Jon's Journal

Back to Jon's page     Student Projects     Laser Center Home

Anirudh's Journal     Danielle's Journal     Matt's Journal     Yaagnik's Journal

Go to First Entry

You will never be happy if you continue to search for what happiness consists of. You will never live if you are looking for the meaning of life. - Albert Camus

Quotes


November 11, 2004

It has been almost two months since my last update, and I've been so long away from Linux that I tried to "cd" an .html file. At any rate, a whole boatload of stuff has happened since September 12, so I'll go in order, ending with today's semi-momentous events.

The week of September 26: week from Hell. On Saturday the 24th, I met with Dr. Noé to discuss my project. With some guidance and a spiffy little sheet of notebook paper which was jam packed with ideas and notes, I began to settle into starting my project. By Saturday evening I'd finished the introduction. Sunday, half the background. Monday, the background and some procedure. Tuesday, the rest of the procedure. Wednesday, the conclusion, future work, references, and abstract. On Thursday, I stayed till 3 AM in the lab converting my paper to LaTeX and finishing things up. On Friday, I made a road trip to Princeton to deliver my finished Siemens paper.

Since then, I've been editing my paper, but to shift focus slightly away from my extraordinary yet vital translation methods, I've begun to really develop the heart of my project: finding the coherence properties of different light sources and how they relate to optical coherence tomography. Recently I've resolved fringes with the sodium gas lamp and measured the visibility in relation to arbitrary micrometer reading.

That brings me to today. I first tried to find fringes with the LED, and when that didn't work, Dr. Noé suggested using a white light source with a narrow (~1-10 nm) bandwidth filter. When we finally found one, I made coarse mechanical measurements with an LED package that was lying around, and then began to scan slowly with the micrometer. Immediately, I resolved fringes at a mirror distance of .390 inches. Observing the peak visibility at .390 inches (rather than .389 or .391), I removed the filter and, much to my surprise, saw white light fringes!

To be continued...

Back to Top


September 12, 2004

Only two weeks to Siemens.

Since last week, I've fixed quite a number of problems. First, it wasn't the software that was broken, nor the hardware. It was both together. The photodetector wassyncing with the multimeter, but the multimeter's connection with the computer must have been flawed or something. I changed multimeters, used updated, Windows software on the Shuttle PC (which is now sitting very beautifully on my work bench next to the beautiful setup), and slowed down the motor to take more data points (this software is able to store 432,000 data points!!). Eventually I got some beautiful data (seeing a trend here?), interferograms which consisted of no less than 15 fringes each.

On top of that, I've completed my beautiful drawing of the setup:

Back to Top


September 12, 2004

Problems abound as we enter the final stretch toward the Siemens deadline. First, and most importantly, the detector is going haywire on me. Starting from Friday (the 3rd of September), every time I tried to set up the motor and record voltage readings, the Radio Shack program displayed "No Connection." Thinking it was the computer, I used DOS to transfer all the Radio Shack software from the old IBM to our new Shuttle SFF PC. The program worked. Or so I thought. When I came in today the first thing I did was hook up the entire photodetector setup to the Shuttle, and see if it would record. It worked just fine when I tested the photodetector out. So, I began data collection and went to go eat a yummy bowl of chips and dip Azure left me. Coming back from microwaving said dip, and eating said chips, I found that the computer was displaying the "No Connection" message again. I removed the photodetector from the setup to see if that was the problem. In regular room light, the photodetector recorded for about two minutes perfectly. Returning it to its mount, I tried again with the motor. No good. I tried tightening the COM connections, which seemed to help, but the error message would pop up at random times. Sometimes, it would be after a minute, and at other times, after as little as 10 seconds. I tightened the COM connections as much as I could have, and there has been no connection left unchecked or unaccounted for. I even replaced the battery. I replaced the computer cables with another one I found in the back room. The software seems to run fine for about two minutes, and then quit. Any more attempts to run the software result in immediate "No Connection" messages. This leads me to believe it is the software, specifically the graphing mode, that is at fault, and not any of the hardware.

