# Integral questionWatch

Announcements
#1
Completely stuck on this one. I tried to solve the integral using the substitution t = e-x but I just got lost in my own workings.

Any help would be much appreciated

0
13 years ago
#2
f(x) = Sqrt[e^(-x) - e^(-2x)] has a defined limit as x -> Inf (zero) and is also defined at the lower limit x = 0. Hence, Int (0 to Inf) f(x) dx will have a defined value and thus is convergent.

I = Int (0 to Inf) Sqrt[e^(-x) - e^(-2x)] dx
= Int (0 to Inf) e^(-x/2).Sqrt[1 - e^(-x)] dx

Let u = e^(-x/2) = > du/dx = -(1/2)e^(-x/2) => dx = -2e^(x/2)
x = 0 => u = 1/e^0 = 1
x = Inf => u = 1/Inf = 0

:. I = Int (1 to 0) e^(-x/2) . Sqrt[1 - u^2] . (-2)e^(x/2) du
= 2 Int (0 to 1) Sqrt[1 - u^2] du

Let u = [email protected] => du/[email protected] = [email protected] => du = [email protected] [email protected]
u = 0 => @ = sin^-1(0) = 0
u = 1 => @ = sin^-1(1) = Pi/2

:. I = 2 Int (0 to Pi/2) cos^2(@) [email protected]
= Int (0 to Pi/2) ([email protected] + 1) [email protected]
= [([email protected])/2 + @] (0 to Pi/2)
= [sin(Pi)/2 + Pi/2]
= Pi/2

EDIT: I made a silly error, correct answer as above, previously was Pi/8.
0
13 years ago
#3
t = e^(-x)
dt/dx = -t

(int from 0 to infinity) sqrt(e^(-x) - e^(-2x)) dx
= (int from 1 to 0) sqrt(t - t^2) * -(1/t) dt
= (int from 0 to 1) sqrt(1/t - 1) dt

t = cos^2(u)
dt/du = -2cos(u)sin(u)

= (int from pi/2 to 0) sqrt(sec^2(u) - 1) * -2cos(u)sin(u) du
= 2 (int from 0 to pi/2) tan(u)cos(u)sin(u) du
= 2 (int from 0 to pi/2) sin^2(u) du
= (int from 0 to pi/2) 1 - cos(2u) du . . . . . cos(2u) = 1 - 2sin^2(u)
= [ u - (1/2) sin(2u) ] (from 0 to pi/2)
= pi/2
0
13 years ago
#4
Ooohhh, two different answers! My money's on Johnny!
0
13 years ago
#5
(Original post by Widowmaker)
Ooohhh, two different answers! My money's on Johnny!
How cheeky of you, just because he's cleverer!
0
13 years ago
#6
(Original post by Widowmaker)
My money's on Johnny!

i did it without seeing the answers... and i got the same answer as Jonny_W's
0
13 years ago
#7
Nima's mistake was here: "u = e^(-x/2) = > du/dx = -2e^(-x/2)"

I made several similar errors in my solution, but thanks to Mathematica corrected them before I posted .
0
13 years ago
#8
(Original post by Jonny W)
Nima's mistake was here: "u = e^(-x/2) = > du/dx = -2e^(-x/2)"

I made several similar errors in my solution, but thanks to Mathematica corrected them before I posted .
again!
i've just found it out too! but i refreshed: and Jonny got it first! hehehe!
0
13 years ago
#9
(Original post by Jonny W)
t = e^(-x)
dt/dx = -t

(int from 0 to infinity) sqrt(e^(-x) - e^(-2x)) dx
= (int from 1 to 0) sqrt(t - t^2) * -(1/t) dt
= (int from 0 to 1) sqrt(1/t - 1) dt

t = cos^2(u)
dt/du = -2cos(u)sin(u)

= (int from pi/2 to 0) sqrt(sec^2(u) - 1) * -2cos(u)sin(u) du
= 2 (int from 0 to pi/2) tan(u)cos(u)sin(u) du
= 2 (int from 0 to pi/2) sin^2(u) du
= (int from 0 to pi/2) cos(2u) + 1 du
= [ (1/2) sin(2u) + u ] (from 0 to pi/2)
= pi/2
cos2x = cos^2x - sin^2x = 1 - 2sin^2x => 2sin^2x = 1 - cos2x => sin^2x = (1 - cos2x)/2
However, you replaced sin^2u with (1 + cos2x)/2.

I corrected my error, and I got Pi/2 as well - But it seems like you made an error in your working?
0
13 years ago
#10
(Original post by Nima)
cos2x = cos^2x - sin^2x = 1 - 2sin^2x => 2sin^2x = 1 - cos2x => sin^2x = (1 - cos2x)/2
However, you replaced sin^2u with (1 + cos2x)/2.

I corrected my error, and I got Pi/2 as well - But it seems like you made an error in your working?
You are right of course.
0
13 years ago
#11
I think we have truly discovered the power of Jonny W. Even if he makes an error in the working, a true rarity, he'll still get the right answer hehe.
0
13 years ago
#12
(Original post by Nima)
cleverer!
No such word.
0
13 years ago
#13
(Original post by Matt_2K)
No such word.
I think you missed the subtle joke at myself by me saying he's cleverer...
0
13 years ago
#14
(Original post by Nima)
I think we have truly discovered the power of Jonny W. Even if he makes an error in the working, a true rarity, he'll still get the right answer hehe.
rofl
0
13 years ago
#15
(Original post by Nima)
I think you missed the subtle joke at myself by me saying he's cleverer...
I feel thick tonight.
0
13 years ago
#16
(Original post by Jonny W)
Nima's mistake was here: "u = e^(-x/2) = > du/dx = -2e^(-x/2)"

I made several similar errors in my solution, but thanks to Mathematica corrected them before I posted .
OK im feeling thicker...why is that wrong?
0
13 years ago
#17
*blank face*

maths.. pah.. who needs that!
0
13 years ago
#18
oh lol cos you're differentiating and not integrating.....please tell me im right!
0
13 years ago
#19
(Original post by posh_git)
oh lol cos you're differentiating and not integrating.....please tell me im right!
You are.
0
13 years ago
#20
lol thank you i don't feel as thick!
0
X

new posts
Back
to top
Latest
My Feed

### Oops, nobody has postedin the last few hours.

Why not re-start the conversation?

see more

### See more of what you like onThe Student Room

You can personalise what you see on TSR. Tell us a little about yourself to get started.

### University open days

• Norwich University of the Arts
Thu, 23 Jan '20
• SOAS University of London
Development Studies, Interdisciplinary Studies, Anthropology and Sociology, Languages, Cultures and Linguistics, Arts, Economics, Law, History, Religions and Philosophies, Politics and International Studies, Finance and Management, East Asian Languages & Cultures Postgraduate
Sat, 25 Jan '20
• University of Huddersfield
Sat, 25 Jan '20

### Poll

Join the discussion

0 (91)
13.13%
1 (89)
12.84%
2 (89)
12.84%
19.05%
22.37%
19.77%