STRUCTURE-MAGNETIC PROPERTY CORRELATIONS OF CROSS SECTIONS OF DC MAGNETRON SPUTTERED CoCr THIN FILMS

(1)

HAL Id: jpa-00229176

https://hal.archives-ouvertes.fr/jpa-00229176

Submitted on 1 Jan 1988

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.

STRUCTURE-MAGNETIC PROPERTY

CORRELATIONS OF CROSS SECTIONS OF DC

MAGNETRON SPUTTERED CoCr THIN FILMS

D. Ravipati, P. Narayan, J. Sivertsen, J. Judy

To cite this version:

(2)

JOURNAL DE PHYSIQUE

Colloque C8, Suppl6ment au no 12, Tome 49, decembre 1988

STRUCTURE-MAGNETIC PROPERTY CORRELATIONS OF CROSS SECTIONS OF DC MAGNETRON SPUTTERED CoCr THIN FILMS

D. P. Ravipati (I), P. B. Narayan ( I ) , J. M. Sivertsen (2) and J. H. Judy (2)

(I) Control Data Corporation, Minneapolis, MN-55435, U.S.A. (2) Unzversity of Minnesota, Minneapolis, MN-55455, U.S.A.

Abstract. - CoCr film structural features, on A1 and Nip (on Al) substrates were established by control of process conditions to optimize structure sensitive properties like perpendicular coercivity. H C ( ~ ) depends strongly on process conditions but relatively independent of substrate type. Rough substrates ~roduced films exhibiting less uniformity in columnar grain diameters.

Introduction

Sputtered CoCr is one of the preferred material sys- tems for high density perpendicular magnetic recording media. The structural features of CoCr films play a dominant role in affecting the magnetic and subse- quently the recording performance. The structural features can be manipulated by controlled variation of the processing variables such as argon pressure, substrate temperature, deposition rate, substrate type etc. The objective of this study is t o report on the correlations between structure and the magnetic properties of DC magnetron sputtered CoCr thin films.

Experimental

DC magnetron sputtering was performed in a dual- sided sputter system. Four different type of substrates were used: Diamond turned Al, Diamond turned A1 and polished, Polished A1 with 5000

A

of carbon and Nip (coated on polished Al). The degree of surface roughness was measured using a Dektak surface pro- filometer. Surface roughness for each of the different substrates used are as shown in table I. Argon pres- sures and substrate temperatures were varied from 1- 25 mtorr and 50-200 OC respectively. The magnetic properties were measured using a computer controlled VSM. Cross-section samples for TEM were prepared using a Sorvall MT-5000 microtome. The A1 base of the samples cut were dissolved in H N 0 3 acid which does not react with the CoCr film. The residual film floated on the surface was picked up and placed in a liquid resin which was left overnight t o solidify. The cross-sections of 800-1 200

A

were prepared using a diamond knife on the microtome.

Results and discussion

Figure 1 shows that perpendicular coercivity H C ( ~ ) ,

increased with film thickness. The effect of the substrate on HC(*) was observed only for the initial 2000

A

thick film for all the substrates studied. Measurements such as H C ( ~ ) and c-axis dispersion,

A450,

are an av- Table I. - Substrate roughness.

Substrate type Arithmetic average

Rough A1 Polished .41 Nip coated .41

Carbon on .41 43

rage measurement of the total film thickness. Mea- surements on thick films (thickness around l pm) do not reveal the true magnetic properties and the orien- tational features of the initial layers. The transition layer between the substrate and CoCr film accomo- dates the transitional stages in the growth and acts as a base for the subsequent epitaxial growth of CoCr with (002) orientation.

The initial layers demonstrate that H c ( I I ) is compa- rable t o H , ( l ) indicating that the initial layers exhibit

a fairly strong longitudinal component. However, a BCC structured underlayer with strong (110) orientation is required for the Co (1011) plane of hcp system

I

1 1 9 2 1 $000 ,100 ,1000

0 0 1 1 1 DO000

o o 8 a Thickness (A)

Fig. 1. - H C ( l ) dependence on film thickness for different substrates used.

to lie in plane. Figure 2 shows that the H c ( I I ) of CoCr film increasing with Cr underlayer thickness. These results demonstrate that the initial layers are magnet- ically isotropic.

