UNCLASSIFIED

AD NUMBER

AD844277

NEW LIMITATION CHANGE

TOApproved for public release, distributionunlimited

FROMDistribution authorized to U.S. Gov't.agencies and their contractors; CriticalTechnology; 03 NOV 1967. Other requestsshall be referred to Foreign TechnologyDivision, Wright-Patterson AFB, OH 45433.

AUTHORITY

FTD USAF ltr, 7 Oct 1971

THIS PAGE IS UNCLASSIFIED

FTD-HT-.67-166

-7z ti

FOREIGN TECHNOLOGY DIVISION

THE RELATION BETWEEN THE GRAIN SIZE AND THE TENDENCY

'OF' AUSTN!TIC CORROSION:-RESITANT' STEELS

TO, INTERCRYSTALLINE CORROSION

by

* Vladimir Cihal

STATEVXNT

s sb ec D DC5 e 0b

This document iS subfrct tok sped3i ' eoS and @881

;,.. /to

t6

" I III al

-L,

FTD HT- 67-166j

EDITD TRANISLATION

THE RELATION BETWEEN THE GRAIN SIZE AND) THE.TENDENCY OFAUSTENITIC CORROSION -RESISTANT -STEELS TO INTERCRYSTAIJLINECORR6OSIO'N

By: Vladimlr Cihal

English pages: 10

SOURCE: Hutnicke Listy (Metallurgical Journal). Vol.No. 1, 1 66, pp. 7-82.

Translatj! d tinder: Contract AF33(657)-i661~8

____ ____ ____ ___ ____ ____ .. TP70QI604

TI6S'T1ANSaLAfION IS A RNMITIOW OF THE 035Wj.MAL FORIWI TEXYWITNOUT ANY ANALyTICAL OnEDITORIAL COMMENT. STATEMENTS HOR gliES PEPRDAD~bCA!'MDOmikPLIEDAR1 TNOSEoP THE SOURCE' PEAE Y

A16 MC, ECESSARILY REPLECT THE PNho TRANSLAT IVO W~Olt 0,PIWW) OF THE FOREIGN YICHMOOS ft OREWN TFICNOL03 ySWY1O W*APS. O0iO.

FTD- HT 67-166D3~_16

S

ITIS INDRXC ONTROL P0OR401 AMc Nr 68 Translation Nr XefAcc Nr Re rneN

P7P016O4 oo70166 18825 1515'97 Header Clas 63 Clas 64 -Control Markings 94 Expans -40 Ctry Info

UNML UNCII, 0 -0 CZ02 Ctry- 03Ref 04 Yr 05.SV01 '06Iss 07 B.P 45 E.Pg.' 0 Dte

CZ 003 66 012 001 007 082 NONETrans-iTerated Title VZTAHY MEZI VELIKOSTI ZRPNA A SKI5ONEM AUSTENITICKYCH

KOR~OZIVZDORNYCH OCELI K.MEZIKRYSTALOVE KOROZI09 English Title 'THE RPLATION BETWEEN THE GRAIN SIZEt AND- THE TENENCY OF

ATT~!F~T~PC~ -- OROETN P T'AM S'IT-. TQ- TN= NE( PYS'AT.T.Th43 Source-(ZEH

HIMTNICKE L~Y CEH

42 kAthor 98 Document Location

16 PD, Author -47 Subjec-t Codes

16 ;.o-_Author 39 Topic-Tags: autriidselNON intekgranular cirrosion, steel corrosion,

160 Co -Author - crosion resistanceNO14E .

16 Co-Author'--NONE , . _ _ _ _ _ _ _ _ _ _ _ _ _ _

ABSTMCt.:-"0fi' the b&sis of colleted knowledge about intercrystallinecorrogion- of austenit-c. tainless steels -a suggestion of evaluating theinifluence of grain size upon intercrystalline corr~osion,. both of' unst~bilized

and,-st~i~ed-'steels, -is -discussed..According t6 well-known iidadapted formulas which express the

-inl1luenc6 of' nickel, chromium and molybdenum, niobium and - above-'all carbon,ieltio wis stated, which includes the influence'of grain

gizel and qevaluates -the regstance of' 'steels,,' type Cr&TNi9, to intercrystalline

e Avv relation- -is r~presented -by a =nmograph- on lo-garithmic scalewith, the aim, to valuae -intercrysaln cors~no tIi:ss steels withdiffferent _ranularity and,.different composition, in' a simple way.

