Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Optical Antennas

  1. Jul 26, 2011 #1
    Hi everyone,

    Light has a dual nature ; wave and particle. Particle part can be explained using Quantum approaches while the wave nature is described by Maxwell Equations. I am interested in light - matter interactions from wave approach which is also my thesis subject. (also I would like to hear people talk about Quantum approach).

    Since light can be assumed as a PHz wave (visible spectrum) it will consist of corresponding wavelength magnetic and electric field oscillations. When a plasmonic based optical antenna is implemented, it will make analog oscillations at PHz frequencies. Currently (practically) we (mankind) can generate coherent light applying electric field to various materials (YAG etc) and filtering out the (beam shaping, polarizing etc) most part to get a true coherence.

    When an optical antenna is implemented and operating, depending on the frequency and power characteristics it is expected to radiate visible light in it's radiation pattern. My curiosity is that will we (human eye) experience it as a common light source, or will there be any difference ?

  2. jcsd
  3. Jul 26, 2011 #2
    Light is light no matter how it is created. It will still appear the same to our eyes. Visible-frequency lasers are in a sense optical antennas, because an electron oscillates between two states and emits radiation with a corresponding frequency. You can actually get pretty far mathematically modeling a laser as a classic oscillator. Lasers emit (nearly) monochromatic, coherent light. So light from an optical antenna would look like light from a diode laser (such as a pen laser).

    By the way, Maxwell's equations describe the wave nature of light, but Quantum electrodynamics describes both the wave and particle nature of light. But there is less dichotomy than people suppose because QED is essentially just Maxwell's equations quantized.
  4. Jul 26, 2011 #3
    Yes that's correct. You say that there will be no difference to our eye and we will be experiencing the radiation like a solid state laser output. But since when an optical antenna becomes practical, we will have chance to precisely generate PHz frequency E and H fields (true coherence) and today laser technology lacks from this feature. Today's light generation technologies are based on physical interactions like potential applied to specific materials and output is not true coherent (and not well controlled). A solid state pump laser (red) varies between 625-635nm wavelength. What we need it to precisely convert electrical signal to optical radiation which will be achieved by optical antennas.

    QED is derived from Maxwell Equations already, of course it introduces much more stuff and complexity. I am just getting into QED and it's a fascinating topic to study.
  5. Jul 29, 2011 #4
    Unfortunately you can't create a monochromatic light, because that would require infinite amount of energy. It's simply as that - check out the Fourier Transform of Dirac function and you will see why it's true.
  6. Jul 29, 2011 #5


    User Avatar
    Staff Emeritus
    Science Advisor
    2018 Award

    Typically monochrome light refers to light of a very small range of wavelengths, but you are correct. You cannot create an exactly monochrome light source.
  7. Jul 29, 2011 #6
    Or an infinite amount of space and an infinite amount of time that the light wave has been in existence.
    Boundary conditions like to muck up pure waveforms. =)
  8. Jul 30, 2011 #7

    Dirac's "The Quantum Theory of the Emission and Absorption of Radiation" paper clearly states FT of monochromatic light will require infinite amount of energy, but on the other hand it is stated at "Lecture Notes in Relativistic Quantum Mechanics" by
    Lars Bergstrom and Hans Hansson that this might be achievable but I think they are talking about applying some boundary conditions to get very close to monochromatic.

    That's correct but this might be achieved with an optical antenna. We can create a pure sinusoidal signal at a specific frequency >> we can create a true monochromatic (purely single frequency) signal at electrical level >> pump it directly to nano antenna. What will be the practical issue that will hold the system back from radiating a monochromatic wave ?
  9. Jul 30, 2011 #8


    User Avatar
    Staff Emeritus
    Science Advisor
    2018 Award

    The uncertainty principle?
  10. Aug 1, 2011 #9
    Sorry, that's wrong. The perfect antenna will create perfect sine waves once it is turned one, but the problem is that it has not always been turned on. When you turn it on, you go from zero signal to sine wave. Fourier transforming the entire waveform, including the zero signal back to negative infinity will reveal more than one frequency. As stated earlier, a perfectly monochromatic wave requires that the antenna be on from a time of negative infinity to positive infinity, and fill all space, clearly not possible. You will find that lasers create waves with non-zero linewidths for some of the same reasons that antennas create such waves: because they are not on forever. A solid-state laser's "on-time" and thus linewidth depends on the quantum state lifetime. An antenna's "on-time" can be as long as the user wishes, constrained by the lifetime of the institution's financial budget.
  11. Aug 1, 2011 #10
    Thank you for the informative reply. Assume there is time 't' when the first half of the first duty cycle (zero to sine maximum value) is completed. What if we analyze signal after 't' ?
  12. Aug 1, 2011 #11
    That's as much biology as physics:


    In other words, the detector has a biological foundation and what is perceived may not accurately record what is emitted....for example, response times might be too slow to detect certain aspects of the wave.
  13. Aug 2, 2011 #12
    Last year we have modeled human vision by implementing antenna arrays instead of rods and cones. Simulations turned out that eye is an amazing structure consisting of lots of other unknown features like error correction and etc. You can check out the paper http://www.jpier.org/PIER/pier.php?paper=08062004.

    As you stated that the behavior also has biological foundations but as far as we are concerned eye components that receive the radiation can be modeled this way up to a high precision but constructing the whole frame is not that easy. It's biological based for sure but the phenomena governs absorption of a specific bandwidth radiation even at the cellular level. Complete human vision is a very fascinating and hard topic to study.
  14. Aug 3, 2011 #13
    Optical antennas are the present rather than the future. It's just that metal losses are so high at these frequencies that designers choose dielectric antennas instead - something already known as low as in UHF band.

    They're already used for THz waves. One or two years ago, ETH Zürich showed the "smallest possible laser", which had atop the pumped semiconductor a rudimentary LC resonator working as the more usual cavity, and this resonator also radiated, so we have to call it an antenna.

    Other designs put tiny metal stripes at Solar cells to concentrate the light's electric field at their edges; again antennas.

    Kraus suggested them a looooong time ago and made sketches with metal feed lines and diodes; these two last features haven't appeared, but with semiconductor technology producing lines 25nm wide, people made optical resonators of them, sure.
  15. Aug 4, 2011 #14
    No they are not present and not off-the-shelf yet. Implementation requires very advanced equipment also improvement in the theory. By optical I mean 'visible spectrum'. Neither UHF or THz are in optical spectrum.

    I think you might be confusing concepts. How long are those metal stripes exactly? Even if there is an induced electrical current at these metal stripes due to sunlight, it would be just energy conversion. I don't think the induced current in these metal stripes can be used to make frequency analysis of sunlight or etc. It might be a raw energy conversion attempt (which might contribute to the efficiency of the array) but can not be used for anything complicated than it is now. What we refer to antenna is that 'the content' it is radiating or emitting plays a role.

    There are optical IC's from various vendors which can process data coming at the specified spectrum. ADNS series are an example of this. Semiconductor technology is not a key component on development of the optical antennas since conductors are much more responsive then semiconductors to em radiation. Yes Kraus had ideas about it (and I liked him) but his contributions to today's optical antenna research is insignificant.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook