Cube roots of unity
Solve over the real numbers and you get exactly one answer, . Solve it over the complex numbers and two more solutions appear out of nowhere — a pair of complex numbers that are just as entitled to be called "cube roots of 1" as the number 1 itself. These extra roots are not a curiosity to memorise and forget: they are called and , they obey two identities short enough to fit on a fingernail, and those identities quietly demolish algebra problems that would otherwise take pages of expansion. This article derives the three cube roots of unity from , pins down the identities and that make them useful, and shows — with a fully worked example — how a monstrous-looking product collapses to a one-line answer once enters the picture. The scope stays exactly where the ISC Class 11 syllabus draws the line: cube roots only, no general -th roots, no De Moivre's theorem as a named tool.
The equation has three roots where , tied together by and , and these two identities let you simplify any expression built from powers or sums of the cube roots of unity without ever touching a decimal.
The core idea
Finding the roots. Factorise using the standard identity with :
where: the first factor gives (the familiar real cube root), and the second factor supplies the other two roots. Solving it by the quadratic formula,
So the three cube roots of unity are
Squaring the expression for does indeed produce the expression for — that check is what justifies calling the third root "" rather than inventing a new symbol.
Two identities that do all the work. Because satisfies (it is a root of that factor), and because is a cube root of 1 by construction,
where: the first says powers of repeat every three steps; the second says the three roots always cancel when added. Together they are the only two facts you ever substitute — every simplification problem in this topic reduces to spotting where one of them applies. It is worth checking once by direct substitution rather than trusting the factorisation alone: . The terms cancel exactly, leaving a purely real sum, which is itself a small confirmation that the identity is not a rounding artefact but an exact algebraic fact.
Geometric picture. Each root has modulus 1, so all three sit on the unit circle in the Argand plane, spaced apart: at , at , at . They form an equilateral triangle inscribed in the circle — three equally spaced unit vectors always cancel when summed, which is the geometric reason . This picture also explains multiplication by : multiplying any complex number by rotates it by about the origin without changing its distance from the origin, since itself has modulus 1 and argument . Multiplying by three times in a row therefore rotates a point by , back to where it started — a geometric restatement of .
Powers cycle with period 3. Since , any power collapses to , , or depending on . For instance , and .
Conjugates and polar form. Compare the explicit values: and differ only in the sign of the imaginary part, so , the complex conjugate of . This is exactly what the picture predicts — reflecting a point at across the real axis lands you at , which is the same location as . In polar form, and , confirming the conjugate relationship without any extra algebra. Knowing is a fast way to double-check a computed value of : just flip the sign on the imaginary part of .
The goal is to show that — a product that looks like it needs six terms of messy expansion — simplifies to the compact real expression , using nothing beyond and ; this identity is the reason is worth learning at all, since it is the engine behind the classic factorisation of .
Step 1. Expand the product term by term, treating like any algebraic symbol for now.
Step 2. Reduce every power of using (so ). This turns into , into , and into . Collecting the surviving terms by which pair of variables they multiply:
Step 3. Apply the second identity. Since , we know . Substitute this single fact into all three bracketed pairs at once:
Step 4. Read off the result and its use. — a real expression with no trace of left, even though was complex throughout. Multiplying this by the factor reproduces , which is why the -notation is a genuine shortcut and not just decoration: it factorises a three-variable cubic that would otherwise resist simple algebra.
Step 5. Sanity-check with . The left side becomes . Using and from the key identity, this is . The right side, . Both sides agree, confirming the general identity on a case simple enough to check by hand.
Worked example
Simplify , then evaluate .
Step 1. Reduce each exponent modulo 3, since . . For , write , so . For , write , so .
Step 2. Add the reduced terms. , using the key identity directly.
Step 3. For the second expression, first rewrite each bracket using and (both immediate consequences of ). , and .
Step 4. Multiply the two simplified brackets. , using in the last step. So and .
Competitive stretch
The same reduction trick handles arbitrarily large exponents without ever computing a decimal power: for any integer , depends only on , so — a fact that turns exam questions with huge exponents into a one-line remainder calculation. The same idea extends to sums: equals whenever is a multiple of (because every term collapses to ), and equals otherwise (because the three terms are just in some order, which always sum to zero). Recognising which case a question falls into — check whether the exponent is a multiple of — is often the entire solution.
Common traps
- Treating as "just some complex number" without fixing which root it is — by convention , the root in the upper half of the Argand plane, and is its conjugate.
- Reducing exponents by dividing instead of taking the remainder — needs , giving , not .
- Misremembering the identity as ; the sum of all three cube roots is always , never .
- Leaving unsimplified in the middle of a calculation instead of replacing it with immediately, which is usually the step that unlocks the rest of the problem.
- Forgetting that only holds for this specific pair — it is a consequence of having modulus 1 and the two roots being reflections in the real axis, not a general rule about squaring complex numbers.
Ask the tutor: If a problem with a huge exponent on or a tangled sum of has you stuck, ask the AI tutor to walk through reducing the exponent mod 3 or substituting step by step alongside your own working.
Why are there three cube roots of unity instead of just 1?
Because is a degree-3 polynomial equation, and over the complex numbers every degree- polynomial has exactly roots (counted with multiplicity), a fact you may see stated later as the Fundamental Theorem of Algebra. Over the reals only works, since has no real solutions (its discriminant is negative). But complex numbers do allow that discriminant, giving the two extra complex roots and . All three are equally valid answers to "what number, cubed, gives 1" — real numbers simply cannot supply the other two.
Do I need to remember the exact values of omega and omega squared, or just the identities?
Both, but for different jobs. Know the identities and cold — they are what you actually substitute in nearly every problem, and most exam questions can be solved without ever writing down a square root. Keep the explicit values and in reserve for the rarer question that asks you to verify an identity from scratch, as in the derivation of by direct substitution, or one that asks for the modulus and argument of .
Is omega squared the same as omega, squared numerically, or a separate root?
It is both at once, and that is exactly the point of the notation. literally means "square the number ," and when you carry out that squaring on you land precisely on the third root of . So the symbol is not a coincidence of naming — it is a genuine algebraic fact that makes the whole system self-consistent, and it is why powers of simply cycle through the same three roots forever: , , , again, and so on without ever producing a fourth distinct value.