2
$\begingroup$

When going through the calculation of the conformal algebra, I find that

$$[J_{ab},J_{cd}]=i(g_{ad}J_{bc}+g_{bc}J_{ad}-g_{ac}J_{bd}-g_{bd}J_{ac})$$

where $J_{xy}$ are the generators of the conformal field theory, which consists of translations, dilations, rotations, and the special conformal group.

Many sources say that the above is equivalent to an $SO(d+1,n+1)$ special pseudo-orthogonal group, where $d$ is the number of space dimensions and $n$ is the number of time dimensions. However, I also know that the generators of the Lorentz group satisfies a similar algebra:

\begin{align} [\mathcal{M}^{\mu \nu},\,M^{\rho \sigma}]=i(g_{\nu \rho}\mathcal{M}^{\mu \sigma}-g_{\nu \sigma}\mathcal{M}^{\mu \rho}-g_{\mu \rho}\mathcal{M}^{\nu \sigma}+g_{\nu \rho}\mathcal{M}^{\mu \sigma}) \end{align}

which is $SO(3,1)$ My question is why do we say the Lorentz group is $SO(3,1)$ while the conformal group is $SO(d+1,n+1)$? Wouldn't the metric be the same in both--i.e., be a diagonal matrix with three +1's and one -1? If we consider a general metric in some arbitrary dimension, wouldn't the Lorentz group also be $SO(d+1,n+1)$?

$\endgroup$

1 Answer 1

3
$\begingroup$

You don't have all algebra to the conformal-group.

The conformal-group have rotations and boost $M_{\nu\mu}$, the Lorentz group ($SO(d,n)$) $$ [M_{\nu\mu},M_{\rho\gamma}]=4\eta_{[\mu[\rho}M_{\sigma]\mu]}. \text{ (This is yours equation for $J_{ab}\leftrightarrow \mathcal{M}_{\nu\mu}$)} $$ Also have translations $P^\mu$, the Poincaré Group $$ [P_\mu, M_{\nu\rho}]=2\eta_{\,\mu[ \nu}P_{\rho]}. $$ But, the conformal group also have, the special conformal transformations $K^\mu$ $$ [K_\mu, M_{\nu\rho}]=2\eta_{\,\mu[ \nu}K_{\rho]};\quad [P_\mu,K_\nu]=2(\eta_{\mu\nu}D+M_{\mu\nu}), $$ and the dilations $$ [D,P_\mu]=P_\mu;\quad [D,K_\mu]=-K_\mu. $$ And this is,of course, the algebra of $SO(D,2)$ whit $D=d+n$ and $n=1$ (I don't remember if this algebra is valid to $n>1$).

Let $\hat\mu=0,\cdots,D,D+1$, the generators can be codified in a compact way by $$ M^{\hat \mu\hat \nu}= \begin{pmatrix} M^{\mu\nu}&\frac{1}{2}(P^\mu-K^\mu)&\frac{1}{2}(P^\mu +K^\mu)\\ -\frac{1}{2}(P^\mu -K^\mu)& 0 &-D\\ -\frac{1}{2}(P^\mu +K^\mu)& D &0 \end{pmatrix}, $$ which the indices are raised by $\hat \eta=\text{diag}(-1,1,1,\cdots,1,-1)$.

Therefore the Lorentz Group only have rotations and boost, and it is a subgroup of the conformal-Group.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.