Solution-Based Engineering of Nanodimensional Phase-Change Materials and Memory Devices

Informations projet

NanoMMs

N° de convention de subvention: 852751

DOI

10.3030/852751

Projet clôturé

Date de signature de la CE 3 Septembre 2019

Date de début 1 Janvier 2020

Date de fin 31 Decembre 2024

Financé au titre de

• EXCELLENT SCIENCE - European Research Council (ERC)

Coût total

€ 1 604 600,00

Contribution de l’UE

€ 1 604 600,00

Coordonné par

EIDGENOESSISCHE TECHNISCHE HOCHSCHULE ZUERICH
Switzerland

Periodic Reporting for period 4 - NanoMMs (Solution-Based Engineering of Nanodimensional Phase-Change Materials and Memory Devices)

Période du rapport: 2024-07-01 au 2024-12-31

Résumé du contexte et des objectifs généraux du projet

Scaling limits of Si-based transistor memory set a stage for new technologies and computer architectures. Phase-Change Memory (PCM) is undoubtedly the most promising contender for future post-Si memory. The work principle of PCM devices is simple: memory material is converted to its crystalline and amorphous phases by heating a memory cell interface above crystallization and melting phase transitions, respectively. The resistivity of the two phases differs by several orders of magnitude, which allows a clear definition of binary “0” (highly resistive) and “1” (highly conductive) logic states. Moreover, the crystalline and amorphous structures are stable at room temperature for a long time, hence defining the PCM chips as non-volatile type of memory.

One important advantage of PCM technology is the possibility to select from a dozen of materials. Characteristics of PCM devices can thus be adjusted by the choice of PCM material, filling the entire gap between Si-based storage-class NAND memory and Dynamic RAM. While PCM devices already outperforms NAND, they hold potential to reach characteristics on DRAM in the future. Among other unique features of PCM technology is the possibility to fine-tune the partial crystallinity of memory material, giving access to many intermediate resistivity states, reminiscent to analog-like brain-mimicking computing. Finally, PCM memory displays data robustness at elevated temperatures and resilience to radiation, being suitable for extreme condition in automotive industry or even in open-space applications.

With all its benefits, two disadvantages of PCM devices remain: the price of PCM devices is notably higher than for Si-based memory technology and the energy consumption of PCM is larger too. As for the price, this is because state-of-the-art fabrication of PCM arrays relies exclusively on expensive and material-inefficient techniques such as sputtering, etching, and lithography. Large power consumption of PCM devices, however, is a permanent pain point due to necessity of local melting upon memory switching. This ERC Starting Grant aims to tackle both problems at once by implementing liquid-phase fabrication methods for PCM technology. We rely on colloidal nanoparticles and molecular inks to deposit thin films and patterned layers of PCM devices via material-efficient, inexpensive, and high-throughput fabrication from the solution. Furthermore, we aim for control over size and thickness of PCM nanomaterials and thin films to study size dependences of phase-change properties and to help design the scaling rules for future PCM memory bits with ultrasmall sub-10 nm physical dimensions.

Travail effectué depuis le début du projet jusqu’à la fin de la période considérée dans le rapport et principaux résultats atteints jusqu’à présent

The objectives of this project have been charted to three parallel work streams, covering chemistry, physics and engineering aspects of liquid-borne PCM materials and devices, respectively.

For the chemistry-related tasks, we have synthesized a large range of ternary tellurides and antimony-rich alloyed nanoparticles. We have achieved all planned compositions of nanoparticles, spanning from classical Ge-Sb-Te (GST) material to the most recent ternary telluride systems (i.e. Sn-Ge-Te or Cu-Ge-Te) and antimony-based bimetallic nanocrystals (i.e. Bi-Sb). For many of these materials, we demonstrate a total control over size, composition, and phase of nanoparticles. In addition, we develop a solution-based approach for molecular ternary tellurides, reaching an accurate composition control and high-quality homogeneous thin films from these molecular inks. Taken together, we have developed means to prepare literally any phase-change material in the form of liquid formulations.

For the physics-related tasks, we study a structure, dynamics, and nanoscale effects of amorphous GeTe nanoparticles. We focus on crystallization phase transition in particularly, combining experimental methods with theory calculations. For the experiments, we employ high-temperature conditions and controlled temperature profile for in-situ X-ray absorption spectroscopy, Differential scanning calorimetry, and X-ray diffraction measurements. We achieve better understanding of amorphous structure of GeTe phase-change material, quantify nanoscale effects, processes at the interface, alloying and crystallization phenomena. Importantly, our results allow building a model of ideal GeTe glass and revealing the mechanism, kinetics, and size dependency of crystallization in chalcogenide phase-change materials.

Finally, for the engineering work package, we aim at integration of solution-based spin-coating and printing fabrication of PCM thin films and patterns. We have optimized spin-coating and printing conditions of PCM inks and complement these results with standard nanofabrication methods to complete PCM devices in several configurations. We have demonstrated stable switching characteristics and excellent energy efficiency of such liquid-borne PCM devices. In addition, we have realized a non-volatile reflective PCM device in which an ultrathin film of telluride nanoparticles is sandwiched between two transparent electrodes on top of metallic mirror.

Progrès au-delà de l’état des connaissances et impact potentiel prévu (y compris l’impact socio-économique et les conséquences sociétales plus larges du projet jusqu’à présent)

The results of this proposal will have an impact on chemistry, materials science, and nanotechnology. We report many new materials in form of high-quality colloidal nanoparticles. Technologically relevant is that our synthetic approaches are technically simple, scalable, reproducible, and inexpensive. Furthermore, successful synthetic method should be expandable to many materials (i.e. universality of the synthesis). To exemplify our chemistry contribution, we have developed the nanoscale amalgamation reaction for the synthesis of bimetallic nanocrystals. Starting from monometallic seeds, we carry out a thermal decomposition of metal-amides to dispatch low-melting metals to the surface of nanocrystals and thus trigger a time-efficient and homogeneous alloying to bimetallic compositions. Our method is capable to combine two very dissimilar metals at the nanoscale and unlock up to a 1000 of new intermetallic nanocrystals with unprecedented quality for size uniformity, composition control, and phase purity.

As many other technologies, PCM memory undergoes device miniaturization. Currently, the device dimension is at 22 nm, but it is expected to scale down to below 10 nm soon. At such small physical dimensions, all phase-change properties attain size dependence, for example melting point decreases, kinetics slows, and crystallization temperature increases as the size becomes smaller. However, little is known about the nanoscale effects on phase-change properties and especially fractured is the knowledge on amorphous structure of phase-change materials. This project will thus contribute to fundamentals of chalcogenide glasses and will facilitate the scaling of PCM memory to sub-10 nm size range.

Last but not least, this project introduces inexpensive liquid-phase deposition methods to the standard fabrication sequence of PCM arrays. Implementing high-throughput non-vacuum solution-based engineering methods will decrease a price of PCM devices and ultimately trigger their broad implementation instead of Si-based transistor memory. Moreover, liquid-phase PCM fabrication may unlock unique configurations, such as ultrathin and high-aspect-ratio devices and widen a selection of materials and substrates.

pcm-chemistry-summary.jpg