When a Direct Substitution Beats Trigonometric Substitution

Look for the Derivative Inside the Numerator

A radical such as \(\sqrt{x^2-a^2}\) often suggests trigonometric substitution. That method is valuable when it turns a difficult radical into a simpler trigonometric expression. But sometimes the integrand already contains the derivative of the expression under the radical. In that case, a direct substitution can do the job with fewer steps.

The key is to look inside the radical first. If its radicand is \(q(x)\), differentiate \(q(x)\), then compare \(q'(x)\) with the numerator. The match may be exact, or it may require only a constant factor or a sign change.

The Direct-Substitution Rule

The rule is an application of the chain rule in reverse. Let \(q\) be differentiable on an interval \(I\), and suppose \(G\) is an antiderivative of \(F\) on the relevant range of \(q\). Then, wherever the expressions are defined,

In practice, the substitution is \(u=q(x)\), so \(du=q'(x)\,dx\). If the numerator is \(c\,q'(x)\), where \(c\) is a constant, take that constant outside the integral. If it is \(-q'(x)\), retain the minus sign.

For a quadratic radicand \(q(x)=Ax^2+Bx+C\), its derivative is \(q'(x)=2Ax+B\). That makes the comparison especially quick: look for a linear numerator with these same coefficients, possibly multiplied by a constant.

Why the Method Works

The substitution rule is not a separate trick: it follows directly from the chain rule. Suppose \(G'(u)=F(u)\). Differentiate the proposed antiderivative \(G(q(x))\):

This is precisely the integrand \(q'(x)F(q(x))\). Therefore \(G(q(x))+C\) is an antiderivative on any interval where the functions and derivatives involved are defined. For a square root, we normally work on an interval where \(q(x)>0\) while applying the derivative formula. An antiderivative may also extend continuously to an endpoint where \(q(x)=0\).

This reasoning explains the diagnostic test: the radicand is the natural substitution because differentiating the resulting outer function recreates the radical expression, while differentiating the radicand supplies the matching numerator.

Worked Examples

A Practical Matching Test

Before choosing trigonometric substitution for a quadratic radical, use this short test.

If the match works, the radical becomes a power of \(u\), which can be integrated directly. This can avoid extra algebra and trigonometric expressions without changing the underlying mathematics.

Common Pitfall: Losing the Sign

When the radicand is \(a^2-x^2\), its derivative is \(-2x\), not \(2x\). Ignoring this sign produces an incorrect antiderivative. For instance, consider \(\int \frac{x}{\sqrt{9-x^2}}\,dx\). A wrong calculation might claim

But differentiating this proposed answer gives

which has the wrong sign. The correct substitution is \(u=9-x^2\), so \(du=-2x\,dx\) and \(x\,dx=-\frac12du\). Thus

Indeed, differentiating the corrected answer gives \(x/\sqrt{9-x^2}\), as required. Checking by differentiation is particularly effective for catching a missing factor or sign.

When Direct Substitution Is Not the Shortcut

The presence of a quadratic radical alone does not guarantee a useful direct substitution. The numerator must supply the derivative of the radicand, up to a manageable factor. For example, in an integrand with \(\sqrt{x^2-a^2}\), the derivative of the radicand is \(2x\). If the numerator has no corresponding \(x\)-factor, substituting \(u=x^2-a^2\) also transforms \(dx\); it does not simply turn the entire integral into a power of \(u\).

In that situation, the matching-derivative shortcut is unavailable. Another method may be appropriate, including trigonometric substitution in cases where it simplifies the radical. The point is not that direct substitution replaces trigonometric substitution in general; it is that the derivative test helps identify cases where the more direct method already fits the integrand.

Why the Diagnostic Matters

Recognizing a derivative inside the numerator reduces an integration problem to the chain rule in reverse. It also makes the sign, constant factor, and domain visible before the calculation begins. Those checks help prevent avoidable errors and keep the work focused on the structure actually present in the integral.

When the substitution is appropriate, state it explicitly, transform every factor consistently, and verify the final expression by differentiation. When the derivative is absent, do not force the substitution: choose a method suited to the remaining integrand.