University of Groningen
Growth and nanostructure of tellurides for optoelectronic, thermoelectric and phase-change
applications
Vermeulen, Paul Alexander
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Vermeulen, P. A. (2019). Growth and nanostructure of tellurides for optoelectronic, thermoelectric and phase-change applications.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
Chapter 10. English summary
Growth and nanostructure of tellurides
for optoelectronic, thermoelectric and
phase-change applications
The studies in this thesis may be divided into two parts: the structural and thermal analysis of bulk tellurides, and the growth and analysis of nanostructured 2D material films. They are connected through the shared knowledge and functional application of these compounds, as well as the analysis techniques used to study them. The main topics of this work include: the growth optimization of textured Bi2Te3, Sb2Te3, GeTe, Ge2Sb2Te5, and WTe2 films using
Pulsed Laser Deposition (PLD), the discovery of strain gradients within epitaxial quasi-2D-bonded group V-VI tellurides, as well as implementation of these telluride layers within a switchable optically functional multilayered film. Furthermore, we conclusively determined the arrangement of distortion domains within the GeTe-alloys and other distorted-cubic systems like LaAlO3. We also
determined the crystallization behaviour at high heating rates of SeTe(As) alloys, and elucidated the necessity of combining calorimetry with electron microscopy techniques. The thesis work as a whole represents a study into the relation between preparation history (either layer growth or thermal treatment of bulk), and resulting (nano)structure of several commonly studied telluride compounds. We offer an outlook on the implications for functional applications. Below we describe the performed studies in some more detail, starting from the bulk.
10.1
Bulk tellurides
Phase Change Materials (PCMs) for memory and storage applications are generally studied using the GeTe and GeSbTe systems. These materials are stable in both the amorphous (disordered) and crystalline (ordered) structure, and crystallize extremely quickly. We investigated a phase-change model system (SeTe alloys) using ultrafast Differential Scanning Calorimetry (DSC).Ch.8 We performed
heating and cooling with speeds up to 40 000 K/s, allowing us to construct an extended phase diagram, showing glass transition, crystallization, and melting temperatures for a range of Se1-xTex alloys. For increasing x, the crystallization
temperature sharply drops. The high cooling rates were sufficient to quench the alloy into an amorphous phase, allowing reversible switching. Furthermore, the fast heating during which the alloy crystallized, revealed the crystallization process is non-Arrhenian.
Chapter 10. English summary
The arsenic-alloyed SeTe systems are particularly of interest to the optics field concerned with the transfer of light through amorphous fibers: crystallization and phase-segregation should be prevented.Ch.9 By alloying the SeTe with 10 at.% As,
crystallization can be strongly retarded. We performed the same thermal analysis as for SeTe, but found that the thermal behaviour of the alloy was more complex than for SeTe. Microscopy revealed the alloys phase-separate into lamellae on crystallization, creating a Te-rich crystal, and As/Se-rich amorphous phase. Increasing the heating rate reduces the lamellar width, requiring the transfer of ultrafast DSC samples to Transmission Electron Microscopy (TEM) for analysis. A relatively simple method to transfer and prepare the micron-sized samples was developed to accomplish this.
The GeTe material is used both in the fields of PCMs and Thermoelectrics (TE). Its crystal structure is usually approximated as a distorted (cubic) rocksalt structure, where one of the four body diagonals is slightly elongated. This gives rise to a highly disordered microstructure, which we first noticed in the TE compound GeTeAgSb (TAGS-85), on which we have also co-authored two publications. Using microscopy on all length scales, from optical down to SEM-EBSD to TEM, we proved the microstructure consisted of a system of mixed [100]-[110] crystal boundaries, which incorporated domains of all four distorted body diagonals. It is assumed this high density of domain boundaries contributes to the low thermal conductivity and therefore the high TE performance of GeTe. Since the microstructure is observed in all rhombohedrally distorted materials like the investigated LaAlO3, we may extend this conclusion to many more materials.Ch.7
10.2
Thin films
Like the progress of materials science itself, we made the transition from studying the bulk materials described before, to the growth of artificial nanostructures. Several tellurides described in this work (Ge2Sb2Te5, Sb2Te3,
Bi2Te3, WTe2) are so-called 2D-materials. At the nanoscale they have the
appearance of stacked sheets, since subsequent unit cells are bound along the c-axis by weak van der Waals (vdW) forces, while in the ab-plane they are more strongly bound. Dissimilar 2D-materials may be stacked in an epitaxial fashion, without strain, due to this weak interlayer interaction. This paradigm of vdW epitaxy, introduced by Koma in 1991, dominates the way we think about heterostructures of these materials. Many research efforts are undertaken to manufacture stacks of 2D materials, to combine the functional properties of these thin films within one device.
A Pulsed Laser Deposition (PLD) system was used to optimize the growth of thin films. The system is relatively easy to use, cheap, and capable of monolayer accuracy and stoichiometric growth using the Reflective High-Energy Electron diffraction (RHEED) unit.Ch.1,2 The growth of textured telluride nanostructures was
10.2 Thin films
knowledge of these processes was present at the ZIAM. Therefore a detailed account of the optimization process is given in this thesis, meant to transfer knowledge and early trial lessons to a new generation of researchers.Ch.3 The PLD
technique involves ablation of the desired material from a bulk target using a laser pulse: this ablated plume of material is deposited on a substrate. The growth process is kept within a high vacuum or controlled gas atmosphere. Many degrees of freedom exist and are explored in this work, such as surface structure and chemistry of the substrate, the temperature, gas pressure, and laser energy. It was found that Bi2Te3 and Sb2Te3 can be epitaxially grown on amorphous smooth
substrates with the use of a seed layer, while for a crystalline vdW substrate like mica, epitaxy is achieved without seed. GeTe, and WTe2 required the use of a
Sb2Te3 or Bi2Te3 seed layer, else the material would grow respectively untextured or
amorphously. Furthermore, while for most materials a powder target sufficed, only a single-crystal target of WTe2 yielded crystalline films. While for the chalcogenides
several other techniques exist to produce films, we are to our knowledge, the first to optimize a low-temperature (210 °C) deposition technique of textured (single-crystal-like) WTe2.Ch.4
After optimization of the growth process of single-layer films, we set out to grow heterostructures. We grew heterostructures of Sb2Te3 and Bi2Te3, and using
RHEED to monitor the in-plane lattice parameter, we discovered the vdW epitaxy paradigm is not valid for these systems. The interface was clearly strained, and the layer growth only relaxed strain after a thickness of tens of nanometers. Similar effects were found when GeTe was incorporated, reaching strains within a vdW material of up to 5%, which is more than enough to tune the functional properties of these materials. We proposed a predictive model for the strain state within a thin film.Ch.5
After the growth optimization of many materials, it was realized that by combining several PCM-like materials, such as Sb2Te3, GeTe, and Ge2Sb2Te5, a
multi-level switchable film could be made. Several structures were created, where each layer was separated by an oxide barrier to prevent intermixing. Using ellipsometry experiments to obtain index of refraction and a model based on the Fresnel equations, the optical reflection profile could be predicted for each crystallizing PCM layer. An attempt to detect the optical anisotropy within single-layer vdWaals materials was not successful, and a hyperbolic dispersion within these films could not be confirmed or ruled out. Observed reflection profiles of multilayered structures matched quite well to their modeled values. Moreover, using dynamic ellipsometry, the phase-changes could be explicitly visualized and we could show the crystallizations were robustly separate events and therefore material states.Ch.6
