Abstract
We investigate the growth of needles from a flat substrate. We focus on the situation when needles suddenly begin to grow from the seeds randomly distributed on the line. The width of needles is ignored and we additionally assume that (i) the growth rate is the same for all needles; (ii) the direction of the growth of each needle is randomly chosen from the same distribution; (iii) whenever the tip of a needle hits the body of another needle, the former needle freezes, while the latter continues to grow. We elucidate the large time behavior by employing an exact analysis and the Boltzmann equation approach. We also analyze the evolution when seeds are located on a half-line, on a finite interval. Needles growing from the two-dimensional substrate are also examined.
Introduction: The Competition for Space
When substances crystallize from a vapor or melt, numerous tiny seeds form simultaneously on a surface. As these seeds grow, they inevitably compete for material and space. This process is governed by "geometrical selection": a seed’s chance of becoming a large crystal depends heavily on its orientation. Generally, crystals growing perpendicular to the surface (more "vertical") tend to outlast those growing at shallower angles.
While this phenomenon is well-observed in materials science, the underlying statistical mechanics are complex. This research simplifies the problem using a "needle model" to determine the large-time behavior of the system: How many needles survive over time, and what is the distribution of their growth angles?
The Needle Model
The authors define a deterministic model with random initial conditions. Seeds are randomly distributed on a 1D line or 2D plane. From these seeds, needles grow as rays at a constant speed but in random directions. The core of the model is the Collision Rule: If a needle tip hits the body of another needle, the "hitting" needle freezes. The "hit" needle is unaffected and continues growing.
This simple rule leads to a complex pruning process where only a fraction of the initial needles survive to reach significant heights, as shown in the evolution from dense initial growth to a sparse final skeleton.
Methods and Analysis
To analyze this system, the researchers employed several mathematical tools. They mapped the 2D needle growth onto a 1D line by tracking the projections of the needle tips, allowing them to relate the problem to known 1D traffic models.
For the "totally asymmetric" case where all needles lean in the same direction, they found an exact solution. For symmetric distributions, they established rigorous upper and lower bounds. Additionally, a mean-field approach using the Boltzmann equation was used to describe how the velocity (angle) distribution evolves over time. The geometric logic of these interactions is captured in Figure 2, illustrating how different intersection scenarios dictate survival.
Key Insights: Survival of the Steepest
The model confirms the intuitive "geometrical selection" rule. Needles with more vertical orientations have a much higher survival probability because they are less likely to run into the "body" of a neighbor.
Furthermore, the research reveals a deep, model-independent universality. When starting with a finite number of needles ($N$), the distribution of the number of ultimate survivors ($n$) is identical to the distribution of surviving clusters in 1D ballistic aggregation. This distribution is described by Stirling numbers of the first kind, suggesting that the pruning mechanism is fundamentally linked to universal combinatorics.
Key Results and Scaling Laws
The paper establishes specific decay laws for the density of surviving needles ($n$) over time ($t$):
- 1D Substrate: The density of needles decays as $n \sim t^{-1/2}$.
- 2D Substrate (Cylindrical Needles): For needles of constant width, the density decays as $n \sim t^{-2/3}$.
- 2D Substrate (Conical Needles): For needles whose width grows linearly with length, the density decays faster: $n \sim t^{-4/3}$.
- Inosculating Needles: In a more realistic 2D scenario where needles "merge" rather than one simply stopping, the density decays as $n \sim t^{-1}$.
Limitations and Future Outlook
The primary limitation identified is the Boltzmann equation's performance in 1D. Because 1D systems are highly sensitive to correlations between neighbors, the mean-field assumption of the Boltzmann approach is technically "uncontrolled," leading to erroneous amplitudes even if the scaling exponents are correct.
Future work could involve numerical validation of the predicted exponential tails of the velocity distribution in 2D systems, or moving beyond the deterministic "ray" model to include stochastic growth rates and fluctuating directions.
Summary updated