Faraday Rotation

Light, passing through heavy glass, has its direction of polarization rotated slightly if a magnetic field is applied to the glass.

What it shows:

In 1845, Michael Faraday discovered experimentally that a magnetic field can affect the polarization of light, providing the first evidence that light and magnetism were somehow connected. Specifically, Faraday found that a polarized beam of light, traveling through heavy glass, had its plane of polarization slightly rotated by a magnetic field applied to the glass. With this experiment we demonstrate that phenomenon.

How it works:

A 1/4" diameter X 4" long (6.4 mm X 10 cm) rod of flint glass1 is positioned along the axis of a solenoid. The solenoid provides a uniform magnetic field along the direction of propagation of light traveling the length of the rod. It is adjustable from zero to a maximum of 1,300 gauss. Light from an incandescent light bulb (linearly polarized by a Polaroid™ filter) travels down the length of the glass rod in the bore of the solenoid. A second Polaroid™ filter at the other end of the solenoid is rotated 90° (with respect to the first filter) to block the light. When the solenoid is energized, the light is no longer extinguished. One can secure extinction again by rotating the second polarizer, thus demonstrating that the plane of polarization has been rotated by the longitudinal magnetic field applied to the glass.

The angle of rotation is given by θ = VlB, where V is the Verdet constant of the glass (given in radians/m·T), l is the length of the glass rod, and B is the magnetic field.

Faraday RotationFaraday Rotation

The Verdet constant is wavelength dependent and for SF-59 glass is advertised to be 23 rad/m·T for 650 nm light. Thus, by applying 1000 gauss (0.1T) to the 0.10 m long rod, we expect a rotation of 0.23 radians, or approximately 13 degrees. The wavelength dependence is quite evident in this demonstration in that one cannot extinguish all the light with the second filter; instead, different wavelengths of light are selectively extinguished, depending on the angular setting of the filter. At around 13° the light passing through appears blue, which means that yellow light has been extinguished from the white light. At 17° the light appears magenta (green light has been removed), and at 30° it appears yellow (blue light has been removed). These angles suggest the Verdet constant is 23, 30, and 52 rad/m·T for yellow, green, and blue light, respectively.

Setting it up:

The entire apparatus (including the dedicated solenoid power supply) is mounted on a board and can be set up on eihter a cart or lecture bench. The power supply (0-290 VDC output2) requires 220-VAC input. A video camera shows the view down the bore of the solenoid.

A DC ammeter monitors the solenoid current and has been calibrated against the measured magnetic field. The calibration curve is on the power supply; 1.6 amps is a suggested current to use and will produce 1000 gauss. Up to 1300 gauss is possible with the set-up, but the solenoid will dissipate about 630 W and heat up quickly. Regardless of current setting, it's important that the solenoid be energized only for brief periods of time to avoid overheating, especially when operating at high fields -- rotate the polarizing filter to show the effect and then turn down the current!

If one wants to show that the direction of rotation of polarization depends on the direction of the applied magnetic field, a current reversing switch on the power supply (next to the current/field calibration curve) allows for easy reversal of the magnetic field. Reduce the current to zero before reversing.


Given how small the rotation is with this long piece of specialty glass and strong magnetic field, it's truly amazing that Faraday observed the effect without the benefit of today's materials and technology—a testimony to his experimental genius.

1http://www.us.schott.com/advanced_optics/english/our_products/materials/data_tools/index.html SF-59 glass (a.k.a. "flint" glass)

2 The output of the power supply is floating and neither the + nor — side can be grounded. Thus, if troubleshooting, make sure that the low side of your test instrument is not ground.