If, according to the Maxwell-Boltzmann distribution


which is the equation from which the whole final equation is derived, then why is the final equation

$$f(v) = 4\pi v^2\left(\frac{m}{2{\pi}kT}\right)^{3/2}\exp\left(-\frac{mv^2}{2kT}\right)?\tag{2}$$

Clearly, here

$${f(v)}\propto v^2\exp\left(-\frac{mv^2}{2kT}\right).\tag{3}$$

Where have I made a mistake in understanding this?

  • 1
    $\begingroup$ The link you shared goes to a not-so-informative website. What's $\epsilon$? Let's say that $\epsilon = mv^2$. The proportionality statement still holds. $\endgroup$ Apr 20, 2022 at 12:52
  • $\begingroup$ The 'additional' terms are obtained when transforming the probability distribution from 3-D distribution into magnitude of the velocity vector (averaged over direction) using spherical coordinates. Referenced from ocw.mit.edu/courses/5-62-physical-chemistry-ii-spring-2008/… $\endgroup$
    – user4723
    Apr 20, 2022 at 13:28
  • 1
    $\begingroup$ Closely related, nearly duplicate // You may confuse Boltzmann and Maxwell-Boltzmann distributions, with the latter based on the former, adding the v^2 factor from dV=4pi.r^2 dr, being integrated for the radius spherical coordinate. $\endgroup$
    – Poutnik
    Apr 20, 2022 at 13:28
  • $\begingroup$ The funny thing is both posts share the same OP. $\endgroup$
    – Poutnik
    Apr 20, 2022 at 13:43
  • 1
    $\begingroup$ @Poutnik I thought it was familiar... $\endgroup$ Apr 20, 2022 at 13:58

1 Answer 1


The $v$ in your first expression (that $f(v) \propto \exp(-\varepsilon/kT)$) most likely refers to the true velocity which is a vector

$$\vec{v} = (v_x, v_y, v_z),$$

whereas the $v$ in the Maxwell–Boltzmann distribution refers to the magnitude of the velocity

$$v = |\vec{v}| = \sqrt{v_x^2 + v_y^2 + v_z^2}.$$

The latter is more useful to a chemist because we don't really care which direction the particle is moving in, only its speed. However, note that there are many possible combinations of $(v_x, v_y, v_z)$ which yield the same magnitude $v$; so the formula needs to be adjusted for this. The factor ends up being $4\pi v^2$, which is the surface area of a sphere with radius $v$ (you can think of the surface of this sphere as representing all possible velocities with magnitude $v$).

My suggestion would be to look for a derivation in a good physical chemistry textbook, which should explain this more thoroughly.

  • $\begingroup$ Actually, I did read the first expression that I gave, in Physical Chemistry by Peter Atkins, but I really couldn't share the link because I have a hard copy, and I couldn't find any similar derivation on the internet. So, I had to link that website. But the book did in fact give $f(v)\propto e^{\frac{-\epsilon}{kT}}$ $\endgroup$ Apr 20, 2022 at 14:08
  • $\begingroup$ Also, I found that the derivation in the same mentioned book is wrong in many places such as equating $f(v)$ with $f(v_x).f(v_y).f(v_z)$. $\endgroup$ Apr 20, 2022 at 14:10
  • $\begingroup$ I think I understand now. We were taking $\vec{v}$ and therefore only one combination of $v_x$, $v_y$ and $v_z$ is present(due to only one direction). Thus we need not add the $4πv^2$ part to accommodate for the total volume in the velocity space. Is this what you meant @orthocresol? $\endgroup$ Apr 20, 2022 at 14:32
  • 1
    $\begingroup$ I think that's pretty much what I said, yeah. $\endgroup$ Apr 20, 2022 at 14:36

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.