Thickness Effect on the Crystallization Kinetics of Electrolessly-Deposited Ni(P) Thin Films
- PDF / 306,969 Bytes
- 6 Pages / 414.72 x 648 pts Page_size
- 99 Downloads / 182 Views
L.T. Shi and E.J.M. O'Sullivan IBM Thomas J. Watson Research Center, Yorktown Heights, NY 10598 ABSTRACT
In order to understand thickness and interfacial effects on the crystallization kinetics of amorphous solids, Ni(P) thin films electrolessly deposited on Cu seed layers were annealed at constant heating rates or at constant temperatures in a DSC to obtain activation energies and Avrami exponents. It was found that the activation energy of crystallization in Ni(P) changes as a function of sample thickness when the sample thickness is less than 1.0 pm. Furthermore, the Avrami exponent was found to change not only as a function of thickness but also as a function of annealing temperature. INTRODUCTION
Because of its relatively low deposition cost, technological importance, and simple amorphous phase, the crystallization of electrolessly-deposited (ELD) studied using various analytical techniques' - 1. It has Ni(P) has been extensively been determined3 7,, 8 that the crystallization kinetics of Ni(P) depend on the phosphorus content in the sample. However, some controversy still exists in the reported results measured from samples with almost identical composition; for example, the activation energy of crystallization for the ELD Ni(P) with 11 wt% P reported in ref. 2 (227 KJ/mole) is very different from that reported in ref. 8 (150-160 KJ/mole). In addition, the interfacial effect on phase transitions in thin films is an interesting subject. Its understanding is essential to many technological applications. In an attempt to account for the above discrepancy and to gain some understanding about the interfacial effect, we have conducted a kinetic study of crystallization in ELD Ni(P) samples with various thicknesses using differential scanning calorimetry (DSC). EXPERIMENTAL
ELD Ni(P) films were deposited from an electrolessly plating bath with a pH value of 8.1 and maintained at 73"C. The bath solution contained NiSO4, citric acid as complexant, boric acid as buffer, sodium hypophosphite as reducing agent and Pb as stabilizer. The plating rate was estimated to be 175 A/min. Films were deposited on a 2 pm thick, sputtered Cu layer on a Si wafer, with 400A Cr as an adhesion promoting layer (to Si wafer). Prior to plating, the Cu surface was catalyzed by immersion in an acidic, aqueous, Pd salt solution. The plated about 11.5 wt% P and exhibited an amorphous structure. Ni(P) films contained 0 It is well known4 9, -1 that as-deposited Ni(P) films containing P greater than 7-8 wt% are generally amorphous. Calorimetry analysis was conducted in a modified DuPont 910 DSC under a dry and pure nitrogen environment. Two different sample forms were used for annealing tests at constant heating rates: one was a free standing film detached from the Si substrate, and the other a film attached to the Si substrate (i.e. in situ). The latter was cut from the wafer in a size which could be fitted in the DSC pan, and was tested (without crimping the pan) with the Ni(P) film side facing down and contacting the pan. The purpose of usin
Data Loading...