-In the conclusiz ~ pr cca api±'abilityt of' the sugjgested methodis 'demonr~ated.-Engl-ish ti, nsiati6L i0 pages.

YI'D FEe 67 0-90

THE RELATION BETWEEN THE GRAIN SIZE AND THE TENDENCYOF AUSTENITIC CORROSION RESISTANT STEELS TO

INTERCRYSTALLINE CORROSION

Vladimir Cihal*

HOW GRAIN SIZE IS RELATED TO INTERCRYSTALLINE CORROSION OF AUS-TENITIC STMIILES-S STEELS

On the basis of.collected knowledge about intercrystallinecorrosion of austenitic stainless steels a- suggestion of evalu-ating the influence of grain size upon intercrystalline corro-aion, both of unstabiZized and stabilized steels, ie discussed.

According to well-known and adapted formulas which expressthe influeizce of nickel, chromium and, molybdenum, niobium andabove a1l carbon, the following relation was stated, which in-Cludes the influence of grain size and evaluates the resistanceof stee', type Crl8Ni9, to intercrystalline corrosion:

ooo5V - .(c - 17) + f . '- 43'%N OO1) - ,0o2. (Ni - 10) - C Z o.-? 4

This relation is represented by a, nomograph on logarithmicscale with the aim, to evaluate intercrystalline corrosion ofstainless steels with different granularity and different compo-

sition, in a simpe way.

In the concts ion, practical applicability of the suggestedmethod is demonstrated.

Along with the development of corrosion resistant steels |

resistant to intercrystalline corrosion -a great deal of attentionwas given recently to, the limiting content of carbon - also calledthe minimum - at which value and below the' steel no longer has atendency toward intercrystalline corrosion. Associated with thisis not only the problemh of development and intpoduction into pro-duction of low carbon content stainless steels but the determina-tion of the minimum carbon content also has a considerable signif-icance in stabilized steels and in steels alloyed with molybdenum.

On the basis of the currently generally recognized theory ofintercrystalline corrosion - depletion of grain boundaries ofchromium created by the deposition of complex carbides of chromium(Cr, Fe, Mo) 23 C6 - almost all factors governing the tendency ofstainless steel toward intercrystalline corrosion are known. Among

FTD-HT-67-166

these factors belong primarily the carbon content of steel, thenythe amount of chromium, nickel, molybdenum and .lso manganese,the grain size, the content of stabilizing elements and their ef-fectiveness, thermal treatment, the content of nitrogen and oftenalso- the manner of molding. The number of factors and the inter-connection of their effects indicate the relatively large diffi-culty of soLution which cannot 76e accomplished without , statis-tical interpretation of a r'arge. number of tests.

Current knowle -dge of ;roperties of austenitic corrosioh -re-gistant steels aillows one, with-the help of available dii.t, todet~ermine with certainty the inifluence of somie~ factors. The scopeof 'this' work is 'primarily the study -of one- of the important fac-tors, that is, the size of the grain which is also important incorrosion problems ocuri'ing in- welding. We encounter in practicein a number of cases-,-for example, in the welding of thick-walledvessels or on repeated heatifig o' the sfeel in the zones of cri-tical tempe~rature a-tendency tbwai-'d intercrystalline corrosion innonstabtlized a&'4611 as in stabi~liV'ed steels. .In the lattersteels the -cause of' initercrystal ,ine cor.66irosio'n that apptArs afterin It'hei critical temperiture regions preceding overhea.ting 'ofsteels i's the':diss~olution of stakle -carbides anid the growth oftlhe' steel grain'.V

