2024 Volume 33 Issue 12
Article Contents

Jing-Yi Li, Jin-Lei Shi, Ying-Ying Wang, Jun-Xing Pan, Jin-Jun Zhang. Capture behavior of self-propelled particles into a hexatic ordering obstacle[J]. Chinese Physics B, 2024, 33(12): 124501. doi: 10.1088/1674-1056/ad84c6
Citation: Jing-Yi Li, Jin-Lei Shi, Ying-Ying Wang, Jun-Xing Pan, Jin-Jun Zhang. Capture behavior of self-propelled particles into a hexatic ordering obstacle[J]. Chinese Physics B, 2024, 33(12): 124501. doi: 10.1088/1674-1056/ad84c6

Capture behavior of self-propelled particles into a hexatic ordering obstacle

  • Received Date: 15/07/2024
    Accepted Date: 10/09/2024
    Available Online: 01/12/2024
通讯作者: 陈斌, bchen63@163.com
  • 1. 

    沈阳化工大学材料科学与工程学院 沈阳 110142

  1. 本站搜索
  2. 百度学术搜索
  3. 万方数据库搜索
  4. CNKI搜索

Figures(5)

Article Metrics

Article views(608) PDF downloads(1) Cited by(0)

Access History

Capture behavior of self-propelled particles into a hexatic ordering obstacle

Abstract: Computer simulations are utilized to investigate the dynamic behavior of self-propelled particles (SPPs) within a complex obstacle environment. The findings reveal that SPPs exhibit three distinct aggregation states within the obstacle, each contingent on specific conditions. A phase diagram outlining the aggregation states concerning self-propulsion conditions is presented. The results illustrate a transition of SPPs from a dispersion state to a transition state as persistence time increases within the obstacle. Conversely, as the driving strength increases, self-propelled particles shift towards a cluster state. A systematic exploration of the interplay between driving strength, persistence time, and matching degree on the dynamic behavior of self-propelled particles is conducted. Furthermore, an analysis is performed on the spatial distribution of SPPs along the y-axis, capture rate, maximum capture probability, and mean-square displacement. The insights gained from this research make valuable contributions to understanding the capture and collection of active particles.

1.   Introduction
  • Active matter, originating from biological systems, is prevalent in nature and refers to substances that can actively absorb energy from the external environment and move spontaneously. This phenomenon is exemplified by the migration behavior of fish shoals on a macro scale and the movement of microscopic cells.[1–8] The collective behavior exhibited by active substances demonstrates rich non-equilibrium characteristics,[9] including motility-induced phase separation,[10–13] hexatic ordering,[14–16] and clustering.[17–19] As a result, active matter holds significant potential for applications not only in life sciences but also in the development of micro-devices and drug transport systems. However, it is noteworthy that active matter often operates in complex environments rather than uniform ones, as seen in the spread and invasion of cancer cells within the body[20–23] and the movement of molecular motor micro-machines shuttling proteins between cells.[24,25] Consequently, the study of active matter’s movement behavior in complex obstacle environments has garnered substantial attention from both physical and biological researchers.[26–36]

    In recent decades, researchers have made significant advances in developing new models and computational techniques aimed at understanding the unique multi-particle effects and aggregation phenomena exhibited by active matters in complex obstacle environments.[37–42] Bhadra Hrishikesh and Ethayaraja Mani identified a capture effect of active particles on passive particles, resulting in the formation of large clusters and phase separation within mixed systems of active and passive particles.[43] Conversely, Ma et al. observed the formation of large clusters by active particles under specific conditions.[44] Furthermore, Pan et al. demonstrated that active particles have the ability to gather and rotate around obstacles in a bath containing large obstacles.[45] Additionally, Shi et al. showed that active objects can navigate through complex media.[27] These findings have generated curiosity surrounding the capture behavior of self-propelled particles in intricate obstacle environments, prompting questions about leveraging obstacles to capture dispersed active particles, elucidating the physical mechanisms underlying capture behavior, and optimizing the capture efficiency. Thus, our study investigates the capture of active particles featuring a repulsive Weeks–Chandler–Anderson (WCA) potential in complex obstacle scenarios, focusing on the impact of driven strength, persistence time, and obstacle spacing ratio on the dynamic behavior and capture probability of self-propelled particles through Brownian dynamics simulation.

2.   Model and methods
  • We examine a two-dimensional model system illustrated in Fig. 1. The obstacle arrays are made up of round particles that share the same size as the self-propelled particles (SPPs). The obstacle is fixed in the middle of the simulation box, forming a hexagonal lattice configuration. To create a dense porous medium, we establish a lattice constant of a = 1.938σ for the array. Before initiating the simulation, round SPPs with mass m and size σ are randomly distributed, along with their initial propelling directions being arbitrarily assigned. Throughout the simulation, all particles interact via the purely repulsive WCA potential[46]

    with a cutoff at rc = 21/6σ. The motion of SPPs is described by the Langevin equations

