3
$\begingroup$

On my last HW set, we were asked to show that the operator $$\hat T = \exp(-ic\hat p /\hbar)$$ acts as a translation operator ($\hat T^\dagger q\hat T=q+c)$. This was simple to show using commutators and other such things. The same thing can be said with a momentum operator.

Can I have a translation operator that will take some operator $$\hat A=(\hat p-p_0)(\hat q -q_0)$$ and translate this to eliminate both of the C numbers? I feel like this could be something useful for a multidimensional 'number' type operator that raises some form of state.

$\endgroup$
2
$\begingroup$

Yes: $$\hat A=(\hat p-p_0)(\hat q -q_0)= e^{-i(q_0\hat p -p_0 \hat q)/\hbar} ~\hat p \hat q ~e^{i(q_0\hat p -p_0 \hat q)/\hbar} .$$

In particular, define $$ \hat T ^\dagger \equiv e^{-i~q_0\hat p /\hbar}, $$ so that $$ \hat T ^\dagger \hat q \hat T =\hat q -q_0, \qquad \hat T ^\dagger \hat p \hat T=\hat p , $$ and similarly $$ \hat S \equiv e^{i~p_0\hat q /\hbar}, $$ so that $$ \hat S \hat p \hat S ^\dagger =\hat p -p_0, \qquad \hat S \hat q \hat S ^\dagger=\hat q . $$

Moreover, $$ \hat T \hat S = \hat S \hat T \exp (i(p_0q_0)/\hbar),\\ \hat T \hat S^\dagger = \hat S^\dagger \hat T \exp (-i(p_0q_0)/\hbar)= e^{i(q_0\hat p -p_0 \hat q -p_oq_0/2)/\hbar}. $$

Consequently, $$ \hat A=\hat T ^\dagger \hat q ~ \hat T \hat S ~ \hat p ~ \hat S ^\dagger= \hat T ^\dagger \hat S ~\hat q \hat p ~ \hat T \hat S ^\dagger e^{ip_0 q_0/\hbar}= \hat T ^\dagger \hat S ~\hat q \hat p ~ \hat S ^\dagger \hat T , $$ whence the compact form given above.

To be sure, this proof is overkill. The above expressions follows directly from the standard combinatorial lemma Ad$_{(e^X)}=e^{\operatorname{ad}_X}$ utilized in proving the CBH expansion--many call it the Hadamard lemma. So, for $X\equiv -i(q_0\hat p-p_0\hat q)/\hbar$, and $[X,\hat p]=-p_0$, $[X,\hat q]=-q_0$, $$ \operatorname{Ad}_{e^X} (\hat p \hat q)= \operatorname{Ad}_{e^X} ( \hat p) \operatorname{Ad}_{e^X}(\hat q) = e^{\operatorname{ad}_X} (\hat p) ~~e^{\operatorname{ad}_X} (\hat q), $$ just like your exercise. Can you see that the order of $\hat p$ and $\hat q$ acted on hardly matters in this operation? Their powers? An arbitrary function $f(\hat q, \hat p)$? To quote Feynman quoting Gibbon v1, ChIV, "The power of instruction is seldom of much efficacy except in those happy dispositions where it is almost superfluous.”

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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