Replacing non-ASCII characters on the eText extract (does not work for hyphens)

Summary:

In the etext template, we used the replace character functionality to remove letters like Á and replace them with A.  but it’s not working with —

Content (required):

Example: '12345 PAG—HRIS' should be '12345 PAG-HRIS'

In the .rtf word doc, (see screenshot where green question mark is next to what we think might be the right character) we tried adding the hyphen, and tried adding a blank space, but it does not work and causes it to not pass validation. Any ideas to try?

Version (include the version you are using, if applicable):

22B

Code Snippet (add any code snippets that support your topic, if applicable):