Class 12 Maths, Worked in Full Past papersGet updates
Chapter 7 · Integrals

Example 42: ∫0πx dxa2cos⁡2x+b2sin⁡2x\int_0^{\pi}\frac{x\,dx}{a^2\cos^2x+b^2\sin^2x}

ExamplesProperties of definite integrals (P4, P6, P2) · Miscellaneous Examples
The question

Evaluate

∫0πx dxa2cos⁡2x+b2sin⁡2x\int_0^{\pi}\frac{x\,dx}{a^2\cos^2 x + b^2\sin^2 x}
Exam answer

Let

I=∫0πx dxa2cos⁡2x+b2sin⁡2xI = \int_0^{\pi}\frac{x\,dx}{a^2\cos^2 x + b^2\sin^2 x}

Using ∫0πf(x) dx=∫0πf(π−x) dx\displaystyle\int_0^{\pi} f(x)\,dx = \int_0^{\pi} f(\pi - x)\,dx⁠, with cos⁡2(π−x)=cos⁡2x\cos^2(\pi - x) = \cos^2 x and sin⁡2(π−x)=sin⁡2x\sin^2(\pi - x) = \sin^2 x⁠,

I=∫0π(π−x) dxa2cos⁡2x+b2sin⁡2xI = \int_0^{\pi}\frac{(\pi - x)\,dx}{a^2\cos^2 x + b^2\sin^2 x}

Adding the two expressions for II⁠,

2I=π∫0πdxa2cos⁡2x+b2sin⁡2x2I = \pi\int_0^{\pi}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x}

Using ∫0πf(x) dx=2∫0π2f(x) dx\displaystyle\int_0^{\pi} f(x)\,dx = 2\int_0^{\frac{\pi}{2}} f(x)\,dx when f(π−x)=f(x)f(\pi - x) = f(x)⁠,

2I=2π∫0π2dxa2cos⁡2x+b2sin⁡2x2I = 2\pi\int_0^{\frac{\pi}{2}}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x}

Splitting the interval at π4\frac{\pi}{4}⁠,

I=π∫0π4dxa2cos⁡2x+b2sin⁡2x+π∫π4π2dxa2cos⁡2x+b2sin⁡2x=π(I1+I2)\begin{aligned} I &= \pi\int_0^{\frac{\pi}{4}}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x} \\ &\quad + \pi\int_{\frac{\pi}{4}}^{\frac{\pi}{2}}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x} \\ &= \pi\left(I_1 + I_2\right) \end{aligned}

In I1I_1⁠, divide the numerator and denominator by cos⁡2x\cos^2 x⁠. Put tan⁡x=t\tan x = t⁠. Then sec⁡2x dx=dt\sec^2 x\,dx = dt⁠.

When x=0x = 0⁠, t=0t = 0⁠; when x=π4x = \frac{\pi}{4}⁠, t=1t = 1⁠.

I1=∫0π4sec⁡2x dxa2+b2tan⁡2x=∫01dta2+b2t2=1b2∫01dtt2+(ab)2=1b2⋅ba[tan⁡−1bta]01=1abtan⁡−1ba\begin{aligned} I_1 &= \int_0^{\frac{\pi}{4}}\frac{\sec^2 x\,dx}{a^2 + b^2\tan^2 x} \\ &= \int_0^1\frac{dt}{a^2 + b^2t^2} \\ &= \frac{1}{b^2}\int_0^1\frac{dt}{t^2 + \left(\frac{a}{b}\right)^2} \\ &= \frac{1}{b^2}\cdot\frac{b}{a}\Big[\tan^{-1}\frac{bt}{a}\Big]_0^1 \\ &= \frac{1}{ab}\tan^{-1}\frac{b}{a} \end{aligned}

In I2I_2⁠, divide the numerator and denominator by sin⁡2x\sin^2 x⁠. Put cot⁡x=u\cot x = u⁠. Then −cosec⁡2x dx=du-\operatorname{cosec}^2 x\,dx = du⁠.

When x=π4x = \frac{\pi}{4}⁠, u=1u = 1⁠; when x=π2x = \frac{\pi}{2}⁠, u=0u = 0⁠.

