Special Functions MAT 4006 M Sc-IV Prof. Dr M.

Afzal Rana

Special Functions -MAT-4006

LAGUERRE‘S EQUATION
GAMMA FUNCTION
Lecture-VIII
2.3.3 LAGUERRE‘S EQUATION
Laguerre’s equation is
𝑥𝑦 ′′ + (1 − 𝑥)𝑦 ′ + 𝜈𝑦 = 0. (2.3.3.1)
This equation and its solutions called Laguerre functions are of interest in quantum mechanics,
for example, the hydrogen problem. The origin 𝑥 = 0 is a regular singular point and so take
𝑦(𝑥) = ∑∞
𝑛=0 𝑐𝑛 𝑥
𝑛+𝑟
. (2.3.3.2)
Substituting solution (2.3.3.2) into Eq. (2.3.3.1) to get
∞ ∞ ∞

𝑥 ∑ 𝑐𝑛 (𝑛 + 𝑟)(𝑛 + 𝑟 − 1)𝑥 𝑛+𝑟−2 + (1 − 𝑥) ∑ 𝑐𝑛 (𝑛 + 𝑟)𝑥 𝑛+𝑟−1 + 𝜈 ∑ 𝑐𝑛 𝑥 𝑛+𝑟 = 0,

𝑛=0 𝑛=0 𝑛=0

or
∞ ∞ ∞ ∞

∑ 𝑐𝑛 (𝑛 + 𝑟)(𝑛 + 𝑟 − 1)𝑥 𝑛+𝑟−1 + ∑ 𝑐𝑛 (𝑛 + 𝑟)𝑥 𝑛+𝑟−1 − ∑ 𝑐𝑛 (𝑛 + 𝑟)𝑥 𝑛+𝑟 + 𝜈 ∑ 𝑐𝑛 𝑥 𝑛+𝑟

𝑛=0 𝑛=0 𝑛=0 𝑛=0
= 0,
or
∞ ∞
𝑛+𝑟−1
∑ 𝑐𝑛 {(𝑛 + 𝑟)(𝑛 + 𝑟 − 1) + (𝑛 + 𝑟)}𝑥 + ∑ 𝑐𝑛 {−(𝑛 + 𝑟) + 𝜈}𝑥 𝑛+𝑟 = 0,
𝑛=0 𝑛=0

or
∞ ∞

∑ 𝑐𝑛 (𝑛 + 𝑟)2 𝑥 𝑛+𝑟−1 + 𝑥 𝑟 ∑ 𝑐𝑛 (𝜈 − 𝑛 − 𝑟)𝑥 𝑛 = 0,

𝑛=0 𝑛=0

or
𝑐0 𝑟 2 𝑥 𝑟−1 + 𝑥 𝑟 ∑∞ 2 𝑛−1
𝑛=1 𝑐𝑛 (𝑛 + 𝑟) 𝑥 + 𝑥 𝑟 ∑∞ 𝑛
𝑛=0 𝑐𝑛 (𝜈 − 𝑛 − 𝑟)𝑥 = 0. (2.3.3.3)

pg. 1
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

The indicial equation is

𝑟 2 = 0.
Then the Eq. (2.3.3.3) reduces to
∞ ∞
2 𝑛−1
∑ 𝑐𝑛 𝑛 𝑥 + ∑ 𝑐𝑛 (𝜈 − 𝑛)𝑥 𝑛 = 0.
𝑛=1 𝑛=0