    Equation (2) describes the translational motion of SPPs and equation (3) depicts the coupled rotational kinetics of the propelling direction. ri is the position vector of the ith SPP and Ui is its total interaction energy. We set the translational friction coefficient ζ = 10, which is large enough so that the motion of SPPs is effectively overdamped. The propelling forces on the SPPs have the same strength Fa but independent propelling directions. For example, the direction of the ith SPP μ^i=(cosθi,sinθi) rotates independently and diffusively with the rate Dr. The inverse of the rotational diffusion rate (1/Dr) characterizes the persistence time of the self-propelled motion. ηi (t) and ξi (t) are unit-variance Gaussian white noises. We choose m, σ and ε as scaling units (hence time unit τ=mσ2/ε) and set the scaled temperature kBT = 1. We keep the inertial terms in Eq. (2) so that the velocity–Verlet algorithm (by the home-modified software LAMMPS) and a large time step could be used.[47,48] The simulation box size is 200 σ × 100 σ. The periodic boundary condition is adopted. Typically, the simulation times are 107τ time steps. Unless otherwise specified, the global average concentration of SPPs ϕ is 20%.

3.   Result and discussion
  • First, we analyze the capture behavior of SPPs under various self-propelled conditions. In the absence of obstacles, the SPPs are uniformly distributed throughout the system (see the inset in Fig. 2(a)). When an obstacle is set into the system, three distinct representative aggregation patterns occur in the system (as shown in Fig. 2(a)). These patterns are associated with the dispersion state (Fig. 2(a1)), transition state (Fig. 2(a2)), and cluster state (Fig. 2(a3)). The distribution of SPPs along the y-axis in obstacles is depicted in Fig. 2(b), denoted as Na, which represents the number of particles. In the case of cluster state, SPPs aggregate in obstacle in the vertical direction and form one large cluster. The particle density distribution exhibits distinct characteristics, with a pronounced peak observed on the Na–y curve (the blue line in Fig. 2(b)). The microstructure of these clusters displays a large-scale hexagonal ordering. Conversely, in the cases of dispersion state and transition state, the particle density distribution lacks clear features without pronounced peak, and SPPs are uniformly spread throughout the obstacle area. Further observation indicates that at small driven strength and long persistence time, SPPs form multiple localized cluster structures within the obstacle. In the transition state, the microstructure of these clusters exhibits a short-range hexagonal ordering and these ordering structures are not observed in the dispersion state.

    We conduct extensive simulations to examine the capturing processes under varying levels of driven strength (Fa) and persistence time (1/Dr). The outcomes of these simulations are presented in Fig. 2(c). When the driven strength is small (Fa ≤ 25) and the persistence time is short (Dr ≥ 0.04), SPPs exhibit low persistence and slow movement. The particles have relatively short directional durations and frequent collisions, which enable them to “freely” navigate through obstacles without forming clusters. Consequently, the SPPs display a dispersed state. As the persistence time decreases to Dr ≤ 0.03 while Fa remains in the range of 5 to 20, the persistence time of SPPs increases, leading to a reduction in particle collisions. This, in turn, raises the likelihood of particles entering obstacles and forming small clusters within specific areas. These small clusters further impede the migration of internal particles. Ultimate, SPPs gather into clusters of different sizes in the obstacle; and the clusters are dispersed throughout the obstacle, resulting in transition states. On the other hand, for Fa values exceeding 25, SPPs quickly enter obstacles and form dense clusters, hindering the migration of internal particles and leading to the formation of a large cluster.

    To assess the capture process quantitatively, we statistically calculate the probability of SPPs being captured by the obstacle, denoted as P = (Nb/Na) × 100%, where Nb represents the total number of SPPs captured by the obstacle, and Na is the total number of SPPs in the system. The evolution of the capture probability of the obstacle to SPPs over time at different driven force (Fa) of 5, 10, and 30 is illustrated in Figs. 3(a)–3(c). The time for the obstacle to reach the steady capture is denoted by tsc.

    In the case of Fa = 5, the time tsc increases with the rise in Dr, as shown in Fig. 3(a). Upon increasing Fa to 10, the time tsc decreases, and with a further increase in Dr, the time tsc also rises, as depicted in Fig. 3(b). For Fa greater than or equal to 30, steady capture can be achieved within t ≤ 2000τ, and the capture time is minimally affected by Dr, as demonstrated in Fig. 3(c). In general, tsc decreases progressively with an increase in Fa and a decrease in Dr. Additionally, we determine the maximum capture probability (Pmax) under various propulsion conditions (Fig. 3(d)). First, when Fa ≥ 10, the maximum capture probability (Pmax) increases with the decrease of Dr, and Pmax is close to 100% at Dr = 0.001. Second, for Fa ≥ 5, Pmax also increases with the decrease of Dr, and it is worth noting that there is an anomaly at Dr = 0.001, indicating that SPPs are more likely to escape from obstacles under this specific condition. Third, Pmax increases with the increase of Fa at the same Dr.