I2=∫π4π2cosec⁡2x dxa2cot⁡2x+b2=−∫10dua2u2+b2\begin{aligned} I_2 &= \int_{\frac{\pi}{4}}^{\frac{\pi}{2}}\frac{\operatorname{cosec}^2 x\,dx}{a^2\cot^2 x + b^2} \\ &= -\int_1^0\frac{du}{a^2u^2 + b^2} \end{aligned}

Changing the sign and swapping the limits,

I2=∫01dua2u2+b2=1a2∫01duu2+(ba)2=1a2⋅ab[tan⁡−1aub]01=1abtan⁡−1ab\begin{aligned} I_2 &= \int_0^1\frac{du}{a^2u^2 + b^2} \\ &= \frac{1}{a^2}\int_0^1\frac{du}{u^2 + \left(\frac{b}{a}\right)^2} \\ &= \frac{1}{a^2}\cdot\frac{a}{b}\Big[\tan^{-1}\frac{au}{b}\Big]_0^1 \\ &= \frac{1}{ab}\tan^{-1}\frac{a}{b} \end{aligned} ∴ I=πab(tan⁡−1ba+tan⁡−1ab)\therefore\ I = \frac{\pi}{ab}\left(\tan^{-1}\frac{b}{a} + \tan^{-1}\frac{a}{b}\right)

For ab>0ab > 0⁠, ba>0\frac{b}{a} > 0 and ab\frac{a}{b} is its reciprocal, so tan⁡−1ba+tan⁡−1ab=π2\tan^{-1}\frac{b}{a} + \tan^{-1}\frac{a}{b} = \frac{\pi}{2}⁠.

∴ I=πab⋅π2=π22ab\therefore\ I = \dfrac{\pi}{ab}\cdot\dfrac{\pi}{2} = \dfrac{\pi^2}{2ab} (for ab>0ab > 0⁠).

The book does not say so, but its answer π22ab\frac{\pi^2}{2ab} is correct only when ab>0ab > 0⁠; in general the integral equals π22∣a∣∣b∣\frac{\pi^2}{2|a||b|}⁠.

Want the reason behind every step, the basics and the checks?

The one idea this problem is built on

The factor xx in the numerator is the obstacle. Property P4 (replace xx by π−x\pi - x⁠) leaves the denominator unchanged but turns xx into π−x\pi - x⁠; adding the two forms, the xx's cancel and the numerator becomes the constant π\pi⁠. What remains, ∫dxa2cos⁡2x+b2sin⁡2x\int\frac{dx}{a^2\cos^2 x + b^2\sin^2 x}⁠, is handled by dividing by cos⁡2x\cos^2 x (to get tan⁡x\tan x⁠) on [0,π4][0, \frac{\pi}{4}] and by sin⁡2x\sin^2 x (to get cot⁡x\cot x⁠) on [π4,π2][\frac{\pi}{4}, \frac{\pi}{2}]⁠.

Given, to find, and the plan
Given
I=∫0πx dxa2cos⁡2x+b2sin⁡2xI = \displaystyle\int_0^{\pi}\frac{x\,dx}{a^2\cos^2 x + b^2\sin^2 x} with non-zero constants a,ba, b⁠. The book does not say so, but its answer π22ab\frac{\pi^2}{2ab} is correct only when aa and bb have the same sign (ab>0ab > 0⁠); for ab<0ab < 0 it would be negative, although the integrand is positive. We take a>0a > 0 and b>0b > 0⁠, the natural reading.
To find
The value of II in terms of aa and bb⁠.
Plan
(1) Continuity. (2) P4 and add: I=π2∫0πgI = \frac{\pi}{2}\int_0^\pi g⁠. (3) P6: I=π∫0π/2gI = \pi\int_0^{\pi/2} g⁠. (4) Split at π4\frac{\pi}{4} by P2. (5) and (6) Substitute t=tan⁡xt = \tan x and u=cot⁡xu = \cot x⁠. (7) Add, using tan⁡−1ba+tan⁡−1ab=π2\tan^{-1}\frac{b}{a} + \tan^{-1}\frac{a}{b} = \frac{\pi}{2}⁠.