However, there's bad news. And worse news. The bad news is, I tried the low-pressure sodium lamp today in the setup (it's bright!) and I read Doug Broege and Brendan Wyker's webpages, and it seems that it's quite difficult to get sodium fringes, and even harder to resolve them onto a viewing screen and record their intensities using a photodetector. The worse news is that Fourier is still not on my side. Devoting hours and hours to reading about this mysterious character and his mysterious transform, and then another lifetime on figuring out why Mathematica's built-in Fourier algorith doesn't work.

Back to Top


September 2, 2004

Quickie Update:
Yaagnik and I tore out the IBM's floppy disk drive and replaced it with one of the ones just lying around the lab. For lunch, we had a talk by Melissa, who did a project on Bose Einsteinian Condensates (I really, really need to e-mail her, because one of her calculations was doing a Fourier transform on her data using Mathematica, something I desperately need to do). Here is my first set of "real" data:

Again, the peaks get wider because the rubber band is approaching an area where it is becoming more and more slack, and therefore it takes more turns of the micrometer to move it the same amount.

Back to Top


September 1, 2004

Welcome to September.
This morning, Ani and I discussed how to measure the path length difference on the interferometer. After thinking about using the spring constant of the rubber band to find the proportionality, which was totally off target, I remembered using the proportionality constants from a previous experiment which attempted to measure minute distances with a Michelson interferometer. Using this concept I first tried to find the proportionality constant between the amount the micrometer turns and the amount the mirror actually moves. In order to do this I used two equations:

Δl = nλ/2 (the change in the length of one mirror relative to the other is equal to ½ the wavelength times the number of fringes scanned; after all, a path length change of ½ wavelength results in movement of a fringe from a maximum to a minimum and back to a maximum - one cycle)

Δl = kΔm (the change in the length of one mirror relative to the other is also equal to the proportionality constant between the mirror and the micrometer times the distance the micrometer moves)

Both these equations can be set equal to one another (kΔm = nλ/2), solved (k = nλ/2Δm), and verified using a simple experimental setup:
Removing the detector and letting the interference pattern fall onto the viewing screen allows for counting of fringes. That is, observation of the number of cycles the interference pattern goes through with a specific amount of travel. By turning the micrometer until 1, 2, 3, 4, and 5 fringes passed and taking the amount of travel necessary for each, I was able to calculate five values for the proportionality constant (which of course should have stayed constant). Instead, the proportionality constants changed, from 6.90x10-5 all the way up to 2.35x10-4. The other numbers were 7.306x10-5, 1.024x10-4, and 1.672x10-5. My theory was that this was due to the loosening of the rubber band. My setup consisted of a rubber band connecting the translating mirror to another micrometer stage, and by turning the micrometer screw, the stage would come back, stretch the rubber band, and in turn move the mirror back a small amount. It was setup so that the rubber band would be at its tightest at the beginning, and by loosening the micrometer and moving it forward, the mirror would move slightly forward.

A number of problems arose from this specific setup. The first of these, I hypothesized, was the loosening of the rubber band. As the rubber band loosened, it took more and more travel of the micrometer to make the mirror move forward. If this is hard to understand, imagine if the rubber band was so loose that it wa totally slack. It would take an infinite amount of turns of the micrometer to move the mirror in that case. So, I saw a trend of longer and longer travel distances with the rubber band as I scanned through more and more fringes.

I set out to solve this by moving both the position of the rubber band and its tension. Before, I had set the rubber band up on the base of the mirror. This lack of leverage forced longer turns of the micrometer to achieve visible results. Now, I moved the rubber band to the top of the mounting base, so that less tension would move the mirror back more, and I could scan more fringes. I also moved the initial position of the mirror further back, so that loss of tension would be less of a problem. This time, I got consistent numbers (1.091x10-4, 1.091x10-4, 1.090x10-4, 1.126x10-4, 1.098x10-4, 1.093x10-4).

