While organic light-emitting diodes (OLEDs) have entered the broad commercial market of displays of all the sizes and are trying to conquer the lighting industry [1], the understanding of their principle of operation by an average person is often very limited. Typically, to show how an OLED works the authors in mass media use as an example inorganic LEDs when explaining the processes involved in light generation. Though on a basic level some simplifications may be necessary, at the same time they introduce false description of the phenomena that take place in OLEDs. The origin of these misconceptions seems to be consideration that all semiconductors are similar, independently if they are organic or inorganic. This is obviously not true, since the both types of materials have some physical properties that are unique for them and they determine the processes present in a device.

In this article we try to give the reader an insight into the OLED’s principle of operation. On one hand we want to keep the discussion as simple as possible, but at the same time we avoid simplifications that are mostly found in the literature for nonscientific society. We start with a chapter that describes the differences between inorganic and organic semiconductors and then OLEDs with architectures from the simplest to more complex are discussed.

Physical background

In order to understand the differences between inorganic and organic semiconductors we begin with a short reminder of the energy band theory. According to the Pauli exclusion principle [2], in a system two particles with all the same quantum numbers can’t exist at the same time. In a case of interaction of identical neighbor atoms, it will cause the shifting and splitting of their energy levels into sublevels in order to introduce the difference in the quantum numbers that describe their states. An inorganic semiconductor consists of strongly bound atoms which interact with each other at long distances. As a consequence, one atom “feels” the presence of many neighbors and the energy levels must split into many sublevels. The distance on the energy scale between the sublevels becomes so small that an electron can move easily between them. In other words, the energy band is created [3]. There are two energy bands in an inorganic semiconductor, the valence band where electrons reside if a material was not excited, and the conduction band which are the energy levels accessible for electrons from the valence band if a semiconductor is excited by some stimulus. Between these bands the energy gap exists, the energy levels are not present there and therefore electrons can’t possess such energies. In semiconductors, the distance on the energy scale between the valence band and the conduction band corresponds roughly to the visible spectral range (380 nm (3.26 eV) – 780 nm (1.59 eV)). It is obvious that in order to excite a semiconductor the energy equal or higher than the one of the energy gap must be provided. On the other hand, the transition of the electron from the conduction band to the valence band will cause emission of a photon with the wavelength corresponding to the energy gap of the semiconductor.

Now let us take a closer look at organic semiconductors. They consist of molecules or polymers which possess so called π-conjugation [4,5], which means roughly that an electron can travel around molecule or polymer within an orbital. Orbitals are created due to interaction of atoms that constitute a molecule or polymer. Similarly like in inorganic semiconductors, the energy levels of the atoms split into sublevels, however the distance between them on the energy scale is usually not so close as in the case of inorganic semiconductor. As a result, we will have the situation where in a non-excited molecule or polymer electrons reside on some orbitals, the highest occupied molecular orbital (HOMO) can be seen here as a valence band from an inorganic semiconductor. There will be also the lowest orbital that can accept an electron if a molecule or polymer is excited, the lowest unoccupied molecular orbital (LUMO), which corresponds to the conduction band in an inorganic semiconductor. Molecules or polymer chains which create an organic semiconductor interact with each other, but contrary to strong interactions in an inorganic semiconductor, here weak van der Waals forces are involved. As a consequence, a “connection” between HOMO and LUMO levels of the adjacent molecules or polymer chains cannot be created (the exception are some organic crystals, but they are not employed in OLEDs and will be not discussed here). From this fact important result is derived: in amorphous organic semiconductors transport of charge takes a form of hops from one molecule to another while in inorganic semiconductors charge can travel freely within the energy band. Another consequence of weak interactions in organic semiconductors is that the properties of the bulk material are similar to the properties of molecules or polymers it consists of.

Finally, we must introduce a bound state of an electron and a hole – an exciton [4-6]. An exciton is created due to the Coulomb attractive force between a negative electron and a positive hole. It can move within a bulk of a semiconductor and either dissociates on free charge carriers or recombines and in consequence light is created. The wavelength of the emitted photon is correlated with the energy gap (the difference between the energy of HOMO and LUMO in an organic semiconductor). In inorganic semiconductors excitons exist only at low temperatures while in organic semiconductors they are commonly observed at room temperature.

