well, that character is (depending on the encoding etc, but most likely) a multi-byte character (see https://en.wikipedia.org/wiki/Degree_symbol and https://www.utf8-chartable.de/ etc).
in utf8, I think your specific character is 0xc2b0 (see c2 b0 in that utf8-chartable). That means that hashcat needs to crack it as 2 separate characters/bytes.
you can use them directly in a mask with --hex-charset e.g. hashcat -m 0 -a 3 -w 3 -O hashes.txt ?d?d?d?dc2b0?d?d?d
but this means you need to know the exact position of both bytes (the first is 0xc2 , followed by the second one 0xb0).
Alternatively, you could also use hcmask files and have several mask within the hashcat mask file with different positions of both 0xc2 followed by 0xb0
Of course, there is also the possibility that a slightly different char and different encoding is used, but I guess 0xc2b0 and utf-8 encoding is the most educated and best guess. I hope this explains what the challenge and "problem" with multi-byte characters could be when using mask attacks etc
btw: even if you use --custom-charset1 to --custom-charset4 (or short -1 to -4), you need to add both/two chars to your mastk, e.g. --hex-charset --custom-charset3 c2b0 hash.txt ?d?d?d?3?3?d?d?d
(note that there are 2 positions, twice ?3 , in that specific mask, because the char itself is a multi-byte character). I hope this explains why your attempt with a custom charset didn't work with a single byte usage within the mask
in utf8, I think your specific character is 0xc2b0 (see c2 b0 in that utf8-chartable). That means that hashcat needs to crack it as 2 separate characters/bytes.
you can use them directly in a mask with --hex-charset e.g. hashcat -m 0 -a 3 -w 3 -O hashes.txt ?d?d?d?dc2b0?d?d?d
but this means you need to know the exact position of both bytes (the first is 0xc2 , followed by the second one 0xb0).
Alternatively, you could also use hcmask files and have several mask within the hashcat mask file with different positions of both 0xc2 followed by 0xb0
Of course, there is also the possibility that a slightly different char and different encoding is used, but I guess 0xc2b0 and utf-8 encoding is the most educated and best guess. I hope this explains what the challenge and "problem" with multi-byte characters could be when using mask attacks etc
btw: even if you use --custom-charset1 to --custom-charset4 (or short -1 to -4), you need to add both/two chars to your mastk, e.g. --hex-charset --custom-charset3 c2b0 hash.txt ?d?d?d?3?3?d?d?d
(note that there are 2 positions, twice ?3 , in that specific mask, because the char itself is a multi-byte character). I hope this explains why your attempt with a custom charset didn't work with a single byte usage within the mask