Deriving del cross A in Electrodynamics

Click For Summary
SUMMARY

The discussion focuses on deriving the equation for the magnetic field (B-field) due to a moving charge as presented in Griffiths' "Introduction to Electrodynamics," specifically Chapter 10, equation 10.66. The participants explore the vector calculus operation of the curl (del cross A) applied to the vector potential A, which is defined as A = (v/c^2)V, where V = (1/4πe₀)(qc/(rc - r·v)). The final derived expression for the curl of A is confirmed to be × A = - (1/c)(q/4πe₀)(1/(u·r)³)r × [(c² - v²)v + (r·a)v + (r·u)a]. The discussion highlights the complexity of the algebra involved in this derivation.

PREREQUISITES
  • Understanding of vector calculus, specifically curl operations.
  • Familiarity with electromagnetic theory, particularly the concepts of electric and magnetic fields.
  • Knowledge of Griffiths' "Introduction to Electrodynamics," especially Chapter 10.
  • Proficiency in manipulating vector quantities and dot products.
NEXT STEPS
  • Study the derivation of the B-field from the vector potential A in Griffiths' "Introduction to Electrodynamics."
  • Learn about the product rule for curl operations in vector calculus.
  • Explore the implications of the derived equation in practical electromagnetic applications.
  • Investigate the relationship between the vector potential and the magnetic field in different reference frames.
USEFUL FOR

This discussion is beneficial for physics students, particularly those studying electrodynamics, researchers in electromagnetic theory, and educators seeking to clarify the derivation of the B-field from the vector potential.

Sparky_
Messages
227
Reaction score
5
Hello,

I am trying to derive the equation for the B-field due to a moving charge. ~ Griffiths Chapter 10, equation 10.66.

I have been trying to “do” the del cross A and simplify . Things get messy and I am uncertain on some of my vector operations.

In searching the internet I find statements like, from del cross A and a little algebra you get …

I have tried 3 different times and can get some of it and a good bit of mess.

Below is an attempt.

The final result is (what I want to end up with):

[tex]\nabla \times \vec{A} = - \frac{1}{c}\frac{q}{4\pi e_0}\frac{1}{(\vec{u}*\vec{r})^3}\vec{r} \times [(c^2 – v^2)\vec{v} + (\vec{r}\bullet \vec{a})\vec{v} + (\vec{r}\bullet \vec{u})\vec{a}][/tex]


Some background equations:

[tex]\vec{A} = \frac{\vec{v}}{c^2} V[/tex]

[tex]V = \frac{1}{4\pi e_0}\frac{qc}{(rc - \vec{r}\bullet \vec{v})}[/tex]

Using a product rule for a cross product:

(1) [tex]\nabla \times \vec{A} = \frac{1}{c^2}\nabla \times (V\vec{v}) - \vec{v} \times (\nabla V)[/tex] (1)



[tex]\nabla \times \vec{v} = ( -\vec{a} \times \nabla t_r)[/tex]

[tex]\nabla t_r = -\frac{\vec{r}}{rc - \vec{r} \bullet \vec{v}}[/tex]


[tex]\nabla V = \frac{1}{4\pi e_0}\frac{qc}{(rc - \vec{r}*\vec{v})^3} [(rc - \vec{r}\bullet \vec{v})\vec{v} - (c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r}][/tex]

Plug these into (1)


[tex]\nabla \times \vec{A} = \frac{1}{c^2}[\frac{1}{4\pi e_0}\frac{qc}{(rc - \vec{r}\bullet \vec{v})}](-\vec{a} \times \nabla t_r))-\vec{v}\times [\frac{1}{4\pi e_0}\frac{qc}{(rc - \vec{r}\bullet \vec{v})^3}[((rc - \vec{r}\bullet \vec{v})\vec{v}) - ((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})]][/tex]


Factor out a [tex]\frac{qc}{4\pi e_0}[/tex]


[tex]\nabla \times \vec{A} = \frac{1}{c^2}\frac{qc}{4\pi e_0} [\frac{1}{(rc - \vec{r}\bullet \vec{v})}](-\vec{a} \times \nabla t_r))-\vec{v}\times [\frac{1}{(rc - \vec{r}\bullet \vec{v})^3}((rc - \vec{r}\bullet \vec{v})\vec{v}) - ((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})][/tex]

Can I factor out a [tex]\frac{1}{(rc - \vec{r}\bullet \vec{v}})[/tex] ?