Single layer OLED

All OLEDs consist of a substrate, an anode and a cathode (at least one of them is transparent) and organic layer(s) sandwiched between them (Figure 1). The total thickness of an OLED without substrate is in the order of hundred(s) of nanometers. A whole device must be encapsulated for protection from oxygen and water since organic semiconductors and electrodes are prone to degradation caused by those factors. The simplest device has only one organic layer between the electrodes and the compound plays a double role as a charge transporter and a light-emitting material.

Figure 1. 

In figure 1 the energy levels of all materials employed in a single layer OLED are presented. Typically, a transparent anode is made of metal oxides (e. g. indium tin oxide (ITO)) because they possess high work functions allowing good hole injection into the HOMO level of an organic semiconductor. On the top of an organic layer is a cathode which consists of a metal with small work function suitable for electron injection into the LUMO level. Usually, a cathode is covered with aluminum layer in order to protect the sensitive metal underneath. When an external electric field is applied to the device (positive bias to the anode, negative bias to the cathode), charge carriers are pushed from the electrodes into the organic layer and move towards the center of the diode. As was mention above, the movement takes a form of hops between molecules or polymer chains. If a hole (red ball in the figure) and an electron (blue ball) are close enough to each other, they can create an exciton (blue ellipse) which may recombine radiatively. Typically, the direction of the generated photon in amorphous organic bulk is random, it may be emitted towards transparent anode and leave the device, another option is emission towards the cathode; in this case it may be reflected back and goes outside the diode. However, the majority of light (75 % – 80 %) never leaves the standard organic diode due to the total internal reflections and absorption within the layers constituting the device [7].

A single layer OLED is easy to build but possesses many drawbacks. Typically, the mobilities of holes and electrons are different in a given material which will introduce imbalance in injected charge carriers. This will be the cause of the leakage current of the majority carriers (they will pass on the other side of the diode) and in consequence decrease the efficiency of the device. Finally, usually an organic semiconductor cannot be a good emitter and transporter at the same time. This means that in a case of poor transporting abilities injected charge will move slowly from the electrode to the center of the emitting layer and therefore may hamper charge injection from that electrode. On the other hand, when the transporting properties are good charge carriers will move fast through the organic layer. As a consequence, a chance that an exciton is created decreases and therefore efficiency of a diode may decrease as well [8].

Doped single layer

An improvement of the idea from the previous chapter is a single layer device with a dopant shown in figure 2.  An introduction of a guest into a host resolves the problems with transport and recombination because proper tasks are assigned to the specialized in their purposes materials.

Figure 2.

A host is able to convoy efficiently the charge injected from the electrodes into the center of the emitting layer and prevents in this way charge accumulation close to the electrodes. On the other hand, an emitting dopant can have the chemical structure optimized for light emission. 

The energy levels scheme of a doped diode shown in figure 2 should not be mistaken with a stack of host and guest layers. It is the simplified two-dimensional representation of the system’s energy landscape which in reality is three-dimensional, where the guest is surrounded by the host. It is important to remember that electrons always search the deepest LUMO level while holes the least deep HOMO level.

Since a host always has the band gap wider than the guest, a hole, an electron or both charge carriers will be trapped on the emitting center. A trapped charge will attract the counter charge and therefore the probability of exciton formation increases and consequently also the radiative recombination. In a case that an exciton recombines on the host, the energy transfer process will occur and the excitation will be relocated on the guest [9,10]. Finally, placing a small amount of a dopant inside a matrix (typically 1 – 10 %) prevents from the concentration quenching, which is the cause of light emission reduction. It is worth to notice that there is a particular class of emitters that reveals the opposite behavior, i. e. an improvement of light intensity with increase of concentration, the aggregation induced emitters (AIE) [11].

Transporting layers

A further improvement is achieved if charge transporting layers are applied (Figure 3). Between the anode and the emitting layer, a hole transporting material is introduced while between the cathode and the emitting layer, an electron transporting compound is employed. In this way the injected charge is transported very efficiently into the center of the diode where the emitting layer is localized. This will facilitate charge injection from the electrodes and places the recombination zone far from the electrodes. The latter is an important factor since the exciton recombination close to the electrode may excite some states within it instead of the light generation. The recombination zone close to the center of the device improves also the microcavity resonance condition which may increase recombination rate of the emitter due to the Purcell effect [12].