Basics you need

1P2, P4, P6 (Section 7.10)

P2: ∫pq=∫pr+∫rq\int_p^q = \int_p^r + \int_r^q⁠. P4: ∫0cf(x) dx=∫0cf(c−x) dx\int_0^c f(x)\,dx = \int_0^c f(c - x)\,dx⁠. P6: ∫02cf(x) dx=2∫0cf(x) dx\int_0^{2c} f(x)\,dx = 2\int_0^c f(x)\,dx if f(2c−x)=f(x)f(2c - x) = f(x)⁠. (The book writes aa for our cc⁠; we rename it because aa is already used in this problem.) Lesson: properties of definite integrals

2Reflections of angles

sin⁡(π−x)=sin⁡x\sin(\pi - x) = \sin x and cos⁡(π−x)=−cos⁡x\cos(\pi - x) = -\cos x⁠, so sin⁡2(π−x)=sin⁡2x\sin^2(\pi - x) = \sin^2 x and cos⁡2(π−x)=cos⁡2x\cos^2(\pi - x) = \cos^2 x⁠. Basics: symmetry and mirror images

3Section 7.4, formula (3)

∫dtt2+k2=1ktan⁡−1tk+C\int\frac{dt}{t^2 + k^2} = \frac{1}{k}\tan^{-1}\frac{t}{k} + C⁠, k>0k > 0⁠. Lesson: integrals of particular functions

4Complementary inverse tangents

For p>0p > 0⁠: tan⁡−1p+tan⁡−11p=π2\tan^{-1}p + \tan^{-1}\frac1p = \frac{\pi}{2}⁠. Proof: θ=tan⁡−1p\theta = \tan^{-1}p lies in (0,π2)(0, \frac{\pi}{2}) because p>0p > 0 (principal range of tan⁡−1\tan^{-1}⁠, Chapter 2). Then π2−θ\frac{\pi}{2} - \theta also lies in (0,π2)(0, \frac{\pi}{2})⁠, and tan⁡(π2−θ)=cot⁡θ=1p\tan\left(\frac{\pi}{2} - \theta\right) = \cot\theta = \frac1p⁠. So tan⁡−11p=π2−θ\tan^{-1}\frac1p = \frac{\pi}{2} - \theta⁠. (In a right triangle with legs aa and bb⁠, the two acute angles have tangents ba\frac{b}{a} and ab\frac{a}{b}⁠.) Basics: trigonometric identities

Solution

  1. The integrand is continuous on [0,π][0, \pi]

    Write D(x)=a2cos⁡2x+b2sin⁡2xD(x) = a^2\cos^2 x + b^2\sin^2 x⁠. Since cos⁡2x+sin⁡2x=1\cos^2 x + \sin^2 x = 1⁠,

    D(x)≥min⁡(a2,b2) (cos⁡2x+sin⁡2x)=min⁡(a2,b2)>0.D(x) \ge \min(a^2, b^2)\,(\cos^2 x + \sin^2 x) = \min(a^2, b^2) > 0 .

    So DD never vanishes, and xD(x)\frac{x}{D(x)} and 1D(x)\frac{1}{D(x)} are continuous on all of [0,π][0, \pi]⁠.

    This is why aa and bb must both be non-zero: if, say, b=0b = 0⁠, then D(x)=a2cos⁡2xD(x) = a^2\cos^2 x is 0 at x=π2x = \frac{\pi}{2} and the integral is not a proper one. With continuity, all the properties and Theorem 2 apply.
    Each of a2cos⁡2xa^2\cos^2 x and b2sin⁡2xb^2\sin^2 x is at least min⁡(a2,b2)\min(a^2, b^2) times cos⁡2x\cos^2 x or sin⁡2x\sin^2 x respectively.
  2. P4 removes the xx from the numerator

    P4 with c=πc = \pi replaces xx by π−x\pi - x⁠. The denominator does not change, because cos⁡2(π−x)=cos⁡2x\cos^2(\pi - x) = \cos^2 x and sin⁡2(π−x)=sin⁡2x\sin^2(\pi - x) = \sin^2 x⁠:

    I=∫0π(π−x) dxa2cos⁡2x+b2sin⁡2x=π∫0πdxD(x)−∫0πx dxD(x)=π∫0πdxD(x)−I.\begin{aligned}I &= \int_0^{\pi}\frac{(\pi - x)\,dx}{a^2\cos^2 x + b^2\sin^2 x}\\ &= \pi\int_0^{\pi}\frac{dx}{D(x)} - \int_0^{\pi}\frac{x\,dx}{D(x)}\\ &= \pi\int_0^{\pi}\frac{dx}{D(x)} - I .\end{aligned}

    Adding II to both sides and dividing by 2:

