Faraday rotation measurement using a lock-in amplifier

Sidney Malak, Itsuko S. Suzuki, and Masatsugu Sei Suzuki

Department of Physics, State University of New York at Binghamton
(Date: May 14, 2011)

Abstract:
This experiment is designed to measure the Verdet constant v through Faraday effect rotation
of a polarized laser beam as it passes through different mediums, flint Glass and water, parallel to
the magnetic field B. As the B varies, the plane of polarization rotates and the transmitted beam
intensity is observed. The angle through which it rotates is proportional to B and the
proportionality constant is the Verdet constant times the optical path length.
The optical rotation of the polarized light can be understood circular birefringence, the
existence of different indices of refraction for the left-circularly and right-circularly polarized
light components. The linearly polarized light is equivalent to a combination of the right- and left
circularly polarized components. Each component is affected differently by the applied magnetic
field and traverse the system with a different velocity, since the refractive index is different for
the two components. The end result consists of left- and right-circular components that are out of
phase and whose superposition, upon emerging from the Faraday rotation, is linearly polarized
light with its plane of polarization rotated relative to its original orientation.
________________________________________________________________________
Michael Faraday, FRS (22 September 1791 – 25 August 1867) was an English chemist and
physicist (or natural philosopher, in the terminology of the time) who contributed to the fields of
electromagnetism and electrochemistry. Faraday studied the magnetic field around a conductor
carrying a DC electric current. While conducting these studies, Faraday established the basis for
the electromagnetic field concept in physics, subsequently enlarged upon by James Maxwell. He
similarly discovered electromagnetic induction, diamagnetism, and laws of electrolysis. He
established that magnetism could affect rays of light and that there was an underlying
relationship between the two phenomena. His inventions of electromagnetic rotary devices
formed the foundation of electric motor technology, and it was largely due to his efforts that
electricity became viable for use in technology.
http://en.wikipedia.org/wiki/Michael_Faraday
________________________________________________________________________
Émile Verdet (1824–1866) was a French physicist. He worked in magnetism and optics,
editing the works of Augustin-Jean Fresnel. Verdet did much to champion the early
theory of the conservation of energy in France through his editorial supervision of the
Annales de chimie et de physique. The Verdet constant is named for him.

http://en.wikipedia.org/wiki/%C3%89mile_Verdet
________________________________________________________________________
1. Introduction
In 1845, Michael Faraday found the diamagnetism in a flint glass contained with PbO.
When it was suspended between two magnetic poles of the magnet, the PbO glass was
aligned along a direction perpendicular to the magnetic field direction. In 1845, Michael
Faraday also discovered that when a block of glass is subjected to a strong magnetic field,
it becomes optically active. When plane-polarized light is sent through glass in a
direction parallel to the applied magnetic field, the plane of vibration is rotated. Since
Faraday's early discovery, the phenomenon has been observed in many solids, liquids,
and gases. The amount of rotation observed for any given substance is found by
experiment to be proportional to the field strength B and to the distance the light travels
through the medium.
Faraday rotation is a principle that relates a change in the plane of polarization of
light as it passes through a material with an external magnetic field present. The Verdet
constant provides the linear coefficient relating the polarization change to the magnetic
field value and is a constant of the material. The relation is shown as

 m  vBd ,

where  m is the angle of rotation of rotation (in radians), B is the magnetic field of
propagation (in Tesla), and d is the length of the path (in m) where the light and magnetic
field interact. v is the Verdet constant for the material. This empirical proportionality
constant (in units of radians per T per m) varies with wavelength and temperature and is
tabulated for various materials.
Fig.1 Faraday effect. The rotation of the polarization vector (E) in the presence
of magnetic field (B0) parallel to the propagation direction (the z axis).
The angle of rotation of the plane of polarization of a light wave for a
transparent material of length d in a magnetic field B is given by: m =
νBd.

