The Effects of Dopants on Surface-Energy-Driven Secondary Grain Growth in Ultrathin Si Films

  • PDF / 183,446 Bytes
  • 4 Pages / 420.48 x 639 pts Page_size
  • 75 Downloads / 199 Views

DOWNLOAD

REPORT


Mat. Res. Soc. Symp. Proc. Vol. 54. 11986 Materials Research Society

730

TABLE I Sample labels, implantation doses, implantation energies, approximate film compositions, and the temperature at which films transform to 50% (by area) secondary grains after 20 minute anneals.

Sample

Film Thickness

Dopant(s)

(A)

Dose2 (cm" )

1140A

P

1.45 x 10 cm-

B

1140

P

C

1140

D

1140

F

G

1140

1140

1140

Calculated Composition

Tx=0.5

3

(Cm- ) 16

A

E

Energy

2

(OC) 21

3

90

1 x 10 cm-

1071

1.16 x 1016

90

8x

1020

1097

P

7.22 x 1015

90

5 x 1020

1149

P and( B

1.45 x 1016

90

1 x 1021

1091

P and{ B P and( B undoped

20

15

3.01 x 10

30

2 x 10

1.16 x 1016

90

8x

1020

3.01 x 101"

30

2 x

1020

7.22 x 10'5

90

5x

1020

15

7.52 x 10

30

1129

>1350

20

5 x 10

>1350

0.2 torr of ultrapure Ar. Normal and secondary grain sizes were determined via transmission electron microscopy. Any grain larger than the average normal grain size by more than twice the standard deviation was considered a secondary grain. The fraction of a film transformed to secondary grains, x, was also determined from TEM micrographs. RESULTS Detailed results and analyses of secondary grain growth in these films is given elsewhere. 5,8-9 Here we summarize the evidence that grain growth is enhanced by P doping and that the effect of P doping can be compensated by codoping with B. Shown in figure 1 are the temperatures at which films that have been subjected to 20 minute anneals have undergone secondary grain growth to the point at which 50% of the film is occupied by secondary grains. These temperatures are plotted versus the number of donors minus the number of acceptors. Undoped films and films doped with 5 x 3 1020 cm-3P and 5 x 1020 cm' B did not undergo secondary grain growth even at temperatures as high as 1350'C.

731

(-)

~-1110Figure

0 ,..

1190

1

Temperature

at

which 50% of the area of a

1170 1050

SI

I

I

IlX02 5x 1020 6x20 7x1020 8X1020 9x 1020 3 - *acceptors (cm- ) *donors

film has been consumed by secondary grains after a 20minute anneal. Plotted versus the number of donor atoms minus the number of acceptor atoms. Squares represent films doped with P alone and circles represent films codoped with P and B (see Table I).

DISCUSSION AND CONCLUSIONS

The above results suggest that grain boundary mobility in secondary grain of the Fermi energy.13 This effect is analogous to doping effects on growth is a function self-diffusion' 0 '12 and dislocation motion in silicon. In these cases it has been postulated that increases in the Fermi energy lead to increased numbers or mobilities of charged defects which in turn lead to enhanced kinetics. The same general mechanism is likely to apply here. There is evidence 9 that the rate limiting step for secondary grain growth in the heavily doped films is P diffusion in bulk silicon, (presumably due to solute drag). Since P diffusivity in Si is a function of the Fermi energy, this might explain the observed Fermi energy dependence of grain boundary mobility.