Showing posts with label ALD simulation. Show all posts
Showing posts with label ALD simulation. Show all posts

Tuesday, December 14, 2021

2022 Atomic Layer Processing Modelling Workshop

Pedersen group is organizing a work shop on modelling of atomic layer processes in Linköping 15-16 March next year with a Tutorial by Ray Adomaitis. Possibility to join via Zoom. 

Welcome to a forum where experimentalists and modellers from academia and industry meet to collaboratively push the boundaries of multi-scale modelling.

Predict the Future of Thin Films

Is complete in silico development of new materials and methods a utopia or just around the corner? Join us on-site or online and discuss state-of-the-art scientific methods to model atomic layer processes such as CVD, ALD and ALE, from reactorscale to atomic level.

The conference will have a mix of contributed talks, describing the latest in ALP modelling (atomic layer processing) and industry lectures presenting areas that need modelling. 


Prof. Raymond Adomaitis, from University of Maryland, will describe his way of modelling ALP in the tutorial “Reaction network analysis of ALD processes: Is this a true ALD cycle? What rates can be measured?”.


Date and time: March 15-16, starting on Tuesday at 12.00. 

Place: Planck, Fysikhuset. You will also be able to participate online via link.

Abstract deadline: February 2022.

The conference is free of charge but to participate you will have to register. Registration and Abstract aplication will soon be available here.

Tuesday, December 7, 2021

How Machine Learning Enables Accurate Prediction of Precursor Volatility

by Simon Elliott, Director of Atomic Level Process Simulation, Schrödinger

Challenges in predicting volatility

A crucial process in manufacturing CPUs and other high-tech devices is the deposition of solid material from reactive vapors. Different precursor vapors are used for chemical vapor deposition, vapor phase epitaxy, atomic layer deposition – and indeed the reverse process of atomic layer etching – with the precursor chemistry carefully designed for each case so as to control material quality at the nanoscale. But what all these techniques have in common is that the precursor chemicals must evaporate or sublime at a low enough temperature. Too much heating when vaporizing a precursor can make it decompose, causing it to be undeliverable to the growing surface.

With volatility playing such a central role in this technology (and in other fields like distillation, refrigeration, inkjet printing, food, and perfumes), it is surprising that we understand so little about it. Volatility is the product of a remarkably fine balance of interatomic forces, dictating the extent to which molecules condense together as a solid or liquid, or bounce apart into a vapor and deliver a certain vapor pressure at any given temperature. These interatomic forces can be computed very precisely with quantum mechanics for one molecule or a group of molecules, but not at the scale of a liquid or solid. Even with today’s computing power, routinely and accurately predicting precursor volatility ‘from first principles’ remains unfortunately out of reach.

Machine learning approaches

Could an alternative more empirical approach prove useful? Does enough experimental data exist to find the relation between volatility and chemical structure? The vaporization of some organic molecules, such as alcoholic fractions or natural fragrances, has been of interest for centuries and high-quality vapor pressure data are available in the literature. Over the last decade, these data have been analyzed with advanced fitting algorithms that come under the umbrella of ‘machine learning’. Schrödinger has leveraged the latest machine learning techniques to develop a highly accurate model that predicts the volatility of organic molecules up to C20.

However, when building machine learning models to predict volatility of precursor molecules, which are typically organometallic complexes, the situation is not so straightforward. New precursor molecules are constantly being proposed and evaluated. Commercial sensitivity sometimes means that data are partially withheld or plagued by experimental configuration differences from laboratory to laboratory. Additionally, for the common aim of material processing, complete pressure-temperature curves are rarely measured, as it is more pragmatic to focus on the temperature for vapor to transport successfully to the reactor. As a result, datasets for building predictive models are sparse and incomplete.

Prediction of volatility for inorganic and organometallic complexes 

Schrödinger scientists embarked on the challenge of building machine learning models to predict the volatility of precursor molecules. Using in-house expertise in machine learning and advanced informatics, Schrödinger scientists collated and digitized information about organometallic precursors from disparate literature sources and applied a variety of machine learning algorithms (such as Random Forest and Neural Networks) in conjunction with different chemoinformatic descriptors and fingerprints. The result is the first capability of its kind for accurately and efficiently predicting the volatility for inorganic and organometallic complexes from their chemical structures. For complexes of the fifty most common metals and semimetals, the model predicts the evaporation or sublimation temperature at a given vapor pressure with an average accuracy of ±9°C (which is about 3% of the absolute temperature). As a trained model, the turnaround time is fast with the ability to compute hundreds of complexes per second.