As- the -steel grain enlarges Lhe surface of 'grain boundariesdecreases, s' --that the concentration of chromium carbides in-preAges on the boundAres at the ,Iame cont~int of carbon in steel.This is confirmed. by morphology a4,d by thee distribution of chro-mi uih arbidis" on th'e grain ~boundari~s determined 'by the electronmicr~oscope. It, is, als6 general!-,, known, and recently proved by spot,microanalysis that the penetration ,- of intercr~stalline corrosion-occurs on'v at 'a certain concentration of the precipitatw ~creates a cdontinuous depletion of grain boundaries of qh -ru _M4.it can thus be 'said, that the f ine-grained steels.. -due' to te4y e_-l1.rged surface. of gr'ain boundaries,:e much moreestattintercrystalline corrosion than coarse-grained, -l ~ethe~oniinuous-network of carbides, and thus -depU..E-qrboundaries of chromium, are' formed mo'e, redi. core 4steels. 'The am6UIint of ca'rbides per unitf grt slhe '

agven content o6f hrbon, will dif f er en~ding- on', the gin -sAcaording to some references [1] the -amn1p f ~r correspond->1 ing to the chromium -carbides' f ozmed, er, -unft surface of grain in-creases approximately 1.141 time's when'the grain dimension changesby- a ,unit ofograin size which is ess~ntally the same by our [2]and by the fordign standards £3, 41].

INTtRCRYSTALLINE CORROSION OF STEELS Crl.8N19

The statistical study of F. Bogin [5] and, the experimentallyfound scattering of results of tests for intercrystalline corro-sion of corrosion resistant steels can be exploited in furtherareds- of thisfield. In practice we also often find steels which

I I have a different susceptibility to intercrysta-l-line corrosion al-though they have the same chemical composition. The cause of thisdiversity was often found to be due to the difference of their'grain sizes and thi's serves as a proof at the same time that thefactor of the grain size cannot be ignored [6].

-2-

FTD-HT-67-16

Since nickel if present i, h~gii *mouznts irn steel enhances Ithe effect of carbon -Prom the point bf view of sen~sitivity to-ward intercrystalline corrosion then thie so-called effectlye con- itent of carbon C', expressed by thie relation

is used for the evaluation., where C Is the canteul. of' carbon~ insteel in weight percent, 1M is the content of ra~tkel irs weightpercent from which It f;*.3.ow" that., when Ni In --lefs thar. 10, themodified content of carbon Ini Otel Is laver them the actual.

For tho purpose of study of the tft'eat of grain cize on thetendency toward irtterarystalihe qarroalor. &hic1 is especially I*-portant In steels whose coposItlon lie% on the border line ofresistance we can,, in agreezont with the worix of 0,L SvarcvaC1IL wrlt. a general dependenee of the oontont of chromium an theminiums 0ontent of 6aron In steel whicb assures the resist-ancetowWr- 1ntererystallirse porronion &V~ a given gr&ir. size

f(N).P -C + C6 t C,(2) 4

iv which f()is the function expresaing the grain site, Cr de-.notes tht content of chrtia at vbich the suscoeptibility towardirterarystalline cokrroSion 4& no loziger dependent on the grainalto, Cre is a con~aant Indicatj~ng the ohromiiuim content at whichthe tendency to int.raryalalli1ne corr'osion to a function of grainSise, Cr is the amount of' oaro diloolVed in austenite, C' Is4'the effective cArbon oantent in ateel according to Eq. (1).

For the expression of' the function f(M) we utilize the dataof routine measurements of' grain size. According to the definitionof grain size (I) the number of grains (xs) per unit area is ex-pressed by the relation n a k-N1 Let uas suppose further thatthe grain has a nexagonal shape rememberinig that two grains aredivided by one border. A general expr'ession for the length ofgain boundaries in a unit area of u mwtallographio ground section(.MM a or In the studied 10 em' at 100'x magnification) is

O..)~F~Accor~ing to this the funkction f (1) in Eq. (2) canthen be expressed by the relation

IA (3)

By substituting into Eq. (W we get with the value of grain sizeI the relation

K. (C - 00;a C(4)

The constants In thic equation hava to be determined experi-mentally. Gobin employs for theirp detrmiration the already men-tioned statistical measurements and h~t 'rrivea at the values K-0.005 and Cr. w 17. Svarcovs. edeterip-ines these constants from the

calculation of the amount of' carbldc of chrom'ium deposited on theunit circumference in several 8tee.2l ;cnd:3 ntabillzed with titaniumand she arrives at values that are on the whole In agreement withhis, K -0.0G625, Cro 26.7. Consider.nF, ths r~t&atiatical charac-Ater of' Gobin's test %e usedI iils valu s, h~wevar miore rounded. The

