Phase-Change and Ovonic Materials (Fourth Edition)

Once again, it is our great pleasure to continue the E\PCOS tradition by presenting this fourth edition of the special issue on Phase-Change and Ovonic Materials that is published each year as part of the European Symposium on Phase-Change and Ovonic Sciences (E\PCOS). We have to admit that the 2022 edition of E\PCOS had a special spirit as it marked the long-awaited return to a face-to-face on-site symposium after a two-year hiatus due to the Covid crisis. Last September, the 2022 edition of E\PCOS was the most successful in terms of attendance in the history of E\PCOS. This is no coincidence and once again, in this editorial, Harish Bhaskaran, Luci Bywater and the Oxford team are sincerely thanked on behalf of the entire E\PCOS community for making this success possible even if unfortunately, some of the E\PCOS major actors could not join us at the Wolfson College in Oxford this year. As in the three previous editions, this special issue again aims to summarize recent and innovative scientific and technological achievements in the field of phase-change materials, as well as their possible new fields of application. In addition to recent advances in this field, the objective is also to present emerging interests in neuromorphic computing, phase-change and nonlinear photonics or plasmonics. This special issue thus provides an overview of the state of the art, both experimental and theoretical, for experienced and young researchers interested in these topics. As usual, let us first recall, for the younger and newer members of our ever-evolving community, that E\PCOS was born in Switzerland in 2001, with the aim to provide a platform to discuss and promote the fundamental science of phase-change materials (PCM). This goal also included their applications in rewritable optical discs (e.g., first with CDs and later with the successfully developed DVD and Blu-ray Disc formats) and thus initially PCOS referred to phase-change optical storage (which was diversified in 2005 to phase-change and ovonic science). In fact, E\PCOS was born from the first PCOS symposium held in Japan in 1990, thanks to Professor Masahiro Okuda, who was the advisor of E\PCOS during its early years. In recent years, the field has diversified considerably. While the scientific and technological fingerprints of the field's founding father, the late Stanford Ovshinsky, are still very recognizable, the number of topics covered has continued to grow significantly with applications including non-volatile electronic memories, optoelectronics, photonics, and neuromorphic computing. The 2022 edition of E\PCOS has confirmed that E\PCOS is the premier international conference on this exciting and evergreen topic. This 2022 edition, which follows the 2021 virtual edition, was somewhat of a challenge for the E\PCOS community. However, its unprecedented success confirmed the close ties between key players in the field, both academic and industrial. By again covering a rich variety of topics beyond phase-change memories, this fourth special issue will again mark the history of E\PCOS. The paper from Park et al. on the use of Sb2Te3/TiTe2 heterostructure to replace the conventional and canonical GST (Ge2Sb2Te5) alloy for storage class memory and neuromorphic computing hardware is a first illustration [pssr.202200451]. Indeed, for these recently introduced novel phase-change memory (PCM) applications, faster SET speed and lower RESET energy than those obtained with the usual GST225 alloy are required. In this study, PCM devices based on multilayers made of amorphous Sb2Te3 and TiTe2 nanolayers deposited by sputtering exhibit fast SET speed (≈30 ns), RESET energy reduction of more than 80% compared to the GST-based reference PCM with also lower resistance drift in the high resistance state. These very promising results will deserve future work in the community, for example to evaluate the endurance of this novel type of heterostructure during programming cycles. A similar goal has also motivated the work presented by Kashem et al. [pssr.202200419] in which they proposed a finite element simulation framework combining amorphization–crystallization dynamics and electro-thermal effects to better describe RESET–SET–READ operations of PCM nanoscale devices. They concluded that their GST alloy-based model could account for the impacts of dynamic changes in crystallinity during device operation and that their results were consistent with experimental observations, providing a better understanding of device dynamics. This model would allow any device geometry to be studied to explore the effect of programming pulse and material engineering, as well as device architecture on device performance. For example, the simulation results predict the impact of thermoelectric effects on RESET current requirements and the significant role of heater height on heat loss and thus RESET current. One of the other challenges of PCM technology for storage-class memory applications is related to improving storage density by using of multilevel cells (MLCs), as shown by Zhao et al. [pssr.202200463]. Although the design of high aspect ratio memory cells in confined-structure PCMs enables controllable MLC operation, it negatively affects the power consumption, which becomes too large to successfully operate the