Atomic Terms,Hund’s Rules, Atomic Spectroscopy

6th May 2008

I. Atomic Terms, Hund’s Rules, Atomic Spectroscopy and Spec-

troscopic Selection Rules

Having defined ways to determine atomic terms (which group various quan-
tum microstates of similar energy) we need to specify a protocol to allow us
to determine qualitatively the relative energetics of the terms.

Up till now, apart from the example of the carbon atom ground state, we
have not considered too deeply the idea of spin-orbit coupling. This is
important for nuclei starting with Z=30 and moving to higher charge. For
these atoms, the various terms arising from L-S (Russel-Saunders) coupling
of orbital and spin angular momenta are further split based on the spin
multiplicity, effectively.
Hund’s Rules:

• The lowest energy term is that which has the greatest spin mul-
tiplicity.

• For terms that have the same spin multiplicity, the term with the
highest orbital angular momentum lies lowest in energy.

• spin-orbit coupling (more pronounced for heavier nuclei) splits terms

into levels.

– If the unfilled subshell is exactly or more than half full, the level
with the highest J value has the lowest energy
– If the unfilled subshell is less than half full, the level with the
lowest J value has the lowest energy.

Splitting of Carbon Atom energy levels in Many-Electron Atoms

1
Spectroscopy and General Selection Rules in the Dipole Approx-
imation Molecular and atomic spectroscopy afford information on various
properties of atoms and molecules:

• Bond lengths (rotational spectroscopy)

• Vibrational frequencies for characteristic dynamical modes in molecules

(vibrational spectroscopy)

• Allowed electronic states (atomic spectroscopy)

Spectroscopy attempts to connect the discrete quantum energy (and an-

gular momentum) states to atomic/molecular properties. Since the states
are quantized, one realizes that the differences in energy between states are
discrete. Moreover, the energy differences can be related to frequencies of
EM radiation (as we have seen before):

hν = |E2 − E1 |

We see that the energy spacing between levels is central to spectroscopy. In

fact, the spacing between levels defines the EM radiation required to probe
the transitions. The following table (adopted from Engel and Reid) shows
the major spectroscopic techniques and the radiation involved.

Table 1. Spectroscopies and Associated EM Radiation

2
 SpectralRange λ(nm) ν (1014 Hz) ν̃ (cm− 1) Energy (kJ/mol) Spectroscopy

Radio 109 1x10−6 0.01 10−8 NMR

Microwave 100,000 0.01

Infrared 1000 3.0

Visible (red) 700 4

Visible (blue) 450 6

Ultraviolet < 300 10

• NMR has smallest energy spacing

• Electronic transitions have largest energy spacings
• Rotational and vibrational energy levels have intermediate energy spac-
ings
• Each type of transition (and thus, spectroscopy) dictates a specific set
of selection rules for ”allowed” transitions.
”Allowed” Transitions: Spectroscopic Selection Rules Within the
Dipole Approximation
Qualitative Discussion:
Spectroscopy involves the interaction of electromagnetic radiation with atoms
and molecules. For certain types of interactions, the dynamical molecule can
be considered (and in fact is) an oscillating electric dipole, possessing a static
and dynamic dipole moment. It is this dipole moment that interacts with
the electric field of EM radiation. If the mode of oscillation and the electric
field have the same frequency and phase, there can be transfer of energy to
the quantum mode of the chemical species, and a spectroscopic transition
can occur. It turns out that the dynamical dipole moment dictates whether
the transition occurs.

Selection rules tell us which transitions (among the many available) will be
experimentally observable.

3
The Dipole Approxmiation
It can be shown that a non-zero probability for a transition from a state n
to m is realized if the transition dipole moment, µ mn
x is non-zero, giving
the requirement:

Z
µmn
x = ψm
∗
(x)µx (x)ψn (x)dτ 6= 0

where µx is the dipole moment along the electric field direction, in this
case the x − direction. As discussed in Engel and Reid for the case of a
quantum harmonic oscillator, the integral vanishes for all vibrational transi-
tions except in which the principal quantum number (for H.O.) changes by
±1. More generally, selections rules are different for different spectroscopies.
Within the dipole approximation, however, selection rules are determined
using the transition dipole and the appropriate total energy eigenfunctions
for any spectroscopy.

Atomic Spectroscopy The practicality of the previous discussions of atomic

terms comes from the association of the formalism to atomic spectroscopy.
We have already seen for the hydrogen atom that for absorption and emission
spectroscopy (based on transitions between principle quantum states) the
relation of the energetics of the transitions to quantum states is determined
by a relation of the form:

!
1 1
ν̃ = RH −
n2initial n2f inal

where RH is the Rydberg constant. We also discussed that the values

of ninitial determine specific spectroscopic series of lines in the EM spec-
trum. The series of spectral lines associated with n initial = 1 is the Ly-
man series, and those coupled with the n initial = 2, 3, 4, 5 are the Balmer,
Paschen, Brackett, and Pfund series, respectively (named for the spectro-
scopists wwho discovered them)
Many-Electron Atom Transitions Grotrian Diagrams map the various en-
ergy states of many-electron systems so as to allow a view of the possible
transitions between states. The following is an example for the Lithium and
Helium atoms.