2. Simple theory
We introduce the two circular polarized electric fields propagating in the z direction,

E ( z , t )  Re[ E0 ( xˆ  iyˆ ) exp[i (kz  t )

 E0 [cos(kz  t ) xˆ  sin(kz  t ) yˆ ]

E  ( z , t )  Re[ E0 ( xˆ  iyˆ ) exp[i (kz  t )

 E0 [cos(kz  t ) xˆ  sin(kz  t ) yˆ ]

It is considered to be right-hand (+), clockwise circularly polarized if viewed by the receiver, and
to be left-hand (-), counter-clockwise circularly polarized if viewed by the receiver. We write
the incident electric field as the superposition

E( z  0, t )  2 Re[ E0 xˆ exp[i (k  0  t )
 2 E0 xˆ cos(k  0  t )

which is a linearly polarized electric field with the polarization along the x axis. Now we
assume that the electric fields E± propagate with indices of refraction n± and attenuation
constant ±, respectively. Then the wavenumber k± is given by
  2
k  n  n .
c 

where  is the vacuum wavelength of the light. The incident electric field propagates
through a sample of length L to give the emergent field

B y

v
I -e x

Fig.2 The electron rotates in counter clockwise. Since the electron charge (-e,
e>) is negative, the orbital current flows in clockwise. So the magnetic
moment  is anti-parallel to the z axis. In this case (quantum
mechanically) the frequency is equal to  + L. The electric field rotates in
a clock-wise if viewed from the receiver (E+, or Er)
 zm

B y
I
-e

v
x

Fig.3 The electron rotates in clockwise. Since the electron charge (-e, e>) is
negative, the orbital current flows in counter clockwise. So the magnetic
moment  is parallel to the z axis. In this case (quantum mechanically) the
frequency is equal to  - L. The electric field rotates in a counter clock-
wise if viewed from the receiver (E-, or El)

2
E( z  L, t )  Re[ E0 exp(   L)( xˆ  iyˆ ) exp[i ( n L  t )

2
 Re[ E0 exp(   L)( xˆ  iyˆ ) exp[i ( n L  t )]


In the usual case of equal attenuations for the two polarizations (+ = - = ), this reduces
to
 2
E( z  L, t )  E0 exp( L){Re[( xˆ  iyˆ ) exp[i ( n L  t )]

2
 ( xˆ  iyˆ ) exp[i ( n L  t )]}

2 2
 E0 exp( L){cos[ n L  t ]  cos[ n L  t ]}xˆ
 
2 2
 E0 exp( L){ sin[ n L  t ]} yˆ
n L  t ]  sin[
 
2L (n  n ) 2L (n  n )
 2 E0 exp( L) cos[ ]xˆ cos[  t ]
 2  2
2L (n  n ) 2L (n  n )
 2 E0 exp( L) sin[ ] yˆ cos[  t ]
 2  2
2L (n  n )
 2 E0 exp( L)[cos xˆ  sin yˆ ] cos[  t ]
 2

where

2L ( n  n )   
  .
 2 2

Note that the phase change during traversal for the right- and left-circularly polarized light is

2L 2L
  n ,   n
 

We have the relation,

  
       , or  .
2
 y

E-

E
f-
B q x
f+

E+

Fig.4 Superposition of left- and right-circularly polarized light into linearly polarized
after traversing the sample. E+ (right-circularly polarized light). E- (left-circularly
polarized light), which is viewed from the receiver. E = E+ + E-.From the
geometry, there is a relation that        , or     2 .

What is the value of n+ and n-? The right and left components of the light appear to rotate
with frequencies of  +L and  - L, where L is the Larmor frequency.
 z
B
n+n L
B
n-n L
n

Fig.5 Lorentz model for the normal Zeeman effect. The motion of the electrons
are shown by arrows. The magnetic field is applied along the z direction.
When the electron rotates in counter-clockwise, the direction of the
corresponding current is clock-wise. In this case the system has a
frequency of  + vL. When the electron rotates in clockwise, the direction
of the corresponding current is counter clock-wise. In this case the system
has a frequency of  - vL. Note that >>L.

