Some fourth-degree polynomial equations hide a quadratic in x + 1/x rather than x². The clue is reciprocal symmetry: the leading and constant coefficients match, and the coefficients of x³ and x match. Because the constant is non-zero, x = 0 is not a root, so you can divide by x² and substitute u = x + 1/x.
For example, x⁴ − 5x³ + 6x² − 5x + 1 = 0 becomes u² − 5u + 4 = 0. This gives u = 1 or 4, but real x requires u ≤ −2 or u ≥ 2. Therefore u = 1 is invalid and u = 4 gives x = 2 ± √3. This guide explains the hidden structure, reciprocal root pairs, and how to check repeated or inadmissible results without confusing this method with an ordinary biquadratic substitution.
Recognise the coefficient pattern
A monic reciprocal quartic has the form P(x) = x⁴ + ax³ + bx² + ax + 1. The coefficient sequence reads the same from either end: 1, a, b, a, 1.
That symmetry is the reason the substitution works. The x⁴ and constant terms form a reciprocal pair, and the x³ and x terms form another pair.
A polynomial with a different x coefficient from its cubic coefficient is not in this particular symmetric form. It may still be solvable by factorisation or another technique, but the simple substitution should not be forced onto an unrelated quartic.
Before expanding or guessing roots, compare the coefficients from the outside inward. Recognising structure is the most important first step, just as spotting x⁴ and x² signals a biquadratic equation.
Why x = 0 must be excluded before dividing
The quartic P(x) = x⁴ + ax³ + bx² + ax + 1 has constant term 1. At x = 0, the polynomial equals 1 rather than zero, so zero is not a root.
Therefore every root satisfies x ≠ 0. It is legitimate to divide the entire equation by x² without losing an actual solution.
The resulting equation contains positive and negative powers of x, which can be grouped symmetrically. This division step is not valid for an arbitrary polynomial unless the possibility x = 0 has been checked.
A good student writes “x ≠ 0 because P(0) = 1” before dividing. That one line makes the domain of 1/x explicit and prevents a hidden division-by-zero assumption.
Turn the equation into reciprocal pairs
Divide x⁴ + ax³ + bx² + ax + 1 = 0 by x². The expression becomes x² + ax + b + a/x + 1/x² = 0.
Group the symmetric terms as (x² + 1/x²) + a(x + 1/x) + b = 0.
Now introduce u = x + 1/x. Squaring gives u² = x² + 2 + 1/x², so x² + 1/x² = u² − 2.
The quartic therefore reduces to u² + au + (b − 2) = 0. This is an ordinary quadratic equation in u. The constant −2 is essential; forgetting the middle term in (x + 1/x)² produces an incorrect transformed equation.
Work the opening reciprocal quartic
Solve x⁴ − 5x³ + 6x² − 5x + 1 = 0. The coefficient pattern is symmetric: 1, −5, 6, −5, 1.
Because x = 0 is not a root, divide by x² to obtain x² + 1/x² − 5(x + 1/x) + 6 = 0.
Set u = x + 1/x. Then x² + 1/x² = u² − 2, so the equation becomes (u² − 2) − 5u + 6 = 0, or u² − 5u + 4 = 0.
Factorise: (u − 1)(u − 4) = 0. The algebraic candidates are u = 1 and u = 4. The next step is to decide whether each could actually equal x + 1/x for a real non-zero x.
Know the real range of x + 1/x
For positive x, the inequality x + 1/x ≥ 2 follows from (x − 1)² ≥ 0 after dividing by x > 0. Equality occurs at x = 1.
For negative x, write x = −t with t > 0. Then x + 1/x = −(t + 1/t) ≤ −2, with equality at x = −1.
Therefore, for real x ≠ 0, the possible values of u = x + 1/x are u ≤ −2 or u ≥ 2. The interval −2 < u < 2 is impossible.
This inherited range is the critical admissibility check. A root of the transformed quadratic can be perfectly valid as a number u but impossible as the sum of a real number and its reciprocal.
Reject u = 1 for the real-number problem
The transformed quadratic gave u = 1 or u = 4. But u = 1 lies strictly between −2 and 2, so no real non-zero x satisfies x + 1/x = 1.
You can verify that algebraically by multiplying by x. The equation x + 1/x = 1 becomes x² − x + 1 = 0. Its discriminant is (−1)² − 4 = −3, which is negative.
The lack of real roots is exactly what the range restriction predicted. Neither approach relies on guessing the sign of x.