    I=π2∫0πdxa2cos⁡2x+b2sin⁡2x.I = \frac{\pi}{2}\int_0^{\pi}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x} .
    The second integral is II itself, so we can move it to the left: 2I=π∫0πdxD2I = \pi\int_0^\pi\frac{dx}{D}⁠. The figure in Step 7 shows the idea: xD\frac{x}{D} and π−xD\frac{\pi - x}{D} are mirror images with equal areas, and they add up to πD\frac{\pi}{D}⁠, so each has half the area under πD\frac{\pi}{D}⁠.
    P4 works because the interval is [0,c][0, c] and x↦c−xx \mapsto c - x maps it onto itself. Basics: symmetry and mirror images
  3. P6 halves the interval

    Let g(x)=1D(x)g(x) = \frac{1}{D(x)}⁠. With 2c=π2c = \pi⁠, c=π2c = \frac{\pi}{2}⁠, the condition of P6 holds, since g(π−x)=g(x)g(\pi - x) = g(x) (the same reflection facts as in Step 2). So

    ∫0πg(x) dx=2∫0π/2g(x) dx,\int_0^{\pi}g(x)\,dx = 2\int_0^{\pi/2}g(x)\,dx ,

    and therefore

    I=π∫0π/2dxa2cos⁡2x+b2sin⁡2x.I = \pi\int_0^{\pi/2}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x} .
    The graph of gg on [0,π][0, \pi] is symmetric about x=π2x = \frac{\pi}{2}⁠, so its two halves have equal areas. Working on [0,π2][0, \frac{\pi}{2}] is needed because the substitutions ahead use tan⁡x\tan x and cot⁡x\cot x⁠, which behave well only inside one quadrant.
    π2⋅2=π\frac{\pi}{2}\cdot 2 = \pi⁠.
  4. Split at π4\frac{\pi}{4} by P2

    I=π[∫0π/4dxD(x)+∫π/4π/2dxD(x)].I = \pi\left[\int_0^{\pi/4}\frac{dx}{D(x)} + \int_{\pi/4}^{\pi/2}\frac{dx}{D(x)}\right] .
    Next we want to divide by cos⁡2x\cos^2 x to bring in tan⁡x\tan x⁠. But cos⁡x=0\cos x = 0 at π2\frac{\pi}{2}⁠, so that division is not allowed at the upper end. Dividing by sin⁡2x\sin^2 x instead (to bring in cot⁡x\cot x⁠) fails at 0. Splitting at π4\frac{\pi}{4} lets us use each division where it is safe: on [0,π4][0, \frac{\pi}{4}]⁠, cos⁡x≥12\cos x \ge \frac{1}{\sqrt2}⁠; on [π4,π2][\frac{\pi}{4}, \frac{\pi}{2}]⁠, sin⁡x≥12\sin x \ge \frac{1}{\sqrt2}⁠. Any split point strictly between 0 and π2\frac{\pi}{2} would work; π4\frac{\pi}{4} gives the neat limit 1 for both tan⁡\tan and cot⁡\cot⁠.
  5. The first piece: divide by cos⁡2x\cos^2 x⁠, put t=tan⁡xt = \tan x

    Dividing top and bottom by cos⁡2x\cos^2 x⁠:

    ∫0π/4dxa2cos⁡2x+b2sin⁡2x=∫0π/4sec⁡2x dxa2+b2tan⁡2x.\begin{aligned}&\int_0^{\pi/4}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x}\\ &= \int_0^{\pi/4}\frac{\sec^2 x\,dx}{a^2 + b^2\tan^2 x} .\end{aligned}

    Put t=tan⁡xt = \tan x⁠, dt=sec⁡2x dxdt = \sec^2 x\,dx⁠; x=0⇒t=0x = 0 \Rightarrow t = 0⁠, x=π4⇒t=1x = \frac{\pi}{4} \Rightarrow t = 1⁠:

    ∫01dta2+b2t2=1b2∫01dtt2+(a/b)2=1b2⋅ba[tan⁡−1bta]01=1abtan⁡−1ba.\begin{aligned}\int_0^1\frac{dt}{a^2 + b^2t^2} &= \frac{1}{b^2}\int_0^1\frac{dt}{t^2 + (a/b)^2}\\ &= \frac{1}{b^2}\cdot\frac{b}{a}\Big[\tan^{-1}\frac{bt}{a}\Big]_0^1 = \frac{1}{ab}\tan^{-1}\frac{b}{a} .\end{aligned}
    tan⁡x\tan x is increasing and differentiable on [0,π4][0, \frac{\pi}{4}]⁠, so the substitution is valid on the whole piece. Section 7.4, formula (3) with k=abk = \frac{a}{b}⁠: 1ktan⁡−1tk=batan⁡−1bta\frac{1}{k}\tan^{-1}\frac{t}{k} = \frac{b}{a}\tan^{-1}\frac{bt}{a}⁠. At t=0t = 0⁠, tan⁡−10=0\tan^{-1}0 = 0⁠.
    cos⁡2xcos⁡2x=1\frac{\cos^2 x}{\cos^2 x} = 1⁠, sin⁡2xcos⁡2x=tan⁡2x\frac{\sin^2 x}{\cos^2 x} = \tan^2 x⁠, 1cos⁡2x=sec⁡2x\frac{1}{\cos^2 x} = \sec^2 x⁠. Basics: substitution
  6. The second piece: divide by sin⁡2x\sin^2 x⁠, put u=cot⁡xu = \cot x

