• Aucun résultat trouvé

EARLY STAGE DECOMPOSITION OF Ni-36at%Cu-9at%Al AN ATOM-PROBE FIM STUDY

N/A
N/A
Protected

Academic year: 2021

Partager "EARLY STAGE DECOMPOSITION OF Ni-36at%Cu-9at%Al AN ATOM-PROBE FIM STUDY"

Copied!
7
0
0

Texte intégral

(1)

HAL Id: jpa-00224461

https://hal.archives-ouvertes.fr/jpa-00224461

Submitted on 1 Jan 1984

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

EARLY STAGE DECOMPOSITION OF

Ni-36at%Cu-9at%Al AN ATOM-PROBE FIM STUDY

Z. Liu, R. Wagner

To cite this version:

Z. Liu, R. Wagner. EARLY STAGE DECOMPOSITION OF Ni-36at%Cu-9at%Al AN ATOM- PROBE FIM STUDY. Journal de Physique Colloques, 1984, 45 (C9), pp.C9-441-C9-446.

�10.1051/jphyscol:1984973�. �jpa-00224461�

(2)

JOURNAL DE PHYSIQUE

Colloque C9, supplément au n012, Tome 45, décembre 1984 page C9-441

EARLY STAGE DECOMPOSITION OF N i - 3 6 a t % C u - 9 a t % A l AN ATOM-PROBE F I M STUDY

Z.G. Liu +++ and R. wagnert

+ ~ n s t i t u t e for Meta2 Phÿsies, University of Gottingen, F.R.G.

+ ~ ~ ~ ~ - ~ ~ s e a r c h Centre, Geesthacht, F. R. G.

*+~onderforschun~sbereich 126, Gottingen-CZausthaZ, F. R. G.

Résumé

-

Les analyses par sonde à atomes/FIM indiquent que la séparation de phase dans Ni-36 at.% Cu-9 at.% Al à T = 500, 540 et 580°C se font via nucléation et croissance de précipités Y ' - ( N ~ ~ C U ~ - ~ ) A ~ . L'analyse de la cinétique de précipitation en terme de modèle de précipitation de Kampmann et Wagner permet d'accéder aux constantes effectives de diffusions, D(T), à l'énergie interfaciale U = (0.052 f 0.002)~/m' et à l'énergie de formation SG': (T)

.

Abstract - Atom-probe FIM analyses indicate that phase separation in Ni-36at%Cu-9at%AL at T=500, 540, and 580 OC occurs via nucleation and growth of y'-(Ni,Cul-,) 3 Alprecipitates. Analysis of the precipitation kinetics in terms of the precipitation mode1 of Kampmann and Wagner allows the effective diffusion constants, D ( T ) , the interfacial energy, 0 = (0.052 f 0.002) ~ / m ~ , and the nucleus formation energy, 6G* ( T ) to be obtained.

1 - INTRODUCTION

According to the phase diagram /1/ supersaturated Ni-Cu alloys with ternary addi- tions of 5 to 10 at% Al decompose during aging between 500 and 600 OC into an Al- depleted matrix and an Al-rich phase (y'). Transmission electron microscopic (TEM) studies of the l a t e r stages of phase separation, have led to the suggestion that the y'-particles evolve £rom a spinodal decomposition reaction /2/;.-this would mean that the Al concentration of the incipient y'-precipitates increases contin- uously with aging time £rom virtually zero to about 25 at%.

An investigation of the v e r y e a r i y stages of decomposition in ~i-Cu-Al, a system in which we have been interested for some time /3, 4/, requires i) the detection of extremely fine solute-enriched regions (radii R 5 1 nm), and ii) the measure- ment of their composition as a function of aging t h e . These requirements, which cannot be met by analytical T m , prompted us to employ the atom-probe FIM.

II - EXPERIMENTAL

Prior to the polishing of the field ion tips (in a solution of 1 g NaCr03. 4H20 and 10 ml acetic acid at about 8 to 20 VDC) from 0.2 mm thick wires, the alloy was homogenized (1 h at 1020 OC in an argon atmosphere) and quenched into brine. Some wires were subsequently isothermally aged at 500, 540, and 580 OC for periods (tA) ranging between 1 and 12180 min. The specimen axis was either parallel to <Ill>

(30 % of al1 tips) or parallel to <100> (20 $ 1 .

The FIM-specimens were imaged at 80 K using neon (5

.

10-~ m bar) as imaging gas. The

++ On leave £rom Nanjing University, P.R. China

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:1984973

(3)

C9-442 JOURNAL DE PHYSIQUE

microstructure of the decomposing alloys and the time evolution of the composition of the precipitating phase was analyzed by three different techniques:

-i) Determination of the size distribution, morphology and number density of the precipitated particles, which appear in bright contrast (e.g. Fig. l), directly from the field ion micrographs employing the persistence size technique /e.g. 5/.

