This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
7/18/2019 Formation of Precipitates in Multiple Microalloyed Pipeline Steels
form 20 July 1988. At the time the work was carried out the authors were in the
Department of Mechanical and Production Engineering, Aston University,
Birmingham. Dr Billington is now a Consultant Metallurgist.
Experimental procedure
The worldwide increase in the demand for energy has led toa continuous increase in the production of steel pipelinesfor the transportation of oil and gas from production site to potential user. This has led to the demand for thinner walled, large diameter pipes which can resist hostileenvironments while increasing their fuel carrying capacitiesto meqt the economic restrictions imposed. These advanceshave~been made possible by the development of high
strength low alloy (HSLA) steels which form the basisof the materials used together with improvements in thethermomechanical treatments applied to achieve the final
properties.Microalloying additions, such as niobium, titanium, and
vanadium which, in the presence of C and nitrogen, achievethe control of austenite grains (and, subsequently, ferritegrains) and precipitation strengthening of the ferrite bycontrolled formation of nitrides/carbides. For improved grain refinement, controlled rolling was introduced to
permit the use of much lower finish rolling temperatures.Thus, modern pipeline steels are produced having relativelysmall ferrite grains, formed by the transformation of theoriginal austenite grains. Final properties can be achieved
by the precipitation of vanadium (and niobium) carbo-nitrides below 900°C improving strength and toughness.
Knowledge of the interactions between the microalloyingelements and aluminium is far from complete, especially interms of their dependence upon processing variables. The
present work was undertaken to elucidate the sequence of formation of precipitates during the rolling of a series of steels containing fixed carbon and niobium contents withvarying (aluminium), titanium, and vanadium concentra-tions. The precipitates formed in as cast specimens and specimens quenched from 1250°C with and without defor-mation were examined.
Five 18 kg pipeline steel ingots were produced as described previouslyl and their compositions are given in Table 1.The ingots were hot rolled in four passes, with interpassreheating, at 1250°C to give an overall reduction of 66%and a final thickness of 11± O'2 mm.
ELECTRON MICROSCOPY
A scanning transmission electron microscope (STEM) withEDX attachment (Philips Model EM 400) was used tostudy the morphology and composition of the precipitates.Conventional carbon extraction replicas were prepared from selected specimens which were etched in 2% nital before being carbon coated, scored, and stripped in 5%nital. Individual precipitates were quantitatively analysed in the Philips microscope for 200 live seconds and theelemental counts corrected for atomic numbers were sub-sequently processed through a computer to give the analy-ses presented in Tables 2-4. Because the objective of thiswork was to study the morphology and chemistry of the precipitates, they were classified into three sizes only:coarse, intermediate, and fine.
To elucidate further the production of precipitates and their compositions during the various thermo mechanicaltreatments applied to the steels, a series of rolling and quenching experiments was made on steels 1 and 4. Twosamples from each cast steel (initial thickness 18 mm) wereheated to 1250°C, one sample was quenched after soakingand the other was hot rolled in a laboratory two-high millto give 17% reduction before quenching. The emergenttemperature of the specimens from the rolls was1150± 10°C. Specimens from the"sematerials were prepared for electron microscopy. The remaining materials werereheated at 1250°C and given a second pass of 23%
Table 1 Compositions of steels used in present investigation, wt-%
P u b l i s h e d b y M a n e y P u b l i s h i n g ( c ) I O M C
o m m u n i c a t i o n s L t d
568 Emenike and Billington Formation of precipitates in pipeline steels
a
b
a spherical precipitates in steel 1; bvarious morphologies in steel 4
(massive cuboids are (NbTi) rich particles)
Precipitate morphologies observed in steels 1 and 4
(carbon extraction replicas)
would form titanium carbide (in the presence of carbon)with a solvus below that of titanium nitride, but approxi-mately the same as that of niobium carbide, at equivalentlow activities of 0·01 referred to the pure solid standard state (see Appendix). It should be pointed out that thealuminium content of the precipitates in the as rolled steels4 and 5 was insensitive to size. However, aluminium nitrideformation and precipitation is the subject of another paper 1
and therefore will not be discussed here.
