Variations of Thin Metallic Zinc Film Resistances with Sputtering Rate

Variations of resistance of evaporated or sputtered films during deposition have been studied for deposition times lower than five minutes and for resistances higher than 100 [2 sq -1 As our main objective is to study the conduction mechanisms in sputtered films over the thickness range from 200 to 1500 A, (i.e. in the 10 to 100 2 sq -1 sheet resistance range), we report in this note our investigations about the electrical resistance R(T) of zinc films for deposition time, T greater than two minutes and for four average deposition rates. Preparation of films has been described in a previous paper; they are deposited by d.c. diode sputtering of a zinc target (99.9% purity) in an atmosphere of U grade argon. As broken sputtering is equivalent to continuous deposition4, sputtering was stopped every minute to measure the resistance R with a multimeter. The average sputtering rate was determined from the deposition time and the film thickness measured by an optical method; this method is adequate for we have observed very slight anisotropic effects, 6 which seem more important for higher sputtering rates; it varied with the intensity Ie of the glow discharge current, the voltage Ue remaining constant (curves 1, 2, 3 on Figure 2). Variations of the deposition rate as a function of the intensity Ie of glow discharge current for a voltage equal to 1500 V are shown in Figure 1. This curve is in good agreement with the results of Laville Saint-Martin who established that sputtering rate v is given by

report in this note our investigations about the electrical resistance R(T) of zinc films for deposition time, T greater than two minutes and for four average deposition rates.
Preparation of films has been described in a previous paper; 4 they are deposited by d.c.diode sputtering of a zinc target (99.9%purity) in an atmosphere of U grade argon.As broken sputtering is equivalent to continuous deposition4, sputtering was stopped every minute to measure the resistance R with a multimeter.The average sputtering rate was determined from the deposition time and the film thickness measured by an optical method; this method is adequate for we have observed very slight anisotropic effects, 6 which seem more important for higher sputtering rates; 7 it varied with the intensity Ie of the glow discharge current, the voltage Ue remain- ing constant (curves 1, 2, 3 on Figure 2).Variations of the deposition rate as a function of the intensity Ie of glow discharge current for a voltage equal to 1500 V are shown in Figure 1.This curve is in good agreement with the results of Laville Saint-Martin 8 who established that sputtering rate v is given by v =IeUe exp(-AUe -BUe) (1) where A, B are constants.
For high voltage values in the range 1000 to 1750 V, eq. ( 1) may be expressed as v o:Ie deposition time T are plotted in Figure 2. Attempts have been made 1'2 to fit these experimental varia- tions to an empirical equation in the form R(T)=Roo exp K1 +K:T where R. is the limiting value of R(T) when T is large anctK1 and K: are constants.
As we have established that films thicker than 5000 A exhibit bulk properties, 4 their resistance Rb 2.5 2 sq -1 is assumed equal to R. Substituting for R in Eq. 3, this yields R(T) Rb exp K1 + K T T(mn) FIGURE 2 Experimental (dotted lines) and theoretical (full lines) variations of zinc film resistance R, with sputtering rate v, equal to: (1) 80 A mn -1 (2) 110 A mn (3) 130 A mn -1 (4) 550 A mn -K1 and K2 are determined by plotting I[Ln[R(T)/ Rb] versus deposition time T (Figure 3).The slopes of the best fit straight lines determine the values of K2, while K1 is calculated from the intercept with the vertical axis.
From Figure 3 it can be seen that K1 is indepen- dent of the deposition rate as indicated by Eq. 4. Thus, KI 1/Ln[Ro [R] (5 where Ro is the substrate resistance. However the value of Ro calculated from the experimental value of K1 is low (about 6 kfZ).As quasi-linear growth occurs only above the first critical thickness 9 we assume that this value corresponds to a smaller thickness for which the granular structure consists of a large number of empty channels distri- buted throughout the film.It has been shown that the value of the resistance of such a film depends essentially on the geometrical arrangement of the sputtering chamber which remained unchanged in our experiments.
Calculations allow one to determine suitable values of K2, leading to a good agreement between the experimental and theoretical curves (Figure 2) in the resistance range 10 to 100 2 sq -A discrepancy of less than 10% is observed except for the lower sputtering rate but experimental accuracy is low in this case (15%).
The observed slight departures from the theoret- ical resistance at low R values occur at low sputtering rates, a whereas considerable departures have previously been observed by Laville Saint-Martin 7 for higher sputtering rate.We observe (Figure 4) that the magnitude of constant K: differs markedly for different deposition rates.Constant K2 has been defined as a velocity constant related to the sputtering rate v and may be given approximately by K2 OP with a 5.10 -4 A -1 It may be concluded that in these experimental conditions the measured resistance fits the theoretical formulae obtained with the deposition rate as a parameter in the range 80 A mn -x to 600 A mn -x a simple way to predetermine thin film resistances is thus available.

( 2 )FIGURE
FIGUREVariations of the average sputtering rate v with the intensity Ie of the glow discharge.