You already integrate sinx and cosx. Hyperbolic integration is the same idea with a different family: sinhx, coshx, tanhx, and their reciprocals.
The name comes from geometry: just as cost,sint trace a circle (x2+y2=1), cosht,sinht trace a hyperbola (x2−y2=1). The integration rules are almost identical to the trigonometric ones, with a few sign changes.
The core definitions
In terms of exponentials:
sinhx=2ex−e−x,coshx=2ex+e−x,tanhx=coshxsinhx
From these come the derivatives:
dxdsinhx=coshx,dxdcoshx=sinhx,dxdtanhx=sech2x
Notice the derivative of coshx is +sinhx (not −sinhx as in trigonometry). That plus sign is the only real difference from the circular case.
The integration formulas
Reversing the derivatives:
∫sinhxdx=coshx+C
∫coshxdx=sinhx+C
∫sech2xdx=tanhx+C
∫csch2xdx=−cothx+C
∫sechxtanhxdx=−sechx+C
∫cschxcothxdx=−cschx+C
Why the sign difference matters
Watch out
Don't treat ∫sinhxdx like ∫sinxdx. ∫sinxdx=−cosx+C, but ∫sinhxdx=+coshx+C — the minus sign is gone.
Check it: differentiate coshx and you get sinhx, not −sinhx, so the integral must be positive.
A worked example
Find ∫(3sinhx−2coshx)dx.
=3∫sinhxdx−2∫coshxdx=3coshx−2sinhx+C
When you use it in exams
Direct integration — apply the standard formulas above.
Substitution — a messy integral like ∫x2+a2dx becomes clean with x=asinht or x=acosht. That's a separate technique, but it relies on these basic integrals. …
The integral ∫ex+e−xdx simplifies by rewriting the denominator as 2coshx, then substituting t=ex to get a standard arctangent form. The correct answer is tan−1(ex)+C, which is option (A).
The key insight here is that the integrand ex+e−x1 looks like a hyperbolic secant function — because ex+e−x=2coshx, so the integrand is 21sech x. But the direct hyperbolic route isn't the simplest. Instead, notice that the denominator is symmetric in ex and e−x, which suggests a substitution that "breaks" this symmetry: let t=ex. This turns the integral into a rational function of t, which is a standard technique for integrals involving exponentials.
Rewrite the integrand
Multiply numerator and denominator by ex to clear the negative exponent:
∫ex+e−xdx=∫e2x+1exdx.
This step is crucial — it transforms the denominator into a simple quadratic in ex.
Substitute t=ex
Then dt=exdx, so the numerator exdx becomes exactly dt. The integral becomes:
Mistake 1: Guessing log(ex+e−x) by the ff′ pattern.
Why it's wrong: dxd(ex+e−x)=ex−e−x, not 1, so the numerator is not the denominator's derivative — the log form is wrong (that is option D, a distractor). Correct approach: multiply by ex and substitute t=ex.
Same / Similar Concept — real previous-year questions on the same or a closely similar concept, not this exact question.
CBSE 2023Set 65/3/13 marks
Q.Evaluate : ∫1e4x2−(xlogx)21dx
›Reveal solutionSolution
The integral simplifies by factoring x2 and substituting u=logx, transforming it into a standard inverse trigonometric integral of the form ∫a2−u21du, which evaluates to 6π.
When faced with a complex integral, the first step is almost always to simplify the integrand. Here, we have terms involving x2 and (xlogx)2 under a square root. Our goal is to manipulate this expression into a recognizable standard form.
Notice the structure inside the square root: 4x2−(xlogx)2. Both terms contain x2. This is a strong hint to factor out x2.
4x2−(xlogx)2=x2(4−(logx)2)
Since the integration limits are from 1 to e, x is always positive, so x2=x.
x2(4−(logx)2)=x4−(logx)2
Now the integral becomes:
∫1ex4−(logx)21dx
The presence of logx and x1dx is a classic indicator for a substitution. Let u=logx. Then du=x1dx. This substitution will transform the integral into a much simpler form.
The resulting integral will be of the form ∫a2−u21du. This is a standard integral that evaluates to arcsin(au). It's important to distinguish this from hyperbolic integrals. Hyperbolic inverse functions (like arsinh or arccosh) arise from integrals involving u2+a2 or u2−a2, respectively. Our form a2−u2 is distinctly trigonometric.
Let's work through the steps:
Simplify the integrand:
We begin by simplifying the expression under the square root.
4x2−(xlogx)2=x2(4−(logx)2)
Since $x$ is in the interval $[1, e]$, $x$ is positive. Therefore, $\sqrt{x^2} = x$.
4x2−(xlogx)2=x4−(logx)2
Substituting this back into the integral, we get:
∫1ex4−(logx)21dx
Perform a substitution:
The structure of the integrand, with logx and x1dx, strongly suggests the substitution u=logx.
Let u=logx.
Then, differentiating both sides with respect to x, we get du=x1dx.
We also need to change the limits of integration according to our substitution:
When x=1, u=log1=0.
When x=e, u=loge=1.
Transform the integral:
Substitute u and du into the integral with the new limits:
The key idea is to rationalise the integrand using the substitution x=asecθ, which converts the square root into a pure trigonometric form. The integral then simplifies to ∫(secθ+1)dθ, and after back-substitution, the answer is logx+x2−a2+sec−1(ax)+C.
The problem asks for ∫x1x−ax+adx. At first glance, the square root of a rational function suggests a substitution that will remove the radical. The expression inside the root, x−ax+a, is reminiscent of trigonometric identities — specifically, the secant function.
Recall that sec2θ−1=tan2θ. If we set x=asecθ, then x2−a2=a2tan2θ, and the square root will simplify nicely. This is a classic technique for integrals involving x2−a2 or similar forms.
Let's walk through it step by step.
Choose the substitution.
Let x=asecθ, where a>0 (we can handle sign later). Then dx=asecθtanθdθ.
The domain: x>a or x<−a; we'll assume x>a for the principal branch, and the constant C will absorb sign adjustments.
Rewrite the square root.
Compute x−ax+a=asecθ−aasecθ+a=secθ−1secθ+1.
Multiply numerator and denominator by secθ+1 to rationalise:
secθ−1secθ+1=sec2θ−1(secθ+1)2=tan2θ(secθ+1)2.
Hence x−ax+a=∣tanθ∣∣secθ+1∣. For θ∈(0,π/2), secθ>0, tanθ>0, so the absolute values drop:
x−ax+a=tanθsecθ+1.
Substitute into the integral.
The integrand becomes:
x1x−ax+adx=asecθ1⋅tanθsecθ+1⋅asecθtanθdθ.
Cancel a, secθ, and tanθ:
=(secθ+1)dθ.
Integrate.
So the integral reduces to:
∫(secθ+1)dθ=∫secθdθ+∫1dθ.
We know ∫secθdθ=log∣secθ+tanθ∣+C, and ∫1dθ=θ+C.
Thus:
∫x1x−ax+adx=log∣secθ+tanθ∣+θ+C.
Back-substitute to x.
Since x=asecθ, we have secθ=ax. …
The integral simplifies using hyperbolic identities to 21∫csch(2x)dx, and evaluating from log2 to log3 gives 41ln23.
The key here is to notice that the denominator is a product of two expressions that look like 2coshx and 2sinhx — but without the factor 2. Let’s rewrite:
ex+e−x=2coshx,ex−e−x=2sinhx.
So the integrand becomes:
(2coshx)(2sinhx)1=4sinhxcoshx1.
Now recall the double-angle identity: 2sinhxcoshx=sinh2x. Therefore: