Form Factors
Form Factors
Form Factors
Shapes
Kyle Foster
November 16, 2011
Form Factors are an intuitive and simple tool used to describe the scattering particles from extended targets. Here we will show how the Form Factor comes about
in the context of the scattering of spinless electrons. A discussion of the more rigorous descriprtion, which includes electron spin a magentic moment, will follow.
The basic experimental approach is consistant with what has already been
discussed in the course: The typical set up has an electron beam with initial
P~f
e
P~i
P~f A ~q
A
P~i
target
A
AU
-
momentum P~i directed at the scattering target. The electrons are deflected through
an angle with a final momentum P~f .We define the momentum transfer as the
vector ~q = P~i P~f .
As with many scattering experiments, the quantity we are interested in is the
d
differential cross section d
of our scattered electrons off our target. This quantity
2
can be measured in the lab and easily connected to QM scattering theory in order
to confirm theory and provide insight to the physical processes at play.
Earlier we had shown that the differential cross section is related to the scattering amplitudes through the relation:
d
k
= |f (, )|2
d
ki
(1)
In the first Born Approximation the initial incoming wave and the outgoing waves
are assumed to be plane waves of the form:
~
k~f (~r) = e
(4)
(5)
We will also define the momentum transfer as ~q~ = k~i k~f . Making use of these
deinitions the first Born Approximation can be written as:
Z
i~
q ~
r
(6)
fB1 = e ~ V (~r)d3~r
This result is still quite general; in order to proceed we will need to assume a specific form for the potential, V (~r). We can describe an extended charge distribution
by Ze(~r) with
Z
(~r)d3~r = 1
(7)
In this case, the potential experienced by an electron located at ~r is given by the
Coloumb potential:
Z
Ze2
(~r 0 ) 3 0
V (~r) =
d ~r
(8)
40
|~r ~r 0 |
3
Substitue this potential into the general expression for the first Born Approximation to the scattering amplitudes f (, ) (eq. 6)
Z
Z
i q~
r
(~r 0 ) 3 0 3
Ze2
~
d ~r d ~r
(9)
fB1 =
e
40
|~r ~r 0 |
~ = ~r ~r 0 and noting that d3 R
~ = d3~r,
Make the substitution R
fB1
i~
q ~
r0
~
(~r 0 ) 3 0 3 ~
d ~r d R
~
|R|
Z
Z i~qR~
i~
q ~
r0
e ~ 3~
Ze2
0
3
0
dR
e ~ (~r )d ~r
=
~
40
|R|
Ze2
=
40
~
i~
q R
~
(10)
(11)
(12)
d
,
d
that:
(13)
The second assumption is that the target is comprised of light nuclei as defined by
2
Z
= Ze
1. This a reasonable assumption for elements up to around Calcium.
137
~c
The form of this differential cross section bears a striking resemblence to that
of Rutherford scattering, the only difference being a factor of 4 cos2 ( 12 ). If an
extended charge distribution is considered as the scattering target, the Mott cross
section becomes:
2
d
Ze cos2 ( 12 )
=
|F (q)|2
(15)
4 1
d
2E sin ( 2 )
Analogous to the spinless electron treatment.
Form Factors
(16)
Upon inspection this equation can be idenitfied as the Fourier transform of the
spatial charge distribution, (~r 0 ). This provides a powerful tool for determining
the spatial charge distribtuions of the nuclei resonsible for the scattering.
The special case of a spherically symmetric charge distribution can be easily
analyzed. In the case of spherical symmetry (and spherical coordinates) we have:
(~r 0 ) = (r0 )
d3~r 0 = r02 d(cos())ddr0
And so eq. 16 can be written as:
Z 2 Z
F (q) =
d
Z
d cos()
= 2
0
Z
= 4
0
iqr cos()
~
(17)
(18)
(r0 )dr0
(19)
0
1
~ iqr cos()
e ~
r02 (r0 )dr0
iqr0
1
sin(qr0 /~) 02 0 0
r (r )dr
qr0 /~
(20)
(21)
This tidy expression is a useful form of the Form Factor given the typical experimental setup and the widespread use of the spherically symmetric assumption.