This was good for preliminary data. However, I moved on after this short side trip on to the real meat and potatoes - full interferogram scanning. With the help of Dr. Noé I set up the IBM computer with the interfaced Radioshack on my work bench, and prepared for automatic data taking. A number of things happened though. First, I set the pulse rate for the stepper motor to one pulse per second, which proved way too slow. At that rate, the stepper motor made one revolution every 200 seconds. Each revolution being only .025 inches, a whole scan would have taken 8000 seconds, or upwards of two hours. The graph at first seemed like a completely flat, straight line, which it probably was considering it barely moved even after 10 minutes. Dr. Noé suggested upping the speed to ten steps per second, which seemed perfect. Scans could now be achieved in a mere 12 minutes. Immediately, very clear curves started appearing. After recording the data, graphing it, setting it to a curve of best fit, and in general just making it beautiful, I noticed something Dr. Noé had pointed out to me before: the maximum interference signal was about 26 mV. The maximum signal from each of the two interferometer branches, however, was about 6.5. How, then, is 6.5 + 6.5 equal to 26? Dr. Noé provided this example:

The magnitudes of two intensities, Ia and Ib, are Ia½ and Ib½. The sum of the two intensities is the sum of the magnitudes squared, or (Ia½ + Ia½)2. The expansion of this statement is: Ia + Ib + 2IaIb
Since Ia and Ib are equal, this is actually equal to 4I. Since "I" was 6.5 in our situation...Ready to become a physicist?...4I is 26! What?! Ah, magic.

Back to Top


August 31, 2004

Dr. Noé informed me this morning that the decreasing visibility effect I looked at yesterday was probably a fluke. The change in visibility is probably not happening over the distances that I'm observing it. After all, each mark on the micrometer stage denotes approximately 1/50000th of an inch, or 508 nanometers. Even turning the micrometer one whole inch results in a translation stage movement of one fiftieth of an inch, or 508 microns. This movement is probably not sufficient to see such obvious changes in visibility.

One concern is the ambient light around the setup, particularly the reflections of the laser light off the optical components. It may be beneficial to test the amount of ambient light without the laser, with the laser, and then somehow test the amount of light generated by the laser's reflections. Motor troubles

Stepper motor stats: 1.8° per step ; minimum speed: as slow as necessary ; maximum speed: 100 ns per step.
Benefits: step it when ya want it
Drawbacks: cannot decrease size of steps ; vibrates

Continuous motor stats: 40 RPH (revolutions per hour) ; 4 degrees per second ; minimum speed: 40 RPH ; maximum speed: as fast as you can take data (for instance, if data could be taken infinitely fast, there could be an infinite number of data points, and an infinitely small degree value per data step)
Benefits: continuous movement ; potential for large data values
Drawbacks: doesn't slow down for nobody ; potential for thousands of data points limited by data taking equipment

Back to Top


August 30, 2004

The time is winding down...today I have very specific goals:

  1. Put a resistor in the photodetector
  2. Do a fringe visibility experiment (visibility vs path length difference)
  3. Talk to Dr. Noé about possibly purchasing the $8.50 40 RPH motor from Surplus Shed for FTS
From el experimente numero uno:
Imax signal: 5.1 mV ; 1.31x10-6A > 1.31x10-6W
Imin signal: 0.3 mV ; 7.69x10-8A > 7.69x10-8W
V = IR ; resistance = 3900Ω ; the photodetector converts one milli-watt to one milli-amp ; thus intensity (in wattage) is the same as the amperage. From this we can then get Vf, fringe visibility:
V = (1.31x10-6W - 7.69x10-8) / (1.31x10-6 + 7.69x10-8) = .889

"Perfect" fringe visibility is 1.00, supposedly occuring when path length is zero. Unfortunately, it is difficult finding this point, especially when I am unable to see the interference pattern when it enters the photodetector, and therefore am unable to center it to create the "flickering" effect of total light and total dark that occurs at zero path length. A fringe visibility of 1 requires a totally dark fringe minima, i.e. absolutely no signal at some point. Despite the lack of ambient light (brought about by darkening the optical bench and surrounding lab space), it is difficult simply to align the apparatus and set it to zero path length.

