1.1 9.0 7.9 3 0.1
c5 435 3.1 4.5 +4.0 0.5 11.6 12.1 3 0.1
C6 410 0.6 0.9 +4.0 =3.1 2.2 -0.9

c7 390 1.4 2.0 +2.8 0.7 5.2 4.5 <3 <0.1
c8 375 0.7 1.0 +3.7 -2.7 2.6 -1.9

 

ar, temperature.
bx, thickness.

CAW,, welght of reacted oxygen; AV, measured weight change; &p, total weight of dissolved
elements; AWy(p), weight of reacted iron and chromium; and AWp, total weight of oxidized metal.

dfiegative number represents physical {mpossibility.

€CR, corrosion rate.

 

Y