Sol Manual Atomic Physics

lOMoARcPSD|17631832
sol manual atomic physics
atomic physics (Seoul National University of Science and Technology)
Studocu is not sponsored or endorsed by any college or university

Downloaded by Bibi Huda (bibihuda936@gmail.com)
lOMoARcPSD|17631832
Solutions to the Exercies of

C. J. Foot’s Atomic Physics
Chenchao Zhao
Department of Physics, Beijing Normal University, Beijing, China
(Dated: June 28, 2011)
1 Early atomic physics

Key formulas:

1 1 1
=R − (1.1)
λ n2 n′2
mH
RH = R (1.2)
me + mH
~2
a0 = = 0.529 × 10−10 m (1.3)
(Ze2 /4πǫ0 )me
Ze2 /4πǫ0 1
E=− ∝ Z2 (1.4)
2a0 n2
α2
∆E = 2 E (1.5)
n
e2 /4πǫ0
α= = 1/137 (1.6)
~c
f
= R∞ ((Z − σK )2 − (Z − σL )2 ) (1.7)
c
~ω 3 1
ρ(ω) = 2 3 (1.8)
π c exp ~ω/kT − 1

N2 N1 ~ω
= exp − (1.9)
g2 g1 kT
2 2
e x0 ω 4
P = (1.10)
12πǫ0 c3
eB
ΩL = (1.11)
2me
6πǫ0 me c3
τ= (1.12)
e2 ω 2

lOMoARcPSD|17631832
1.1 Isotope shift

To find wavelengths of Balmer-α transitions, we set n = 2 and n′ = 3 as
36
in Eq (??) Then λ = 5R , hence
36
λH − λD = (1/RH − 1/RD )
5
36me 1 1
= −
5R∞ mH mD
18me
≈ = 0.18 nm
5mH R∞
where mD ≈ 2mH .
1.2 The energy levels of one-electron atoms

Since (1.4), and let m, n be the quantum numbers of He+ and H, ne-
glecting the isotope shifts, the energy levels in agreement are those with
1/n2 = 4/m2 , namely m = 2n.
Those wavelengths should have the ratio
λH RHe
=
λHe RH
mHe (mH + me )
=
mH (mHe + me )
4mH (mH + me )
≈
mH (4mH + me )
≈ 1.00041
which is in accord with the data set, 1.00041.
1.3 Relativistic effects

With n = 4, Eq (1.5) gives
λ E 4
= = 2 = 75076 (1.13)
∆λ ∆E α
for the fact that λ = ch/E and
dλ = −ch dE/E 2
= −(ch/E) (dE/E)
= −λ (dE/E)
This corresponds to a grating of ∼ 105 grooves.

lOMoARcPSD|17631832
1.4 X-rays
√
Eq (1.7) reduces to f = 3cR∞ Z 2 /4 when Z ≫ σ. Therefore, f ∝Z
1.5 X-rays
Since E = hf = 3/4(hcR∞ )Z 2 = 13.6 eV × 0.75Z 2 , then it predicts
absorptions at around 6.4 keV and 6.9 keV.
1.6 X-ray experiments

See http://www.physics.ox.ac.uk/
history.asp?page=Exhibit10
1.7 Fine structure in X-ray transitions

Energy of the electron in the L-shell should be
E = (82/2)2 × 13.6 = 22.9 keV
and Eq (1.5) gives
α2
∆E = E = E/75076 = 0.3 eV
n2
But Kα transition means an energy of
Eα = 10.2(Z − 1)2 eV = 66.9 keV
then
∆E
= 4.5 × 10−4 %
Eα
1.8 Radiative life time

Eq (1.10) provides the power of dipole radiation which is rate of change
of energy, for the circular motion, the power doubles since a circular motion
can be decomposed into two linear oscillations,
12πǫ0 c3
τ = E/P = ~ω ∝ 1/(er)2 ω 3
2e2 r 2 ω 4
For photons of wavelength 650 nm, ω = 2.89 × 1015 rad/s; let r = a0 , the
life time will be 2.7 × 10−7 s.

lOMoARcPSD|17631832
Table 1: Frequency shifts

B [T] 3 × 10−5 1
ΩL [MHz] 2.6 8.8 × 104
ΩL /(1014 Hz) 2.6 × 10−8 8.8 × 10−4
1.9 Black-body radiation

Setting N2 = 0.1N, N1 = 0.9N, g1 = 1, g2 = 3, Eq (1.9) gives
exp(~ω/kT ) = 27
provided the wavelength λ = 600 nm, ω = 2πc/λ = 3.13×1015 rad/s. There-

fore, T = 7.23 × 103 K and Eq (1.8) gives the density
ρ = 4.70 × 10−16 J s/m3
1.10 Zeeman effect

The Larmor frequency is given by Eq (1.11), and the Earth magnetic
field is about 3 × 10−5 T, then the frequency shifts are listed in Table 1.
1.11 Relative intensities in the Zeeman effect

One circular motion can be decomposed into to two orthogonal sinusoidal
motions. Let I denote the intensity of three eigen-oscillations of the electron.
Then, we have
• Along the magnetic field, only circularly polarized lights can be ob-
served and the intensities are Iσ+ = Iσ− = I;
• Perpendicular to the magnetic field, motion along z direction and pro-

jected horizontal motions can all be observed, and
• 2Iσ+ = 2Iσ− = Iπ = I since the projected horizontal motions are only

half of the circular motions.
Therefore,
(a) Total intensity perpendicular to the field is 2I;
(b) Ratio of intensities received along to perpendicular to the field is

2I/2I = 1.

lOMoARcPSD|17631832
1.12 Bohr theory and the correspondence principle

Energy of hydrogen atoms takes the form
1
E = K + V = −K = − me v 2
2
But
v2 e2 /4πǫ0
me =
r r2
Then
1 e2 /4πǫ0 e2 /4πǫ0
E=− , dE = dr
2 r 2r 2
Hence
e2 /4πǫ0
ω = ∆E/~ = ∆r
2r 2 ~
The angular frequency is also given by
v2 e2 /4πǫ0
ω2 = =
r2 me r 3
Equating the two expressions of ω, we have
s
r~2 √
∆r = 2 2
= 2 a0 r
me e /4πǫ0
which is equivalent to
∆r
= 2(a0 r)1/2 .
∆n
Approximate the equation above by the corresponding differential equa-
√
tion, namely r ′ = 2 a0 r, and the solution turns out to be
r = a0 n 2 . (1.14)
1.13 Rydberg atoms

Eq (1.4), then
dE e2 /4πǫ0 d (n−2 ) e2 /4πǫ0

=− = = Ry/n3
dn 2a0 dn a0 n 3
and for n = 50, ∆E = 1.1 × 10−4 eV. The radius of such atoms is around
2500a0 or 132 nm according to Eq (1.14).

lOMoARcPSD|17631832
2 The hydrogen atom

Key formulas
s
(2l + 1) (l − m)! imφ m
Ylm (θ, φ) = (−1)m e Pl (cos θ) (2.1)
4π(l + m)!
Z π
dθ eimθ e−inθ = 2πδmn (2.2)
−π
Z π
cos mθ cos nθ
dθ = πδmn (2.3)
−π sin mθ sin nθ
Z π
dθ sin mθ cos nθ = 0 (2.4)
−π
Z Z 2π Z π Z 2π Z 1
dΩ = dφ sin θdθ = dφ d(cos θ) (2.5)
0 0 0 −1
β
Es-o = β hS · Li = (j(j + 1) − l(l + 1) − s(s + 1)) (2.6)
2
~2 e2 1
β= (2.7)
2me c 4πǫ0 (na0 ) l(1 + 21 )(l + 1)
2 2 3

2 1 ∂ ∂ 1 ∂2
L =− sin θ + (2.8)
sin θ ∂θ ∂θ sin2 θ ∂ 2 φ
2.1 Angular-momentum eigenfunctions

From the table of spherical harmonics and Eq (2.2), (2.5), we have
• h l1 m| l2 ni ≡ 0 if m 6= n, therefore
h 11| 00i = 0
And also
Z 1
h 10| 00i = (constant) × cos θ d(cos θ) = 0
−1
• For l = 1, 2, we only need to show that h 10| 20i = h 11| 21i = 0.