    Dividing top and bottom by sin⁡2x\sin^2 x⁠:

    ∫π/4π/2dxa2cos⁡2x+b2sin⁡2x=∫π/4π/2cosec⁡2x dxa2cot⁡2x+b2.\begin{aligned}&\int_{\pi/4}^{\pi/2}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x}\\ &= \int_{\pi/4}^{\pi/2}\frac{\operatorname{cosec}^2 x\,dx}{a^2\cot^2 x + b^2} .\end{aligned}

    Put u=cot⁡xu = \cot x⁠, du=−cosec⁡2x dxdu = -\operatorname{cosec}^2 x\,dx⁠; x=π4⇒u=1x = \frac{\pi}{4} \Rightarrow u = 1⁠, x=π2⇒u=0x = \frac{\pi}{2} \Rightarrow u = 0⁠:

    −∫10dua2u2+b2=∫01dua2u2+b2=1ab[tan⁡−1aub]01=1abtan⁡−1ab.\begin{aligned}-\int_1^0\frac{du}{a^2u^2 + b^2} &= \int_0^1\frac{du}{a^2u^2 + b^2}\\ &= \frac{1}{ab}\Big[\tan^{-1}\frac{au}{b}\Big]_0^1 = \frac{1}{ab}\tan^{-1}\frac{a}{b} .\end{aligned}
    The minus sign from dudu is absorbed by swapping the limits (P1). The integral is the same as in Step 5 with aa and bb interchanged, so the answer is too.
    ddxcot⁡x=−cosec⁡2x\frac{d}{dx}\cot x = -\operatorname{cosec}^2 x⁠; cot⁡π4=1\cot\frac{\pi}{4} = 1⁠, cot⁡π2=0\cot\frac{\pi}{2} = 0⁠.
  7. Add and simplify

    I=π⋅1ab[tan⁡−1ba+tan⁡−1ab]=πab⋅π2=π22ab.I = \pi\cdot\frac{1}{ab}\left[\tan^{-1}\frac{b}{a} + \tan^{-1}\frac{a}{b}\right] = \frac{\pi}{ab}\cdot\frac{\pi}{2} = \frac{\pi^2}{2ab} .
    Since a,b>0a, b > 0⁠, p=bap = \frac{b}{a} is positive and ab=1p\frac{a}{b} = \frac1p⁠, so Basic 4 gives tan⁡−1ba+tan⁡−1ab=π2\tan^{-1}\frac{b}{a} + \tan^{-1}\frac{a}{b} = \frac{\pi}{2}⁠. (If pp were negative, the sum would be −π2-\frac{\pi}{2}⁠.)
    x y O π/4 π/2 3π/4 π 1 2 3 a = 1, b = 2: D(x) = cos²x + 4 sin²x blue: y = x/D (shaded, area I); dashed gold: y = (π − x)/D violet: y = π/D = blue + gold; I = π²/4 ≈ 2.4674
    Drawn for a=1a = 1⁠, b=2b = 2⁠. The blue curve y=xD(x)y = \frac{x}{D(x)} and the dashed gold curve y=π−xD(x)y = \frac{\pi - x}{D(x)} are mirror images in the line x=π2x = \frac{\pi}{2}⁠, so they enclose equal areas. Their sum is the violet curve y=πD(x)y = \frac{\pi}{D(x)}⁠. So the shaded area II is half the area under the violet curve: Step 2.
Result

∫0πx dxa2cos⁡2x+b2sin⁡2x=π22ab(a,b>0)\int_0^{\pi}\frac{x\,dx}{a^2\cos^2 x + b^2\sin^2 x} = \frac{\pi^2}{2ab} \qquad (a, b > 0) This is the textbook's answer. For any non-zero a,ba, b the integral depends only on a2a^2 and b2b^2⁠, so in general it equals π22∣a∣∣b∣\frac{\pi^2}{2|a||b|}⁠; this agrees with π22ab\frac{\pi^2}{2ab} exactly when ab>0ab > 0⁠.