5
Using this definition of the root-mean-square radius, it can be shown that for many
useful spherically symmetric charge distributions, the Form Factor can be written
as an expansion:
F (qa) = 1 (
q 2 a2
) +
6
(24)
This approximation to the form factor holds as long as the quantity qa remains
small. This result is not entirely unexpected. For low energy collisions (small
qa), we do not expect the electron to penetrate the nucleus and the scattering is
goverend by the electromagnetic interaction. Therefore, all spherically symmetric
charge distribution are going to appear similar from the outside. As a result, the
scattering will be similar as well.
This point is emphasized by Figure 1, which is a plot of the (form factors)2 vs
(qa)2 . The details of the specified charge distributions can be found in Hofstadters
1956 review paper (Rev. Mod. Phys. 28, 214 254 [1956]). They include
Uniform, Gaussian, Exponential, Yukawa and other potentials.
We have shown that, in principle, a measured differential cross section can yield
the Form Factor, F (q). As what pointed out earlier, the Form Factor is simply
the Fourier transform of the spatial charge distribution. So, if the Form Factor is
known, the inverse Fourier transform will result in an expression for the spatial
charge distribution. Therefore, taking measurements in the laboratory of the differential cross section of electrons scattered off a finite target can yield the spatial
charge distribution of the target!
Z
i~
q ~
r0 3
(25)
(~r) F (~q)e ~ d q~
Figure 1: The square of the Form Factor vs (qa)2 for several commonly used and
useful spherically symmetric charge distributions
It should be noted that in order to explicitly invert the Form Factor using the
d
Fourier transform, a complete measure of d
is required. In practice this is a
very difficult measurement to take. Especially at large deflection angles (high
momentum transfer collisions) it is difficult to accurately measure the differential
cross section. As a result, fits and approximations are often used to fill in the
missing data and perfom the inverse transform.
In the case of the spherically symmetric assumption, the resulting charge distribution is known as the Fermi model. Mathematically, the Fermi model is defined as:
(r) =
1
exp [(r c)/z1 ] + 1
(26)
After performing electron scattering experiments using many different elemental nuclei as scattering targets, it became clear that there were an approximate
relations that could be used to describe the parameters of these spherical nuclei.
The relation for the radius parameter c was determined as:
1
c = (1.07 0.02)1013 A 3 cm
(27)
This relation is in agreement with the earlier estimates for the size of the nucleus as
obtained by Rutherford and others. The relation for the skin thickness parameter
t was determined as:
t = (2.4 0.3)1013 cm
(28)
Which is simply a constant. This relation clearly holds for most light nuclei but
should only be regarded as an approximation. Figure 3 (also presented by Hofstadter in 1961) gives a summary of approximate charge density distributions for
8
Up to this point we have been considering the case of elastic scattering, but it
bears noting that inelastic scattering can also offer valuable information about the
nuclear structure of the scattering target.
During an inelastic scattering event, the electron will deposit some extra energy in the target in the form of a nuclear excitation. By measuring the energy of
the electrons after scattering, it is possible to discern some of the atomic structure.
Figure 4 is a plot of scattered electron intensity vs. electron energy. The elastic
scattering peak is clearly visible around 185 MeV(shifted down slightly from the
original beam energy due to target recoil), while some of the inelastic scattering
peaks are also visible at lower energies. This technique is best used to identify
peaks with sufficiently different energy levels (limited by the energy resolution of
your detector), as discerning between elastic and inelastic peaks can be difficult.
10
Figure 4: A plot of electron intensity vs. scattered electron energy of 187 MeV
electrons scattered off a carbon-12 target. Note the distinct peaks with energy differences corresponding to the energies associated with excited states of the target.
11