Through inspections, the integrands as functions of cos θ are both odd,
hence the integrals vanish.
2.2 Angular-momentum eigenfunctions

• According to Eq (2.1),
s
l−1 2l + 1 i(l−1)φ l−1
Yl,l−1 = (−1) e Pl (cos θ)
4π(2l − 1)

lOMoARcPSD|17631832
• It is convenient to write,
h l, l − 1| l − 1, l − 1i = hl, l| (Lˆ− )† |l − 1, l − 1i
= hl, l| Lˆ+ |l − 1, l − 1i
=0
2.3 Radial wavefunctions

With n = 2, l = 1 the integral reads,
Z ∞
1 1 2
= R (r)r 2 dr
r 3
0 r 3 2,1

Z 5 2 2 2 −r/a0
Z ∞
dr
= √ r e
0 r 2a0 3
= 1/(24a30 )
Invoking the rhs formula, namely

2
1 1 Z
= 1 (2.9)
r3 l(l + 2 )(l + 1) na0
yields the same result 1/(24a30 ).
2.4 Hydrogen
The probability is given by
Z rb Z rb
(r/a0 )2 d(r/a0 ) −2r/a0
r 2 dr |ψ(r)|2 = e
0 0 π
Z r
ǫ:= ab
y 2 dy −2y 0
= e
0 π
≈ ǫ3 e−2ǫ /π ∼ (rb /a0 )3
The electronic charge density of this region is
e(1 − rb /a0 )
ρe ≈ e|ψ(rb /2)|2 =
a30 π
2.5 Hydrogen: isotope shift, fine structure and Lamb shift

The mass ratio of electron to proton is ∼ 5 × 10−4 and isotope shift is
of the same order, namely, if λ = 600 nm, order of isotope shift will be
δν̄isotope ∼ (5 × 10−4 ) · (5 × 105 GHz) = 250 GHz

lOMoARcPSD|17631832
Relativistic effect is of the order of α2 ≈ 5 × 10−5 , then the wave number

difference reads
δν̄fs ∼ (5 × 10−5 ) · (5 × 105 GHz) = 25 GHz
Lamb shift is 1/10 of the fine structure shift, that is
δν̄Lamb ∼ 2.5 GHz
A Fabry-Perot étalon of finesse F = 100, width d = 1 cm is supposed to

resolve
1
δν̄ = = 0.5 × 3 × 108 ∼ 0.1 GHz (2.10)
n(2d)F
but Doppler effect attenuates the resolution to ∼ 0.7 GHz ∼ 1 GHz.
Therefore, isotope shift, fine structure can be resolved but Lamb shift
approaches the limit of the apparatus and hence cannot be accurately ob-
served.
2.6 Transitions
From Eq (1.12), we have
Ultraviolet 100 nm 0.45 ns
Infrared 1000 nm 450 ns
2.7 Selection rules

Following similar arguments as in Problem 2.1 and notice there is an
additional cos θ in the integrand.
2.8 Spin-orbit interaction

Calculations based on Eq (2.6) give
β β
Ej = l Ej ′ = − (l + 1)
2 2
and the mean of the two is
Ē = (2j + 1)Ej + (2j ′ + 1)Ej ′ = β[(l + 1)l − l(l + 1)] = 0
2.9 Selection rule for the magnetic quantum number

The integral is readily obtained in text of section 2.2.1, and the result
follows the same arguments of Problem 2.1.

lOMoARcPSD|17631832
2.10 Transitions
(a) The wavefunction takes the form Ψ(t, r, θ) = Aψ1 eiE1 t/~ + Bψ2 eiE2 t/~
with A ≫ B,
ψ1 = R1,0 Y00 , ψ2 = R2,1 Y1,0
and
|Ψ|2 ≈ A2 |ψ1 |2 + 2|AB||ψ1∗ ψ2 | cos(ω12 t)
The second term can be written as
f (r)r cos θ cos(ω12 t) = f (r)r · ẑ cos(ω12 t)

= f (r) z cos(ω12 t)
The sketch of the orbital is as follows (Figure 1)
time
Figure 1: Contour of electron density, or the orbital of wavefunction

Ψ(t, r, θ) = Aψ1 eiE1 t/~ + Bψ2 eiE2 t/~ with A ≫ B, ψ1 = R1,0 Y00 , ψ2 =
R2,1 Y1,0 during one period of oscillation.
(b) If Π̂ |Ψi = ±1 |Ψi, namely the state exhibits parity, then
hri = hΨ| Π̂† r Π̂ |Ψi = − hri
hence hri = 0. But the Hamiltonian of hydrogen atom commutes with

Π̂, then the eigenstates are of specific parities. The only possibility for a
none-vanishing hri is to require ψ1 and ψ2 to possess opposite parities.
(c) Now set a0 = 1, the radial integral yields

Z ∞ 5
dr 3 1 2
√ r 4 e− 2 r = √
0 6 6 3
The angular integral is exactly 1. The total electric dipole moment is
5
1 2
−eD = − √ ea0 cos ωt ẑ
6 3
where a0 is put back through dimension analysis.

lOMoARcPSD|17631832
(d) The density distribution should be more or less similar to Figure 1 for
a constant φ but it becomes apparent if one writes c = ωt − φ or
φ = ωt − c
that a flock of charge is circulating about the z-axis.
(e) The case of (a) is akin to the vertical motion of the electron while (d)
corresponds to right-handed circular motion. The motions are characterized
by the ml but the role of ground state 1s is crucial as pointed out in (b). The
1s state is an exponential decay which binds the electron to the region around
nucleus and hence it is the quantum analogue of the classical restoring force
−ω 2 r.
2.11 Angular eigenfunctions: Yll

(a) Raising operator is given as
L+ = eiφ (∂θ + i cot θ∂φ ) (2.11)
Then we write
(∂θ + i cot θ∂φ ) (Θeimφ ) = 0
it is equivalent to
Θ′ cos θ
=m
Θ sin θ
(b) The solution is Θ = sinm θ. Applying L2 (see (2.8)) yields
L2 (Θeimφ ) = m(m + 1)Θeimφ
2.12 Parity and selection rules

If l1 − l2 is even, then
Iang = (−1)l1 −l2 +1 Iang = −Iang
which implies the integral vanishes.

