Quantum Physics (I)

Answer Sheet for Homework 5

Da-Ming Huang1
B00502124
1

Department of Electrical Engineering, National Taiwan University, Taipei 106, Taiwan

Problem 1. Inside the region, it is like the nite well potential

V (x) =

V0 , x > 0
0,
x<0

where V0 = 50MeV.
The wave function can be written as
{

(x) =

where k :=

Aeikx + Beikx , x < 0

F eilx ,
x>0

2mE
,
~

l :=

2m(E+V0 )
.
~

There are four boundary conditions:

Continuity of at 0 :

Continuity of

d
dx

A+B =F

at 0 : ik (A B) = ilF

Solve these equations, we obtain the reection coecient R is

)2
)2
(
(
 2
B 
E + V0 E
lk



R=  =
=
A

l+k
E + V0 + E
)2

55 5
47 12 11

=
0.2880
25
55 + 5

Problem 2. (a) Now we consider the innite well potential

The excited energy is

(n + 1)2 2 ~2
En =
2ma2
where a is the radius of the nucleus.
Using a 1015 m to estimate, we have
E0 =

(
)
2 ~2
11

3.7603

10
eV
2me (1015 )2

and

m0 c2 = me c2 5.1100 104
Hence

eV

m0 c 2
1.3589 106 1 E0 m0 c2
E0

(b) Since we consider the ground state, so the wavelength = 2a = (2 1015 ) m

So the energy E is
E = pc =

(
)
h
c 6.1992 108 eV

Problem 3. (a) The time-independent Schrodinger equation is

[

1
~2 d2 (x)

+ m 2 x2 + (x) (x) = E(x)

2
2m dx
2
Dene :=

m
x,
~

k :=

E
~

and :=

m
~2

~
,
m

(1)

then Eq.(1) can be rewritten as

)
d2 () (
2
() 2()() = 0
+
2k

d 2

(2)

With = 0, Eq.(2) is recognized as the Weber-Hermite dierential equation.

The solution which had been introduced in the textbook is
(

m
n () =
~

)1
4

2
2
1
Hn ()e 2 Hn ()e 2
2n n!

(3)

So we have
() =

cn n () := w()e 2

(4)

n=0

At the same time, it is a common practice to write 2k = 2 + 1, where R.

This allow us to transform Eq,(2) into
d2 w
dw
2
+ 2w 2()w = 0
2
d
d

(5)

which is Hermites dierential equation when = 0.

Apply a further substitution, := 2 , so we have
2
d2 w
dw dw
dw
2d w
=
4
+2 ,
= 2
2
2
d
d
d d
d

This substitution transform Eq.(5) into the Kummers equation,

dw
d2 w
aw = 0
2 + (b )
d
d

(6)

with a = 2 and b = 21 .
Thus, the general solution of Eq.(6) is
(

1
1
w() = A M , , + B U , ,
2 2
2 2

(7)

where A and B are arbitrary complex constants and is an arbitrary number.

(

The solution U 2 , 21 , can be expressed in terms the function M 2 , 12 ,

(

M 2 , 12 ,

1
U , , = (1) (1 )
2 2
2 22

1
2 , 32 ,
2
( ) (
)
32 2

(8)

Using Eq.(8), it is possible to show if is either zero or a positive integer

number (denoted by n), the solution in Eq.(7) can put in the following form
( > 0):
w()

)
(

n 1

,
,

,
M

2 2

M 1
2

n 3
, ,
2 2

for n even
)

(9)
, for n odd

Note that the solution M (a, b, ) with non-integer blows up as +

since
(
)]
(b) ab [
1 + O ||1
e
+ (a)
(
)]
(b) 2 ab [
= lim
1 + O ||1 +
e
+ (a)

M (a, b, ) = lim

(10)

So the solution is not what we want.

On the contrary way, the function U 2 , 12 , with a non-integer does not

(

blows up as +. In fact, it can be shown that U 2 , 12 ,

as

+ by using Eq.(8) and Eq.(10).

Thus, the function

1
() = Ae U , , 2
2 2
could in principle be an acceptable wave function for any value of , since it
2

is both an even parity function of y and square integrable (therefore normalizable). Why is it then that this solution has been casted away from textbooks? The weaker answer would be because it does not provide quantized
energy values, which is known to exist in any conned quantum system. The
stronger answer is however that the function () violates the boundary con

dition <
(0) = >
(0) for any non-integer , which must be obeyed by the

even parity wave functions () when = 0. We are now about to see that
(0+ ) = (0 ). It is the latter property of () which allows the solution
of the quantum problem(5) with nite .
We have now gathered all the information needed to nd the solutions of the
eigenvalue problem(5). Since the Hamiltonian