I did this experiment again, this time moving the translation stage forward by a whole inch (which should take the movable mirror back ~1/50th of an inch). This time...
Imax signal: 5.8 mV ; 1.49x10-6A > 1.49x10-6W
Imin signal: 0.2 mV ; 5.13x10-8A > 5.13x10-8W
V = (1.49x10-6W - 5.13x10-8) / (1.49x10-6 + 5.13x10-8) = .933
This result is much more pleasing, as it is much closer to the ideal difference of 1. This leads me to believe that by moving the mirror back, I've come closer to the point of zero path difference, rather away from it. Next...points in between.

I realized after taking a number of measurements (at .8 inches, .6 inches, and .4 inches away from the initial) that, well, the results are disappointing. There should be a trend, either moving toward better visibility or poorer visibility. The problem, I ascertain, is the fact that the microscope objective expands the beam so much that even when the "dark" of constructive interference does not take up the entire interference pattern, when projected into a larger shape, the darkness does take up the entire effective measuring field of the photodetector. Thus, even when the visibility shouldn't be a value of 1, it is, because 0 signal is still possible. With this, I will try to move the photodetector close enough to the beam that the beam perfectly coincides with the opening of the photodetector. Actually, the fact is the beam is larger than the opening of the photodetector right now. I'll have to think this over a little more and determine what is causing the problem of "false" perfect visibility.

EDIT: The conversion ratio of the photodetector is actually 0.44 Amps/Watt, not 1.00 like I assumed (for whatever reason). Regardless, the visibility readings taken still remain the same; the amount of light calculated from the conversion changes.

Back to Top


August 27, 2004

From the Melles Griot glossary:

---

Coherence Length

If a laser beam is split into two parts in a Michelson interferometer and then recombined, interference fringes result. A measure of the quality of the interference fringes is the "visibility," defined as

V = (Imax -   Imin)/(Imax +   Imin),

where Imax is the maximum fringe intensity and Imin is the minimum fringe intensity. If the paths traversed by the two beams are equal in length, the fringe visibility is approximately 1. If one path changes while the other remains stationary, the fringes will degrade and the visibility will decrease. For practical purposes, the coherence length is usually defined as the path length difference that corresponds to a fringe visibility of 50%.

A mathematical definition, which corresponds closely to the above and is often used in laser specifications, is

L = c / Dn

where L is the coherence length, c is the speed of light, and Dn is the linewidth of the laser light.

For a common helium neon laser with a linewidth of approximately 1.5 GHz, the coherence length is approximately 20 cm. For a stabilized, single-frequency laser with a linewidth of 1 MHz, the coherence length is approximately 300 m.

Coherence length is a critical parameter in long-path-length holography, because is essentially determines the allowable size and depth of the subject.

---

Using this concept, I can plot fairly accurately the relationship between fringe visibility and path length difference.

Back to Top


August 26, 2004

A short description of my current setup:
The HeNe is setup with a pinhole to create a diverging airy pattern. The light enters the beamsplitter, splits into two paths, and re-emerges on the screen (or photodetector) as an airy pattern with a small interference pattern in the center where the main airy disc should normally be. Using a microscope objective, which also doubles as an iris that filters out the unwanted airy pattern, we can enlarge the central interference pattern to a visible size. Because the pinhole makes the laser beam divergent, it is relatively easy to find where the path length difference between the two mirrors is zero, or close to zero - the rings of the interference pattern become so thick that the center portion fills the microscope objective. Rather than "scanning" in and out, the interference, now visible only as a central dark or light image, simply "flash" on and off.

In order to decrease the sensitivity of the fringe scanning process, a lever connecting two spring-loaded micrometer stages was used. A pivot was installed nearer to the stage that the mirror was mounted on so that with a relatively large movement of the long side of the lever, the short side would travel proportionately less. As of now, a 50x proportionality has been attained, though further increases in effectiveness would require a prohibitively long lever, or a pivot unattainably close to the lever's mount on the translating mirror stage.

