- Why does my subscript disappear when I paste from Word into Google Sheets?
- Because in Word the subscript isn't part of the text — it's a formatting flag stored beside it, the same way bold and italic are. When you copy, the flag travels only as long as the destination understands Word's formatting. A spreadsheet cell, a chat box, a product listing field, and most web forms store plain characters and throw the formatting away, so H₂O lands as H2O and your formula is quietly wrong. Unicode subscript works the opposite way: the little 2 is its own character, U+2082, as much a character as the letter H. There's no formatting to lose, so it survives every paste that accepts text at all.
- Why can't I subscript the letter b?
- Because it doesn't exist. Unicode assigns a code point to every character it supports, and subscript letters were added piecemeal over decades for phonetics and mathematics rather than as a complete alphabet. Seventeen lowercase letters got one — a, e, h, i, j, k, l, m, n, o, p, r, s, t, u, v and x. The other nine — b, c, d, f, g, q, w, y and z — never did. No tool can generate them, here or anywhere, because there is no character to generate. Any site that appears to do it is showing you an image, a font trick that breaks on paste, or a lookalike character from another script. This tool tells you by name which of your letters came out full size instead of pretending it worked.
- Are there uppercase subscript letters?
- No, not one. Unicode has no uppercase subscript at all, which is awkward for chemistry because element symbols are capitalized. That's exactly why the Formula row exists and comes first: it subscripts only the digits, so H2O becomes H₂O with the H at full size, which is what a chemical formula is supposed to look like anyway. The Everything row handles capitals by falling back to the lowercase subscript glyph — H comes out as ₕ — which is the right answer when you want the whole string small, and the wrong answer for a formula. Pick the row that matches the job.
- What's the difference between the Formula row and the Everything row?
- Formula converts digits and nothing else. Every letter, space, sign and punctuation mark comes out byte-identical to what you typed. That makes it lossless and predictable: C6H12O6 becomes C₆H₁₂O₆, and Fe3+ becomes Fe₃+ with the plus sign untouched, because in chemistry that plus is a charge and subscripting it would change the meaning. Everything converts all of it — digits, the signs + − = ( ), and the 17 covered letters — which is what you want for math indices like x₁ and x₂, or when you just want small text. Formula is first because it's the row most people arrive needing.
- Where can I paste subscript characters?
- Google Sheets and Excel cells, Slack and Discord messages, WhatsApp and Telegram, YouTube titles and descriptions, product listings on eBay and Etsy, Instagram bios and captions, exam answer fields, Notion, and most web forms. It won't work in the handful of fields that accept only plain ASCII — usernames on some platforms, email addresses, and certain database inputs that strip anything outside a-z0-9. If a field rejects it, that's the field's rule, not a broken copy. Try the Formula row there, since it leaves everything except the digits alone.
- Will a screen reader or a search box handle subscript text?
- Handle, yes. Handle well, not always. Screen readers such as VoiceOver and NVDA typically announce each subscript character by its full Unicode name — "subscript small letter e" — so a subscripted sentence becomes exhausting to listen to. Search is the other cost: to a search index, H₂O and H2O are different strings, so a subscripted product name usually won't be found by someone typing the plain version. A formula, a chemical name, or a single index is a fair trade. A whole paragraph isn't. Keep your headline, your brand name, and anything you want found in plain characters.
- Can I paste text that's already subscripted?
- Yes, and nothing bad happens. None of the subscript characters this tool produces is itself something it converts, so running H₂O back through gives you H₂O — the same single U+2082, never doubled up, never expanded back to a full-size 2. That means you can paste a formula you found somewhere, check it, and copy it again without degrading it. It also means the Formula row is safe to apply repeatedly, which matters if you're piping text through the tool as part of a larger workflow.
- Does it handle emoji, other alphabets, and multi-line text?
- Emoji pass through untouched, including the complicated ones — flag pairs, skin-tone modifiers, and ZWJ family sequences come out exactly as they went in, because the text is split into grapheme clusters rather than raw code units. That's the bug behind most of the mangled emoji you'll see on other subscript sites. Line breaks and tabs are preserved, so each line converts on its own. Cyrillic, Devanagari, and Arabic have no subscript characters in Unicode and pass through unchanged; Greek has exactly five (β γ ρ φ χ), and those convert. Whatever can't convert gets named under the Everything row.
- Does anything I type get sent anywhere?
- No. The conversion is a lookup table running in your browser, in JavaScript, on the text in the box. Nothing is uploaded, stored, or logged, and the page keeps working with the tab offline. That's worth saying plainly for a tool like this, because people paste half-finished messages, unpublished product names, and lab notes into it without thinking twice.