Then the index of refraction are given by

n  n (   L ) , n  n (   L )

Using the Taylor expansion, we have

 n  n  n(   L )  n(   L )
dn dn
 [n( )   L ]  [n( )   L ]
d d
dn
2 L
d

We note that the Larmor frequency for the electron is given by

eB
 L  2 L  >0 (e>0, -e is the charge of electron)
2m

Then we have

 eB   dn
2
 eB  dn
n  n  2   2 
 4m  d  4m  c d

where we use the relation   c /  , c is the velocity of light, and dm/dl is the rotary
dispersion. Note that

dn 2 dn

d c d

Then we have

2L (n  n ) 2L  eB  2 dn e dn

     BL
 2   4m  c d 2mc d

The rotation angle m is defined by

 m    vBL

since <0 in usual way. Here we have

e dn dn
v (  )  293.34 (  ) [rad/(T m)]
2mc d d

or
 dn
 = 1.0083  [min/G cm)]
d

with

m = 9.10938 x 10-31 kg, e = 1.60218 x 10-19 C, and c = 2.99782 x 108 m.

Note that v is positive when the direction of the rotation for polarization vector is the
same as the direction of current producing the magnetic field B (along the z axis).

((Cauchy's equation))
The index of refraction n depends on the wavelength , and is described by a
Cauchy's equation

b dn b
na  2 2 0
2
d 

Note that the plot of n vs /2 for SF-59 (from the Ref. of glass is shown below.

2.02 n
2.00 SF-59
1.98

1.96

1.94

1.92

106 lnm2
1.90

1 2 3 4 5 6

Fig.6 The wavelength dependence of the index of refraction (From the Ref. H.
Bach and N. Neuroth).

The straight line is well described by

106
n  (1.8833  0.003)  (0.0246  0.0008) 
 (nm)2
We use the least-squares fit of the data to the straight line. The index of refraction for SF-
59 glass decreases with increasing 

((Calculation of the Verdet constant for SF-59))

dn 293.34  2  0.024611  106 1.4.442  107

 m  2934( )  .
d [nm]2 [nm]2

v 80

60 SF-59 glass

40

20

lnm
500 600 700 800 900 1000 1100

Fig.7 Verdet constant v (rad/(T m)) vs the wavelength  (nm) for SF-59 glass.
Experimental data (blue solid circles). The calculation curve (red line)
using the Cauchy's equation.

________________________________________________________________________
3. Schematic diagram for the DC method
Fig.8 Apparatus of Faraday rotation (TeachSpin). FR1-A apparatus.

Fig.9 Schematic diagram of DC Faraday rotation method. He-Ne laser (3 mW,

650 nm). Schott SF-59 glass is used as a sample. The DC magnetic field is
generated by a DC power supply. The DC current flowing in the solenoid
is monitored by a DC voltage (digital volt meter) across a resistance R.

4. Maulus's law
 An electric field component parallel to the polarization direction is passed (transmitted) by a
polarizing sheet. A component perpendicular to it is absorbed.

Fig.10 Maulus's law: I  I 0 cos 2 ( ) .

The electric field along the direction of the polarizing sheet is given by

E y  E cos

Then the intensity I of the polarized light with the polarization vector parallel to the y axis is
given by

I  I 0 cos2  (Malus’ law)

where

2
Erms
I  S avg   I 0 cos 2 
c0

Suppose that the light passes through the first polarizer and enters into the solenoid.
The light goes out from the solenoid and passes through the analyzer (the second
polarizer). In the polarizer, the direction of the electric field Ei of the incident light is the
same as that of the polarization vector Pi.

Ei Pi

Fig.11 The direction of the electric field Ei is the same as that of the polarizing
vector (Pi) for the first polarizer

Ef
Pf

Fig.12 The direction of the electric field Ef is different from that of the polarizing
vector (Pf) for analyzer

Suppose that there is no magnetic field. In this case, the light intensity measured by
the photo-diode detector is given by

1
I  I 0 cos 2 ( )  I 0 [1  cos( 2 )]
2

according to the Maulus's. The output voltage from the photo-diode is given by
 1
VDC  kI  kI 0 [1  cos( 2 )]
2

where k is a constant.
VDC kI0 
Malus law

2.0

1.5

1.0

0.5

f radian 
p p 3p 5p 3p 7p
0 4
p 4 2 4
2p
4 2

Fig.13 Malus's law. VDC/kI0 vs the rotation angle .

