When debugging issues like this, experience says "round" numbers in a delayed request (like 120s for two minutes, or 300s for five) are indicative of hitting a timeout somewhere in the hosting stack.
This might for example be -
- a reverse DNS lookup that is failing (server lacks working DNS)
- an HTTP or SMTP request triggered within CiviCRM or custom code which fails after some time (eg: email is timing out, server can't connect to your analytics engine, POSTs make connection to a spam analytics tool which isn't responding, ...)
- "things like that ..."
Something is wrong here; that's not how CiviCRM works out of the box (try it on a demo site). But from the details here I can't tell you what's wrong.
Making specific guesses risks leading you astray even as it is likely to be of assistance. That said: You might want to check if this bug affects all situations where CiviCRM sends an email, or where CiviCRM processes a form submission.
This may require some debugging expertise I think - check out civicrm.org/providers if you don't have an existing team to lean on for this.
Do report back on the solution you find!