or

∞ ∞
2 𝑛−1
∑ 𝑐𝑛 𝑛 𝑥 + ∑ 𝑐𝑛 (𝜈 − 𝑛)𝑥 𝑛 = 0,
⏟
𝑛=1 ⏟
𝑛=0
[ 𝑘=𝑛−1 𝑘=𝑛 ]
or
∞ ∞

[∑ 𝑐𝑘+1 (𝑘 + 1) 𝑥 + ∑ 𝑐𝑘 (𝜈 − 𝑘)𝑥 𝑘 ] = 0,
2 𝑘

𝑘=0 𝑘=0

or
∑∞ 2 𝑘
𝑘=0{𝑐𝑘+1 (𝑘 + 1) + 𝑐𝑘 (𝜈 − 𝑘)}𝑥 = 0. (2.3.3.4)
Therefore, the recurrence relation is
𝑐𝑘+1 (𝑘 + 1)2 + 𝑐𝑘 (𝜈 − 𝑘) = 0,
or
(𝑘−𝜈)
𝑐𝑘+1 = (𝑘+1)2 𝑐𝑘 , 𝑘 = 0,1,2, ⋯. (2.3.3.5)

When 𝜈 is a positive integer 𝑛, then

−𝑛 (−1)1 𝑛 −(𝑛−1) (−1)2 (𝑛−1)𝑛
𝑐1 = 𝑐0 = 𝑐0 𝑐2 = 𝑐1 = 𝑐0
12 12 22 (2!)2

−(𝑛−2) (−1)3 (𝑛−2)(𝑛−1)𝑛 −(𝑛−3) (−1)4 (𝑛−3)(𝑛−2)(𝑛−1)𝑛

𝑐3 = 𝑐2 = 𝑐0 . 𝑐4 = 𝑐3 = 𝑐0 .
32 (3!)2 42 (4!)2

⋮
(𝑛−𝑘+1)(𝑛−𝑘+2)(𝑛−𝑘+3)⋯(𝑛−1)𝑛
𝑐𝑘 = (−1)𝑘 (𝑘!)2
𝑐0 . (2.3.3.5)

We usually choose
𝑐0 = (−1)𝑛!,

pg. 2
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

then
(𝑛 − 𝑘 + 1)(𝑛 − 𝑘 + 2)(𝑛 − 𝑘 + 3) ⋯ (𝑛 − 1)𝑛
𝑐𝑘 = (−1)𝑘+1 𝑛!.
(𝑘!)2

∞
(𝑛 − 𝑘 + 1)(𝑛 − 𝑘 + 2)(𝑛 − 𝑘 + 3) ⋯ (𝑛 − 1)𝑛
𝑦(𝑥) = ∑ 𝑐𝑘 𝑥 𝑘 = (−1)𝑘+1 𝑛!. 𝑥 𝑛 .
(𝑘!)2
𝑘=0
∞ ∞
(𝑛 − 𝑘 + 1)(𝑛 − 𝑘 + 2)(𝑛 − 𝑘 + 3) ⋯ (𝑛 − 1)𝑛
𝑦(𝑥) = ∑ 𝑐𝑘 𝑥 = ∑(−1)𝑘+1
𝑘
𝑛! 𝑥 𝑘
(𝑘!)2
𝑘=0 𝑘=0

𝑦(𝑥)
(𝑛 + 1)(𝑛 + 2)(𝑛 + 3) ⋯ (𝑛 − 1)𝑛
=− 𝑛!
1
∞
𝑛(𝑛 + 1)(𝑛 + 2) ⋯ (𝑛 − 1)𝑛 (𝑛 − 𝑘 + 1)(𝑛 − 𝑘 + 2)(𝑛 − 𝑘 + 3) ⋯ (𝑛 − 1)𝑛
+ 2
𝑛! 𝑥 ∑(−1)𝑘−1
𝑘
𝑛! 𝑥 𝑘
(𝑘!) (𝑘!)2
𝑘=0

𝑛(𝑛 − 𝑘 + 2)(𝑛 − 𝑘 + 3) ⋯ (𝑛 − 1)𝑛

𝑐1 = 𝑛!.
(𝑘!)2

The first few Laguerre polynomials are

(1)
(2)
(3)

(4)

(5)

