Chemotaxis of an elastic flagellated microrobot.

Machine learning algorithms offer a tool to boost mobility and flexibility of a synthetic microswimmer, hence may help us design truly smart microrobots. In this work, we design a two-gait microrobot swimming in circular or helical trajectory. It utilizes the coupling between flagellum elasticity and resistive force to change the characteristics of swimming trajectory. Leveraging a deep reinforcement learning (DRL) approach, we show that the microrobot can self-learn chemotactic motion autonomously (without heuristics) using only several current and historical chemoattractant concentration and curvature information. The learned strategy is more efficient than a human-devised shortsighted strategy and can be further greatly improved in a stochastic environment. Furthermore, in the helical trajectory case, if additional heuristic information of direction is supplemented to evaluate the strategy during the learning process, then a highly efficient strategy can be discovered by the DRL. The microrobot can quickly align the helix vector to the gradient direction using just several smart sequential gait switchings. The success for the efficient strategies depends on how much historical information is provided and also the steering angle step size of the microrobot. Our results provide useful guidance for the design and smart maneuver of synthetic spermlike microswimmers.

Exploring the link between coffee matrix microstructure and flow properties using combined X-ray microtomography and smoothed particle hydrodynamics simulations.

Coffee extraction involves many complex physical and transport processes extremely difficult to model. Among the many factors that will affect the final quality of coffee, the microstructure of the coffee matrix is one of the most critical ones. In this article, we use X-ray micro-computed (microCT) technique to capture the microscopic details of coffee matrices at particle-level and perform fluid dynamics simulation based on the smoothed particle hydrodynamics method (SPH) with the 3D reconstructured data. Information like flow permeability and tortuosity of the matrices can be therefore obtained from our simulation. We found that inertial effects can be quite significant at the normal pressure gradient conditions typical for espresso brewing, and can provide an explanation for the inconsistency of permeability measurements seen in the literature. Several types of coffee powder are further examined, revealing their distinct microscopic details and resulting flow features. By comparing the microCT images of pre- and post-extraction coffee matrices, it is found that a decreasing porosity profile (from the bottom-outlet to the top-inlet) always develops after extraction. This counterintuitive phenomenon can be explained using a pressure-dependent erosion model proposed in our prior work. Our results reveal not only some important hydrodynamic mechanisms of coffee extraction, but also show that microCT scan can provide useful microscopic details for coffee extraction modelling. MicroCT scan establishes the basis for a data-driven numerical framework to explore the link between coffee powder microstructure and extraction dynamics, which is the prerequisite to study the time evolution of both volatile and non-volatile organic compounds and then the flavour profile of coffee brews.

Hydrodynamic clustering of two finite-length flagellated swimmers in viscoelastic fluids.

Clustering of flagellated microswimmers such as sperm is often mediated by hydrodynamic interactions between them. To better understand the interaction of microswimmers in viscoelastic fluids, we perform two-dimensional simulations of two swimming sheets, using a viscoelastic version of the smoothed dissipative particle dynamics method that implements the Oldroyd-B fluid model. Elasticity of sheets (stiff versus soft) defines two qualitatively different regimes of clustering, where stiff sheets exhibit a much more robust clustering than soft sheets. A formed doublet of soft sheets generally swims faster than a single swimmer, while a pair of two stiff sheets normally shows no speed enhancement after clustering. A pair of two identical swimmers is stable for most conditions, while differences in the beating amplitudes and/or frequencies between the two sheets can destroy the doublet stability. Clustering of two distinct swimmers is most stable at Deborah numbers of De = τω ≈ 1 (τ is the relaxation time of a viscoelastic fluid and ω is the beating frequency), in agreement with experimental observations. Therefore, the clustering of two swimmers depends non-monotonically on De. Our results suggest that the cluster stability is likely a dominant factor which determines the cluster size of collectively moving flagellated swimmers.

Hypernetted-chain-like closure of Ornstein-Zernike equation in multibody dissipative particle dynamics.

We have derived a hypernetted-chain-like (HNC-like) approximate closure of the Ornstein-Zernike equation for multibody dissipative particle dynamics (MDPD) system in which the classic closures are not directly practicable. We first point out that the Percus's method is applicable to MDPD system in which particles interact with a density-dependent potential. And then an HNC-like closure is derived using Percus's idea and the saddle-point approximation of particle free energy. This HNC-like closure is compared with results of previous researchers, and in many cases, it demonstrates better agreement with computer simulation results. The HNC-like closure is used to predict the cluster crystallization in MDPD. We determine whether the cluster crystallization will happen in a system utilizing the widely applicable Hansen-Verlet freezing criterion and by observing the radial distribution function. The conclusions drawn from the results of the HNC-like closure are in agreement with computer simulation results. We evaluate different weight functions to determine whether they are prone to cluster crystallization. A new effective density-dependent pairwise potential is also proposed to help to explain the tendency to cluster crystallization of MDPD systems.

Application of the dissipative particle dynamics method to the instability problem of a liquid thread.

We investigate the application of the dissipative particle dynamics method to the instability problem of a long liquid thread surrounded by another fluid. The dispersion curves obtained from simulations are compared with classic theoretical predictions. The results from standard dissipative particle dynamics (DPD) simulations at first have a tendency of gradually approaching to Tomotika's Stokes flow prediction when the Reynolds number is decreased. But they then abnormally deviate again when the viscosity is very large. The same phenomenon is also confirmed in droplet retraction simulations when also compared with theoretical Stokes flow results. On the other hand, when a hard-core DPD model is used, with the decrease of the Reynolds number the simulation results did finally approach Tomotika's predictions when Re≈0.1. A combined presentation of the hard-core DPD results and the standard DPD results, excluding the abnormal ones, demonstrates that they are approximately on a continuum when labeled with Reynolds number. These results suggest that the standard DPD method is a suitable method for investigation of the instability problem of immersed liquid thread in the inertioviscous regime (0.1

Mesoscopic simulation of a thinning liquid bridge using the dissipative particle dynamics method.

In this research, the dissipative particle dynamics method was used to investigate the problem of thinning and breakup in a liquid bridge. It was found that both the inertial-force-dominated thinning process and the thermal-fluctuation-dominated thinning process can be reproduced with the dissipative particle dynamics (DPD) method by varying the simulation parameters. A highly suspect viscous thinning regime was also found, but the conclusion is not irrefutable because of the complication of the shear viscosity of DPD fluid. We show in this article that the DPD method can serve as a good candidate to elucidate crossover problem in liquid bridge thinning from being hydrodynamics dominated to being thermal fluctuation dominated.