Difficulties with derivative of a vector [Landau Textbook]

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
3 replies · 2K views · last reply
GrimGuy
Messages
11
Reaction score
2
Hi guys, I'm having trouble computing a pass 1 to 106.15. It's in the pictures.

Typeset equations from a physics textbook: h_0α = 4k/c³ Σ_a m_a v_aα/|r−r_a| − (1/c³) ∂²f/∂t∂x^α (eq. 1), with Δf = φ = −Σ_a k m_a/|r−r_a|, f = −(k/2) Σ_a m_a|r−r_a|, and h_0α = k/(2c³) Σ_a m_a/|r−r_a| [7v_aα + (v_a·n_a)n_aα] (eq. 106.15).


So, what a have to do is the derivative of ##f## with respect to time and coordinates. Then I need to rearrange the terms to find the equation 106.15. I am using the following conditions. ##r## vector varies in space and the vector ##r_a## varies in time.

I did that:
Handwritten equation: partial derivative with respect to t of |r - r_a| equals minus (r - r_a) dot v_a, divided by |r - r_a|. Below it, v_a = (dx_a/dt) i + (dy_a/dt) j + (dz_a/dt) k.


This calculus is from. (Section 106)
Cover title page of The Classical Theory of Fields, Third Revised English Edition, Course of Theoretical Physics Volume 2, by L. D. Landau (Deceased) and E. M. Lifshitz, Institute of Physical Problems, USSR Academy of Sciences.
 
Attachments
  • Cover title page of The Classical Theory of Fields, Third Revised English Edition, Course of Theoretical Physics Volume 2, by L. D. Landau (Deceased) and E. M. Lifshitz, Institute of Physical Problems, USSR Academy of Sciences.
    1618494413115.png
    89.1 KB · Views: 237
Physics news on Phys.org
The spatial dependence of ##f## is in the argument ##\mathbf{r}##, whilst the time dependence is in the source positions ##\mathbf{r}_a##\begin{align*}
\partial_{\alpha} |\mathbf{r} - \mathbf{r}_a| &= \frac{x_{\alpha} - x_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} = n_{a \alpha}

\end{align*}Then taking the time derivative\begin{align*}
\partial_t \partial_{\alpha} |\mathbf{r} - \mathbf{r}_a| = \partial_t \left( \frac{x_{\alpha} - x_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} \right) &= \frac{-v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} + (x_{\alpha} - x_{a \alpha}) \partial_t \left( \frac{1}{|\mathbf{r} - \mathbf{r}_a|} \right) \\

&= \frac{-v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} + \frac{-(x_{\alpha} - x_{a \alpha})}{|\mathbf{r} - \mathbf{r}_a|^2} \left( \frac{(\mathbf{r} - \mathbf{r}_a) \cdot (-\mathbf{v}_a )}{|\mathbf{r} - \mathbf{r}_a|} \right) \\

&= \frac{-v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} + \frac{n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a)}{|\mathbf{r} - \mathbf{r}_a|}

\end{align*}Hence if ##f = -(k/2) \sum_a m_a |\mathbf{r} - \mathbf{r}_a |##, then\begin{align*}

\frac{\partial^2 f}{\partial t \partial x^a} &= \frac{k}{2} \sum_a m_a \left[ \frac{v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} - \frac{n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a)}{|\mathbf{r} - \mathbf{r}_a|} \right]\end{align*}and consequently\begin{align*}
h_{0 \alpha} &= \frac{8k}{2c^3} \sum_a \frac{m_a v_{a\alpha}}{|\mathbf{r} - \mathbf{r}_a|} - \frac{k}{2c^3} \sum_a \left[ \frac{m_av_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} - \frac{m_a n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a)}{|\mathbf{r} - \mathbf{r}_a|} \right] \\

&= \frac{k}{2c^3} \sum_a \frac{m_a}{|\mathbf{r} - \mathbf{r}_a|} \left[ 7v_{a \alpha} + n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a ) \right]\end{align*}
 
Last edited by a moderator:
etotheipi said:
The spatial dependence of ##f## is in the argument ##\mathbf{r}##, whilst the time dependence is in the source positions ##\mathbf{r}_a##\begin{align*}
\partial_{\alpha} |\mathbf{r} - \mathbf{r}_a| &= \frac{x_{\alpha} - x_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} = n_{a \alpha}

\end{align*}Then taking the time derivative\begin{align*}
\partial_t \partial_{\alpha} |\mathbf{r} - \mathbf{r}_a| = \partial_t \left( \frac{x_{\alpha} - x_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} \right) &= \frac{-v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} + (x_{\alpha} - x_{a \alpha}) \partial_t \left( \frac{1}{|\mathbf{r} - \mathbf{r}_a|} \right) \\

&= \frac{-v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} + \frac{-(x_{\alpha} - x_{a \alpha})}{|\mathbf{r} - \mathbf{r}_a|^2} \left( \frac{(\mathbf{r} - \mathbf{r}_a) \cdot (-\mathbf{v}_a )}{|\mathbf{r} - \mathbf{r}_a|} \right) \\

&= \frac{-v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} + \frac{n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a)}{|\mathbf{r} - \mathbf{r}_a|}

\end{align*}Hence if ##f = -(k/2) \sum_a m_a |\mathbf{r} - \mathbf{r}_a |##, then\begin{align*}

\frac{\partial^2 f}{\partial t \partial x^a} &= \frac{k}{2} \sum_a m_a \left[ \frac{v_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} - \frac{n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a)}{|\mathbf{r} - \mathbf{r}_a|} \right]\end{align*}and consequently\begin{align*}
h_{0 \alpha} &= \frac{8k}{2c^3} \sum_a \frac{m_a v_{a\alpha}}{|\mathbf{r} - \mathbf{r}_a|} - \frac{k}{2c^3} \sum_a \left[ \frac{m_av_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|} - \frac{m_a n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a)}{|\mathbf{r} - \mathbf{r}_a|} \right] \\

&= \frac{k}{2c^3} \sum_a \frac{m_a}{|\mathbf{r} - \mathbf{r}_a|} \left[ 7v_{a \alpha} + n_{a \alpha} (\mathbf{n}_a \cdot \mathbf{v}_a ) \right]\end{align*}
Thx man, I'll take a time to understand and check it. Thank you a lot.

EDIT: I checked it and everything is fine and i could undestand the whole explanation, except for one little thing. What u were considering to write it '##n_{a \alpha}##'. It is the modul of the unit vector, or it is just a simplifyed notation to write ##\frac{x_{\alpha} - x_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|}## ?
 
Last edited:
GrimGuy said:
EDIT: I checked it and everything is fine and i could undestand the whole explanation, except for one little thing. What u were considering to write it '##n_{a \alpha}##'. It is the modul of the unit vector, or it is just a simplifyed notation to write ##\frac{x_{\alpha} - x_{a \alpha}}{|\mathbf{r} - \mathbf{r}_a|}## ?
Not the modulus, it's the second option: the ##\alpha## component of the vector ##\mathbf{n}_a := (\mathbf{r} - \mathbf{r}_a)/|\mathbf{r} - \mathbf{r}_a|##, as L&L define below the bit you posted. So ##n_{a \alpha} := (\mathbf{r} - \mathbf{r}_a)_{\alpha}/|\mathbf{r} - \mathbf{r}_a| = (x_{\alpha} - x_{a \alpha})/|\mathbf{r} - \mathbf{r}_a|##

I'll admit it's a slightly confusing notation on the authors' part, using ##a## to enumerate the different sources and ##\alpha## to denote vector components 😜