To have non-vanishing angular integrals, l1 − l2 must be odd. But the
parity of spherical harmonics is
ΠYlm = (−1)l Ylm (2.12)
Therefore, the initial and final states must exhibit different parities.
2.13 Selection rules in hydrogen

The wavelengths corresponds to energy 0.306, 0.663, 1.890 eV and they
are transitions from 5 to 4, 4 to 3 and 3 to 2 respectively.
10

lOMoARcPSD|17631832
3 Helium
3.1 Estimate the binding energy of helium
(a) The total Hamiltonian is the sum of two individual and one interaction
Hamiltonians,
Ĥ = Ĥ1 + Ĥ2 + Ĥint (3.1)

~2 ˆ 2 Z ẽ2
Ĥi = − ∇ − (3.2)
2m i ri
ẽ2
Ĥint = (3.3)
|r 1 − r 2 |
where ẽ2 = e2 /4πǫ0 .
(b) The energy

~2 Z ẽ2
E(r) = −
2mr 2 r
assumes minimum at
~2
rm =
Zmẽ2
and it is
Z 2 mẽ4
E(rm ) = −
2~2
(c) The repulsive energy, namely electron-electron interaction, is

ẽ2 ẽ2 Zmẽ4
Eint = ≈ =
r12 rm ~2
The ionization energy of one electron, according to the estimated energies,
is Eion = 0 But experiment gives Eion = 24 eV, then the average distance
between the two electrons should be greater than rm .
(d) For Si12+ , Eion = (142 − 14 × 2) × 13.6 = 2285 eV, excluding the
repulsion we have E = 2666 eV. Comparing with results of helium, repulsion
gets irrelevant for larger Z.
3.2 Direct and exchange integrals for an arbitrary system

(a) The direct and exchange integrals are respectively,
Z
ẽ2
J = d3 x1 d3 x2 |uα (r1 )|2 |uβ (r2 )|2 (3.4)
r12
Z 2
ẽ
K = d3 x1 d3 x2 u∗α (r1 )u∗β (r2 ) uα (r1 )uβ (r2 ) (3.5)
r12
11

lOMoARcPSD|17631832
Yet, |Ai = (2)−1/2 (|αβi − |βαi),

1
hA| Ĥ ′ |Ai = (hαβ| − hβα|)Ĥ ′ (|αβi − |βαi) (3.6)
2
hαβH ′ αβi − hαβH ′ βαi + α ↔ β
= (3.7)
2
Note that
αβH ′ αβ = βαH ′ βα
∗
αβH ′ βα = βαH ′ αβ
Therefore, it holds for real-valued u(r) that
hA| Ĥ ′ |Ai = αβH ′ αβ − αβH ′ βα = J − K
(b) The symmetric wavefunction is construct as
|Si = (2)−1/2 (|αβi + |βαi)
The inner product reads
(1 − 1 + h αβ| βαi − c.c)

h A| Si = = Im h αβ| βαi
2
Again, for real u(r), h A| Si are orthogonal.
(c) Since Ĥ ′ is invariant under the interchange of particle labels, let Σ

denote such an operation, then
hA| Ĥ ′ |Si = hA| Σ† Ĥ ′ Σ |Si = − hA| Ĥ ′ |Si
Hence hA| Ĥ ′ |Si = 0.
3.3 Exchange integrals for a delta-function interaction

(a) The Hamiltonian is simply the kinetic energy,
~2 2
H=− ∂
2m x
For the fact that
u′′0 = −(π/l)2 u0 u′′1 = −(2π/l)2 u1 ,
then the energies are
~2 π 2 2~2 π 2
E0 = E1 =
2ml2 ml2
12

lOMoARcPSD|17631832
(b) The direct integral is

2 Z
2 πx1 πx2
J= dx1 dx2 sin2 ( )aδx1 −x2 sin2 ( )
l l l
2 Z l
2
=a sin4 (πx/l)dx
l 0
Z
4 π 4 ′ ′ 3a
=a sin x dx =
πl 0 2l
The exchange integral is the same thing, K = J. Then the energy shift will
be 3a/l. There is only one state, the symmetric one.
(c) All the eigenstates takes the form,

r
2 nπx
un (x) = sin
l l
and trivially Jmn = Kmn since the delta potential identifies the two coordi-
nates. Hence, the antisymmetric state gives no rise to energy shift. Actually,
the antisymmetric part does not exist at all when interaction is considered,
because delta interaction rules out the possibility of ‘no-touch’, otherwise
there is no interaction, and hence no energy shift!
(d) See Figure 2.
0.8
0.6
0.4
0.2
0
-0.2
1 -0.4
-0.6
0.6 -0.8
0.2
-0.2
-0.6
-1
1
0.8
1 0.6
0.8 0.4 x1
0.6
0.4 0.2
x2 0.2
00
Figure 2: The horizontal axes are the coordinates x1 and x2 , and the vertical
axis marks the values of function ψ = u1 (x1 )u2 (x2 ).
(e) Possible total spins are 0 and 1. The symmetric spatial functions
corresponds to 0-spin, and the antisymmetric spatial functions should have
spin-1 or spin-0 but they do not exist.
13

lOMoARcPSD|17631832
(f ) The wavefunctions are independent of particle masses, therefore all the

mathematics are invariant once the states are given. The energy levels are
still
E+ = E1 + E2 + 2J E− = E1 + E2
but the difference is that the levels are not related to the exchange symme-
tries of the particles.
3.4 A helium-like system with non-identical particles

The exoticon-exoticon system (identical fermions) is indifferent with the
electron-electron system except for richer spin configurations for the two
symmetries. Yet no restriction upon symmetries of particle exchange is
placed on exoticon-electron system, therefore, the spatial orbitals can be
freely occupied by the two fermions, and for the fact that spin is not included
in the Hamiltonian, hence the energies levels are left unaltered.
3.5 Integrals in helium

Set a0 = 2Z, the integral in the curly brackets is
−r2 e−2r2 /2 − e−2r2 /4 + 1/4
and
ẽ2 5
J= Z = 34 eV
2a0 4
3.6 Calculation of integrals for 1s2p configuration
6 R
R10
R20
21
3
Rnl
-10 2 4 6 8 10
r / a0
Figure 3: The plot of R10 , R20 and R21 with Z = 2.
The integral
ẽ2
J1s2p = − × 0.00208 = 0.0283 eV
2a0
14

lOMoARcPSD|17631832
3.7 Expansion of 1/r12

Expansion in terms of spherical harmonics is the following
∞ k
1 1 X r1 k 4π X ∗
= Ykq (θ1 , φ1 )Ykq (θ2 , φ2 ) (3.8)
r12 r2 r2 2k + 1
k=0 q=−k
(a) Setting k = 0, 1, we have