The distribution of niobium between fine and coarse precipitates in the as rolled steels 2 and 3 in Table 3suggests that the niobium containing particles formed during the initial stages of precipitation, i.e. at the higher
temperatures, had a high probability of nucleating on tita-nium nitride particles. These high temperature precipitateswould be more vulnerable to particle growth and, in fact,the presence of aluminium in these precipitates has beensuggested to accelerate coarsening.8 The fine particles weredevoid of aluminium and probably formed during the later stages of precipitation and tend to support this claim. This
explanation might be advanced also for the behaviour of niobium in steels 4 and 5 (Table 3), but, in addition, theniobium was distributed apparently equally between fineand coarse precipitates. These assertions have beensubstantiated further in the Appendix.
The affinity of vanadium for small particles can be under-stood from the calculations summarised in the Appendixwhich indicate the bulk of its precipitation as carbide in theferrite. Such late forming particles were likely to remain fine
because of the improbability of Ostwald ripening.Generally, it can be stated that the original precipitates
and/or those out of solution at the soaking temperaturehad a greater tendency to coarsen and this was accentuated
by contamination (e.g. the presence of aluminium in tita-
nium nitride). This is partially supported by the analysis inthe Appendix which indicates that the titanium nitridesolvus was higher than the soaking temperature of l250°C,approaching a maximum value for steel 4 corresponding toa maximum of titanium-nitrogen solubility product.
The strong relationship between the micro alloyingelements in the steels and the precipitates is reflected in the
parameter Kh
the precipitate yield quotient, defined as theratio of the weight percentage of a micro alloying element inthe precipitates to that in the steel. Evidently, K 1 is ameasure of the participation of the microalloying elementsin precipitate formation. The values of K 1 are plotted as bar charts in Figs. 2-4. Of great interest are the high values of K 1 ~ 8000 for the lowest titanium containing steel (0'007%),especially in the quenched condition and shown in Figs. 4a
and b, compared with that of 0·074%Ti (Fig. 4d ) which hasa value of K 1 ~ lOOO.This high degree of participation in precipitation exhibited by the former may provide anexplanation for the advocation of low titanium additions~O'Ol% for elevated temperature austenite grain pinning.This dimensionless parameter which can be applied to anyreaction that obeys the solubility relationship givenll,12could therefore prove to be an important design require-ment, because it provides information concerning thechemical identity of the alloy additions and particles.
EVOLUTION OF PRECIPITATE ANALYSES AND
SIZESIn Table 4 are given the analyses of complex precipitatesobtained in steel 1 after soaking at 1250°C and quenching.They portray fine, dense, spherical, titanium rich particles
f in e in te rm e d ia te in te rm e d ia tel a ) I b ) Id )
2 Variation of K. for steels 1, 2, and 4 in as cast condition
2 0 0 0
1 8 0 0 S tee l S tee l1 2
1 6 0 0
1 4 0 0
1 2 0 0
~1 0 0 0
8 0 0
6 0 0 N b Ii N b Ii N b Ii N b A l
4 0 0
2 0 0A l A l
Ii
N b
Ii
fin eIe)
S tee l4
A l
N b
Ii
A l
in term ed iateI f )
Materials Science and Technology June 1989 Vol. 5
7/18/2019 Formation of Precipitates in Multiple Microalloyed Pipeline Steels
P u b l i s h e d b y M a n e y P u b l i s h i n g ( c ) I O M C
o m m u n i c a t i o n s L t d
Emenike and Billington Formation of precipitates in pipeline steels 569
2000S te e l S te e l S te e l S teel
1800 1 2 2 4
1600
1400
~ - 1 20 0N b N b N b
1000
8 00 .
