DLA growth is closely related to the Laplace equation: the probability field of where a random walker will first touch the cluster satisfies Laplace's equation with zero boundary conditions on the cluster surface. This means growth is fastest at protruding tips (high probability density) and slowest in interior fjords (screened regions). The resulting self-similar structure obeys a power law N ~ R^D_f, where D_f ≈ 1.71 in 2D and ≈ 2.5 in 3D. Researchers use sandbox dimension, box-counting, and correlation methods to measure D_f precisely.

DLA patterns appear in a surprisingly wide range of physical systems. Electrochemical deposition of metals such as zinc or copper from solution produces branching dendrites with DLA-like morphology. Dielectric breakdown in thin films creates fractal discharge channels. Mineral dendrites in rocks (often mistaken for fossils) form through DLA-like precipitation. Viscous fingering, where a less-viscous fluid pushes into a more-viscous one in a Hele-Shaw cell, also produces DLA-like branching. Even some bacterial colonies and plant root networks display growth patterns consistent with DLA dynamics.

Yes. When stickiness is set below 1.0, particles do not necessarily stick on first contact and may bounce off and continue diffusing. This gives particles more opportunity to penetrate into the interior of the cluster before sticking, leading to a denser, more compact aggregate with a higher effective fractal dimension. At very low stickiness values the structure approaches a solid disc (D approaching 2). This regime is sometimes called the "Eden growth" limit. The transition from DLA (low stickiness = high penetration) to compact growth illustrates how attachment probability fundamentally shapes morphology.

DLA was introduced by Thomas Witten and Leonard Sander in their landmark 1981 paper "Diffusion-Limited Aggregation, a Kinetic Critical Phenomenon" published in Physical Review Letters. Witten and Sander performed computer simulations that showed random-walk particles aggregating into fractal structures, and measured the fractal dimension of the clusters. Their work launched an entire field of research into kinetic growth phenomena and non-equilibrium pattern formation. The model is still widely studied today both theoretically and in experiments.

DLA belongs to a family of Laplacian growth models alongside viscous fingering and dielectric breakdown models. It is distinct from iterated function system (IFS) fractals such as the Sierpinski triangle or Mandelbrot set, because DLA is a stochastic, growth-based process rather than a deterministic recursive rule. The Eden growth model is a related process where particles attach to any perimeter site with equal probability, producing a compact cluster. Ballistic Aggregation is another variant where particles travel in straight lines rather than random walks, producing different fractal dimensions. This simulation also connects to random walks and Brownian motion explored in other physics simulations.

Engineers and material scientists exploit or suppress DLA-like growth depending on context. In battery technology, understanding and controlling dendritic growth (a DLA-like phenomenon) is critical: lithium dendrites growing on battery electrodes can cause short circuits and fires, so researchers develop electrolyte additives and nanostructured anodes to inhibit fractal growth. Conversely, DLA-inspired geometry is used to design fractal antennas with wideband characteristics and to create high-surface-area electrodes for supercapacitors and fuel cells. DLA algorithms also appear in computer graphics for generating realistic tree branches, coral, lightning, and river networks.

Active research directions include multi-particle DLA with interactions and excluded-volume effects, DLA on curved surfaces and complex networks, and the exact computation of D_f in 2D (which has resisted rigorous proof for over 40 years). Researchers also study off-lattice DLA, anisotropic variants that model crystal symmetry, and DLA under external fields (electric, gravitational). In biology, DLA frameworks are being applied to model tumour growth, vascular network formation, and neurite branching. Machine learning is increasingly used to predict cluster morphology from growth parameters, a problem that remains difficult to solve analytically.