Algebra

Exponent and Radical Rules

Zero, negative, and fractional exponents are not extra definitions to memorise. They are the only values that keep one rule from breaking.

By the Gauth editorial teamHow this content is produced and checkedUpdated 10 September 2026

The idea

For a positive whole number $n$ , the notation $a^{n}$ is shorthand for $n$ copies of $a$ multiplied together. That definition covers $a^{3}$ and $a^{17}$ and nothing else. It says nothing about $a^{0}$ , because zero copies of something is not an obvious idea, and it says nothing at all about $a^{- 2}$ or $a^{1/2}$ , because you cannot multiply a thing by itself minus two times or half a time.

So those cases have to be defined. The question is how, and the answer is what makes this topic coherent: they are defined by whichever value keeps the existing rules working. The counting definition immediately gives $a^{m} \cdot a^{n} = a^{m + n}$ , since laying $m$ copies beside $n$ copies produces $m + n$ copies. Insist that this keeps holding for every exponent, and each extension becomes forced.

That is the difference between memorising nine rules and understanding one. You do not have to remember whether $a^{0}$ is 0 or 1 — you can rebuild it in five seconds from the quotient rule. The derivations below take two lines each and they are the reason this page is short.

When you need it

Exponent manipulation is rarely the question; it is the step inside the question that decides whether you finish:

Simplifying anything with a variable in a denominator. Rewriting $\frac{1}{x ^{3}}$ as $x^{- 3}$ turns a fraction into a factor you can combine.
Differentiating or integrating a root. The power rule needs an exponent, so $x$ has to become $x^{1/2}$ before you can touch it.
Scientific notation and unit conversion, where the whole calculation is adding and subtracting powers of ten.
Solving exponential equations, where the first move is usually to write both sides with the same base and equate the exponents.
Factoring an expression with mixed powers. Pulling $x^{2}$ out of $x^{5} + x^{2}$ is the quotient rule running backwards.

The recognition cue for a mistake is just as useful: any time you find yourself applying an exponent across a plus sign, stop. That operation does not exist, and it is the source of a large fraction of lost marks in expanding and factoring problems.

The method

Take the counting rule as given and watch the other definitions fall out. First, $a^{0}$ . Any non-zero number divided by itself is 1, and the quotient rule gives a second name for the same expression:

1 = \frac{a ^{n}}{a ^{n}} = a^{n - n} = a^{0}

So $a^{0} = 1$ is not a convention someone picked; it is the only value consistent with the rule. It also shows why $0^{0}$ is left undefined — the argument divides by $a^{n}$ , which is illegal at $a = 0$ .

Next, negative exponents. Multiply $a^{n}$ by $a^{- n}$ and apply the product rule:

a^{n} \cdot a^{- n} = a^{n + (- n)} = a^{0} = 1

A number that multiplies $a^{n}$ to give 1 is the reciprocal of $a^{n}$ . Hence $a^{- n} = 1/ a^{n}$ . Notice that nothing in this makes the value negative: $2^{- 3}$ is $\frac{1}{8}$ , comfortably positive.

Finally, fractional exponents. Raise $a^{1/ n}$ to the power $n$ and use the power rule:

(a^{1/ n})^{n} = a^{n / n} = a^{1} = a

A number whose $n$ th power is $a$ is exactly what $n a$ means. Roots and fractional exponents are the same object in two notations, which is why every radical rule is an exponent rule in disguise. Combining the two gives $a^{m / n} = (n a)^{m}$ , and taking the root first keeps the arithmetic small.

Rationalising a denominator is the one radical technique that is not simply a rule applied. A root in the denominator is not wrong, but answers are conventionally written without one so that two people can tell whether their results agree. For a single radical, multiply top and bottom by that radical, since $a \cdot a = a$ :

\frac{1}{2} = \frac{1}{2} \cdot \frac{2}{2} = \frac{2}{2}

That is where the familiar $2 /2$ in trigonometry comes from. When the denominator is a sum, one radical is not enough — multiply by the conjugate instead, because the cross terms cancel and the roots disappear:

(a + b) (a - b) = a - b^{2}

So $\frac{3}{5 - 1}$ becomes $\frac{3 ( 5 + 1 )}{5 - 1} = \frac{3 ( 5 + 1 )}{4}$ . Check numerically: $5 \approx 2.236$ , so the original is $3/1.236 \approx 2.427$ , and the rationalised form gives $3 (3.236) /4 \approx 2.427$ .

Every rule, with its derivation

Tap any formula with a derivation to see where it comes from.

