Additional Mathematics Synthesis Guide 16: Hidden Quadratics, Substitution, Structural Factorisation and Equation Families

Recognise repetition → name the repeated object → solve the simpler structure → restore restrictions → return to the original variable → verify.

1. What makes a quadratic “hidden”?

A standard quadratic has the form au²+bu+c=0. A hidden quadratic has the same relationship, but the repeated object u is itself an expression in the original variable.

For example,

x⁴−5x²+4=0

is quadratic in u=x²:

u²−5u+4=0.

The useful observation is not merely that the powers are even. It is that x⁴=(x²)², creating a repeated object and its square.

2. Worked example — a biquadratic equation

Solve

x⁴−5x²+4=0.

Let u=x². Because x is real, u≥0. Then

(u−1)(u−4)=0.

So u=1 or 4. Returning to x gives

x=±1, ±2.

All four values satisfy the original equation. The temporary variable solved one quadratic, but each positive u-value produced two x-values.

3. The range of the substitution matters

Consider

x⁴+x²−6=0.

With u=x²≥0, we obtain

u²+u−6=(u+3)(u−2)=0.

The quadratic has u=−3 or u=2, but only u=2 is compatible with u=x² over the reals. Therefore

x=±√2.

A correct quadratic solution can still contain values that the substitution cannot produce.

4. Exponential equations can be quadratic in a positive quantity

Solve

4^x−5·2^x+4=0.

Since 4^x=(2^x)², let u=2^x. Then u>0 and

u²−5u+4=(u−1)(u−4)=0.

Thus 2^x=1 or 4, giving

x=0 or x=2.

The positive-range restriction is important even when both roots happen to be valid.

5. A negative quadratic root may be impossible before returning to x

Solve

4^x+2^x−6=0.

Again set u=2^x>0:

u²+u−6=(u+3)(u−2)=0.

u=−3 is impossible because 2^x is positive for every real x. Only u=2 remains, so

x=1.

The restriction is not an end-of-solution formality. It reduces the equation before the logarithmic or exponential return step.

6. Trigonometric equations can be quadratic in sinx or cosx

Solve

2sin²x−3sinx+1=0

for 0°≤x<360°.

Let u=sinx, so −1≤u≤1. Then

(2u−1)(u−1)=0.

Thus sinx=1/2 or sinx=1. Over the stated interval,

x=30°, 90°, 150°.

The substitution solved the algebraic structure; the trigonometric stage then supplied all angles in the required interval.

7. A trigonometric substitution carries a closed range

Suppose a transformed equation produces u=3/2 where u=cosx. That value should be rejected immediately because cosx cannot exceed 1 over the reals.

This is analogous to rejecting u<0 when u=e^x or u=x². Different substitutions create different allowable ranges. The temporary variable is not automatically an unrestricted real number.

8. Radical equations can be quadratic in √x

Solve

x−5√x+6=0

over the reals.

Because √x requires x≥0, let u=√x≥0. Then x=u² and

u²−5u+6=(u−2)(u−3)=0.

So u=2 or 3, giving

x=4 or 9.

No squaring of the original equation was necessary. The substitution uses the square-root expression already present.

9. A repeated algebraic expression can be named temporarily

Solve

(x²+x)²−5(x²+x)+6=0.

Let u=x²+x. Then

u²−5u+6=(u−2)(u−3)=0.

Case 1: x²+x=2, so x²+x−2=0 and x=1 or −2.

Case 2: x²+x=3, so x²+x−3=0 and

x=(-1±√13)/2.

The full solution set has four real values. Expanding the original fourth-degree expression first would obscure the repeated x²+x structure that makes the question simple.

10. Structural factorisation can sometimes avoid naming u

The previous equation can be factorised directly:

[(x²+x)−2][(x²+x)−3]=0.

This is the same mathematics as substitution. Naming u is useful when it clarifies the quadratic structure, but it is not compulsory if the factorisation is already visible.

The deeper skill is recognising the repeated object and treating it consistently.

11. Reciprocal symmetry can create a new variable

Solve

x+1/x=3, x≠0.

Multiplying by x gives

x²−3x+1=0.

Thus

x=(3±√5)/2.

The two roots are reciprocals of each other because their product is 1. This symmetry is built into the original equation.

