∫ 0 ∞ arctan ( x 2 1 ) d x
If the value of the integral above can be represented as B π A , where A and B are integers, find the value of A + B .
This section requires Javascript.
You are seeing this because something didn't load right. We suggest you, (a) try
refreshing the page, (b) enabling javascript if it is disabled on your browser and,
finally, (c)
loading the
non-javascript version of this page
. We're sorry about the hassle.
Similar solution with @Aaghaz Mahajan's
Using Feynman's integration trick of differentiation under the integral sign as follows:
I ( a ) ∂ a ∂ I ( a ) ⟹ I ( a ) I ( 0 ) = ∫ 0 ∞ tan − 1 ( x 2 a ) d x = ∫ 0 ∞ a 2 + x 4 x 2 d x = 4 a 1 ∫ 0 ∞ 1 + u u − 4 1 d u = 4 a 1 B ( 4 3 , 4 1 ) = 4 a Γ ( 1 ) Γ ( 4 3 ) Γ ( 4 1 ) = 4 a ( 0 ! ) 2 π Γ ( 2 1 ) = 8 a π = ∫ 8 a π d a = 2 π a + C = 0 + C = 0 Let u = a 2 x 4 ⟹ a 2 d u = 4 x 3 d x Beta function: B ( m , n ) = ∫ 0 ∞ ( 1 + u ) m + n u m − 1 d u B ( m , n ) = Γ ( m + n ) Γ ( m ) Γ ( n ) , where Γ ( ⋅ ) denotes gamma function Legendre duplication: π Γ ( 2 z ) = 2 2 z − 1 Γ ( z ) Γ ( z + 2 1 ) with z = 4 1 and by Γ ( n ) = ( n − 1 ) ! Since Γ ( 2 1 ) = π where C is the constant of integration. ⟹ C = 0
Therefore, I ( 1 ) = ∫ 0 ∞ tan − 1 ( x 2 1 ) d x = 2 π , ⟹ A + B = 1 + 2 = 3 .
**References:
@Chew-Seong Cheong Sir, could you please tell me how to align my integrals to the center...??
Log in to reply
The LaTex code I used for my solution above. Hope it is helpful.
Let J = ∫ 0 ∞ arctan ( t 2 1 ) d t Substituting t 2 1 = u the integral changes to J = 2 1 ∫ 0 ∞ u − 2 3 arctan u d u = 2 1 ∫ 0 ∞ u − 2 3 k = 0 ∑ ∞ 2 k + 1 ( − 1 ) k u 2 k + 1 d u
= 2 1 ∫ 0 ∞ u − 2 1 k = 0 ∑ ∞ k ! ( − u 2 ) k ⋅ 2 k + 1 Γ ( k + 1 ) d u
Using substitution u 2 = x this changes to J = 4 1 ∫ 0 ∞ x − 4 3 k = 0 ∑ ∞ 2 k + 1 Γ ( k + 1 ) k ! ( − x ) k d x
Using Ramanujan's Master theorem this integral equals to J = 4 1 Γ ( s ) ϕ ( − s ) where s = 4 1 and ϕ ( k ) = 2 k + 1 Γ ( k + 1 )
Substituting the values you get the answer as J = 2 π
Hence A = 1 and B = 2
How did you align the integrals in the center of the page??
Log in to reply
@Aaghaz Mahajan I don't know.... I mean thats the basic code I usually write... So I don't have an idea which part of the code aligns it to the center.
But you didn't change the integration domain when substituted and used a series for arctan which is only valid for |x|<1 🤔
@Bert Vitan For the Ramanujan's Master theorem, the limits need not to be according to the radius of convergence of the function in the integral. It just needs that the function can be represented in form of infinite series near neighbourhood of x = 0 . See here
You say it is a very easy integral but u calculated for a long page!
therefore it has Contradiction
therefore this problem is not true
For those who wish to do this without applying RMT, let
F
(
a
)
=
∫
0
∞
arctan
(
x
2
a
)
d
x
Now, differentiating under the integral sign leads us to
F ′ ( a ) = ∫ 0 ∞ x 4 + a 2 x 2 d x
Substituting x 2 = a tan t we get,
F ′ ( a ) = 4 a 1 ∫ 0 2 π 2 ( sin t ) 2 1 ⋅ ( cos t ) − 2 1 d t
This is a simple application of Beta function along with Euler's Reflection Formula and hence, we get,
F ′ ( a ) = 2 2 a π
Now, integrating the expression, and using F ( 0 ) = 0 we get
F ( a ) = π 2 a
@Darkrai ~Rayquaza See my approach.......
Log in to reply
@Aaghaz Mahajan
For
F
′
(
a
)
there was no need to use substitution to form Integral that would give out the Beta function. Instead you could have simply applied Glasser's Master theorem as follows.
F
′
(
a
)
=
∫
0
∞
x
4
+
a
2
x
2
d
x
=
2
1
∫
−
∞
∞
(
x
−
x
a
)
2
+
2
a
d
x
=
2
1
∫
−
∞
∞
x
2
+
2
a
d
x
=
∫
0
∞
x
2
+
2
a
d
x
=
[
2
a
1
arctan
(
2
a
x
)
]
0
∞
=
8
a
π
Log in to reply
Ohhh!!! I didn't think of that!! Thanks!!
Log in to reply
@Aaghaz Mahajan – @Aaghaz Mahajan You are welcome...
Problem Loading...
Note Loading...
Set Loading...
Let J denote the integral. Taking integration by parts, J = ∫ 0 ∞ arctan ( 1 / x 2 ) d x = = 0 [ x arctan ( 1 / x 2 ) ] 0 ∞ + ∫ 0 ∞ 1 + x 4 2 x 2 d x = ∫ − ∞ ∞ 1 + x 4 x 2 d x . At this point, the last integral can be solved completely by elementary means. However, let us give a quicker solution using Glasser's master theorem . Rewriting the integrand, J = ∫ − ∞ ∞ x 2 + x − 2 1 d x = ∫ − ∞ ∞ ( x − x − 1 ) 2 + 2 1 d x Now, by the Glasser's master theorem, this equals J = ∫ − ∞ ∞ u 2 + 2 1 d u = 2 π .
Addendum. For the reader interested in a completely elementary solution, here is a way to go: Write J = ∫ 0 ∞ x 2 + x − 2 2 d x = ∫ 0 ∞ x 2 + x − 2 1 − x − 2 d x + ∫ 0 ∞ x 2 + x − 2 1 + x − 2 d x = ∫ 0 ∞ ( x + x − 1 ) 2 − 2 1 − x − 2 d x + ∫ 0 ∞ ( x − x − 1 ) 2 + 2 1 + x − 2 d x . Now we substitute v = x + x − 1 for the first integral and u = x − x − 1 for the second integral. Then J = ∫ ∞ ∞ v 2 − 2 1 d v + ∫ − ∞ ∞ u 2 + 2 1 d u . The limits of the first integral may be puzzling at the first glance, but they simply reflect the fact that, as x moves from 0 to ∞ , v = v ( x ) descends from ∞ to the minimun v ( 1 ) = 2 and then go back to ∞ . So in effect, this is simply the sum of two convergent improper integrals which cancel out each other: ‘ ‘ ∫ ∞ ∞ v 2 − 2 1 d v ” = ∫ ∞ 2 v 2 − 2 1 d v + ∫ 2 ∞ v 2 − 2 1 d v = 0 . So we are led to the same last step as in the previous solution, hence obtaining the answer 2 π .