WKB Approximation for Tunneling Rates: Odd-Shaped Barriers and Cold Emission, Study notes of Quantum Physics

The use of the wkb approximation to calculate tunneling rates through odd-shaped barriers and the application to cold emission from a metal. Topics include the transmission coefficient for delta-function and rectangular barriers, the wkb solutions in the classically forbidden region, and the fowler-nordheim tunneling rate calculation.

Typology: Study notes

Pre 2010

Uploaded on 07/30/2009

koofers-user-1ga-1
koofers-user-1ga-1 🇺🇸

10 documents

1 / 2

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Lecture 39 Highlights
We are now going to use the WKB approximation to calculate tunneling rates
through odd-shaped barriers. You encountered tunneling before in Phys 401, including:
1) Section 2.5 where tunneling through a delta-function barrier of
height )()( xxV
δ
α
=. The result for the transmission coefficient is EE
T/1
1
0
+
=, where
E
is the energy of the particle and is a characteristic energy in the
problem.
22
02/ h
α
mE =
2) Problem 2.33 Tunneling through a rectangular barrier of height and width
. The transmission coefficient is
0
V
a
() ()
+
=
EVm
a
EVE
V
T
0
2
0
2
02sinh
4
1
1
h
valid for
.
0
VE <
Consider the WKB approximation for particles in the “classically forbidden”
region, . In this region the kinetic energy is negative, which can be
interpreted as resulting from an imaginary momentum
)(xVE <mp 2/
2
(
)
)(2 xVEmp = . In this case
the WKB solutions we found last time become:
±=
x
class
class
dxxp
xp
D
x')'(
1
exp
)(
)( h
ψ
,
with
()
)(2 xVEmpclass = . Notice that the complex exponential has now become a
positive or negative exponential because the solutions are no longer running waves.
Consider a barrier of width awith an arbitrary potential on top, , as discussed
on pages 320-322 of Griffiths. For energies
)(xV
E
less than the minimum of , and for
barriers that are sufficiently tall and thick, the transmission probability is dominated by
the negative exponential term;
)(xV
, where
γ
2
eT
=
a
class dxxp
0
')'(
1
h
γ
Going back to the flat-top barrier of problem 2.33 to test this result, the
transmission probability in the WKB approximation (after doing the integral
with ) is
0
)( VxV =
()
EVm
a
eT
0
2
2
h. If we expand the exact result given above in the limit
of tall and wide barrier (i.e.
()
12 0>> EVm
a
h), the result is
()
()
EVm
a
e
V
EVE
T
0
2
2
2
0
0
4h. The pre-factor is on the order of 1, so the exponential
dominates, and is in agreement with the WKB approximate result.
Now consider the problem of cold emission from a metal. A metal can be
modeled as a potential well of depth
Φ
+
F
E, where is the Fermi energy and is the
work function of the metal (see the Figure below). The work function is the energy
F
EΦ
1
pf2

Partial preview of the text

Download WKB Approximation for Tunneling Rates: Odd-Shaped Barriers and Cold Emission and more Study notes Quantum Physics in PDF only on Docsity!

Lecture 39 Highlights

We are now going to use the WKB approximation to calculate tunneling rates through odd-shaped barriers. You encountered tunneling before in Phys 401, including:

  1. Section 2.5 where tunneling through a delta-function barrier of

height V ( x )= α δ( x ). The result for the transmission coefficient is E E

T

= , where

E is the energy of the particle and is a characteristic energy in the

problem.

2 2

E 0 = m α / 2 h

  1. Problem 2.33 Tunneling through a rectangular barrier of height and width

. The transmission coefficient is

V 0

a

mV E

a EV E

V

T

0

2 0

2 (^0) sinh 2 4

h

valid for

E < V 0.

Consider the WKB approximation for particles in the “classically forbidden”

region,. In this region the kinetic energy is negative, which can be

interpreted as resulting from an imaginary momentum

E < V ( x ) p^2 / 2 m

p = 2 m ( E − V ( x )). In this case

the WKB solutions we found last time become:

x class class

p x dx p x

D

x ( ') '

exp ( )

h

with p class = 2 m ( E − V ( x )). Notice that the complex exponential has now become a

positive or negative exponential because the solutions are no longer running waves. Consider a barrier of width a with an arbitrary potential on top, , as discussed

on pages 320-322 of Griffiths. For energies

V ( x ) E less than the minimum of , and for

barriers that are sufficiently tall and thick, the transmission probability is dominated by the negative exponential term;

V ( x )

T ∝ e −^2 γ , where = ∫

a pclass x dx 0

h

Going back to the flat-top barrier of problem 2.33 to test this result, the transmission probability in the WKB approximation (after doing the integral

with V ( x )= V 0 ) is

a m ( V E ) T e

− − ∝

(^2) h 2 0

. If we expand the exact result given above in the limit

of tall and wide barrier (i.e. 2 m ( V 0 − E ) >> 1

a h

), the result is

( ) a m (^ V E )

e V

EV E

T

(^220) 2 0

(^4 0) h (^). The pre-factor is on the order of 1, so the exponential

dominates, and is in agreement with the WKB approximate result. Now consider the problem of cold emission from a metal. A metal can be modeled as a potential well of depth EF +Φ, where is the Fermi energy and is the

work function of the metal (see the Figure below). The work function is the energy

E F Φ

required to remove an electron from the Fermi energy in the metal and to set it free. Work functions are typically on the order of a few electron volts for metals.

filled Fermi sea

V(x)

E F

E F+ Φ

No Electric Field Applied

x

If an electric field is applied to the surface of a metal, the potential is modified and it becomes possible for electrons to tunnel out of the metal into free space (this is called cold emission by Fowler-Nordheim tunneling).

Note that the tunnel barrier is not a “flat top” but instead triangular shaped. The potential is given by V ( x )= EF +Φ− eEx , where E now (confusingly!) is the applied

electric field. The calculation of the Fowler-Nordheim tunneling rate is a job for the WKB approximation!

filled Fermi sea

V(x)

0

E (^) F

E (^) F+ Φ

EF

Electric Field Applied

x x=0 (^) x= Φ /eE

T ∝ e −^2 γ , where = ∫

a pclass x dx 0

h

γ ,

and ∫ ( )

Φ = Φ−

eE m eExdx

/

0

h

γ. The upper limit of the integral is the point in x where the

Fermi energy is equal to the potential outside the metal (a classical turning point). The integral can be solved by standard methods and the result is;

⎥ ⎦

eE

m T

2 3 /^2

exp h

The tunneling current is proportional to the transmission probability. Hence the log of the tunnel current should be proportional to the inverse of the electric field strength. This is the characteristic of Fowler-Nordheim tunneling, and data showing this behavior is posted on the class web site.