Performance of a-Si-H Solar Cells at Higher Growth Rates

  • PDF / 343,891 Bytes
  • 6 Pages / 382.5 x 615.6 pts Page_size
  • 51 Downloads / 177 Views

DOWNLOAD

REPORT


Mat. Res. Soc. Symp. Proc. Vol. 557 ©1999 Materials Research Society

6

6

(b)

0

(a)

5-

SP,D,F

5

2

Cu

0u

#3

1.5P 2F, D,

...2D, 2D P P. FF

•1

1.5P, F, 0 .5,F0

"

P, 2 F, D

.

SP,5DFF

P.....

P/2, F,2D 0

0

1

I,

I

2

3

.

4

5

0

Relative Plasma Power (C)

1

2

3

4

5

Relative Plasma Power (C)

Fig 1 (a) & (b). The growth rate of i-layers plotted as function of the relative plasma current at different relative pressures (P), hydrogen dilutions (D), and flow rates (F). The dashed lines are best fits to data through the origin. The continuous lines indicate saturation due to gas depletion. 0.27 cm 2 devices, but the short circuit current densities (Jsc) were obtained from quantum efficiency measurements. RESULTS The growth rate of a-Si:H is expected to vary as the concentration of SiH 3 radicals where [7] [SiH 3 ] = N 3 Ve3 0e3 [Si4 1] T3 /

(Net Vet aet +

{l/T4}),

where the [ ] denote concentration, Ne, ve and ae are the electron density, velocity and cross section respectively, and Tris the time constant. The subscript 3 denotes SiH 3, 4 denotes Sill4 and t denotes total. The data in Fig. 1 are consistent with the expression above in that the dashed lines show that the growth rate increases with plasma power (Ne3 v.3 cra3) when the power level is low (Net Vet Get > {1/r4}), the growth rates saturate as shown by the continuous lines in Fig. 1(b). It has been suggested that three potential causes for deterioration of performance with increasing plasma power are higher silane related radicals (SinH2n+i, n>1), short-lifetime radicals (SiHl, x=1,2), and ion-bombardment energies [8]. While discussing the results we will evaluate the possible contributions of these three factors. The initial efficiencies of single junction devices are shown as function of i-layer growth rates in Fig. 2. It is clear that increasing the growth rate reduces the initial efficiency of the devices. Furthermore, increasing hydrogen dilution and/or decreasing the pressure leads to a faster decrease of the efficiency with increasing growth rate. One common factor between these two variables is that they decrease the partial pressure of silane. This would imply an increase of the contribution to growth of short-lifetime radicals relative to Sill3 and/or an increase of ion-bombardment energies. We find that the decrease in initial efficiency with increasing growth rate is mainly due to decrease of short-circuit current (Jsc). In Fig. 3(a) we

128

10

......P . .. -....... ------------------. ................

..... ...... 9..

".... .. .. . :: X i;. . . . . .. .. .. .. .. .. .. ..----

v°"8

c_ .A

A

.

.. . . . . .. . . ........ ..... ...........- ....

. ...............- .- .. --. .. .. . .. ..... ................

6

P; 2 *P;2D

. . . .. . .. ..... ....... •:.......... . ..:..........................

P; 5D

oU

4

w5 A

.. -..--..- .. -.. .. .. .. .. .. .. .. .. .. .. ..--..- -

3

.. ...

...

. . . -. . . . .

. . .... ...

I

0

.. . . . . ..........

. .

.. . .

I

1

2

x

. . . . . ./

I

3 4 Growth Rate (A/s