Game plan:

  • Re-read and re-clarify the fringe visibility vs. path length change business
  • Use a rubber band to scan the fringes
  • Figure out what to do with the sodium lamp if I decide to use it

Back to Top


August 23, 2004

General update:

  • Dr. Noé tried the demo with the short laser where the interference generated by the beating of the two modes washes in and out; unfortunately, attempts to recreate the experiment were unsuccessful
  • Attempts were made to build a lever that could translate the cube corner retroreflector
  • Another week looking for a motor to translate the lever was unsuccessful
  • Dr. Noé sent us (Ani, Yaagnik, Matt, and me) an e-mail full of links; one of them demonstrated a lever setup
  • The retroreflectors were tossed out and replaced with a spring-loaded stage / lever setup
  • Liquid nitrogen was brought into great use
  • Working interferogram scanning was achieved
I have yet to secure a motor that will step the stage in a controlled manner. This is essential to getting an interferogram. Something I learned when looking at the lever setup was that as the arms approach zero path length difference, the rings scan much faster than they do when the arms have a greater path length difference. My question is, wouldn't this throw off any distance measurements someone was trying to make with an interferometer? An amount of movement when path length difference is relatively little compared to the same amount of movement when path length difference is much greater would yield many more changes in the number of fringes.

Back to Top


August 12, 2004

Tomorrow is the last day of Simons. Today, I built the retroreflector setup in under 10 minutes and aligned it in a similarly short amount of time. At some point I should write a guide to alignment, because this has almost become second nature for me. Soon, I hope to use liquid nitrogen to find the paths of the laser and align it perfectly. For now, however, there are still lots of artifacts in the image, perhaps because of the front surface reflections. The retroreflectors I'm using are not hollow. They are constructed less like mirrors glued together in a corner shape and more like mirrors glued around a prism. For this reason, the front surface reflections off the retroreflectors make their very own interference pattern, which may or may not be a reason for the "ghost" images that I see superimposed over the "real" interference pattern. One thing that would make the image much nicer to look at (and clearer and larger) is a large beamsplitter. A new 1x1 inch beam splitter would help alignment greatly, as more of the interference magnified by the diverging lenses would be shown. However, for the ultimate measurement of light, the larger beam splitter would only help in that it would have a clearer image than the one that I'm using now.

Today I also discovered what could be the answer to the entire problem of the motor and the photodetector: what seems to be a combination stepper motor photomultiplier tube sitting in a box around the lab. I'm almost 100% sure these two connected instruments were meant to be computer interfaced and used in conjunction with one another.

EDIT: the strange apparatus is just a PMT, not a PMT-stepper. From the Times Science section: "...Yoshimasa Hayashi, a member of the Japanese Parliament from the ruling party, crystallized the Japanese position...'In Japan we have pet dogs,' he said. 'But we don't tell the Koreans to stop eating dogs. Nor should people tell us to stop eating whales.'"

Back to Top


August 11, 2004

An uneventful morning was capped off by a chapter in Introductory Fourier Transform Spectroscopy and building and taking apart stuff in the setup. Thankfully, the morning was followed by an amazing hospital lunch (rotisserie chicken, creamed spinach, mac and cheese, and chicken gumbo...mmm...).

The afternoon was kicked off by Ani and Yaagnik's continual debate over how vortices work, joined occasionally by Danielle and me. Seeing that we needed to settle things, we paid a visit to Professor Metcalf. He told me about leaf springs, something Dr. Noé had previously told me about, as a way to translate one of my mirrors. While it makes a lot of sense, is easy to build, is hands on, and can be a self-standing experiment (testing different types of sheet springs/hacksaw blades and seeing how well they translate a mirror), a double-block leaf spring translator can never be perfect. It requires perfect screwing, alignment, and equally sized blocks. On top of that, how much it translates can never be measured.

