A metadynamics-based approach to sampling crystallisation events

Quigley, David; Rodger, P. Mark

doi:10.1080/08927020802647280

Cited by 39 publications

(45 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar variables have been used in previous classical molecular dynamics studies of crystallization. [ 36 ] The use of a local CV enables us to preselect the region where the nucleus will form during the metadynamics simulation (although, in principle, crystallites may also form outside this region). It also helps to reduce spurious effects due to the periodic boundary conditions by favoring the formation of quasi-spherical nuclei.…”

Section: Growth Of Crystalline Nuclei Generated Using Metadynamicsmentioning

confidence: 99%

Crystallization Properties of the Ge₂Sb₂Te₅ Phase‐Change Compound from Advanced Simulations

Ronneberger

Zhang

Eshet

et al. 2015

Adv Funct Materials

View full text Add to dashboard Cite

6407wileyonlinelibrary.com strong dependence of the crystallization speed on temperature has recently been linked to the high fragility of PCMs. [4][5][6][7][8][9][10] More specifi cally, this behavior has been attributed to a pronounced change in the activation energy and the prefactor for growth velocity as a function of temperature, which is a fi ngerprint of fragility.The GeSbTe compounds lying on the pseudobinary line GeTe-Sb 2 Te 3 are the most widely studied family of PCMs, due to their applications in DVD-RAM, rewritable Blu-Ray discs, and phase-change memories. [11][12][13][14][15][16][17][18][19][20][21] Recrystallization of GST is known to be triggered by nucleation events. However, in state-of-the-art memory cells, it may be governed by crystal growth at the interface with the crystalline matrix, due to the small size of the amorphous mark. For system sizes of the order of nanometers, crystallization occurs on the sub-nanosecond time scale at elevated temperatures, opening up the possibility of investigating this phenomenon by ab initio molecular dynamics (AIMD) based on density functional theory (DFT). [22][23][24][25][26] In previous studies [22][23][24] amorphous models of the prototypical compound Ge 2 Sb 2 Te 5 (GST) containing 60-180 atoms were crystallized using AIMD. These works provided valuable information about the microscopic mechanisms of crystallization, although they did not assess fi nite size effects, which are expected to play an important role in such small samples. Larger models of GST containing 460 and 640 atoms were later studied: [ 25,26 ] in one study, [ 25 ] crystallization was facilitated by inserting a crystalline seed inside the amorphous network, whereas, in another [ 26 ] extremely long simulations (4 ns) of 460-atoms models without structural constraints were carried out.The stable crystalline phase of GST is hexagonal and consists of alternating Ge, Sb, and Te layers. [ 27 ] However, there is experimental evidence that, upon fast crystallization of the glass or the supercooled liquid, a metastable rocksalt phase is formed. In this phase, Te atoms occupy one face-centered cubic sublattice, whereas Ge, Sb, and vacancies are randomly arranged in the second one. The amount of disorder in this sublattice is difficult to ascertain. Recently, it has been shown experimentally, [ 28 ] and subsequently corroborated by theory, [ 29 ] that the degree of randomness in crystalline GeSbTe and similar compounds (such as Ge 1 Sb 2 Te 4 ) can be reduced by thermal annealing, so as to induce disorder-driven insulator-metal transitions.In this work, we carry out AIMD simulations to study the crystallization kinetics of GST at high temperature (≈600 K) and estimate its growth velocity, as well as to determine the structural properties of the recrystallized phase, including the layer stacking Ge 2 Sb 2 Te 5 (GST) is an important phase-change material used in optical and electronic memory devices. In this work, crystal growth of GST at 600 K is investigated by ab initio molecular dynamics. S...

show abstract

Section: Growth Of Crystalline Nuclei Generated Using Metadynamicsmentioning

confidence: 99%

Crystallization Properties of the Ge₂Sb₂Te₅ Phase‐Change Compound from Advanced Simulations

Ronneberger

Zhang

Eshet

et al. 2015

Adv Funct Materials

View full text Add to dashboard Cite

show abstract

“…Our implementation of the metadynamics method is described in detail in Ref. 28 and briefly summarized below for convenience. A more general description of the method has recently been presented by Laio and Gervasio.…”

Section: Metadynamicsmentioning

confidence: 99%

“…Further details on the computation of these order parameters can be found in another paper. 28 We also employed a sixth order parameter, the contribution to the potential energy from all real-space terms involving a calcium or carbonate/bicarbonate ion. In the case of long range electrostatic terms ͑which are treated in our work by the Ewald method͒, the reciprocal space terms are not as easily partitioned, hence are excluded from this order parameter.…”

Section: Metadynamicsmentioning

confidence: 99%

Metadynamics simulations of calcite crystallization on self-assembled monolayers

Quigley

Rodger

Freeman

et al. 2009

The Journal of Chemical Physics

Self Cite

View full text Add to dashboard Cite

We show that recent developments in the application of metadynamics methods to direct simulations of crystallization make it possible to predict the orientation of crystals grown on self-assembled monolayers. In contrast to previous studies, the method allows for dynamic treatment of the organic component and the inclusion of explicit surface water without the need for computationally intensive interfacial energy calculations or prior knowledge of the interfacial structure. The method is applied to calcite crystallization on carboxylate terminated alkanethiols arrayed on Au (111). We demonstrate that a dynamic treatment of the monolayer is sufficient to reproduce the experimental results without the need to impose epitaxial constraints on the system. We also observe an odd-even effect in the variation of selectivity with organic chain length, reproducing experimentally observed orientations in both cases. Analysis of the ordering process in our simulations suggests a cycle of mutual control in which both the organic and mineral components induce complementary local order across the interface, leading to the formation of a critical crystalline region. The influence of pH, together with some factors that might affect the range of applicability of our method, is discussed.

show abstract

“…The most significant challenge is the identification of the important slowly changing variables. These collective variables (CV) can be used in metadynamics searches to provide directions for the exploration of the free energy surface (FES) [357,358]. They therefore, must lead to a reasonable projection of the motion of the system along the minimum reaction path (not necessarily known) from the reactive state to the product state.…”

Section: Free Energy Methodsmentioning

confidence: 99%

Simulating Interface Growth and Defect Generation in CZT – Simulation State of the Art and Known Gaps

Henager¹,

Gao²,

Hu³

et al. 2012

View full text Add to dashboard Cite

A metadynamics-based approach to sampling crystallisation events

Cited by 39 publications

References 35 publications

Crystallization Properties of the Ge₂Sb₂Te₅ Phase‐Change Compound from Advanced Simulations

Crystallization Properties of the Ge₂Sb₂Te₅ Phase‐Change Compound from Advanced Simulations

Metadynamics simulations of calcite crystallization on self-assembled monolayers

Simulating Interface Growth and Defect Generation in CZT – Simulation State of the Art and Known Gaps

Contact Info

Product

Resources

About

A metadynamics-based approach to sampling crystallisation events

Cited by 39 publications

References 35 publications

Crystallization Properties of the Ge2Sb2Te5 Phase‐Change Compound from Advanced Simulations

Crystallization Properties of the Ge2Sb2Te5 Phase‐Change Compound from Advanced Simulations

Metadynamics simulations of calcite crystallization on self-assembled monolayers

Simulating Interface Growth and Defect Generation in CZT – Simulation State of the Art and Known Gaps

Contact Info

Product

Resources

About

Crystallization Properties of the Ge₂Sb₂Te₅ Phase‐Change Compound from Advanced Simulations

Crystallization Properties of the Ge₂Sb₂Te₅ Phase‐Change Compound from Advanced Simulations