2.3.3.1 THE ORTHOGONAL LAGUERRE FUNCTIONS

The Laguerre polynomials 𝐿𝑛 (𝑥) do not by themselves form an orthogonal set. But the
functions 𝑒 −𝑥⁄2 𝐿𝑛 (𝑥) are orthogonal in the interval (0, ∞). For any two Laguerre polynomials
𝐿𝑚 (𝑥) and 𝐿𝑛 (𝑥), we have from Laguerre equation
𝑥𝐿′′𝑚 + (1 − 𝑥)𝐿′𝑚 + 𝑚𝐿𝑚 = 0,
𝑥𝐿′′𝑛 + (1 − 𝑥)𝐿′𝑛 + 𝑛𝐿𝑛 = 0.

pg. 3
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

Multiplying these equations by 𝐿𝑛 (𝑥) and 𝐿𝑚 (𝑥) respectively and subtracting to get
𝑥(𝐿𝑛 𝐿′′𝑚 − 𝐿𝑚 𝐿′′𝑛 ) + (1 − 𝑥)(𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 ) + (𝑚 − 𝑛)𝐿𝑚 𝐿𝑛 = 0,
or
(1 − 𝑥) (𝑛 − 𝑚)𝐿𝑚 𝐿𝑛
(𝐿𝑛 𝐿′′𝑚 − 𝐿𝑚 𝐿′′𝑛 ) + (𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 ) = ,
𝑥 𝑥
or
𝑑 (1 − 𝑥) (𝑛 − 𝑚)𝐿𝑚 𝐿𝑛
(𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 ) + (𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 ) = .
𝑑𝑥 𝑥 𝑥
Now, integrating factor is
1−𝑥 1
)𝑑𝑥
𝐼𝐹 = 𝑒 ∫( 𝑥 = 𝑒 ∫(𝑥−1)𝑑𝑥 = 𝑒 (𝑙𝑛𝑥−𝑥) = 𝑒 𝑙𝑛𝑥 𝑒 −𝑥 = 𝑥𝑒 −𝑥 .
Multiplying above equation by integrating factor to have
𝑑
𝑥𝑒 −𝑥 (𝐿 𝐿′ − 𝐿𝑚 𝐿′𝑛 ) + 𝑒 −𝑥 (1 − 𝑥)(𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 ) = (𝑛 − 𝑚)𝑒 −𝑥 𝐿𝑚 𝐿𝑛 .
𝑑𝑥 𝑛 𝑚
or
𝑑
[𝑥𝑒 −𝑥 (𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 )] = (𝑛 − 𝑚)𝑒 −𝑥 𝐿𝑚 𝐿𝑛 .
𝑑𝑥
Integrating from 0 to ∞ to have
∞
[𝑥𝑒 −𝑥 (𝐿𝑛 𝐿′𝑚 − 𝐿𝑚 𝐿′𝑛 )]|∞ −𝑥
0 = (𝑛 − 𝑚) ∫ 𝑒 𝐿𝑚 𝐿𝑛 𝑑𝑥 ,
0

or
∞
(𝑛 − 𝑚) ∫ 𝑒 −𝑥 𝐿𝑚 𝐿𝑛 𝑑𝑥 = 0.
0

Thus, if 𝑚 ≠ 𝑛, then
∞
∫0 𝑒 −𝑥 𝐿𝑚 𝐿𝑛 𝑑𝑥 = 0 (𝑚 ≠ 𝑛), (2.3.3.1.1)

which proves the required result.

pg. 4
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

2.3.3.2 THE ASSOCIATED LAGUERRE POLYNOMIALS 𝐋𝐦

𝐧 (𝐱)

Differentiating Laguerre’s equation (2.3.3.1) 𝑚 times by the Leibnitz theorem we obtain

𝑑𝑛 𝑛(𝑛 − 1) 𝑛−2 2
{ 𝑛
(𝑢(𝑥)𝑣(𝑥)) = 𝑢𝑛 𝑣 + 𝑛𝑢𝑛−1 𝑣 1 + 𝑢 𝑣 + ⋯ + 𝑛𝑢1 𝑣 𝑛−1 + 𝑢𝑣 𝑛 }
𝑑𝑥 2!

