In several fields of basic and applied research, progresses and advances are strongly connected to a joint effort involving both experiments and modelling activities. Historically, simulations are based on the applications of theory to relatively simple models of complex systems. In the past years, however, high-performance computing infrastructures have enabled the realization of computational experiments, which are able to produce new knowledge by replicating reality in a completely virtual environment. In this context, the virtualization of scanning probe microscopy investigations is potentially able to boost the experimental work, by enhancing for example resolution and high-throughput screening capabilities. At the same time, a suitable approach for the simulation of microscopy activities will contribute to the rationalisation and to the design of complex experiments. In this work, we present a computational framework for the simulation of scanning probe microscopy experiments and, in particular, based on atomic force microscopy (AFM) techniques. In our approach, we implement strategies for the definition of realistic, full- scale models, including structural features at the atomistic scale and taking the effect of complex environments into account. The need for including details of the systems under investigation is especially relevant to the application of AFM to biological systems. In these latter, the intrinsic complexity of the system (in terms of structure, morphology, dynamical phenomena, interaction with the substrate, role of solvent and ionic solutions) requires detailed and realistic model systems. Our modelling framework reproduces the morphology of the AFM tip, mimicking the experimental set-up, and the interactions with a substrate in terms of topology (surface scanning) and forces (tip-substrate interactions). This set-up provides access to several properties of the system that are usually not accessible from experiments, including dynamical phenomena and details with sub-nanoscale (atomistic) resolution. The computational framework relies on the application of large-scale molecular dynamics (MD) simulations, based on quasi- atomistic coarse grained (CG) model potentials. Our Virtual Microscopy framework is applied to investigations of the surface properties of materials for applications in technology, such as polymers, and biological materials in complex environments. In particular, we simulate AFM experiments on supported lipid bilayers (SLBs) and membranes, supported extra-cellular vesicles (EVs), and synthetic nano-micelles. We show that our modelling tool is able to simulate quantitatively the mechanical response of SLBs and EVs with different composition, in excellent agreement with the available experimental data, thus allowing a computational screening of different materials, set-up and environments. Work is in progress to extend the virtualization framework for including further features of interest in AFM experiments, such as time-dependent properties (dynamical forces, frequency spectra, etc.).
The Virtual Microscope: towards realistic simulations of scanning probe experiments
F Mercuri
2019
Abstract
In several fields of basic and applied research, progresses and advances are strongly connected to a joint effort involving both experiments and modelling activities. Historically, simulations are based on the applications of theory to relatively simple models of complex systems. In the past years, however, high-performance computing infrastructures have enabled the realization of computational experiments, which are able to produce new knowledge by replicating reality in a completely virtual environment. In this context, the virtualization of scanning probe microscopy investigations is potentially able to boost the experimental work, by enhancing for example resolution and high-throughput screening capabilities. At the same time, a suitable approach for the simulation of microscopy activities will contribute to the rationalisation and to the design of complex experiments. In this work, we present a computational framework for the simulation of scanning probe microscopy experiments and, in particular, based on atomic force microscopy (AFM) techniques. In our approach, we implement strategies for the definition of realistic, full- scale models, including structural features at the atomistic scale and taking the effect of complex environments into account. The need for including details of the systems under investigation is especially relevant to the application of AFM to biological systems. In these latter, the intrinsic complexity of the system (in terms of structure, morphology, dynamical phenomena, interaction with the substrate, role of solvent and ionic solutions) requires detailed and realistic model systems. Our modelling framework reproduces the morphology of the AFM tip, mimicking the experimental set-up, and the interactions with a substrate in terms of topology (surface scanning) and forces (tip-substrate interactions). This set-up provides access to several properties of the system that are usually not accessible from experiments, including dynamical phenomena and details with sub-nanoscale (atomistic) resolution. The computational framework relies on the application of large-scale molecular dynamics (MD) simulations, based on quasi- atomistic coarse grained (CG) model potentials. Our Virtual Microscopy framework is applied to investigations of the surface properties of materials for applications in technology, such as polymers, and biological materials in complex environments. In particular, we simulate AFM experiments on supported lipid bilayers (SLBs) and membranes, supported extra-cellular vesicles (EVs), and synthetic nano-micelles. We show that our modelling tool is able to simulate quantitatively the mechanical response of SLBs and EVs with different composition, in excellent agreement with the available experimental data, thus allowing a computational screening of different materials, set-up and environments. Work is in progress to extend the virtualization framework for including further features of interest in AFM experiments, such as time-dependent properties (dynamical forces, frequency spectra, etc.).I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
