Physics 17 Spring 2003

Lab 5 - Electron Spin Resonance

Theory

The application of an external magnetic field to an atom will split the atomic energy levels due to
an interaction between the magnetic moment of the atom and the external magnetic field. If m is
the magnetic moment and Bo is the external magnetic field, the interaction potential energy is

U = -m ⋅ B o = -m Bo cos q (1)

where q is the angle between m and Bo . This energy simply adds to, or subtracts from, the
unperturbed energy of the atomic energy level.

For the simplified and hypothetical case of a single electron atom with no electron spin, there is
a simple relation between the orbital angular momentum of the electron, L , and the magnetic
moment, m ,

m = - e L . (2)
2m

If we substitute equation (2) into equation (1) and let the external field Bo define the Z direction,
equation (1) simplifies to

U = - e Bo LZ . (3)
2m

Figure 1 shows the orientation of Bo , L , m and LZ .

Figure 1
For single electron atoms, however, there is quantization of orientation. If an electron is in a
state denoted by the quantum numbers n and l , then LZ = mlù where ml can have the 2 l + 1
values l, l - 1 , ...., -1 . Thus, U can only have the values

eù
U = Bo ml . (4)
2m

Therefore, when a single electron (no spin) atom is placed in an external field Bo the energy level
n, l is split into 2 l + 1 components, each component separated in energy by an amount ∆U = mB
Bo where mB is a fundamental constant called the "Bohr magneton" which has the value e ù/2m
.

Energy level splitting of a single electron (no spin) atom placed in an external magnetic
field, Bo .

Figure 2

In an actual atom, we cannot treat the problem as being simply a case of one electron with no
spin. In general, for multielectron atoms in an atomic state specified by the quantum numbers s , l ,
j and assuming that LS coupling is valid (this assumes that Bo is much less than the internal
magnetic field of the atom; the internal field is typically of the order of 10,000 Gauss), we find that

U = m B Bo g mj (5)

where

j j + 1 + s s + 1 - l l + 1
g = 1 + .
2j j + 1

g is called the Lande g factor. For the case of purely orbital angular momentum such that
s = 0 and j = l , we have g = 1 . For purely spin angular momentum such that l = 0 and
j = s , we have g = 2 . The quantity mj can assume the 2 j + 1 values j , j - 1 , .... , -j . An atomic
energy level s, l , j is thus split into 2 j + 1 components with each component separated by an
energy
 DU = m BBog . (6)

General case. Energy level splitting of a multielectron atom or molecule with electron spin
placed in an external magnetic field Bo .

Figure 3

A special case of energy level splitting occurs when a multielectron atom or molecule has one
optically active (i.e. unpaired) electron outside a closed subshell. In such a case, we have s = 1/2.
If, in addition, the ground state is an l = 0 , j = 1/2 state (a state denoted as 2S1/2 ), then the
unperturbed energy state would split into just two energy levels when placed in an external magnetic
field Bo . Under these circumstances, g = 2 and the energy difference between the levels, as
expressed by equation (6), becomes

DU = 2 m B Bo . (7)

Since l = 0 , the 2S1/2 state can be considered to be a purely spin angular momentum state. In
other words, the energy level splitting is due solely to the interaction between the spin magnetic
moment, ms, of the electron and the external magnetic field. Therefore, equation (7) can be
rewritten as

DU = 2 m s Bo . (8)
 Energy level splitting in an atom or molecule with one optically active electron outside a
closed subshell - pure spin angular momentum state: s = 1/2, l = 0 , j = 1/2 - placed in an
external magnetic field.

Figure 4

Materials with this atomic structure can be used to experimentally measure the spin magnetic
moment of the electron. When the material is placed in an external magnetic field, the energy level
of the unpaired electron will split into two levels with the energy level separation proportional to the
magnitude of the magnetic field. The spin magnetic moments of those electrons would now have
just two possible spatial orientations, either parallel or antiparallel with the external magnetic field.
The electrons whose spin magnetic moments become aligned antiparallel to the direction of the
external field would have an energy of +msBo relative to the unperturbed energy level, the electrons
whose spin magnetic moments become aligned parallel to the external field would have a relative
energy of -msBo and the energy difference between the two levels would be 2 msBo . Consider
what would happen if the material were now placed in a region containing an oscillating
electromagnetic field, such as a region containing radio waves. The photons in the radio waves
would each have an energy equal to hn where n is the frequency of the radio waves. If the
frequency n were such that the energy hn was equal to the energy level separation 2 msBo , the
photons would induce electron transitions from the lower energy level to the upper and vica versa.
The transitions from the lower energy state to the upper state would absorb energy from the
electromagnetic field while those transitions from the upper state to the lower state would return
energy to the field. Since there are more electrons in the lower energy level, more electrons would
absorb energy than emit energy and the net result would be an energy absorption from the field.
(One can use the Boltzmann factor e-E/kT to calculate the relative populations of the two states.)
When this condition occurs, the photons are said to be in resonance with DU . For this case, the
resonance applies to transitions of electrons in a purely spin state (l = 0). Therefore, we have the
term electron spin resonance. From classical arguments, we can show that, to induce the transitions,
the magnetic field of the radio waves must be perpendicular to the external magnetic field.
Thus, to measure the spin magnetic moment of the electron in such an atom or molecule, one
applies an external magnetic field Bo to a sample to split the ground state energy level in two. One
then places the sample in a region containing small amplitude radio frequency (rf) waves oriented
so that the magnetic field of the waves is perpendicular to Bo . One then varies either Bo or n
(it is usually more convenient to vary Bo ) until resonance occurs as indicated by a sharp increase
in the absorption of energy from the radio wave field. Knowing n , the energy level separation ∆U
can be computed from hn = ∆U . Knowing ∆U and Bo , one can then calculate ms from
equation (8). In this lab, you will use electron spin resonance to measure the spin magnetic moment
of the election.