[𝑥𝑦 𝑚+2 + 𝑚𝑦 𝑚+1 ] + [𝑦 𝑚+1 − 𝑥𝑦 𝑚+1 − 𝑚𝑦 𝑚 ] + 𝜈𝑦 𝑚 = 0 (𝜈 = 𝑛),

or
𝑥𝑦 𝑚+2 + (𝑚 + 1 − 𝑥)𝑦 𝑚+1 + (𝑛 − 𝑚)𝑦 𝑚 = 0.
If we write
𝑧 = 𝑦𝑚
then above equation becomes
𝑥𝑧 ′′ + (𝑚 + 1 − 𝑥)𝑧 ′ + (𝑛 − 𝑚)𝑧 = 0. (2.3.3.2.1)
This is Laguerre’s associated equation and it possesses polynomial solution
𝑧 = 𝐷𝑚 𝐿𝑛 (𝑥) = 𝐿𝑚
𝑛 (𝑥) (𝑚 ≤ 𝑛), (2.3.3.2.2)
Called the associated Laguerre polynomial of degree (𝑛 − 𝑚).
Now, the Laguerre polynomial are given by Rodrigues’ formula
𝑑𝑛
𝐿𝑛 (𝑥) = 𝑒 𝑥 𝑑𝑥 𝑛 (𝑥 𝑛 𝑒 −𝑥 ). (2.3.3.2.3)

Using formula (2.3.3.2.3) we obtain

𝑑𝑚 𝑑𝑚 𝑑𝑛
𝐿𝑚 𝑥 𝑛 −𝑥
𝑛 (𝑥) = 𝑑𝑥 𝑚 𝐿𝑛 (𝑥) = 𝑑𝑥 𝑚 {𝑒 𝑑𝑥 𝑛 (𝑥 𝑒 )}, (2.3.3.2.3)

Which is very useful in establishing further properties of the associated Laguerre polynomials
𝐿𝑛 (𝑥). The first few polynomials are listed below:
𝑑0 𝑑0 𝑑 0 0 −𝑥
𝐿00 (𝑥) = 0 𝐿0 (𝑥) = 0 {𝑒 𝑥 (𝑥 𝑒 )} = 1,
𝑑𝑥 𝑑𝑥 𝑑𝑥 0

𝑑0 𝑑0 𝑑1 1 −𝑥
𝐿01 (𝑥) = 𝐿 (𝑥) = {𝑒 𝑥 (𝑥 𝑒 )} = 1 − 𝑥,
𝑑𝑥 0 1 𝑑𝑥 0 𝑑𝑥1

𝑑1 𝑑1 𝑑1 1 −𝑥
𝐿11 (𝑥) = 1 𝐿1 (𝑥) = 1 {𝑒 𝑥 (𝑥 𝑒 )} = −1,
𝑑𝑥 𝑑𝑥 𝑑𝑥1

pg. 5
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

𝑑0 𝑑0 𝑑2
𝐿02 (𝑥) = 0 𝐿2 (𝑥) = 0 {𝑒 𝑥 (𝑥 2 𝑒 −𝑥 )} = 2 − 4𝑥 + 𝑥 2 ,
𝑑𝑥 𝑑𝑥 𝑑𝑥 2

𝑑1 𝑑1 𝑑2
𝐿12 (𝑥) = 𝐿 (𝑥) = {𝑒 𝑥 (𝑥 2 𝑒 −𝑥 )} = −4 + 2𝑥,
𝑑𝑥1 2 𝑑𝑥1 𝑑𝑥 2

𝑑2 𝑑2 𝑑 2 2 −𝑥
𝐿22 (𝑥) = 𝐿 (𝑥) = {𝑒 𝑥 (𝑥 𝑒 )} = 2.
𝑑𝑥 2 2 𝑑𝑥 2 𝑑𝑥 2

2.3.4 GAMMA FUNCTION

Euler’s integral definition of the gamma function is
∞
Γ(𝑥) = ∫0 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡. (2.3.4.1)

Convergence of the integral requires that 𝑥 − 1 > −1 or 𝑥 > 0.

