Multiphase diffusion in some related ternary metal systems

Multiphase diffusion in some related ternary metal systems

Citation for published version (APA):

Loo, van, F. J. J., & Bastin, G. F. (1983). Multiphase diffusion in some related ternary metal systems. In F. J. Kedves, & D. L. Beke (Eds.), Diffusion in metals and alloys : international conference, 1982, Tihany, Hungaria: proceedings (pp. 580-592). (Diffusion and Defect Monograph Series; Vol. 7). Trans Tech Publications.

Document status and date: Published: 01/01/1983

Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers)

Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

MULTIPHASE DIFFUSICN

m

SCME REIATED TERNARY MErAL SYSTEMS. F.J.J. van IDo and G.F. Bastin

Eindhoven University of Tec1mology,

Ialx:>rato:ry of Physica.l Chemi.stry, Eindhoven, '!he Netherlands.

Intrcx1uction

Recently the authors have published their results of investigations into nultiphasè diffusion phenarena in the systen1s Ti-Ni-CU [1],

Ti-Ni-Fe [2] and Ti-Ni~[3]. Especially the product layer norphology in these diffusion couples in which a TiNi

3-like phase was fonred

~ te he ve:ry interesting. '1herefore, a systematie investigaticn

was started in order to carpare the resulting m:>rphologies in couples of the type TiNi

xMe1_x versus NiyMe1-y in which TiNi3-like phases are

expected to occur (the syni:x:>l Me represents ene of the elemants Cu, Fe or Co). Fran a fundanental as weIl as fran an experinental point of view i t appeared to he advantageous te annea.l the couples at ene fixed taIperature, Le. 900oC. It was, therefore, necessa:ry te

detennine the 9000C cross-section through the Ti-Ni

-cu

phase diagram since in earlier investigations this was dane by us only at 800 and

8700C [1].

'!he experilrental procedure

For the detenni.naticn of the 9000C isothenn of the Ti-Ni-CU system as weIl as for the use as a tenni.nal material in diffusicn

eooples, alloys have been prepared by repeatecl. argon are nelting. After nelting they wem hatDgeni.sed for at least 100 hours at 9000C in

sealed evacuatecl. silica capsules. As reported by Zwicker et al. [4] the

nelting point of binary Ti

-cu

alloys is ve:ry nuch influenced by traces of oxygen • '1herefore, in order to prevent axygeh contamination as a result of direct contact between the alloy and the silica capsule sare alloys

were

wrélfPed

in titanium sheet.

Diffusion couples were made in a special furnace in which three slices of the tenni.nal alloys were hot-pressed as a sandwich by rreans

of a set of wei.ghts for 70 te 150 hoors at 9000

e

in a vacuum better than

10-5 Torr. After diffusian annealing the cooples were eniJedded, ground

and polished parallel te the diffusion direction and, if

necessary,

etched with a mixture of 10% ~02' 5%HF and 85% ~O.

'llie alloys and oouples were then investigated by cptical microscqJy

using polarised light, by X-ray diffractien and by micrt\'robe analysis

using a Jeol Superprobe 733 [3].

'!he detennination of the 9000

e

isotherm of the Ti-Ni-eu

J;ilase

diagram

In Table 1 a nt.miJer of equilibrated alloys are given together

with the

Ifuises

which were·found. '!he presence of a liquid

Fhase

in the Ti~-regialmade the detennination of that part of the isotherm rather difficult. 0Ur results are in l.ine with the phase diagram of the binary

Ti

-cu

system as given by let>ffat [5].

In fig. 1 the resulting isotherm is given.

Table 1.

Phases present in Ti-Ni-eu alloys equilibrated at 900o

e.

For the designation of the

Fhases see

Fig. 1.

Alloy ca!J?C?Sition

TiCu4 , TiCu2 and Ti2~ Ti25Ni50~5 Ti31Ni8Cu61 Ti33Ni7Cu6Ó Ti33Ni9Cu58 Ti33Ni

n

Cus6 Ti33Ni13Cu54 Ti34Nill.5Cl1s4.5 Ti34Ni54.5CUll.S Ti38Ni8CUS4 Ti39Ni4Cu57 Ti44Ni7Cu49 Ti60Ni23Cu17 Ti7SNiI2.SCUI2.S Phases present Melting phenatena

e

E

+

nrri2~JI E E E+O D D+-E+"~~n DiJI'iNi 3+TiNi D+-''Ti_2--jCL"+Ti Cu_{3 4} E+"Ti2~" TiNi+Ti 3Cu4 TiNi+Ti~i+Ti₂Cu (3.JI'i+Ti 2Ni+Ti2Cu

p

-Ti

Ni

Fig. 1. The 9000C isotherm of the Ti-Ni-Cu phase diagram.

Results of the diffusion couple experlltents

In figs. 2, 3 and 4 our results·en diffusicn couples are sh,a,m

in a qualitative Wér;/. 'llle thicknesses of the product layers are

nonnali-sed for an armealing ti1te of 100 hours by nultiplying the actua1

thicknesses by a factor

(IOO/t)~,

where t = annealing tine in hours. For a m.mber of couples the parabolic gr<:Mt:h behaviour bas been verified.

