Rutherford Scattering0
Rutherford Scattering0
SCATTERING
Orientation
A point-like charged projectile approaches a charged point-like target along a
straight line that would pass by the nucleus at a distance b, the impact parameter,
if there were no interaction between them. However, the incoming particle is
deflected by Coulomb forces such that its outgoing direction, a long way away
from the nucleus, is changed by an angle θ, the scattering angle (see diagram
below).
Assumptions
For now the following assumptions will be made; some can be relaxed as
discussed at the end:
(i) Consider only elastic scattering where neither the projectile nor the target is
raised into an excited state by the collision.
(ii) Consider only the case where the target is very much heavier than the alpha
particle; the CM frame is identical to the LAB frame.
(iii) Consider only electromagnetic interactions between the projectile and target
using the Coulomb force.
(iv) Analyse using classical dynamics.
(v) Assume that the projectile only scatters once as it transits the thickness of
the target material.
Before the collision, a long way from the target, the projectile has negligible
electrostatic potential energy and therefore its total energy is purely kinetic E =
1
2
mv02 . It has an angular momentum |r × mv| = mv0 b relative to the target.
During its approach to the target, it reaches a minimum separation distance
rmin , which depends on b. The smallest value of rmin , the distance of closest
approach d, is for a head-on collision with b = 0. At this point its energy is
purely electrostatic potential energy. By conservation of energy:
1 2 zZe2
mv0 =
2 4π0 d
where ze and Ze are the charges of projectile and target. Thus
zZe2 zZe2
d= =
4π0 E 2π0 mv02
1
separation of target and projectile and v is the projectile speed at the particular
point of interest.
This should be equal to the net impulse of the component of the Coulomb
force in the direction of ∆p:
Z Z
zZe2 Z cos α
∆p = dp = F∆p dt = dt.
4π0 r2
Here the integral is taken from t = 0 to the final position t = ∞. At the latter
point, α = π/2 − θ/2 and at the former, α = −(π/2 − θ/2).
In order to do the integral it is useful to transform to an integral over angle
rather than time. The tangential component of the angular momentum about
the target at a point along the trajectory is mr2 dα/dt and by conservation of
angular momentum:
dα
mv0 b = mr2
dt
2
dt dα
2
= .
r v0 b
The final integral is then:
cos 2θ
2
dR = Φπd dθ.
4 sin3 2θ
All these particles are scattered into a range of angles from θ to θ + dθ, which
presents a solid angle of dΩ = 2π sin θdθ. The differential cross section is the
scattering probability per unit incident flux per solid angle for one target, hence:
dσ 1 dR cos 2θ 1
= = d2 3 θ dθ
dΩ Φ dΩ 4 sin 2 2 sin θdθ
!2
d 1
=
4 sin4 θ
2
!2
zZe2 1 1
=
4π0 4E sin4 θ
2
This is known as the Rutherford scattering formula.
3
If the target is not massive compared to the projectile, E and θ are reinter-
preted as being measured in the centre-of-mass system and the above formula
then gives the cross section in that centre-of-mass frame. In this case, the labo-
ratory beam energy is related to E via
mt
E = ELAB ×
mp + mt
Discussion
At very small angles the Rutherford formula suggests that the cross section
becomes infinite. This turns out not to be a realistic problem. Firstly, it is
difficult to measure at very small angles due to the physical width of any real
beam. Secondly, small angles correspond to large impact parameters. When
the impact parameter gets comparable to atomic radii, atomic electrons shield
the nuclear charge from the projectile and the cross section departs from the
Rutherford prediction.
The derivation above uses classical mechanics, but it turns out to be the same
in the case of non-relativistic quantum mechanics by mathematical chance! It
does need revising in relativistic conditions.
In reality, a projectile could scatter multiple times within the target thickness.
If the mean angle per single scatter is Θ, then a random-walk analysis√suggests
that after n multiple scatters the total angular deviation would be Θ n. This
suggests that the amount of multiple scattering at a particular angle will increase
according to the square root of the target thickness, whereas single scattering
varies linearly because the chance of scattering is directly proportional to the
number of nuclei available to scatter from. This allows an experimental dis-
tinction between the two which indicates that plural or multiple scattering only
effects the angular distribution at small angles to the beam axis.