This is from Thursday, June 4th, as I am late on updating for that date.
Today I spent the morning with Antoine, which is a student who I will be working under for this project. We went over some basic things such as how data is collected and how everything is analyzed. This included looking at sample signals on his computer and getting a feel for what data collection might be like. We spent more time with the mathematics, though.
The basis of what we talked about was the Fourier transform and what certain results might occur when performing a FT on certain functions (for instance the cosine function). This included how we are able to combine and separate out the frequencies when we do spectroscopy. From what I understand, a FT on f(t) will give us F(freq), which is an integral along the real axis of f(t)*exp(-2*pi(*nu*t)dt.
We also talked about different convolutions involving Gaussian curves enveloping sine/cosine signals. We may understand a convolution to be the integral of the product of two functions after one is reversed and shifted (wiki definition). This is useful in understanding how to get a spatial profile of the beam and why we may get a certain thing. Basically, most laser beams have a Gaussian distribution of intensity or electric field amplitude as a function of distance for the center of a cross-sectional disk of the beam. Within this distribution are the sine/cosine waveforms which describe the propagation of the actual THz signal. One characteristic of the THz signal is that we are limited to frequency by the closeness of the Gaussian beam width and actual THz beam width. This limits us to a 5fs pulse, and that seems to be the limit as of now (though research on this is needed to confirm). Finally, we typically experience a THz signal which has frequency bandwidth greater that the maximum frequency, which is not very common.
It was next possible to look at sources of THz radiation and another review of how we create/analyze the beam.
From Antoine, I was told that my focus should be on three things:
1. Determining the shape of the beam
2. Trying out a technique involving (I think) a non-THz laser and a Si wafer (which I will mention soon)
3. Using a speaker to measure the beam's spatial dimensions
The technique that we talked about with the Si wafer is that a THz radiation is incident on a thin Si wafer. The semiconductor is changed to a metal (because of the excitation of electrons to the conduction band) and the light passes though. What I need to investigate is what happens if a laser pointer shines a dot on this disc. From what I understand, we expect the THz radiation to come out of the other side of the wafer at only the point where the laser pointer (again, different than THz, I think). This results in a thinner beam with diameter approximately the size of the laser pointer beam.
The last thing that we mentioned together was the use of a speaker to move a set of reflecting mirrors back and forth to let the computer sample the beam point-by-point. We only mentioned this and I did not look it up yet, so that is one thing I need to do.
Finally, I had a chance to go into the lab today and look at a sample of data off a Si wafer. What we saw was a time-domain profile of the wave reference beam and then the reflected (and flipped) wave off the surface of the sample.
Other news for the day: I got my badge and spent the rest of the day reading over material.
Subscribe to:
Post Comments (Atom)
Alex,
ReplyDeleteI see you are quite busy doing lots of reading (and writing). Looks good.
I do admit to having a little confusion about the experiment with the Si wafer. High-resistivity Si (i.e., with no intentional doping) has a bandgap at room temp of about 1.12 eV. This corresponds to a wavelength of about 1100 nm, indicating that photon energies of light colors of shorter wavelength will excite the electron/hole-pair carriers to which you refer (and make the semiconductor more metal-like). Longer wavelengths will essentially see a high transmission through the Si.
The THz beam is definitely different than the laser pointer beam. Using the explanation above, the THz beam (and its THz pulses) will have a photon energy of a few milli-electron volts (meV), which is much less than that energy needed to excite electrons to the conduction band (unless it is a Si material that could somehow have carriers trapped very near the conduction band edge - but this would be a very cutting-edge experiment by itself, and I am pretty sure this is not what LOB is working on). Thus, it sounds as if the THz beam is being used as a probe of the material, with the variable in the experiment being the status of an incident laser beam on the Si sample. The laser pointer, on the other hand, has two key characteristics: it is cw (i.e., not pulsed), and it is presumably visible, meaning that its photons could excite carriers in the Si.
Now, since the laser pointer is cw and low power, it doesn't have much of a capacity to excite a large carrier concentration like that which can be produced by a pulsed laser. It may have some effect on the transmission of an incident THz beam, but probably not much. And, the effect that it would have is to make the semiconductor more conductive (and thus reflective), which should block more of the THz beam and not act as a source.
I can think of a few other things they might have in mind, like creating a small pool of photoexcited carriers in the Si with the laser pointer, and then observing a THz output only when the pulsed bias from an incident THz beam also is present. However, it didn't exactly seem like that was the goal from your description.
At any rate, hope this eventually enhances your understanding of what's going on with that Si experiment. It could be that I am not completely understanding what is going on, too. In that case, I will be interested to find out exactly how the intent of the experiment is.
Best wishes,
- John