The initial layers. (i.e. up to 2000 A) exhibited highest H , ( l ) for CoCr films grown on coated sub-

strates (Nip and carbon on Al) than the bare A1 substrates. The temperature of the film increases contin- uously with the deposition thickness as there was no heat sinking applied. The conductive nature of the substrate becomes crucial in affecting H c ( ~ ) . It can be said qualitatively that Nip and carbon coated A1 sub-

Fig. 2. - In-plane coercivity, H c ( I I ) dependence on the CR thickness.

(3)

C8 - 2000 JOURNAL DE PHYSIQUE strates are less conductive than the bare A1 substrates.

The relatively high conductive substrates loose heat to a greater rate and results in lower effective substrate temperature and eventually lower coercivity [I].

Figure 3 shows the dependence of H c ( I ) on substrate

temperature. The substrates were preheated to 100 OC and CoCr was sputtered on top of the heated substrates. H c ( l ) increases from 76 Oe to 186 Oe and

is independent of the substrate when the substrates were preheated. These results demonstrate an already established fact that increasing substrate temperature increases H C ( l ) due to increase in surface column diameter [l, 21. H C ( ~ ) decreases with increased argon pressure and can be attributed to the lowering of the residual stress in the film [3].

I , , , , , _{. .} I

1 1 1 2 1 $ 0 0 0 7 0 1 0 0 0 0 DDIOI

0

: :

:

; Thickness (A)

Fig. 3 . - H C ( l ) dependence on substrate temperature. Figure 4a is a cross-sectional feature of a rough -41 substrate. The peaks and valleys on the substrate can be observed. Figure 4b demonstrates that the columns growing in the valleys grow with a higher nucleation rate than the columns growing on the peaks. Nucle- ation and growth conditions favor the growth of larger columns on the peaks and smaller columns in the valleys. The films growing on rough substrate replicate the roughness of the substrate as shown in figure 4c. Figure 4d exhibits columnar structure for films grown on substrate without any preheat. A fairly thick transition layer (1000 a ) was observed for films deposited on substrates without preheat. Films deposited on

>em

-

an* 2% _as*

ts>

Fig. 4. - TEM cross-sections 1 pm thick CoCr films on rough A1 substrate. (a) Cross-section of A1 substrate. (b) Columns growing on peaks and valleys. (c) CoCr film repli- cating substrate surface. (d) CoCr film grown on surface heated to 100 OC.

substrates with preheat the thickness of the transition layer was observed to decrease with substrate preheat. Figure 5 shows the cross-sectional features of 1 pm thick CoCr films deposited on A1 substrate with 5000

A

of carbon. Figure 5a exhibits columnar cross-section for an equiaxed grain structure between columnar grain structure, terminating about 3000

a

beneath the surface. A void can be clearly observed above the equiaxed structure. The reason for the equiaxed growth can be attributed due to the carbon (on top of Al) film exhibiting certain localised areas where the thermal fluctuations associated with the transforma- tion process and nonuniformity in temperatures can be expected. Equiaxed growth is promoted by higher substrate temperatures. Figure 5b demonstrates columns

I<!"

C I 1cm

d W imy Ib)

Fig. 5. - CoCr films of 1 pm thickness grown on 5000

A

of carbon. (a) Equiaxed structure in between columnar structure. (b) Columns growing in a "rose flower" pattern on a carbon particle.

growing in a "rose flower" pattern on a carbon precipitate. The columns grow initially with a preferred orientation controlled by the precipitate surface and then eventually grow vertically in order to accomodate the incoming particle flux. However, correlations between

Hc(L) and any of the above mentioned growth char-

acteristics was not attempted due to the fact that the volume fraction of the total area exhibiting the above features is unknown.

Conclusions

The effect of the substrate on H C ( ~ ) was observed only for the initial 2000

a

film thickness. The initial layers on any type of substrate exhibit H c ( I I ) as strong as H , ( J ) and with proper underlayer having

(110) orientation can exhibit a fairly high in-plane coercivity. The columns grown on rough substrates exhibit a greater variation of the column diameter about the mean. Comparisons of type of the substrate with average measurements of H C ( l ) and A450 are not jus- tified as the top more highly oriented layers shield the more randomly oriented layers.

[I] Ravipati, D. P., Sivertsen, J. M. and Judy, J. H., IEEE Trans. Magn. (Septembre 1986).

[2] Ouchi, K. and Iwasaki, S., J. Appl. Phys. (April 1986).