New avenues for precursor development

This predictive model opens a new path for designing novel precursors with improved performance, not only improving their deposition or etch chemistry, but also optimizing the temperature at which they evaporate or sublime and can be delivered as a vapor. This advance will allow a much wider range of structural modifications to be screened computationally than before and will produce candidate precursors for experimental synthesis and testing that are both less risky and more innovative. This volatility model, together with Schrödinger’s quantum mechanics-based workflows for computation of reactivity and decomposition, gives scientists a complete design kit for vapor-phase deposition or etch, delivering a faster pace of research into materials and processes for new technologies.

* The banner image is from Tyndall National Institute.

About the author

Dr. Simon Elliott is Director of Atomic Level Process Simulation at Schrödinger. From 2001-2018 he led a research group at Tyndall National Institute, Ireland. Prior to that, he studied chemistry at Trinity College Dublin and Karlsruhe Institute of Technology. He qualified as a Project Management Professional and is a Fellow of the Royal Society of Chemistry. He was co-chair of the 16th International Conference on Atomic Layer Deposition and chair of a 175-member COST network on the same topic.

About Schrödinger

Schrödinger is an industry-leading computational solutions provider for both life science and materials science, with a mission to improve human health and quality of life by transforming the way therapeutics and materials are discovered.

With the goal to accelerate the discovery and optimization of novel materials by a digital chemistry platform governed by physics-based modeling, amplified by machine learning, and optimized through team-based intelligence, Schrödinger’s Materials Science platform offers unprecedented insights into the mechanisms and properties of materials and chemical systems in a wide range of technological applications: Organic Electronics, Polymeric Materials, Consumer Packaged Goods, Catalysis & Reactive Systems, Semiconductors, Energy Capture & Storage, Complex Formulations, Metals, Alloys & Ceramics.

Learn more

Request more information

Tuesday, December 22, 2015

A fresh review in Advanced Materials on ALD modelling

Here is a fresh review in Advanced Materials on ALD modelling from Simon Elliott and co-workers at Tyndall National Institute University College Cork. It contains an rather interesting part on MLD/ALD combination.

The article lists the following four major challenges for modelling ALD:
  1. Computing Precursor Volatility
  2. The Effect of Weak Interactions on Adsorption
  3. Modeling Plasma-Enhanced ALD
  4. Simulating Processes Over Long Time Scales

Modeling Mechanism and Growth Reactions for New Nanofabrication Processes by Atomic Layer Deposition

Simon D. Elliott, Gangotri Dey, Yasheng Maimaiti, Hayrensa Ablat, Ekaterina A. Filatova1 and Glen N. Fomengia
Article first published online: 21 DEC 2015
DOI: 10.1002/adma.201504043


Recent progress in the simulation of the chemistry of atomic layer deposition (ALD) is presented for technologically important materials such as alumina, silica, and copper metal. Self-limiting chemisorption of precursors onto substrates is studied using density functional theory so as to determine reaction pathways and aid process development. The main challenges for the future of ALD modeling are outlined.

Queen Elisabeth investigating The Tyndall Effect sporting a clean room hat and Class 100 stealth clean room gloves at a visit to Tyndall Institute, University College Cork in 2011.


Saturday, October 3, 2015

Workshop Simulation of chemistry-driven growth phenomena for metastable materials

CECAM/Psi-k/HERALD Workshop
Simulation of chemistry-driven growth phenomena for metastable materials

The controlled growth of thin films based on metastable materials by chemistry-driven processes is of high technological importance for topics like semiconductor devices or optical coatings. Computational modelling of this inherently multiscale process is crucial for an atomistic understanding and enables a decoupling and separate optimization of the growth-determining factors of non-equilibrium materials. This workshop will result in a joint effort by experts from different modelling communities covering the necessary length and time scales.
The workshop will be held at

nearby Marburg in Germany from November 08-11, 2015.