FITD-T-67-166

1* ' . . i

/

Fig. 1. The dependence of the rate of intercrystelliae corrosion(mm/hours) on the carbon content. 1) According to Bain, Rutherford,Aborn; 2) corrected curve according to Heger and Hamilton £73. a)Rate of corrosi on in thousandths of Inches/hour; 2) mm/hour.

value of the constant C, 0.015 differs from the data obtain~edfrom mathematical statistical analysis of Bain, Rutherford andAborn's diagram (Fig. 1) according to whom the m~ni--UM- raiLe -:finterrystalline corrosion is reached only at a carbon contentbelow 0.01% and which has a much larger effect on the distortionof results compared with the constant Cr,. We have t ereforedecided not to Include it in the relation Eq. (1). THl symbol C'denotes a total free (not bound) carbon modified i±th regard tothe nickel content in the steel.

On the basis of the above-mentioned fact we have seltiedthe following relation determining the resistance o1 corrosionresistant steels of the type Crl8Ni toward interrysta.rie cor-rosion

which have expressed nomographically in a dIagram in Fig. 2.The application of this diagram can be demonstrated with a simpleexample. If we have a steel containing 20% chromium and 10% nick-el, of grain size N 5, then, when carbon content is 0.06% (andlower) the steel will still be resistant toward intercrystallinecorrosion. On the other hand for a given content of chromium andcarbon we can read from the graph the grain size which still en-sures the resistance of steel to intererystalline corrosion. Aswas mentioned above the relation of the grain size and intercrys-talline corrosion is important even in steels alloyed with molyb-denum and in 3tabilized steels. We shall, therefore, attempt torearrange the obtained relation so that we could apply the givennomogram even to the latter types of corrosion resistant steels.

I

144

/L.10I

Fi.2.Te nleneofgan ie1nte ucetbliyofcr

4M as w n 4 W40w 40&

Ce - C + bO.(NI-o; +002 Ta I -C0 . -. r

A) Or; B) Ti in isolated material; C) Ti in TIC.

THE EFFECT OF MOLYBDENUM ON THE RESISTANCE TO INTERCRYSTALLINEfOFROSION

The admixture of molybdenum into corrosion resistant steelshas a favorable effect on the steels by improving their overallcorrosion resistance and by increasing the value of carbon contentat which the steel is still resistant to Intercrystalline corro-sion.Thus in the presence of molybdenum the value for the chro-mium content in Eq. (5) increases and it can be expressed by therelstion

4e = c'cr +b. M01 (6)

in which Cr is the percentage of chromium in steel, Mo is thepercentage of molybdeium in steel, b is the coefficient of equ!-valency considered by Gobin to be equal to 1, C'i is the effectiveI content of chromium as plotted in the graph.

14 ic i 20 XO-

0 1 2 5 "

S %I

Fig. 3. The effect of molybdenum (0) and chromium (9) on the orit-nal passivation current density j of the steel type Crl8Nil2Mo2in the solution 1M - HC1 + 0.01% KCNS at 201C.

We believe that there is not as yet sufficient justificationfor the value of coefficient b and for the present we must be sat-

} isfied that b = 1 and to ascertain its value in the future.

Basically it is correct to consider its additive effect withchromium since in tne standard austenitic corrosion resistantsteels it is deposited as complex carbides with chromium and notas a special carbide; in a solid solution it improves the abilityof passivation much more distinctly than chromium (Fig. 3).

6 -

We lack the exact data of the percent distribution of molyb- J

denum in a comlex carbide, but considering its effect in the

solid solution it seems probable, that the value of coefficientb is higher than 1. This would also be born out by the fact [83that the percentage Ti = 4. (0C) roughly suffices for the stabil-izaton of steel Cr!8NilOMo2T! so that molybdenum actually be-

1 .

haves indirectly as a stabilizer although not as an effective one

as titanium, because the complex carbides containing molybdenum

dissolve in austenite already at 11000C.

