Dirac-delta function and different Limit representation

of Dirac-delta function

Shyamal Bhar

Assistant Professor
Department of Physics
Vidyasagar College for Women
Kolkata - 6
 Dirac-delta function and different Limit representation of Dirac-
delta function

r̂
first calculate the divergence of
r2
 r̂
Let E 2
r

This vector function is directed radially outward and has a large positive
divergence
 
What happens, if we want to calculate the divergence of E ?

 1   2 1
E  2 r 2 
r r  r 

1 
 2 1 
r r
0 ..................1

???????
Surface integration of the vector function over a sphere of radius R ,
centered at the origin

   rˆ 
   R2  sin   d d rˆ 
 2
E  dS   R

 2
  s in   d  
0 0
d

 4 .......  2 

volume integral of the divergence of the vector function is


   E dv  0 ................  3 
 According to Divergence theorem equation no. (2) and (3) must
be same
???????
 Source of the problem r0

At this point the vector function E blows up.


  E  0 except at r  0
The surface integral is independent of R . So the surface integral will
be  for any sphere centered at the origin, whatever be the radius of
the sphere.

Entire contribution of the surface integral comes from the point r  0 .

For the consistency of the divergence theorem


   E dv  0 except r  0
 4 otherw ise
 
  E has an unusual property

It vanishes everywhere except at one point and its integral over any
volume containing that point is  .

This type of function is called Dirac Delta function.

One dimensional Dirac Delta function:

 0, if x  0
 x  
, if x  0
and

   x  dx  1
-
 An important property of a Delta function:

f  x    x   f  0   x 

 
 f  x    x  dx  f  0     x  dx
 

 f 0

 One can take the limits of integration from   to   instead of  to  

 if the peak of the delta function be at xa

0, if x  a
 x  a   
 , if x  a
and

 -
 x  a dx 1

 The product of this delta function with an arbitrary function f  x 

f  x    x  a  f  a    x 


 f  x    x  a  dx  f  a 

Dirac Delta function in three dimension

3
  r     x   y    z 

where r  xiˆ  yjˆ  zkˆ

  
3 
   r  d       x    y    z  dxdydz  1
all space   

And

 3   
 f  r    r  a  d  f  a 
all space
 Resolving the paradox introduced beginning of the talk

 rˆ  3 
  2   4  r 
r 

It follows that

1
2 rˆ 3 
    2  4  r 
r r
 Limiting representation of Dirac Delta function

 Technically   x  is not a function, since its value is not finite at r  0.

 In mathematical literature it is known as a generalized function or

distribution

Dirac delta function basically is the limit of sequence of functions

 There are many limit representation of Dirac delta functions

Examples:…………….
Limit of sequence of Rectangles R  x 

  x   lim R  x 
 0

 
0 for x  
2

1  
where, R  x    for   x 
 2 2
 
0 for x 
 2
Width of rectangles 
1
height of rectangles

 Limit of sequence of isosceles triangles T  x  :

  x  limT  x
 0

0 for x  

 x 
where, T  x  1 
  
  for x  

Here the base of the triangle T  x  is 2

1
and height

The area of the triangle is unity
Limit of sequence of Gaussian functions G  x 

  x   lim G   x  This is the normalized

0
x2 Gaussian distribution
1  2
w h ere, G  x   e 
  function. The area under the
curve is unity and the peak
1
value
 
Limit of sequence of Exponential functions E  x 

  x   lim E   x 
 0
x
1 
where, E  x   e 
2

The area under the curve is

1
unity and the peak value 2
Limit of sequence of sinc functions S  x 
This function arises
  x   lim S   x  frequently in signal
 0 processing and the
sin  x /   theory of Fourier
where, S  x   transforms. The full
x name of this function is
since cardinal but it is
commonly referred to
“sinc” function.
Limit of sequence of modified sinc function Sm  x

 x   lim S m 
0
x 
 2  
 1    x 
s in    
1 1
2
  The peak value   /
1 
  2  
w h e re , Sm x  
2  x 
sin  
 2 
Limit of sequence of Lorentzian L  x 

  x   lim L  x  1
 0 The peak value

1 1
where, L  x  
 x2
1 2

Limit of sequence of Inverse Cosh Square functions C  x 

  x   lim C  x 
 0

1 1
where, C  x  
2  x
2
cosh  
 

1
The peak value
2
Note: Though  is not a legitimate function but the integrals over  are
perfectly acceptable.

 
If
 f  x  d  x  dx   f  x  d  x  dx

1

2
for all function f  x 

then d1  x   d 2  x 
Example

Show that 1
  kx    x Where k is nonzero constant
k

Answer

Consider an arbitrary function f  x  and consider the following integral




 f  x    kx  dx
Change of variable y  kx
1
dx  dy
k
 
1  y
ve sign for k  0  f  x   kx dx   k  f  k   y dy
-ve sign for k  0
 1
  f 0 
k
1
 f 0 
k

1

k  f  x    x  dx


 
1 
  f  x    kx  dx   f  x     x   dx
  k 

1
Hence  k x    x 
k
Example

d
Show that x   x      x 
dx
Solution

 d
 f x  x    x    d x
  dx 


 d
 xf  x    x     dx
 xf  x     x  dx


 
df
 0  x   x  dx   f  x    x  dx

dx 

 0  0  f 0 
 
   f  x    x  dx


d
 x   x      x 
dx
Example

If   x  be a step function:

1, if x  0
  x  
0, if x  0
d
Show that   x
dx
Solution 
d
 f x dx

dx

 df
 f  x    x       x  dx

dx

df
 f      x dx
0
dx

 f     f    f 0 

 f 0 

  f  x    x  dx


d
   x
dx
Problem-1 6
  3x  2 x  1   x  3 dx  ?
2
2

6
  3x  2 x  1   x  3 dx
2
Solution
2

6
  f  x    x  3  dx f  x    3 x 2  2 x  1
2

 f 3 

 20
2
Problem-2
  2 x  3   3x  dx
2

2 1 f  x    2 x  3
Solution   f x  x dx
2 3
 1
 f 0 
3

1
a
Problem-3
   x  b  dx  ?


Solution
a
fo r a  b :    x  b  dx  1


a
fo r a  b :
   x  b  dx  0

 Reference:

Introduction to Electrodynamics

David J. Griffiths