1 1
≈ {4π/4π
r12 r2

r1 4π 3
+ (cos θ1 cos θ2 + sin θ1 sin θ2 cos(φ1 + φ2 ))
r2 3 4π

1 r1
= 1 + cos θ12
r2 r2
(b) Mathematically, the expectation values of 1/r12 are always sandwiched

by ‘bra-ket’ where the phase factors cancels; this is seen in the expression
for K1s nl . Physically, the quantum number m is responsible for magnetic
interactions but Coulomb repulsions have no interests in that.
(c) Due to the orthogonality relations of spherical harmonics, the terms

in 1/r12 are eliminated excluding the one with quantum numbers lm. For
l = 1, this corresponds to the second term in the expansion as shown in part
(a).
(d) This follows the arguments in part (c), the given l samples out the
order k in Eq (3.8).
4 The alkalis
4.1 Configuration of the electrons in francium
Fr = [Rn]7s1
Rn = [Xe]5d10 5f 14 6s2 6p6 6d10
Xe = 1s2 2s2 2p6 3s2 3p6 3d10 4s2 4p6 4d10 4f 14 5s2 5p6
4.2 Finding the series limit for sodium

Energy levels of sodium obey the rule
Ry 13.6 eV
E=− =− (4.1)
(n∗ )2 (n − δ)2
15

lOMoARcPSD|17631832
Then
∆E ∝ E 3/2 (∆E)2 ∝ E 3
The topmost level must have ∆E = 0, and that energy corresponds ioniza-
tion energy
p which was found to be 5.1 eV. The effective principal number
∗
n = 13.6/5.1 = 1.63.
4.3 Quantum defects of sodium

From formula Eq (4.1), the quantum defects are
3s 1.37 4s 1.34 5s 1.33 6s 1.35

the average quantum defect is 1.35 ± .02. Assume that the quantum defect
for 8s is still 1.35, the binding energy will be
ENa 8s = −13.6/(8 − 1.35)2 = 0.31 eV
while for hydrogen EH 8s = 0.21 eV.
4.4 Quantum defect

Quantum defect of Rb 5s is 3.19 with which that of 7s is approximated.
Therefore the energy difference reads

1 1
∆E = 13.6 − = 3.22 eV
(5 − 3.19)2 (7 − 3.19)2
Then the wavelength of the two identical photons are
hc
λ= = 771 nm
∆E/2
4.5 Application of quantum defects to helium and helium-

like ions
Throught a direct calculation, the wavelength from 1s3d to 1s2p is ob-
tained as 625 nm, compared with 656 nm of hydrogen Balmer-α.
Quantum defects can be evaluated by Eq (4.1), they are listed in Table
2. It is readily seen that δs > δp > δd .
Table 2: Quantum defects of helium

1s2s 1s2p 1s3s 1s3p 1s3d
0.2356 0.0275 0.2265 0.0287 0.0026
To estimate binding energy of 1s4l states, assume the quantum defects

for l = 1, 2, 3, 4 as 0.23, 0.28, 0.0026, 0.00. Then the binding energies read
0.96, 0.98, 0.85, 0.85 eV
16

lOMoARcPSD|17631832
The second ionization energy of Li+ is
Z 2 Ry
IE2 = −(E ′ − E1s4f ) = 72.24 + = 75.64 eV
42
where Z = 2.
4.6 Quantum defects and fine structure of potassium

The wave numbers can be put into three groups (3 row vectors), and
take average and difference within each group. The data set becomes ν =
(hνi , δν). The energies are calculated through E = hc/λ, and the data
set for energies reads E = (hEi , δE). Given that the ionization energy of
potassium IE = 4.34 eV, the energy levels are E − IE from with we can find
out the effective principal quantum numbers and quantum defects. They
are n∗ = 2.23, 3.26, 4.28 and δl = 1.77, 1.73, 1.72; the pattern confirms that
they corresponds to 4p, 5p, 6p. The fine structure splitting δE/E divided
by α2 are, for 4p, 5p, 6p, 49.18, 33.45, 26.38.
According to Landé formula,
Zi2 Zo2
∆Efs = α2 Ry (4.2)
(n∗ )3 l(l + 1)
the ratio of ∆E/(n∗ )3 should be 1:1:1 and it turns out to be 1.00 : 0.99 :
1.03. The transitions of 7p should produce spectral lines of wavelengths
321.84 nm, 321.93 nm.
4.7 The Z-scaling of fine structure

From Eq (4.2), fine structure splitting of ions scales as Z 4 , then for
Na+10 ,it is 1.3 × 10−5 × 114 = 0.19 eV; for neutral ions it scales as Z 2 ,
namely, for sodium atom it should be 0.0016 eV.
4.8 Relative intensities of fine-structure components

(a) In this problem J = 1/2, 3/2, 5/2, hence the weight ratio should be
2 : 4 : 6 = 5 : 10 : 15 = 5 : (1 + 9) : (1 + 9 + 5).
(b) Possible electrical dipole transitions are D5/2 −F5/2 , D5/2 −D7/2 , D3/2 −
F5/2 with intensities 1 : 20 : 14.
Hint: Plot out energy levels and then count and take ratios.
4.9 Spherical symmetry of a full sub-shell

Pl
Show that m=−l |Ylm (θ, φ)|2 is spherically symmetric. For l = 1, the
sum reads
3 3
s1 = (cos2 θ + sin2 θ) =
4π 4π
17

lOMoARcPSD|17631832
Associated Legendre polynomials are defined as follows

m
d
Plm (x) = (−1)m (1 − x2 )m/2 Pl (x) (4.3)
dx
and they are related to spherical harmonics in Eq (2.1).
5 The LS-coupling scheme

5.1 Description of the LS-coupling scheme
Central-field approximation is achieved by imposing spherical symmetry
on electron distributions where electrons are seated in several spherically
symmetric layers. The central field Hamiltonian is the sum of individual
electron Hamiltonian and hence the Shrödinger equations are decoupled.
The arrangements of pseudo-independent electrons constitute the electron
configurations. (a) The residual electrostatic interactions couples the elec-
trons and give rise to splitting of energy levels. (b) Magnetic spin-orbit
interactions include another degree of freedom, the spin, which ‘rotates’ the
original eigenstates resulting in further splittings.
5.2 Fine structure in the LS-coupling scheme

The spin-orbit Hamiltonian reads
hs1 · Si hl1 · Li hs2 · Si hl2 · Li

Hs−o = β1 S· + β2 S· := βLS S · L
S(S + 1) L(L + 1) S(S + 1) L(L + 1)
(5.1)
3
For 3s4p P configuration, S = 1 = L, therefore the relation βLS = β4p /2
holds if
2
1X
β4p = βi hsi · Si hli · Li
2
i=1
but consider the symmetry of electrons, the right hand side is just
X
βi sik lik = β4p
i
5.3 The LS-coupling scheme and the interval rule in calcium

The ground configuration of calcium: 1s2 2s2 2p6 3s2 3p6 4s2 .
The triple lines comes from 3 P term with J = 0, 1, 2 with and interval
ratio ∼ 2. The first three of muiltiplet of six lines come term 3 DJ=1,2,3 with
an interval ratio ∼ 1.5; the rest must be 3 P, 3 D and 3 F from selection rules.
18

lOMoARcPSD|17631832
5.4 The LS-coupling scheme in zinc

The zero comes from 1 S0 , then it leaps to triplet 4s4p3 P0,1,2 , next it
jumps to singlet 4s4p1 P1 and 4s5s 3 S1 , 1 S0 .
5.5 The LS-coupling scheme

The interval rule accurately indicates that the four levels of Mg are 3 P0,1,2
and 1 P1 of 3s3p. Heavier Fe14+ shows a weaker interval pattern (2.4 rather
than 2) due to intercombination and therefore ∆S 6= 0 transitions occurs in
Fe14+ but not Mg.
5.6 LS-coupling for configurations with equivalent electrons

(a) For np2 configuration (2 out of 6 states, 15-fold degenerate), ML =
ml1 + ml2 and MS = ms1 + ms2 , then ML = ±2, ±1, 0 and MS = ±1, 0.
Considering Pauli exclusion, the 3 D and 3 S can be directly eliminated pre-
serving 1 D (5-fold) and 1 S (1-fold). Then what’s left is 15 − 5 − 1 = 9 and
3 P (9-fold) fits.
(b) The first 3 lines are the spin-orbit splittings of 3 P term, the latter
two are 1 D and 1 S. The weak emissions lines near 1 D indicates deviations
from LS-coupling scheme.
(c) In order to make MS = 2 and ML = 2 for six d-electrons, ms =
( 2 , 2 , 2 , 2 , 2 , − 12 ) and ml = (2, {0}5 ) where curly bracket means ‘all combi-
1 1 1 1 1
nations.’ To meet Pauli exclusion principle and maximize MS , the longest

sequence of aligned ms is up to 5 since ld = 2, namely, five electrons are
equally spread out, therefore ms takes the form above and ML = ml (ms =
1
2 ) ≤ ld = 2.
5.7 Transition from LS- to jj-coupling

For configuration 3p4s, Hre > Hs-o due to a higher correlation energy (ex-
change integral) from inner core, while 3p7s is just the opposite, the electrons
seems more independent. Therefore, LS-coupling is proper for 3p4s and jj-
coupling for 3p7s. The former gives two terms 3 P and 1 P while the latter
gives four levels: (3/2 = (1+1/2), 1/2)J=2 , (1/2 = (1−1/2), 1/2)J=1 , (1/2 =
(1 − 1/2), −1/2)J=0 , (3/2 = (1 + 1/2), −1/2)J=1 . It is shown in Figure 5.10
in the text.
5.8 Angular-momentum coupling schemes

Take the 0-2 and 1-1 difference of np(n + 1)s for n = 3, 4, 5, then take
their ratios, I find ∆02 /∆11 = 0.23, 0.70, 0.89. As n grows, the LS-coupling
transitions to jj-coupling, namely ∆02 /∆11 → 1.
The g-factor gJ = 1.06 should belong to 1 P (spin-0), the deviation from
1 is due to the mixing with other 3 P wavefunctions.
19

lOMoARcPSD|17631832
5.9 Selection rules in the LS-coupling scheme

(a) No, ∆l = 2; (b) no, ∆J = 2; (c) yes; (d) no, ∆l = 0; (e) no, ∆J = 2.
4d9 5s5p is the candidate that mixes with 4d10 5p.
5.10 The anomalous Zeeman effect

Electric dipole transition rules require ∆MJ = 0, ±1. The energy is
given by
EZM = gJ µB BMJ (5.2)
The transitions 3 S1 (gJ = 2)-3 P2 (gJ = 23 ) take place at MJ : [1, 0, −1] →
[2, 1, 0, −1, −2], then (gJ MJ ) : [2, 0, −2] → [3, 23 , 0, − 32 , −3]. Therefore,
∆(gJ MJ ) = [(1, 32 , 2); (− 21 , 0, 12 ); (−2, − 23 , −1)] The spacing is µB B/2 =
14 × 1/2 = 7 GHz with B = 1 T

Formula for the g-factor:
3 S(S + 1) − L(L + 1)
gJ = + (5.3)
2 2J(J + 1)
then gJ [ 3 S1 ] = 2 and gJ [ 3 P1 ] = 23 . Compared with the previous problem

(or Figure 5.13), since ∆J = 0, ∆MJ 6= 0, there are only six lines left.

With previous experience and the fact J = 0 states have no Zeeman
splitting, possible transitions are:
3
P1 → 3 D2 (1), 3
P2 → 3 D1 (2)
and the g-factors are

3 3
P1 = 3/2 P2 = 3/2
3 3
D1 = 1/2 D2 = 7/6
The quantities
(gJ MJ )1 : [3/2, 0, −3/2] → [7/3, 7/6, 0, −7/6, −7/3]
and
(gJ MJ )2 : [3, 3/2, 0, −3/2, −3] → [1/2, 0, −1/2]
Therefore, 6∆(gJ MJ )1 = [+9, +7, +5; +2, 0, −2; −5, −7, −9] fits the data.
20

lOMoARcPSD|17631832
5.13 The anomalous Zeeman effect in alkalis

(a) gJ [( 2 S1/2 , 2 P1/2 , 2 P3/2 )] = (2, 2/3, 4/3)
(b) The g-factors are
gJ [3s 2 S1/2 ] = 2 gJ [3p 2 P3/2 ] = 4/3
It is a set of ‘2 to 4’ transitions, then there are six lines. The intervals are
4/3×[3/2, 1/2, −1/2, −3/2]−2×[1/2, −1/2] = [(5/3, 1)(1/3, −1/3)(−1, −5/3)]
and they are equally spaced by 2/3 µB B = 2/3 × 14 × 1 = 9.33 GHz

(c) There are only two lines of same intensity, no ∆MJ = 0 transition.
(d) The energy of fine structure is 517.96 GHz this corresponds to a huge
magnetic field 517.96/9.33 = 55.5 T.
5.14 The Paschen-Back effect

Spin, in this case, is a spectator variable, then J = L, hence gJ =
3/2 − 1/2 = 1, same as in the normal Zeeman effect.
6 Hyperfine structure and isotope shift

6.1 The magnetic field in fine and hyperfine structure
The magnetic field is given by
2
B e = − µ0 gs µB |ψns (0)|2 s (6.1)
3
where
Z3
|ψns (0)|2 =
πa30 n3
Setting Z = 1, s = 1/2 and n = 1, 2 for 1s and 2s, we have magnetic field
flux density at center as 16.7 T and 2.1 T respectively. The magnetic field
felt by an orbiting electron is B = βl where β is the spin-orbit interaction
constant given by Eq (2.7). Then the field that 2p-electron in hydrogen
experiences is 0.2 T.
6.2 Hyperfine structure of lithium

Hyperfine and fine structure splittings are both proportional to the mag-
netic moment (spin) and in turn, proportional to the mass. Therefore, hy-
perfine structure is of order me /Mp smaller than fine structure.
Let P = max{I, J} and Q = min{I, J}, then F = {P + Q, P + Q −
1, . . . , P − Q + 1, P − Q} and hence there are 2Q + 1 values of F .
21

lOMoARcPSD|17631832
Since 2s has no fine structure, the fine structure totally comes from
2p 2 P3/2,1/2 with maximum J = 3/2 on the top. For 6 Li (at J = 3/2),
2I + 1 = 3, then I6 = 1 and for 7 Li, 2I + 1 = 4, I7 = 3/2, and notice both
are no larger than 3/2.
Interval Rule With computer, I find the splittings for I = 1 is 3 : −2 : −5,

the interval ratio is then 5 : 3; for I = 3/2, the splittings are 4.5 : −1.5 :
−5.5 : −7.5 with interval ratio 3 : 2 : 1. The data confirms our prediction of
nuclear spins given the hyperfine constants for e to g negative.
The hyperfine shift of J = 1/2 level are 1 : −2 and 1.5 : −2.5, then the
gap ratio will be 3 : 4. From the 2s hyperfine structures, we can find out the
gI ratio of two isotopes, it is (228.2/803.5)(4/3) = 1/2.641. Then we have
X = 4 × (26.1/3) × 2.641 = 91.899 = 91.9 MHz
6.3 Hyperfine structure of light elements