Ni

~

~4-~_

- -

:i.:F~i:f..i.·f·f=i.·i:t:t.i.·

2»))}))}})l!

:)·:14)x:wX',:)(:':x.1::%..i.

~.-Cu

melting

phenomena

o

znm

E

rnmn

C

~ A sO).Jm

583

Fig. 2. Survey of the morphologies and thicknesses of the reaction

layers in Ti-Ni-Cu diffusion couples after annealing 100 hours

at 900

0

C

Ni

-N"

F'o 180, 20

•

...

-Ni Fe

50 5 -...,...LI;f~?J~~it{.~fi~ï4..

~,;.;.".'.~.:

_{- .}..:.:.:

:.~.~

..

,:.~

..

~,.:~.:.:.~~.:.:.:~.J.:~.;

..

~~.:-.:.~:,~.'.:.;u.:';

~_ ili'tW{-;··';\

TiFe

₂ sO}Jm

Fig. 3. Survey of the morphologies and thicknesses of the reaction

layers in Ti-Ni-Fe diffusion couples after annealing

100

hours

at

900

oC.

Ti

_so

CO

_so

Ni

Co

_

ill:!lfS'à

mrrm

~ ~ 4 •

TiCo

₂

Chex)

TiCo

₂

(cub)

_TiCo3

TiCN!5C~S)3

_TiNi3

s0J,Jm

Fig.

4. Survey of the morphologies and thicknesses of the reaction

layers

i

n

Ti

-Ni -Co diffusion couples after anneal ing

100

hours

at

900oC.

product noqilology in each of the 15 diffusion couples investigated in the Ti-Ni

-cu

system. '1be TiNixCu

l_x tenni..nal alloy is always the

upper part, whereas NiyCul-y is the bottan part in eachdrawing. In these

schelres no infonration can be found conoeming the concentration gradients existing in the various phases. In all couples these have been det:enni.ned by micrcprcbe analysis, but in the franework of this article al1y two exanples are shewn in Fig. S,a and b. Especially in

the TiNi te:cninal alloy a large zene is found in which the concentration

of the variOJS elem:mts gradually change, in contradiction te the al-ways steep concentraticn gradients in all ether te:rm:inal allays. Typical

for these systems is the maxilrurn in the Ti concentration profile in

the NiOl tenninal alloy, and the max:i.nml in the Ni concentration profile

in the TiNi tenni..nal alloy in couples of the type TiNi-NiyFel-y and

TiNi-Ni Col •

Y

-y

In the Figs. 6, 7 and 8 a nunber of diffusion paths is presented on the respective i.sotherins.

In all types of couples 'the original welding interfaoo (Kirkendall

interface) was found

near

the Nil'fel-y tenninal alloy

as indicated

by the arrov1 in Figs. 2, 3 and 4.

Evaluaticn of the exper:inental results

In each vertical colmn in Figs. 2, 3 and 4 a clear relatioo can

be found conoeming the variOJS no~ologiesand thicknesses of the

product layeri;. In the harizontal rCMS this relation is also present althcugh less prona.mced.

'1be lOOSt interesting point for us is the questicn why in certain

oouples ene ar nore interfaces are no longer' straight and becare

serrated.

In all three systems this ~cnencn

occurs,

but clearly

not in the sane way. '1lle kind of pertw:baticn is also different: the

needles of the TiNi

3 I;t1ase, protruding into Ni/,el-y' for exanple, are clearly nuch nore regular than the needles of TiFe₂ protruding inte Ti49Ni24Fe27' In the first case an orientation relatioo exists between

the releVant phases, leading te parallel needles or fixed angles

between the Deedles.

Petween

_{TiFe2 and Ti49Ni24Fe27 such a relation} bas

Fig. 5. Concentration profiles in the diffusion couples TiN;-NiCu (a)

and TiNi-NiFe (b), both annealed 100 hours at 900oC.

D • Ti o • Ni + .Cu

400

- - - I• • firn

300

200

587

+~+.++ t +,+ +r+ I

100

o ,

"Ni Cu" TiNVA 0 "TiNi"

•

+

• •

100

-at%

Î

0 00 +

50 -

•

•

++ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 8 9 1 2 1 '&0 DDDDDDDDDDDD 0 0 DDD

utr

"TiNi" 2phases "'NiFe" ,

r

T

100

-I I • Ti • D atOJo

_'.

, _.Ni

Î

•

0

,,

• Fe I ₊ I I I

..

50 -

•

• •

'&~: ij 0

a

8

o 0 0 0 0 0 0 _{8 B 9 121} D D D D D D D D 0 0 ' 0

q

0 I

,

I I

,

I I I _{+ +} I + + +

o ,

,

Do,D~

I I + + +I I

t

+, + , + , + I

100

200

300

400

• ).Irn

( b)

p

-Ti

Ni

Fig. 6. Same diffusion paths on the 900

0

C isotherm of the Ti-Ni-Cu

phase diagram.

carplicated by the fact that also in sate grain boundaries in this a1loy the TiFe

2 phase is fOnted.