60 0 T i T i N b A I N b A I N b N b N b
40 0 T i T i T i T i
20 0 A IA I T i T i A IA I A I A I
f in e in t e rm ed ia t e f in e c o a rs e c o a rs e f in e c o a rs e c o a r s e(a ) ( b ) (c ) ( d ) (e ) (f ) ( g) ( h )
3 Variation of K1 for steels 1, 2, and 4 in as rolled condition
(70 at.-%Ti and 30 at.-%Nb). This analysis contrasts strik-
ingly with those of the as cast and the as rolled specimens of the same steel (see Tables 2 and 3) in which the niobiumand titanium values are reversed. These precipitates areshown in Fig. 5a.
After 17% deformation followed by quenching from the
rolls, a new set of fine, spherical, but less dense, precipitatesemerged and the resulting microstructure contained amixture of these precipitates and titanium rich precipitates.The new precipitates were niobium rich (about 70 at.-%Nband 30 at.-%Ti) and their chemistry was close to that of the
precipitates in the as cast steel 1 given in Table 2. These particles a"re shown in Fig. 5b . The increased number of precipitates can be seen from Fig.5b, thus confirming thework of Yamamoto et al.13 Therefore, it is probable that, as
well as increasing the population of precipitates, rollingaccelerated the precipitation of niobium rich particles in thetemperature range 1250-1150°C. Even the lower limit ismuch greater than the finish precipitation temperature of
NbC or Nb(C,N)oog7 (see Table 5).
Steel 4, also quenched after soaking at 1250°C, contained
titanium rich precipitates (about 83 at.-%Ti and 16 at.-%Nb) which deviated markedly from those obtained in the as cast or as rolled material. The quenched particlesare shown in Fig. 6.
The effect of a second thermomechanical deformation
and quenching cycle for steel 1 on the precipitate size and distribution is shown in Figs.7a and b. By comparingFig. 7a with Fig. 5a, it can be inferred that the intermediateheat treatment nullified the previous rolling. By comparing
Fig. 7b with Fig. 5b it can be deduced that titanium nitrideor titanium-niobium nitride grew slowly, hence, there wasno major difference in particle size and in Fig. 7b the
presence of more dense spherical precipitates can beobserved, suggesting that the intermediate reheating did
not completely redissolve all the prior niobium rich precipi-tates. In both cases, the distribution of the precipitatesappeared to be even with some of the particles pinningthe prior austenite grain boundaries or lath martensite
boundaries.
a after soaking at 1250°C, WQ; b, C after soaking at 1250°C, rolling
(17%), WQ; d after soaking at 1250°C, wa
4 Variation of K1 for steels 1 and 4 after water quench
from 1250°C
f i n e , d e n s e(a)
8000
7 5 00
7 000
65 00
6 000
5 000
~ - 2 0 00
1800
1600
1400
1200
1000
80 06 00 N b
40 0
20 0
T i
A I
S teel
Ii
N b
A I
f in e , d en se( b J
N b T i
A I
f in e , l e s s d e ns e(c)
S te el 4
N b T i
A I
c o a r s e(d)
a
b
a after soaking at 1250°C, wa, insoluble dense precipitates and less
dense soluble precipitates can be observed; b after soaking at
1250°C, rolling (17%), wa, increase in number of dense precipitates
can be observed
5 Microstructure of steel 1 after different heat
treatments {carbon extraction replicas}
Materials Science and Technology June 1989 Vol. 5
7/18/2019 Formation of Precipitates in Multiple Microalloyed Pipeline Steels
Nb(C,N)o'87 0·0625 1812 0·5 1124 0·0416 37'4 Nb 0·048 53·8 55 AIN 0·0930 1812 0·03 724 0·0064 5·8 AI 0·0042 4·7 1·0
TiC 0·1 1212 0·01 732 0·0274 24·6
NbC 0·1 1210 0·003 718 0·0163 14·7
(Fe3C 1·0 720 1'1834)
Total 0·1112
* Mean of 'as rolled' precipitates.