5. Verdet constant v
When the DC magnetic field B is applied along the z axis, the rotation angle of the
Faraday effect is given by

  vdB  vdB

where v is a Verdet constant and d is the length of the system along the z axis. Then we
have

1
VDC  I 0 cos 2 (   )  kI 0 [1  cos(2(   )]
2
1
 kI 0 [1  cos(2 ) cos(2 )  sin( 2 ) sin( 2 )]
2
1
 kI 0 [1  cos(2 )  sin( 2 )2 ]
2


We choose   ; sin(2 ) =1 and cos(2 )  0 . Then we get
4

1
VDC  kI 0 (1  2 )  kI 0  kI 0
2
 kI 0  kI 0 (vdB)
6. Experimental procedure
1. We use Schott SF-59 glass as sample. Put the sample inside the solenoid. In the
absence of an external magnetic field, measure VDC as a function of the rotation
angle of polarizer, . Confirm that the Malus's law is valid. Determine the
maximum value of VDC: Vmax = 2 kI0.
2. Apply the magnetic field B along the z axis (axis of the solenoid). The magnetic
field B (TeachSpin) is related to the flowing current I0 through

B  111.0 I 0 (A) (Oe)

Then the magnitude of B is related to the voltage across the resistance R = 10  as

V0
B  111.0 (Oe)
R

Note that 1 T = 104 Oe.

3. At B = 0, find the rotation angle  at which VDC = kI0. In fact, when 2   / 2 (or  =
/4): the angle between the analyzer and polarizer directions is 45º (or 315º), we
have

VDC  kI0  kI0 (vdB) .

4. Measure the DC voltage VDC as a function of B. Show that VDC is propotional to B

as shown in Fig.

VDC kI0
Verdet constant
2.0

1.5

1.0

0.5

0.0 vdB  radian 

0.0 0.2 0.4 0.6 0.8 1.0
Fig.14 VDC/kI0 vs vdB, showing the straight line with a slope 1 and a y-intercept 1.
d is the length of the system and B is the external magnetic field along the
z axis.

7. Schematic diagram of AC method

Fig.15 Schematic diagram of the AC Faraday method using a lock-in amplifier (digital,
dual-phase Stanford SR850). A red laser pointer (3 mW,  = 650 nm). The
small oscillation on top of the DC signal in the oscilloscope (photo-diode
output) is due to the Faraday effect. The audio amplifier is used to amplify
the reference signal from the lock-in amplifier. The frequency of the AC
magnetic field is the same as that of the reference signal. Schott SF-59
glass is used as a sample. R0 = 1 k, 3 k, and 10 k. We use a power audio
amplifier (TeachSpin).

The light intensity measured by the detector is given by

I  I 0 cos2[   (t )]
Suppose that the AC magnetic field is applied along the z axis.

B(t )  B0 cos(t  0 )

Then the rotation angle is given by

 (t )  vdB(t )  vdB0 cos(t  0 )  m cos(t  0 )

where v is a Verdet constant and d is the distance. Then we have

I0
I  I 0 cos 2 [   m cos(t  0 )]  {1  cos[2  2 m cos(t  0 )]}
2
I0
 {1  cos(2 ) cos[2 m cos(t  0 )]  sin(2 ) sin[2 m cos(t  0 )]}
2

Since

2m <<1

I is approximated by

I0
I [1  cos( 2 )  2 m sin( 2 ) cos(t  0 )]
2

The output voltage of the detector is proportional to I, as

kI 0
V  kI  [1  cos( 2 )  2 m sin( 2 ) cos(t  0 )]
2

The output signal consists of the DC voltage and the AC voltage

VAC  kI0m sin(2 ) cos(t  0 )

and

kI 0
VDC  [1  cos( 2 )]
2
Using the lock-in amplifier, we get the AC amplitude (root-mean square value)

VAC rms  kI0m sin(2 )

2

Then the ratio VAC / VDC is obtained as

1
kI 0 m sin( 2 )
VAC rms 2 sin( 2 )
  2 m  2 m tan 
VDC kI 0
[1  cos( 2 )] 1  cos( 2 )
2

When we use  = /4 for the experiment, we have

VAC rms VAC  peak

 2 m , or  2 m
VDC VDC

8. The possible use of second harmonics for AC method

We discuss the possibility of using the second harmonics signal to determine the Verdet
constant. As we show above, the output intensity of analyzer is given by

