Exp 7: Faraday Rotation
1 Aim of the Experiment
To determine Verdet’s constant for light of 2 different wavelenths for the material Schott
N-SF57.
2 Theory
Faraday effect is a magneto-optic in which the plane of palarized light is rotated as it passes
through a medium that is placed in a magnetic field. The angle of rotation is dependent on
the length of sample through which light passes l, strength of the magnetic field B and the
proportionality constant called Verdet’s constant V .
θ = V Bl
The strength of the magnetic field created by an electromagnet with N number of turns per
unit length when current I passes through it is given by,
B = µ0 N I
Verdet’s constant depends upon the medium, its temperature and the frequency of light.
3 Analysis
The values of the rotation of the plane of the polarized light θ for different values for the
magnetic field B for both the cases of 532 nm and 650 nm has been calculated in table 1.
532 nm 650 nm
Current B = µ0 · N · I Angle readings θ Angle readings θ
(A) (T) (radians) (radians) (radians) (radians)
0.00 0.00000 5.0964 5.2185
0.40 0.00840 5.1182 0.0218 5.2360 0.0175
0.80 0.01681 5.1444 0.0480 5.2491 0.0306
1.20 0.02521 5.1705 0.0741 5.2665 0.0480
1.60 0.03362 5.2011 0.1047 5.2753 0.0568
2.00 0.04202 5.2360 0.1396 5.2927 0.0742
2.40 0.05043 5.2360 0.1396 5.3058 0.0873
2.80 0.05883 5.2709 0.1745 5.3189 0.1004
3.20 0.06724 5.2971 0.2007 5.3363 0.1178
Table 1: θ v/s B values
1
Least count of the current source was 0.01 A, which was the uncertainty in the current
measurements. Therefore, uncertainty in B,
δB = µ0 N δI = 0.00021
The uncertainty in the angle readings in θ was 0.25◦ . Therefore the uncertainty in the angle
readings when they are expressed in radians would be,
δ(angle) = 0.0174533 × 0.25 = 0.0044
Since θ is the difference of two such angle readings which are uncorrelated,
√
δθ = 2 × δ(angle) = 0.0062
Now, the θ readings for different B values all depend on the angle reading taken at B = 0. So
the θ readings are correlated. However, I couldn’t find a way to perform linear regression on
the data which takes into account that correlation in the uncertainties for different θ values.
So the uncertainties are assumed to be uncorrelated.
Weighted linear regression was performed on the 2 datasets and is plotted in fig. 1 and 2.
Figure 1: θ v/s B values for light of wavelength 532 nm
Values of the slope obtained in each case,
1. V532 l = 3.00 ± 0.14 rad/T
2. V650 l = 1.68 ± 0.11 rad/T
∵ l = 10 cm, ∴ the values of the Verdet’s constant are,
3.00 ± 0.14
V532 = = 30 ± 1.4
0.1
1.68 ± 0.11
V650 = = 16.8 ± 1.1
0.1
2
Figure 2: θ v/s B values for light of wavelength 650 nm
4 Possible sources of error
1. The minima of the current signal of photodiode persisted over a range of angle readings
and one of the readings was chosen arbitrarily.
2. Dust accumulated on the surfaces of the glass rod sample and the polarizers.
5 Results
For the given glass rod sample, the values of the Verdet’s constant are,
1. For 532 nm light, V = 30 ± 1.4 rad/T m
2. For 650 nm light, V = 16.8 ± 1.1 rad/T m
The values of Verdet’s constant for Schott SF 57 for lights of wavelengths 546.1 nm and 632.8
nm given in literature are 29 and 20 rad/T m.[1] Our calculated values are closeby.
References
[1] M. J. Weber, Handbook of optical materials (Laser & Optical Science & Technology),
1st ed. CRC Press, 2002, isbn: 9780849335129; 0849335124.