Discussion
MORPHOLOGIES
Most of the morphologies observed in Figs. 5-7 may be
explained in terms of the compositions of the steels asevident from Tables 2 and 3 and Figs. la and b. Thisopinion is supported by the observation that steel 1displayed predominantly spherical precipitates virtuallydevoid of aluminium. Other shapes appeared in the struc-tures when titanium and/or aluminium-nitrogen solubility
products were relatively higher in the alloys (steels 2-5). Inthe present work, the thermomechanical treatments weresimilar and therefore could not throw any further light onthis proposition.
Materials Science and Technology June 1989 Vol. 5
ANALYSES
The isomorphology between most of the precipitates leadsto intersolubilities and could facilitate the formation of carbonitrides. Similarities in the atomic sizes of carbon and nitrogen and in the magnitudes of their diffusivities can be
conducive to the interchangeability of carbon and nitrogenwithin the non-stoichiometric lattices of these complex
precipitates (except for aluminium nitride). In addition tointerstitial diffusion which may take place, the existence of non-stoichiometry creates vacancies with associated enhanced diffusion, possibly increased by mutual strainingof precipitate lattices. Such changes in lattice parameter with precipitate composition have been documented earlier.14 A precipitation model based on such mixed com-
positions has been attempted,7 but cannot be adapted to
7/18/2019 Formation of Precipitates in Multiple Microalloyed Pipeline Steels
P u b l i s h e d b y M a n e y P u b l i s h i n g ( c ) I O M C
o m m u n i c a t i o n s L t d
572 Emenike and Billington Formation of precipitates in pipeline steels
(iii) some precipitates may not have been detected
(iv) premature termination of the VC precipitation
reaction as a result of the onset of Fe3C formation.
The consistently maximum difference exhibited by the
vanadium containing steels (i.e. steels 2 and 3), the magni-
tude of which increased with increasing vanadium content
of the steels, may be explained by (iii) and (iv) above. It can
be inferred from this discrepancy between the values that
the participation in higher temperature precipitation still
left excess vanadium for low temperature precipitation.
EVOLUTION OF PRECIPITATE COMPOSITIONS AND SIZES
The analyses of the precipitates of steel 1 quenched from
1250°C reflected a stoichiometry of Tio .7 Nbo .3(NjC). These
complex high temperature precipitates were likely to be rich
in nitrogen and low in carbon, consistent with the accepted
view. Further, the solvus of the complex precipitate was
probably higher than 1250°C, because these precipitates
were insoluble at 1250°C.
The precipitates which were produced during or
immediately after rolling, but before quenching, i.e. thosehaving a stoichiometry Tio.3 Nbo.7(NjC), must have h ad a
solvus below 1250°C. They were likely to be low in nitrogen
and rich in carbon, which could only have been confirmed
by the use of electron energy loss spectroscopy (EELS),
which was beyond the scope of the present investigation.
The driving force for these niobium rich precipitates was
likely to be the difference in free energy over the tempera-
ture gradient within the roll gap, because of the recrystalli-
sation of the austenite at this temperature reported
previously21.22 and confirmed in the present work. A
similar evolution of complex particles should be exhibited
by steel 4 given the same thermomechanical treatment, i.e.
the transition from titanium rich to niobium dominant
particles.
The evolution of the observed spectrum of particle sizes
was due to the propensity for precipitate coarsening by the
well known Ostwald mechanism which would have a
decreasing effect in the order:
(i) insoluble precipitates at the soaking temperature
(ii) precipitates formed at high temperatures and during
rolling, e.g. titanium nitride or Ti,Nb(NjC)
(iii) low temperature precipitates, e.g. vanadium carbide
or nitride.
The presence of niobium in the titanium rich precipitates at
1250°C suggests that a transport phenomenon is involved
in the formation of complex precipitates, thus supporting
model 2 of Houghton et al.7
That model is based on themechanism of mixing of phases in the thermodynamic
treatment, but becomes very complex for multiple micro-
alloyed steels, because of their multicomponent nature.
