Unicode character <U+009C> (U+009C)

[Wolfgang Engelmann]

If I run the attached lyx file,I get an error:
Missing character: There is no œ (U+009C) in font 
[LibertinusSerif-Regular.otf]

show output anyway gives the attached pdf:

nuclear ancient Ca2+
(Na+ /Ca2+ exchanger)
Ca 2+ homeostasis
�Ca2+ dependent kinase

The exported luatex tex file was tested with this bash line (oneline!, 
thanks to people from the Linux User Group Tuebingen)

grep -P "[\xc2\x9c]" A-07-C.tex | grep --color=auto -P 
".{0,2}[\xc2\x9c].{0,2}" > A.txt

and tells

%%Package inputenc Error: Unicode character œ (U+009C)
œ$\mathrm{Ca^{2+}}$ dependent kinase

Could someone tell me, why in the first and the second line of the pdf 
file the Ca^2+ is shown correctly, whereas it is only shown correctly if 
I do not (3rd line) put it in mathrm (as done in the 4th line)

I also wonder whether this square/cross in front of the 4th line can be 
searched for under lyx, since it seems to mark extended regular expressions.

Wolfgang
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ATEST.lyx
Type: application/x-lyx
Size: 7493 bytes
Desc: not available
URL: <http://lists.lyx.org/pipermail/lyx-users/attachments/20220107/6a5547b8/attachment-0001.lyx>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: ATEST.pdf
Type: application/pdf
Size: 10888 bytes
Desc: not available
URL: <http://lists.lyx.org/pipermail/lyx-users/attachments/20220107/6a5547b8/attachment-0001.pdf>
-------------- next part --------------
%%Package inputenc Error: Unicode character œ (U+009C)
œ$\mathrm{Ca^{2+}}$ dependent kinase