4
5
Spectroscopic Selection Rules: Electronic Transitions in Many-
Electron Atoms In atomic absorption and emission processes, only certain
transitions are allowed. This makes intuitive sense since the transitions have
to maintain states that comply with the commutativity relation between
angular momenta and the system Hamiltonian. Thus, the spectroscopic se-
lection rules for atomic transtions based on the dipole approximation are:

• ∆l = ±1
• ∆L = 0, ±1
• ∆J = 0, ±1

6
 • ∆S = 0 (no transition between singlet and triplet states in Helium)

NOTE: The first selection rule refers to the angular momentum of a single
electron; the other rules pertain to the vector sum of all electrons in the
atom. J refers to the total electron angular momentum.

Magnetic Field Effects on Spin States

Orbital and spin angular momenta of electronic states lead to magnetic mo-
ments that can interact with applied external magnetic fields.
The orbital magnetic moment is related to the orbital angular momentum
by:

e ~
µL = − L
2me
eh̄ q q
|µl | = − l(l + 1) ≡ −βo l(l + 1)
2me
e eh̄
|µlz | = − Lz = − m = −βo m
2me 2me

The electron spin magnetic moment is given by:

e ~
µs = − gS
2me
eh̄ q q
|µs | = − g s(s + 1) = −βo g s(s + 1)
2me
e eh̄
µ sz =− gSz = − gms = −βo gms = ±βo
2me 2me
g ≡ ”electronic g f actor” = 2.0023

Energetics of Magnetic Field Effects

~ :
Potential energy in magnetic field B

˙eBz
~ = −µz Bz =
E = −µB (Lz + gSz )
2me

where the Lz and Sz are the orbital and spin angular momentum z-components.
The total Hamiltonian operator is thus:

7
 eBz  
Ĥ = Ĥo + L̂z + g Ŝz
2me
eBz  
ĤΨnlmms = Ĥo Ψnlmms + L̂z Ψnlmms + g Ŝz Ψnlmms = EΨnlmms
2me
o
E = Enlmm s
+ β o B(m + gms )

Thus, the energy in a magnetic field depends on the orbital and spin mag-
netic quantum numbers! Every electron orbital state is split into spin sub-
levels mx = ±1/2.

Optical transitions between different electronic levels follow selection rules:

δm = ±1 δms = 0

Electron Spin Resonance: ESR Low-frequency radiation stimulates

transitions between electron spin sublevels.

• Selection rule: δms = ±1.

 
1
• δE = gβo B 2 − −1
2 = gβo B = hν

• Radiation frequency in microwave (GHz) region

• ESR: applicable to unpaired electron spins.

• ESR: information on electronic orbital structure (line positions and

splittings depend on coupling between unpaired electron spin and other
(including nuclear) spins.

• Which nuclei does the electron interact with ”most”

Nuclear Spin and Magnetic Field Effects: Nuclear Magnetic Res-

onance Spectroscopy

• Electrons (fermions) have intrinsic spin magnetic moment (spin 1/2)

• Some nuclei possess intrinsic nuclear magnetic moment

8
 • Nuclear magnetic moments give rise to transitions exploited by NMR

• Nucleus with non-zero magnetic moment is sensitive probe of local

electron distribution in a molecule

• Uses: structure determination, electron distribution, chemical kinetics,

organ imaging(MRI)

Energy of Spin Active Nuclei in Magnetic Field

The nuclear magnetic moment is related to the nuclear angular momentum
via:

e I~ I~
µ
~ = gN I~ = gN βN = γ
2mproton h̄ h̄

• γ is the magnetogyric ratio: gN βN

eh̄
• βN is the nuclear magneton, βN = 2mproton (see above for the analogue
for the electron)

• The nuclear g-factor, gN , is specific to a particular nucleus, and is a

dimensionless number.

• Most abundantly occuring nuclie are not spin-active (have no nuclear

magnetic moment) and are thus not suitable for use in NMR. Thus,
one encounters NMR with isotopes of common elements, such as 1 H
and 1 3C.

NOTE: The operators and commutation relations for nuclear spin are the
same as for electron spin.

The energies associated ”up” and ”down” nuclear spins in a magnetic field
aligned in the z-direction are determined as:

˙
E = −~µB~o = −γBo mz h̄

With the z-component of the nuclear spin taking on values of ±1/2, there
are two energies for the ”up” and ”down” spins in a z-oriented magnetic field:

9
 1 1
   
E = − ± gN βN Bo = − ± γBo
2 2

NOTE: There are 2 distinct energy levels; however, the separation is a

continuous function of the magnetic field!