1
~2 d2
+
m 2 x2 + (x)
2m dx2 2

is invariant over a parity the transformation, their eigenstates () are either

even or odd parity states. In the case of odd states we have (0) = 0 and
therefore they do not see the presence of the delta-function at the origin. Thus,
the odd parity wave functions odd odd () are the states n () of the ordinary
harmonic oscillator, with = n = 1, 3, 5, . . . and the eigenvalues are k = n+ 12 .
To latter result we call the trivial solution of the eigenproblem (5).
The solution of the even parity eigenfunctions even even () is not as simple,
since these states feel the presence of the delta function at the origin. We need
to nd now the boundary condition the function w() must obey at the origin.
Proceeding as we did at the introductory section, we integrate Eq.(5) around
= 0 obtaining:

dw 
dw 
= 2w(0)
lim



0
d =+
d =

(11)

Eq.(11) enable us to nd the quantized energies of the even parity eigenstates

we are seeking. Thus, the correct wave function for even () is () and not

n () with n = 0, 2, 4, . . ., as in the case = 0, for the latter wave function

violates the boundary condition (15). Using the results

)
(
1 2

lim+ U , , = ( 1 )
x0
2 2
22

(
)
1 2

)
lim U , , = (
x0+
2 2
1 2

(12)
(13)

the eigenvalues associated with even parity eigenstates of Eq.(2) are given by
the numerical solution of the transcendent equation

F () := ( 1
2

2
)

=0

(14)

which follows from the boundary condition(11).

The absolute value of this negative energy state increases with the strength of
the -function potential and, in the limit , the connement imposed
by the harmonic potential becomes irrelevant and the wave function transforms

and with the same

into the bound state given by (x) = e|x| , = m
~2
eigenenergy. Indeed, using Stirlings formula
)
(
1
1
139
571
z z 12
(z) e z
2 1 +
+

+
(15)
12z 288z 2 51840z 3 2488320z 4

(+ 1 )
it can be proved that as +, ()2 . For = 1 + 12 || this implies
that as ,

2||. Since E ||~, the former result implies that

(

1
m2
E =
~ = 2
2
2~
which is the energy value we have obtain before for the simple case of an
2

isolated attractive -function potential.

0
0.0
2.0
4.0
6.0
8.0

-0.25
-0.1557
1.9288
3.9469
5.9558
7.9614

Eigenvalues
+0.25
-1.0
0.1281 -0.8424
2.0693 1.7208
4.0525 3.7912
6.0439 5.8258
8.0384 7.8473

for several values of

+1.0
-2.5
+2.5
-5.0
0.3927 -3.5865 0.6434 -12.9900
2.2546 1.4285 2.5042 1.2305
4.2002 3.5420 4.4274 3.3227
6.1699 5.6051 6.3772 5.3833
8.1501 7.6473 8.3412 7.4285

Hence the energy spectrum is

E=

)
(
1

n
+
~,

(x) is odd, n = 1, 3, 5, . . .

(
)

+ 1 ~,
2

(x) is even

+5.0
0.7691
2.7003
4.6364
6.5887
8.5509

(b) When , even solution n +

(

spectrum E 2n +

3
2

2
even (x) = lim e 2
0

1
2

, n = 0, 1, 2, . . ., so the energy

. Even solution:
(

||
) 1 +
= f () 2 (
n + 21
j=1

(n + j 1) ( 2 ) 2
)
e 2
(1
j!
+j
2

where f () is a nearly vanishing term.

Thus, when , even solutions still survive, and
{

even (x) |odd (x)|

even (0) 0

the energy spectrum then becomes

E=

1
1
1
2 ||
3
) M n, , 2
M n, , 2 ( 1
(n + )
2
2
2
2 n

)
(
1

~,
2n
+
1
+

for odd (x),

(
)

2n + 1 + 1 ~,
2

for even (x), n = 0, 1, 2, . . .

n = 0, 1, 2, . . .

which implies the energy spectrum is doubly degenerate.

Problem 4. (a) Since R = 1 T , we have
=

1T
R
=
T
T

i. T = 0:

lim = lim

T 0

T 0

ii. T = 1:

1T
T

1 T 
|T =1 =
=0

T
T =1
(b) According to Problem 1.(c) of Recitation 4, we have
1
|M22 |2


 M21 2
2
2


R = |rl | = |rr | = 
M22 
T = |tl |2 = |tr |2 =

So we have


 M21 
M 
R
= 22
= |M21 |2
=
1
T
|M |2
22

= (M21 ) (M21 ) = (M12 ) (M12 )

= |M12 |2

(c) By the result of Problem 2.(a) of Recitation 4 and multiplication property, we

can derive the relation
Mn+1 = Mn Ma M1 M+a
(

where
Ma =

e+ika
0
ika
0
e

here k is from Recitation 4.

Dene the elements of Mn and M1 to be
(

Mn :=
(

M1 :=

n
n
M11
M12
n
n
M21
M22
1
1
M11
M12
1
1
M21
M22

)
)

So we have
Mn+1 = Mn Ma M1 M+a
(

=
(

=
(

n
n
M11
M12
n
n
M21
M22
n
n
M11
M12
n
n
M21 M22

)(

)(

eika
0
+ika
0
e

)(

1
1
M11
M12
1
1
M21
M22

1
1 2ika
M11
M12
e
1 +2ika
1
M21 e
M22

)(

e+ika
0
ika
0
e

n
1
n
1 +2ika
n
1 2ika
n
1
M11
M11
+ M12
M21
e
M11
M12
e
+ M12
M22
n
1
n
1 +2ika
n
1 2ika
n
1
M21
M11
+ M22
M21
e
M21
M12
e
+ M22
M22

Thus, the resistance n+1 is



 n
n+1 
1
n
1 +2ika 
n+1 = M21
 = M21
M11
+ M22
M21
e

(

n
1
n
1 +2ika
= M21
M11
+ M22
M21
e

)(
(

n
1
n
1 +2ika
M21
M11
+ M22
M21
e

)
(

n
1
n
1
n
n
1
1
n
n
1
1
= M12
M11
 + M11
M12
 + M12
M22
M21
M22
e+2ika + M11
M21
M11
M12
e2ika

(d) The average of n+1 is

avg
n+1 =

k k
n+1 da
0

 n
n
1 
1 
= M12
M11
 + M11
M12


= n (1 + 1 ) + (1 + n ) 1
= (1 + 21 ) n + 1

According to the relation above, we discovered that

avg
n+1 +
avg
n +
avg
2

1
2
1
2

1
2

= n +

1
2

= n1 +
..
.
(

= 1 +

1
2

1
2

(1 + 21 )
)

(1 + 21 )

1
1
avg
= + (1 + 21 )n1 1 +
n

2
2

(1 + 21 )

So

1
n
avg
n = [(21 + 1) 1]
2
where we use the assumption avg
n n , n.
i. for small n:
1
n
avg
n = [(21 + 1) 1]
2
n1

1
= [(21 + 1) 1]
(21 + 1)i
2
i=0
= 1

n1

(21 + 1)i

i=0

Since n is small,
n1

(21 + 1)i

i=0

n1

1=n

i=0

(Assumed that 1 1)
Then we have
avg
n n1
ii. for big n: Since n is big, so we have (21 + 1)n 1, then we have
1
n
avg
n = [(21 + 1) 1]
2
1
(21 + 1)n
2
Problem 5. The potential reaches zero at x = a, and starts to decay at x = 23 a. So when a
particle reaches x = 23 a, it starts to accelerate and leak out to the region on the far
right.
The time-independent Schrodinger equation is
(

1
~2 d2 (x)
1
+
m 2 x2 m 2 x3 (x) = E(x)

2
2m dx
2
2a

Dene :=

m
x
~

and

2E
~

:= K we have

d2 () m 3
2E
2 (x)
() =
+
~
d 2
a

d () 2
=
d 2
Since a

~
,
m

~
m

3 K ()

we have
)
d2 () ( 2

K
()
d 2

in the region 0 < x < 23 a.

This the form of harmonic oscillator, so the ground state is
0 (x)

m
~

)1

e 2~ x
m

and the eigenvalue (energy) is

1
E0 ~
2
Then we have two turning points: inner turning point := x1 ; outer turning point
:= x2
(

1
1
~ = m 2 x21
2
2
(
1
1
2
~ = m x22
2
2

1 3
1
x1 m 2 x21 x1
a
2
)
1 3
x
a 2

It is hard to nd the outer, but since a

small real number. (Obviously, x1 <

2
a
3

~
,
m

~
m

we assume x2 a , where is a

< x2 < a)

So
[

1
~
(a )2 (a )3
m
a
3

= a 22
a
a

So we have

~
,
ma

then
{

2
T
= e2 = exp
~

= exp

= exp

2m [V (x) E]dx

x1

2 a ma

exp

~
=
~
m

x2

1
m2 2 x2 m2 2 x3 m~dx
a

~
a ma

2a m
~
~
m

x2
a2

x3
a3

~
m
a2

( )
x
d

[
]

1
2a2 m
2a2 m 1 2
2
3
3

y y dy = exp
y y dy
~
m
~
~
0
a
~
m
a2

1

2a m 2 (3y + 2) [ (y 1) y ] 
= exp


~
15y 3


3
2

2a m
2 (3y + 2) [ (y 1) y ]
= exp
lim

~ y0
15y 3
(
)

= exp

3
2

8m 2
a
15~

So it takes time to leak out and is

=
where

e2

is the period of the particle.

8m 2
2
exp
a

15~