Do not keep complex roots in a question that asks for real solutions. The correct final list must come only from admissible transformed values.
Return from u = 4 to x
The admissible transformed root is u = 4, so solve x + 1/x = 4 with x ≠ 0.
Multiply by x to obtain x² + 1 = 4x, or x² − 4x + 1 = 0. The quadratic formula gives x = [4 ± √(16 − 4)]/2 = [4 ± 2√3]/2.
Therefore x = 2 + √3 or x = 2 − √3. Both are positive real numbers and neither is zero.
These are the only real solutions of the original quartic. They come from one admissible u-value, which produces two real x-values because the equation x + 1/x = 4 is quadratic in x.
Check the answer by factorisation
The original polynomial factors as (x² − x + 1)(x² − 4x + 1). Multiply the factors to check: the product is x⁴ − 5x³ + 6x² − 5x + 1.
The first quadratic has discriminant −3 and therefore no real roots. The second has discriminant 12 and real roots 2 ± √3.
This factorisation agrees with the substitution method. It also reveals how each value of u turns into an x-quadratic: u = 1 gives x² − x + 1, while u = 4 gives x² − 4x + 1.
Comparing the two routes improves understanding. The reciprocal substitution is effectively a systematic way of discovering the pair of quadratic factors without guessing their coefficients.
Reciprocal roots occur in pairs
A reciprocal quartic obeys the identity P(x) = x⁴P(1/x) for x ≠ 0. Therefore if a non-zero number r is a root, its reciprocal 1/r is also a root.
In the opening example, (2 + √3)(2 − √3) = 4 − 3 = 1. Thus 2 + √3 and 2 − √3 are reciprocals.
This property provides an independent check. If a student reports one real root near 3.732 but a second root not near 0.268, the pair may be inconsistent with the reciprocal symmetry.
The property follows from the coefficient pattern, not from the particular numerical values. Every correctly formed reciprocal quartic with non-zero roots exhibits the reciprocal pairing, with possible repeated roots at 1 or −1.
Why u = 2 gives x = 1 only once
Suppose the transformed quadratic has a root u = 2. Returning to x gives x + 1/x = 2, so x² − 2x + 1 = (x − 1)² = 0.
This produces the single distinct real solution x = 1, though it is a repeated root of the corresponding quadratic factor.
Similarly, u = −2 produces x² + 2x + 1 = (x + 1)² = 0, giving the single distinct real value x = −1.
These boundary cases explain why the range endpoints u = ±2 are included. They also show why one admissible u-value does not always generate two distinct x-values.
A second example with a negative admissible u-value
Consider x⁴ − x³ − 4x² − x + 1 = 0. Its coefficients are symmetric, so divide by x² and use u = x + 1/x.
The transformed quadratic is u² − u − 6 = 0, which factors as (u − 3)(u + 2) = 0. Thus u = 3 or u = −2.
Both lie in the admissible real range: 3 ≥ 2 and −2 ≤ −2. For u = 3, solve x² − 3x + 1 = 0 to obtain x = (3 ± √5)/2.
For u = −2, solve (x + 1)² = 0 to obtain x = −1. Therefore there are three distinct real solutions, with x = −1 repeated algebraically. This example contrasts with the opening problem, where one transformed root had to be rejected.
Check the second quartic by quadratic factors
The second polynomial factorises exactly as (x² − 3x + 1)(x + 1)². Expanding gives x⁴ − x³ − 4x² − x + 1.
The first factor has two positive real roots, (3 + √5)/2 and (3 − √5)/2, whose product is 1. The second factor has the repeated negative root −1.
Their total multiplicity is four, matching the quartic’s degree, while the number of distinct real values is three.
This check illustrates two related counting rules: reciprocal solutions appear as paired values, while a root equal to its own reciprocal, namely 1 or −1, can appear with multiplicity and does not form a second distinct value.
A reciprocal quartic can have no real roots
Consider x⁴ + x² + 1 = 0. This polynomial is symmetric, with a = 0 and b = 1.
Dividing by x² and substituting u = x + 1/x gives u² + (1 − 2) = 0, or u² − 1 = 0. The transformed values are u = 1 and u = −1.
Neither lies in u ≤ −2 or u ≥ 2, so the original equation has no real solutions.
A direct sign proof reaches the same result: x⁴, x² and 1 are all non-negative, with the constant term strictly positive. Their sum can never equal zero for real x. This is a useful case where the sign argument is shorter than the full substitution.
What happens when the transformed u-root repeats?
Consider x⁴ − 6x³ + 11x² − 6x + 1 = 0. The corresponding u-quadratic is u² − 6u + 9 = (u − 3)².
Its only transformed root is u = 3, repeated. Since u = 3 is admissible, the x-equation is x² − 3x + 1 = 0.
Thus x = (3 ± √5)/2. Both real values occur with multiplicity two in the original quartic, because the entire polynomial is (x² − 3x + 1)².
There are only two distinct real solutions. A repeated u-root changes multiplicity but does not necessarily change the admissibility rule or create extra distinct x-values.
Keep the distinction from a biquadratic equation
A biquadratic has terms in x⁴, x² and a constant, such as x⁴ − 5x² + 4. It is quadratic in x², so the substitution is v = x² with v ≥ 0.
A reciprocal quartic, by contrast, can contain x³ and x terms in matching coefficients. Its natural substitution is u = x + 1/x, with admissible real values u ≤ −2 or u ≥ 2.
The restrictions are therefore different. A biquadratic substitution accepts any v ≥ 0, while reciprocal substitution accepts two exterior intervals for u.
Both methods turn degree-four equations into quadratics, but the choice depends on the actual coefficient structure. Correct method selection begins by inspecting the polynomial rather than applying the same substitution to every quartic.
Why the leading and constant coefficients must match
The symmetry P(x) = x⁴ + ax³ + bx² + ax + 1 makes the reciprocal pairs combine neatly. More generally, a quartic of the form A(x⁴ + 1) + B(x³ + x) + Cx² can be divided by x² and rewritten using x² + 1/x² and x + 1/x.
If the coefficients of x⁴ and the constant are different, or the coefficients of x³ and x are different, those simple reciprocal pairings fail.
A different structure may still be factorable, but the direct reduction to a quadratic in u may leave unmatched powers of x.
It is better to identify an appropriate pattern than to insist on a substitution that does not eliminate the original variable. An algebraic technique is useful only when it actually simplifies the problem.
A quick test for the general monic form
For P(x) = x⁴ + ax³ + bx² + ax + 1, the substitution yields u² + au + (b − 2) = 0.
This formula is a compact way to check the transformed coefficients. In the opening example, a = −5 and b = 6, so the u-equation is u² − 5u + 4.
In the second example, a = −1 and b = −4, so the equation becomes u² − u − 6. These match the detailed calculations.
Students should still understand that the −2 comes from x² + 1/x² = u² − 2. Memorising the final formula without recognising its derivation can lead to errors when the coefficients are scaled or the polynomial has a different structure.
Negative x-values correspond to u ≤ −2
If x is negative, both x and 1/x are negative. Their sum u is therefore negative, and the reciprocal inequality gives u ≤ −2.
For example, x = −1 has u = −2. The negative value x = −2 has u = −2 − 1/2 = −5/2.
This is useful when a transformed quadratic gives u = −3. That u-value is admissible and leads to x² + 3x + 1 = 0, whose two roots are both negative: x = (−3 ± √5)/2.
Do not discard all negative u-values. The inadmissible region is −2 < u < 2; values at or below −2 are as legitimate as values at or above 2 for real x.
A negative-u example with two distinct roots
Suppose the transformed equation has u = −3. Solve x + 1/x = −3, multiply by non-zero x and rearrange to x² + 3x + 1 = 0.
The roots are x = [−3 ± √(9 − 4)]/2 = (−3 ± √5)/2. Both are negative because √5 < 3.
Their product is 1, so they are reciprocal. Their sum is −3, matching u. These two checks prove that the original relation is satisfied.
The same reasoning works for any u < −2: the corresponding x-quadratic has two distinct negative real roots. At u = −2, those roots merge at x = −1.
A reciprocal quartic inequality needs a zero check
Consider P(x) = x⁴ − 5x³ + 6x² − 5x + 1 and the inequality P(x) < 0. For x ≠ 0, divide by x², which is strictly positive, so the sign of the inequality is unchanged.
In terms of u = x + 1/x, the expression becomes u² − 5u + 4 < 0, or (u − 1)(u − 4) < 0. Thus 1 < u < 4.
The real-u restriction cuts this to 2 ≤ u < 4, since values between 1 and 2 are impossible. Thus x must be positive, and x + 1/x < 4. That inequality holds for 2 − √3 < x < 2 + √3.
At x = 0 the original polynomial equals 1, so zero is not a solution. The complete strict solution is (2 − √3, 2 + √3). The direct factorisation confirms it because x² − x + 1 is always positive, leaving the sign of x² − 4x + 1.