'1be pcssibility of the presence of an orientation relatioo is, in

any case, not sufficient in order te create interface perturbatiOlS, since then the effect had te he seen in m::>re couples .

we

believe, therefare, that the diffusioo nechani.sm plays an inp:>rtant role. It is a Well-kncwn fact that in single phase ternary alloys it already takes foor independent diffusim ccefficients in order te describe the

~:ri

TiNi

Ti Fe

~~~~

a-Fe

Y-{Ni,Fe}

Fig. 7.

Some diffusion paths on the

gDDOe

isothenm of the Ti-Ni-Fe

phase diagram.

diffusion kinetics • In an actual multiphase diffusion couple in which

e.g. UNo phases are fonted this leads te 16 independent diffusial

cref-ficients which can all be coo.centratial dependent. ~ do oot believe, however, that the question of interface instability can be solved by analysing the diffusion nechanism along theSe lines,apart fran the

experinental difficulties.

In a qualitative way, al the other" hand, i t might he pa;sible te find keys to a soluticn of the proolem in a similar way as shown in a forner studyen the Mo-Si-C systern [6] and which follows essentially

~-T

i

Ni

/ / / / /

//

Tie

_O2

C

c u

bJ

//

,/

z~~iC02Chex'>

fzzi/~ZZZZZ!mZ~~~~~~~~77m'!fÎlÎ,~

Tie

03

Y-CNi.Co)

Fig. 8. Same diffusion paths on the 900°C isothenm of the Ti-Ni-Co

phase diagram.

a line given by wagner [7] for oxidation processes. Catpare, for instanee , in Fig. 9 the couples .TiNi~iFe ànd TiNi-NiCu and let us cx::ncentrate en the interface TiNi]-NiMe.

In bath cases·Ti will ba the less nohile CXI'lpOI1ent in the TiNi 3

phase in view of the locatien of the Kirkendall interface, and Ni is

pmfere.ntly withdrawn fran the NiMe alloy. If the arrival of Ni atcm; at the interface is rate-det:eJ:mining for the grcwth of the TiNi

3 phase, then

_.

thi.S layer·_. gr~ faster at point 1 than at point 2, so an accidental perturbation will grON. If the Supply of Ti is

rate-TiNi

TiNi

o

A

Ti N

iJ

_~Ti

---fNi

---

--

--

---591

Fig. 9. Schematic representation of the diffusion layers in the couples

TiNi-NiFe and TiNi-NiCu.

detennining, then the layer will grcM faster at point 2 than at point 1 and the interface will stay straight.

Since the diffusim rate of Ti in TiNi

3 can he expe.cted te be roughly the sane for the Ti-Ni-Fe and Ti-Ni-CU system, the difference

between both systems shoold then he the faster diffusion of Ni in a

NiCu alloy than in a NiFe alloy. 'lhis is indeed repOrted in the litera-ture [8] and bas been verified by us in diffusicn e:xperi.IIents on the c:ouples NiFe-Fe and NiCu-CU. As a consequence, the prcduct layer in the

Ti-Ni

-cu

system should be thicker than in the Ti-Ni-Fe system which is indeed the case. '!he perturbation effect IID.lSt beccne m:>re pronounced if the SUH?ly of Ni gets less, i.e. when using alloys poorer in Ni. This is indeed true as ean be seen frem Figs. 2 and 3 where at the end

NiCu

NiFe

even in the Ti-Ni

-cu

system the interface is net straight anynore. 5emi.-quantitativelyour results satisfy wagners canditic:n, which

states that straight interfaces occur if the quantity Q> 1, whereas

serrated bamdaries may occur if Q < 1, where Q is a functian of the

relevant diffusion coefficients and cc:ncentratians

[7J.

\'ë believe that these relatively sinple consideratioos will be

very useful in predicting the occurrence of perturbations in reactic:n interfaces in temary diffusic:n ooupl.es. 'Ihis nay be of teclmological inportance in view of the adherence of e.g. coatings on Substrates •

Refemnces

1. F.J.J. van IDa, G.F. Bastin and A.J.H. Ieenen,

J.

Less-oammcn Met., 57 (1978) 111.

2. F.J.J. van !Do, J.W.G.A. Vrolijk and G.F. Bastin,

J.

Less-oammcn

Met.,

77 (1981) 121. 3. F.J.J. van !Do and G.F. Bastin,

J.

Less-acmmcn Met., 81 (1981) 61.

4. U. Zwicker, E. Kalsch, T. Nishinura,

o.

Ott and H. seilstarfer, Metall 20 (1966) 1252.

5. W.G. M:>ffat, '!he H.andbcx>k of Binary Phase Diagrans , General Electric Catpany,'SChenectady, N. Y., 1978. 6. F.J.J. van IDa, F.M. Srret, G.O. Rieck and G. Verspui,

Proc. Plansee-seminar 1981, Vol. 1, p. 141. 7. C. Wagner,

J.

El~.

Soc.,

103 (1956) 571.

8. Y. J\dda and J. Philibert, Ia Diffusic:n dan les Solides , Tc:1te 11, Presses Universitaires de France, Paris 1966.