Computational modeling of diffusive dynamics in a bouncer system with an irregular surface
Luiz Antonio Barreiro · 2025
Read Computational modeling of diffusive dynamics in a bouncer system with an irregular surface by Luiz Antonio Barreiro (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2509-10253.txt.
Introduction
Luiz Antonio Barreiro studies a bouncing ball on a rough, irregular surface in this solo 2025 physics paper. The ball's sideways motion, though governed by ordinary deterministic rules, behaves like a random walk. Barreiro supports that reading by computing a permutation entropy, a measure of how disordered the motion's patterns are.
Over time, the spread of likely ball positions settles into a Gaussian, bell-shaped pattern. The width of that spread grows with time following a power law, the signature of diffusion. The lesson reaches beyond bouncing balls: a fully deterministic system with complex shapes or nonlinear rules can produce behavior that statistics cannot tell apart from true randomness. The paper closes with suggested problems for extending the analysis.
Catalog record
| CREATOR | Luiz Antonio Barreiro |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2509-10253
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Computational modeling of diffusive dynamics in a bouncer system with an irregular surface Luiz Antonio Barreiro Abstract The horizontal dynamics of a bouncing ball interacting with an irregular surface is investigated and is found to demonstrate behavior analogous to a random walk. Its stochastic character is substantiated by the calculation of a permutation entropy. The probability density function associated with the particle positions evolve to a Gaussian distribution, and the second moment follows a power-law dependence on time indicative of diffusive behavior. The results emphasize that deterministic systems with complex geometries or nonlinearities can generate behavior that is statistically indistinguishable from random. Several problems are suggested to extend the analysis. I Introduction Diffusion is a fundamental physical process that describes the movement of particles from regions of higher concentration to regions of lower concentration due to random motion. This process is not only a key concept in physics but is also crucial in chemistry, biology, and materials science. From the spreading of ink in water to the exchange of gases in the lungs, diffusion is an essential mechanism that governs numerous natural and technological phenomena. Crank 1975 ; Berg 1993 The study of diffusion offers an excellent opportunity to introduce students to core concepts in statistical mechanics and transport phenomena. The mathematical framework of diffusion is often described by Fick’s laws, first formulated in the 19th century, which provide a quantitative understanding of how particles spread over time. ( Fick 1855 ) Furthermore, the study of Brownian motion, first observed by Robert Brown in 1827 and later explained by Albert Einstein in 1905, presents an engaging way to connect diffusion with the kinetic theory of matter and stochastic processes. ( Einstein 1905 ) Beyond its applications in physics, diffusion is a critical topic across multiple disciplines. In chemistry, it governs reaction kinetics and mass transport in solutions and gases. ( Atkins and de Paula 2018 ) In biological systems, diffusion enables essential processes such as oxygen transport and cellular respiration. ( et al. 2014 ) In engineering and materials science, diffusion influences the behavior of semiconductors, corrosion rates, and the efficiency of batteries and fuel cells. ( Shewmon 2016 ) Additionally, diffusion plays a crucial role in atmospheric science, helping to explain pollutant dispersion and climate dynamics. ( Seinfeld and Pandis 2016 ) Due to its interdisciplinary significance, diffusion serves as a valuable framework for physics education, offering a natural bridge between theoretical concepts and real-world phenomena. This work examines the foundational principles of diffusion, with an emphasis on its mathematical formulation and computational modeling as effective pedagogical tools for deepening students’ understanding of transport processes. To explore diffusive behavior without resorting to stochastic techniques such as random number generation, we investigate a deterministic system based on classical free fall. Although the dynamics of a falling particle under gravity are well understood, additional complexity arises when the particle interacts with a structured surface. The coupling between the vertical descent and lateral displacement introduces rich dynamical behavior that can lead to statistical dispersion. We focus on a particularly illustrative scenario in which a particle undergoes successive collisions with a sinusoidal surface. This setup induces a progressive spreading of the particle’s horizontal position over time. In the following we will simulate this motion, analyze the resulting distribution of the horizontal displacements, and highlight key features associated with chaotic dynamics and diffusion, thus providing insights