The three rules everything else comes from

Forced consequences, not new rules

Fractional exponents and radicals

Traps that look like rules

To simplify an expression, work in this order:

Distribute outer powers first. An exponent on a bracket applies to every factor inside, coefficient included. $(3 x^{2})^{2}$ is $9 x^{4}$ , and forgetting to square the 3 is the most frequent slip in the whole topic.
Move negative exponents across the bar. A factor with a negative exponent moves to the other side of the fraction and the sign flips. This applies to factors only — a negative exponent inside a sum cannot be moved.
Combine like bases. Add exponents when multiplying, subtract when dividing. The bases must match exactly; $x^{2} \cdot y^{3}$ combines into nothing.
Handle the numbers separately. Simplify the coefficient as its own small problem so an arithmetic error cannot corrupt the algebra.
Convert fractional exponents last. Take the root before the power, and only when the root is a whole number.
Test with $x = 2$ . Substituting a small number into the original and into your answer is a complete check, and it takes fifteen seconds.

Solve a algebra problem now

First solution free

Three worked examples

The first is a straightforward power-of-a-product simplification. The second stacks a negative exponent onto a fractional one. The third is the messy case: negative exponents in three places, two of them inside brackets that are themselves raised to a negative power.

Worked example

\frac{( 3 x ^{2} y ^{3} ) ^{2} ( 2 x y ) ^{3}}{6 x ^{4} y ^{5}}

1
Distribute the outer powers before anything else. The exponent applies to the coefficient too, so the 3 becomes 9 and the 2 becomes 8.
$(3 x^{2} y^{3})^{2} = 9 x^{4} y^{6}, (2 x y)^{3} = 8 x^{3} y^{3}$
2
Multiply the two numerators. Coefficients multiply; exponents on matching bases add.
$9 x^{4} y^{6} \cdot 8 x^{3} y^{3} = 72 x^{4 + 3} y^{6 + 3} = 72 x^{7} y^{9}$
3
Now divide. Handle the coefficient on its own, then subtract exponents base by base.
$\frac{72}{6} = 12, x^{7 - 4} = x^{3}, y^{9 - 5} = y^{4}$
4
Assemble the result.
$12 x^{3} y^{4}$
5
Check at x = y = 1: the original is (3²)(2³)/6 = 9 × 8 / 6 = 12, and the answer is also 12.
$\frac{9 \cdot 8}{6} = 12$

Answer

12 x^{3} y^{4}

Worked example

(\frac{8}{27})^{- 2/3}

1
Deal with the negative sign first. A negative exponent means reciprocal, so invert the fraction and make the exponent positive.
$(\frac{8}{27})^{- 2/3} = (\frac{27}{8})^{2/3}$
2
Split the exponent using the power rule. Take the root before the square so the numbers stay small.
$(\frac{27}{8})^{2/3} = (3 \frac{27}{8})^{2}$
3
The cube root applies to numerator and denominator separately, and both are perfect cubes.
$327 = 3, 38 = 2 ⟹ \frac{3}{2}$
4
Square the result.
$(\frac{3}{2})^{2} = \frac{9}{4}$
5
Check by the other route: squaring first gives 729/64, whose cube root is 9/4 because 9³ = 729 and 4³ = 64. Same answer, worse arithmetic.
$3 \frac{729}{64} = \frac{9}{4}$

Answer

\frac{9}{4}

Worked example

(2 x^{- 3} y^{2})^{- 2} \cdot (4 x^{5} y^{- 1}) \div (8 x^{- 2} y^{3})

1
Start with the bracket carrying the outer negative power. Every factor inside is affected, including the 2, and each inner exponent is multiplied by −2.
$(2 x^{- 3} y^{2})^{- 2} = 2^{- 2} x^{6} y^{- 4}$
2
Note the sign arithmetic: (−3)(−2) = +6 and (2)(−2) = −4. Write the result as a fraction to keep track.
$2^{- 2} x^{6} y^{- 4} = \frac{x ^{6}}{4 y ^{4}}$
3
Multiply by the second factor. The 4 in the numerator cancels the 4 in the denominator exactly.
$\frac{x ^{6}}{4 y ^{4}} \cdot \frac{4 x ^{5}}{y} = \frac{x ^{11}}{y ^{5}}$
4
Dividing by an expression means multiplying by its reciprocal. Invert the third bracket, which flips the sign of every exponent in it.
$\div (8 x^{- 2} y^{3}) = \times \frac{x ^{2}}{8 y ^{3}}$
5
Combine. The x exponents add to 13 and the y exponents in the denominator add to 8.
$\frac{x ^{11}}{y ^{5}} \cdot \frac{x ^{2}}{8 y ^{3}} = \frac{x ^{13}}{8 y ^{8}}$
6
Check at x = y = 1. The original is (2)^-2 × 4 ÷ 8 = 0.25 × 4 ÷ 8 = 0.125, and the answer gives 1/8. They agree.
$\frac{1}{4} \cdot 4 \div 8 = \frac{1}{8}$

Answer

\frac{x ^{13}}{8 y ^{8}}

Where people go wrong

Leaving the coefficient unpowered

$(3 x^{2})^{2}$ is $9 x^{4}$ , not $3 x^{4}$ . The exponent applies to every factor inside the bracket, and a number is a factor. The same slip in reverse produces $9 x^{2} = 3∣ x ∣$ written as $9 x$ . Substituting $x = 2$ exposes it immediately: $(3 \cdot 4)^{2} = 144$ , and $9 \cdot 16 = 144$ , while $3 \cdot 16 = 48$ .

