% IMPORTANT: The following is UTF-8 encoded. This means that in the presence
% of non-ASCII characters, it will not work with BibTeX 0.99 or older.
% Instead, you should use an up-to-date BibTeX implementation like “bibtex8” or
% “biber”.
@PHDTHESIS{Zhou:1014292,
author = {Zhou, Yiming},
othercontributors = {Wuttig, Matthias and Waser, Rainer},
title = {{C}rystallization mechanism and switching kinetics of
$\mathrm{{I}n_{3}{S}b{T}e_{2}}$ based phase change
materials},
school = {RWTH Aachen University},
type = {Dissertation},
address = {Aachen},
publisher = {RWTH Aachen University},
reportid = {RWTH-2025-06007},
pages = {1 Online-Ressource : Illustrationen},
year = {2025},
note = {Veröffentlicht auf dem Publikationsserver der RWTH Aachen
University; Dissertation, RWTH Aachen University, 2025},
abstract = {Phase change materials (PCMs) can be switched between their
high-resistive amorphous phase and low-resistive crystalline
phase. The prominent property contrast of PCMs exists not
only in their electrical resistance but also in their
switching kinetics. While both phases of PCMs demonstrate
stability for more than a decade at ambient temperatures,
they can be reversibly switched within nanoseconds at
elevated temperatures through electrical pulses. Such high
non-linearity in temperature dependent stability and high
resistance contrast enables PCM-based non-volatile data
storage known as phase change memory. As the first
commercialized technology for storage class memory (e.g. 3D
XPoint), their large-scale deployment in practical
applications has highlighted key optimization targets for
novel PCMs. Among all performance metrics, the switching
kinetics, especially the crystallization kinetics defines
the competitiveness of PCMs. Two major parts have been
accomplished in this work. Since the electrical switching of
PCMs requires both device dimensions below the 100 nm scale
and electrical pulse generating and sensing system on the
nanosecond scale, a characterization platform has been
established, including developing the fabrication process
for confined PCM devices (30-100 nm critical dimensions) and
a semi-automatic probing system from scratch. To achieve
accurate measurements of crystallization kinetics, the
endurance of the device was engineered up to 108 cycles,
enabling detailed study of switching process. The protocol
of the kinetics measurement employed a two-step write-verify
strategy which minimized the artifact of cycle-to-cycle
variability. With this protocol, the switching kinetics of
prototype PCM Ge2Sb2Te5 reveals a strong similarity between
the optical crystallization speed from the different
location of an amorphous film and the electrical
crystallization speed from a single device from
cycle-to-cycle. Therefore, the characterization platform
presented great potential for the investigation of
crystallization kinetics. Second, this work systematically
investigates the crystallization kinetics of In3SbTe2 -based
PCMs. The investigation of the crystallization behavior of
In3SbTe2 thin film through structure characterization and
microstructural analysis revealed a spherulitic growth
mechanism in In3SbTe2, traced to the excessive tetrahedral
indium motif that exists in amorphous In3SbTe2 depending on
its density. While spherulitic growth is a relatively slow
crystallization mechanism due to the limited diffusion,
reducing device thickness to 50 nm suppressed growth-front
nucleation, boosting switching speeds of intrinsic In3SbTe2
to 18 ns, which is 5 times faster than the state-of-the-art
In3SbTe2 with tailored doping. Also, SnTe was introduced
into In3SbTe2 to reduce tetrahedral motifs in the amorphous
phase, though phase separation challenges persist despite
both components have the same rock-salt structure and low
lattice mismatch. Using comprehensive advanced transient
electrical response characterization and statistical
analysis of stochastic switching, we demonstrate that
nucleation is not only an initiating process but also an
accelerating factor in the SET operation of SnTe-doped
In3SbTe2 -devices. At voltages slightly above the threshold
voltage, stochastic nucleation and subsequent crystal growth
serve as the switching mechanism in device. The measured
minimum nucleation time is as low as 2 ns, although it
happens randomly. Fitting with a Gompertz function provides
statistical calculations of nucleation probability and
yields a typical nucleation time of 8 ns for $30\%$ SIST.
With high applied voltages near the RESET voltage, the
switching mechanism of the device mainly comes from the
deterministic growth at crystalline-amorphous interfaces
near top electrode with spatial limitation. The crystal
growth rate is lower than the nucleation-growth case. By
directly correlating transient electrical responses to
physical nucleation and growth events, we establish a new
paradigm for understanding crystallization mechanisms in
nanoscale PCM devices which have implications for
next-generation memory technologies exhibiting ultrafast
operation.},
cin = {131110 / 130000},
ddc = {530},
cid = {$I:(DE-82)131110_20140620$ / $I:(DE-82)130000_20140620$},
pnm = {SFB 917 Z04 - Technologieplattform für nanoskalige ReRAM-
und PCM-Bauelemente (Z04*) (426850996) / SFB 917: Resistiv
schaltende Chalkogenide für zukünftige
Elektronikanwendungen: Struktur, Kinetik und
Bauelementskalierung "Nanoswitches" (167917811) /
G:(DE-82)BMBF-16ME0398K Verbundprojekt: Neuro-inspirierte
Technologien der künstlichen Intelligenz für die
Elektronik der Zukunft - NEUROTEC II (BMBF-16ME0398K)},
pid = {G:(GEPRIS)426850996 / G:(GEPRIS)167917811 /
G:(DE-82)BMBF-16ME0398K},
typ = {PUB:(DE-HGF)11},
doi = {10.18154/RWTH-2025-06007},
url = {https://publications.rwth-aachen.de/record/1014292},
}
Gast ::