The hyperfine structure can be estimated as
Zi Zo2 me 2
EHFS ∼ α Ry (6.2)
(n∗ )3 Mp
For ground state hydrogen, the hyperfine structure should be around 95 MHz
and for ground state lithium (Zi = 3, Zo = 1, n∗ = 1.59), 71 MHz. Both
structures are badly underestimated.
6.4 Ratio of hyperfine splittings

For 1s state, J = 1/2, put in notation {element : I; F } we have relevant
quantum numbers for H, D and 3 He+ :
{H : 1/2; 1, 0} {D : 1; 3/2, 1/2} {3 He+ : 1/2; 1, 0}
Given A ∝ gI µN Z 3 , gI ∝ µI /I and interval rule, we have
AH 3/2 × 2.79
EH /ED = = = 4.3
3/2 AD 3/2 × 0.857
AH 2.79
EH /E 3 He+ = = = −0.16
A 3 He+ −2.13 × 8
6.5 Interval for hyperfine structure

(a) The interaction can be expressed in terms of F, I, J,
A
EF = A hI · J i = {F (F + 1) − I(I + 1) − J(J + 1)} (6.3)
2
22

lOMoARcPSD|17631832
Therefore, EF − EF −1 = A2 {F (F + 1) − F (F − 1)} = AF , the interval rule.

(b) Apply the same method as the text demonstrated, F = 4.9, 6.1, 6.9, 8.1,
respectively for c, d, e, f. The nuclear spin is I = Fmax −J = 8−11/2 = 5/2.
(c) From the peaks, it can be inferred that nuclear spin is 5/2 for both
isotopes, then A( 8 S7/2 ; 153) = 20 × (4.86 − 2.35)/(6.42 − 0.77) = 8.9 MHz.
6.6 Interval for hyperfine structure

From the argument in (6.2), I postulate that 2I +1 = 6 and I = 5/2. The
splittings make the ratio 3.1 : 2.6 : 2.0 : 1.5 : 1.0 consistent with prediction
3.0 : 2.5 : 2.0 : 1.5 : 1.0.
6.7 Hyperfine structure

From the data: (a) 70/5 = 42/3 = 14, (b) 70/42 = 5/3, I conclude that
70 and 42 are two levels of 39 K and 5 and 3 the corresponding two levels
of 40 K. Since the intensities depend on degeneracies 2F + 1, then the two
levels are F = 1 and F = 2. Then nuclear spin is just I = 2 − 1/2 = 3/2,
consistent with the number of splittings (I > J). And the magnetic moment
(µI = gI I, I the same) ratio (39 over 40) reads 1.6/(1.9 − 1.0) = 1.8.
6.8 Zeeman effect on HFS at all field strengths

(a) As B goes up, the atom moves from F - to J-scheme, and from
the numbers of splittings, we have 2J + 1 = 4, J = 3/2 and 2F1 + 1 =
5, 2F2 + 1 = 3, F = 2, 1. Hence I = 2 − 3/2 = 1/2.
(b) MJ is the good quantum number.
(c) The weak field Zeeman energy is
EZM, weak = gF µB BMF (6.4)
Same separations is equivalent to same gF which is
hJ · F i F (F + 1) + J(J + 1) − I(I + 1)
gF = gJ = gJ (6.5)
F (F + 1) 2F (F + 1)
J = 3/2, I = 1/2, F = 2, 1, from calculation the gF are different.

(d) The strong field Zeeman energy is
EZM, strong = gJ µB BMJ + AMI MJ (6.6)
If µB B ≫ A, and J is same for both hyperfine states, then Zeeman splittings

are the same.
(e) The cross-over can be defined at µB B = A = 3.4 GHz, or B =
3.4/14 = 0.24 T.
23

lOMoARcPSD|17631832
6.9 Isotope shift

The mass effect shifts the energy by
me δA
∆ν̃Mass ≈ ν̃∞ (6.7)
Mp A′ A′′
while volume effect gives

2
rN δA Z 2
∆ν̃Vol ≈ R∞ (6.8)
a20 A (n∗ )3
and radius of nucleus is
rN ≈ 1.2 × A1/3 fm (6.9)
For rubidium isotopes A′ = 85 and A′′ = 87, then total isotope shifts,
mass plus volume effects, read (a) .186 + .590 = .777 m−1 (b) .202 + .590 =
.792 m−1 .
6.10 Volume shift

From Eq (6.8), the relative uncertainty that rN contributes is
δ[∆ν̃Vol ] = 2δ[rN ], d[∆ν̃Vol ] = 2% ∆ν̃Vol
For hydrogen 1s configuration, the volume shift is about 5 × 10−9 eV ≈

1 MHz. Therefore, d[∆EVol ]/ELamb ≈ 1% × 1/1057.8 ∼ 1 ppm.
6.11 Isotope shift

Assume Z = A/2, n∗ = 2, ν̃∞ = 1/500 nm−1 , then
me
A11/3 = ν̃∞ R∞ /a30 × 32/(1.2 × 10−15 )2 A = 71
Mp
6.12 Specific mass shift

P
Substitute the nucleus momentum for pN = − N i pi , then kinetic en-
ergy becomes
 

 


 

 N N 
1  me X 2 X 
T = 1+ pi + 2 pi · pj
2me  MN 


 i=1 i<j 

 {z
| normal mass effect
} | {z } 

specific mass effect
24

lOMoARcPSD|17631832
6.13 Muonic atom

The radius is easily found to be
n2 ~2
r= = a0 /2277
Zmµ ẽ2
since n = 1, mµ = 207me and Z = 11. The ‘muonic Bohr radius’ is way
much smaller than electronic Bohr radius. The binding energy is
Z 2 ẽ4 mµ
En = − = −112 × 207 Ry/n2
2n2 ~2
If n = 1, E1 = −25047 Ry. Volume effect must be extremely evident since
the size of orbit is comparable to nucleus scale. Actually, it is about 4% of
the transition energy.
7 The interaction of atoms with radiation

7.1 Averaging over spatial orientations of the atom
(a) We need to show that | hxi |2 = | hri |2 /3.
Proof.
Z Z 1 Z 2π
2 2 1 + cos(2φ)
dΩ (cos φ sin θ) = d(cos θ) (1 − cos θ) dφ
−1 0 2
1
y3 2π
= y− ×
3 −1 2
Z
4π 1
= = dΩ.
3 3
(b) Since there is no specific preference of coordinate system, the result
in (a) is valid for all three axes and hence any direction in space.
7.2 Rabi oscillations