PCM. Therefore, they investigated the trade-off between different sizes of the GST nanowire aspect ratio using COMSOL simulation and experiments. Their conclusion is that the best trade-off between not too much RESET power consumption and MLC enhancement can be obtained for a GST nanowire length below 1.5 μm and an aspect ratio of about 10:1. Increased storage density could also be enabled by vertical 3D integration of the PCM in 3D cross-point memory arrays. This could be achieved by adding to each PCM element at each cross point of the memory cell array an Ovonic Threshold Switching (OTS) selector. The latter is made of amorphous chalcogenide material with OTS behavior under high electrical-field application. OTS device is thus a nonlinear two-terminal selector device allowing to program and read the selected PCM cell in the cross-point array but at the same time limiting or suppressing the unavoidable leakage current paths through the adjacent unselected cells. Performance of OTS selector devices are characterized by the threshold voltage (Vth) required to switch the amorphous into its metastable highly conductive state and the subthreshold current below Vth that directly defines the size of the achievable memory array. In this context, Ravsher et al. [pssr.202200417] investigate one of the challenges for OTS selectors that is related to Vth instability. To this end, they have studied, through electrical characterization of OTS devices, the polarity-induced Vth shift in quaternary SiGeAs(Te/Se) materials as well as in binary GexSe1−x compounds to investigate in the latter case the impact of polarity, bipolar cycling, thickness, interface oxidation, and stoichiometry on the Vth shift and subthreshold conduction. They concluded that the magnitude of the Vth shift depends on the material composition and increases up to nearly one order of magnitude when going from SiGeAsTe, SiGeAsSe to GeSe. They also showed that the SiGeAs(Te/Se) and GexSe1−x families of materials exhibit a reversed direction of the shift with a different dependence on operating current. From a more theoretical and fundamental point-of-view Perego et al. [pssr.202200433] answer a very relevant question, the structure and crystallization kinetics of GeTe films. To this end, they perform molecular dynamics simulations with up to 8000 atoms. They compare the atomic arrangement of melt-quenched samples with as-deposited samples and conclude that both are very similar, at least if the atoms are deposited with sufficiently high particle energies. This is very important since it is very difficult to create macroscopic samples of melt-quenched phase-change materials and often as-deposited amorphous samples are investigated to unravel the atomic arrangement of the glassy phase of PCMs. Finally, they conclude that the as-deposited films contain a larger number of Ge-Ge bonds and Ge-atoms in tetrahedral geometries. The impact and nature of defects, studied using ab initio molecular dynamics, is discussed by Konstantinou and Elliott [pssr.202200496]. They explore four different GeSbTe compounds, i.e. GeTe, Sb2Te3, GeTe4 and Ge2Sb2Te5 and find charge-trapping centers. The atomic arrangement of these trapping centers ranges from over- and under-coordinated atoms, tetrahedral and distorted octahedral sites, four-fold rings and homopolar bonds. These charge-trapping centers are an intrinsic property, which localizes electrons or holes and should thus be relevant for device operation. van-der-Waals (vdW) gaps play an important role in various chalcogenide materials and thus PCMs. Strepanov et al. in [pssr.202200430] focus on GaSe and employed density functional theory (DFT) to analyze the influence of uniaxial compressive stress perpendicular to the vdW gaps. It is shown that high stress levels (up to 30 GPa) do not close the vdW gaps, i.e. going from 2D to 3D, but on the contrary generates a quasi-1D structure with GaSe chains oriented along the applied pressure. This structural change of course also strongly alters the electronic structure. The authors think that this behavior is more generic and currently test if similar effects occur for Sb2Te3. Ge-rich GST alloys are attractive for embedded phase-change memory since they fulfill the tough temperature specifications required by automotive applications. However, these alloys are challenging due to their intrinsic tendency to phase separate. In the paper by Petroni et al. [pssr.202200458], an advanced statistics based methodology is presented that quantifies the segregation and clustering of multiple chemical species during processing, also highlighting specific correlations that occur between the elements as the material evolves. The methodology uses here elemental maps obtained by STEM from focused ion beam prepared cross-sectional samples from actual integrated PCM cells. The present work not only demonstrates the strong generic capabilities of this methodology, but it also enabled optimization of the BEOL process causing improved Ge-rich GST with less segregation and undesirable clustering. The impact of Ge on crystallization is also addressed in the study of multistep crystallization of Ge-rich GeSbTe (GST) compounds. Rahier et al. investigate the crystallization process employing in-situ synchrotron radiation and scanning transmission electron microscopy (STEM) [pssr.202200450]. Ge-rich