Centre International de Formation et de Recherche Avancées en Physique

Activity 1. Theoretical/experimental training activities.

As part of this activity, seminars, courses, meetings with the project director and other members of the team were organized. A study database was created with recipes, theoretical and experimental protocols, results and documents. Also at this stage, gold nanoparticles (AuNPs) were synthesized and characterized.

AuNPs synthesis

The AuNP synthesis recipe involved a stock solution of chloroauric acid (HAuCl4) of 0.507M and a solution of 34 mM sodium citrate in demineralized water. Solutions of 0.101 mM HAuCl4 were obtained by diluting the stock solution in demineralized water.

A volume of 50 mL of the 0.101 HAuCl4 solution was heated at 160℃, on a hot plate, and a volume of 5 mL of sodium citrate solution was added once the solution reached the boiling temperature. The HAuCl4 solution is slightly yellowish in color and after the addition of the citrate solution it changed its color to purple, indicating that nucleation has been initiated. The solution was kept at boiling temperature for 15 minutes, after which it was allowed to cool to room temperature. Once cooled, the solution was stored at 4 ℃, wrapped in Al foil, until the spectral characterization was performed.

Morphological characterisation

In order to characterize the AuNPs by the UV-Vis spectroscopy method, samples were prepared from each type of nanoparticle solution obtained (in the synthesis solution, washed once, washed twice) as follows: a concentrated nanoparticle solution; a fourfold diluted nanoparticle solution; an eightfold diluted nanoparticle solution. Spectra were recorded with a Cary UVWIN 5000 Spectrophotometer at wavelengths between 300 and 800 nm in absorption mode. All samples were characterized using 1 cm x 1 cm PMMA cuvettes at room temperature (25 ℃±1). The characterization was performed immediately after washing and diluting the AuNP solutions to avoid the possibility of AuNP aggregation due to the decrease in concentration/removal of sodium citrate, which acts as stabilizer. From the analysis of the graphical representations of the absorption spectra of the gold nanoparticles for all concentrations and regardless of the number of washes performed, the absorption maximum is found at the value of 528 (± 1 ) nm, which demonstrates that the AuNP size is not influenced by the dilution or washing.

Figure 1. Absorption spectra of the AuNPs solutions in citrate solution, washed once and washed twice (concentrated solution, diluted four times and eight times).

Scanning electron microscopy
A scanning electron microscope (SEM) Gemini 5000 (Carl-Zeiss) was used for the morphological characterization of the AuNPs. For the comparison of AuNPs, the concentrated solutions obtained from the synthesis (sodium citrate), washed once and washed twice were analyzed. The samples were deposited on a silicon/silicon dioxide substrate. A volume of 20 µL of each prepared solution was placed with the help of a semi-automatic pipette on silicon wafers by the drop-casting technique. The plates were allowed to dry at room temperature. These results show that a more even distribution of AuNPs can be observed with the number of washes, but also a better differentiation of the nanoparticles. The mean diameter of the synthesized AuNPs was found to be 20 nm.

Figure 2. SEM images obtained at 10k magnification of AuNPs deposited on silicon substrate by drop-casting from the synthesis solution (A), washed once (B) and washed twice (C)

Figure 3. SEM image obtained at 200k magnification of AuNPs deposited on silicon substrate by drop-casting after 2 washes

The synthesis procedures will be continued with different ratios between the precursors to determine the influence of their concentration on the size of the nanoparticles but also on their stability.

Activity 2. Analysis of previous results and identification of reference ones to be reproduced through numerical simulations (benchmarking).

Numerical analysis of fluid flow through microfluidic channels

Using COMSOL Multiphysics, the flow in microfluidic channels was evaluated in 2D geometry. In the case of microfluidic devices, fluid flow through channels is characterized by very low Reynolds numbers (Re << 1), and the inertial forces in the Navier-Stokes equation becomes negligible. Thus, for the numerical modeling of the process, the Stokes flow regime was chosen, in which the theoretical description includes the equations for momentum conservation (1.1) and mass conservation (1.2):

The flow through the channels was simulated using incompressible fluids, so that the density will be a constant and the first term of equation (1.2) becomes 0. The material parameters used were specific to water (density = 1000 km m^-3 and dynamic viscosity at 25 ℃ = 8.9 x 10^-4 Pa s).

The chosen geometry includes two inlets, a reaction zone, an analysis zone and an outlet, Figure 4A). For the simulation of the process, a discretization grid (mesh) adapted to the geometry was chosen, so that in the intersection areas of the channel elements the discretization step is as small as possible, Figure 4B). The simulation conditions were imposed on the inlet and outlet areas as follows: flow of 1 mm/s for the two inlets and zero pressure for the outlet. The flow variation in the chosen geometry, the presence of vortices, the pressure distribution in the channel and the velocity distribution in the intersection areas were evaluated, Figure 5.

Figure 4. Channel geometry in three-dimensional form A) and divided into discretization elements for simulation B).

In the case of microfluidic channels with simple geometries, mixing of two or more fluids occurs predominantly under the action of molecular diffusion. Thus, sufficiently large flow distances are necessary for the homogenization or the reaction of the components. By including structures in the channel geometry that induce regions of turbulent flow,  in which collisions between particles can be stimulated in a passive way, thus inducing mixing and reaction of the components.
The numerical simulation results show the presence of vortexes in both circular areas of the channel, but their impact is more pronounced in the reaction zone, thus proving its application in the catalysis of the reaction between two compounds in aqueous phase. In the case of the analysis area, the presence of vortexes occupies a reduced area. This fact, combined with the low flow rates, implies the possibility of performing measurements with low noise.

Figure 5. A) Flow rate distribution and representation of vortexes in the channel. B) Pressure distribution over the entire surface of the channel and velocity distribution at the intersection areas.

Molecular Dynamics Simulations

LAMMPS is a classical molecular dynamics (MD) software that models assemblies of particles in liquid, solid or gaseous states. It can model atomic, polymeric, biological, solid-state (metals, ceramics, oxides), granular, coarse-grained or macroscopic systems using a variety of interatomic potentials (force fields) and boundary conditions. It can model 2D or 3D systems with sizes ranging from just a few particles to billions.

LAMMPS can be built and run on a single laptop or on desktop computers, but it is designed for parallel computers. It will also run serially on any parallel machine that supports the MPI message passing library. These include multicore servers with shared memory, multi-CPU and clusters, and distributed memory supercomputers. Parts of LAMMPS also support OpenMP multi-threading, vectorization and GPU acceleration. LAMMPS is written in C++ and requires a compiler that is at least compatible with the C++-11 standard. Previous versions were written in F77, F90 and C++-98. All versions can be downloaded as source code from the LAMMPS website. LAMMPS is designed to be easily modified or extended with new capabilities such as new force fields, atom types, boundary conditions or diagnostics. More details can be obtained by consulting the dedicated manuals on the web page (https://www.lammps.org/) and the manual of this software (https://docs.lammps.org/Manual.html#).

A simple tutorial example of the software package is the self-assembly of small lipid-like molecules as mycellae: