College Algebra

Upper and Lower Bounds on Zeros

Last updated: March 2026 · Advanced
Before you start

You should be comfortable with:

The Rational Root Theorem gives a list of candidates. Descartes’ Rule tells you how many positive and negative zeros to expect. The bound theorems narrow the search further by establishing a range [L,U][L, U] that contains all real zeros. Any candidate outside this range can be skipped entirely.

The Upper Bound Theorem

If you divide f(x)f(x) by (xc)(x - c) using synthetic division where c>0c > 0, and all numbers in the bottom row are non-negative (no sign changes), then cc is an upper bound for the real zeros of ff. No real zero of ff is greater than cc.

Why It Works

When you perform synthetic division of f(x)f(x) by (xc)(x - c) with c>0c > 0 and get all non-negative entries in the bottom row, the quotient polynomial has all non-negative coefficients. For any x>cx > c, both the quotient q(x)q(x) and the divisor (xc)(x - c) are positive, so f(x)=(xc)q(x)+r>0f(x) = (x - c)q(x) + r > 0 (where r0r \geq 0 is the non-negative remainder). Therefore f(x)>0f(x) > 0 for all x>cx > c, meaning no zero exists beyond cc.

Worked Example 1: Finding an Upper Bound

Find an upper bound for the zeros of f(x)=x45x3+3x2+7x10f(x) = x^4 - 5x^3 + 3x^2 + 7x - 10.

Try c=5c = 5:

5153710501511010322100\begin{array}{c|ccccc} 5 & 1 & -5 & 3 & 7 & -10 \\ & & 5 & 0 & 15 & 110 \\ \hline & 1 & 0 & 3 & 22 & 100 \end{array}

Bottom row: 1,0,3,22,1001, 0, 3, 22, 100 — all non-negative.

Conclusion: 55 is an upper bound. No real zero is greater than 55.

Can we find a tighter bound? Try c=4c = 4:

41537104441211132\begin{array}{c|ccccc} 4 & 1 & -5 & 3 & 7 & -10 \\ & & 4 & -4 & -4 & 12 \\ \hline & 1 & -1 & -1 & 3 & 2 \end{array}

Bottom row: 1,1,1,3,21, -1, -1, 3, 2 — contains negative values. So 44 is NOT confirmed as an upper bound.

Since c=4c = 4 failed and c=5c = 5 succeeded, 55 is the smallest integer upper bound for this polynomial.

The Lower Bound Theorem

If you divide f(x)f(x) by (xc)(x - c) using synthetic division where c<0c < 0, and the numbers in the bottom row alternate in sign (counting zeros as either sign), then cc is a lower bound for the real zeros. No real zero of ff is less than cc.

The alternation means: positive, negative, positive, negative, … or negative, positive, negative, positive, … (zeros can be assigned to either sign to continue the pattern).

Worked Example 2: Finding a Lower Bound

Find a lower bound for the zeros of f(x)=x45x3+3x2+7x10f(x) = x^4 - 5x^3 + 3x^2 + 7x - 10.

Try c=2c = -2:

2153710214345417172744\begin{array}{c|ccccc} -2 & 1 & -5 & 3 & 7 & -10 \\ & & -2 & 14 & -34 & 54 \\ \hline & 1 & -7 & 17 & -27 & 44 \end{array}

Bottom row: 1,7,17,27,441, -7, 17, -27, 44 — alternating signs: +,,+,,++, -, +, -, +.

Conclusion: 2-2 is a lower bound. No real zero is less than 2-2.

Can we tighten it? Try c=1c = -1:

1153710169216928\begin{array}{c|ccccc} -1 & 1 & -5 & 3 & 7 & -10 \\ & & -1 & 6 & -9 & 2 \\ \hline & 1 & -6 & 9 & -2 & -8 \end{array}

Bottom row: 1,6,9,2,81, -6, 9, -2, -8 — the pattern is +,,+,,+, -, +, -, -. The last two entries are both negative (no alternation), so 1-1 is NOT confirmed as a lower bound.

So the lower bound is 2-2: all real zeros lie in [2,5][-2, 5].

Complete Zero-Finding Strategy

Combining all the tools from this cluster:

Step 1: Degree and Leading Coefficient

Determine end behavior and the maximum number of zeros.

Step 2: Descartes’ Rule

Count possible positive and negative real zeros to build the possibilities table.

Step 3: Rational Root Theorem

List all pq\frac{p}{q} candidates.

Step 4: Bound Theorems

Use synthetic division to find upper and lower bounds. Eliminate candidates outside the bounds.

Step 5: Test Remaining Candidates

Use synthetic division on the remaining candidates, reducing the polynomial’s degree each time you find a zero.

Step 6: Finish

When the polynomial is reduced to degree 2, use the quadratic formula.

Worked Example 3: Full Strategy in Action

Find all zeros of f(x)=2x4x314x2+7x+6f(x) = 2x^4 - x^3 - 14x^2 + 7x + 6.

Step 1: Degree 4, leading coefficient 2>02 > 0. Both ends go up.

Step 2 — Descartes’ Rule:

f(x)f(x): signs +,,,+,++, -, -, +, +. Sign changes: ++\to - (1), +++\to + (no). Wait, let me recount: +,,,+,++, -, -, +, +. Changes: ++\to- (1), -\to- (no), +-\to+ (2), +++\to+ (no). Two sign changes.

Positive zeros: 2 or 0.

f(x)=2x4+x314x27x+6f(-x) = 2x^4 + x^3 - 14x^2 - 7x + 6: signs +,+,,,++, +, -, -, +. Changes: +++\to+ (no), ++\to- (1), -\to- (no), +-\to+ (2). Two sign changes.

Negative zeros: 2 or 0.

Step 3 — Candidates: pq\frac{p}{q} where p6p \mid 6 and q2q \mid 2:

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

Step 4 — Bounds:

Test c=3c = 3:

32114766153302511036\begin{array}{c|ccccc} 3 & 2 & -1 & -14 & 7 & 6 \\ & & 6 & 15 & 3 & 30 \\ \hline & 2 & 5 & 1 & 10 & 36 \end{array}

All non-negative: 33 is an upper bound. Eliminate 66.

Test c=3c = -3:

321147662121422771448\begin{array}{c|ccccc} -3 & 2 & -1 & -14 & 7 & 6 \\ & & -6 & 21 & -21 & 42 \\ \hline & 2 & -7 & 7 & -14 & 48 \end{array}

Alternating: +,,+,,++, -, +, -, +. So 3-3 is a lower bound. Eliminate 6-6.

Remaining candidates: ±1,±2,±3,±12,±32\pm 1, \pm 2, \pm 3, \pm\frac{1}{2}, \pm\frac{3}{2}.

Step 5 — Test:

f(1)=2114+7+6=0f(1) = 2 - 1 - 14 + 7 + 6 = 0. Found x=1x = 1.

121147621136211360\begin{array}{c|ccccc} 1 & 2 & -1 & -14 & 7 & 6 \\ & & 2 & 1 & -13 & -6 \\ \hline & 2 & 1 & -13 & -6 & 0 \end{array}

Reduced: 2x3+x213x62x^3 + x^2 - 13x - 6.

Test x=12x = -\frac{1}{2}:

1221136106.5\begin{array}{c|cccc} -\frac{1}{2} & 2 & 1 & -13 & -6 \\ & & -1 & 0 & 6.5 \end{array}

2,0,13,0.52, 0, -13, 0.5 — remainder 0.500.5 \neq 0. Not a zero.

Test x=2x = 2:

221136410625312\begin{array}{c|cccc} 2 & 2 & 1 & -13 & -6 \\ & & 4 & 10 & -6 \\ \hline & 2 & 5 & -3 & -12 \end{array}

Remainder 120-12 \neq 0.

Test x=2x = -2:

22113646142378\begin{array}{c|cccc} -2 & 2 & 1 & -13 & -6 \\ & & -4 & 6 & 14 \\ \hline & 2 & -3 & -7 & 8 \end{array}

Remainder 808 \neq 0.

Test x=3x = 3:

3211366212427818\begin{array}{c|cccc} 3 & 2 & 1 & -13 & -6 \\ & & 6 & 21 & 24 \\ \hline & 2 & 7 & 8 & 18 \end{array}

Remainder 18018 \neq 0.

Test x=3x = -3:

321136615625212\begin{array}{c|cccc} -3 & 2 & 1 & -13 & -6 \\ & & -6 & 15 & -6 \\ \hline & 2 & -5 & 2 & -12 \end{array}

Remainder 120-12 \neq 0.

Test x=12x = \frac{1}{2}:

1221136116221212\begin{array}{c|cccc} \frac{1}{2} & 2 & 1 & -13 & -6 \\ & & 1 & 1 & -6 \\ \hline & 2 & 2 & -12 & -12 \end{array}

Remainder 120-12 \neq 0.