GRK 1782 ESFDock/Chemicals

Wednesday, September 16, 2015

Surface Chemistry of Copper Metal and Copper Oxide ALD

To produce continuous non-island forming films of copper by ALD is extremely difficult. Here is a good article from Fraunhofer ENAS in Chemnitz, Germany, on the mechanism behind ALD of Copper and coper oxide using the rather well studied Cu(acac)2 precursor but not so easy process. 

Surface Chemistry of Copper Metal and Copper Oxide Atomic Layer Deposition from Copper(II) Acetylacetonate: A Combined First-Principles and Reactive Molecular Dynamics Study

(Physical Chemistry Chemical Physics) Monday September 14th 2015
Author(s): Xiao Hu, Joerg Schuster, Stefan Schulz, Thomas Gessner

Atomistic mechanisms for the atomic layer deposition using the Cu(acac)2 (acac = acetylacetonate) precursor are studied by first-principles calculations and reactive molecular dynamics simulations. The results show that Cu(acac)2 chemisorbs on the hollow site of the Cu(110) surface and decomposes easily into a Cu atom and the acac-ligands. A sequential dissociation and reduction of the Cu precursor [Cu(acac)2→Cu(acac)→Cu] is observed. Further decomposition of the acac-ligand is unfavorable on the Cu surface. Thus additional adsorption of the precursors may be blocked by adsorbed ligands. Molecular hydrogen is found to be nonreactive towards Cu(acac)2 on Cu(110), whereas individual H atoms easily lead to bond breaking in the Cu precursor upon impact, and thus release the surface ligands into the gas-phase. On the other hand, water reacts with Cu(acac)2 on a Cu2O substrate through a ligand-exchange reaction, which produces gaseous H(acac) and surface OH species. Combustion reactions with the main by-products CO2 and H2O are observed during the reaction between Cu(acac)2 and ozone on CuO surface. The reactivity of different co-reactants toward Cu(acac)2 follows the order H > O3 > H2O.

Saturday, September 5, 2015

Role of Surface Termination in Atomic Layer Deposition of Silicon Nitride

Yet another fundamental publication from Eindhoven and Oxford Instruments on one of the most important (PE)ALD processes for scaled semiconductor devices - silicon nitride. This time Tyndall has helped them out to sort out the growth mechanism to better understand growth promotion and inhibition that has been reported previously - BTBAS Silicon nitride PEALD by TU Eindhoven, Oxford Instruments and ASM Microchemistry

Role of Surface Termination in Atomic Layer Deposition of Silicon Nitride

Chaitanya Krishna Ande†, Harm C. M. Knoops†‡, Koen de Peuter†, Maarten van Drunen†, Simon D. Elliott§, and Wilhelmus M. M. Kessels*†

† Department of Applied Physics, Eindhoven University of Technology, Den Dolech 2, 5600 MB Eindhoven, The Netherlands
‡ Oxford Instruments Plasma Technology, North End, Bristol BS49 4AP, United Kingdom
§ Tyndall National Institute, University College Cork, Dyke Parade, Lee Maltings, Cork, Ireland
J. Phys. Chem. Lett., 2015, 6, pp 3610–3614
DOI: 10.1021/acs.jpclett.5b01596

There is an urgent need to deposit uniform, high-quality, conformal SiNx thin films at a low-temperature. Conforming to these constraints, we recently developed a plasma enhanced atomic layer deposition (ALD) process with bis(tertiary-butyl-amino)silane (BTBAS) as the silicon precursor. However, deposition of high quality SiNx thin films at reasonable growth rates occurs only when N2 plasma is used as the coreactant; strongly reduced growth rates are observed when other coreactants like NH3 plasma, or N2–H2 plasma are used. Experiments reported in this Letter reveal that NHx- or H- containing plasmas suppress film deposition by terminating reactive surface sites with H and NHx groups and inhibiting precursor adsorption. To understand the role of these surface groups on precursor adsorption, we carried out first-principles calculations of precursor adsorption on the β-Si3N4(0001) surface with different surface terminations. They show that adsorption of the precursor is strong on surfaces with undercoordinated surface sites. In contrast, on surfaces with H, NH2 groups, or both, steric hindrance leads to weak precursor adsorption. Experimental and first-principles results together show that using an N2 plasma to generate reactive undercoordinated surface sites allows strong adsorption of the silicon precursor and, hence, is key to successful deposition of silicon nitride by ALD.

