Set Theory Symbols
Ninety set-theory characters, each with the codepoint and official Unicode name it actually carries. Copy the one you need; the notes below cover the pairs that look identical and are not.
Tap any tile to copy it. Nothing is sent anywhere — the copy happens in your browser.
The characters that carry ordinary set notation
Eight of the hundred and three do most of the work: ∈ U+2208 ELEMENT OF, ∉ U+2209 NOT AN ELEMENT OF, ∪ U+222A UNION, ∩ U+2229 INTERSECTION, ⊂ U+2282 SUBSET OF, ⊆ U+2286 SUBSET OF OR EQUAL TO, ∖ U+2216 SET MINUS and ∅ U+2205 EMPTY SET. Most of the rest are variant forms — dotted, barred, doubled, closed — and none carries a meaning outside the document defining it.
⊂ does not tell you whether the inclusion is proper
⊂ U+2282 SUBSET OF is silent on whether the two sets are allowed to be equal. ⊆ U+2286 SUBSET OF OR EQUAL TO allows it explicitly, and ⊊ U+228A SUBSET OF WITH NOT EQUAL TO forbids it explicitly. If the distinction matters, one of the explicit characters carries it instead of a sentence of prose.
The empty set has two convincing impostors
∅ U+2205 EMPTY SET is a maths symbol, general category Sm. U+00D8 LATIN CAPITAL LETTER O WITH STROKE is a letter, category Lu, with a lowercase mapping and a job in Danish and Norwegian. U+2300 DIAMETER SIGN is a technical symbol, category So. All three are drawn as a slashed oval, none normalises into any other, and a search for one returns none of the others. Only the first is in the grid above; the other two are named so you can identify them when they arrive pasted in from elsewhere.
Membership runs both ways
∈ U+2208 ELEMENT OF puts the element on the left. ∋ U+220B CONTAINS AS MEMBER turns it round and puts the set there instead. ∍ U+220D SMALL CONTAINS AS MEMBER is the small form, and it is the character Unicode provides for “such that” in set-builder notation. A colon or a vertical bar does that job too, and both are ASCII, which is the argument for using them wherever the text has to survive a form field.
Width, when the column has to line up
∈, ∩, ∪, ⊂, ⊆ and ⊇ have East Asian width Ambiguous: their cell count is not a property of the character but of the terminal, which may give them one column or two. The variant operators from U+2A40 upward are Neutral and always take one. Mix the two families in a monospace table and alignment depends on a setting at the far end.
What a field does to them
None of these characters is ASCII or Latin-1, so a column, an export or a mail header that is genuinely latin-1 cannot hold any of them — that is the failure to plan for, not a font one. Normalisation is mostly not a risk: NFKC leaves ∈, ∪, ∩, ⊆ and ∅ exactly as they are.
ℵ is the exception to everything above
ℵ U+2135 ALEF SYMBOL, the cardinality sign for aleph-null, is the one character here that is not a symbol at all: category Lo, a letter. It is also the only one NFKC rewrites — into U+05D0 HEBREW LETTER ALEF, in a different block. The two look alike, so the substitution is invisible, and not cosmetic: U+2135 has bidirectional class L and U+05D0 has class R. Normalise a formula containing an aleph and you have put a right-to-left character into it, which can reorder how the text after it displays. Keep it away from anything that applies NFKC.
The rest of the wing is indexed on the maths and science symbols list. Quantifiers and connectives are on logic symbols, and the operators of arithmetic and calculus are on math symbols.