Scribd
Upload
36 views6 pages

5uebung: 1 Ubung 5 - Computational Physics

This document discusses solving the time-independent Schrodinger equation for a particle in a linear potential well. It defines the problem, writes the dimensionless Schrodinger equation, develops a Numerov method code to solve it numerically, and uses the code to estimate the first three eigenvalues to two decimal places by plotting the wavefunctions for varying energy values and locating where the asymptotic behavior changes from positive to negative infinity.

Uploaded by

killmelorry1627
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
36 views6 pages

5uebung: 1 Ubung 5 - Computational Physics

This document discusses solving the time-independent Schrodinger equation for a particle in a linear potential well. It defines the problem, writes the dimensionless Schrodinger equation, develops a Numerov method code to solve it numerically, and uses the code to estimate the first three eigenvalues to two decimal places by plotting the wavefunctions for varying energy values and locating where the asymptotic behavior changes from positive to negative infinity.

Uploaded by

killmelorry1627
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 6

5Uebung

May 22, 2014


1

Ubung 5 - Computational Physics
Luca Gonzalez
Fernando Nu~nez
In [71]: import matplotlib.pyplot as plt
import numpy as np
%config InlineBackend.figure_format = svg
1.1 Exercise
We rst start with the time independent Schrodinger equation:

(z) +
2m

2
(E V (z)) (z) = 0
in our case the potential is given by:
V (z) = mgz
for z 0. If we now dene a scaling factor:
R :=


2
2m
2
g

1/3
and the dimensionless quantities:
x(z) :=
z
R
; :=
E
mgR
So that we get the equation:

(x) + ( x) (x) = 0
Where we can identify k(x) = x. We now proceed to write the corresponding code. We will create a
function called numerov so that we can use it later without having to write the code again.
In [69]: def numerov(E):
fig = plt.figure(figsize=(10, 12)) # Final figure size
for j in range(len(E)):
e = E[j] # Epsilon
n = 1000 # Number of steps
d = 15.0/n # x = 15 limit will be changed for the different parts
z = zeros([n], float)
k = zeros([n], float)
1
psi = zeros([n], float)
z[0] = 0
k[0] = e-z[0]
psi[0] = 0
z[1] = z[0]+d
k[1] = e-z[1]
psi[1]= 1
# Numerov code
for i in range(1,n-1):
z[i+1] = z[i]+d
k[i+1] = e-z[i+1]
psi[i+1] = (2*(1-5/12*d**2*k[i])*psi[i]-(1+d**2*k[i-1]/12)*psi[i-1])/(1+d**2*k[i+1]/12)
# Plots
ax = plt.subplot(5,2,j+1)
plt.plot(z, psi, label = r$\epsilon = %.2f$%e)
plt.axhline(0, color = k)
plt.xlabel($z$)
plt.ylabel($\psi(z)$)
plt.legend(loc=upper left)
show()
We can now select two dierent values for and see the asymptotic behaviour asked:
In [68]: numerov([2,6])
As we can see, for = 2 we have an asymptotic behaviour going to positive innity and for = 6 to
negative innity.
1.2 Exercise
We will now try to estimate the three rst eigenvalues
n
to 2 decimals behind the comma. The eigenvalues
make (x) 0 when x . These will be inside an interval cointaining the change of the asymptotic
behaviour of the solution for large x from positive innity to negative or vice versa. We rst plot a large
range so we can locate approximately where these changes will be produced. A later plot will be made to
give the accuracy asked.
2
In [60]: numerov(np.linspace(1, 10, 10))
In [50]: numerov(np.linspace(4.2, 4.29, 10))
3
We can now see how the tendency of the function changes when raising from 4.24 to 4.25, we can then
asume that
1
(4.24, 4.25).
In [55]: numerov(np.linspace(7.4, 7.49, 10))
4
With the same reasoning:
2
(7.42, 7.43).
In [70]: numerov(np.linspace(10, 10.09, 10))
5
And so for
3
(10.03, 10.04).
6

You might also like