Note on precession of individual nuclei

The discussion so far has considered only magnetic moments parallel to an
applied external magnetic field. If the nuclear magnetic moment is not par-
allel to the externail field, then it experiences a torque (much like a spinning
top as it wobbles in a gravitational field). The torque leads to a precessive
motion of the nuclear magnetic moment that can be described as motion
on a cone about the z-direction (magnetic field direction). The frequency of
this precession is called the Larmor frequency given by:

1
ν = γo
2π

We see that the Larmor frequency scales linearly with the applied mag-
netic field. It has characeristic values for particular nuclei. Finally, we
note that in a given ”macroscopic” sample, there are Avogadros’ number
of nuclear magnetic moments precessing. Individually, the moments have
different transverse components, but collectively, they cancel one another to
leave only the z-component. Thus, in a practical sense, we consider the to-
tal magetization, M~ defined as the sum of the individual magnetic moments.

Since NMR, as any spectroscopy, requires transitions between states, let’s

consider the populations of states at normal conditions to get an idea of
one practical aspect of the NMR technique. First, let’s consider the energy
separation for a magnetic field of 5.5 Tesla:

1
E = ± gN βN Bo
2
E = ±7.76x10−26 J
∆E = 2(7.76x10−26 J) = 1.55x10−25 J

10
The frequency assciated with this transition is in the radiofrequency range
(you may hear NMR spectroscopists talk about RF frequencies associated
with inducing transitions via ”pulse sequences”). The relative popula-
tions of the ”up” and ”down” spin states are:

β −α
nβ −
= exp kB T
= 0.999962
nα

This shows us that at 300K, with a 5.5 T magnet, the rate of upward and
downward transitions will be roughly the same, and so a relatively weak
NMR signal is obtained. In general, the energy absorbed is proportional to
the product of the energy difference and populations of the two spin states,
both of which depend on the magnetic field. Thus, NMR measurements
are invariably performed at high magnetic fields employing superconducting
magnets.

Practical Application: Chemical Shifts

Up to now, we have implicitly considered the effects of a ”bare” nucleus in
a magnetic field. In reality:

• Electron density surrounds nucleus and contributes to an induced mag-

netic dipole at the nucleus
• Other nuclei and electrons neighboring the nucleus of interest also
contribute to modulating the local magnetic field at the nucleus of
interest.
These effects must be explicitly included in the theoretical framework in
order to connect to actual measurement. With respect to the effects of local
electron density associated with the atom of interest, we need to consider
the response of the electron density to the nuclear magnetic moment. This
response is in the form of an induced magnetic field that opposes the
external field at the nucleus. The induced field is proportional to the exter-
~ induced = −σ B~o . The proportionality constant is the shielding
nal field: B
constant. Thus, the total field at the nucleus is the sum of applied external
~ = (1 − σ)B~o . The resonance frequency (the fre-
and induced fields, Btotal
quency of the RF radiation required to induce transitions) is also modulated
by the shielding constant:

γBo (1 − σ)
ν =
2π

11
This modulation of the resonance frequency of the nuclear spin is the basis
of chemical shift in NMR spectroscopy. Keep in mind that the shift is due
to the response of the electron density in the neighborhood of the nuclues;
the shielding is a local effect.

In a molecule, the presence of other atoms and chemical functionalities gives

rise to chemical shifts based on the influences of electronic density surround-
ing other nuclei, and the interaction with magnetic fields generated by other
nuclei. Groups that tend to draw electron density away from nucleus de-
crease the shielding for that nucleus, and reduce the chemical shift. Electron
donating groups have the opposite effect.

Finally, NMR signals from nuclei in different chemical environments can

”split” due to effects of spin-spin interactions. For the non-interacting spin
case (no explicit coupling in the Hamiltonian), the Hamiltonian can be
written as:

Ĥ = −γBo (1 − σ1 )Iˆz1 − γBo (1 − σ2 )Iˆz2

The eigenfunctions of this Hamiltonian are products of the eigenfunctions

of the individual operators Iˆz1 and Iˆz2 :

ψ1 = α(1)α(2)
ψ2 = β(1)α(2)
ψ3 = α(1)β(2)
ψ4 = β(1)β(2)

Four energy eigenvalues are obtained after solving the Schrodinger equation
for this system, and these lead to two unique transition frequencies; thus,
the NMR spectrum simulated for this case would give two peaks. Introduc-
ing a spin-spin coupling term into the Hamiltonian:

hJ12 ˆ ˆ˙
Ĥ = −γBo (1 − σ1 )Iˆz1 − γBo (1 − σ2 )Iˆz2 + I1 I2
h̄2

gives rise to a quartet (4 unique transiton frequencies). Thus, though we

do not show explicitly here, more detailed calculations demonstrate that

12
spin interactions and shielding effects are important and at the heart of
extracting meaningful information from NMR methods.

13