I0
I  I 0 cos 2 [   m cos(t )]  {1  cos[2  2 m cos(t )]}
2
I I
 0 [1  ( m  1) cos(2 )]  0 [ m cos(2  t )   m cos(2  t )
2

2 2
1 3 1 3
  m cos(2  t )   m cos(2  t )]
2 2
I
 0  m [cos(2  2t )  cos(2  2t )]
2

4
I
 0  m [cos(2  3t )  cos(2  3t )]
3

12

(a) When  = /4, we have

I0 1 3
[1   m ( m  2) sin(t )]   m sin(3t )]
2
I
2 3
I
 0 [1  2 m sin(t )]
2
which means no second harmonics signal in I.

(b) When  = /3

I0
[3  3 m  3 3 m (2   m ) sin(t )  3 m cos(2t )  3 m cos(3t )]
2 2 2 3
I
12
I
 0 [3  3 m  6 3 m sin(t )  3 m cos(2t )]
2 2

12

In this case we have both the first harmonics and the second harmonics signals. The first
harmonics which is in phase with the reference signal, is described by

I0
I1in  k 6 3 m .
12

The second harmonics (out-of phase) is described by

I0
k 3 m
2
I 2 out 
12 ,

where k is a constant. Then the ratio is obtained as

I 2 out 3 3
 m  m .
I1in 6 3 6

This implies the simplest way to determine the value of  m . We do not need any information of
the DC component of the intensity. We do not have to take into account of the factor 2 for the
root mean square value of the lock-in amplifier (DC output).

9. Experimental Procedure (Conventional AC method)

1. We use the low frequency as a reference signal; typically f ≈ 60 Hz. The output signal
from the photodiode (voltage across the resistance R0) is examined by using the
oscilloscope. We find that the signal voltage consists of the AC component and DC
component. The AC component (the small oscillation on top of the DC component) is
due to the Faraday effect. The phase of the AC component signal is the same as that of
the AC magnetic field. The rotation angle of the polarization vector is proportional to the
AC magnetic field and length of path. Using the lock-in amplifier (in-phase, first
harmonics, we use the digital dual-phase lock-in amplifier, Stanford SR850), measure the
DC amplitude of VAC signal (the output of the photodiode detector). We note that the DC
 output voltage of the lock-in amplifier is equal to the root-mean square value of the real
AC amplitude [( VAC )rms].

((Note))
When the lock-in technique is applied, care must be taken to calibrate the signal,
because lock-in amplifiers generally detect only the root-mean-square signal of the
operating frequency. For a sinusoidal modulation, this would introduce a factor of 2
between the lock-in amplifier output and the peak amplitude of the signal, and a different
factor for non-sinusoidal modulation. In the case of extremely nonlinear systems, it may
in fact be advantageous to use a higher harmonic for the reference frequency, because of
frequency-doubling that takes place in a nonlinear medium.

2. Next find the maximum value of ( VAC )rms by rotating the analyzing polarization. In fact,
when 2   / 2 (or  = /4): the angle between the analyzer and polarizer
directions is 45º (or 315º), we have

VAC rms  kI 0 m sin(2 )  kI 0 m . (1)

2 2

3. Using the oscilloscope, measure the DC voltage. When 2   / 2 , we have

kI 0
VDC  . (2)
2

4. Calculate the ratio between the AC and DC voltages,

VAC rms
 2 m  2vdB0 ,
VDC

or

VAC  peak
 2 m  2vdB0
VDC

5. The magnetic field B0 (TeachSpin) is related to the flowing current i through

B  111.0i (A) (Oe)

((Note)) 1 T = 104 Oe.

6. Find the root-mean square amplitude of the AC current flowing the solenoid from
the other oscilloscope (see the schematic diagram). The AC current is related to
the AC voltage across the resistance R (= 10 ) by

VBR  Rim cos(t ) , or V BR rms  Rim

2

Then the amplitude of the AC current is evaluated as

2 V BR rms
im 
R

7. We have the relation (TeachSpin)

B0  111.0im

8. Then we get the ratio

VAC rms 2vdB0


VDC 2
 2vd [111.0im ]
2 V BR rms
 2vd [111.0 ]
R
vd
 2(111.0) V BR rms
R

Note that the values of VAC rms V BR rms are the RMS values. In other words, these values
are digitally monitored using lock-in amplifier and oscilloscope.

