What is the law of reflection in special relativity?

Join the discussion
Registration is free. Start your own thread to ask a follow-up.
1 reply · 2K views
spaceofwaste
Messages
3
Reaction score
0
Hi,

I'm working through an exercise in Sean Carrol's spacetime and geometry book. The question asks you to consider an inertial frame S with coordinates [tex]x^\mu=(t,x,y,z)[/tex] and a frame S' with primed coordinates. Which is related to S by a boost v in the y direction. Imagine a wall( or mirror) lying along the line x'=-y'. From the point of view of S, what is the relationship between the angle of incidence (assuming ball travels in x-y plane only) and angle of reflection? Also what is velocity before and after?

Unfortunately there are not even one word answers let alone solutions in this book, so I don't even know if what I have is correct.

The way I proceeded, was firstly to consider what angle the mirror would appear at in S. Since the y-direction will be Lorentz contracted, but the x -direction will remain unchanged, the mirror should appear to be at a greater angle than 45 to someone in the S frame. This angle would precisely be:
[tex]\theta_M=arctan(\gamma \frac{\Delta x'}{\Delta y'})=arctan(\gamma)[/tex], the last equality following from the angle being 45 degrees in the primed frame meaning [tex]\Delta x'=\Delta y'[/tex]

OK so that is my step 1. Now we need to work out how the incident velocity looks in S.

e.g. [tex]u_x(init)=\frac{\Delta x}{\Delta y}=\frac{\Delta x'}{\gamma (\Delta t'+v\Delta y')}= \frac{u_x'(init)}{\gamma(1+vu_y'(init))}[/tex] Similarly, [tex]u_y(init)=\frac{u_y'(init)+v}{(1+vu_y'(init))}[/tex].

From this we can work out the angle of the incident ray (wrt -x-axis) in the S frame, this is therefore [tex]\alpha=\frac{u_y(init)}{u_x(init)}[/tex] (which you can sub into from above). I work out then the actual incident angle, [tex]\theta_I=\alpha+\theta_M-90[/tex]. (not 100% sure if this is correct)

Does this look like I'm going down the right path?
 
Physics news on Phys.org
I come out with this monster finally anyway:

[tex]\theta_r-\theta_i=-2tan(\gamma)+arctan(\frac{u_(y')(i)+vu_(x')(i)u_(y')(i)}{\gamma(u_(x')(i)+v+v[u_(x')(i)]^2+v^2u_(x')(i)})+arctan(\frac{u_(x')(i)+vu_(x')(i)u_(y')(i)}{\gamma(u_(y')(i)+v+v[u_(y')(i)]^2+v^2u_(y')(i)})[/tex]

where the [tex]u_(y')(i)[/tex] etc, represents the component of the incident velocity in the S', wrt to y' axis. [tex]\gamma[/tex] is just the Lorentz factor. v is the boost between S and S'.

Not sure if anyone knows what the expression shouled be for such a shift, and if this looks on the right track?