Test x=32x = \frac{3}{2}:

32211363610.524716.5\begin{array}{c|cccc} \frac{3}{2} & 2 & 1 & -13 & -6 \\ & & 3 & 6 & -10.5 \\ \hline & 2 & 4 & -7 & -16.5 \end{array}

Not a zero. Test x=32x = -\frac{3}{2}:

3221136331522109\begin{array}{c|cccc} -\frac{3}{2} & 2 & 1 & -13 & -6 \\ & & -3 & 3 & 15 \\ \hline & 2 & -2 & -10 & 9 \end{array}

Not a zero. So the cubic 2x3+x213x62x^3 + x^2 - 13x - 6 has no rational zeros from our list (other than what we may have missed). Let us re-check x=12x = -\frac{1}{2} more carefully:

1221136106.520130.5\begin{array}{c|cccc} -\frac{1}{2} & 2 & 1 & -13 & -6 \\ & & -1 & 0 & 6.5 \\ \hline & 2 & 0 & -13 & 0.5 \end{array}

Not zero. Let me try x=2x = -2 again on the original: f(2)=2(16)(8)14(4)+7(2)+6=32+85614+6=240f(-2) = 2(16) - (-8) - 14(4) + 7(-2) + 6 = 32 + 8 - 56 - 14 + 6 = -24 \neq 0.

Try the original polynomial with x=1x = -1: f(1)=2+1147+6=120f(-1) = 2 + 1 - 14 - 7 + 6 = -12 \neq 0.

The remaining cubic 2x3+x213x62x^3 + x^2 - 13x - 6 must be solved numerically or has rational zeros we have not found. Let me verify by trying to factor directly: grouping 2x3+x213x62x^3 + x^2 - 13x - 6. Testing x=2x = 2 gave remainder 12-12, x=2x = -2 gave 88. By the IVT, there is a zero between 22 and 33, and also between 2-2 and 3-3. These are irrational zeros. The quadratic formula finishes the job after finding one more rational zero or confirming there are none.

This example illustrates an important point: not all polynomials with integer coefficients have rational zeros, even when the Rational Root Theorem gives many candidates.

Real-World Application: Engineering Tolerance Bands

In manufacturing, a polynomial model describes how a measurement varies with a parameter xx. If all real roots must lie within a tolerance band [a,b][a, b], the bound theorems confirm this algebraically. For instance, if a vibration amplitude is modeled by:

A(f)=f312f2+44f48A(f) = f^3 - 12f^2 + 44f - 48

where ff is frequency in Hz, the bound theorems can verify that all resonant frequencies (zeros) lie within the operational range.

Testing c=12c = 12:

1211244481205281044480\begin{array}{c|cccc} 12 & 1 & -12 & 44 & -48 \\ & & 12 & 0 & 528 \\ \hline & 1 & 0 & 44 & 480 \end{array}

All non-negative: 1212 is an upper bound.

Testing c=1c = -1:

111244481135711357105\begin{array}{c|cccc} -1 & 1 & -12 & 44 & -48 \\ & & -1 & 13 & -57 \\ \hline & 1 & -13 & 57 & -105 \end{array}

Alternating signs: 1-1 is a lower bound.

All resonant frequencies lie in [1,12][-1, 12]. (In fact, this factors as (f2)(f4)(f6)(f - 2)(f - 4)(f - 6), with zeros at 22, 44, 66 Hz.)

Practice Problems

Test your understanding with these problems. Click to reveal each answer.

Problem 1: Use synthetic division to show that 44 is an upper bound for the zeros of f(x)=x33x22x+4f(x) = x^3 - 3x^2 - 2x + 4.

4132444811212\begin{array}{c|cccc} 4 & 1 & -3 & -2 & 4 \\ & & 4 & 4 & 8 \\ \hline & 1 & 1 & 2 & 12 \end{array}

Bottom row: 1,1,2,121, 1, 2, 12 — all non-negative.

Answer: Since c=4>0c = 4 > 0 and all entries are non-negative, 44 is an upper bound.

Problem 2: Show that 3-3 is a lower bound for the zeros of f(x)=x3+2x25x6f(x) = x^3 + 2x^2 - 5x - 6.

312563361120\begin{array}{c|cccc} -3 & 1 & 2 & -5 & -6 \\ & & -3 & 3 & 6 \\ \hline & 1 & -1 & -2 & 0 \end{array}

Bottom row: 1,1,2,01, -1, -2, 0. The signs are +,,,?+, -, -, ?. The zero can be assigned negative to continue: +,,,+, -, -, -. This does NOT alternate.

Actually, 3-3 is a zero (remainder is 0), so it IS a lower bound (trivially — it is itself a zero, and we should check if any zeros are smaller).

