DETERMINATION OF THE GROWTH RATES OF Au2Al DIFFUSIONALLY FORMED IN THIN STEP-SHAPED Au-Al COUPLES

(1)

HAL Id: jpa-00223477

https://hal.archives-ouvertes.fr/jpa-00223477

Submitted on 1 Jan 1983

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.

DETERMINATION OF THE GROWTH RATES OF Au2Al DIFFUSIONALLY FORMED IN THIN

STEP-SHAPED Au-Al COUPLES

I. Reda, A. Wagendristel, H. Bangert, P. Schattschneider

To cite this version:

I. Reda, A. Wagendristel, H. Bangert, P. Schattschneider. DETERMINATION OF THE GROWTH

RATES OF Au2Al DIFFUSIONALLY FORMED IN THIN STEP-SHAPED Au-Al COUPLES. Jour-

nal de Physique Colloques, 1983, 44 (C10), pp.C10-95-C10-99. �10.1051/jphyscol:19831020�. �jpa-

00223477�

(2)

JOURNAL DE PHYSIQUE

Colloque CIO, suppl6ment au n012, Tome 44, dicembre 1983 page C10-95

DETERMINATION OF THE GROWTH RATES OF AuZAl DIFFUSIONALLY FORMED IN THIN STEP-SHAPED Au-A1 COUPLES

I.M. Reda, A. Wagendristel, H. Bangert and P. Schattschneider I n s t i t u t e of AppLied and TechnicaZ Physics, TeehnicaZ University, Vienna, Austria

R6sum6

-

L'or et ses compos6s Au2A1 et Au A1 presentent des dif f 6rences de r6f lectivit8 spec?ra?e importantes dans le visible. Ces diff6rences, observables 2 l'oeil nu, ont St6 utilis6es dans un couple de couches minces Au-A1 pour dstecter le mouvement vers la surface d'Au du front de phase des compos6s formgs par diffusion.

A partir d'une ssrie continue de tels couples 6vapor6s avec des gpaisseurs differentes sur le meme support, c'est-a-dire des couples de diffusion en forme de mar- chef avec un rapport atomique total 2Au/A1, les vitesses de croissance ont 6tG d6terminBes. La temp6rature de recuit n6cessaire 6tait comprises entre 333 et 373 K.

Les valeurs des vitesses de croissance sont en parfait accord avec des donnges r6centes de la littsrature, seulement dans la partie la plus 6paisse du couple : D = 2,l exp(-l/kT) cm /s; des vitesses l6qSrement plus 6lev6es ont 6t6 obtenues dans la partie la plus mince.

Abstract

-

Gold and its compounds Au A1

,

Au A12 show marked differences in their spectral re?lectivity in the visible range. These differences, observable with the naked eye, have been used to detect the approach of the phase front of diffusionally grown compounds against the gold surface of a Au-A1 thin film couple.

From a continuous sequence of such couples evaporated in different thicknesses onto the same substrate,i.e.

a step-shaped diffusion couple, of a total atomic ratio 2Au/A1, the growth rate constants were obtained The temperature for the diffusion anneal ranged from 333 to 373 K. Excellent agreement with recent lite- rature data for the rate constants was found only in the thicker part of the couple: D z 2 . l expf-l/kT)cm2/s whereas the thinner part showed slightly higher rates.

I

-

INTRODUCTION

Contaots between Au and A1 metallisation are often found in micro- electronic circuitry. Unfortunately this bimetallic system forms al- ready well below 1 0 0 ~ ~ compounds revealing undesired properties such as high electrical resistance and brittleness. For this im- portant practical reason the thin film diffusion system Au-A1 was the subject of extensive research in the past / I - 4 / as well as up to recent times, where more elaborate methods could succesfully clarify the kinetics of phase formation in this system under different conditions 5 - 1 1 Mainly the papers of Vandenberg / 1 1 / , Campisano 5 Majni / 9 , 1 0 / and their coworkers showed that in

Article published online by EDP Sciences and available at http://dx.doi.org/10.1051/jphyscol:19831020

(3)

CIO-96 JOURNAL DE PHYSIQUE

couples with a composition according to 2Au/A1 and annealed somewhat below 1 0 0 9 ~ the phase Au A1 always appears first as a layer parallel to the origina15ingerface. On the aluminium interface of this compound very soon Au A1 appears as a second intermediate layer, growing with a comparagle rate. If the couple is set up such that the atomic ratio is 2Au/A1, the Au5A12 phase front reaches its limit at the Au surface, then this phase reacts with the remaining A1 to form also Au2A1 until all Au5A12 and A1 is used up. The final state is given by a homogeneous fllm of Au Al. A scheme of this process is shown in fig.1. The growth rate constants have been found

t = O

F i g l : P h a s e f o r m a t i o n i n A u - A 1 t h i n f i l m c o u p l e s / l o , 1 I/

to be rather-lQigh (in the temperature range from 6 0 to 1 0 0 ~ ~ between 3 x 10 and 1 x cm2s-I) and depending on the film structure. Polycrystalline reaction partners showed higher rate constants. In a previous paper / 7 / we reported growth rates in very thin couples (60 nm) which exceeded those for thicker films. For this reason we reexamined this diffusion system in a wide range of couple thicknesses by means of a rather simple method making use of the discolouring of the gold part of the couple when it is replaced by the newly formed alloy.

