Band gaps from conductivity measurements

Hi,

I have a question on calculating a band gap from conductivity measurements.

$G = G_0exp(-E_g/2k)$

So

$lnG = lnG_0 - \frac{-E_g}{2k}$

So plotting lnG against 1/T gives a gradient of -Eg/2k

I've plotted the data in the question as above and have a gradient of 6.0x10^-5 which gave a band gap of 1.66x10^-27 J which doesn't seem right!

Assuming my units of J are correct, converting to eV gives 1.04 x 10^-4 eV which still doesnt seem right!

The data is:

G (ohms^-1 cm^-1) T (K)

0.47 ----------------- 2573
1 ------------------- 2953
1.64 ----------------- 3223

Thanks for any help! And sorry if this is meant to be in the Physics section!