The George Washington University

Physics Department
Measurement of the Landé factor using the Electron Spin Resonance − Lab 6∗

+ 3/2
Learning Objectives: Measurement of the Landé fac-
tor using AC and DC methods. Understand how the 2P
+ 1/2
3/2
coil size is essential for the experiment. Understand the − 1/2
purpose of a calibration of an instrument and how it is − 3/2
applied to data (arc calibration). Estimation of the sys-
tematics. + 1/2
2P
1/2
− 1/2

I. PHYSICAL BACKGROUND

Electron Spin is the term applied to the intrinsic 2S

+ 1/2
1/2
angular momentum of the electron. It was discovered − 1/2
in 1925 by the Dutch physicists Goudsmit and Uhlen-
beck when they explained the multiple splitting of spec- FIG. 1: Term Diagram of the Sodium Resonance Doublet.
tral lines in t magnetic field — an effect which bears the
name of their fellow countryman Pieter Zeeman. Until
then, only the normal Zeeman Effect was understood, ac- rectly with some types of test-pieces. It involves the use
cording to which a spectral line in a magnetic field splits of a high-frequency alternating magnetic field to induce
into three lines. According to the classical concept of an transitions in the Zeeman components, obtained from the
electron oscillating within an atom, any one linear os- splitting effect of a constant magnetic field. This way of
cillation can be broken down into one linear component measuring, known as Electron Spin Resonance (ESR),
parallel to the magnetic field and two opposing circular is a means of learning about the intrinsic angular mo-
components vertical to the field. The frequency of the mentum and about the magnetic moment of the electron
circular components rises or falls due to the effect of the associated with it. These are basic properties of the elec-
Lorentz Force in the magnetic field, while the parallel tron in the same way as electrical charge and rest mass.
component is unaffected. For demonstration purposes, the organic compound
The normal Zeeman Effect can only be observed when DPPH [1] is very well suited as a test-piece due to its
the total of the intrinsic angular momenta in the atom is narrow, intense resonance line. The chemical structure of
zero, i.e., when the total angular momentum is made up Diphenyl-picryl-hydrazine (DPPH) is depicted in Fig. 2.
solely of all the orbital momenta. Much more common, It’s a radical in which an unpaired electron appears with
however, is the “anomalous” Zeeman Effect in which a nitrogen atom. This electron, which has no orbital an-
more than three lines are produced. For example, the two gular momentum, is used to create resonance. Sample
yellow sodium lines split into four or six lines (Fig. 1). has two pieces: DPPH (dark purple which looks as
On the assumption that the electron has an intrinsic black) and cotton (white). In the case of paired electron
angular momentum with two possible attitudes in a mag- sample or cotton, no resonance effect can come. Sample
netic field, Goudsmit and Uhlenbeck were able to explain of paired electron will have a white color [1] as cotton
the splitting of the 2 S1/2 and 2 P3/2 states into two energy does.
levels. The 2 P3/2 state splits into four levels because the
orbital angular momentum is added to the intrinsic an-
gular momentum, so that the sodium double splits into A. Electron Spin Resonance Formula
four or six lines. Thus, the anomalous Zeeman Effect
could now also be understood. There is a magnetic moment combined with the elec-
The resonance method to be used here enables the tron spin. This is understandable when one considers the
splitting, due to the Zeeman Effect, to be measured di- electron as a rotating electrical charge and remembers
that an eclectic circuit has a magnetic moment. Because
of the negative charge of the electron, the magnetic mo-
ment opposes the spin. The analogy cannot be taken too
∗ S. Strauch, 2002; Y. Ilieva, 2007; I. Strakovsky, 2008, 2009, 2010 literally, however, since no qualitatively correct result can
 2

