Random walk

Take 10 minutes to prepare this exercise.

Then, if you lack ideas to begin, look at the given clue and start searching for the solution.

A detailed solution is then proposed to you.

If you have more questions, feel free to ask them on the forum.

Because of the thermal motion, the particles in a gas or a liquid have broken lines trajectories.

Starting from the initial point O at \(t = 0\), a particle makes \(N\) moves \({\vec \delta _i}\) (\(i=1\) to \(N\)).

After each collision, it goes towards another direction, independent from the previous one.

Question

What is the average distance \(r\) between O and the particle after \(t=N\tau\) (\(N\) is huge), where \(\tau\) is the average time between to following collisions ?

Express \(r\) with \(t\), \(L\) (free mean path) and \(u\) (mean square velocity).

Solution

After \(N\) moves \({\vec \delta _i}\), the particle is in :

\(\vec r = \sum\limits_{i = 1}^N {{{\vec \delta }_i}}\)

So, in modulus :

\({\vec r\;^2} = {r^2} = \sum\limits_{i = 1}^N {\vec \delta _i\;^2} + 2\sum\limits_i {\sum\limits_{j < i} {{{\vec \delta }_i}.{{\vec \delta }_j}} }\)

And, as the definition of the free mean path states :

\({L^2} = \frac{1}{N}\sum\limits_{i = 1}^N {\vec \delta _i\;^2}\)

Furthermore, for \(N\) really huge, the random characteristic of scattering makes cosinus equally distributed between \(-1\) and \(+1\), so :

\(\sum\limits_i {\sum\limits_{j < i} {{{\vec \delta }_i}.{{\vec \delta }_j}} } = 0\)

It remains, with \(L = u\tau\) :

\({r^2} = N{L^2} = \frac{t}{\tau }u\tau L = Lut\)

So :

\(r = \sqrt {Lut}\)

Question

Numerical application : estimate the time required by a fragrance to diffuse after the opening to a distance \(d = 10\; cm\) ; comment on it.

Solution

Numericale application :\( t=3 \;min \;20 \;s\).

Scattering phenomenons are really slow.

In reality, the molecules are heavier than air, so \(u\) is smaller and \(t\) is longer, and here is the interest of convection.