    In order to understand the dynamic mechanism of the above phenomena, we calculate the ensemble average of the mean-square displacement (MSD) as a function of the time lag time Δt. MSD is calculated as follows:[49,50]

    where Tmax is the total simulation run time, r(t) represents the position of the tracer at an initial time t, r (t + Δt) denotes the tracer position after the lag time Δt, and N is the total number of independent trajectories. The time exponent α(t) is defined as

    The plots of MSD against t on the log–log scale and the corresponding time exponents (t) on log–linear scale are shown in Fig. 4. In the early times, SPPs exhibit a super-diffusive behavior where the growth rate of MSD is high (1 < α(t) < 2). In scenarios with long persistence times (Dr = 0.001) and low driven strength (Fa = 5) (refer to the black line in Fig. 4), SPPs consistently display super-diffusive characteristics throughout their movement. However, upon encountering obstacles, the SPPs’ movement is impeded, causing a slowdown in the growth rate of MSD, leading to particles aggregation within the obstacle and resulting in a transition state (refer to Fig. 2(a2)). By t = 1× 107τ, the aggregation state inside the obstacle stabilizes, and the rate of change in MSD decreases to its minimum. Meanwhile, SPPs that are unable to penetrate the obstacle obtain a higher growth rate of MSD, compensating for the loss of MSD exhibited by SPPs within the obstacle, and the exponent α(t) slightly increases (refer to Movie S1). Conversely, when Fa = 5 and Dr = 0.1 (refer to the red line in Fig. 4), SPPs that manage to enter the central region of the obstacle become effectively “pinned” within, inducing a swift transition of the system to a sub-diffusive state (α(t) < 1). This pinning of SPPs impedes the entry of additional particles, culminating in the system forming a dispersed state (refer to Fig. 2(a1)). In cases of Fa = 30 and Dr = 0.001 (refer to the green line in Fig. 4), nearly all SPPs swiftly infiltrate the obstacle, leading to a rapid decrease in α. Following the entry of SPPs into the obstacle, there ensues a period of structural reorganization within the cluster, evolving from a dispersed state into complete hexagonal ordering (refer to Fig. 2(a3)). Throughout this transition, the change in α approaches zero, and SPPs exhibit normal diffusive behavior at α(t) = 1. Subsequently, except for a few particles positioned outside the obstacle and at the cluster’s periphery, the MSD of SPPs dwindles close to zero, signifying a shift in the dynamics of SPPs towards sub-diffusive behavior (refer to Movie S2). For Fa = 30 and Dr = 0.1 (refer to the blue line in Fig. 4), the MSD continues to decrease until t = 1× 107τ. Subsequently, the system reaches a dynamic equilibrium, with α stabilizing as a constant.

    Furthermore, we consider the effect of obstacle on the capture behavior. We define the matching degree as R = (a–d1)/d2, where a is the distance between the centers of two adjacent obstacle particles, d1 is the diameter of the obstacle particle and d2 is the diameter of the self-propelled particle.

    The maximum capture probability (Pmax) as a function of R is statistically calculated under different matrix structures in the case of Fa = 30 and Dr = 0.001, as shown in Fig. 5. When the obstacle particles are arranged in a hexagonal structure. The maximum capture probability follows a non-monotonous trend with the increase in the matching degree. When R < 0.9 (indicating obstacle particles spacing (a – d1) smaller than the particle diameter d2), the SPPs tend to cluster at the edge of the obstacle, and it is difficult to enter the obstacle. As R ranges from 0.9 to 1.0 (indicating obstacle particles spacing (a – d1) equivalent to the particle diameter d2), SPPs tend to aggregate within the obstacle, forming a stable cluster state with a maximum capture probability Pmax nearing 100%. Subsequently, as R further increases, the migration of SPPs in obstacles becomes easier, which also means that it is easier for SPPs to escape from the obstacle, resulting in a decrease in Pmax. When R ≥ 1.7, the obstacle particles spacing is large enough to make the SPPs through freely, consequently resulting in a substantial decrease in Pmax. In addition, when the obstacles are arranged in a square structure, the trend of the maximum capture probability changing with the matching degree is similar to that of hexagonal obstacles. However, square obstacles exhibit a lower Pmax and a narrower range of matching degrees that can achieve higher capture probabilities. This is because square obstacles create channels that facilitate easier traversal for SPPs. Meanwhile, we also consider random structure obstacles and find that due to the scattered distribution of obstacle particles, SPPs are unable to form a dense stacking structure in the obstacles.

4.   Conclusion
  • In this paper, the capture behavior of self-propelled particles in a hexatic ordering obstacle is investigated by using molecular dynamics. The study reveals the following key findings. (i) Self-propelled particles exhibit three distinct states, dispersion state, transition state, and cluster state, depending on the prevailing self-propelled conditions. (ii) The optimal capture state and maximum capture probability (Pmax) closed to 100% can be obtained with the condition of large Fa and small Dr. (iii) SPPs demonstrate varied dynamic behaviors across different aggregation states, transitioning from super-diffusion to sub-diffusion behavior upon encountering obstacles. (iv) The capture efficiency is significantly influenced by the matching degree between the spacing of obstacle and the SPPs. A higher capture probability is observed within the range of 0.9 < R < 1.0. It is worth mentioning that this study exclusively considers SPPs governed by idealized WCA potential, and our results provide the theory and model reference for further research. Subsequent studies can introduce more complex interaction potentials to make it closer to the real system.

Figure (5)  Reference (50)

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return