From the evidence of insoluble titanium rich precipitates
at 1250°C for steels of very low (0·007%) titanium content
(see Fig. 3), it can be inferred that these particles,
Ti,Nb(NjC), were available for impeding grain boundary
motion at the soaking temperature (i.e. Zener's concept).
The usefulness of such high temperature austenite pinning,
especially during soaking or welding, has given great
impetus to the practice of making low titanium additions to
steel.
Summary
1. The precipitates in multiplemicroalloyed steels are
complex.
2. A parameter K., precipitate yield quotient, which is
indicative of solute participation in precipitation pheno-
menon has been established.
Materials Science and Technology June 1989 Vol. 5
3. Specimens quenched from 1250°C were found to
contain titanium rich particles.
4. Hot rolling increased the population of precipitates
and effected a transition in their chemistry.
5. Microalloying elements showed a tendency to form
precipitates differing in size. Titanium and aluminium
appeared in the coarse precipitates, but vanadium was
detected chiefly in the fine particles. Niobium showed no
particular preference, but had a dominant chemical effect
on the other micro alloying elements.
6. Steel composition dictated the morphology of the
precipitates for the same casting and thermomechanical
conditions.
Acknowledgments
The authors acknowledge financial support by the Nigerian
Government to one of the authors (COlE) and the provi-
sion of research facilities by Aston University, Birmingham.They would like also to thank Professor M. H. Loretto,
University of Birmingham. Useful suggestions by Dr F. B.
Pickering, Sheffield City Polytechnic, were appreciated.
References
1. c. o. 1. EMENIKE and J. C. BILLINGTON: Mater. Sci. Technol.,
1989, 5, (5), 450-456.
2. c. o. 1. EMENIKE: PhD thesis, University of Aston, Birmingham,1987.
3. I. WEISS, G. L. FITZSIMONS, K. MIELITYINEN-TITTO, and A. J.
DeARDO: in 'Thermomechanical processing of microalloyed austenite' (Proc. Conf), (ed. A. J. DeArdo et al.), 33; 1981,
Pittsburgh, PA, ASTM.4. M. J. WHITE and w. S. OWEN: Me tall. Trans., 1980, l1A, 597.
5. w. ROBERTS: 'HSLA steels, technology and applications' (Proc.Conf), (ed. M. Korchynsky), 33; 1983, Philadelphia, PA,ASTM.
6. T. SIWECKI, A. SANDBERG, W. ROBERTS, and R. LAGNEBORG: in'Thermomechanical processing of microalloyed austenite'(Proc. Conf), (ed. A. J. DeArdo et al.), 163; 1981, Pittsburgh,PA, ASTM.
7. D. C. HOUGHTON, G. C. WEATHERLY, and J. D. EMBURY: in'Thermomechanical processing of microalloyed austenite'(Proc. Conf), (ed. A. J. DeArdo et al.), 267; 1981, Pittsburgh,PA, ASTM.
8. B. LOBERG, A. NORDEN, J. STRID, and K. E. EASTERLING: Metall.Trans., 1984, 15A, 33.
9. T. SIWECKI, A. SANDBERG, and W. ROBERTS: 'HSLA steels,technology and applications' (Proc. Conf), (ed. M.Korchynsky), 19; 1983, Philadelphia, PA, ASTM.
10. Z. CHEN, M. H. LORETTO, and R. C. COCHRANE: Mater. Sci.
Techno!., 1987, 3, 836.
11. K. NARITA: Trans. Iron Steel Inst. Jpn, 1975, 15, 147.
12. K. J. IRVINE, F. B. PICKERING, and T. GLADMAN: J. Iron Steel
Inst., 1967, 205, 161.
13. s. YAMAMOTO, C. OUCHI, and T. OSUKA: in 'Thermomechanical processing of microalloyed austenite' (Proc. Conf), (ed. A. J.DeArdo et al.), 613; 1981, Pittsburgh, PA, ASTM.
