# Lorentz transformation matrix

1. Sep 20, 2010

### Brian-san

1. The problem statement, all variables and given/known data:
Consider a two dimensional Minkowski space (1 spatial, 1 time dimension). What is the condition on a transformation matrix $\Lambda$, such that the inner product is preserved? Solve this condition in terms of the rapidity.

2. Relevant equations:
Rapidity Relations:
$$\beta=tanh\theta, \gamma=cosh\theta$$

Inner Product:
$$u^T \eta u$$

3. The attempt at a solution:
From the definition of inner product, to preserve inner product when $u'=\Lambda u$, we must have $\Lambda^T\eta\Lambda=\eta$

In matrix form:
$$\left[ \begin{array}{cc} \lambda_1 & \lambda_3 \\ \lambda_2 & \lambda_4 \end{array} \right]\left[ \begin{array}{cc} 1 & 0 \\ 0 & -1 \end{array} \right]\left[ \begin{array}{cc} \lambda_1 & \lambda_2 \\ \lambda_3 & \lambda_4 \end{array} \right]=\left[ \begin{array}{cc} 1 & 0 \\ 0 & -1 \end{array} \right]$$

This gives three relations:
$$\lambda_1^2-\lambda_3^2=1, \lambda_2^2-\lambda_4^2=-1, \lambda_1\lambda_2=\lambda_3\lambda_4$$

After substituting and solving the equations, letting $\lambda_1=\lambda$, I get the final form of the matrix as:
$$\Lambda=\left[ \begin{array}{cc} \lambda & \pm\sqrt{\lambda^2-1} \\ \pm\sqrt{\lambda^2-1} & \lambda \end{array} \right]$$

The two matrices are inverses of each other which can be shown easily. Since the Lorentz transformations are like rotations that mix space and time dimensions, I know the final result in terms of rapidity should be:
$$\Lambda=\left[ \begin{array}{cc} cosh\theta & \pm sinh\theta \\ \pm sinh\theta & cosh\theta \end{array} \right]$$

However, I'm not sure how to get the final step I need, by showing $\lambda=cosh\theta$. All I can say for sure is based on how the transformations behave at v=0 (returns identity matrix), and v=c (rapidity is infinite), is that λ(0)=1 and the function is strictly increasing to infinity. Obviously hyperbolic cosine fits that description, but so do a lot of other functions. So, I'm not sure what specifically will let me get the function I need.

Thanks.

2. Sep 22, 2010

### Ben Niehoff

What you have so far is correct. You have shown that you can write

$$\Lambda = \begin{pmatrix} \cosh \rho & \pm \sinh \rho \\ \pm \sinh \rho & \cosh \rho \end{pmatrix}$$

for some quantity $\rho$. What is left is to show that $\rho$ is, in fact, the rapidity. To accomplish that, you should take a particle at rest and then boost it to some velocity $\beta = v/c$; then show how $\rho$ is related to $\beta$.

I.e., you should know independently what the velocity 4-vector (2-vector in this case) should look like after a boost from rest, so use that to find how to interpret the parameter $\rho$.