Levitating glass bead proves Einstein wrong - physics-math - 20 May 2010 - New Scientist

Levitating glass bead proves Einstein wrong - physics-math - 20 May 2010 - New Scientist

The velocity of a microscopic bead has been measured for the first time, contradicting a century-old claim by Albert Einstein. The same technology used to make the measurement may eventually be used to force such beads to exhibit quantum mechanical behaviour that is normally seen in subatomic objects.

Microscopic particles in liquid or gas undergo Brownian motion – jittery, random movements that are the result of countless collisions with neighbouring molecules.

Albert Einstein studied this motion, and in 1907, he predicted that a microscopic particle's kinetic energy – and thus the square of its velocity – should be proportional to the temperature of its surroundings.

But directly testing this idea, which is called the equipartition theorem, is difficult to do for Brownian particles. That's because the many collisions experienced by the particle cause it to change speed and direction extremely quickly.

Trapped by light
If the position of a particle can be measured rapidly enough, it might be possible to measure its velocity before collisions knock it off course. But Einstein predicted the short time scales between collisions would render the measurements "impossible".

Now a team led by Mark Raizen of the University of Texas at Austin has found a way to do it – at least in air. The density of air is lower than water, so collisions are less frequent and microscopic particles change direction on longer time scales.

To measure the velocity, the team used two laser beams to trap a dust-sized, 3-micrometre-wide glass bead in mid-air. By measuring how much the laser light was deflected by the glass bead as it moved around, the team could make multiple measurements of a particle's position before collisions caused it to veer off course. These position measurements enabled them to obtain a measure of the velocity every 5 microseconds and directly demonstrate that the equipartition theorem holds.