Check it with numbers

For a=1a = 1⁠, b=2b = 2⁠: numerical integration gives 2.46740110032.4674011003 to ten decimal places, and π24≈2.4674011003\frac{\pi^2}{4} \approx 2.4674011003⁠. The two pieces of Step 4 are 12tan⁡−12≈0.553574\frac12\tan^{-1}2 \approx 0.553574 and 12tan⁡−112≈0.231824\frac12\tan^{-1}\frac12 \approx 0.231824 (numerical integration agrees with both); their sum 0.785398≈π40.785398 \approx \frac{\pi}{4}⁠.

For a=b=1a = b = 1 the integrand is just xx⁠, and ∫0πx dx=π22\int_0^\pi x\,dx = \frac{\pi^2}{2}⁠, which matches π22ab\frac{\pi^2}{2ab}⁠. For a=3a = 3⁠, b=12b = \frac12⁠: numerical value 3.2898683.289868⁠, and π23≈3.289868\frac{\pi^2}{3} \approx 3.289868⁠.

Common mistakes
  1. Dividing by cos⁡2x\cos^2 x over all of [0,π2][0, \frac{\pi}{2}]⁠. Then the substitution t=tan⁡xt = \tan x has upper limit tan⁡π2\tan\frac{\pi}{2}⁠, which does not exist. Split first, or treat it as an improper integral.
  2. Forgetting that P4 leaves the denominator unchanged only because of the squares. cos⁡(π−x)=−cos⁡x\cos(\pi - x) = -\cos x⁠, but cos⁡2(π−x)=cos⁡2x\cos^2(\pi - x) = \cos^2 x⁠.
  3. Losing the factor π\pi that P4 brings out, or the 2 from P6. Keep track: π2⋅2=π\frac{\pi}{2}\cdot 2 = \pi⁠.
  4. Mixing up the property's letter aa with the constant aa of the question. Here P4 is used with upper limit π\pi⁠.
What to take away
  1. If f(c−x)=f(x)f(c - x) = f(x)⁠, then ∫0cx f(x) dx=c2∫0cf(x) dx\int_0^c x\,f(x)\,dx = \frac{c}{2}\int_0^c f(x)\,dx⁠. This "remove the xx" trick is P4 plus adding.
  2. For 1a2cos⁡2x+b2sin⁡2x\frac{1}{a^2\cos^2 x + b^2\sin^2 x}⁠, divide by cos⁡2x\cos^2 x to make sec⁡2x\sec^2 x and tan⁡x\tan x appear; split the interval if cos⁡x\cos x vanishes in it.
  3. ∫0π/2dxa2cos⁡2x+b2sin⁡2x=π2ab\int_0^{\pi/2}\frac{dx}{a^2\cos^2 x + b^2\sin^2 x} = \frac{\pi}{2ab} for a,b>0a, b > 0⁠: worth remembering.

Every fact used, and where it comes from

FactUsed inSource
Continuity of the integrandStep 1Remark (iv) after Theorem 2, Section 7.8.3
P4, P6, P2, P1Steps 2, 3, 4, 6Section 7.10 (Lesson: properties)
sin⁡2(π−x)=sin⁡2x\sin^2(\pi - x) = \sin^2 x⁠, cos⁡2(π−x)=cos⁡2x\cos^2(\pi - x) = \cos^2 xSteps 2, 3Class XI trigonometry
Substitutions t=tan⁡xt = \tan x⁠, u=cot⁡xu = \cot x with new limitsSteps 5, 6Section 7.9
∫dtt2+k2=1ktan⁡−1tk+C\int\frac{dt}{t^2 + k^2} = \frac1k\tan^{-1}\frac{t}{k} + CSteps 5, 6Section 7.4, formula (3) (Lesson: particular functions)
tan⁡−1p+tan⁡−11p=π2\tan^{-1}p + \tan^{-1}\frac1p = \frac{\pi}{2} for p>0p > 0Step 7Proved in Basic 4 from the principal range of tan⁡−1\tan^{-1} (Chapter 2); the identity itself is not stated in this reprint
Found an error, or want more?

Tell us if something on this page is wrong or unclear, or if you would like a fuller explanation, another method or a new feature. Send feedback on this problem →