Why dividing by x² preserves inequality direction
When x ≠ 0, x² is strictly positive. Dividing an inequality by a positive quantity preserves its direction.
That makes the reciprocal substitution safe for sign comparisons on the non-zero domain. But a student must not forget to inspect x = 0 separately, because division by x² removes it from the transformed problem.
For the polynomial P above, P(0) = 1, so x = 0 does not satisfy P(x) < 0 or P(x) ≤ 0. The transformed solution is therefore complete without restoring zero.
This example generalises an important habit: algebraic transformations involving division require checking whether excluded values could have satisfied the original relation. In the monic reciprocal quartic, the non-zero constant ensures that zero is not a root of the equation.
An exact numerical check of reciprocal roots
The first example gave x₁ = 2 + √3 and x₂ = 2 − √3. Their product is 1 and their sum is 4.
The product confirms reciprocal pairing, while the sum confirms the transformed value u = 4. Each value is positive, and neither is zero.
If desired, substitute the factor x² − 4x + 1. At either root, x² − 4x + 1 = 0 by the quadratic formula. Consequently the full factored quartic also equals zero.
This is more reliable than substituting rounded decimal roots into a fourth-degree polynomial, where rounding can produce small nonzero residues even when the exact values are correct.
A reliable five-stage solution
First inspect the quartic for reciprocal coefficient symmetry. Second check x = 0 against the original polynomial before dividing by x².
Third group x² + 1/x² and x + 1/x, substitute u = x + 1/x and replace the first group with u² − 2. Fourth solve the resulting quadratic in u.
Fifth retain only u ≤ −2 or u ≥ 2 and solve x² − ux + 1 = 0 for each valid u. List the distinct real x-values and verify them in the original or factored equation.
The method converts an apparently complicated quartic into two familiar quadratic stages, but the real-domain filter is essential between them.
Practice questions with checked answers
Question one: solve x⁴ − 5x³ + 6x² − 5x + 1 = 0 over the reals. Answer: x = 2 ± √3.
Question two: solve x⁴ − x³ − 4x² − x + 1 = 0. Answer: x = −1 or x = (3 ± √5)/2.
Question three: solve x⁴ + x² + 1 = 0 over the reals. Answer: no real solutions.
Question four: solve x⁴ − 6x³ + 11x² − 6x + 1 = 0. Answer: x = (3 ± √5)/2, each repeated.
Question five: solve x + 1/x = −3 for real non-zero x. Answer: x = (−3 ± √5)/2.
Question six: solve x⁴ − 5x³ + 6x² − 5x + 1 < 0. Answer: 2 − √3 < x < 2 + √3.
Show the admissible u-values in every question, even when the factoring appears simple, and check the reciprocal pairing of nonzero real roots.
How tutors should separate method errors
A student who cannot spot symmetric coefficients needs practice identifying algebraic patterns. One who transforms correctly but accepts u = 1 has missed the range of x + 1/x. Another may reject negative u-values even when u ≤ −2.
A fourth may correctly obtain u = 4 yet keep only one of the quadratic roots. That calls for careful solution of x² − ux + 1 = 0 and interpretation of its discriminant.
These are different errors and deserve different targeted corrections. A changed quartic should be solved without prompts after each explanation.
The method can serve as mathematical enrichment where a student’s current assessed syllabus focuses on simpler quartics or quadratics. Appropriate challenge depends on prerequisite algebra and the student’s readiness, not just on the polynomial’s degree.
Syllabus context and connected reading
Reciprocal quartics combine polynomial structure, substitution, discriminants and real-domain reasoning. These symmetric fourth-degree extensions should be used with regard to the student’s school sequence and applicable syllabus rather than presented as compulsory for every A-Math pathway.
Singapore families can consult SEAB’s 2026 O-Level Additional Mathematics 4049, 2027 SEC G3 K341 or 2027 SEC G2 K232 as appropriate.
For connected learning, see How to Solve a Biquadratic Equation Using x² Substitution, How to Spot a Hidden Quadratic in an Exponential Equation and How to Recover Polynomial Coefficients from Two Remainders.
Explore more in the eduKateSG Additional Mathematics archive, or enquire about suitable maximum three-student tutorials near Sixth Avenue MRT. The key skill is choosing a substitution from the polynomial’s actual symmetry and preserving the set of real inputs it represents. Properly taught kids shine a bright light into the future.