Math Core

Lesson 2.2 · Polynomial and Rational Functions

Real and complex zeros

A graph shows you a polynomial's real zeros as xx-intercepts, but it can't give you exact values, and it hides the complex zeros completely. This lesson collects the algebra tools that find every zero exactly: the rational zero theorem to generate candidates, synthetic division to test them and shrink the polynomial, Descartes' rule of signs to know what to expect, and the fundamental theorem of algebra to know when you're done.

The rational zero theorem

Guessing zeros at random is hopeless. Fortunately, for polynomials with integer coefficients, any rational zero must come from a short list.

Rational zero theorem

If f(x)=anxn+⋯+a1x+a0f(x) = a_n x^n + \cdots + a_1 x + a_0 has integer coefficients and pq\dfrac{p}{q} is a rational zero in lowest terms, then

  • pp is a factor of the constant term a0a_0, and
  • qq is a factor of the leading coefficient ana_n.

Here's why. If f(pq)=0f\left(\dfrac{p}{q}\right) = 0, multiply through by qnq^n:

anpn+an−1pn−1q+⋯+a1pqn−1+a0qn=0.a_n p^n + a_{n-1}p^{n-1}q + \cdots + a_1 p q^{n-1} + a_0 q^n = 0.

Every term except the last contains a factor of pp, so pp must divide a0qna_0 q^n. Since pp and qq share no factors, pp divides a0a_0. The same argument with the first term shows qq divides ana_n.

The theorem only lists candidates. You still have to test them, and a polynomial may have no rational zeros at all.

Testing candidates with synthetic division

Synthetic division by x−cx - c does two jobs at once. The last number is the remainder, which equals f(c)f(c), so a remainder of 00 means cc is a zero. The other numbers are the coefficients of the quotient, a polynomial one degree lower, called the depressed polynomial. Keep dividing until you reach a quadratic, then finish with factoring or the quadratic formula.

Worked example: Finding all rational zeros

Find all zeros of f(x)=2x3−3x2−11x+6f(x) = 2x^3 - 3x^2 - 11x + 6.

Candidates. Factors of 66: ±1,±2,±3,±6\pm 1, \pm 2, \pm 3, \pm 6. Factors of 22: ±1,±2\pm 1, \pm 2. So the possible rational zeros are

±1, ±2, ±3, ±6, ±12, ±32.\pm 1, \ \pm 2, \ \pm 3, \ \pm 6, \ \pm \tfrac{1}{2}, \ \pm \tfrac{3}{2}.

Test. f(1)=2−3−11+6=−6f(1) = 2 - 3 - 11 + 6 = -6, so 11 is not a zero. Try 33 with synthetic division:

22−3-3−11-1166
336699−6-6
2233−2-200

The remainder is 00, so 33 is a zero and f(x)=(x−3)(2x2+3x−2)f(x) = (x - 3)(2x^2 + 3x - 2).

Finish. 2x2+3x−2=(2x−1)(x+2)2x^2 + 3x - 2 = (2x - 1)(x + 2). The zeros are 33, 12\dfrac{1}{2} and −2-2.

Irrational zeros

Once the depressed polynomial is a quadratic, the quadratic formula finds its zeros whether or not they are rational. Irrational zeros of a polynomial with rational coefficients come in pairs a±ba \pm \sqrt{b}, just like the quadratic formula produces them.

Worked example: A rational zero and an irrational pair

Find all zeros of f(x)=x3−3x2−2x+4f(x) = x^3 - 3x^2 - 2x + 4.

The leading coefficient is 11, so the candidates are just the factors of 44: ±1,±2,±4\pm 1, \pm 2, \pm 4. Try 11: f(1)=1−3−2+4=0f(1) = 1 - 3 - 2 + 4 = 0. ✓

11−3-3−2-244
1111−2-2−4-4
11−2-2−4-400

The depressed polynomial is x2−2x−4x^2 - 2x - 4, which doesn't factor over the integers. Use the quadratic formula:

x=2±4+162=2±252=1±5.x = \frac{2 \pm \sqrt{4 + 16}}{2} = \frac{2 \pm 2\sqrt{5}}{2} = 1 \pm \sqrt{5}.

The zeros are 11, 1+51 + \sqrt{5} and 1−51 - \sqrt{5}.

Descartes' rule of signs

Before you start testing, it helps to know how many positive and negative zeros to expect. Write f(x)f(x) in descending order and count the sign changes between consecutive nonzero coefficients.

Descartes' rule of signs

Let ff have real coefficients.

  • The number of positive real zeros equals the number of sign changes in f(x)f(x), or is less than that by an even number.
  • The number of negative real zeros equals the number of sign changes in f(−x)f(-x), or is less than that by an even number.

Zeros are counted with multiplicity.

"Less by an even number" is because non-real zeros come in conjugate pairs, so real zeros disappear two at a time.

Worked example: Counting sign changes

What does Descartes' rule say about f(x)=x4−2x3+x2+2x−6f(x) = x^4 - 2x^3 + x^2 + 2x - 6?