Professor Metcalf did mention using two light sources (one HeNe or another known wavelength, and the other a "test" wavelength), and then comparing how many fringes of each one changed as the interferometer scanned (using polarizers...I'm not 100% sure how that would work). Knowing the wavelength of the HeNe, the change in distance would be measurable, and then the changing signal of the test source could be compared to path length difference. Simple setup, complex data analysis. Then we discussed how a retroreflector works, and how it might be used for my setup. However, a corner cube can't be used specifically in an interferometer like mine because although the outgoing beam is parallel to the incoming one, they are not incident, so interference would be impossible.

A leaf spring doesn't allow for measuring path length difference. A corner cube retroreflector doesn't allow for incident beam paths. What a conundrum.

After playing with the corner cube retroreflector for a while, and realizing and being told th beauty of it (no matter what way you turn it, outgoing light is always parallel to incoming light, until it's turned so much that it can't enter the retroreflector), I settled in and wrote down everything that had just happened. A little reading and a little browsing later...voilá! a solution! This is the answer to my prayers, however brief they might have been in the hour or so I looked for a viable FTS solution. All I need to carry it out is a way to mount the corner cubes and of course some kind of arm that can be turned with a knob. The only problem that might occur is that if one of the arms turns so much that the retroreflector doesn't meet the beam...well there wouldn't be interference. So, if I were to use two arms like in a parallelogram, or the things in common desk lamps, I could still get the translational stability of a leaf-spring setup and the versatiliy of a corner cube retroreflector. The perfect solution.

But, as with every perfect solution, there's a problem. Not all of my sources are going to be perfect beams, or at least I hope not (i.e. I hope that I'll have a chance to test low coherence souces). Yes, while some sources can be low coherence and directional, like superluminescent diodes, some may not be, like LEDs. I don't quite know how LED light will be directed when reflected off a corner cube. First things first. Mounts. A stable test base. Building this complex setup. Good stuff.

Back to Top


August 10, 2004

Notes:

  • for weak light sources, a photo multiplier tube (PMT) might be needed.
  • an interferogram is a path length-signal graph (I assume path length meaning path length difference because that's all that matters)
  • a computer-linked PD might have to be used, because turning a translation stage is not only tedious but horribly inaccurate
After lunch I moved the sad ball breadboard onto the optical table. Ani and Yaagnik were laughing it off, and I had some reservations in the end, but what the hell, I said, the interferometer will be portable (not much of a benefit considering I am trying to gain stability), and the sad balls can't possibly hurt. In the end I'm not sure they did much good, but I found that remounting all the optics was simple, and realigning everything was easy enough. After putting the Michelson back together I put a rail on the table so that I could translate one of the arms far enough back that the fringes would disappear. Bad idea. The fringes disappeared before I got a chance to move it back, naturally. As soon as I mounted the business on the rail, it went out of alignment and the fringes disappeared. I manually pointed it together, but of course this was useless, because the fringes shook. Moving it further back and realigning it was impossibly annoying. On top of that, as I moved the mirror back the beam coming from the moving arm diverged.

Collimating the beam from the HeNe was never a big concern in my mind, because I never saw the effects of not collimating it, but this really made me think. In a lot of setups online the laser/light source beam is collimated as to avoid the irksomeness of a diverging beam. Of course, I had to use two lenses to observe the diverging effect, but nevertheless, I wondered if it had an effect on how the interferogram would be recorded (of course it would). (Note: think about interference as that between two spherical wave fronts).

I think I'm going to ditch the rail thing. Finding the coherence length isn't worth it, especially since it's probably pretty long (Note: calculate the coherence length of a HeNe). On another hand, I need a way to translate one of the Michelson arms, measure the amount of translation, and record the interferogram. While doing all this, I need to relate path length difference to the signal. Somehow, the translator and the photodetector have to be synced on a computer. Or, there would have to be some way to compare both using a third variable, most likely time. If the translator could tell at what position the mirror is at any given time, and the photodetector could display what signal was taken at any given time, a graph comparing path length difference and signal could still be obtained. I'll find a way.

