Computer simulations associated denser crowding with a smaller, more compact polymer, but the structural pattern was linked to the size of the particles around the chain. Small crowders mainly affected short-wavelength structure, while large crowders were associated with stronger compaction at long wavelengths. The chain’s motion also differed between the two simulation approaches: LD showed strong dependence on crowder size, whereas LBMD was less dependent on it.
Two ways to model the solvent
The study compared Langevin dynamics, or LD, with lattice-Boltzmann molecular dynamics, or LBMD. LD used an implicit-solvent description, while LBMD incorporated long-range hydrodynamic interactions. The LD calculations used polymer lengths of 96 and 192 monomers; LBMD used only the 96-monomer chain. The modeled crowders had radii of 2, 4, 12 and 18 nanometres, and their volume fractions ranged from 0.05 to 0.60.
The uncrowded chain set the baseline
In the uncrowded reference, the 96-monomer polymer had a radius of gyration, a measure of its overall spread, of 6.6σ in LD and 6.7σ in LBMD. The 192-monomer polymer measured 10.0σ. The longer chain was therefore larger in the reference system, while the two methods gave similar sizes for the shorter chain.
Crowding changed the scale of the polymer’s shape
At higher crowder volume fractions, the polymer’s radius of gyration was smaller and its conformation was more compact across both chain lengths and all crowder sizes. The structural pattern was scale-dependent: small crowders mainly affected short-wavelength structure, whereas large crowders were associated with reduced long-wavelength scaling and stronger compaction at long scales.
One way the simulations described long-scale structure was through an effective Flory exponent. In ordinary terms, this is a number used to describe how polymer size scales at long wavelengths. For large-crowder systems, the estimate reached as low as 0.42, while small-crowder systems remained closer to 0.588. These were effective scaling estimates, subject to finite-size and accessible-wavelength limits.
A related analysis supported a confinement-blob picture for the large-crowder cases. In this interpretation, sections of the polymer behave as blobs within colloidal voids. The closest-colloid distance had a fairly linear relationship with the blob monomer count raised to the 0.6 power, supporting that interpretation.
Motion depended on the solvent model
The clearest contrast between the methods appeared in diffusion. In LD, polymer diffusion depended strongly on both crowder volume fraction and crowder size. The factor ln(1 + R/r) gave a reasonable collapse across crowder sizes, meaning the results lined up more closely when expressed with that factor. Additional scaling by the polymer’s radius of gyration reduced the scatter further, although the paper said it was difficult to determine the statistical significance of that extra reduction.
In LBMD, diffusion followed an approximately exponential dependence on volume fraction over a wide range and was less dependent on crowder size than in LD. The pattern deviated at high volume fraction near random close packing. The only condition described as significantly subdiffusive at long times, with displacement spreading more slowly than in ordinary diffusion, was LBMD with volume fraction 0.6 and crowder radius 18 nanometres.
Relaxation did not follow one pattern
The study also tested Rouse-mode relaxation, which tracks how different patterns of motion along a chain die away. In LD, the expected scaling was less consistent at low volume fraction, especially for short-wavelength modes, and more consistent at sufficiently high volume fraction. In LBMD, the expected Rouse-mode scaling held for every mode number and across all volume fractions, including bulk conditions.
The reported dynamics pointed to more than one regime. In LBMD, the dynamic exponent z was about 3 at low to intermediate volume fractions, then increased at high volume fraction for very large crowders. In LD, z approached 3 at moderate to high volume fractions when the crowders were comparable to monomers. The estimates showed substantial statistical fluctuations because of limited runtimes.
A result with a defined scope
Only 96- and 192-monomer chains were studied, with LBMD run only for N = 96. The study reports no formal uncertainty intervals for the size and diffusion trends, and describes the Flory values as effective estimates subject to finite-size and accessible-wavelength limits. The document is an arXiv version 1 preprint dated 26 August 2026.
Financial support was reported from NSERC and the Ontario Graduate Scholarship, with computing resources from SHARCNET and the Digital Research Alliance of Canada. The authors reported no conflicts to disclose.
Paper data and sources
Original title: Dynamical and conformational behavior of a polymer in a crowded solution
Authors: Setarehalsadat Changizrezaei, Colin Denniston
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-26
DOI: Not available
Original paper · Full text