14. H. J. GOLDSCHMIDT: 'Interstitial alloys'; 1967, London Butter-worth.
15. R. LAGNEBORG: Scand. J. Metall., 1985, 14, 289.16. N. K. BALLIGER and R. w. K. HONEYCOMBE: Met. Sci., 1980, 14,
121.
17. A. P. COLDREN, v. BLISS, and T. G. OAKWOOD: in 'Thermo-mechanical processing of micro alloyed austenite' (Proc.Conf), (ed. A. J. DeArdo et al.), 591; 1981, Pittsburgh, PA,ASTM.
18. T. N. BAKER and R. L. REUBEN: 'Advances in the physicalmetallurgy and applications of steels', 213; 1982, London, TheMetals Society.
19. Y. C. HIRSCH and B. A. PARKER: 'Advances in physical metal-
7/18/2019 Formation of Precipitates in Multiple Microalloyed Pipeline Steels
u b l i s h e d b y M a n e y P u b l i s h i n g ( c ) I O M C
o m m u n i c a t i o n s L t d
Emenike and Billington Formation of precipitates in pipeline steels 573
lurgy and applications of steels', 26; 1982, London, The Metals
Society.20. F. B. PICKERING: personal communication, Sheffield City
Polytechnic, 1987.21. R. K. AMIN and F. B. PICKERING: in 'Thermomechanical
processing of microalloyed austenite' (Proc. Con f.), (ed. A. J.DeArdo et at.), 1; 1981, Pittsburgh, PA, ASTM.
22. R. K. AMIN and F. B. PICKERING: in 'Thermomechanical processing of microalloyed austenite' (Proc. Conf.), (ed. A. J.DeArdo et at.), 377; 1981, Pittsburgh, PA, ASTM.
Appendix
Modelling of sequence of precipitation inmultiple microalloyed steels
TiN precipitation starts at 2162 K (l889°C). When almostall the nitrogen has been removed, i.e. N ~ 0 (say10-6 wt-%), the titanium will have been reduced to 0'0397%(stoichiometric ratio (SR) of Ti : N = 3'43) and the end of
TiN precipitation is given by:
log (0'0397)(10-6) = -15 200/1finish+ 3·9
where 1finish= 1345 K (l072°C). In the event of a solid solution with another compound being formed, the activityof TiN may be reduced relative to the pure solid compound. For QTiN= 0·1 the end of precipitation would be 1203°C with Ti = 0'0397% and N = 10-6%, for bothfinish temperatures and the quantity of TiN produced
would be 0'0443%.
Niobium carbonitride formation
where QXY is the Raoultian activity of the compound referred to the pure solid XV. In the absence of any
practical data on the activities in the mixtures presumed to form, assumed activities are given in the followingcalculations.
where X and Yare the dissolved microalloying elementsand A and B are constants, T is the absolute temperature,and it is assumed that the pure solid compound XY isformed. When the chemical potential is reduced by theformation of solid solutions, equation (2) becomes
Limited attempts to achieve this have been made,?,23 but
none incorporated multiple microalloying. Consequently,the present analysis is an extension and a modification tothe approach of Houghton et al.7 An example of the presentmethod is given below.
Alloy 4 comprised the following: 0'09C, 0'059AI,0'043Nb, 0'074Ti, 0'00430, 0'010N (see Table 1). It wasassumed that the order of precipitate formation in this steel
would be: TiN, Nb(C,N)o.87' AIN, NbC, TiC, Fe3C. It isfurther assumed that mixtures of these compounds maycoprecipitate forming mutual solid solutions with a result-ant decrease in chemical potentials of the individualcompounds referred to the pure solid state. The chemical
potential fli is defined as
fli = flo + R T In Qi (1)
where flo is the chemical potential of the pure solid state, Qiis the activity of the compound referred to the pure solid substance, R is the gas constant, and T is the absoluteternpera ture.
