1
$\begingroup$

I am having trouble with a term that arises in a physics equation (deriving the Lagrange equation for one particle in one generalized coordinate, $q$, dimension from one Cartesian direction, $x$).

My struggle essentially boils down to manipulating the following expression in one of the terms, $$\frac{d}{dt}\left[\frac{dx(q)}{dq}\right],$$ where $x$ is a function of only $q$. I know the order of partial derivatives is interchangeable; is it also true here for total derivatives?

$\endgroup$
1

1 Answer 1

2
$\begingroup$

if $x=x(q$ only$)$ then \begin{equation}\frac{dx}{dq}=\frac{\partial x}{\partial q} \end{equation} therefore, \begin{equation}\frac{d}{dt}\frac{dx}{dq}=\frac{d}{dt}\frac{\partial x}{\partial q}= \frac{\partial \dot{x}}{\partial q}=\frac{\partial}{\partial q} \frac{dx}{dt}\neq\frac{d}{dq}\frac{dx}{dt} \end{equation}

Inequality in the end is because $\frac{dx}{dt}$ can be an explicit function of both $q$ and $\dot{q}$, in which case we cannot swap total derivative w.r.t $q$ with a partial one.

Be careful, even though $x$ is a function of $q$ only, $\frac{dx}{dt}$ will still not be zero as it's a total derivative and $x$ may have some implicit time dependence through $q$. That being said, $\frac{\partial x}{\partial t}$ is equal to zero as $x$ does not depend on time explicitly.

$\endgroup$
12
  • $\begingroup$ thanks for your response. How did you proceed from left to right in the second and third equalities (in the lower set of equations)? $\endgroup$ Commented Oct 6, 2022 at 18:03
  • $\begingroup$ $\dot{x}=\frac{dx}{dt}$. If $f$ is an explicit function of generalized coordinates only, the partial derivative of $f$ w.r.t any of the generalized coordinates commutes with partial/total derivative of $f$ w.r.t time. $\endgroup$
    – user292464
    Commented Oct 6, 2022 at 18:58
  • $\begingroup$ I see; I suppose this is what I was unaware of. Where can I find an explanation or proof of this property, please? $\endgroup$ Commented Oct 6, 2022 at 20:00
  • $\begingroup$ math.stackexchange.com/a/3830795 $\endgroup$
    – user292464
    Commented Oct 6, 2022 at 21:21
  • 1
    $\begingroup$ After all these years I’ve finally understood the difference between an ordinary derivative and a total derivative (despite them, infuriatingly, having the same notation!) $\endgroup$ Commented Oct 7, 2022 at 5:57