relativistic quantum mechanics. It has a value of ge = 2

to within 0.1%, in accordnace with the experiment.
The resonance formula is obtained from the results of
quantum mechanics, according to which ~
p the value, j of
angular momentum is given by |~j| = j(j1 ) · ~, where j
is the quantum number of angular momentum.
Furthermore, the observable components of the the an-
gular momentum are quantized in a preferred direction
using the formula jz = m~, where m is the magnetic
field. For orbital angular momenta with an orbital quan-
tum number of 1, m takes the whole-number values of
m = 0, ±1, ±2, . . . , ±ℓ. The total number of values is
therefore the odd number 2ℓ + 1, which is the same as
the number of energy levels. From the manner in which
a fundamental state S splits into two Zeeman compo-
nents, it can therefore be concluded that the spin quan-
tum numbers s has a value s = 1/2. The magnetic spin
quantum number ms can therefore only take the values
ms = ±1/2, which explains the splitting-up into two lev-
els.
The amount of electron spin is given by
p
|~s| = s(s + 1)~, where s = 1/2 (3)
FIG. 2: The structure of Diphenyl-picryl-hydrazine (DPPH)
[1]. and its z-component by
sz = m~, where ms = ±1/2. (4)
be deduced from it. In particular, the halt-integrity of Since there are only two possible spin attitudes there
the electron spin cannot be obtained. are only two possible attitudes for the electron’s mag-
The relationship between the electron spin ~s and the netic moment, which is coupled to the intrinsic angular
electron’s magnetic moment µ ~ s is written in the form momentum. From
ge µB µB
µs = −
~ ~s, (1) ~µ = −ge ~s and sz = ±~/2 (5)
~ ~
where µb is the Bohr magneton, the fundamental con- it follows that, for the z-component of the magnetic mo-
stant for atomic magnetic moments, and ~ (called “h- ment
h
bar”; ~ = 2π ) is the Dirac constant, which is similarly 1
the fundamental constant for atomic angular momenta. µz = ± ge µB . (6)
2
The constant ge is a characteristic value of the electron,
known as the Landé factor g. From the behavior of spec- The potential energy Em of a magnetic moment µ~ in
tra in magnetic fields, Landé concluded in 1923 that such ~ is given by:
a magnetic field of flux density B
a factor must exist. It shows the relationship between the ~ = µz B.
Em = µ
~ ·B (7)
values for magnetic moment and angular momentum in
their respective atomic units: Thus the energy, E0 , of an electron not in a magnetic
field, is split into two levels, i.e.,
|~
µs |/µs 1
ge = (2) Em = E0 ± ge µB B. (8)
|~s|/~ 2
Considering the corresponding relationship between Resonance absorption occurs when the energy of the
the orbital angular momentum and the magnetic moment photons being beamed in, E = hf , is the same as the dif-
of an orbiting electron, the result of the classical calcu- ference in magnetic energy between the two levels, where
lation gives the experimentally correct value of g = 1 f is the frequency. This gives the resonance formula
and the Bohr magneton µB = e~/(2me ). However, the E = hf = ge µB B. (9)
g-factor of a rotating electron cannot be obtained from
such a simplistic approach; it can only be obtained from
 3

determine the corresponding f (B) which satisfies the

Em resonance formula (9), the distribution B vs. f (f vs.
B) should be linear. The Landé factor g for the electron
can be determined then from the slope of this line.
The test-piece of Diphenyl-picryl–hydrazine (DPPH)
E0 E=hf exhibits a paramagnetic electron spin system, as de-
scribed in Section I. When placed in the coil of a
high-frequency oscillating circuit in a static magnetic
field, it absorbs high-frequency energy when resonance
is reached. This is shown by the change in impedance of
the oscillating circuit.
In order to monitor this effect using an oscilloscope,
B it is necessary to convert the impedance-change into a
Voltage-change and this is done by repeated periodic
FIG. 3: Zeeman-splitting and resonance requirement. traversal of the resonance point. This is achieved by us-
ing a low frequency (60 MHz) alternating magnetic field
instead of a static field.
II. EQUIPMENT The use of a two-channel oscilloscope enables the mag-
netic field to be monitored as well as the resonance pulses
produced whenever the resonant field intensity is reached.
• Pair Leybold 555 06 Helmholtz Coils,
• Leybold 514 56 ESR Adapter,
A. AC Method
• Leybold didacting GMBH,
In this method, the magnetic field of the Helmholtz coils
• Sorensen DC Power Supply, is produced by applying a low-frequency AC current to
• Standard Electrical Adjust-A-Volt, them. The AC current is supplied by a Variac power
supply followed by a transformer, through a 1-Ω resistor.
• Keithley 175A Autorenging Multimeter, The transformer is required in order to reduce to ade-
quately low for the experiment values the magnitude of
• Tektronix TDS 2000B digital storage Oscilloscope the AC Voltage produced by the Variac. A schematic of
on its own cart (manual is available), the setup is shown in Fig. 4.
• Keithley 175A Autorenging Multimeter, • Frequency to be measured: f .
• Elenco GF-8026 Function Generator, f
• Frequency of the ESR Adapter output: 1000 .
• Hewlett-Packard 5314A Universal Counter,
• Helmholtz coils should be connected in series. The
• Mesuring Resistor, 1 Ω - Lab made, 1-Ω resistor is rated at 4 W. Do not exceed maxi-
mum current Imax = 2 A, i.e., maximum Voltage
• Leybold DPPH Test Piece, drop Vmax = 2 V !!!
• Mounting rail, saddle bases, mounting stand. • By mounting the Helmholtz Coils are spaced one
coil radius apart.