References

The following are the sections in Tipler's Modern Physics which are pertinent to this lab. They
should be read before coming to lab.
1. Chapter 7 sections 7-4 and 7-5
2. Chapter 5 problem 18

Experimental Purpose

The purpose of this lab is to experimentally measure the spin magnetic moment of the electron
using the technique of electron spin resonance.

Procedure

The apparatus you will use is shown in figure 5 on the next page. The external magnetic field,
Bo , is provided by the Helmholtz coils. These coils produce a very uniform magnetic field in the
area between the coils which is oriented perpendicularly to the cross sectional area of the coils. The
magnitude of the field at the center of the coils is given by

B o = 0.47 x 10– 3 I (9)

where I is the current flowing in the coils. If I is in amps, then Bo is in tesla. The Daedalon
ESR power supply has been built in such a way that a voltage of one volt read from the scope
display equals one amp of current flowing in the Helmholtz coils. [Note: Equation (9) is not a
general result. It is peculiar to the Helmholtz coils used in the lab. In general, the field at the center
of a set of Helmholtz coils is given by

8 mo N I
B = 3
a 52
where N is the number of turns of wire in the coils, I is the current in the coils, a is the radius of
the coils (and also the intercoil distance), and mo is a constant called the permittivity of free space.
When the coils you will be using were constructed, N and a were chosen so that this equation
simplified to equation (9). ]

The material containing the optically active electrons is a red crystalline powder called DPPH
(dipheny/picry/ hydrazyl). The sample of DPPH is inserted inside a simple helical coil which is
located at the tip of the aluminum probe. The coil and sample are used as the inductive portion of
an LC "tank circuit" as shown schematically in figure 6a. The tank circuit

(a)

Marginal Oscillator

(b)

Figure 6

has a resonant frequency nR = (1 / (2p LC))1/2. If such a tank circuit is incorporated into an
oscillator circuit as shown schematically in figure 6b, the oscillator will have a tendency to oscillate
at the frequency nR. The current oscillations in the tank circuit will produce a small amplitude (of
radio frequency in our case) electromagnetic field inside the coil wrapped around the sample. If the
feedback resistor R is adjusted so that the oscillator is barely oscillating (the net gain in the circuit
exceeds the net loss by a very small amount), then the amplitude of oscillation will be very sensitive
to any changes in absorption of radio frequency energy in the circuit. In particular, the absorption
of energy due to electron spin resonance in the sample will give a large change in the amplitude of
oscillation. Such a circuit is called a marginal oscillator. The amplitude of the oscillations is
displayed on channel A of the oscilloscope. The capacitor C is variable, which allows us to change
nR . The frequency of oscillation of the oscillator is measured by the frequency counter. The
sample and coil are oriented so that the magnetic field , Brf , of the electromagnetic waves in the rf
field is parallel to the long axis of the sample coil which in turn is parallel to the long axis of the
probe. The rf coil and sample are placed between the Helmholz coils such that Brf is
perpendicular to Bo .

In general then, one chooses the frequency nR by adjusting C . One adjusts R so that the
oscillator is barely oscillating. Bo is varied at a rate of 60 Hz, and whenever Bo is such that hnR
= 2 ms Bo there is a decrease in the amplitude of the radio frequency oscillation, as displayed on
the oscilloscope, which represents electron spin resonance. Bo can be computed from the voltage
displayed on channel B of the scope and nR can be read from the frequency counter. Knowing Bo
and nR , ms can be computed from equation (8) or from a plot of nR vs Bo .

1. Check to see that the electronic equipment is wired as shown in figure 5.

2. Turn on the oscilloscope and the Daedalon ESR power supply. Press the SETUP button on the
scope. A menu will appear at the bottom of the scope display. Press the left most button under
the scope display screen until the number 1 is displayed in the left most part of the menu
display. Press the third button from the left to recall the scope setup stored in memory location
1. This should set all of the controls on the scope to values that will give you a good display of
the data being fed into the scope from the ESR equipment. The display on the oscilloscope will
be either a noisy horizontal line or to a line with two or four absorption peaks in it as shown
below.

Oscillator not oscillating. Oscillator oscillating.

