• PDF / 1,026,125 Bytes
• 5 Pages / 417.6 x 639 pts Page_size
• 107 Downloads / 218 Views

DOWNLOAD

REPORT

---

251

Mat. Res. Soc. Symp. Proc. Vol. 564 © 1999 Materials Research Society

The CVD experiments are performed using a laboratory scale, warm-wall pedestal reactor. The Cu(fod) 2 or Cu(hfac) 2 precursor is introduced into a H2 carrier gas stream using a temperature-controlled evaporator. The precursor partial pressure is calculated from the weight loss of the evaporator, the carrier gas flow rate, and the total pressure in the reactor. Films are deposited onto TiN-coated Si(100) substrates. Growth rates and induction times are determined from plots of substrate weight change vs. deposition time for different operating conditions. For convenience, the growth rates are reported in units of equivalent thickness (i.e., 1 mg cm-2 hr-1 = 18.6 nm min-1 for a fully dense film). Plan view scanning electron microscope images and four-point probe resistivities are obtained for each sample. RESULTS Cu(fod) 2 Kinetic Results The deposition kinetics observed using Cu(fod) 2 are summarized in Figure 1, which plots the steady state deposition rate as a function of Cu(fod) 2 partial pressure for two different values of H2 partial pressure and a fixed substrate temperature of 300 'C. The Cu(fod) 2 partial pressure is varied from 1.6 to 2.9 Torr, using evaporator temperatures over the range 140-150 'C. The highest growth rate, 70 nm min-', is obtained using a Cu(fod) 2 partial pressure of 2.9 Torr and H2 partial pressure of 80 Torr. The apparent kinetic reaction orders are also demonstrated in Figure 1. For both values of H 2 partial pressure, the dependence of growth rate on Cu(fod) 2 partial pressure can be described by a linear fit passing through the origin. This indicates the reaction kinetics are first-order with respect to Cu(fod) 2 . The deposition rates using H 2 partial pressure of 80 Torr are about 4 times greater than for a H 2 partial pressure of 40 Torr. This indicates the kinetics are second-order with respect to H 2. We also briefly examined the influence of substrate temperature. Using a temperature of 350 °C with Cu(fod) 2 partial pressure of 1.6 Torr and H 2 partial pressure of 80 Torr, we obtained a growth rate of 70 nm min-. Compared to the growth rate of 30 nm min-' observed at 300 'C, this gives an apparent activation energy of 50±10 kcal/mole.

70 60

Ts = 3000C

"LE 50

PH = 40 Torr PH2 = 80 Torr

E 40 a 30i

o 20 C.

10. 0~

0

0.5

1

1.5

2

2.5

Cu(fod)2 Partial Pressure (Torr)

Figure 1

Dependence of copper deposition rate on Cu(fod) 2 partial pressure (Other conditions: Ts = 300 0C; U: PH2 = 80 Torr, EIl: PH2 = 40 Torr).

252

= 2gm

5 min (0.05 gm) Figure 2

15 min (0.3 gm)

30 min (0.8 jLtm)

Plane view SEM images of copper films deposited using Cu(fod) 2 . Deposition conditions: Ts = 300 °C, PCu(fod)2 = 1.7 Torr, PH2 = 80 Torr.

Cu(fod) 2 Film Properties The evolution of film microstructure is illustrated by the sequence of SEM images shown in Figure 2. The growth conditions for these films were Cu(fod) 2 partial pressure = 1.7 Torr, H 2 partial pressure = 80 Torr, and substrate temperature = 300 °