next up previous contents
Next: B7. Error Function Up: B. Optimization Previous: B5. Gauß Function


B6. Pearson Function

For a realistic description of dopant profiles the Pearson Type IV distribution is well established today. The Pearson function family is based on the differential equation


\begin{displaymath}
\frac{df(y)}{dy} = \frac{y - a}{b_0 + b_1 y + b_2 y^2} f(y),
\quad y=x-R_p
.
\end{displaymath} (B9)

with the coefficients


\begin{displaymath}
a = - \frac{\gamma_1 \sigma_p (\beta_2 + 3)}
{10 \beta_2 - 12 \gamma_1^2 - 18}
\end{displaymath} (B10)


\begin{displaymath}
b_0 = - \frac{\sigma_p^2 (4 \beta_2 - 3 \gamma_1^2)}
{10 \beta_2 - 12 \gamma_1^2 - 18}
\end{displaymath} (B11)


\begin{displaymath}
b_1 = - \frac{\gamma_1 \sigma_p (\beta_2 + 3)}
{10 \beta_2 - 12 \gamma_1^2 - 18}
\end{displaymath} (B12)


\begin{displaymath}
b_2 = - \frac{2 \beta_2 - 3 \sigma_1^2 - 6)}
{10 \beta_2 - 12 \gamma_1^2 - 18}
\end{displaymath} (B13)

where Rp, $\sigma$, $\gamma$, and $\beta$ are the projected range, standard deviation, skewness and kurtosis, respectively. The solution for the Pearson Type IV function is

\begin{displaymath}
f(y) = K \left ( - \left ( b_0 + b_1 y +
b_2 y^2 \right )...
... b_2 y - b_1}{\sqrt{4 b_2 b_0 - b_1^2}} \right ) }
\right ) }
\end{displaymath} (B14)


\begin{displaymath}
0 < \gamma_1^2 < 32
\end{displaymath} (B15)


\begin{displaymath}
\beta_2 > \frac{39 \gamma_1^2 + 48 + 6(\gamma_1^2 + 4)^{\frac{3}{2}}}
{32 - \gamma_1^2}
\end{displaymath} (B16)

The maximum value is reached for x = Rp. The projected range Rp is the average depth of the distribution. Skewness $\gamma$ measures the asymmetry of the distribution, a positive value places the peak closer to the surface than the projected range. The kurtosis $\gamma$ describing the flatness of the top of the distribution.


next up previous contents
Next: B7. Error Function Up: B. Optimization Previous: B5. Gauß Function

R. Plasun