Proof of Γ(1/2)

The gamma function is defined as

Z ∞
Γ(α) = xα−1 e−x dx.
0

Making the substitution x = u2 gives the equivalent expression

Z ∞
2
Γ(α) = 2 u2α−1 e−u du
0

A special value of the gamma function can be derived when 2α − 1 = 0 (α = 12 ). When α = 12 , Γ 12

simplifies as
  Z ∞
1 2
Γ =2 e−u du
2 0
1
To derive the value for Γ 2 , the following steps are used. First, the value of Γ 21 is squared. Second,

the squared value is rewritten as a double integral.
Third, the double integral is evaluated by
transforming to polar coordinates. Fourth, the Γ 12 is explicitly solved for.
First, square the value for Γ 12 and rewrite as a double integral. Hence,

  2      
1 1 1
Γ = Γ Γ
2 2 2
 Z ∞  Z ∞ 
2 2
= 2 e−u du 2 e−v dv
0 0

Z ∞ Z ∞
2
+v 2 )
=4 e−(u dvdu (1)
0 0

The region R which defines the first quadrant, is the region of integration for the integral in (1). The
bivariate transformation u = r cos θ, v = r sin θ will transform the integral problem from cartesian
coordinates to polar coordinates, (r, θ). These new variables will range from 0 6 r 6 ∞ and 0 6 θ 6 π2
for the first quadrant. The Jacobian of the transformation is
 ∂u ∂u   
 ∂r ∂θ  cos θ −r sin θ
 
|J| =  ∂v ∂v  =  = r cos2 θ + r sin2 θ = r
  sin θ r cos θ 
∂r ∂θ

Hence, (1) can be written as

  2 Z ∞Z ∞ Z π2 Z ∞
1 2 2 2
Γ =4 e−(u +v ) dvdu = 4 e−r rdrdθ
2 0 0 0 0

"Z π
#"Z #
2 ∞ h π i  1 Z −∞ 
2
−r u
=4 dθ e rdr =4 − e du
0 0 u = −r 2 2 2 0
du = −2rdr

= −π [0 − 1]

=π
2
Finally, since e−u > 0 for all u > 0, then Γ 12 > 0. Hence,

  2
√
 
1 1
Γ = π =⇒ Γ = π
2 2