Parameter Screening¶
Simulation parameters can be screened in order to study device performance under different conditions, or to search for desired material properties.
Create Materials¶
We start by creating the materials that are used in the stack.
NPD has a HOMO energy of -5.45 eV and a LUMO energy of -1.4 eV
mCBP has a HOMO energy of -6 eV and a LUMO energy of -1.5 eV
Ir(dmp)3 has a HOMO energy of -5 eV and a LUMO energy of -1.7 eV
For this tutorial, we will focus on charge transport only. The Transport template can be used when creating new materials. We will use a Gaussian DOS with a standard deviation of 0.1 for both polarons. Excitonic processes will be omitted for now. The exciton energy levels can therefore be put to 0 for all materials.
Create Compositions¶
Navigate to the Compositions tab to access the compositions that make up the device layers. Pure compositions for each of the compounds should be available.
In addition, we are going to create a new host-guest mixture. Select the New Composition option. Provide a name for the composition and select the Basic template. Selecting the Save option will bring you to the composition editor.
You will see a warning in the editor stating that the mole fractions do not sum to 1. This warning will be displayed until the composition is fully defined.
We will create a mixture of 0.9 mCPB and 0.1 Ir(dmp)3. To add a component to the mixture, simply select the material and the desired fraction. The Create Fraction button will add the material to the mixture. It is also possible to adjust or remove fractions that were previously included by selecting the editor icons in the fractions list.
Create a Stack¶
We will create a stack using 3 layers. The outer contact layers are composed of pure NPD. The inner emitter layer contains the host-guest mixture defined earlier.
Create a 10 nm NPD layer
Create a 60 nm layer containing the mCPB-Ir(dmp)3 mixture
Create another 10 nm NPD layer
You will end up with an 80 nm stack.
Create a Parameter Set¶
For this simulation, we are interested in investigating the device performance at different voltages. We will therefore use the Voltage Sweep template when creating our parameter set.
Give a name to the parameter set and add the stack that was created in the previous step. The voltage will be set as part of the parameter screening and does not need to be chosen at this step.
The Fermi levels of the electrodes are automatically matched to the material parameters. These levels can be adjusted to those of the external contacts. We will use an electrode energy level of -5.25 at the anode and -1.2 at the cathode. By using an energy barrier of 0.2 eV compared to the NPD polaron energy levels, we are reducing the rate of polaron exchange processes with the electrodes. This biases the kMC simulation, increasing the number of samples that contain transport processes compared to electrode exchange.
Note
Care should be taken that this artificial barrier does not affect the device statistics. This can be achieved by analyzing the change in device properties when varying the barrier height. Because electrode exchange processes are localized at the edges of the device, a single-layer simulation can be used to perform these screenings, significantly reducing the simulation costs.
Note
When operating under high currents, the device sensitivity to the barrier height may become voltage-dependent. Make sure to verify the validity of your data at both extremities of the investigated voltage range.
The Modules tab allows you to enable different processes that are allowed to occur during the simulation. You can disable the excitonics module as we will focus on polaron transport for this tutorial.
We will set the number of simulation steps to 1.000.000.000 in the Termination Criteria tab. On the Output tab, we set the report interval to 100.000 and the output interval to 1.000.000. The parameter set can now be saved.
Starting the Simulation¶
Navigate to the Simulations tab. Specify a name for the simulation and check that the correct server is selected for running the job. Then, select the voltage sweep parameter set created in the previous step. For tutorial purposes, we can set the disorder instances to 1.
Parameter screenings can be specified as part of the simulation setup. The screening allows evaluation of the device behavior for different parameter values. Aside from the sweep parameter, all other conditions will remain unaltered from the default parameter set.
The screening values are obtained through linear interpolation. Minimum and maximum values for the screening parameter are selected and the total number of screening steps is chosen. A uniform stepsize between parameter values is calculated.
For this example, we select the voltage as our screening parameter. The minimum and maximum voltages are set to 1 and 5 V respectively. By choosing 3 voltage steps, this will prompt the screening to perform simulations at 1, 3 and 5 V. A preview of these screening conditions is provided in the web interface.
Note that the number of disorder instances is applied to each screening step. The default of 5 disorder instances would therefore yield 3x5 independent jobs. By using only a single disorder instance, the number of simulation jobs is limited to 3.
We now submit the simulation and wait for the screening steps to complete.
Simulation Output¶
Current-voltage characteristics can be viewed in the OLED Report section of the Sweep Report panel.
Tip
After performing an initial screening, it is possible to add additional sample points. After selecting a simulation, the Sweep tab contains the Add Sweep option to request additional jobs. The output from these jobs is then automatically collected and appended to the current simulation. This option was used in the previous tutorial to improve simulation statistics, but can also be used to extend the sweep range or to add additional points to the parameter screening.