Most of the data in the literature on the formation of precipitates in steels use Henrian activities for the alloyingelements in steel and the standard state chosen is the oneweight percentage (when ax = 1 at 1 wt-%). Thus, the datais in the form
log (%X%Y)= -A/T+B
log (%X%Y)/QXY = - A/T +B
(2)
(3)
It is assumed that any niobium carbonitride to be formed would be cubic l5-Nb(C,N)o'87 (Ref. 24) varying in com- position from nitrogen rich to high carbon with decreasingnitrogen in the steel. The C-curve can be represented by theequation
log (wt-%Nb) + 0·87log (wt-%C
+12/14wt-%N)=-6770/T+2'26 ..... (5)
assuming QNb(C,N)o'87= 1 for pure solid carbonitride. When Nb = 0'043% and N = 0'010% with C = 0'09%, ~tart for Nb(C,N)o.87 is 1169°C. However, if the Nb(C,N)o.87 should be dissolved in TiN to form a complex mixture or solid solution, its chemical potential would be reduced and itsformation temperature would increase. For example, for thecarbonitride to be coprecipitated with TiN at 1889°C,its activity would have to be 0·043 relative to the
pure solid Nb(C,N)o'87' assuming 0'043%Nb, 0'010%N, and 0·09%C. Thus, it is possible that the titanium nitride and niobium carbonitride could coprecipitate at temperatures~ 1889°C. It can be shown that the end of TiN precipi-tation can occur at 1072 or 1203°C, depending upon itsactivity (1'0 or 0'1, respectively, referred to the pure solid compound) with the nitrogen content being 10-6% and titanium at 0'03972%. However, during coprecipitationwith Nb(C,N)o'87 some nitrogen would react to produce thecarbonitride. From the data presented in Tables 2-4, it isnot possible to calculate how much nitrogen is used to
produce nitride or carbonitride and therefore it is necessaryto assume a partition of nitrogen between the compoundsto obtain their individual completion temperatures. Twovalues for the partition of nitrogen going to form TiN areassumed in the following calculations: 70 and 50%. Further,the final mixture of compounds or solid solutions will causea decrease in the chemical potentials of the individualcompounds which cannot be deduced from data availablein the literature. Therefore, to illustrate the effect of different activities of the compounds (referred to their puresolid states), it is assumed that for both the 70 and 50%nitrogen used by titanium the activities of TiN in the
product should be 0·7 and 0·5 and, for Nb(C,N)o.87' the
TiN Formation
Using Narita's derived solid data 11 for pure solid TiN (withTin K)
log(wt-%Ti)(wt-%N)=-15200/T+3'9 .... (4)
i.e.
log (0'074)(0'010)= -15 200/~tart + 3·9
Table 6 Final precipitation temperatures for TiN and
Nb(C,N)o'87 formation
Finish temperature, °C
N u s ed b y T i , w t -% 8riN aNb(C,NlO'87 TiN Nb(C,N)o'87
u b l i s h e d b y M a n e y P u b l i s h i n g ( c ) I O M C
o m m u n i c a t i o n s L t d
574 Emenike and Billington Formation of precipitates in pipeline steels
corresponding activities should be 0·3 and 0·5. The end of precipitation as represented by equilibrium finishingtemperatures is given in Table 6. It seems that changes innitrogen usage by titanium have little effect upon the final
precipitation temperature of titanium nitride, but signifi-cantly change the temperatures for completion of carbo-nitride precipitation. Nevertheless, it does appear that
coprecipitation is a valid concept for these two compounds.
Niobium carbide and titanium carbideformation
At the end of carbonitride formation the carbon content of the steel would be 0·09- 0·00343 = 0·0866%, the titanium0,0603%, and the niobium 0·0125%.
