previous up next contents Previous: 5.2 Effect of Pre-amorphization Up: 5. Applications Next: 5.4 Implantation into Topological


5.3 Pollution of Silicon by the Implantation through
SiO$ _2$ Layers

By implanting through a silicon dioxide layer oxygen atoms are knocked into the silicon substrate by the implanted ions. These oxygen atoms can influence the electric behavior of a transistor because they are located in the active area. Additionally the oxygen atoms can convert parts of the substrate into SiO$ _2$ during an annealing step which follows the ion implantation. Thereby, for instance, the thickness and the shape of the gate can be modified as a consequence of an ion implantation.

Figure 5.20: Distribution of oxygen atoms in the silicon substrate as a consequence of the implantation of arsenic ions with a dose of $ 3{\cdot }10^{15}$cm$ ^{-2}$ and with energies ranging from 10 keV to 150 keV through a 11 nm thick SiO$ _2$ layer. A tilt of 7 $ \,^\circ\;$and a twist of 90 $ \,^\circ\;$was used.
\begin{figure}\begin{center}
\psfrag{Depth \(um\)}[c][c]{\LARGE\sf Depth ($\math...
...}{\includegraphics{fig/appli/Oxygen.eps}}}\end{center}\vspace*{-4mm}\end{figure}

Figure 5.21: Distribution of arsenic atoms in the silicon substrate as a consequence of the implantation of arsenic ions with a dose of $ 3{\cdot }10^{15}$cm$ ^{-2}$ and with energies ranging from 10 keV to 150 keV through a 11 nm thick SiO$ _2$ layer. A tilt of 7 $ \,^\circ\;$and a twist of 90 $ \,^\circ\;$was used.
\begin{figure}\begin{center}
\psfrag{Depth \(um\)}[c][c]{\LARGE\sf Depth ($\math...
...{\includegraphics{fig/appli/Arsenic.eps}}}\end{center}\vspace*{-4mm}\end{figure}

Figure 5.22: Depth of the concentration level of $ 1{\cdot }10^{21}$cm$ ^{-3}$ of the oxygen distribution after an implantation with arsenic ions with a dose of $ 3{\cdot }10^{15}$cm$ ^{-2}$ through a 10 nm thick SiO$ _2$ layer. The depth is plotted as a function of the energy of the arsenic ions.
\begin{figure}\begin{center}
\psfrag{Depth \(nm\)}[c][c]{\LARGE\sf Depth (nm)}
\...
...includegraphics{fig/appli/Lev_1e21.eps}}}}\end{center}\vspace*{-4mm}\end{figure}

Figure 5.23: Distribution of oxygen atoms in the silicon substrate as a consequence of the implantation of arsenic ions with a dose of $ 3{\cdot }10^{15}$cm$ ^{-2}$ and with an energy of 75 keV through a 10 nm thick SiO$ _2$ layer. A tilt of 7 $ \,^\circ\;$ and a twist of 90 $ \,^\circ\;$was used. For the displacement energy of the oxygen atom from a stable bound in the silicon dioxide is varied from 15 eV to 75 eV.
\begin{figure}\begin{center}
\psfrag{Depth \(um\)}[c][c]{\LARGE\sf Depth ($\math...
...\includegraphics{fig/appli/Displace.eps}}}\end{center}\vspace*{-4mm}\end{figure}

In Fig. 5.20 to Fig. 5.22 the oxygen distribution in the silicon substrate as a result of an implantation of arsenic ions with a dose of $ 3{\cdot }10^{15}$cm$ ^{-2}$ and with energies varying from 10 keV to 150 keV is analyzed.

Fig. 5.20 shows the distribution of the oxygen atoms. One interesting aspect is that the oxygen distribution just slightly varies with the implantation energy of the arsenic ions. While the arsenic ions penetrate significantly deeper into the target as show in Fig. 5.21 by increasing the implantation energy, the depth of the oxygen atoms even decreases above an implantation energy of 30 keV. The dependence of the penetration depth of the oxygen atoms on the implantation energy of the arsenic ions is analyzed in Fig. 5.22 by evaluating the depth of an iso-concentration level as a function implantation energy. A concentration level of $ 1{\cdot }10^{21}$cm$ ^{-3}$ is used, which corresponds to the tail region of the oxygen distribution.

The depth of the tail region has a maximum at an implantation energy around 30 keV as shown in Fig. 5.22. The stars in Fig. 5.22 represent the simulation results while the red line denotes an analytical regression of the simulated data. This analytical function averages the statistical noise in the simulation results which is inevitable due to the simulation method. Since the statistical noise of the results presented in Fig. 5.22 is of the order of 1 nm, two implantation with a slightly different number of ions have been performed for some ion energies. Due to higher number of results a more accurate regression function could be extracted. The analytical function for the depth $ d(E)$ of the concentration level of $ 1{\cdot }10^{21}$cm$ ^{-3}$ in nm which describes the regression is presented in (5.6). The energy unit applied in (5.6) is keV and the function is valid in the energy range from 10 keV to 150 keV.

\begin{displaymath}\begin{split}\frac{d(E)}{\mathrm{nm}} = & \;5.34586 + \left(-...
...left(-0.0690593\cdot \frac{E}{\mathrm{keV}} \right) \end{split}\end{displaymath} (5.6)

For the above simulations a displacement energy of 25 eV for removing an oxygen atom from a stable position in silicon dioxide was assumed. This value is just an estimated value as suggested in [44], while the actual value can be in the range up to 200 eV. Since the displacement energy is suggested to have a strong influence on the oxygen distribution, this influence has been analyzed by simulation. The result of this analysis is presented in Fig. 5.23. Simulations have been performed with five different displacement energies in the interval from 15 eV to 75 eV and an implantation energy of 75 keV has been used. According to the simulations mainly the oxygen concentration near the silicon-silicon dioxide interface is reduced by increasing the displacement energy, while the concentration in the tail region remains almost unchanged.

Worth mentioning concerning this application is that the real oxygen distribution is probably slightly deeper than the simulated distribution, because a knock-on of oxygen atoms located in silicon by succeeding ions is neglected in the simulations, but the oxygen concentration in silicon is very high and thereby the probability for a collision with an oxygen atom is also high.

previous up next contents Previous: 5.2 Effect of Pre-amorphization Up: 5. Applications Next: 5.4 Implantation into Topological


A. Hoessiger: Simulation of Ion Implantation for ULSI Technology