Diffusive Limits for the Random Lorentz Gas

Definitions and mathematical statement

Let $X^{r,\rho}(t)$ denote the random Lorentz-gas particle trajectory named in the supplied record, and let $\mathcal N$ denote the event described there as "the particle is not trapped." The record does not define the environment distribution, the dynamics, the spatial dimension, the parameters $r,\rho$, the initial conditions, or the event $\mathcal N$. These are unresolved parts of the formulation.

For $T>0$, define the rescaled trajectory by \[ Y_T(t)=\frac{X^{r,\rho}(Tt)}{\sqrt T},\qquad 0\leq t\leq 1. \] The supplied question asks whether, after conditioning on $\mathcal N$, the laws of $Y_T$ converge weakly as $T\to\infty$ to the law of a Wiener process with nondegenerate variance, in the quenched or averaged-quenched setting.

More explicitly, once the missing model data and conditioning laws have been specified, the intended convergence has the schematic form \[ \mathcal L_{\star}\left(Y_T\mid\mathcal N\right) \Longrightarrow \mathcal L(W), \qquad T\to\infty, \qquad \star\in\{\text{quenched},\text{averaged-quenched}\}. \] Here $\mathcal L_{\star}$ denotes the trajectory law in the indicated setting; its precise definition is not supplied. If the spatial state space is ${\mathbb{R}}^d$, the usual meaning of a Wiener process with nondegenerate variance is a process $W$ starting at zero, with continuous paths and independent Gaussian increments satisfying \[ \mathbb E[W(t)-W(s)]=0,\qquad \operatorname{Cov}(W(t)-W(s))=(t-s)\Sigma, \quad 0\leq s\leq t\leq1, \] where $\Sigma$ is a positive definite $d\times d$ matrix. The dimension and intended covariance convention require confirmation; isotropy is not stated.

This is a statement of the supplied target, subject to unresolved qualifications. The record does not specify a path space or topology for weak convergence, parameter quantifiers, an environment-almost-sure quantifier for the quenched assertion, or the order of averaging and conditioning in the averaged-quenched assertion. The conditioning also requires an appropriate positive-probability nontrapping event, or another explicitly defined conditional law. No complete theorem with these missing hypotheses is asserted.

Short English statement

The supplied problem asks whether a random Lorentz particle, conditioned never to be trapped, has a nondegenerate Brownian limit under the scaling $X^{r,\rho}(Tt)/\sqrt T$ on $[0,1]$, in a quenched or averaged-quenched formulation. The model, conditioning, parameter ranges, and convergence topology need clarification before this becomes a precise theorem statement. The supplied open status is dated curation, not an independently verified conclusion.

Sources

• Supplied local UnsolvedMath registry record, permanent ID $30003959$, title: Diffusive Limits for the Random Lorentz Gas; collection: Oberwolfach Reports Open Problems; exact source problem code: catalogue entry. The record gives proposed year 2018 and names no individual proposer.
• Supplied report citation: Scaling Limits in Models of Statistical Mechanics (2019), Oberwolfach Reports, pp. 2535--2581, DOI 10.4171/owr/2018/41. Source URL and supplied literature-source link labeled "Oberwolfach Report 41, Lorentz particle": https://doi.org/10.4171/owr/2018/41. The supplied background describes Oberwolfach Reports as documenting workshops and their open-problem discussions.
• Supplied literature-source link labeled "Lutsko--Toth, Lorentz gas invariance principle": https://doi.org/10.1007/s00220-020-03852-8. No further bibliographic details are supplied.
• Literature assessment dated exactly 2026-08-21: the supplied curation labels the problem open and describes the conditioned nontrapped Lorentz-particle invariance principle as a major open problem. It also says that later Lorentz-gas work records a corresponding periodic higher-dimensional invariance problem as open. This is supplied dated curation, not independently verified; its time-sensitive status, cited claims, and applicability to this formulation require independent checking.