And end up with:

[tex]\nabla \times \vec{A} = \frac{1}{c^2}\frac{qc}{4\pi e_0}\frac{1}{(rc - \vec{r}\bullet \vec{v})} (-\vec{a} \times \nabla t_r))-\vec{v}\times [\frac{1}{(rc - \vec{r}\bullet \vec{v})^2}[((rc - \vec{r}\bullet \vec{v})\vec{v}) - ((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})]][/tex]


Using the equation for [tex]\nabla t_r[/tex]

[tex]\nabla \times \vec{A} = \frac{1}{c^2}\frac{qc}{4\pi e_0}\frac{1}{(rc - \vec{r}\bullet \vec{v})} [(-\vec{a} \times -\frac{\vec{r}}{(rc - \vec{r} \bullet \vec{v})})-\vec{v}\times [\frac{1}{(rc - \vec{r}\bullet \vec{v})^2}[((rc - \vec{r}\bullet \vec{v})\vec{v}) - ((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})]]][/tex]

Can I factor out another [tex]\frac{1}{(rc - \vec{r}\bullet \vec{v}})[/tex] ?

Yielding:

[tex]\nabla \times \vec{A} = [\frac{1}{c^2}\frac{qc}{4\pi e_0}\frac{1}{(rc - \vec{r}\bullet \vec{v})^2}] [(-\vec{a} \times -\vec{r}) -\vec{v}\times [\frac{1}{(rc - \vec{r}\bullet \vec{v})}[((rc - \vec{r}\bullet \vec{v})\vec{v}) - ((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})]]][/tex]

I see that I have a
[tex]\vec{v} \times \vec{v}[/tex] in the middle which results in zero.

[tex]\nabla \times \vec{A} = [\frac{1}{c^2}\frac{qc}{4\pi e_0}\frac{1}{(rc - \vec{r}\bullet \vec{v})^2}] [(-\vec{a} \times -\vec{r}) + \vec{v}\times [ \frac{((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})}{(rc - \vec{r}\bullet \vec{v})}]][/tex]

Introducing [tex]\vec{u} = c\hat{r} – \vec{v}[/tex]

[tex]\vec{r} \bullet \vec{u} = (rc - \vec{r} \bullet \vec{v})[/tex]


[tex][/tex]
Assuming I am correct at this point, here is where for me things start getting messy.
[tex]\nabla \times \vec{A} = [\frac{1}{c^2}\frac{qc}{4\pi e_0}\frac{1}{(\vec{r} \bullet \vec{u})^2}] [(-\vec{a} \times -\vec{r}) + \vec{v}\times [ \frac{((c^2-v^2 + \vec{r}\bullet \vec{a})\vec{r})}{(\vec{r} \bullet \vec{u})}]][/tex]

Frustration is showing up and I could use some help.

Bottom line is I am trying to derive and see how to get the del cross A result at the top. Searching the internet I find that equation and how it is used for the B field but not how it was derived.

Thanks
Sparky_
 
Physics news on Phys.org
It works better of you change the size of your brackets as you go out, and use \cdot instead of \bullet for a dot-product. vis:
[tex]\nabla \times \vec{A} = - \frac{1}{c}\frac{q}{4\pi e_0}\frac{1}{(\vec{u}*\vec{r})^3}\vec{r} \times \left [ (c^2 – v^2)\vec{v} + (\vec{r}\cdot \vec{a})\vec{v} + (\vec{r}\cdot \vec{u})\vec{a}\right ][/tex]
... I don't know what the asterisk is supposed to represent.

Some background equations:

[tex]\vec{A} = \frac{\vec{v}}{c^2} V[/tex]

[tex]V = \frac{1}{4\pi e_0}\frac{qc}{(rc - \vec{r}\cdot \vec{v})}[/tex]

... but you seem to be missing a ##\vec{u}##.

The product rule for curl is:
$$\nabla \times (\vec{u}\cdot\vec{v}) = (\nabla \cdot \vec{u}) \times \vec{v} + \vec{u}\cdot (\nabla \times \vec{v})$$
... so, on the whole, I'm not sure I follow what you are doing here.

In general, when a text says "... and then, after some algebra, you get..." the "some algebra" tends to be pretty ugly.
 
I went through the exercise (on pages 435-437, Griffiths) of taking the gradient of V and the time derivative of A and arriving at E. The text then states that after del cross A you arrive at:

[tex]\nabla \times \vec{A} = - \frac{1}{c}\frac{q}{4\pi e_0}\frac{1}{(\vec{u}*\vec{r})^3}\vec{r} \times \left[(c^2 – v^2)\vec{v} + (\vec{r}\bullet \vec{a})\vec{v} + (\vec{r}\bullet \vec{u})\vec{a}\right][/tex]

I want to turn the crank and arrive at the above equation.

In my scratch work I did not introduce or bring in u at the start-

[tex]\vec{u} = c\hat{r} – \vec{v}[/tex]

u sort of cleans up the [tex](rc - \vec{r}\cdot \vec{v})[/tex]

I brought in u after a step or two into my work.


also – I missed the asterisk – that also is a dot product.

Thanks
-Sparky_
 

Similar threads

  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 0 ·
Replies
0
Views
1K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 14 ·
Replies
14
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
4
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K