Plastic spintronics : spin transport and intrinsic magnetoresistance in organic semiconductors

W. Wagemans

Onderzoeksoutput: ScriptieDissertatie 1 (Onderzoek TU/e / Promotie TU/e)

330 Downloads (Pure)

Samenvatting

A specific class of plastics and other organic materials is (semi)conducting and can be used for electronic applications. In this thesis, we investigated their use in certain applications using magnetic fields.These applications belong to the field of spin electronics, or ‘spintronics’, where next to the charge of the electrons, also their spin is utilized.This spin can be considered as a small magnetic moment, pointing either up or down. Organic semiconductors have several advantages that make them interesting for applications in spintronics.They are relatively cheap, are easy to process, and can be chemically adapted. Two different, but related, topics that combine organic materials with spintronics have been studied both experimentally and theoretically. The first topic is the recently discovered change in the current through an organic layer when a small magnetic field is applied. This effect is called organic magnetoresistance (OMAR). Large changes in the current (10–20%) have been found at relatively small magnetic fields (~10mT) and at room temperature.These properties make the effect interesting for applications like, for instance, cheap magnetic field sensors, opening the possibility to add magnetic functionality to existing organic electronic applications. Moreover, if its mechanisms are understood, the effect could be used to investigate processes in conventional organic electronic devices by studying their magnetic response. OMAR is observed in a wide range of organic materials, from small molecules to polymers. It is believed that OMAR originates from the interactions of a pair of charge carriers (for instance, electron–electron, hole–hole, or electron–hole).More specifically, it is believed that the possible interactions depend on the relative orientation of the spins of these two charges. Without an external magnetic field, small intrinsic magnetic fields in the organic layer (resulting from hyperfine coupling to nuclear spins) randomize the orientations of the two spins. This allows a change froma spin configuration that is less favorable for the current into amore favorable configuration. However, applying amagnetic field larger than these hyperfine fields results in a strong reduction of this spin randomization or spin mixing, causing a pair to remain locked in a less favorable spin configuration. Although there is agreement on the crucial role of hyperfine fields, the exact mechanisms behind OMAR are still heavily debated. Several models were proposed in literature that explain OMAR in terms of different charge pairs. In this thesis, we investigated a model based on pairs of equal carriers, called the bipolaron model. We used an elementary model of two neighboring sites, where, depending on the spins, one carrier might be preventing another one to pass. With this theoretical model we were able to successfully reproduce several characteristics of OMAR. Both a decrease and an increase in current, as found in experiments, could be obtained and also the universal shapes of the experimental OMAR curves could be reproduced. Additionally, we performed new types of experiments to gain better understanding of OMAR. We showed that when an oscillating magnetic field is applied,OMAR is reduced beyond a certain frequency threshold.This occurs when the slowest charges can no longer follow the oscillations, as we showed by measuring the frequency dependence of the capacitance. These findings are in agreement with recent interpretations in which these slowest carriers are expected to induce the largest OMAR effect. In literature, it was claimed that OMAR is independent of the orientation of the magnetic field. However, via sensitive measurements we demonstrated a small but systematic dependence on the angle between the magnetic field and the sample. We showed theoretically that this angle dependence can be explained in the different models by including an interaction between the spins. This interaction has to be direction dependent in order to explain the angle dependence.We identified dipole–dipole coupling or an anisotropy in the hyperfine fields as themost likely candidates. Furthermore,we outlined a first exploration of an alternative approach to describe OMAR curves. We introduced a function that allows us to extract information both about the hyperfine fields and about an additional broadening of the curves. Thereby, this approach could allow for amore quantitative analysis of changes in the OMAR curves resulting from changes in the operating conditions or the material properties. In the second topic, the spin of electrons is used in a different way. In many spintronics applications a difference between the number of spin-up and spin-down electrons, called spin polarization, is used to transport information through a device. For the functioning, it is essential that this polarization persists while transporting the charges.The main mechanism for loss of polarization inmost inorganic semiconductors, which is related to spin–orbit coupling, is negligible in organic materials. The absence of this loss mechanismmakes organic materials ideal candidates for these types of spintronics applications. However, there might still be other mechanisms that cause a smaller but non-zero loss of spin polarization. We conjecture that the hyperfine fields are the main source of polarization loss in organic materials, which results from mixing between the spin-up and spin-down electrons by precession of spins about these random fields. We theoretically investigated this effect of the hyperfine fields on the spin polarization. We explicitly included the hopping transport characteristic for organic semiconductors. Due to spatial and energetic disorder, the charges hop from one localized site to another. The longer the time they spend on a site, the larger the loss of spin polarization. We showed that an external magnetic field larger than the typical hyperfine-field strength reduces the loss of spin polarization. Hence, such an external field causes the polarization to persist over a larger distance, leading to an increase of the spin-diffusion length.We thus found a magnetic-field dependent spin-diffusion length. In addition, we found the spin-diffusion length to depend only weakly on temperature. A spintronics device that makes use of spin-polarized transport is the spin valve. Using the magnetic-field dependent spin-diffusion length obtained from our theory, we could very accurately fit experimental data on the magnetoresistance of organic spin valves reported in literature. However, there is still a hot debate on the interpretation of these and similar experiments.The question has been raised whether spins are indeed transported through the whole layer, or only through thin regions. A discriminating experiment would be the manipulation of spins during their transport through the organic layer.This can be done by applying a magnetic field perpendicular to the direction of the spin polarization. Using our spin-transport model, we made predictions about the results that can be expected from such an experiment. We showed that, in the case of transport through the organic layer, an effect of a perpendicular field should be observable, but that the strong oscillations in the signal that are typically seen in inorganic semiconductors will be absent. Finally, as an extension of the work presented in this thesis, we made predictions about possible future experiments in which spin polarization is combined with OMAR. Because for a spin-polarized current the majority of the spins point in the same direction, most charge pairs will have parallel spins.Therefore, within the bipolaron model, we expect an increase in the magnitude of OMAR when the injected current is spin polarized. Moreover, we showed that the shapes of the OMAR curves will also be changed.These experiments would provide a means to both prove spin-polarized transport and to validate the bipolaron model. In conclusion, both theoretical and experimental results on OMAR and spin polarized transport have been presented in this thesis. Contributions have been made to a new model for OMAR and new type of experiments have been performed that have added further insights to the puzzle of OMAR. The limiting role of the hyperfine fields on spin-polarized transport has been investigated theoretically, providing an explanation for the experimentally observed magnetoresistance curves of organic spin valves and providing suggestions for future experiments. Although the present work has led to better understanding of OMAR and spin-polarized transport, the field of organic spintronics still poses many theoretical and experimental challenges that should be resolved before a widespread emergence of organicspintronics applications will occur.
Originele taal-2Engels
KwalificatieDoctor in de Filosofie
Toekennende instantie
  • Faculteit Technische Natuurkunde
Begeleider(s)/adviseur
  • Koopmans, Bert, Promotor
  • Swagten, Henk J.M., Promotor
  • Bobbert, Peter, Co-Promotor
Datum van toekenning14 jun 2010
Plaats van publicatieEindhoven
Uitgever
Gedrukte ISBN's978-90-386-2255-2
DOI's
StatusGepubliceerd - 2010

    Vingerafdruk

Citeer dit