Squaring a surd equation can produce algebraic candidates that do not satisfy the original equation. The safe method is to check where every real square root is defined, isolate one radical when appropriate, check the signs of both sides, square carefully and test every candidate in the unsquared equation.
For example, √(x + 5) = x − 1 leads after squaring to x² − 3x − 4 = 0, with candidates x = 4 and x = −1. Only x = 4 is valid: √9 = 3 equals 4 − 1. At x = −1, the left side is 2 but the right side is −2. This guide shows how to identify such extraneous roots before they become an incorrect final answer and how to handle equations with two radicals.
A square root has a domain and a sign
In real-number Additional Mathematics, √A denotes the principal, non-negative square root when A ≥ 0. The inside expression must be non-negative, and the square-root output itself cannot be negative.
Therefore an equation √A(x) = B(x) requires two preliminary conditions: A(x) ≥ 0 and B(x) ≥ 0. If either fails at a candidate input, that candidate cannot solve the original equation.
Students sometimes remember only the condition inside the radical. But a non-negative left side also forces its equal right side to be non-negative.
These sign conditions are particularly useful after squaring. The squared equation A(x) = [B(x)]² no longer distinguishes B from −B, which is exactly how extraneous roots can arise. Record both conditions before transforming the equation.
Why squaring is not automatically reversible
If two real numbers are equal, their squares are equal. But the converse is not generally true: 3² = (−3)² even though 3 ≠ −3.
Squaring both sides of an equation is therefore a necessary transformation but may not be sufficient. It can enlarge the set of candidate solutions by losing sign information.
For √(x + 5) = x − 1, squaring creates x + 5 = (x − 1)². The new equation can hold when x − 1 is negative even though the original square-root side is non-negative.
A complete solution either preserves the required sign conditions throughout or checks each candidate in the original equation at the end. Both approaches are mathematically sound; doing both when the calculation is short offers a useful safeguard.
Worked example with one genuine and one extraneous candidate
Solve √(x + 5) = x − 1. The radical requires x ≥ −5. The right side must be non-negative, so x ≥ 1. Together these restrict potential solutions to x ≥ 1.
Square both sides: x + 5 = x² − 2x + 1. Rearrange to x² − 3x − 4 = 0. Factorise: (x − 4)(x + 1) = 0.
The algebraic candidates are x = 4 and x = −1. Only x = 4 satisfies the preliminary requirement x ≥ 1.
Check directly: at x = 4, √(4 + 5) = √9 = 3 and x − 1 = 3. At x = −1, √4 = 2 while x − 1 = −2. Thus the original equation has exactly one real solution, x = 4.
Why the rejected candidate was generated
At x = −1, the squared equation gives x + 5 = 4 and (x − 1)² = (−2)² = 4. Both squared sides agree.
But the original equality asks whether √4 = −2, which is false. Taking the principal square root gives +2, not −2.
The transformation from the original equation to the squared equation lost the sign of the right-hand expression. That is the exact cause of the extraneous candidate; it is not an unpredictable calculator error.
A student who can explain why x = −1 appeared and why it must be rejected understands more than one who simply crosses it out because the answer key lists only x = 4.
A domain check can reject a candidate before substitution
Since √(x + 5) = x − 1 implies x ≥ 1, a candidate x = −1 can be rejected immediately without recomputing the square root.
However, explicitly substituting it into the original equation can still be educational. It shows that the two sides have opposite signs and therefore cannot be equal, even though their squares match.
A sensible examination solution may write: “Since x − 1 ≥ 0, x ≥ 1. The squared equation gives x = 4 or −1, but −1 is invalid. Hence x = 4.”
This is concise and rigorous. The learner should be able to reconstruct the sign restriction independently rather than memorising a rule that negative answers are always wrong. Negative x-values can be perfectly valid in many surd equations.
A second example with the root on the left
Solve √(2x + 3) = x. The radical condition is 2x + 3 ≥ 0, and the right side must satisfy x ≥ 0. Therefore potential solutions have x ≥ 0.
Squaring gives 2x + 3 = x². Rearranging produces x² − 2x − 3 = (x − 3)(x + 1) = 0.
The candidates are x = 3 and x = −1. Reject −1 because the right side of the original equation would be negative.
At x = 3, √(2·3 + 3) = √9 = 3, which matches the right side. The only real solution is x = 3.
This is another case where the transformed quadratic has two roots but the original surd equation has just one.
A negative input can be a valid surd solution
It would be wrong to conclude that every negative x must be rejected in a square-root equation. The restriction applies to the radicand and the sign of any other expression equal to the square root, not to the sign of x by itself.
For example, √(x + 4) = 1 has solution x = −3. The radicand is then 1, which is non-negative, and the square root equals the stated positive number.
Similarly, √(x + 1) = 0 has solution x = −1. The radicand equals zero, which is permitted.
This contrast is important. A mathematical condition such as x ≥ 1 arises from the structure of a particular equation, not from a universal rule forbidding negative x-values whenever a surd appears.
When the right side has a stronger restriction
Consider √(x − 2) = x − 4. The radical requires x ≥ 2, but the right-hand side must also be non-negative, requiring x ≥ 4. The combined necessary domain is x ≥ 4.
Square the equation: x − 2 = (x − 4)² = x² − 8x + 16. Rearranging gives x² − 9x + 18 = 0, or (x − 3)(x − 6) = 0.
The candidates are x = 3 and x = 6. Reject 3 because it violates x ≥ 4. At x = 6, √(6 − 2) = √4 = 2 and 6 − 4 = 2.
Thus x = 6 is the only solution. The strict preliminary reading saves time by revealing which candidate can survive before direct substitution.
Always preserve zero as an allowable square-root value
The principal square root is non-negative, not strictly positive. A radicand equal to zero is allowed, and the square root then equals zero.
For √(x + 1) = 0, x = −1 is valid. If a learner writes x + 1 > 0 instead of x + 1 ≥ 0, that legitimate endpoint is lost.
The distinction changes when a square root appears in a denominator. For 1/√(x + 1), the denominator cannot vanish, so x + 1 must be strictly positive.
Read the role of the radical before writing a domain inequality. The mathematical restriction is not identical for √A, 1/√A and log A, even though each expression contains an argument whose sign matters.
Two radicals require a common starting domain
Consider √(x + 4) + √(x − 1) = 5. Both square roots must be real, giving x + 4 ≥ 0 and x − 1 ≥ 0. Their intersection is x ≥ 1.
The sum of two non-negative roots is non-negative, so equality with the positive constant 5 is possible. Unlike a problem with a negative right-hand side, there is no immediate impossibility.
The standard approach is to isolate one radical and square. After that, a remaining radical may be isolated and squared again.
At every stage, record any new sign requirement before squaring. Equations with two radicals invite more transformations, so the risk of losing signs or introducing unwanted candidates increases unless the working remains organised.
Work the two-radical equation in full
Solve √(x + 4) + √(x − 1) = 5 for real x. The domain is x ≥ 1. Isolate the first radical: √(x + 4) = 5 − √(x − 1).
The right-hand side must be non-negative. Squaring gives x + 4 = 25 − 10√(x − 1) + (x − 1), so x + 4 = x + 24 − 10√(x − 1).
Cancel x and rearrange: 10√(x − 1) = 20, hence √(x − 1) = 2. Square to obtain x − 1 = 4, so x = 5.
Check the original: √9 + √4 = 3 + 2 = 5. Therefore x = 5 is a valid solution. The calculation never assumed that a square root of a sum equals a sum of square roots.
A quick method using the difference of squares
The previous equation has two radicals whose squared expressions differ by a constant: (x + 4) − (x − 1) = 5. Let A = √(x + 4) and B = √(x − 1).
The equation says A + B = 5. Since A² − B² = 5, factorising gives (A − B)(A + B) = 5, so A − B = 1.
Now solve the simple simultaneous equations A + B = 5 and A − B = 1. They give A = 3 and B = 2. Hence x + 4 = 9 and x − 1 = 4, so x = 5.
This alternative avoids repeated squaring and makes good use of the fixed difference of the radicands. It is valid because A and B are non-negative real square roots and the sum is known.