III. ELECTRON SPIN RESONANCE • The potential difference, V , across the 1-Ω mea-
EXPERIMENTS suring resistor is proportional to the coil current,
I, and therefore to the magnetic flux density B, of
In order to determine the Landé factor g for the free the almost homogeneous magnetic field B,~ between
electron using Electron Spin Resonance, one uses the the Helmholtz coils.
resonance formula (9) which is satisfied when resonance
occurs. The correlation between the magnetic field B,
which causes the splitting of the electron energy levels, 1. Resistor Calibration
and the frequency f of the photon to be absorbed by
the electron, which causes a transition between the two The 1-Omega-resistor is the crucial element of both AC-
energy levels, is linear. Thus, if we vary B (or f ), and and DC- experiments. Using a simple circuit from Fig. 5,
 4

MHz

Digital Counter
ESR basic
unit

ESR Adapter
Connections of the Helmholtz coils
f
1000
+12V
0
−12V Y
Z A

+− 12V power supply

Z A
−12V 0 +12V
1Ω

VARIAC Oscilloscope
+ transformer 1Ω

Ch 1 Ch2 EXT
DVM

FIG. 4: Schematic diagram of the setup for AC-method measurement of the Electron Spin Resonance. The blue solid lines
show connecting cables. The Voltage-signal across the measuring 1-Ω resistor can be connected either to the EXT, or to Ch2,
input of the oscilloscope. Connection to Ch2 is a better choice since it gives a possibility to observe the Voltage signal on the
scope.

measure it resistance. Do several (5 - 7) measurements along this axis the magnetic field is homogeneous.
vs. different Voltage or different well known resistance of
the high quality resistor. 3. Turn the Variac knob to 0 V and disconnect it from
the power network, if found connected.
a 200 Ω c b 4. Connect the apparatus, as shown in Fig. 4.
R Rx 5. Call the instructor to check if the setup is connected
correctly.

6. Check that all the components which need it are

connected to the power network (scope, power sup-
plies, DVM, frequency counter.
FIG. 5: A simple circuit for R-calibaration. 7. Switch the power on on these components which
require it (do not forget to turn on the ESR basic
unit).

2. Measurements 8. Set the frequency on the ESR basic unit to ∼ 40

MHz. Start to increase slowly the amplitude of the
1. Connect the medium plug-in coil to the ESR basic AC Voltage by turning the Variac’ knob clockwise.
unit, and insert the DPPH sample inside it. Observe on the scope the Voltage signal from the
oscillating circuit and look for the resonance signal.
2. Mount the ESR basic unit in such a way that the Once the resonance signal is observed (see Fig. 6),
DPPH sample is in middle of the coils and is per- stop increasing the Variac Voltage and adjust the
pendicular to their common axis. This is because settings of the oscilloscope properly.
 5

9. Turn the frequency on the ESR basic unit to min- 16. Continue increasing the frequency f in steps of
imum (this is approximately 20 MHz) and keep it 5 MHz till you reach the maximum possible value
constant. (approximately 80 MHz), and for each f repeat the
procedure from the previous item.
10. Decrease slowly the Variac Voltage, i.e., the AC
current through the Helmholtz coils, till resonant 17. Insert the small coil with the DPPH sample in, and
signals are observed on the oscilloscope. repeat the measurements as for the medium coil
starting from the lowest available frequency and
11. Tune finely the amplitude of the AC Voltage, so going to the highest available frequency in steps
that the minima (resonances) of the oscillating- of 5 MHz.
circuit signal are equally placed with respect to
18. Insert the large coil with the DPPH sample in, and
each other (see Fig. 7). When this is done, the Volt- repeat the measurements as for the medium coil
age shown by the DVM is directly proportional to
starting from the lowest available frequency and
the resonant magnetic field intensity B which sat- going to the highest available frequency in steps
isfies the resonant formula (9) for the chosen fre-
of 5 MHz.
quency f .