The signs of the coefficients are + − + + −+ \, - \, + \, + \, -. The changes are ++ to −-, −- to ++, and ++ to −-: three changes. So ff has 33 or 11 positive real zeros.

For negative zeros, replace xx with −x-x. Even powers keep their sign and odd powers flip:

f(−x)=x4+2x3+x2−2x−6.f(-x) = x^4 + 2x^3 + x^2 - 2x - 6.

The signs are + + + − −+ \, + \, + \, - \, -: exactly one change. So ff has exactly 11 negative real zero. That's worth knowing: once you find one negative zero, stop testing negative candidates.

Complex zeros and complete factorization

The fundamental theorem of algebra guarantees that a polynomial of degree nn has exactly nn complex zeros, counted with multiplicity. Combined with the factor theorem, that means every polynomial splits completely into linear factors.

Linear factorization theorem

A polynomial of degree n≥1n \ge 1 with leading coefficient ana_n can be written as

f(x)=an(x−c1)(x−c2)⋯(x−cn),f(x) = a_n(x - c_1)(x - c_2)\cdots(x - c_n),

where c1,…,cnc_1, \ldots, c_n are its complex zeros (some may repeat). If the coefficients are real, the non-real zeros come in conjugate pairs a±bia \pm bi.

So your search is finished when you've found nn zeros. Over the reals, each conjugate pair stays together as an irreducible quadratic factor.

Worked example: Real and non-real zeros

Find all zeros of f(x)=x4−x3+2x2−4x−8f(x) = x^4 - x^3 + 2x^2 - 4x - 8, and write ff as a product of linear factors.

Candidates: ±1,±2,±4,±8\pm 1, \pm 2, \pm 4, \pm 8. Test: f(2)=16−8+8−8−8=0f(2) = 16 - 8 + 8 - 8 - 8 = 0 ✓ and f(−1)=1+1+2+4−8=0f(-1) = 1 + 1 + 2 + 4 - 8 = 0 ✓. So (x−2)(x+1)=x2−x−2(x - 2)(x + 1) = x^2 - x - 2 is a factor. Divide (or use synthetic division twice) to get

f(x)=(x2−x−2)(x2+4).f(x) = (x^2 - x - 2)(x^2 + 4).

Check by expanding: x4−x3−2x2+4x2−4x−8=x4−x3+2x2−4x−8x^4 - x^3 - 2x^2 + 4x^2 - 4x - 8 = x^4 - x^3 + 2x^2 - 4x - 8. ✓

From x2+4=0x^2 + 4 = 0, x=±2ix = \pm 2i. The four zeros are 22, −1-1, 2i2i and −2i-2i, and

f(x)=(x−2)(x+1)(x−2i)(x+2i).f(x) = (x - 2)(x + 1)(x - 2i)(x + 2i).

The graph shows only the two real zeros. The factor x2+4x^2 + 4 is always positive, so it never creates an xx-intercept.

f(x) = x⁴ − x³ + 2x² − 4x − 8 crosses the x-axis only at x = −1 and x = 2. Its zeros ±2i don't appear on the graph.Open in grapher →

Common mistake

Don't stop when you run out of xx-intercepts. A degree-44 polynomial always has 44 complex zeros. If you've found only two real ones, the other two are either another real pair or a complex conjugate pair, and the depressed quadratic will tell you which.

Tip

Use every shortcut the theorems give you. The sum of the coefficients is f(1)f(1), so if they add to 00, then 11 is a zero, and if not, 11 is out. If all coefficients are positive, Descartes' rule says there are no positive zeros, so test only negative candidates.

Practice

Practice 1

Which number is not a possible rational zero of f(x)=3x3−5x2+x−2f(x) = 3x^3 - 5x^2 + x - 2?

Practice 2

Find all zeros of f(x)=x3−7x−6f(x) = x^3 - 7x - 6.

Separate answers with commas, e.g. 2, -5

Practice 3

Find all zeros of f(x)=2x3+3x2−8x+3f(x) = 2x^3 + 3x^2 - 8x + 3.

Separate answers with commas, e.g. 2, -5

Practice 4

Find all zeros of f(x)=x3−5x2+5x+3f(x) = x^3 - 5x^2 + 5x + 3. Use sqrt() for square roots.

Separate answers with commas, e.g. 2, -5

Practice 5

What does Descartes' rule of signs say about f(x)=x5−3x4+2x2+x−7f(x) = x^5 - 3x^4 + 2x^2 + x - 7?

Practice 6

The polynomial f(x)=x3+x2+3x−5f(x) = x^3 + x^2 + 3x - 5 has one real zero and two non-real zeros a±bia \pm bi with b>0b > 0. Enter (a,b)(a, b).

Enter a point like (2, -3)

Practice 7

Given that 2i2i is a zero of f(x)=x4+x3+2x2+4x−8f(x) = x^4 + x^3 + 2x^2 + 4x - 8, find all the real zeros of ff.

Separate answers with commas, e.g. 2, -5

Practice 8

How many non-real complex zeros does f(x)=x5+3x3−4xf(x) = x^5 + 3x^3 - 4x have?

Enter a number. Fractions like 3/4 and sqrt(2) are OK.