Katakana can represent many foreign sounds that do not exist in native Japanese words, either by using special characters such as ヴ or by using special katakana combinations such as フュ. They are collectively called "extended katakana", although few native speakers know this term. Thanks to them, young Japanese people can learn to pronounce foreign words more or less closely to their original pronunciation. Although they may not be perfect, フューチャー is a lot better than ヒューチャー and レモンティー is a lot better than レモンチー. Few Japanese people mistake 'Venus' as 'Benus' because it is commonly spelt as ヴィーナス in katakana.
However, I am not aware of any previous attempts to distinguish, for example, English 'read' and 'lead' using modified/extended katakana. Come to think of it, this seems strange to me. Japanese people have invented special kana to express Okinawan sounds, Ainu sounds and bidakuon, so I think there must have been similar attempts also for the English sound pairs that are known to be difficult to Japanese speakers.
- Historically, were there any attempts to distinguish "L" and "R", "S" and "TH", etc., using katakana? How commonly were they used? It could be special katakana like "ラ with (han)dakuten" or "small ラ", and it could be special combinations like "ゥラ".
- If there were such attempts in the past, why were they all unsuccessful despite obvious advantages? I doubt "because they looked unnatural" is a good reason here, because ティ and ヴ should have looked equally unreasonable and unnatural at first.
I know some recent English dictionaries targeted at Japanese middle school students use hiragana らりるれろ/さしすせそ to denote the sound of R and TH. I also found this 2ch thread, which was not really helpful.