Back to Top


August 9, 2004

This morning, Dr. Noé went over my abstract with me for a little while, clearing up some clunky parts in the description of my apparatus and in the history of the Michelson interferometer and the applications of FTS in optical coherence tomography. He also gave me suggestions related to what I should do with the Michelson now that it was properly aligned.

One suggestion was to equalize path lengths for both mirror arms. Doing this part of the alignment was a task in itself, because when I went to measure the distance between the mirrors and the beam splitter, I could only use a ruler, which was somewhat inaccurate to say the least. On the other hand, there is a way to tell when the two path lengths have equalized in length by looking at the interference pattern. As the two arms' lengths approach the same number, the distances between the dark and light fringes on the interferogram increase. However, there is no surefire way to tell when the path lengths are 100% equal, and therefore accuracy can only be achieved to a couple of microns, if that. I had a lot of trouble finding the center of the interference pattern, which is where the photodetector would be placed in order to observe fringe changes (although Dr. Noé did mention something about being able to put the photodetector on the rings rather than dead in the center.)

Using the translation stage, I changed the length of one arm until the center of the interference pattern disappeared (i.e. the center became so large due to the close path lengths that I couldn't tell whether I was in the middle of the center or simply did not have an interferogram. And I had an interferogram). Dr. Noé had also suggested moving one arm way out of sync with the other until the interference pattern disappeared completely. Instead of doing that, I followed one of the other suggestions given to me: to find a primitive way to translate the arms smoothly. Dr. Noé suggested using a translation stage and a rubber band, and, misunderstanding him, I tied the rubber band around the translation stage barrel. Duh. Corrected, I put a mount base onto the translation stage, and strung the rubber band between the mount base and the base of the mirror. There were a number of encountered problems: the rubber band would bend the mirror backwards, tilting it upwards, and throwing it out of alignment, and if left for a while, the rubber band would skew the alignment of the mirror not only in the vertical axis but in the entire x-y plane.

Danielle finally brought Dave Matthews. That's two hours of guaranteed good music. Sarah Chang, eat your heart out.

Back to Top


August 6, 2004

Wow. I know that this site is in my weblinks, but I can't stress how much it has helped in figuring out how the setup is going to work. Today I really, really wanted to play with the computer, but since it was the last day of REU everyone was busy and I wasn't exactly sure which program it was that controlled the CCD. Not like it matters right now anyway, considering the CCD shouldn't be considered as a detector for my project.

Last night I also wrote my abstract, which was a little disconcerting. The writing was okay, but when I started writing the paragraph that started "Current progress..." the gravity of the lab's project situation hit me like Mike Tyson on cocaine. After a sickening lunch (the food was okay, I fulfilled the sickening paramater), I made an effort to put one of the arms of the Michelson on a translating stage. Unfortunately, this was one of those stages that doesn't have a clear mount. At least it wasn't to me anyway. After about fifteen minutes of fumbling with the headless bolt threads (the ones that can't be screwed) I finally figured out that there were holes in the top of the stage matched to the ones in the bottom. Despite the discovery, I left this task for another time, and decided instead to measure the path lengths of the two Michelson arms and try to equalize them. Before I knew it, it was time to say farewell and attend the last REU keg...ever.

Back to Top


August 5, 2004

Another week, another Pizza Thursday. Today's was special not only for the pizza (which is always there), but also because the other fellows in the lab (the REU students) gave us powerpoint presentations on their projects. After the presentations we went out and bought $40 worth of ice cream, courtesy of Lidiya. I came back looking for the CCD, and Azure helped me get the computer next to my table, but for some reason or other the CCD could not be found. For now I actually just wanted to see if I could Fourier transform something (that would be a huge step, in my mind at least, toward seeing progress). Seeing that everything I had on the table was already aligned, if and when I want I could have just moved everything to the sad-ball stabilized breadboard, or just move the breadboard onto the table (NoteEDIT!: the sad ball mounting slab is 7cm deep, which is much too high for the laser already mounted on the table [on the other hand the sad ball stage fits on the optical table perfectly, so all the components - laser, mirrors, lenses, beamsplitter - can fit on the breadboard]).