12. The minima of the oscillating-circuit have to be

spaced equally because a DVM is used to measure B. DC Method
the alternating Voltage V1Ω . The value measured
by the voltmeter
√ is the root mean square Voltage The idea of this method is the same as in the case of the
rms 0
(V1Ω = V1Ω / 2). Thus, the Voltage shown by AC-method measurements, but here DC-current, with a
the DVM is directly proportional to the resonant very small AC-component, is applied to the Helmholtz
magnetic only when coils, and the resonance is observed on the oscilloscope
rms being in XY-mode. A schematic diagram of the setup
V1Ω = Vres . (10) for the DC-method is shown in Fig. 8. The small AC-
component is necessary in order to be able to sweep the
Because V1Ω is alternating, the following applies magnetic field intensity around the resonant value while
0 0 keeping f constant. In XY-mode, the oscilloscope shows
V1Ω (t) = V1Ω sin 2πνt = V1Ω sin φ, (11) two Voltage magnitudes on both, X and Y, axes. Note:
In standard mode, the X axis shows time, and the mag-
where ν is the frequency and φ is the phase angle of nitude of the Voltage is shown on the Y axis. In the
the Voltage. The latter means that V1Ω (t) = Vres DC-method setup, the Voltage output of the oscillat-
at a time t such that φ = 45◦ (which happens four ing circuit (which goes to a minimum when resonance
times during one period, every 90◦ ). In order to happens) is on the Y axis, this is why that signal is con-
0
have V1Ω (φ = 45◦ ) equal Vres , the amplitude V1Ω nected to Chennel 2 of the oscilloscope. The Voltage
must be tuned. As can be seen in Fig. 6, varying the across the 1-Ω measuring resistor, corresponding to the
amplitude of the alternating current applied to the current applied to the Helmholtz coils, is on the X axis
Helmholtz coils varies the phase angle φ (and the of the oscilloscope, and this signal is input to Chennel 1
time t correspondingly) at which the current passes of the oscilloscope.
through the resonant value. This, of course, varies
the length of the time intervals between consecu-
tive minima of VRF displayed on the oscilloscope. 1. Measurements
Because the amplitude of the alternating Voltage is
adjusted so that V1Ω (t) passes through Vres when
1. Helmholtz coils should be connected in series. The
φ = 45◦ (every 90◦ ), the displayed VRF minima are
1-Ω resistor is rated at 4 W. Do not exceed maxi-
equidistant with respect to each other.
mum current Imax = 2 A, i.e., maximum Voltage
13. Record in a table the values of f (MHz) and V (V). drop Vmax =2 V!!!
2. By mounting the Helmholtz Coils are spaced one
14. Record the instrumental uncertainties σf (MHz) coil radius apart.
and σV (V) in the same table.
3. The potential difference, V , across the 1-Ω mea-
15. Increase the frequency by 5 MHz, and repeat the suring resistor is proportional to the coil current,
last 3 steps above. I, and therefore to the magnetic flux density B, of
 6

VRF

0
V 1Ω
Vres
V (a.u)

−Vres
0
−V 1Ω
V 1Ω

0 0
V 1Ω = V 1Ωsin2πνt = V 1Ωsinφ , ν=60 Hz

t (a.u), φ (a.u)

FIG. 6: Diagram showing the Voltage signals on the oscilloscope when resonance takes place. Black is the Voltage signal
from the oscillating circuit, VRF ; red is the alternating magnetic field (the Voltage across the measuring resistor, V1Ω ). Note:
resonance is observed no matter if the resonant minima are equally spaced.

the almost homogeneous magnetic field B,~ between 10. Switch the power on on these components which
the Helmholtz coils. The power to all the electronic require it (do not forget to turn on the ESR basic
modules should be OFF. For the Variac power sup- unit).
ply this means the knob turned to 0. Begin with
the medium coil and the DPPH sample placed in 11. Set the frequency on the ESR basic unit to ∼ 40
the magnetic field of the Helmholtz coils. MHz. Increase the AC amplitude by few small
divisions.
4. Connect the medium plug-in coil to the ESR basic 12. Start to increase slowly the amplitude of the
unit, and insert the DPPH sample inside it. DC Voltage by turning its Voltage-regulating knob
clockwise. Observe on the scope the Voltage signal
5. Mount the ESR basic unit in such a way that the
from the oscillating circuit and look for the reso-
DPPH sample is in middle of the coils and is per-
nance signal.
pendicular to their common axis. This is because
along this axis the magnetic field is homogeneous. 13. Once the resonance signal is observed, adjust the
Variac Voltage to some small amplitude so that the
6. Turn the Variac knob to 0 V and disconnect it from edges of the resonant signal have small, but finite
the power network, if found connected. lengths. Adjust the scope so that the edges of the
resonant signal are completely symmetric with re-
7. Connect the apparatus, as shown in Fig. 4.
spect to the central vertical line of the screen.
8. Call the instructor to check if the setup is connected 14. Turn the frequency on the ESR basic unit to min-
correctly. imum (this is approximately 20 MHz) and keep it
constant.
9. Check that all the components which need it
are connected to the power network (oscilloscope, 15. Decrease slowly the DC Voltage, i.e., the DC cur-
power supplies, DVM, frequency counter. rent through the Helmholtz coils, till the resonant
 7

V RF

V res
V (a.u)

−V res

rms V 1Ω
V 1Ω = V res

t (a.u.), φ (a.u.)

