Simple OLED

Introduction

In this tutorial we will simulate the operation of an idealized OLED using lightforge with parametric input. The OLED consists of an hole transport layer (HTL), a emission layer consisting of a host material and dye, an electron transport layer (ETL) and two electrodes. For simplicity we use the same material as host HTL and ETL. We will analyse IV characteristics, internal quantum efficiecies as well as exciton formation and quenching.

Tutorial files:

The settings file can be created manually with the instructions given below or be downloaded here.

Prerequisites

Make sure the environment varibales neccesary to run lightforge are set as shown here.

Settings

Create a file called "settings" with the following content:

pbc: [False, True, True]
excitonics: use presets
connect_electrodes: True
coulomb_mesh: True

Periodic boundary conditions are switched off in transport direction as electrodes are attached. Exciton transport is treated by chosing from typical predefined types of molecules. Coulomb_mesh is required to treat long range periodic coulomb interaction, if two metallic contacts are present.

particles:
 holes: True
 electrons: True
 excitons: True

morphology_width: 20

We require all types of particles. The system has a width of 20 nm.

materials:
- name: host
  input_mode_transport: "PAR: eaip,sig,l"
  molecule_parameters:
    exciton preset: fluorescent
    energies:
    - [5.8,2.0]
    - [0.1,0.1]
    - [0.2,0.2]

- name: emitter
  input_mode_transport: "PAR: eaip,sig,l"
  molecule_parameters:
    exciton preset: phosphorescent
    energies:
    - [5.5,2.3]
    - [0.09,0.09]
    - [0.2,0.2]

We define two materials. One host, which we will also use as transport material and one emitter. The exciton preset "phosphorescent" will lead enable triplet decay on the microsecond scale, very fast inter system crossing and strong TPQ/TTA. The IP of 5.5 eV and EA of 2.3 eV are chosen so that the emitter is a charge trap. 0.09 eV is the energy disorder 0.2 eV the reorganization energy.

layers:
- thickness: 10
  morphology_input_mode: cubic
  molecule_species:
  - material: host
    concentration: 1.0

- thickness: 18
  morphology_input_mode: cubic
  molecule_species:
  - material: host
    concentration: 0.85
  - material: emitter
    concentration: 0.15

- thickness: 10
  morphology_input_mode: cubic
  molecule_species:
  - material: host
    concentration: 1.0

We define three layer constructed form a simple cubic lattice. The second layer containts two materials, the predefined "host" and "emitter" with an concentration of 85% and 15% respectively.

neighbours: 26
transfer_integral_source: Miller-Abrahams

Charge transport is possible to the 26 nearest neighbours. Transfer integrals are parametric.

electrodes:
- electrode_workfunction: -5.35
  coupling_model: parametrized
  electrode_wf_decay_length: 0.3
  electrode_coupling: 0.001

- electrode_workfunction: -2.45
  coupling_model: parametrized
  electrode_wf_decay_length: 0.3
  electrode_coupling: 0.001

Two electrodes are defined. The first electrode is attached before the layer that was defined first, the second one after the last layer.

pair_input:
- molecule 1: host
  molecule 2: host
  transfer_integral_parameters:
    hole_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.001
    electron_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.001
    Dexter_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.0001

- molecule 1: host
  molecule 2: emitter
  transfer_integral_parameters:
    hole_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.001
    electron_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.001
    Dexter_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.0001

- molecule 1: emitter
  molecule 2: emitter
  transfer_integral_parameters:
    hole_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.001
    electron_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.001
    Dexter_transfer_integrals:
      wf_decay_length: 0.1
      maximum_ti: 0.0001

We need to define the parameters for the transfer integrals between all kind of neighbouring pairs in the OLED.

experiments:
- simulations: 7
  measurement: DC
  Temperature: 300
  field_direction: [1, 0, 0]
  field_strength: 0.11 0.12 0.13
  initial_holes: 0

We do a simulation at constant Voltages. The field strength has to compensate for the in-built potential which is created due to the different workfunctions. 2.45V-5.35V = -2.9V. The length of the device is 38 nm, and electrodes are attached 0.8 nm from the layer. 2.9V/39.6nm = 0.073 V/nm. The effective fields chosen are thus 0.11-0.073 = 0.037 V/nm, etc. We run 7 simulations per field for better statistics.

iv_fluctuation: 0.001
max_iterations: 800000
new_wano: True

We run the simulation for 800000 Monte-Carlo steps

Simulation using lightforge

Run lightforge by typing:

# For V4
$OPENMPI_PATH/bin/mpirun -x OMP_NUM_THREADS --bind-to none -n 1 --mca btl self,vader,tcp python -m mpi4py $LFPATH/lightforge.py -s settings
# For V3 and below
$MPI_PATH/bin/mpirun -np 1 python -m mpi4py $LFPATH/lightforge.py -s settings

or to simulate the mobilities for all data points and the three specified fields simultaneously:

# For V4
$OPENMPI_PATH/bin/mpirun -x OMP_NUM_THREADS --bind-to none -n 22 --mca btl self,vader,tcp python -m mpi4py $LFPATH/lightforge.py -s settings
# For V3 and below
$MPI_PATH/bin/mpirun -np 22 python -m mpi4py $LFPATH/lightforge.py -s settings

Monitoring the calculation

In the case of a serial calculation the status of the calculation is printed out to the terminal. In case of a parallel calculation the status of the simulation of each data point is written to the files output_job_i.

Inspecting the results

In the folder materials we get a overview of the energy levels in the device in the files energy_crossection_f_x.png It might also be interesting to look at the bimolecular Förster hopping rates for triplets between emitters (molecule id 1) in the file T1T1_1_1.png.

The folder experiments contains several files with the results of the simulation. Field dependent internal quantum efficiencies are shown in IQE.png As we can see the IQE's in this OLED are pretty bad. Lets look at overall quenching and charge balance for the second field ( 0.12 eV/nm) in the file "quenching_density_average_1.png". Further details about excitonic processes in the 0.12eV/nm simulation can be seen in the file "exciton_decay_density_average_1_species_total.png" The files "exciton_decay_density_average_1_species_0.png" and "exciton_decay_density_average_1_species_1.png" further detail the same kind of information molecule type resolved. Looking at these we see that most excitonic processes happen on the emitter. Although included in the former plots, exciton formation (recombination) and radiative decay are also plotted seperately in the file photon_creation_density_average_1.png.

The results of the search are