GST compounds are very attractive for the next generation of embedded digital memories since they offer increased thermal stability of their amorphous phase. Upon crystallization, incongruent phase formation is observed. Initially, the amorphous phase already starts to undergo phase separation leading to regions of different Ge content. Subsequently, GeTe grains start to crystallize, which trigger the crystallization of a cubic Ge phase. Cubic Ge2Sb2Te5 is only formed at higher temperatures upon incorporation of Sb into GeTe grains. Finally, the work of Humphreys et al. [pssr.202200474] highlights the strong added value PCMs have for the field of optical metamaterials. Here, GST is added to periodic arrays of sub-wavelength-scale holes in plasmonic metal films that generate resonant transmission peaks via the extraordinary optical transmission (EOT) effect. Adding GST, with its switching between amorphous and crystalline phases, allows active control, in particular shifting of the spectral position of such transmission peaks and is thus highly relevant for various applications. The whole cycle of design, fabrication, and characterization of active EOT devices targeted at various important regions of the optical spectrum from terahertz to infrared is presented. In conclusion, please enjoy this fourth and new issue of 2022 presenting some of the latest research on phase-change and ovonic science. As in previous years, we also hope that this new issue will help strengthen our community in preparation for the next E\PCOS meeting to be held in Rome on September 18-20, 2023. In many countries, there is the saying: “All roads lead to Rome”, so we look forward to welcoming you again for a promising conference to celebrate together the 22 years of E\PCOS following this last edition in 2022, which showed once again that phase-change materials and their research are still as dynamic and thriving as ever. Pierre Noé, Bart J. Kooi and Matthias Wuttig Guest editors B.J.K. and P.N. acknowledge funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No.824957 ('BeforeHand': Boosting Performance of Phase Change Devices by Hetero- and Nano-structure Material Design). P.N. also thanks the support of the ANR through projects MAPS ANR–20–CE05 and OCTANE ANR–20–CE24–0019, the Auvergne-Rhône-Alpes region through “Pack Ambition Recherche 2019”-NanoCHARME project as well as the European Union's Horizon Europe research and innovation programme under grant agreement No.101070238 (‘NEUROPULS’: Neuromorphic Energy-Efficient Secure Accelerators Based On Phase-Change Materials Augmented Silicon Photonics). P.N. also thank the French research group “GdR CHALCO”, recently created (https://gdrchalco.cnrs.fr/). M.W. acknowledges funding from the Deutsche Forschungs-Gemeinschaft (DFG) via the collaborative research center Nanoswitches (SFB 917). At last, we highly appreciate the time and efforts of all contributors to this Focus Issue. Pierre Noé began his career at the CEA's Basic Research Directorate in 2000 as a materials engineer working on Si-based nanostructured materials for microelectronics, photonics and spintronics. In 2011, he became a permanent researcher at CEA-Leti and obtained a PhD in physics in 2013 at the University of Grenoble-Alpes. Since then, he has created his own research group on advanced chalcogenide materials within Leti's Silicon Technologies division. His research focuses on the development and science of innovative chalcogenide materials at the frontier between fundamental knowledge and technological applications (memories, photonics, thermoelectricity, etc.). He is currently involved in two French Research Agency projects, two European HORIZON projects; he is one of the six permanent members of the committee of the Groupement de Recherche (GdR) CHALCO and guest editor since 2019 of the Special Focus Issue of E\PCOS which will be published annually in the journal pssRRL. In 2022, he received his "Habilitation à Diriger des Recherches" from the University of Grenoble-Alpes. Bart J. Kooi Obtained his PhD degree in materials science in 1995 from Delft University of Technology, Netherlands. Worked since then at the University of Groningen (Netherlands) as assistant, associate and full professor, starting in 2009 his own research group Nanostructured Materials and Interfaces within the Zernike Institute for Advanced Materials. His main research interests are nanostructure-property relations, advanced transmission electron microscopy, interfaces, phase transformations and tellurium and antimony based materials for thermoelectric and phase-change memory applications. Matthias Wuttig received his Ph.D. in physics in 1988 from RWTH Aachen/Forschungszentrum Jülich. He was a visiting professor at several institutions including Lawrence Berkeley Laboratory, CINaM (Marseille), Stanford University, Hangzhou University, IBM Almaden, Bell Labs, DSI in Singapore and the Chinese Academy of Sciences in Shanghai. In 1997, he was appointed Full Professor at RWTH Aachen. Since 2011, he has been heading a collaborative research center on resistively switching chalcogenides (SFB 917), funded by the German Science Foundation (DFG). [pssr.202200451] [pssr.202200419] [pssr.202200463] [pssr.202200417] [pssr.202200433] [pssr.202200496] [pssr.202200430] [pssr.202200458] [pssr.202200450] [pssr.202200474]