12. A quartic can hide the reciprocal object x+1/x

Consider

x⁴−5x²+1=0, x≠0.

Divide by x²:

x²+1/x²=5.

Now let u=x+1/x. Since

u²=x²+2+1/x²,

we get u²=7, so u=±√7.

For u=√7, solve x²−√7x+1=0. For u=−√7, solve x²+√7x+1=0. These give four real roots:

x=(√7±√3)/2 and x=(-√7±√3)/2.

The reciprocal symmetry is now explicit: if x is a root, so is 1/x.

13. The range of x+1/x is restricted over the reals

For real x≠0, u=x+1/x cannot lie strictly between −2 and 2.

One way to see this is to rearrange x+1/x=u into

x²−ux+1=0.

For real x, the discriminant must satisfy u²−4≥0, so

u≤−2 or u≥2.

Therefore a transformed equation producing u=1 can be rejected before solving another quadratic.

14. Completing the square can expose a hidden variable range

Let u=x²+x. Then

u=(x+1/2)²−1/4≥−1/4.

This range matters. If an equation in u produces u=−1, that branch gives no real x even though the quadratic in u itself was solved correctly.

For example,

(x²+x)²−(x²+x)−2=0

gives u=2 or u=−1. The u=−1 branch is impossible because u≥−1/4. Only x²+x=2 remains, producing x=1 or −2.

15. Parameter families reveal how the number of roots changes

Consider the family

x⁴−(k+1)x²+k=0.

Let u=x²≥0. Then

u²−(k+1)u+k=(u−1)(u−k)=0.

The possible u-values are 1 and k. The number of distinct real x-roots depends on k:

• k>0, k≠1: two distinct positive u-values, giving four distinct real roots ±1 and ±√k.
• k=1: one repeated u-value 1, giving two distinct real roots ±1.
• k=0: u=1 or 0, giving three distinct real roots −1, 0, 1.
• k<0: u=k is impossible over the reals, leaving only ±1.

This is more than a parameter calculation. It shows how a family can change its number of real solutions when the parameter crosses structural boundaries.

16. Discriminants can be used after substitution

Suppose a transformed equation is

u²−ku+4=0

with u=e^x>0. The discriminant condition k²−16≥0 tells us when real u-candidates exist, but positivity supplies an additional condition.

If k≥4, both roots are positive because their sum k is positive and product 4 is positive. If k≤−4, both roots are negative, so neither can equal e^x. Thus real x-solutions exist only for

k≥4.

The discriminant alone would incorrectly include k≤−4 if the substitution range were ignored.

17. A substitution should reduce complexity

Not every new variable helps. If the substitution creates an equation just as complicated as the original, it has not earned its place.

Useful substitutions normally compress a repeated expression, reduce the number of different powers, or turn an inverse relationship into a familiar algebraic form.

Before substituting, ask: “What structure will become simpler?” This keeps substitution as a reasoning tool rather than a ritual response to unfamiliar notation.

18. Never forget the return journey

If a question asks for x and the quadratic solves for u, the work is incomplete until every valid u-value has been converted back into x-values.

This is especially important in trigonometry, where one u-value can produce several angles; in x² substitutions, where one positive u-value can produce two signs; and in exponential substitutions, where u>0 but each positive u determines one real x through a logarithm.

The number of quadratic roots is therefore not automatically the number of final solutions.

19. Verification should return to the original equation

A candidate may satisfy the transformed quadratic and still fail the original domain or branch condition. Substitute final x-values into the original equation when practical.

For trigonometric equations, also check the required interval. For radical equations, check the square-root domain. For logarithmic or exponential substitutions, check positivity. For reciprocal equations, check x≠0.

A transformed equation is a temporary workspace. The original equation remains the final authority.

20. Common failure patterns

• Expanding a repeated structure before noticing the quadratic pattern.
• Treating the substitution variable as unrestricted.
• Solving for u and forgetting to return to x.
• Assuming one u-root gives exactly one x-root.
• Using a discriminant condition without applying the substitution’s range.
• Dividing by x in a reciprocal problem without retaining x≠0.
• Accepting a trigonometric u-value outside [−1,1].
• Failing to verify candidates in the original equation or interval.