Do not use a difference-of-squares shortcut when the sum is zero
The shortcut A − B = (A² − B²)/(A + B) divides by A + B. It therefore requires A + B ≠ 0.
If two non-negative square roots have sum zero, both must individually be zero. That special case should be handled directly rather than dividing by zero.
In the example A + B = 5, division is safe because 5 is non-zero. The relation A² − B² = 5 then legitimately yields A − B = 1.
This small condition illustrates a general mathematical principle: every division step needs a non-zero divisor. A shortcut that is efficient under the proper assumption can be invalid if used without checking that assumption.
Another two-radical equation with an exact fractional answer
Solve √(x + 5) + √(x + 1) = 4. The common real domain is x ≥ −1. Let A = √(x + 5) and B = √(x + 1).
Then A + B = 4 and A² − B² = 4. Dividing gives A − B = 1. The simultaneous equations yield A = 5/2 and B = 3/2.
Since B² = x + 1, we have 9/4 = x + 1, giving x = 5/4. The other expression confirms x + 5 = 25/4 and A = 5/2.
Check the original: 5/2 + 3/2 = 4. The exact fractional solution is preferable to rounding 1.25 prematurely, particularly if the question later uses it in another algebraic expression.
An equation can have no solution by a simple lower bound
Consider √(x + 4) + √(x − 1) = 1. The common domain is x ≥ 1.
On that domain, √(x + 4) is at least √5, which is greater than 2. Meanwhile √(x − 1) is non-negative. Therefore their sum is greater than 1 for every permitted x.
No real solution exists. There is no need to square repeatedly or solve a lengthy transformed equation.
This sign-and-bound check is a useful alternative method. Before moving into algebra, ask whether the requested right-hand value is even attainable. A direct inequality may prove impossibility more clearly than any squared equation.
Differences of radicals need a sign check
Consider √(x + 9) − √x = 1, with x ≥ 0. Isolate the first root: √(x + 9) = 1 + √x.
Square both sides to obtain x + 9 = 1 + 2√x + x, so 2√x = 8. Hence √x = 4 and x = 16.
Check: √25 − √16 = 5 − 4 = 1, so the solution is valid.
The left side is positive because x + 9 > x and the square-root function is increasing on non-negative inputs. Thus an equation √(x + 9) − √x = −1 would have no real solution. Such qualitative sign reasoning can prevent unnecessary algebra.
Squaring both sides of two radicals can conceal a new sign condition
Suppose an equation has the form √A(x) + √B(x) = C(x). Squaring yields A + B + 2√(AB) = C², so the isolated remaining radical must satisfy 2√(AB) = C² − A − B.
The left side is non-negative. Therefore the right side must be non-negative before another squaring can be justified as a reversible step.
A second squaring may create algebraic candidates that make C² − A − B negative. Those candidates satisfy the double-squared equation but not the previous equation, let alone the original sum.
To remain safe, keep the real domains A ≥ 0 and B ≥ 0, require C ≥ 0, preserve the intermediate sign restriction and test every resulting candidate in the starting equation. This is the correct logic even when the resulting polynomial is high degree.
Equal square roots can be squared safely with their domains
Consider √(x + 1) = √(2x − 1). The radicands require x ≥ −1 and x ≥ 1/2, so the common domain is x ≥ 1/2.
Both sides are non-negative wherever they are defined. Squaring is therefore equivalent to equating their radicands: x + 1 = 2x − 1, giving x = 2.
Check: the left is √3 and the right is √3. The solution is valid.
This example is different from √A = B with an unrestricted signed expression B. When both sides are principal square roots, their non-negativity is built into the domain. Squaring them is reversible on the common real domain.
The difference between √(x²) and x
For real x, √(x²) = |x|, not x for every real number. The principal square root is always non-negative.
At x = 3, √9 = 3, so |x| = x. At x = −3, √9 = 3, while x = −3, so the expressions differ.
This matters when simplifying equations before or after squaring. Replacing √(x²) with x without a domain assumption x ≥ 0 can lose valid negative inputs or create an incorrect identity.
A student should translate the square root of a square into an absolute value, then use the given domain to decide whether the absolute value simplifies to x or −x. That preserves the sign information the radical carries.