Treating a negative exponent as a negative number

$2^{- 3}$ is $\frac{1}{8}$ , not $- 8$ and not $- \frac{1}{8}$ . The sign of the exponent controls which side of the fraction bar the factor lives on, never the sign of the value. A power of a positive base is always positive, whatever the exponent does.

Moving a term instead of a factor

In $\frac{1}{x ^{- 2}}$ the negative exponent can move, giving $x^{2}$ . In $\frac{1}{x ^{- 2} + y}$ it cannot, because $x^{- 2}$ is one term of a sum rather than a factor of the whole denominator. The correct move there is to rewrite the denominator as a single fraction first. Confusing terms with factors is the same error that makes people cancel across a plus sign.

Distributing a power over addition

$(a + b)^{2}$ is $a^{2} + 2 ab + b^{2}$ . The cross terms are real and they are usually the largest part of the answer. The radical version fails the same way: $9 + 16 = 5$ , while $3 + 4 = 7$ . Exponents distribute over multiplication only, and one numerical counterexample is enough to settle it permanently. The precedence reasoning behind which parts of an expression an exponent attaches to is worked through in the guide to order of operations.

Practice

Write powers with $^$ and give fractions in lowest terms.

Four simplifications

Answers are checked here — nothing is sent anywhere.

1
$Simplify: (x^{3})^{4} \cdot x^{- 2}$
2
$Evaluate: 2 5^{3/2}$
3
$Evaluate: 4^{- 3/2}$
4
$Simplify: \frac{12 a ^{5} b ^{- 2}}{3 a ^{- 1} b ^{2}}$

Where this goes next

Recognising perfect squares and cubes is what makes the four factoring patterns visible — a difference of squares is invisible until you see $9 x^{2}$ as $(3 x)^{2}$ . Rewriting roots as fractional powers is also the step that makes calculus possible, and it appears on the checklist in the algebra you need before calculus, where every entry is a skill from this page or the previous one. For an expression that will not come out, the algebra solver names the rule it applies at each step rather than jumping to a simplified form.

Frequently asked questions

Why does anything to the power zero equal 1?

Because the quotient rule forces it. a^n divided by a^n is obviously 1, and the rule says it is also a^(n-n) = a^0. Two names for the same number, so a^0 = 1. The argument needs a to be non-zero, since 0^n / 0^n is 0/0 and undefined, which is exactly why 0^0 is left undefined too.

Why is a negative exponent a reciprocal rather than a negative number?

Apply the product rule to a^n times a^(-n). The exponents add to zero, so the product is a^0 = 1. A number whose product with a^n is 1 is by definition the reciprocal of a^n, so a^(-n) = 1/a^n. Nothing about the sign of the exponent makes the value negative: 2^(-3) is 1/8, a positive number.

What does a fractional exponent actually mean?

a^(1/n) is the number that gives a when raised to the power n, because (a^(1/n))^n = a^(n/n) = a^1 by the power rule. That is the definition of an nth root, so a^(1/n) and the nth root of a are the same object written two ways. From there a^(m/n) is that root raised to the power m.

Is the square root of x squared equal to x?

Only when x is non-negative. The radical symbol denotes the principal root, which is never negative, so sqrt(x^2) = |x|. Check it with x = -3: the square is 9 and its principal square root is 3, not -3. This is why solving x^2 = 9 gives x = 3 or x = -3 while sqrt(9) alone gives only 3.

Why can't I split the square root of a sum?

Because the rule for splitting a radical applies to products, not sums. sqrt(9 * 4) = 3 * 2 = 6 works, but sqrt(9 + 16) = sqrt(25) = 5, whereas 3 + 4 = 7. One numerical test kills the false rule permanently, and it is worth doing once so the pattern sticks.

Keep going

Algebra solverSimplification and equation solving with the rule named at every step.Order of operationsWhich part of an expression an exponent actually attaches to.Factoring quadraticsWhere recognising perfect squares and cubes pays off immediately.The algebra you need before calculusA diagnostic that maps each gap to the fix for it.