11. METHOD

The characteristic yellow colour of gold is caused by a low reflectivity below 500 nm wavelength, whereas the colour of Au2A1 is typi- cally neutral-gray, showing no significantly structured reflection spectrum in the visible

range but rather generally

-1.0

low reflectivity around 55%.

Au5A12

,

however

,

appears ( l2 V C

-

darkgray with a slight

purple shade corresponding R

to a continuous decrease in ^{A,, A I} reflectivity with decreasing

wave length. In fig. 2 the

i n c e d e n c e o f Au, A 1 a n d t h e i r i n t e r m e t a l l i c

c o m p o u n d s Au2A1, A u 5 A 1 2

lo.,

^, ^, ^, ^, A L ~ I ~ I , a s a f u n c t i o n o f p h o t o n 0.4 0.5 0.6 0.7 0.8

w a v e l e n g t h . -I

(4)

reflectivities of Au, Al, Au2A1 and Au5A12 are plotted for compari- son. The different colours of Au and Au2A1 or Au5A12can clearly be distinguished with the naked eye and offer a simple possibility to observe the approach of the phase front of the reacted layers towards the surface of the gold film during the anneal of a Au-A1 thin film couple. Of course one must expect quite a large lack of definition of the end of the growth process as the remaining part of the gold film becomes successively more transparent and hence its colour is changing continuously. But since the human eye is quite sensitive to difference in colour and brightness, when it can compare neighbouring areas, this lack of definition can be drastically reduced if there is a Au2A1 standard next to the couple.

For this reason we have used films where this possibility exists automatically. Our films consisted of individual couples all with an atomic ration of 2Au/A1 but of increasing thickness and each is deposited spacelessly next to the other. In such a stair like bimetallic couple (fig.3) the phase boundary of the reaction products

Schematic illustration of layered structure caused

b y diffusion in a step-

shaped 2Au/Al t h i n film couple.

successively reaches one step after the other, thus making each previous one a standard Au2A1 for comparing it with the next.

111. EXPERIMENT

Films have been prepared by evaporation of a stair film Al, 5,10, 15. ..nm on a glass substrate-which was overlayed by Au 10,20,30

...

^nm

The thickness ratio 1A1:2Au of the individual couples corresponds directly to the atomic ratio since structure and lattice constant are a1 ost the same for Au and Al. The pressure during evaporation was

lo-'

torr. These samples were annealed by pressing them with the substrate side onto a heated copper block. The temperature was measured at the film side and kept constant within a range of

1°c.

The gold film at the outer surface protects the couples automatically against oxidation of samples which was checked by comparing the results of samples annealed under atmosphere and in situ. An- nealing temperatures were 60, 74, 7 8 , 8 0 , 88 and 9 8 O ~ . During the anneal the successive discolouring of the gold layer from step to step in the film was observed. The end of the reaction (t*) in the original films was found from the disappearance of the boundary between the step just reacted and the one reacted before. The un- certainty to define this moment is about 20% and arises mainly from the similarity in colour of Au5A1 and that of Au2A1.

The distance to be crossed by the diffusing atoms is the thickness of the formed compounds (fig.3). Hence the growth rates constant is given by x/t. For the entire reaction this relation yields the mean rate constant according to d2/t*.

(5)

JOURNAL DE PHYSIQUE

I V . RESULTS

From e a c h s a m p l e s i x v a l u e s a c c o r d i n g t o t h e s t e p h e i g h t s w e r e o b - t a i n e d i I n a l l c a s e s we f o u n d h i g h e r r a t e c o n s t a n t s ( g i v e n by t h e s l o p e o f d 2 v s t ) f o r t h i n n e r f i l m s .

F i g . 4 : [ T h i c k n e s s o f r e a c t e d -50

l a y e r ) s q u a r e d v s t i m e f o r a s t e p - s h a p e d ZAu/

A 1 t h i n f i l m a n n e a l e d i n s i t u a t 353K.

t(min1

G r o w t h r a t e c o n s t a n t s b e t w e e n 3 3 3 a n d 371 K a r e d i s p l a y e d i n f i g . 5 t o g e t h e r w i t h l i t e r a t u r e d a t a i n f o r m o f a n A r r h e n i u s p l o t . The l e n g t h o f t h e b a r s i s g i v e n by t h e i n i t i a l a n d t h e f i n a l g r o w t h r a t e . The l o w e r v a l u e s ( c o r r e s p o n d i n g t o t h e t h i c k e r c o u p l e s ) c o m p a r e v e r y w e l l w i t h t h o s e f o u n d by o t h e r a u t h o r s w h e r e a s t h e h i g h e r l i m i t t e n d s t o w a r d s r a t e c o n s t a n t s w h i c h we h a v e r e p o r t e d f o r v e r y t h i n f i l m s / 7 / . The f i n a l r a t e o b e y s D = 2 . 1 e x p ( - l / k T ) c m 2 / s .

