nextnano³ - Tutorial

next generation 3D nano device simulator

1D Tutorial

GaAs / AlGaAs  -  inverted High Electron Mobility Transistor (HEMT)

last updated: 17-02-10

Author: Stefan Birner
Please send comments to nextnano3 (-at-) wsi.tum.de.

GaAs / AlGaAs  -  inverted High Electron Mobility Transistor (HEMT)

• Here is the input file: invertedHEMT.in

1. Step 7: GaAs / AlGaAs  -  inverted High Electron Mobility Transistor (HEMT) - MBE doped
2. Please edit your keywords.in file and set the name of the file to be read in to invertedHEMT.in.
3. The sample is 1137 nm, pseudomorphically grown on GaAs.
   The interesting area is between 400 nm and 900 nm.
4. Again, we perform a one-dimensional simulation.
5. Just a reminder: If you need additional information about the keywords and their specifiers, you can look it up here.
6. The quantum region is over the whole device.
7. The inverted HEMT device looks like this:
   
   _________________________________________________________________
        1             2            3               4                5           6            7           8
     GaAs       GaAs    AlGaAs   AlGaAs:Si   AlGaAs    GaAs   GaAs:Si   metal
    200 nm       300        260            66               61        300         15          50
   _________________________________________________________________
                                                                                                           ohmic contact
   
   
8. Al_xGa_1-xAs
   We choose x to be 0.34.
   In principle, we have a short period lattice (SPS)
   SPS GaAs/AlAs (2 nm / 1 nm)
   but we consider it due to the short period to be stoichiometric equivalent to Al_0.34Ga_0.66As which should simplify our calculations.
   It improves the thermic resistence of the structures which underly a thermic annealing step for activation of the doping materials.
9. The flow scheme is 2:
      1. calculate nonlinear Poisson as specified in input
     (2. calculate current as specified in input)
   For this example, we don't calculate the current.
10. Output
    - The band structure will be saved into the directory band_struc1/
- The densities will be saved into densities1/
- The strain will be saved into strain1/
- Raw data will be saved into raw_data1/
- kp data will be saved into kp_data1/
11. We are interested in plotting conduction band 1 and the resulting electron density to visualize the two 2DEGs (two dimensional electron gas) which should look somewhat like this:
    
    The black curve shows conduction band 1 and the red curve shows the electron density. Clearly, one can see that there are two triangular "bags" in the conduction band at GaAs/AlGaAs interfaces. If these bags lie below the Fermi energy, 2DEGs (two dimensional electron gases) can be formed. By variation of the doping concentration one should be able to generate zero, one or two 2DEGs. By increasing the doping concentration one should observe an increase in the number of electrons in the left GaAs/AlGaAs interface, which is not desired. The task is to avoid this and to get an 2DEG on the right side by choosing an appropriate doping profile.
12. The 15 nm GaAs layer is doped with a constant n-type Si doping (2.0*10¹⁸ cm^-3). The donor level of Si in GaAs lies 5.8 meV below the conduction band.
    
    $doping-function

 doping-function-number = 2             ! acts as separator
 impurity-number        = 1             ! properties of this impurity type have
                                        ! to be specified later
 doping-concentration   = 2d0           ! 2.0*10^18 cm^-3
 only-region            = 1187d0 1202d0 ! actually, only boarders of a line,
                                        ! rectangle, cube are allowed
$end_doping-function


$impurity-parameters

 impurity-number             = 2          ! number, 1 or 2 ... (impurity numbers labeled in doping-function)
 impurity-name               = Si-in-GaAs ! a name (for later use - planned to read parameters from data base)
 impurity-type               = n-type     ! n-type, p-type, trap
 number-of-energy-levels     = 1          ! number of energy levels of this impurity
 energy-levels-relative      = 0.0058d0   ! energy relative to 'nearest' band edge
                                          ! (n-type -> conduction band, else valence band)
                                          ! Si in GaAs: 5.8 meV below conduction band
 degeneracy-of-energy-levels = 2          ! degeneracy of energy levels (2 for n-type, 4 for p-type)

$end_impurity-parameters
13. The interior doping looks like this (to be revised!!).
    Experimental details: Si is implanted by FIB (focused ion beam) on an amorphous As layer which will be removed after the implantation process.
    
    
    
    $doping-function

 doping-function-number     = 1
 impurity-number            = 1
 base-function-1            = gauss-1d                          ! a valid base function name
 apply-function-1-along-dir = 0 0 1                             ! (0 0 1) , (0 1 0) , (1 0 0)
 parameters-base-function-1 = 792.677d0 71.064d0 0.0d0 16.955d0 ! center-coordinate width minimum-value maximum-value
 doping-concentration       = 1.5d0                             ! doping concentration refers to that position
 only-region                = 672d0 822d0
 position                   = 792.677d0

$end_doping-function

$impurity-parameters

 impurity-number             = 1
 impurity-name               = Si-in-GaAs
 impurity-type               = n-type
 number-of-energy-levels     = 1
 energy-levels-relativ       = 0.0058d0
 degeneracy-of-energy-levels = 2

$end_impurity-parameters
14. A plot of the potential looks like this:
    

• Please help us to improve our tutorial. Please send comments to nextnano3 (-at-) wsi.tum.de.