FIG. 7: Diagram showing the Voltage signals on the oscilloscope when resonance takes place and the Voltage value shown
rms
by the voltmeter, V1Ω , corresponds to the intensity of the resonant magnetic field B. Black is the Voltage signal from the
oscillating curcuit, VRF ; red is the alternating magnetic field (the Voltage across the measuring resistor, V1Ω ).

signal is observed on the oscilloscope. going to the highest available frequency in steps
of 5 MHz.
16. Tune finely the amplitude of the DC Voltage, so
that the minimum of the oscillating-circuit signal 22. Insert the large coil with the DPPH sample in, and
is centered at the central vertical line of the oscil- repeat the measurements as for the medium coil
loscope’s screen. When this is done, the Voltage starting from the lowest available frequency and
shown by the DVM is directly proportional to the going to the highest available frequency in steps
resonant magnetic field intensity B which satisfies of 5 MHz.
the resonant formula (9) for the chosen frequency
f.
17. Record in a table the values of f (MHz) and V (V). IV. DATA ANALYSIS

18. Record the instrumental uncertainties σf (MHz) • For each method (AC and DC), for each coil, cal-
and σV (V) in the same table. culate the intensity of the resonant magnetic field
B ± σB from the recorded values of V. The func-
19. Increase the frequency by 5 MHz, and repeat the
tional dependence of B vs. V is linear:
last 3 steps above.
20. Continue increasing the frequency f in steps of B = a + bV, (12)
5 MHz till you reach the maximum possible value
(approximately 80 MHz), and for each f repeat the where
procedure from the previous item.
a ± σa = (−0.2563 ± 0.2088) × 10−4 T, (13)
21. Insert the small coil with the DPPH sample in, and
repeat the measurements as for the medium coil
starting from the lowest available frequency and b ± σb = 0.4109 × 10−2 ± 0.39895 × 10−4 T /V, (14)
 8

MHz

Digital Counter
ESR basic
unit

ESR Adapter Connections of the Helmholtz coils

f
1000
+12V
0
Z A
−12V Y

+− 12V power supply

Z A
−12V 0 +12V AC
+ 1Ω
DC

VARIAC Regulated Oscilloscope

+ transformer DC power 1Ω
supply
Ch 1 Ch2 EXT
DVM

FIG. 8: Schematic diagram of the setup for DC-method measurement of the Electron Spin Resonance. The blue solid lines show
connecting cables. The Voltage-signal across the measuring 1-Ω should be connected to Chennel 1 input of the oscilloscope.
The oscillating-circuit Voltage signal should be connected to Chennel 2. The oscilloscope should be set to operate in XY-mode.
Observe the Voltage signal on the oscilloscope.

and the corresponding the Landé factor g.

cov(a, b) = −7.05628 × 10−10 T 2 /V. (15) • For each calculated g-factor value above gi , calcu-
late its uncertainty σgi .
σB is calculated using error propagation – σa and
σb are propagated, V is treated as a constant. Since
2
a and b are correlated the expression for σB should • Present graphically your results: plot gi ± σgi as a
∂B ∂B
contain a term 2 ∂a ∂b cov(a, b). function i.

• For each method (AC and DC), for each coil, de- • Using weighted mean, calculate a final value g ± σg .
termine the parameters of the linear dependence B
vs. f by fitting lines to the corresponding data [2]. • Compare a canonical (standard) Landé factor g
Why does one fit the dependence B vs. f and not (ge = −2.0023193043622 ± 0.0000000000015) [3]
f vs. B? with your determination.
• From the slope of each linear fit B vs. f calculate

[1] E.N. Hristea et al., Int. J. Mol. Sci. 7, 130 (2006). [3] http://physics.nist.gov/cgibin/cuu/Category?view//
[2] Lecture 9, How to use Physica to make Analysis and =html&All+values.x=80&All+values.y=11
present Results.