FEIT Research Project Database

Soft matter informatics: intelligent computational material simulation tool for polymer-colloid assemblies

Project Leader: Ellie Hajizadeh
Primary Contact: Ellie Hajizadeh (ellie.hajizadeh@unimelb.edu.au)
Keywords: artificial intelligence; computational fluid dynamics; computational materials science; rheology; surfactant and polymer structure in solution
Disciplines: Chemical & Biomolecular Engineering,Computing and Information Systems,Mechanical Engineering

The discovery of materials possessing a desired attribute is a holy grail of the materials science. ‘Forward design’ – searching through parameter space to find favourable combinations that yield materials with some chosen characteristic – is commonly adopted and has benefited from recent advances in multiscale materials simulation techniques and computational power. However, combinatorial space increases exponentially with dimensionality; therefore, such approaches can still be hampered by the cost associated with exploration of vast swathes of parameter space as well as potential couplings between tunable parameters [1].

With recent developments in machine learning (ML) optimisation techniques, we now have a powerful arsenal to tackle issues associated with the common ad hoc trial-and-error forward design approaches. ‘Materials 4.0’ has emerged as a transformational discipline that integrates ML-based optimisation algorithms into the realm of materials design [2]. ‘Materials 4.0’ enables development of novel ‘inverse design’ methodologies, where a desired material property is explicitly targeted, and the parameters necessary to achieve it are found by a way of solving a constrained optimisation problem.

Recently, several methods have emerged across disciplines that draw upon machine learning-based optimisation and computational materials simulation to create computer programs that tailor material responses to specified behaviours [3]. However, so far, the majority of the methods developed either involve black-box techniques, in which the optimiser operates without explicit knowledge of the material’s configuration space or requires carefully tuned algorithms with applicability limited to a narrow subclass of materials.

To address this critical gap in current approaches in adaptation of ML optimisation methods into the realm of materials design, in this project we aim to develop a formalism that can generate optimisers automatically by extending statistical mechanical models into the realm of design. The strength of this approach lies in its capability to transform statistical models that describe materials into optimisers to tailor them [4].

We will build on our previous work [5–10] and develop such a formalism for a model multicomponent soft assembles resembling that of paint suspensions, which involves interacting polymer chains (flow modifier) and paint colloidal latex particles in order to optimise the formulation parameters to achieve the desired paint flow behaviour/viscosity.


[1] Greg van Anders, Daphne Klotsa, Andrew S. Karas, Paul M. Dodd, and Sharon C. Glotzer, “Digital Alchemy for Materials Design: Colloids and Beyond”, ACS Nano, 9, 9542–9553 (2015).

[2] Debra J. Audus and Juan J. de Pablo, “Polymer Informatics: Opportunities and Challenges”, ACS Macro Lett., 6, 1078–082 (2017).

[3] MZ Miskin MZ and HM Jaeger, Evolving design rules for the inverse granular packing problem, Soft Matter 10, 3708–3715 (2014).

[4] Marc Z. Miskina, Gurdaman Khairab, Juan J. de Pablob, and Heinrich M. Jaeger, “Turning statistical physics models into materials design engines”, PNAS, 113, 34ndash;39 (2016).

[5] Elnaz Hajizadeh, Shi Yu, Shihu Wang, and Ronald G. Larson, “A novel hybrid population balance—Brownian dynamics method for simulating the dynamics of polymer-bridged colloidal latex particle suspensions”, J. Rheol. 62, 235–247 (2018).

[6] Elnaz Hajizadeh, Billy Todd, Peter Daivis, Journal of Rheology, 58, 281–305 (2014)

[7] Elnaz Hajizadeh, Billy Todd, Peter Daivis, J. Chem. Phys, 142, 174911 (2015)

[8] Elnaz Hajizadeh, Billy Todd, Peter Daivis, J. Chem. Phys, 141, 194905 (2014)

[9] Guorui Zhu, Hossein Rezvantalab, Elnaz Hajizadeh, Xiaoyi Wang, Ronald G Larson, Journal of Rheology, 60, 327–343 (2016)

[10] Elnaz Hajizadeh, Ronald G Larson, Soft Matter, 13, 5942–5949 (2017)

Towards Development of an Intelligent Computational Materials Engineering Tool