into the transition from simple mechanical motion to complex transport phenomena. II The model We follow the approach presented in Ref. Boscolo et al. 2023 and consider a particle falling on a non-flat surface described by y=\beta\left(\sin(\alpha x)+1\right). (1) The particle undergoes successive collisions with the surface, being reflected at each impact and subsequently following a parabolic trajectory until the next encounter. In the absence of dissipation, this process repeats indefinitely. To analyze this sequence of collisions, it is essential to determine the spatial coordinates of the impact points as well as the corresponding velocity vectors at the moments of contact. The reflection velocity vector at each collision is obtained by considering the local slope of the surface at the point of impact. The motion can be analyzed graphically as depicted in Fig. 1 . Figure 1: The collision details are shown in the circular region, including the normal and tangential vectors to the curve at impact. The dashed line illustrates the trajectory of the particle, and the collision and their corresponding normal vectors are indicated by dotted arrows. The law of reflection relates the incident velocity vector \boldsymbol{v}_{i}^{(\rm in)} to the reflected velocity vector \boldsymbol{v}_{i}^{(r)} as in Eq. ( 6 ). Following the i th collision at (x_{i},y_{i}) , the particle is reflected with velocity \boldsymbol{v}_{i}^{(r)}=\left(v_{x_{i}}^{(r)},v_{y_{i}}^{(r)}\right) and subsequently follows a ballistic trajectory under the influence of gravity until it reaches the next impact (x_{i+1},y_{i+1}) . The corresponding time of flight for this trajectory is denoted by \Delta t_{i,i+1}=t_{i+1}-t_{i} , such that the total elapsed time after N_{\rm col} collisions is given by t_{N_{\rm col}}=\sum_{i=1}^{N_{\rm col}}\Delta t_{i-1,i} , with the initial condition t_{0}=0 . The position of the particle during the free-flight phase evolves according to \displaystyle x_{i+1} \displaystyle=x_{i}+v_{x_{i}}^{(r)}\,\Delta t_{i,i+1}, (2) \displaystyle y_{i+1} \displaystyle=y_{i}+v_{y_{i}}^{(r)}\,\Delta t_{i,i+1}-\frac{g}{2}\left(\Delta t_{i,i+1}\right)^{2}. (3) These expressions are evaluated iteratively. To advance the simulation, it is necessary to determine the time of flight \Delta t_{i,i+1} , which is obtained by imposing the condition that the coordinate y_{i+1} of the subsequent collision lies on the surface profile defined by Eq. ( 1 ). Under this constraint, Eq. ( 3 ) for y_{i+1} leads to the transcendental equation \displaystyle\beta\sin\left[\alpha\left(x_{i}+v_{x_{i}}^{(r)}\,\Delta t_{i,i+1}\right)\right] \displaystyle= \displaystyle\beta\sin(\alpha x_{i})+v_{y_{i}}^{(r)}\,\Delta t_{i,i+1} (4) \displaystyle-\frac{g}{2}\left(\Delta t_{i,i+1}\right)^{2}, which must be solved numerically to determine \Delta t_{i,i+1} , given the known position x_{i} and the reflected velocity vector at the i th collision. The Appendix outlines the secant method for solving this transcendental equations numerically. In the limit \beta\ll 1 , we can assume that the height of the particle at a collision satisfies y_{i}\simeq y_{i+1}\simeq 0 , while the local slope of the surface may still be nonzero. This approximation eliminates the need to solve Eq. ( 4 ), significantly simplifying the analysis. For \beta\rightarrow 0 , Eq. ( 4 ) reduces to an algebraic form, and the flight time is readily obtained as \Delta t_{i,i+1}=\frac{2v_{y_{i}}^{(r)}}{g}. (5) To determine the post-collision velocity \boldsymbol{v}_{i}^{(r)} , it is assumed that the tangential component relative to the surface remains unchanged, while the normal component reverses sign. At the instant of collision, the law of reflection relating the incident velocity vector \boldsymbol{v}_{i}^{(\rm in)} to the reflected velocity vector \boldsymbol{v}_{i}^{(r)} is \boldsymbol{v}_{i}^{(r)}=\left(\boldsymbol{v}_{i}^{(\rm in)}\cdot\hat{\boldsymbol{t}}_{i}\right)\hat{\boldsymbol{t}}_{i}-\gamma\left(\boldsymbol{v}_{i}^{(\rm in)}\cdot\hat{\boldsymbol{n}}_{i}\right)\hat{\boldsymbol{n}}_{i}, (6) where \hat{\boldsymbol{t}}_{i} and \hat{\boldsymbol{n}}_{i} are the unit tangent and normal vectors, respectively. Inelastic collisions can be taken into account if the dissipative factor \gamma is included in the normal component of the reflected velocity. For elastic collisions, \gamma=1 . From Fig. 1 , the unit tangent and normal vectors can be expressed as follows \hat{\boldsymbol{t}}_{i}=\frac{1}{\sqrt{1+\lambda_{i}^{2}}}\left[\begin{array}[]{c}1\\ \lambda_{i}\end{array}\right],\quad\hat{\boldsymbol{n}}_{i}=\frac{1}{\sqrt{1+\lambda_{i}^{2}}}\left[\begin{array}[]{c}-\lambda_{i}\\ 1\end{array}\right], (7) where \lambda_{i} is the local slope of the surface, which from Eq. ( 1 ) is given by \lambda_{i}=\left.\frac{dy}{dx}\right\lfloor_{x_{i}}=\alpha\beta\cos\left(\alpha x_{i}\right). (8) The velocity vector incident at collision i+1 , is related to the velocity vector reflected at the previous collision i as \boldsymbol{v}_{i+1}^{(\rm in)}=\left[\begin{array}[]{c}v_{x_{i}}^{(r)}\\ v_{y_{i}}^{(r)}-g\,\Delta t_{i,i+1}\end{array}\right]. (9) Therefore, the reflected