Figure 3.

Finally, the application of transporting materials allows to balance the number of electrons and holes within the diode since we can choose compounds with suitable charge mobilities.

Blocking layers

A charge carrier injected into an OLED in principle is able to pass through the whole diode without recombination with the counter charge. Such a leakage current will be the cause of the device efficiency decrease and therefore should be avoided. To this end charge blocking layers are introduced at the interfaces between the emitting layer and the charge transporting layers (figure 4).

Figure 4.

A compound used in the hole blocking layer should possess the LUMO level similar to the LUMO energies of an electron transporting material and a host in the emitting layer, while the HOMO level must be deeper than the neighbor materials. In this way holes that arrive from the anode will be stopped at the interface and cannot enter the electron transporting layer. Similarly, the LUMO level of the electron blocking layer must be higher than the LUMO levels of the hole transporting material and the matrix. Charge blocking layers keep electrons and holes within the emitting layer for a long time therefore the exciton creation probability increases and as a consequence the efficiency of a device improves. Typically, blocking layers are thin (~ 10 nm) just enough to stop charge within the emitting layer but without increase much the thickness of the device. In some cases, a transporting layer can be at the same time the blocking layer if an alignment of the energy levels is possible.

Epilog

In this short article we tried to explain in a simple way the basics of an OLED operation. We hope that it will encourage the reader for further research in the vast field of organic semiconductors and devices based on these materials.

Editorial Board’s Note: OLED stanowią znaczący krok naprzód w dziedzinie wyświetlaczy i oświetlenia, oferując większą elastyczność, wyższą jakość obrazu i efektywność energetyczną w porównaniu do tradycyjnych technologii. W Polsce, rozwój badań nad OLED oraz ich aplikacji w różnych dziedzinach przemysłu, od elektroniki po oświetlenie, może stanowić szansę na wzmacnianie polskiego udziału w globalnym rynku technologicznym.

Bibliography:

1. Shi-Jie Zou, Yang Shen, Feng-Ming Xie, Jing-De Chen, Yan-Qing Li, Jian-Xin Tang, Mater. Chem. Front., 4 (2020) 788, https://pubs.rsc.org/en/content/articlepdf/2020/qm/c9qm00716d
2. Wolfgang Pauli, Nobel lecture (December 13, 1946), https://www.nobelprize.org/uploads/2018/06/pauli-lecture.pdf
3. https://en.wikipedia.org/wiki/Electronic_band_structure
4. M. Pope, C. E. Swenberg, “Electronic Processes in Organic Crystals and Polymers”, Oxford University Press, 1999
5. A. Köhler, H. Bässler, “Electronic Processes in Organic Semiconductors. An Introduction”, Wiley-VCH, 2015
6. J. Frenkel, Phys. Rev. 54 (1938) 647, https://doi.org/10.1103/PhysRev.54.647
7. W. Brütting, J. Frischeisen, T. Schmidt, B. Scholz, C. Mayr, Phys. Status Solidi A 210 (2013) 44, https://onlinelibrary.wiley.com/doi/epdf/10.1002/pssa.201228320
8. J. Kalinowski, J. Phys. D 32 (1999) R179, https://iopscience.iop.org/article/10.1088/0022-3727/32/24/201/meta
9. T. Förster, Discuss. Faraday Soc. 27 (1959) 7, https://pubs.rsc.org/en/content/articlelanding/1959/df/df9592700007/unauth
10. S. E. Braslavsky, E. Fron, H. B. Rodrìguez, E. San Romàn, G. D. Scholes, G. Schweitzer, B. Valeur, J. Wirz, Photochem. Photobiol. Sci. 7 (2008) 1444, https://link.springer.com/content/pdf/10.1039/b810620g.pdf?pdf=button
11. M. Yu, R. Huang, J. Guo, Z. Zhao, B. Z. Tang, PhotoniX 1 (2020) 11, https://photonix.springeropen.com/articles/10.1186/s43074-020-00012-y
12. https://en.wikipedia.org/wiki/Purcell_effect