(a) The proof is straightforward for equivalence of eqns 7.25 and 7.26.
The equation 7.26 allows for a family of solutions with arbitrary constant
phases, without loss of generality, set

|Ω| Wt
c2 = sin = const × (eiW t/2 − e−iW t/2 )
W 2
and plug it into the equation 7.26
2
Ω
c̈2 + i(ω − ω0 )ċ2 + c2 = 0 (7.1)
2
25

lOMoARcPSD|17631832
The first exponential gives
−(W/2)2 − (ω − ω0 )W/2 + |Ω/2|2 = 0
the second exponential (with minus sign) gives
(W/2)2 + (ω − ω0 )W/2 − |Ω/2|2 = 0
Both cases are consistent, if ω ≈ ω0 , with definition of W ,
W 2 = Ω2 + (ω − ω0 )2 (7.2)
(b) The plot.
-2 0 2 4 6 8 10 12 14
1.2 1.2
F1
1 F2 1
0.8 0.8
0.6 0.6
0.4 0.4
0.2 0.2
0 0
-0.2 -0.2
-2 0 2 4 6 8 10 12 14
Figure 4: F1 for Ω and F2 for 3Ω
7.3 π- and π/2-pulses

(a) Try the ansatz c(t) = c exp(iλt/2), I find λ = Ω, then with initial
conditions c(0) = [1, 0],
c1 (t) = cos(Ωt/2) c2 (2) = sin(Ωt/2)
(b) A general solution takes the form,

Ωt Ωt
|ψi = cos + φ |1i − i sin + φ |2i (7.3)
2 2
Initially,
|ψ(0)i = cos φ |1i − i sin φ |2i
26

lOMoARcPSD|17631832
Now set Ωt = π, then
|ψ(tπ )i = − sin φ |1i − i cos φ |2i = −i{cos φ |2i − i sin φ |1i}
(c) Set Ωt = 2π, |ψ(t2π )i = − |ψ(0)i.

(d) Set φ = 0 and Ωt = π/2,
√
φ(tπ/2 ) = (|1i − i |2i)/ 2
(e) First set φ = 0, Ωt = π/2, then the state becomes

√
1′ = {|1i − i |2i eiϕ }/ 2
with an orthogonal state

√
2′ = {−ie−iϕ |1i + |2i}/ 2.
Here comes a second π/2-pulse, the state evolves into
1 − e−iϕ 1 + eiϕ
|1i − i |2i
2 2
Therefore, the probabilities in |1i and |2i√are sin2 (ϕ/2) and cos2 (ϕ/2).
(f )√Since cos(φ + π/4) = (cos φ − sin φ)/ 2 and sin(φ + π/4) = (cos φ +
sin φ)/ 2, then
1 1 −i
Û π2 = √
2 −i 1
Similarly,
0 −i
Ûπ =
−i 0
and the phase shift in |2i
1 0
Ûϕ =
0 eiϕ
Then the successive operator reads
Û = Û π2 Ûϕ Ûπ Û π2

iϕ/2 cos ϕ2 sin ϕ2
= −e
− sin ϕ2 cos ϕ2
7.4 The steady-state excitation rate with radiative broaden-

ing
(a) The differential equation reads ċ2 = − Γ2 c2 , the solution reads, from
inspection, c2 (t) = c2 (0)e−Γt/2 .
27

lOMoARcPSD|17631832
(b) Since
d Γt/2 Ω∗
c2 e = −ic1 e−i(ω−ω0 +iΓ/2)t
dt 2
integration gives
Ω∗ e−i(ω−ω0 +iΓ/2)t
c2 (eΓt/2 − 1) = −i
2 Γ/2 − i(ω − ω0 )
where c1 = 1 in the weak excitation. In the limit Γt ≫ 1,
Ω2 /4
|c2 |2 =
(ω − ω0 )2 + Γ2 /4
7.5 Saturation of absorption

(a) Transmission in a weak resonant field takes the simple form
I(ω, z) = I(ω, 0)e−κ(ω)z = I(ω, 0)e−N σ(ω)z (7.4)
For transmission 1/e, N σ = 1. The absorption cross-section reads
g2 π 2 c2
σ(ω) = A21 gH (ω) (7.5)
g1 ω02
where g1 , g2 are the degeneracies of two states and gH is the Lorentzian line
shape function
1 Γ
gH (ω) = (7.6)
2π (ω − ω0 )2 + Γ2 /4
For a simple two level system (Γ = A21 ) with g2 /g1 = 3 at resonance,
2πc2 A21 3λ20

σ(ω0 ) = 3 × = (7.7)
ω02 Γ 2π
Sodium s-p transition has λ0 = 589 nm, invoking Eq (7.7), we have

N = 6 × 1012 .
(b) The saturation intensity is defined as
~ωA21
Is (ω) = (7.8)
2σ(ω)
and the fact that Isat = Is (ω0 ), Isat /Is = ω0 σ(ω0 )/ωσ(ω). The absorption
coefficient is defined
N σ(ω)
κ(ω, I) = (7.9)
1 + I/Is (ω)
then we can find κ(ω, Isat ). Notice τ = 1/Γ = 6 ns for sodium at λ = 589 nm.
28

lOMoARcPSD|17631832
7.6 The properties of some transitions in hydrogen

(b) The A21 for 3s, 3p and 3d are 6.3 × 106 s−1 , (17 + 2.2) × 107 s−1 and
6.5 × 107 s−1 , the lifetimes τ = 1/A21 are then 160, 5.5, 15 ns.
(c) Shorter lifetime means larger spontaneous transition rate. Since 1s-
2p overlaps much more than 1s-3p, transition rate is higher for the former
and hence shorter lifetime.
(d) The Einstein coefficients satisfy the following relations
~ω 3
A21 = B21 g1 B12 = g2 B21 (7.10)
π 2 c3
and radial matrix element D12 is contained in
πe2 |D12 |2 g1 4α 3
B12 = ⇒ A21 = ω |D12 |2 (7.11)
3ǫ0 ~2 g2 3c2
Then the D12 ’s for transitions 2p-3s, 1s-3p, 2s-3p, 2p-3d and 1s-2p are
0.53, 0.52, 3.0, 3.8, 1.3, measured in a0 .
(e) The saturation intensity is given by
π hc π hcΓ
Isat = = (7.12)
3 λ3 τ 3 λ3
For the case where spin and fine structures are ignored, Γ = A21 .
Now calculate the saturation intensities for 2p-3s and 1s-3p. The A21 are
(6.3 × 106 , 1.7 × 108 ) s−1 , the wavelengths are 1./R∞ (5/36, 8/9), therefore
Isat = (0.46, 3.7 × 103 ) mW/cm2 .
7.7 The classical model of atomic absorption

The dynamical equation takes the form
F (t)
ẍ + β ẋ + ω02 x = cos ωt (7.13)
m
The steady state ansatz takes the general form x(t) = Aei(ωt+δ) and plug
it into the dynamical equation (I have secretly replaced cos ωt by eiωt ), we
have
F (t) −iδ
−ω 2 + iβω + ω02 = e
mA
(a) Take the real part of the ansatz,
x = A cos(ωt + δ)
= A(cos ωt cos δ − sin ωt sin δ)
: = u cos ωt − v sin ωt
29

lOMoARcPSD|17631832
(b) Notice that

p
u2 + v 2 = A
F0 /m e−iδ
=
ω 2 − ω02 + iβω
F0 /m
=p
(ω − ω02 )2 + (βω)2
2
−1/2
F0 2 β2
≈ (ω − ω0 ) +
2mω 4
The peak is located at ω 2 = ω02 − β 2 /2.