During the early stages of precipitation, the particle number density was too high for individual particles to be recognized separately on the field ion micrographs.

Under these conditions the various precipitation parameters had to be determined by employing technique iii).

-ii) Selected area atom-probe analyses of the composition of individual (visible) precipitates. This mode was also used for analysis of the composition profiles

across the precipitate/matrix interphase boundary.

-iii) Recording of composition profiles with the atom-probe along a <Ill> or <100>

direction into depths ranging between 200 and 1000 field evaporated atomic layers, i.e. between -40 and -200 nm (dlll=0.207 nm). With the known diameter of the atom- probe aperture (2.5 nm) and the known probing depth, the particle number density

(Nv) could be determined directly £rom the composition profiles. However, because of the small volume probed in this type of analysis (in general, only -10 to "20 particles were recorded), at least in the later stages of decomposition Nv

-

and,

hence, the precipitated volume fraction and the remaining supersaturation - could be determined more accurately with technique i) where a considerably larger volume

(containing -100 precipitates) is sampled.

The composition profiles were subjected to autocorrelation analyses from which the homogeneity of the solution-treated specimens, the mean diameter (2 R ) of the pre- cipitates, and the mean interparticle distance could-be directly determined /5, 6/.

For al1 variously heat treated specimens, for which R could be measured with both techniques ( i) and iii) ) , agreement was found within IO % /4/.

Al1 analyses were carried out in the atom-probe FIM described earlier /5/. Atom- probe analyses of homogenized specimens which were carried out with a VP/Vnc ratio larger than 0.15 (Vp: field evaporation pulse amplitude; VDC: imaging voltage) yield- ed good agreement with wet chemical analyses; for vP/Vnc C0.15, preferential field evaporation of copper atoms led to an apparent Cu concentration which was found to be too low, and which decreased further when still smaller values of VP/VnC were chosen /4/.

III - RESULTS AND DISCUSSIOJ

Autocorrelation analyses of composition profiles of solution-treated specimens re- vealed the alloy to be homogeneous. However, already after aging for only 1 min/5800C or 3 min/540 OC or 10 mi>/500 OC, nucleation was suf ficient to produce a large den- sity of brightly imaging precipitates (Fig. 1).- As expected from classical nuclea- tion theory, the mean radius of the particles ( R ) which, during the nucleation peri- od, is close to the critical radius (R*) of the nuclei /7/, decreases with decreas- ing temperature, i.e. with increasing supersaturation of the alloy (Fig. 2a). Cor- respondingly, the nucleation rate (J) is the higher the larger the supersaturation;

hence, the maximum number density of precipitates (Nvmax) increases with decreasing temperature (Fig. 2b); as the kinetics are more sluggish at lower temperatures,the incubation period is more extended /7/ resulting in a shift of Nqax on the time scale. At 580 OC the incubation time is extremely short, confining the nucleation period to aging times of less than one minute. Therefore, at 580 OC the maximum of Nv(t~) occurs at tA<lmin; for longer aging times a Lifshitz-Slyozov-Wagner (LSW)- type dissolution and coarsening reaction of the precipitates commences /cf.7/leading to a decrease of Nv already for tA>l min (though it should be noted that at these precipitation stages the coarsening rate da3/dt~ is completely different from the one predicted by LSW-theory, see /7/). Thus, in the present study observation of t k nucleation period has only been possible for aging at 500 and 540 OC.

(4)

1 - Onr

lOnm

C - )

lunm -

Fig. 1

-

Neon field ion image of

Ni -36%Cu

-

9%Al aged for a) 20 min, b) 190 min, C) 3000 min and d) 12180 min at 500 OC.

Fig. 2

-

Time evolution of the mean radii ( a) ) and the number density ( b) ) of Y'-precipitates in N i - 3 6 % C u - 9 % A 1 during aging at 500, 540 and 580 OC.

(5)

JOURNAL DE PHYSIQUE

Fig. 3

-

Composition of y'-precipitates as a

function of time during aging at 580 OC.

The composition of the precipitates as analyzed with the atom-probe (technique ii) )

does not change with1time at either aging temperature (Fig. 3). This result sug- gests that phase separation is inikiated via nucleation and growth of y'-precipi- tates, rather than via spinodal decomposition. For al1 temperatures, the particles have a stoichiometric composition y'-CNixCul-x)3Al with x=0.8 at 580 OC and x=0.85 at 540 and 500 OC.

Interpretation of the experimental data displayed in Fig. 2 in terms of the 'Nume- rical precipitation mode1 (N-model)' devised by Kampmann and Wagner /7/ allows both the yl/matrix-interfacial energy ( 5 ) and the effective diffusivity ( D ) to be deter- mined rather accurately for each aging temperature. As shown in Fig. 4, the N-mode1 which comprises nucleation, growth and coarsening and, thus, describes the kinetics in al1 stages of the precipitation reaction, fits the experimental data rather well for the 5 and D values given in'Table 1 (more details will be given in a forth- coming paper /9/).

