How to solve this integral without using substitution method? As I'm curious that is there another method to solve integral? I did integration by parts.

$$\int \sqrt{4-x^2} dx=x\sqrt{4-x^2}+\int \frac{x^2}{\sqrt{4-x^2}}dx$$ $$=x\sqrt{4-x^2}-\int \frac{4-x^2}{\sqrt{4-x^2}}dx+\int \frac{4}{\sqrt{4-x^2}}dx$$ $$=x\sqrt{4-x^2}-\int \sqrt{4-x^2}dx+4\sin ^{-1}(\frac{x}{2})$$ $$2\int\sqrt{4-x^2}dx=x\sqrt{4-x^2}+4\sin ^{-1}(\frac{x}{2})$$ $$\int \sqrt{4-x^2}dx=\frac{1}{2}x\sqrt{4-x^2}+2\sin ^{-1}(\frac{x}{2})$$

Is there any other method can solve integral other than substitution and this? I think Riemann Sum also can be used to solved. But people riemann sum is not considered a method of integration. I wonder why. Thanks a lot.

  • $\begingroup$ By knowing what integrals of $\sqrt{a^2-x^2}$ are, allowed ;)? $\endgroup$
    – Chinny84
    Mar 6 '15 at 15:00
  • $\begingroup$ Please edit your question to include a description of what you've tried and where you're stuck. In this case, why are you disallowing substitutions? $\endgroup$ Mar 6 '15 at 15:00
  • 2
    $\begingroup$ Why do you want to do this? If you had $\int_0^{2}\sqrt{4-x^2}\,\mathrm{d}x = \pi$, since it is part of a circle with radius $2$. Otherwise you need to do some work to figure out the integral, where substitution is the easiest. $\endgroup$ Mar 6 '15 at 15:00
  • $\begingroup$ @N3buchadnezzar, actually there is a very nice geometric solution, fleshing out your observation about the definite integral, that avoids substitution. I would say more, but I think the OP needs to say more first. $\endgroup$ Mar 6 '15 at 15:22
  • $\begingroup$ Yeah! I see it now as well using a drawing! clever However I think I will let my answer be until OP says more. $\endgroup$ Mar 6 '15 at 15:32

Method 1: Integration by parts

$$ \int \sqrt{a^2-x^2\,} \,\mathrm{d}x = \frac{x}{2}\sqrt{a^2-x^2\,} + \frac{a}{2} \int \frac{a}{\sqrt{a^2-x^2\,}\,}\mathrm{d}x \tag{1} $$

Pick $v' = x$ and $u = \sqrt{a^2-x^2}$ then solve with respect to $\int \sqrt{a^2-x^2\,} \,\mathrm{d}x$. Next part is to remember the derivative of the inverse sine function. $$ \frac{\mathrm{d}}{\mathrm{d}x} \arcsin x = \frac{1}{\sqrt{1-x^2\,}\,} \tag{2} $$

Hence by using the chain rule, you can show that

$$ \frac{\mathrm{d}}{\mathrm{d}x} \arcsin \left( \frac{x}{a} \right) = \frac{a}{\sqrt{a^2-x^2\,}\,} \tag{3} $$

Integrating $(3)$ w.r.t $x$ and then inserting the result into $(1)$ completes the calculations. I will leave it to you to fill in the details. Just ask if any part were particularly confusing.

Method 2: Geometric considerations

here is a proof without words. I was able to discover this through the help of Barry Cipra in the comments above.

enter image description here

$$ \int_0^x \sqrt{a^2-t^2\,} \,\mathrm{d}t = \color{blue}{\frac{x}{2}\sqrt{a^2-x^2\,}} + \color{green}{\frac{a^2}{2} \arcsin \left(\frac{x}{a}\right)} $$

Be warned, spoilers ahead.

The total area of a circle is $\pi a^2$. The area of the sector can be obtained by multiplying the circle's area by the ratio of the angle and $2 \pi$ (because the area of the sector is proportional to the angle, and $2 \pi$ is the angle for the whole circle, in radians): $$ A = \pi a^2 \cdot \frac{\theta}{2 \pi} = \frac{a^2}{2} \theta $$ The rest follows since $\sin \theta = x/a$, where $a$ of course is the hypotenuse since it is the radius of the circle $x^2 + y^2 = a^2$. The area of the triangle should be straigt forward to figure out.

  • $\begingroup$ what happens if $x < 0?$ $\endgroup$
    – abel
    Mar 8 '15 at 14:33
  • $\begingroup$ you would then mirror the figure, and the calculations would be identical =) $\endgroup$ Mar 8 '15 at 14:35
  • $\begingroup$ wont you get negative areas? $\endgroup$
    – abel
    Mar 8 '15 at 14:38
  • $\begingroup$ You are still above the x-axis. You would get $\int_0^{-x} \sqrt{a^2-t^2\,} \,\mathrm{d}t = \color{blue}{\frac{-x}{2}\sqrt{a^2-(-x)^2\,}} + \color{green}{\frac{a^2}{2} \arcsin \left(\frac{-x}{a}\right)}$ Because $x$ is now negative.. Is this suprising to you? $\endgroup$ Mar 8 '15 at 14:45
  • $\begingroup$ if $x < 0, $ then you will need $\pi + \sin^{-1}(x)$ for the angle of the sector. it includes the quarter of a circle in addition to the triangle. you will have to argue much more carefully. but you can avoid all this if you keep $x \ge 0$ and deal with $x < 0$ by some sort of symmetry. $\endgroup$
    – abel
    Mar 8 '15 at 14:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.