PCL5 printers are still everywhere, even on the production side (sometimes an option you need to buy though). PCL6 has a compiled stream so you can't just add in a code. As for Postscript, you might be able to but it's source code for an application that will be executed on the printer, as such structure and where calls are inserted is sometimes important.
This is one of the benefits of PCL5, that you can simply insert commands with ease at anytime, at any place.
With Postscript, you might be able to; however, it might be inserted as part of the 'text to print' and simply show up on the page. Some devices have their own codes for duplex so you'll want to 'print to file' to confirm your device usage but it might look something like this: "<< /Duplex true >> setpagedevice"
I had to do something similar in Postscript in the past and what I did was enter some unique text on each page that needed a specialized command inserted. I then printed through Redmon using Redrun to execute a command line tool I wrote that parsed the string and inserted the command at the proper location. Redmon is a little old but there are similar tools our there including a simple print to file and execute the tool manually after. Doing this allowed me to control where in the file I inserted the Postscript commands.
PCL5 will still be supported for sometime due to legacy requirements. Countless clients of ours have embedded PCL5 in their mid-range applications printing large volumes. My recommendation...stick with PCL5 and simply find devices that support it. On the production side its usually done in a separate or even 3rd party RIP anyways, the bulk of which should support PCL on the input.
Another option might be to insert the same print command you are using now and write a macro that inserted blank pages within that section to compensate for your want to print simplex. The paper would still flip inside the printer slowing your performance but could produce the same result.