2.5 Examples

To demonstrate the simulation of an AVC scan and to test the implemented MINIMOS-NT models two devices are simulated. These are a p⁺n and an n⁺p diode with the same geometry. The doping concentration profile is the same for both devices only the doping type is exchanged. For the p⁺n diode the acceptor doping is constant in vertical direction and varies in lateral direction with a maximum value of $\mathrm{N_{A}}= \ensuremath{\mathrm{10^{19}~cm^{-3}}}$. The donor doping is constant throughout the device with a value of $\mathrm{N_{D}}= \ensuremath{\mathrm{10^{16}~cm^{-3}}}$. The metallurgical junction is located at $x = \ensuremath{\mathrm{0.4~\mu m}}$. Fig. 2.9 shows the doping distribution for the p⁺n diode.

The simulated test device consists of a block of silicon which is contacted from both sides and connected to a reference potential. The probing electron beam impinges on the top surface of the semiconductor. Fig. 2.10 shows the geometry of the test devices.

For the simulation of each AVC scan approximately 60 device simulations are performed. A constant electron beam energy of 3 keV is assumed for all calculations and the beam current is varied between 10 pA and 1 nA. The beam diameter has a value of 70 nm. There is nearly no difference between the extracted potential for a beam current of 10 pA and the equilibrium potential. Beam currents of the order of 1 nA cause a considerable change in the potential and simulation convergence becomes very slow or even impossible because of the strong carrier concentration variations.

Figure 2.9: Doping distribution of the simulated p⁺n diode. The metallurgical junction is located at x = 0.4 ^. 10^-6 m.

Figure 2.10: Geometry of the simulated test devices.