RATE CONSTANT U

F i g . 5 : T e m p e r a t u r e d e p e n d - e n c e o f g r o w t h r a t s c o n s t a n t s i n A1-Au t h i n f i l m c o u p l e s .

To e n v i s a g e t h e r e a c t i o n s p e e d e . g . a t room t e m p e r a t u r e o n e f i n d s by e x t r a p o l a t i o n a t h i c k n e s s o f t h e r e a c t e d l a y e r o f 1 0 nms a n d 1 urn a f t e r a t i m e o f 5 . 4 h a n d 3.15 h r e s p e c t i v e l y .

(6)

V

.

DISCUSSION

Majni et a1 /lo/ reported growth rates for Au A1 grown into a thin film and a single crystal of gold which differ %y about one order of magnitude. This makes clear that the speed of the reaction is strongly dependent on the supply of the diffusing atoms via grain boundaries. The transport of the atoms across the reacted layer can hardly be understood by lattice diffusion since in this case the stoichiometry of Au2A1 and Au5A12 would have to be broken. Fast grain boundary diffusion, however, explains the high growth rates as well as their decrease with time: The density of grain boundaries in the just grown layer is higher in the initial state and is gradually reduced by recrystallisation during the extended anneal. Especially in the Au2A1 layer significant changes from a highly disordered or a very fine grained structure to the crystalline state are taking place during the reaction /8/. (Diffusion coefficients reduced by defect annihilation during the heat treatment is very often found in freshly deposited thin film couples). To recognize grain boundary diffusion as the driving mechanism of the reaction one may also consider the activation energy o f 1 eV typical for grain boundary diffusion in metals. In our previous paper we had also to assume

"grain boundary spikes" in the reaction fronts to explain the x-ray- optical reflection curves. S o , in conclusion, we may consider the growth process of the reacted intermediate layers as limited by the diffusion of Au and A1 via the grain boundary network in the reacted layers. This fast transport supplies these elements t o t h e p h a s e - front where they react with their residual counterpart Au5A12at the Au side and Au A1 at the A1 side. Finally, when all Au is used up the little remainder of A1 has to be brought to the Au A12/Au A1 interface which moves towards the other surface. The possi?illty %or the easy transport of atoms to the actual reaction front is decrea- sed as the structure of the reacted layer is improved during the heat treatment.

This work was sponsored by the Austrian Fonds zur Forderung der wissenschaftlichen Forschung, Projekt Nr.2152 and the Jubilaums- stiftung der Stadt Wien.

REFERENCES

1. W.J. TAKA1 and H.M. FRANCOMBE, Solid State Electronics (Pergamon, New York, 1968), Vol.11, P.202

2 . D.P. KOLESNIKOV. A.F. ANDRUSHKO and Y.E.J. SUKHININIA,

Fiz .Metall.

3,

^'529 ( 1972)

3. C. WEAVER, in Physics of Thin Films 6 , ed. by M.H. FRANCOMBE and R.W. HOFFMANN (Academic, New York, 1971) P.301

4. C. WEAVER and L.C. BROWN, Phil-Mag.

1,

¹ ⁽¹⁹⁶²⁾

5. T. INAGAKI, E.T. ARAKAWA and T.A. CALLCOTT, J.Appl.Phys. 52(9), 5597 (1981)

6. B. LOISEL and E.T. ARAKAWA, Applied Optics,

2,

1959 (1980) 7. A. WAGENDRISTEL, H. SCHURZ, E. EHRMANN-FALKENAU and H. BANGERT,

J.Appl.Phys. 51(9), 4808 (1980)

8. J.M.

VANDENBEE

and R.A. HAMM, J.Vac.Sci.Techno1.

9,

1 , 8 4

' (1981)

9. S.U. CAMPISANO, G. FOTI, E. RIMINI, S.S. LAU and J.W. MAYER,

Phil.Mag. 31, 903 (1975)

10. G. MAJNI,

C.

NOBILI, G. OTTAVIANI, M. COSTATO and E. GALLI, J.Appl.Phys. 5 2 ( 6 ) , 4047 (1981 )

1 1 . G. MAJNI, G. OTTAVIANI and E. GALLI, J.Crysta1 Growth

9,

583 (1979)