Can Converting Secant to Cosine Simplify Integration by Parts?

Join the discussion
Registration is free. Ask a follow-up in this thread, or start your own.
3 replies · 1K views
AStaunton
Messages
100
Reaction score
1
problem is to integrate the following by parts:

[tex]\int x\sec^{2}xdx[/tex]

my feeling is convert the secant term to cosine by:

[tex]sec^{2}x=cos^{-2}x\Rightarrow\int\sec^{2}xdx=\int\cos^{-2}xdx[/tex]

then:

[tex]u=\cos^{-2}x\implies du=2\sin x(\cos^{-3}x)[/tex]

and also:

[tex]dv=xdx\implies v=\frac{x^{2}}{2}[/tex]

however plugging all this into int. by parts equation ends up with:

[tex]\frac{x^{2}}{2}\cos^{-2}x-\int2\frac{x^{2}}{2}\sin x\cos^{-3}xdx[/tex]

which seems to be an even more complicated integral...

is my idea of converting the secant to cosine a good or does it make more complicated?

any advice appreciated
 
Physics news on Phys.org
While recognizing the antiderivative of sec^2(x) is the real key to this problem, many times it will make the problem easier if you can eliminate the polynomial portion by choosing that as your u.
 
osnarf said:
While recognizing the antiderivative of sec^2(x) is the real key to this problem, many times it will make the problem easier if you can eliminate the polynomial portion by choosing that as your u.
In this case: both of these ideas will work together.

[tex]\text{Let: }u=x\quad\text{ and let: }\quad dv=\sec^2(x)\,dx\ .[/tex]

This works out nicely!