The formation of pure titanium carbide to be coprecipi-tated is given byll
log (%Ti)(%C)=-10475/T + 5·33 (7 )
Aluminium nitride formation
Another compound which could be formed from thismicro alloyed steel is aluminium nitride (AIN) the C-curveof which can be represented by12
The initial weight per cent of aluminium is given by the
analysed %AI (from Table 1)minus the %AI associated withoxygen in the form of alumina, e.g. for steel 4
0·059- 54/48 x 0·0043=0·0542%AI
which, with 0·010%N, gives ~tart(AIN)=1303°C.For coprecipitation with TiN at 1889°C, the activity of
AIN referred to the pure solid compound would have to be0·0686 and it is assumed that this is probable. However,unlike c5-Nb(C,N)o.87'aluminium nitride has a close packed hexagonal crystal structure and therefore is not isomor- phous with titanium nitride. Consequently, coprecipitationmay not be easy to achieve and this may lead to lower aluminium in the final precipitate than the equilibrium
conditions predict. Nevertheless, some nitrogen would beexpected to react with aluminium to form AIN. If this wereassumed to be 20% and the final a A1N were 0·05, the final
precipitation temperature would be given by
log (0.0503)(10-6)/0.05 =-6770/Tfinish+ 1·03
Tfinish= 690°C, leaving 0·0503%AI and producing0·0059%AIN.
Summarising the coprecipitation' of TiN, Nb(C,N)o'87'and AIN, it follows the initial precipitation of pure TiN at1889°C, and finishes (it is assumed) when nitrogen decreasesto 10-6%. Assuming the final activities of TiN, Nb(C,N)o'87'and AIN are 0·5, 0·4, and 0·05, respectively, and the nitrogen
is divided 40% as TiN, 40% as Nb(C,N)o.87' and 20% as
AIN, the completion temperatures would be 1132, 1235,and 690°C respectively (a theoretical estimate for AIN, butit is probable that it will have stopped precipitating athigher temperatures as a result of the lower nitrogen and kinetic problems at lower temperatures). The weight per-centages of the precipitates formed would be 0·0137%TiN,
0·0378%Nb(C,N)o'87' and 0·0059%AIN leaving 0·0603%Ti,0·0125%Nb, and 0·0503%AI in the steel.
log (wt-%AI)(wt-%N) = -6770/T + 1·03 (6 )
~tart(TiC)=1103°C. For titanium carbide to be coprecipi-tated with its nitride at 1889°C, its activity would have to
be only 0·00171 referred to pure solid TiC and therefore it
is assumed that it is unlikely to coprecipitate at such a hightemperature, although it has a cubic crystal structure.However, when its activity is decreased to only 0·01, it is
probable that it can form a significant part of the final precipitate at 1593°C. When the titanium content reaches10-6% and the carbon 0·07155%, the end of TiC precipi-tation will be almost complete at 567°C at an activity of 1
or at 727°C for an activity of 0·01 (the total weight would be 0·0754 g per 100 g of steel).
Using the same data,! 1 pure NbC solubility product datais given by
10g(%Nb)(%C) = -7900/T+3·42 . . . . . . (8)
and when Nb =0·0125% and C =0·0866%, ~tart=964°Cat a NbC activity of 1 or 1615°C at an activity of 0·01referred to pure solid NbC.
Therefore, it would seem that the carbides of titaniumand niobium could coprecipitate with the nitrides and carbonitride at temperatures < 1500°C to produce a very
mixed precipitate of titanium, niobium, aluminium,nitrogen, and carbon which would continue until the }'~ (J .
transformation in the steel at 720°C. The weight of niobiumcarbide formed would be 0·014% leaving 0·0700% carbonwhich would form 1·05% of cementite in the steel. Summar-ies of the equilibrium weights of precipitates and their temperatures of formation at the assumed activities aregiven in Table 5 for steels 1-5. These calculations haveconfirmed that TiN forms in the liquid iron, but the veryhigh temperature of formation of precipitates in steel 4(1889°C) brings into question the validity of Narita's solid data 11when titanium-nitrogen solubility product exceeds acertain value. Further research is required in this area.
References
23. s. R. KEOWN and w. G. WILSON: in 'Thermomechanical processing of micro alloyed austenite' (Proc. Conf.), (ed. A. J.DeArdo et al.), 319; 1981, Pittsburgh, PA, ASTM.
24. R. C. SHARMA, V. K. LAKSHMANAM, and J. s. KIRKALDY: Metall.