Going Green with Weak Values
Post date: May 21, 2013 2:56:20 PM
Paper number two deals with a theme of research I have been involved with now for several years: Weak value amplification. The idea behind this metrological technique is that when a "pre- and post- selected" quantum system weakly interacts with a meter, the meter is shifted by an anomalously large amount. This large shift can be interpreted as an amplification of a parameter that is involved with the weak coupling. This idea has been spectacularly applied to measure a tiny shift of an optical beam depending on the polarization of the light by Onur Hosten and Paul Kwiat, and an even tinier shift using an interferometric based technique to measure the tilt of a mirror by the Howell group and yours truly (not that I was turning the knobs in the lab - more like moral support). How much tinier, you ask? They measured the tilt of a laser beam a hair's width at the moon's distance.This brings us to the present paper. One generic weakness of these weak value amplification techniques is that the "post-selection" mentioned earlier means you intentionally throw away most of the measurement events, or in this case, most of the photons. Now it is an incredible thing that you can throw away most of your photons, and still do better than conventional techniques that use all of the photons! Without getting into an explanation of why that is so, it is a natural question if you can do even better if you found a way to detect and use those photons, in effect, recycling them.
This is what we consider in today's paper, coauthored with Justin Dressel, Kevin Lyons together with Trent Graham and Paul Kwiat. I am a fan of Paul, mainly for his devoted use of the bow tie, that unmistakable badge of the professor, but also because he does really great physics.We teamed up to look into how one could capture all the photons. We devise a scheme that is able to take the unused portion of the light beam coming out of the interferometer and return it back to the interferometer. Eventually, all the light is detected, or post-selected, giving much more power on the detector with an amplified signal. This gives more than order of magnitude greater precision of the measured quantity.
Some interesting results along the way: if you just do recycling and capture all the photons, the amplification effect goes away, since the beam profile you measure at the end of the day is exactly the same as the one you send in! Consequently, you need to do something to "refresh" the profile after every round. We consider a beam-profile flipping technique, as well as using the quantum Zeno effect to project the slightly altered beam shape back into its original shape with an optical filter.
This work also sets the stage for combining this technique with bona fide quantum states of lights, such as squeezed light, or entangled photons.