21. A reliable hidden-structure routine

1. Scan for a repeated expression and its square.
2. Choose a substitution that genuinely reduces complexity.
3. Write the range or domain of the new variable immediately.
4. Solve the transformed quadratic or factorisation.
5. Reject transformed roots outside the allowable range.
6. Return each surviving value to the original variable.
7. Apply interval or sign restrictions.
8. Verify final candidates in the original equation.

22. Practice set

1. Solve x⁴−13x²+36=0.
2. Solve x⁴+4x²−5=0 over the reals.
3. Solve 9^x−10·3^x+9=0.
4. Solve 9^x+2·3^x−3=0.
5. Solve 2cos²x−cosx−1=0 for 0°≤x<360°.
6. Solve x−7√x+12=0 over the reals.
7. Solve (x²−x)²−5(x²−x)+4=0.
8. Solve x+1/x=4.
9. State the real range of x+1/x for x≠0.
10. Solve x⁴−7x²+1=0 using reciprocal structure.
11. Find the real range of u=x²+x.
12. Solve (x²+x)²+2(x²+x)−3=0 over the reals.
13. For x⁴−(k+1)x²+k=0, how many distinct real roots are there when k=9?
14. How many when k=−2?
15. For u=e^x, explain why u²+5u+4=0 gives no real x.
16. Find k-values for which e^2x−ke^x+4=0 has at least one real solution.
17. Explain why u=sinx must satisfy −1≤u≤1.
18. Why can u=x² create two x-values from one positive u?
19. Why can a transformed quadratic have two roots but the original equation have only one real solution?
20. State one sign that a proposed substitution is probably unhelpful.

Answers

1. Let u=x²: (u−4)(u−9)=0, so x=±2,±3.
2. u²+4u−5=(u+5)(u−1)=0; u≥0, so x=±1.
3. Let u=3^x>0: (u−1)(u−9)=0, so x=0 or 2.
4. u²+2u−3=(u+3)(u−1)=0; only u=1, so x=0.
5. (2u+1)(u−1)=0 with u=cosx; cosx=−1/2 or 1, so x=0°,120°,240°.
6. Let u=√x≥0: (u−3)(u−4)=0, so x=9 or 16.
7. Let u=x²−x: (u−1)(u−4)=0. This gives x²−x−1=0 or x²−x−4=0, so x=(1±√5)/2 or (1±√17)/2.
8. x²−4x+1=0, so x=2±√3.
9. (−∞,−2]∪[2,∞).
10. Divide by x²: x²+1/x²=7. Let u=x+1/x; u²=9, so u=±3. Hence x=(3±√5)/2 or x=(−3±√5)/2.
11. u≥−1/4.
12. u²+2u−3=(u−1)(u+3)=0. Since u≥−1/4, only u=1. Then x²+x−1=0, so x=(−1±√5)/2.
13. k=9 gives u=1 or 9, hence x=±1,±3: four distinct real roots.
14. k=−2 gives u=1 or −2; only u=1 is allowed, so two roots ±1.
15. The quadratic roots are −1 and −4, but e^x is always positive.
16. k≥4.
17. Because sine is the y-coordinate of a point on the unit circle and cannot exceed 1 in magnitude.
18. Because x²=u has the two real solutions x=±√u when u>0.
19. Because one transformed root may lie outside the substitution range, or several transformed roots may map differently when returned to the original variable.
20. If it does not reduce the number of distinct structures or produces an equation no simpler than the original.

23. What mastery looks like

Mastery means the learner sees repeated structure before surface notation, chooses substitutions with an explicit purpose, carries the range of the new variable, understands that transformed roots and final solutions are different objects, and can move back to the original equation without losing branches or restrictions.

The transfer test is to place the same quadratic relationship inside different wrappers: x², e^x, sinx, √x or x+1/x. If the learner can identify what changes and what remains the same, substitution has become structural reasoning rather than a memorised trick.

---

Continue through Batch 04

• Guide 13: Radians, Arc Length, Sector Area and Trigonometric Modelling
• Guide 14: Rational Functions, Algebraic Fractions, Asymptotes and Domain Control
• Guide 15: Exponential and Logarithmic Change, Differentiation, Growth and Decay
• Additional Mathematics Directory
• BTT Mathematics Hub