STABILIZED CORROSION RESISTANT STEELS IThe Droblem of the effect of grain size on the tendency to-

ward intercrystalline corrosion is much more complicated in sta-bilized steels. An admixture of a stabilizing element in suitable

quantities imparts to stainless steels a resistance to intercrys-talline corrosion. In practice various formulae for the minimum

contents of stabilizing elements are used that relate to rhecarbon content in the steel and that ensure its resistance to

intercrystalline corrosion. Most of them are for steels stabilized

with titanium and these are mostly used also in this country. It

was found, however, that the addition of a stabilizing element in

relation to the carbon content, especially if it approaches the

theoretical value, is quite insufficient if the stabilized steels

are subjected to unfavorable heat loads, for example, in welding.

The carbon bond to the stabilizing element is disrupted at the

overheated sites, and the carbon is capable again of binding

chromium in the critical temperature region and of developing a

tendency of stabilized steels toward intercrystalline corrosion.

Exactly in these steels the grain size is an important factor

which determines secondarily whether a steel is suitable from the

point of view of intercrystalline corrosion. Once the relation

expressing the effect of the nitrogen content in steel, and the

effect of the grain size are established more precisely it will

be possible in many instances to find the resistance of such steels

only on the basis of chemical and metalographical analysis by cal-

culation, or from a diagram, without the need to carry out the

tests for intercrystalline corrosion.

The relations derived for nonstabilized steels may be used to

an advantage also for stabilized steels with such a modification

that for the calculation of the free carbon content we apply the

data for the distribution of the stabilizing element in steel,that is, for carbon, nitrogen and solid solution. Since the val-

ue of tne stabilizipg element in a solid solution is a function

of temperature, time of heating and its amount in the steel, and fsince we do not have as yet a sufficient foundation for its math-

ematical expression it will be advantageous, also considering

future statistical analysis of the studied relations, to substi-

tute this value with the so-called coefficient of effectiveness

f7, which we can assume to be equal to 1 for steels stabilized by

annealing.

Substituting the relation thus obtained for free carbon con-

tent in Eq. (5) we arrive at an expression

-7-

FTD-HT-67-166

0 005. V . (Cr'- 17) +f. Ti 3,43. (% N- 0,001) _ 0,002.(7)

(Ni- 10)- C' 0.,

for steel stabilized by titanium which determines its resistancetoward intercrystalline corrosion, in which is also denrted Tias the total content of titanium in steel, %N as the aaount ofnitr6gen in steel in perceht weight, f as the coefiicAent of ef-fectiveness of the stabilizing element.

Similar relation may be written also for steels stabilizedwith niobium. The above expression-presupposes an analysis ofnitr6gen in the calculation of the free carbon content. If we dont -know its content we can simplify the appropriate portion ofEq. (7),, for example,, to the form f, Ti/4; at the same time, how--ever, the coefficient. f has a considerable effect on the accuracyof the results. If, on the contrary, we know either the percent-age of titanium in the isolated residue or outright the amount oftitan iumbound to carbon and-nitrogen, besides knowing the chem-ical composition of the steel, the value of the free carbon canbe expressed .more accurately,. for example in the second case,without the use of the coefficient of effectiveness. Isolation ofcarbideso.and their analysis,, however, is very demanding and isnot routinely carried out in all laboratories. For this reasonwe are giving several methods of evaluation of free carbon insteel so that laboratories with less equipment can evaluate thestabilized corrosion resistant steels froih the above relations.

CONCLUSION

The decisive significance of this solution of the grain sizeeffect on the tendency of corrosion resistant steels toward inter-crystalline corrosion is in the fact that it allows one to eval-uate steels with medium carbon content which would often be clas-sified as unsuitable and rejected but which can be quite suitableprovided they are fine-grained. Similarly the stabilized steelswhich .would not be suitable by the usual criteria, or vice versa,can be evaluated more accurately by the given process and manyfine-grained steels can be kept in production or improved as faras intercrystalline corrosion is connerned by heat processing(stabilizing annealing). The application of this method is notless significant in the evaluation of welded corrosion resistantsteels especially in the region of the weld in which the grainsize changes due to the temperature effect (for example, thegrowth of the grain in the close proximity to the molten metal).Steel that has in the original -tate a perfect resistance towardintercrystalline corrosion here! becomes often susceptible as thegrain grows and the effectiveness of the stabilizing element de-creases. Exactly these phenomena of intercrystalline corrosion,associated also with "knife" corrosion, can be detected with theusual tests only with gre-at difficulty.

