Influence of packing protocol on fractal exponents in dense polydisperse packings

Artem A. Vladimirov, Alexander Yu. Cherny, Eugen M. Anitas, Vladimir A. Osipov · 2025

Read Influence of packing protocol on fractal exponents in dense polydisperse packings by Artem A. Vladimirov, Alexander Yu. Cherny, Eugen M. Anitas, Vladimir A. Osipov (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-2504-10095.txt.

Introduction

Artem Vladimirov and three co-authors study how disks pack into dense piles in a 2025 physics paper. Disk sizes follow a power law, from tiny to large. Such piles can show fractal patterns that repeat across scales. The team built piles with three different packing methods.

One method tiles the plane with triangles, another squeezes at steady pressure. A third grows in the classic Apollonian style. The ratio of largest to smallest disk proved to be a key control. Wide ratios made the fractal measures agree across methods. The pressure method left oddly large pores that skewed its measures. The authors link those pores to lower randomness in the pile. Their pore counting method may help read real granular materials.

Catalog record

CREATORArtem A. Vladimirov, Alexander Yu. Cherny, Eugen M. Anitas, Vladimir A. Osipov
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE26 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2504-10095

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Influence of packing protocol on fractal exponents in dense polydisperse packings Artem A. Vladimirov, Alexander Yu. Cherny, Eugen M. Anitas, Vladimir A. Osipov Abstract We study fractal properties of a system of densely and randomly packed disks, obeying a power-law distribution of radii, which is generated by using various protocols: Delaunay triangulation (DT) with both zero and periodic boundary conditions and the constant pressure protocol with periodic boundary conditions. The power-law exponents of the mass-radius relation and structure factor are obtained numerically for various values of the size ratio of the distribution, defined as the largest-to-smallest radius ratio. It is shown that the size ratio is an important control parameter responsible for the consistency of the fractal properties of the system: the greater the ratio, the less the finite size effects are pronounced and the better the agreement between the exponents. For the DT protocol, the exponents of the mass-radius relation, structure factor, and power-law distribution coincide even at moderate values of the size ratio. By contrast, for the constant-pressure protocol, all three exponents are found to be different for both moderate (around 300) and large (around 1500) size ratios, which might indicate a biased rather than random spatial distribution of the disks. Nevertheless, there is a tendency for the exponents to converge as the size ratio increases, suggesting that all the exponents become equal in the limit of infinite size ratio. I Introduction Compact packings with power-law size distributions are of great interest for various applications in science and engineering, in particular for understanding the structural properties of materials Torquato 2018 . This understanding is critical for applications ranging from ultra-high performance concrete Lyu et al. 2021 to biological systems Sato et al. 2018 and soils Wang et al. 2022 , as it enables the design and optimization of materials with desired mechanical properties and stability. The spatial correlations in such systems exhibit fractal-like behaviour Cherny et al. 2022 ; Cherny et al. 2023 ; Monti et al. 2023 ; Cherny et al. 2024 and the description of the emerging fractal properties is one of the most interesting and intriguing problems. In numerical simulations, dense packing can be achieved by various methods (see Torquato 2018 for a review), including random successive addition (RSA) Rényi 1958 ; Widom 1966 . In RSA, particles are added one at a time in random positions. If a particle overlaps with existing ones, it is discarded. This process continues until no more particles can be added without overlap. Recently, the authors suggested Cherny et al. 2024 Delaunay triangulation (DT) protocol as a modification of RSA. DT creates a network of triangles connecting the centers of packed particles, which optimizes the process of finding empty space for newly added particles. Another method of dense packings is the constant pressure (CP) protocol, which uses LAMMPS (large-scale atomic/molecular massively parallel