Let us try c=4c = -4:

41256481212318\begin{array}{c|cccc} -4 & 1 & 2 & -5 & -6 \\ & & -4 & 8 & -12 \\ \hline & 1 & -2 & 3 & -18 \end{array}

Alternating: +,,+,+, -, +, -. Yes, 4-4 is a lower bound.

Since f(x)=(x+3)(x+1)(x2)f(x) = (x + 3)(x + 1)(x - 2) (zeros at 3,1,2-3, -1, 2), 3-3 is the smallest zero and 4-4 is confirmed as a lower bound.

Answer: 4-4 is a lower bound (alternating signs). In fact, 3-3 is the smallest zero.

Problem 3: Find the smallest positive integer upper bound for the zeros of f(x)=x37x+6f(x) = x^3 - 7x + 6.

Try c=2c = 2: 1,2,3,01, 2, -3, 0. Has a negative, not all non-negative. Not confirmed.

Try c=3c = 3: Synthetic division:

3107639613212\begin{array}{c|cccc} 3 & 1 & 0 & -7 & 6 \\ & & 3 & 9 & 6 \\ \hline & 1 & 3 & 2 & 12 \end{array}

All non-negative: 33 is an upper bound.

Answer: 33 is the smallest positive integer upper bound. (The zeros are 1,2,31, 2, -3, confirming 33 works.)

Problem 4: Determine the bounds for zeros of g(x)=2x3+5x24x3g(x) = 2x^3 + 5x^2 - 4x - 3 and list all rational zeros.

Upper bound: try c=1c = 1:

125432732730\begin{array}{c|cccc} 1 & 2 & 5 & -4 & -3 \\ & & 2 & 7 & 3 \\ \hline & 2 & 7 & 3 & 0 \end{array}

Bottom row all non-negative AND remainder 0. So x=1x = 1 is both a zero and an upper bound.

Quotient: 2x2+7x+3=(2x+1)(x+3)2x^2 + 7x + 3 = (2x + 1)(x + 3).

Zeros: x=1x = 1, x=12x = -\frac{1}{2}, x=3x = -3.

Lower bound: try c=4c = -4:

425438123223835\begin{array}{c|cccc} -4 & 2 & 5 & -4 & -3 \\ & & -8 & 12 & -32 \\ \hline & 2 & -3 & 8 & -35 \end{array}

Alternating: +,,+,+, -, +, -. Confirmed, 4-4 is a lower bound.

Answer: Upper bound 1, lower bound 4-4. Zeros: x=1,12,3x = 1, -\frac{1}{2}, -3.

Problem 5: All real zeros of p(x)=x410x2+9p(x) = x^4 - 10x^2 + 9 lie in what interval? Factor p(x)p(x) completely to verify.

Try upper bound c=4c = 4:

4101009416249614624105\begin{array}{c|ccccc} 4 & 1 & 0 & -10 & 0 & 9 \\ & & 4 & 16 & 24 & 96 \\ \hline & 1 & 4 & 6 & 24 & 105 \end{array}

All non-negative: 44 is an upper bound.

Try lower bound c=4c = -4:

4101009416249614624105\begin{array}{c|ccccc} -4 & 1 & 0 & -10 & 0 & 9 \\ & & -4 & 16 & -24 & 96 \\ \hline & 1 & -4 & 6 & -24 & 105 \end{array}

Alternating: +,,+,,++, -, +, -, +. Confirmed, 4-4 is a lower bound.

All zeros lie in [4,4][-4, 4].

Factor: x410x2+9=(x21)(x29)=(x1)(x+1)(x3)(x+3)x^4 - 10x^2 + 9 = (x^2 - 1)(x^2 - 9) = (x-1)(x+1)(x-3)(x+3).

Zeros: ±1,±3\pm 1, \pm 3, all within [4,4][-4, 4].

Answer: All zeros lie in [4,4][-4, 4]. The zeros are 3,1,1,3-3, -1, 1, 3.

Key Takeaways

  • Upper Bound Theorem: If synthetic division by (xc)(x - c) with c>0c > 0 gives all non-negative entries, no zero exceeds cc
  • Lower Bound Theorem: If synthetic division by (xc)(x - c) with c<0c < 0 gives alternating signs, no zero is less than cc
  • The bounds narrow the search interval — eliminate candidates outside [L,U][L, U] immediately
  • The complete strategy combines degree analysis, Descartes’ Rule, Rational Root Theorem, bounds, and synthetic division
  • Start with small candidates (±1,±2\pm 1, \pm 2) before testing larger ones
  • When reduced to degree 2, use the quadratic formula to finish
  • Not all polynomials have rational zeros — the tools tell you when to stop looking

Return to College Algebra for more topics in this section.

Last updated: March 29, 2026