Never distribute a square root over a sum
A common false shortcut is √(a + b) = √a + √b. Taking a = b = 1 shows it is wrong: √2 is not 2.
Likewise, √(x + 5) cannot be separated into √x + √5 for arbitrary real x. Doing so introduces a different expression and often a different domain.
Valid surd identities include √(ab) = √a√b when a and b are non-negative real numbers, and √(a²) = |a| for real a. The conditions are part of the rule.
In surd equations, use careful isolation and squaring rather than splitting sums inside a radical. An algebraically plausible but false transformation may lead to candidate values that satisfy none of the original conditions.
A good reason to substitute candidates into the original equation
After squaring, the final polynomial may have roots that satisfy the transformed equation but not the starting surd relation. Checking only the polynomial repeats the same potential error.
For √(x + 5) = x − 1, both x = 4 and x = −1 solve the squared quadratic. Only substitution into the original distinguishes equality 3 = 3 from the false relation 2 = −2.
Checking the original also catches arithmetic mistakes in sign restrictions or earlier isolation. It is the final authoritative test of whether a candidate solves the actual question.
The check should be exact where possible. A calculator approximation can be useful for a complicated root, but exact integer or surd equality gives clearer evidence in these examples.
A compact five-stage solution
First find the real domain of every radical. Second identify any non-negativity requirement on a side equated to a principal square root. Third isolate a radical and square carefully, keeping track of the sign conditions needed for the transformation.
Fourth solve the resulting equation, repeating the process only if another radical remains and another sign condition has been checked. Fifth substitute every candidate into the unsquared original equation and retain only those that satisfy it.
This sequence is short enough for a school answer yet logically complete. A student who merely lists the roots of the squared polynomial has not necessarily finished solving the surd equation.
The final answer should contain only valid real solutions and should state explicitly when no real solution exists.
Independent practice with checked answers
Question one: solve √(x + 5) = x − 1. Answer: x = 4; reject x = −1.
Question two: solve √(2x + 3) = x. Answer: x = 3; reject x = −1.
Question three: solve √(x − 2) = x − 4. Answer: x = 6; reject x = 3.
Question four: solve √(x + 4) + √(x − 1) = 5. Answer: x = 5.
Question five: solve √(x + 5) + √(x + 1) = 4. Answer: x = 5/4.
Question six: solve √(x + 9) − √x = 1. Answer: x = 16.
Question seven: solve √(x + 1) = √(2x − 1). Answer: x = 2.
For each, record domain and sign conditions before squaring. Explain why any rejected candidate fails the original equation.
Errors that need different kinds of practice
A student who consistently keeps extraneous roots may not understand that squaring loses sign information. Start with the elementary comparison 2² = (−2)² and a one-radical example with a rejected candidate.
Another student understands that principle but makes mistakes expanding (x − 1)². That calls for algebraic expansion practice, not a long explanation about radicals. A third forgets that two radicals must both have non-negative radicands, requiring a domain-intersection exercise.
These gaps should be diagnosed at their first occurrence. A changed equation after the correction tests whether the student can perform the reasoning independently.
At eduKateSG’s maximum three-student tutorial setting near Sixth Avenue MRT, with generally 1.5-hour weekly lessons where suitable spaces are available, the tutor can inspect the exact decision that caused the mistake rather than relying only on final-answer marking.
Syllabus scope and connected guides
Surds, algebraic equations and domain restrictions connect to many Additional Mathematics topics, including functions, logarithms and calculus. The precise examination demands differ by level and syllabus year, so advanced multi-radical extensions should be chosen appropriately.
Singapore families can consult SEAB’s 2026 O-Level Additional Mathematics list, 2027 SEC G3 K341 list or 2027 SEC G2 K232 list, as appropriate.
Read the companion guides How to Disprove a False Algebraic Rule with One Counterexample, How to Split an Absolute-Value Equation into Valid Cases and How to Track Domain Restrictions Through Composite Functions.
More material is available in the Additional Mathematics archive or through an eduKate Singapore parent–student consultation. The objective is not just solving a transformed polynomial, but proving that every final value satisfies the original surd equation. Properly taught kids shine a bright light into the future.