Table

-

1

O/J

-

m-2 ~ /

-

msec ~- 1

TA/OC R*

-

6 / nm ô ~ * / k ~ ~ cgl / at %

From an Arrhenius plot of D(TA) the activiation energy for diffusion and the pre- exponential factor have been obtained as Q = 50 kcal/mol and Do = 8.

IO-^

m2/sec, respectively. Within the error bars, a = 0.052 ? 0.002 J / I U ~ is independent of the aging temperature. 5 is considerably larger in the ternary Ni- 36at%Cu-9 at%Al than in the binary Ni- 14 at%Al alloy, for which 5 was determined to be only 0.016

. . .

0.019 ~ / /7, m ~8/. Presently, we do not understand the large increase of 5, once about 20 at% of the Ni atoms within the y'-particles are substituted by Cu atoms. However, we have found some indications /4, 9 / that both the Ni and the

(6)

Cu concentrations in the matrix close to the yl/matrix interphase boundaries do not yet correspond to the equilibrium concentrations, but rather show a depletion of Ni and an enrichtment of Cu in these regions. It is conceivable that the large value of o in the ternary alloy is associated with these non-equilibrium concen- trations around the y'-precipitates.

IO-' IO' 103 los IO-' 10' 1o3 los

tA Irn~nl te, lrninl

Fig. 4 - Time evolution of the normalized supersaturation (cgl: equilibrium solu- bility) (left), of the mean radius

(R)

(middle), and of the number density of y'-particles (right) in Ni- 3 6 % C u - 9 % A 1 at T ~ = 5 0 0 OC ( a) 1 , 540 OC ( b) ) and 580 OC ( c) ) . The solid curves are the predictions from the Numerical mode1 /7/

with the D and o values given in Table 1. The nucleation rate J is calculated with the nucleus formation work 6GX/kTA = 4 r R*' o/3kT~ /7/ and with R*=R, see Table 1.

(7)

C9-446 JOURNAL DE PHYSIQUE

Acknowledgements: FJe thank D r . H. Wendt and P r o f . P. Haasen f o r many f r u i t f u l d i s - c u s s i o n s . We a r e e s p e c i a l l y g r a t e f u l t o R. Kampmann f o r h i s computations of t h e p r e - c i p i t a t i o n k i n e t i c s i n terms of t h e Numerical Model.

REFERENCES

1. w.0. Alexander, J. I n s t . Metals ( 1 9 3 8 ) , 163.

2. J. Kagawa, T. Miyazaki and H. Mori, Trans. J I M

18

( 1 9 7 7 ) , 707.

3. Z.G. Liu and R. Wagner, Proc. 3 0 t h I n t . F i e l d Emission Symp., P h i l a d e l p h i a , 1983, p. 113.

4. Z.G. L i u , Ph. D. t h e s i s , Univ. of Gottingen, 1984.

5 . R. Wagner, F i e l d Ion Microscopy i n M a t e r i a l s S c i e n c e , Vol. 6 ,

CRYSTALS (springer-Verlag, B e r l i n , 1982).

6. J. P i l l e r and H. Wendt. Proc. 29th i n t . F i e l d Emission Symp., Eds. H.-O. Andrén, H. Nordén, Goteborg, 1982, p. 265.

7. R. Kampmann and R. Wagner, i n 'Decomposition of Alloys:

The E a r l y S t a g e s ' . Eds. P. Haasen, V . Gerold, R. Wagner, M.F. Ashby (Pergamon P r e s s , 1984); i n t h e p r e s s .

8. H. Wendt and P. Haasen, Acta Met.

31

(1983), 1649.

9. Z.G. Liu and R. Wagner, t o appear i n Acta Met.

Références

Documents relatifs

Résume - Nous re-examinons la théorie des énergies d'apparition et de déficits d'énergie s'applicant aux mesures de resolution à haute énergie dans la sonde à atomes puisée

Our detailed study has shown ,beyond any doubt, that Ni segregates to the surface when the bulk Ni concentration is in the range of 0 to 16at%: the first definitive experimental

As-quenched specimens (e.g. Ni41Pd35P18) showed on first heating in the DSC-scan one glass temperature and an exothermic peak which is identified with the enthalpy of a

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des

The A1-Nb interfaces formed at room temperature exhibited an atomically abrupt interface and no mixed layer was found.. The A1 atoms at the abrupt atomic interface

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des

Since the enviroment of individual atoms in these micro-clusters could be different from those in the matrix, the field evaporation behavior of micro-clusters may well

ATOM PROBE FIM STUDY OF SHORT RANGE ORDER IN Ni RICH Ni-W AND Ni-Mo-W