(c) The phase is given by
v
tan δ =
u
βω
=
ω2 − ω02
The first line comes from (a) and the second line comes from the complex
algebraic equation above.
(d) Combine (b) and (c).
(e) Notice that P ∝ |A|2 and simply invoke (b) we have,
1
P ∝
(ω − ω0 )2 + (β/2)2
7.8 Oscillator strength

(a) Integrate the cross-section in Eq (7.5) gives
Z ∞
g2 π 2 c2 g1 4α
σ(ω)dω = |D12 |2 ω 3
−∞ g1 ω02 g2 3c2
π 2 c2 4α
≈ 2 |D12 |2 ω03
ω0 3c2
= 2π 2 r0 cf12
where, as you may check, r0 = α~/me c = 2.8 × 10−15 . Note, I computed

the complicated integral with computer algebra system and made approxi-
mations by hand.
(b) The cross-section can be calculated from
P (ω) e4 ω 4 8π/c
σ(ω) = 2 =
(c/8π)E0 3m3 c (ω − ω02 )2 + β 2 ω 2
2 3 2
30

lOMoARcPSD|17631832
(c) For an harmonic oscillator E = K + U = 2K, if we assume that

oscillation is much faster than the damping, then this formula holds and we
have
E = m ẋ2 = mx20 e−βt ω ′4 cos2 (ω ′ t) = mx20 e−βt ω ′4 /2
Then, according to the classical damping formula (1.23), the damping rate
is
1 Ė e2 ω 2
=− =β=
τ E 6πǫ0 me c3
(d) Combine the result in (b) and (c).
(e) From Eq (7.11), we have
g2 3λ3
|D12 |2 = A21 (7.14)
g1 4αc(2π)3
and for sodium 3s-3p transition, A21 = 2π × 107 s−1 , λ = 589 nm, therefore
|D21 |2 = 5.3 × 10−20 C2 m2 . Absorption oscillator strength is, by definition,
2
f12 = 2me ω0 D12 /(3~) = 0.980 (7.15)
(f ) For hydrogen 1s-2p and 1s-3p the oscillator strengths follow from the
formula
3 mc2 A21
f12 =
2 ~αω 2
and they are 0.418, 0.0801.
7.9 Doppler broadening

In the rest frame and moving frame the frequency differs by δ = ω −ω0 =
kv. And the sinc2 behaves like a delta, then v = (ω0 − ω)/k in f (v).
7.10 An example of the use of Fourier transforms

I recommend first complexify the expression, then apply Fourier trans-
form and finally take the real part. It is straightforward if you look it up in
the table and the final result will be a Lorentzian with peak at ω.
7.11 The balance between absorption and spontaneous emis-

sion
It comes from the Einstein equation
Ṅ2 = (N1 B12 − N2 B21 )ρ(ω12 ) − N2 A21 (7.16)
at equilibrium and if we neglect the contribution from spontaneous emission,

the absorption is
κI = (N1 B12 − N2 B21 )ρ(ω12 )~ω = N2 A21 ~ω
31

lOMoARcPSD|17631832
Note that N2 = N ρ22 and substitute eqn (7.69) for ρ22 , then we have eqn
(7.87) invoking the definition
I 2Ω2
= (7.17)
Isat Γ2
7.12 The d.c. Stark effect

(a) The perturbation is similar to “intermediate field strength” in the
discussion of Zeeman effect. The eigenvalues are
p ǫ V2
λ1,2 = ± V 2 + ǫ2 /4 ≈ ±( + )
2 ǫ
(b) See part (a).
(c) For 3s-3p transition, ǫ = hc/λ = 2.11 eVThe displacement will be
of the order a0 , then the electrostatic energy will be of the order V =
eE0 a20 /2 ∼ 10−18 eV. Then ∆E = V 2 /ǫ ∼ 10−37 eV.
32

Sol Manual Atomic Physics

Uploaded by

Copyright:

Available Formats

Sol Manual Atomic Physics

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Sol Manual Atomic Physics

Uploaded by

Copyright:

Available Formats

lOMoARcPSD|17631832

sol manual atomic physics

atomic physics (Seoul National University of Science and Technology)

Studocu is not sponsored or endorsed by any college or university

Solutions to the Exercies of

(Dated: June 28, 2011)

1 Early atomic physics

Downloaded by Bibi Huda (bibihuda936@gmail.com)

1.1 Isotope shift

1.2 The energy levels of one-electron atoms

which is in accord with the data set, 1.00041.

1.3 Relativistic effects

This corresponds to a grating of ∼ 105 grooves.

Downloaded by Bibi Huda (bibihuda936@gmail.com)

1.6 X-ray experiments

1.7 Fine structure in X-ray transitions

E = (82/2)2 × 13.6 = 22.9 keV

and Eq (1.5) gives

Eα = 10.2(Z − 1)2 eV = 66.9 keV

1.8 Radiative life time

Downloaded by Bibi Huda (bibihuda936@gmail.com)

Table 1: Frequency shifts

1.9 Black-body radiation

provided the wavelength λ = 600 nm, ω = 2πc/λ = 3.13×1015 rad/s. There-

ρ = 4.70 × 10−16 J s/m3

1.10 Zeeman effect

1.11 Relative intensities in the Zeeman effect

• Perpendicular to the magnetic field, motion along z direction and pro-

• 2Iσ+ = 2Iσ− = Iπ = I since the projected horizontal motions are only

(a) Total intensity perpendicular to the field is 2I;

(b) Ratio of intensities received along to perpendicular to the field is

Downloaded by Bibi Huda (bibihuda936@gmail.com)

1.12 Bohr theory and the correspondence principle

1.13 Rydberg atoms

dE e2 /4πǫ0 d (n−2 ) e2 /4πǫ0

Downloaded by Bibi Huda (bibihuda936@gmail.com)

2 The hydrogen atom

2.1 Angular-momentum eigenfunctions

• For l = 1, 2, we only need to show that h 10| 20i = h 11| 21i = 0.

2.2 Angular-momentum eigenfunctions

Downloaded by Bibi Huda (bibihuda936@gmail.com)

2.3 Radial wavefunctions

Invoking the rhs formula, namely

yields the same result 1/(24a30 ).

The electronic charge density of this region is

2.5 Hydrogen: isotope shift, fine structure and Lamb shift

δν̄isotope ∼ (5 × 10−4 ) · (5 × 105 GHz) = 250 GHz

Downloaded by Bibi Huda (bibihuda936@gmail.com)

Relativistic effect is of the order of α2 ≈ 5 × 10−5 , then the wave number

δν̄fs ∼ (5 × 10−5 ) · (5 × 105 GHz) = 25 GHz

Lamb shift is 1/10 of the fine structure shift, that is

δν̄Lamb ∼ 2.5 GHz

A Fabry-Perot étalon of finesse F = 100, width d = 1 cm is supposed to

Ultraviolet 100 nm 0.45 ns

Infrared 1000 nm 450 ns

2.7 Selection rules

2.8 Spin-orbit interaction

Ē = (2j + 1)Ej + (2j ′ + 1)Ej ′ = β[(l + 1)l − l(l + 1)] = 0

2.9 Selection rule for the magnetic quantum number