It is apparent that this method cannot completely substituteother ways of control, that is, tests for intercrystalline corro-sion. It will be necessary to check thoroughly the reliability ofthe calculated and nomographically expressed data and mainly tocheck the precision of the constants in the relations on the basis

FTD-HT.67-166

of a larger number of statistically evaluated results. We have,therefore, applied to some of our laboratories which deal withcorrosion resistant steels to help us obtain as much additionaldata as possible.

In the evaluation it is necessary to heed the subjectiveviews in the determining of various quantities, also to consider

local changes in the composition of steel, changes in steel struc-ture, etc.

REFERENCES

1. Svarc, G.L., K voprosu o mezkristallitnoj korrozijinerzavjejuscich stalej stabilizirovannych titanom [Inter-granular Corrosion of Titaniu'°-Stabilized StainlessSteels]. Collection entitled: Korrozija i borba s nej,NIICHIMMAS 12 [Corrosion and Corrosion Prevention,Scientific Research Institute for Chemical Machinery 12],Moscow, 1952, pages 27-47.

2. CSN [Czechoslovak Standard] 420 463: Stanoveni volikostizrna oceli [Determination of the Grain Size of Steel].

3. N.F.A 04-102. Mesure de la grosseur du groin A. M~thodemicrographique, B. Mgthode macrographique, par compa-raison avec des casure sipes [Measurement of GrainSize. A. Micrographic Method; B. Macrographic Method,by Comparison with ... Fracture.

4. ASTME 19 46, Standard Classification of Austenite GrainSize in Steels.

4 5. Gobin, F., Etude statistique de la corrosion intercris-talline des aciers inoxydables du type 18-8 [StatisticalStudy of Intergranular Corrosion of type 18-8 StainlessSteels], Corrosion et Anticorrosion, 9, 1961, No. 4,pages 119-131.

6. Srejder, A.V., 0 nekotorych osobjenostach korrozijiaustenitnoj nerzavejusceJ stali [Certain Peculiaritiesof Corrosion of Austenitic Stainless Steel], Z. prikl.chimiji, 28, 1955, No. 6, pages 608-615.

7. Heger, J.J., "amilton, J.L., Effect of Minor Consti-tuents of the Intergranular Corrosion of AusteniticSteels. "Corrosion," 11, 1955, pages 6t-lot.

8. Pospisil, R., Antikorozni a zaruvzdorne oceli [Anti-corrosion and Heat Resistant Steels], Prague, 1956,240 pages, 239 illustrations.

Received 19 December 1964.

-9-

FKD-HT-67-..66

'I ~I 4

I

Manu-script (Footnote]

No.

1 Dr. Chem. Engrg., State Research Institute for the Pro-tection of Materials named for G.V. Akimov, Prague.

i

.1

FTD-HT-67- 166'

DISTRIBUTION LIST

Organization Nr. Cys. Organization Nr. Cys,

AIR FORCE OTHER DOD AGENCIES

Hq USAF DDC 20OAR (RRYA) I Hq USAREUR 1.AFCRL (CRXLR) 1ARL (ARB) WP,AB 1 Msl Intei Dir RDST 1Hq AFSC (SCFT) 1 U.S. Army (FSTC) 3ASD (ASFS-2) 9 Army Security Agency 1SAMSO (SMFA) 1 Harry Diamond Lab 15"TD NOTS China Lake 1

TDBAE 3 PAC sl Range 1ATD (2) U. S. Navy (STIC/N2EA) 1PIO (1)

TDBAS-2 2TDBID-2 2TDBR 1TDDEI 1 OTHER GOVERNMENT AGENCIESTDP (PHE) 6

* PAD (1) AEC ('Term) 2PEK (1) AEC (Wash) 2PF (4) FAA (SS-1O) 1

NAFEC 1NASA (ATSS-T) 1

DISTRIBUTION TO BE MADE BY

CLA (DIACO-3) 18DIAAP-1C1 (i)DIAAP-1H2 (I)DIAST-A (i)

DIAST-5C (1)U.S. Navy (ONR) (1)

OP922F2 (i)CIA (SD) (5)NSA (CREF/CDB) (6)

FTD-HT-67-166 I*7 --