Tuesday, December 2, 2014

Figure 2 shows that P(x) becomes sharply peaked as the variance ( σ ) goes to zero.

 
The delta function potential
Some properties of the Delta function in one dimension  are
 
                             ∫ f(x) δ(x- x0 ) dx = f (x 0 )                    ,        -∞ < x < +∞       (5)
 
      
 
                                              ∫ δ(x- x0 ) dx =1                    ,                              (6)
 
 
and                                          ∫ δ( a x ) dx =1/a                .                               (7)
 
The dimensions of   δ( x ) are 1/length  .
A dirac sequence is established when certain functions approach the delta function  with the help of an adjustable parameter. Two example are given. First consider the gaussian function
 
                              P(x) = (1/σ) ( 1/(2π)1/2 ) exp[ -(x-μ)2 /(2σ2 )]                 ,        (8)
 
μ is the mean and σ2 the variance. In other words
                                           ∫ x P(x) dx =  μ       ,      -∞ < x < +∞                      (9)
 
and                                     ∫ x2 P(x) dx =  σ2  .                                             . (10)                                           
                               
 
 
 
 
Fig 2. The gaussian function P(x) will resemble Dirac delta function as σ→ 0 .
 
Figure 2 shows that P(x) becomes sharply peaked as  the variance ( σ ) goes to zero.
 
 
 
A second  sequence that converges to the Dirac delta function is 
                                                                                        (11)          
It resembles the dirac delta function as n becomes large .
 
Its integral   ∫ δn (x) dx = 1    ,  - ∞ < x < + ∞ .
 
 
 
Let the potential be V(x) =- η δ(x)  where η is the strength of the delta function. This peculiar potential produces a single energy eigenvalue  given by E = - {h'2 /(2m)} η2  . To show this start with
 
                      d2 Ψ /dx2  - (2m/h'2 ) V(x) Ψ = - (2m/h'2 ) E Ψ                         .   (12)
Integrate the whole expression around the origin  from  -ε to  +ε . We get 
 
 
 ∫ (d2 Ψ /dx2 )  dx =  ( d Ψ( +ε ) /dx   - d Ψ( -ε ) /dx ) = 2 ( d Ψ( +ε ) /dx )           ,  (13)
see next figure 3.
Fig 3. Shape of  the eigenfunction of the delta function.
 
The potential term integration gives
 
- (2m/h'2 )∫ V(x) Ψ dx = - (2m/h'2 )∫  -η δ(x)  Ψ dx =  (2m/h'2 ) η  Ψ(0)    .      (14)
 
The last term is zero , i.e.
 
- (2m/h'2 ) E ∫ Ψ  dx =  - (2m/h'2 ) E  Ψ( 0) (2ε ) → 0                           .     (15)
 
From (13) and (140 it follows that
 
                     d Ψ( +ε ) /dx )  = (-1/2)(2m/h'2 ) η  Ψ(0)  =  -(m/h'2 ) η  Ψ(0)  . (16)
The function near the origin ( x > 0) is thus of the form   
 
                 Ψ   ~     exp{-(m η /h'2 ) x }  .                                                  (17)
To get the energy go back to (12) at x near the origin (x>0) and use (17)
 
      E Ψ  = - {h'2/(2m) } d2 Ψ /dx2 = - (1/2) (m/h'2 ) η2 Ψ                ,        (18) 
 
                   E = -m η2 /(2 h'2 )  ,
or                E = - η2 /2     in units where m=1,h'=1.  
 
This is the only allowed energy eigenvalue by an attractive delta function of strength η.
Figure 4 shows a delta like potential. Its eigenvalue is E = - η2 /2 = -50.

No comments:

Post a Comment