simulator) with GRANULAR package Thompson et al. 2022 . It allows the modeling of granular materials under controlled pressure conditions. Constant external pressure slowly compresses an initially dilute configuration of particles, leading to compact jammed packing. It was shown with RSA Cherny et al. 2023 and DT Cherny et al. 2024 protocols that dense random packings of disks with a power-law size distribution exhibit fractal-like properties with the fractal dimension D_{\mathrm{f}} being equal to the exponent of the power-law size distribution D . On the other hand, Monti et al. Monti et al. 2023 obtained jammed packings of the same distribution with CP protocol and argued that the packings, although exhibiting fractal properties, nevertheless have fractal dimensions different from the exponent of the power-law size distribution. The authors of the paper Monti et al. 2023 concluded that the fractal exponent depends on the packing protocol. Fractal dimension D_{\mathrm{f}} of a fractal is defined through the exponent in the mass-radius relation Mandelbrot 1982 ; Gouyet 1996 : M(r)\sim r^{D_{\mathrm{f}}} , which coincides with the exponent D_{\mathrm{f}} of decay of the structure factor S(q)\sim 1/q^{D_{\mathrm{f}}} . The relation between both exponents is based on Erdélyi’s theorem for asymptotic expansion of Fourier integrals Erdélyi 1956 , because the structure factor is the Fourier transform of the pair distribution function, which is proportional to \frac{\partial M(r)}{\partial r}\frac{1}{r} , see Appendix A . In practice, however, the fractal power-law behaviour is always realized within a finite fractal range, and thus the relation between the power-law exponent of the structure factor and fractal dimension should be used with some caveats: it is valid only if the fractal range in momentum space is sufficiently long . This “rule of thumb” is well known in experimental studies of fractal aggregates using small-angle scattering Schmidt 1991 , which directly yields the structure factor of the system under study. As shown in the previous paper Cherny et al. 2023 , the fractal range of densely packed disks or spheres with a power-law size distribution is determined by their largest-to-smallest size ratio, and this result is still valid in the thermodynamic limit Cherny et al. 2024 . In the paper Monti et al. 2023 , the ratio was taken in two dimensions from 100 to 300 for various exponents D of the distribution 1 1 1 Note that in the paper Monti et al. 2023 , the notation \beta=D+1 is used. . It is shown below in Sec. IV that these ratios are not large enough to obtain the correct value of the fractal dimension of the system. To this end, the data of jammed packings with CP protocol of Ref. Monti et al. 2023 are reproduced for a specific value of the exponent D=1.5 and the size ratio 292 . We calculate both structure factor and mass-radius relation and obtain the discrepancy between the exponents \alpha [ S(q)\sim 1/q^{\alpha} ] and D_{\mathrm{f}} [ M(r)\sim r^{D_{\mathrm{f}}} ] in the corresponding fractal ranges: \alpha\simeq 1.31 and D_{\mathrm{f}}\simeq 1.419 , see Fig. 2 and Table 1 below. This discrepancy is due to the insufficient size ratio. Increasing the size ratio to R/a=1575 and keeping the other parameters unchanged (see Sec. IV.2 below), we obtain \alpha\simeq 1.41 and {D_{\mathrm{f}}}\simeq 1.441 , which are much closer to each other, although its value is still different from D=1.5 predicted by our model of dense random packings Cherny et al. 2023 ; Cherny et al. 2024 . Nevertheless, this result suggests that the both exponents slowly converge to D as the size ratio R/a increases. This paper is organized as follows. In Sec. II , we explain in detail a set of disk to be packed and packing methods. In the next section, the fractal properties of dense random packing with DT protocol is studied for various values of the size ratio. In Sec. IV , the results of packing with CP protocol are represented. In Conclusions, we summarize the main results and outline prospects for future research. II A set of disks to be packed and methods Figure 1: Comparison of spatial distributions of N=125000 disks obtained from different protocols: (a) DT with zero boundary conditions (packing fraction 0.983). (b) CP with periodic boundary conditions (packing fraction 0.967). In both cases the ratio between the largest and smallest radii is 1575. One can notice some inhomogeneities and voids in the configuration generated with

Read the complete text