5.2 Effect of Pre-amorphization

Another method to reduce the channelling effect is to perform two ion implantations immediately one after the other without an annealing step in between. The first implantation is used to destroy the crystalline structure of the substrate. For the implantation ion species are used which are not electrically active. Otherwise a very high doping concentration would be generated due to the high implantation dose which is necessary to achieve an amorphization. Silicon and germanium are the preferred ion species. The second implantation introduces the required dopant atoms.

Fig. 5.16 and Fig. 5.17 demonstrate the influence of the amorphization by the first implantation on the doping profile generated by the second implantation. For this analysis the pre-amorphization was performed by silicon ions with energies ranging from 50 keV to 400 keV. Additionally the dose is varied in Fig. 5.16 and Fig. 5.17, from $5{\cdot }10^{14}$ cm $^{-2}$ to $5{\cdot }10^{16}$ cm $^{-2}$ . Finally the dopant atoms are introduced by a boron ion implantation with an energy of 90 keV.

According to the simulation results the penetration depth of the boron ions is reduced by increasing the energy and the dose of the first implantation. The change in the penetration depth saturates as well by increasing the dose as the energy above a certain threshold level as shown in Fig. 5.18. The top figure of Fig. 5.18 shows the depth of a concentration level of $1{\cdot }10^{15}$ cm $^{-3}$ , while the bottom figure shows the depth of a concentration level of $5{\cdot }10^{16}$ cm $^{-3}$ as a function of the ion energy for various doses of the silicon implantation. This saturation effect can be explained by a useful limit in the size of the amorphous area. The maximal useful size of the amorphous area is of the order of the penetration depth of the ions of the second implantation.

**Figure 5.16:** Simulated boron concentration resulting from an implantation with boron ions with an energy of 90 keV and a dose of $5{\cdot }10^{13}$ cm $^{-2}$ after an implantation of silicon ions with doses of $5{\cdot }10^{14}$ cm $^{-2}$ (top) and $1{\cdot }10^{15}$ cm $^{-2}$ (bottom) and with various ion energies. Both ion beams were tilted by 7 $\,^\circ\;$ and rotated by 90 $\,^\circ\;$ .
$\begin{figure}\begin{center} \psfrag{Depth $um$}[c][c]{\LARGE\sf Depth ($\math... ...\includegraphics{fig/appli/Pre_1e15.eps}}}\end{center}\vspace*{-4mm}\end{figure}$

**Figure 5.17:** Simulated boron concentration resulting from an implantation with boron ions with an energy of 90 keV and a dose of $5{\cdot }10^{13}$ cm $^{-2}$ after an implantation of silicon ions with doses of $5{\cdot }10^{15}$ cm $^{-2}$ (top) and $1{\cdot }10^{16}$ cm $^{-2}$ (bottom) and with various ion energies. Both ion beams were tilted by 7 $\,^\circ\;$ and rotated by 90 $\,^\circ\;$ .
$\begin{figure}\begin{center} \psfrag{Depth $um$}[c][c]{\LARGE\sf Depth ($\math... ...\includegraphics{fig/appli/Pre_1e16.eps}}}\end{center}\vspace*{-4mm}\end{figure}$

**Figure 5.18:** Depth of the boron concentration levels of $1{\cdot }10^{15}$ cm $^{-3}$ (top) and $5{\cdot }10^{16}$ cm $^{-3}$ (bottom) resulting from an implantation with boron ions with an energy of 90 keV and a dose of $5{\cdot }10^{13}$ cm $^{-2}$ after an implantation with silicon ions. The depth is plotted as the function of the energy and for different doses of the silicon implantation.
$\begin{figure}\begin{center} \psfrag{5e14}{\LARGE \sf $\mathsf{5{\cdot}10^{14}}$... ...ncludegraphics{fig/appli/Range_5e16.eps}}}\end{center}\vspace*{-4mm}\end{figure}$

**Figure 5.19:** Simulated boron concentration resulting from an implantation of boron ions into a MOS transistor structure after the formation of the spacer. The implantation was performed with an energy of 80 keV and a dose of $1{\cdot }10^{15}$ cm $^{-2}$ . In the bottom figure the boron implantation is preceded by an implantation with silicon ions with an energy of 200 keV and a dose of $1{\cdot }10^{16}$ cm $^{-2}$ . Both ion beams were tilted by 10 $\,^\circ\;$ and rotated by 270 $\,^\circ\;$ . The figure shows the doping profile in the vicinity of the gate corner.

By increasing the dose, first the size of the amorphous area increases, staring from a small area around the peak of the point defect distribution, which is close to the peak of the distribution of the implanted silicon ions. Until the amorphous area reaches the surface of the substrate the size of the amorphous area changes quite rapidly by increasing the dose. At higher doses the amorphous area still slightly extends to the maximal penetration depth of the silicon ions.

By increasing the energy mainly the depth of the center of the amorphous area is moved deeper into the target. As long as the amorphous area is smaller than and lies within the range of the atoms of the second implantation, a dependence of the distribution of the atoms of the second implantation on the energy and the dose of the first implantation can be observed.

This explains why the influence of the dose of the first implantation increases until an implantation energy of 200 keV as show in Fig. 5.18. At this energy the center of the amorphous area lies approximately at the same depth as the peak of the boron distribution. The maximal useful dose of the silicon implantation is about $1{\cdot }10^{16}$ cm $^{-2}$ . At this dose the amorphous area has filled the whole useful area. A further increase of the dose just extends the amorphous area outside of the useful area and has therefore no influence on the distribution of the boron ions.

The influence of pre-amorphization on the shape of the doping profile in the vicinity of a gate corner is studied in Fig. 5.19. An implantation with boron ions into a MOS transistor structure after the formation of the spacer is performed with (bottom) and without (top) pre-amorphization by an implantation with silicon ions. The boron ions are implanted with an energy of 80 keV and a dose of $1{\cdot }10^{15}$ cm $^{-2}$ , and the silicon ions are implanted with an energy of 200 keV and a dose of $1{\cdot }10^{16}$ cm $^{-2}$ . While the vertical shape and the depth of the peak of the doping profile remains almost unchanged, the gradient in the tail region is significantly increased as expected already from the one-dimensional simulations. The distance of the iso-concentration line of $5{\cdot }10^{17}$ cm $^{-3}$ from the gate corner is 125 nm for the boron implantation without pre-amorphization while it is 140 nm for the implantation after pre-amorphization, if the lateral size of the spacer is 72 nm. Without pre-amorphization the depth of this iso-concentration line is 545 nm and with pre-amorphization the depth is 365 nm. This results in an aspect ratio of 4.36 and 2.6 respectively. The peak of the boron concentration is approximately at a depth of 240 nm in both cases.