________________________________________________________________________
10. Results and Discussion
We measured the Verdet constants for SF-59 glass (diameter 5mm, length 10 cm,
TeachSpin) and liquid.

v = 19.99 ± 0.05 rad/(T m) for SF-59 glass

v = 4.86 ± 0.02 rad/(T m) for liquid(distilled water)

________________________________________________________________
REFERENCES
Sidney Malak unpublished paper (Senior Laboratory, Phys.429, Binghamton University,
2011).
J.F. Reichert, Faraday Rotation Instructor's Guide to TechSpin FR1-A.
User manual of Tektronix, TDS 2000-Series, Digital Storage Oscilloscope
A.C. Melissinos and J. Napolitano, Experiments in Modern Physics, second edition
(Academic Pressm New York, 2003).
F.L. Pedrotti and P. Bandettini, Am. J. Phys. 58, 542 (1990).
A. Jain, J. Kumar, F. Zhou, L. Li, and S. Tripathy, Am. J. Phys. 67, 714 (1999).
V.K. Valev, J. Wouters, and T. Verbiest, Am. J. Phys. 76, 626 (2008).
R.M. Duffy and R.P. Netterfield, Applied Optics 22, 1272 (1983).
V.K. Valev, J. Wouters, and T. Verbiest, Eur. J. Phys. 29, 1099 (2008)
E. Hecht and A. Zajac, Opticsn (Addison_Wesley Publishing Comapny, Reading MA,
1979). p.261-3.
F.A. Jenkins and H.E. White, Fundamentals of Optics, McGraw-Hill Book Company
1957.
H. Bach and N. Neuroth, Springer-Verlag, Berlin, 1998, edited by, The Optical Properties of
Optical Glass

_______________________________________________________________________
APPENDIX
Sample: SF-59 glass

A-1 Verdet constant

The Faraday effect is chromatic (i.e. it depends on wavelength) and therefore the
Verdet constant is quite a strong function of wavelength.
The Verdet constant decreases monotonically with increasing wavelength in nearly all
cases. For glasases of the SF type the Verdet constant increases monotonically with the
PbO content. A large values of Verdet are measured for the heavy flint glass type SF59.
This glass has a extremely large PbO content, namely about 80 % by weight. Because of
its large Verdet constant, this glass type is excellently suited as Faraday rotator materials.
The symbol ν is defined as the Verdet constant. For the SF-59 glass rod sold with the
TeachSpin apparatus, the Verdet constant for 650 nm light is

v = 23 rad/T m.

Table
Verdet constant for distilled water

(nm) v (rad/(Tm)
590 3.81
600 3.66
800 2.04
1000 1.28
1250 0.84

A-2 Verdet constant for SF-59 glass

The Verdet constant decreases monotonically with increasing wavelength  in nearly all
cases. For glasses of the SF type the Verdet constant increases monotonically with the PbO
content. Very large values if Verdet constant have been measured for the heavy flint glass type
SF 59. This glass has an extremely large PbO content namely about 80% by weight. Because
of its large Verdet constant, this type of glass is excellently suited as a Faraday rotator material.
It is to be mentioned that almost all glasses of these types show diamagnetic. On the other hand,
special glasses with a large content of rare-earth ions and large magnetic moments such as Tb3+
have been developed, providing large Verdet constants. Since these glasses are paramagnetic,
the Verdet constant and consequently the Faraday rotation angle depend on temperature.
Taking advantage of the rotation in opposite directions, yielding a Verdet constant close to
zero. This has been almost perfectly achieved for the optical glasses SF L16 and SF L56 at
room temperatures. Thus, in these glasses, the plane of polarization of an electromagnetic
wave is nearly unaffected by magnetic fields.

A-3 Index of refraction and Verdet constant for SF-59 glass

Table

Wavelength (nm) Index of refraction Verdet constant

435.8 2.0156 69.8
480 1.989
546.1 1.9635 37.2
632.8 1.9432 25.9
1060 1.9074 8.1