Figure 7

wired.] If the display is the former, then adjust the feedback resistor until the two or four peaks
are displayed. If you adjust the feedback resistor throughout its entire range and cannot obtain
the four peaks, then there is a problem with either the marginal oscillator or the probe. See your
TA. If you can obtain the four peaks, adjust the feedback resistor to get the largest absorption
peaks. The frequency reading displayed on the front of the Daedalon ESR power supply
should be somewhere between 20000 and 40000 KHz.

Actually, you should be seeing just two peaks. Consider what theoretically should be
happening. Assume that the scope trace is at the center of the display screen when the current
in the Helmholtz coils is zero. As the current in the Helmholtz coils increases from zero, the
magnetic field strength increases proportionally and the energy level of the optically active
electron splits into two levels with the amount of separation proportional to the strength of the
field. The trace starts tracking horizontally to the right (or left) reflecting this increase in the
magnetic field strength. There is no vertical deflection of the trace since, initially, the energy
level separation, 2moBo , is less than the energy of the photons in the rf field. Under those
conditions, little or no energy absorption can take place. When the the magnetic field strength
reaches a value such that energy level separation exactly equals the photon energy (hn =
2moBo), the electrons in the DPPH absorb energy from the rf field and the vertical deflection of
the trace becomes large. As the magnetic field continues to grow larger, the energy level
separation becomes greater than the energy of the photons in the rf field and absorption stops.
The signal amplitude returns to its nonabsorption level. The trace to this point then would be a
flat horizontal section followed by an absorption peak followed by another horizontal flat
section. As the magnetic field strength reaches its maximum and decreases toward zero, the
trace should retrace itself. When hn = 2moBo an absorption peak should again occur. It
should, however, occur in the same horizontal position as it did when the magnetic field strength
was increasing. (Why?) The pattern should then repeat itself on the other side of the zero field
position resulting in just two peaks. The reason that you see four peaks is because the
electronic circuitry used to monitor the system introduces a phase shift in the signal coming
from the 0.1 ohm resistor and from the rf amplitude output on the oscillator. The function of
the phase shifter is eliminate this phase shift and thereby make the Bo measurements easier.

Adjust the PHASE NULL control on the Daedalon ESR power supply until the four peaks
merge into just two peaks as shown below.

Figure 8

At this point, you should have a scope display similar to figure 8. Adjust the TUNING C
control on the oscillator counterclockwise until the two peaks just disappear. Adjust the
TUNING C clockwise from that position until the two peaks reappear. Record the frequency
 of oscillations. This is the lowest value of nR which your oscillator can produce. Using a
similar procedure, determine the largest value of nR which your oscillator can produce.

If all of the equipment functioned as described above, the system is fully operational and you
are ready to start taking data.

3. Adjust the frequency of the oscillator to its lowest value as determined in step 2 above.
Measure the the resonant magnetic field, Bo-Res. [Note: As discussed above, there will be two
absorption peaks, one on each side of Bo = 0 , because we are using an alternating external
magnetic field. The distance between the two peaks is thus equal to 2 Bo-Res . To measure
Bo-Res , measure 2 Bo-Res directly from the scope and then divide by two to get Bo-Res .
This is more accurate than trying to measure Bo-Res directly. (Why?)]

Figure 9

Repeat this step for ten more oscillator frequencies spread uniformly over the entire range of the
oscillator frequencies.

4. As noted in the introduction, the electron transitions are induced only by perpendicular
oscillating magnetic fields. Since only the component of the field Brf of the sample coil
perpendicular to the external field is effective in causing transitions, the observed signal height
should be proportional to sin2 f where f is the angle between the axis of the oscillator coil
(the probe) and the direction of the external magnetic field Bo as shown in figure 10.

Choose an oscillator frequency somewhere near its midrange and adjust the feedback resistor to
get the largest peaks possible. Using the provided angle apparatus, make signal height vs f
measurements for a minimum of five different angles.
 Figure 10

5. Observe the effect on the positions and signal heights of the absorption peaks when a small
fixed magnetic field is superimposed on the alternating external magnetic field Bo . This may
be done by bringing up a small permanent magnet. Describe the effect on the absorption peaks
as the permanent magnet is brought up from every possible perpendicular direction x, y, z with
every possible orientation x, y, z . Reverse the polarity of the magnet and repeat this step.

Lab Report

Follow the usual lab notebook format. Your lab report should include the answers to all of the
questions asked in the introduction or procedure, all raw and derived data, and an estimate of the
magnitude and sources of error in any data recorded. When answering any question or when
giving any comparison or explanation, always refer to specific data to support your statements.
For this lab, also do the following.

1. Plot the oscillator frequency versus resonant magnetic field data. From the plot, obtain an
experimental value for the spin magnetic moment of the electron and for the Lande g factor.
Compare these experimental values with their accepted values.

2. Plot the signal height vs f data taken in step 4 of the procedure and verify the sin2 f
dependence.

3. Explain why the addition of the small fixed magnetic field to the external alternating magnetic
field changed the position and signal height of the absorption peaks as observed in step 5 of the
procedure. [Hint: Vectorially combine the various magnetic fields to explain the overall effect.]

4. Given that the probability of a molecule at temperature T having energy E is proportional to