Wednesday, July 29, 2015

Tyndall propose an new metallocene reducing agent pathway for ALD of copper

The Materials Modelling For Devices group headed by Simon Elliott at Tyndall National Institute, University College Cork propose using metallocene compounds as reducing agents for atomic layer deposition (ALD) of the transition metal Cu from metalorganic precursors resulting in a new pathway for ALD of copper. The screening results of 10 different compunds has been published in Dalton Transactions recently and generated the status - Hot Article!

Quantum chemical and solution phase evaluation of metallocenes as reducing agents for the prospective atomic layer deposition of copper 

Gangotri Dey, Jacqueline S. Wrench, Dirk J. Hagen, Lynette Keeney and Simon D. Elliott
Dalton Trans., 2015,44, 10188-10199
DOI: 10.1039/C5DT00922G 

We propose and evaluate the use of metallocene compounds as reducing agents for the chemical vapour deposition (and specifically atomic layer deposition, ALD) of the transition metal Cu from metalorganic precursors. Ten different transition metal cyclopentadienyl compounds are screened for their utility in the reduction of Cu from five different Cu precursors by evaluating model reaction energies with density functional theory (DFT) and solution phase chemistry.

Saturday, June 13, 2015

Simulation of a Multihole outlet Savannah ALD reactor

For those of you interested in ALD reactor simulations and especially the classic cross flow design you should definitely check out this publication. University of Wisconsin-Milwaukee, University of Alaska Anchorage and Macquarie University, Sydney.

Investigating atomic layer deposition characteristics in multi-outlet viscous flow reactors through reactor scale simulations

Mohammad Reza Shaeri,Tien-Chien Jen, Chris Yingchun Yuan, Masud Behnia
Available online 6 June 2015

The Atomic Layer Deposition system at the Laboratory for Sustainable and Nano-Manufacturing in Milwaukee was required in October 2009 from Cambridge Nanotech Inc. (Today Ultratech / Cambrideg Nanotech). The Model is a Savannah S100 with capabilities for 100 mm wafers up to 10 wafer batch processing in a single deposition ( Either this is actually a picture from 2009 when the tool was new or these guys take really good care of their equipment as indicated that the protective shipping foil is still on.

In order to minimize the operational time of atomic layer deposition (ALD) process, flow transports and film depositions are investigated in multi-outlet viscous flow reactors through reactor scale simulations. The simulation process is performed on depositions of Al2O3 films using trimethylaluminum and ozone as the precursors, and inert argon as the purge gas. The chemistry mechanism used includes both gas-phase and surface reactions. Simulations are performed at a fixed operating pressure of 10 Torr (1330 Pa) and at two substrate temperatures of 250 °C and 300 °C, respectively. Flows inside the reactors are following the continuum approach; as a result, the Navier–Stokes, energy and species transport equations can be used to simulate transient, laminar and multi-component reacting flows. Based on the chemistry mechanism adopted in this study, the amount of oxygen atoms produced from the ozone decomposition is found to be the major reason for discrepancies in oxidation times and deposition rates at different ALD processes. A reactor with fewer outlets minimizes the ALD operational times by reducing both oxidation time and second purge time. In addition, higher deposition rates at a shorter time are obtained by using a reactor with fewer outlets. However, assigning a long enough time for the ozone exposure results in independency of ALD characteristics from the number of outlets such that the growth rates of around 3.78 angstrom/cycle and 4.52 angstrom/cycle are obtained for the substrate temperatures of and , respectively.

At Lund Nano Lab in Sweden we also operate a Savannah from about the same time and it is one of the more popular tools judging by the frequent user bookings. Since Ultratech came in the picture some things have happened. Today you get a slim white design Generation 2 and the delivery time in Europe can be as fast as 8 weeks which is pretty fast if you ask me. Check out the new product page at Cambridge Nanotech here.

The Savannah is available in three configurations: S100, S200, and S300 and is capable of holding substrates of different sizes (up to 300mm for the S300).