2.3.4.1 SOME PROPERTIES OF GAMMA FUNCTION

In section some properties of gamma function are illustrated.
1) Γ(𝑥 + 1) = 𝑥Γ(𝑥)
2) Γ(𝑛 + 1) = 𝑛!, 𝑛 is a positive integer
1
3) Γ ( ) = √𝜋.
2

Proof: 1) By definition (2.3.4.1), we have

∞
Γ(𝑥 + 1) = ∫ 𝑡 𝑥 𝑒 −𝑡 𝑑𝑡,
0

Integrating by parts to obtain

∞
Γ(𝑥 + 1) = −(𝑡 𝑥 𝑒 −𝑡 )|∞
0 + ∫ 𝑥𝑡
𝑥−1 −𝑡
𝑒 𝑑𝑡
0
∞
Γ(𝑥 + 1) = 𝑥 ∫ 𝑡 𝑥−1 𝑒 −𝑡 𝑑𝑡 = 𝑥Γ(𝑥).
0

Thus
Γ(𝑥 + 1) = 𝑥Γ(𝑥). (2.3.4.1.1)
2) Setting 𝑥 = 1 in Eq. (2.3.4.1) to obtain
∞
Γ(1) = ∫ 𝑒 −𝑡 𝑑𝑡 = (−𝑒 −𝑡 )|∞
0 = 1.
0

pg. 6
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

Thus, Eq. (2.3.4.1.1)

Γ(2) = Γ(1 + 1) = 1Γ(1) = 1,

Γ(3) = Γ(2 + 1) = 2Γ(2) = 2 × 1 = 2!,
Γ(4) = Γ(3 + 1) = 3Γ(3) = 3 × 2! = 3!,
Γ(5) = Γ(4 + 1) = 4Γ(4) = 4 × 3! = 4!,
⋮
Hence, for and positive integer 𝑛,

Γ(𝑛 + 1) = 𝑛!. (2.3.4.1.2)

For this reason the gamma function is called the generalized factorial function.

Alternate Approach:
We know that Γ(𝑥 + 1) = 𝑥Γ(𝑥), therefore

Γ(𝑛 + 1) = 𝑛Γ(𝑛) = 𝑛Γ((𝑛 − 1) + 1) = 𝑛(𝑛 − 1)Γ(𝑛 − 1) = 𝑛(𝑛 − 1)(n − 2)Γ(𝑛 − 2).

Continuing this process, we have
Γ(𝑛 + 1) = 𝑛(𝑛 − 1)(n − 2) ⋯ 2 ∙ 1Γ(1) = 𝑛(𝑛 − 1)(n − 2) ⋯ 2 ∙ 1 = 𝑛!.

3) Setting 𝑥 = 1/2 in Eq. (2.3.4.1) to obtain

∞
1 1
Γ ( ) = ∫ 𝑡 −2 𝑒 −𝑡 𝑑𝑡.
2 0

Let

𝑡 = 𝑢2 ⟹ 𝑑𝑡 = 2𝑢𝑑𝑡.
Also

𝑡 = 0 ⟹ 𝑢 = 0, 𝑎𝑛𝑑 𝑡 = ∞ ⟹ 𝑢 = ∞,
then above equation becomes

∞ ∞
1 2 2
Γ ( ) = 2 ∫ 𝑢−1 𝑒 −𝑢 𝑢 𝑑𝑢 = 2 ∫ 𝑒 −𝑢 𝑑𝑢.
2 0 0

Also

pg. 7
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

∞ ∞
2 2
∫ 𝑒 −𝑢 𝑑𝑢 = ∫ 𝑒 −𝑣 𝑑𝑣.
0 0

Now

1 2 1 1 ∞
−𝑢2
∞
−𝑣 2
∞
2 2
[Γ ( )] = Γ ( ) Γ ( ) = (2 ∫ 𝑒 𝑑𝑢) (2 ∫ 𝑒 𝑑𝑣) = 4 ∫ 𝑒 −(𝑢 +𝑣 ) 𝑑𝑢 𝑑𝑣.
2 2 2 0 0 0

Switching to polar coordinates

𝑢 = 𝑟 cos 𝜃 ,
} ⟹ 𝑢2 + 𝑣 2 = 𝑟 2 .
𝑣 = 𝑟 sin 𝜃 ,
Now
𝑢 = 0 ⟹ 𝑟 cos 𝜃 = 0 ⟹ 𝑟 = 0,
}
𝑢 = ∞ ⟹ 𝑟 cos 𝜃 = ∞ ⟹ 𝑟 = ∞.
Similarly,
𝑣 = 0 ⟹ 𝑟 sin 𝜃 = sin 𝜃 = 0 ⟹ 𝜃 = 0,
}
𝑣 = ∞ ⟹ 𝑟 sin 𝜃 = ∞ ⟹ sin 𝜃 = 1 ⟹ 𝜃 = 𝜋⁄2 .
𝜕𝑢 𝜕𝑢
𝜕(𝑢, 𝑣) 𝜕𝜃 | = |cos 𝜃 −𝑟 sin 𝜃
𝐽= = | 𝜕𝑟 | = 𝑟(cos2 𝜃 + sin2 𝜃) = 𝑟.
𝜕(𝑟, 𝜃) 𝜕𝑣 𝜕𝑣 sin 𝜃 𝑟 cos 𝜃
𝜕𝑟 𝜕𝜃
Therefore,
𝜋⁄
1 2 ∞
−(𝑢2 +𝑣 2 )
∞ 2 2
[Γ ( )] = 4 ∫ 𝑒 𝑑𝑢 𝑑𝑣 = 4 ∫ ∫ 𝑒 −𝑟 𝑟 𝑑𝑟 𝑑𝜃 ,
2 0 0 0
𝜋⁄ 𝜋⁄
1 2 −4 2
−𝑟 2
∞ 2 𝜋⁄
2
𝜋
[Γ ( )] = ∫ [𝑒 ]0 𝑑𝜃 = 2 ∫ 𝑑𝜃 = 2𝜃|0 = 2 [ ] = 𝜋.
2 2 0 0 2
Thus
1
Γ (2) = √𝜋. (2.3.4.1.3)
1
EXAMPLE 2.3.4.1: Show that Γ (− 2) = −2√𝜋.

PROOF: Set 𝑥 = − 12 in Eq. (2.3.4.1.1) to obtain

1 1 1
Γ (− + 1) = − Γ (− ) ⟹
2 2 2
1 1 1
Γ (− ) = −2Γ ( ) = −2√𝜋 ∵ Γ ( ) = √𝜋.
2 2 2

pg. 8
 Special Functions MAT 4006 M Sc-IV Prof. Dr M. Afzal Rana

EXAMPLE 2.3.4.1: Show that Γ(5) = 4!.

PROOF: Using formula (2.3.4.1.2), i.e.
Γ(𝑛 + 1) = 𝑛!.
set 𝑛 = 4, then
Γ(5) = 4!.

Alternate:
Using formula (2.3.4.1.), i.e.
Γ(𝑥 + 1) = 𝑥Γ(𝑥).
We get

Γ(5) = Γ(4 + 1) = 4Γ(4) = 4Γ(3 + 1) = 4 × 3Γ(3) = 4 × 3 × 2Γ(2) = 4 × 3 × 2 × 1Γ(1)

= 4 × 3 × 2 × 1 = 4!

pg. 9