Back to Top


August 4, 2004

Today was the tour of the LTC (and Danielle's birthday), given to the other Simons Fellows. Their interest at first was dubious at best, but when Dr. Noé broke out the rainbow glasses and diffraction gratings, we gained some ground. In the conference room each of us showed the Fellows our websites and gave introductions to what our projects consisted of. Our guests, from the nuclear physics group, also gave short and quite cryptic summaries of their projects. Inside we gave individual, short presentations on our projects, including demos of our setups. Danielle showed off her Talbot effect setup, I demonstrated interference with the Michelson interferometer, Matt illustrated cavity expansion with his polarized laser, and Ani played around with his tweezers in an effort to find the laser beam.

In the afternoon, I talked to Azure a little bit about how I could find the spectrum of the light sources I had in mind, and she told me that with a CCD, the intensity values across an interference pattern can be obtained, and that taking the measurement of pixel values will result in a nice graph, which can be Fourier transformed. (Note: I believe at this point I was misguided in what I believed happened in the process of Fourier transform spectroscopy; the interference pattern isn't recorded, but rather the change in the interference pattern). At any rate, looking in Mathematica I saw that there was a Fourier transform function, and using the data collected from whatever interference pattern I got, I could find the conventional frequency-intensity spectrum using the FT algorithm. Finding this out got me a little too excited. Before I went to be merry (who knows why this was so jaw dropping), I asked Azure if a photodetector would work for the same purpose, and apparently she was as clueless as I was about scanning an interferogram, because she said a photodetector could be used, but only if it were translated in the x-y plane. The important thing is, I know now what to do with the photodetector.

Back to Top


August 3, 2004

After suffering another jolting wakeup call from the malfunctioning residence hall fire alarm, Danielle and I joined Yaagnik and Azure on a journey...to the depths of (cue scary music) the SAC. Unfortunately, while usually we shove our stolen food in our clothing just as squirrels load their cheeks with acorns, today was a different story. Not only was a guard around, but she shooed us away. Unable to accept defeat, we snuck around back, on a mission to steal the forbidden fruit (and bagels and muffins and coffee). At first, the coast was clear, and we proceeded toward the entrance to an enticing complementary culinary cove. Having spotted the ever-watchful guardian of the lode, however, we scattered, much like incoherent light does after entering tissue for optical coherence microscopy analysis. Yaagnik, sadly, was left in the open to be reprimanded by said sleepless sentinel, whose shrieks of "Ya'll can wait outside as long as you want but you ain't gettin' any of this food!" chased us all the way back to the lab.

On another note, I did do work today.

Starting out this morning by browsing the 101+ physics experiment page to look for some insight on my project. One thing I saw was the gas tube/vacuum idea that would change the index of refraction of Michelson arm path lengths (I mentioned it in yesterday's entry). For a bit I tried a couple of lasers in a simple Michelson setup in the back room on a table supported by sad balls. I guess they're sad because they're rubber balls that don't bounce. At any rate, when I begin constructing my final Michelson, I should support my table/slab/plate with the balls, which are there to absorb shock and therefore create a much more stable interference image.

After some lunch, and after giving up on the rubber-band supported laser that I had been using (and that Dr. Noé suspected was arcing in its cavity), I tried to align the Michelson with cross-hairs. Now, this might have worked if both cross hairs were perfectly centered and raised at the right height on their respective mounts, but unfortunately, as good as it sounded for aligning the laser to the primary axis of the Michelson, it didn't work. I realized it would have to be perfectly aligned in order to work. Then again it would also breed perfection in alignment. What a conundrum.

Eventually I gave up on the sad ball stage and moved onto Lidiya's table, which already had a laser clamped perpendicular to the mounting surface. This removed one dimension of alignment. Eventually, I put together the entire Michelson (with much help from the already-aligned laser that Lidiya mounted) and leveled the laser. Unfortunately, despite the fact that the laser beam reflected back into the source and was perpendicular at all places, there were a lot of artifacts in the resulting interferogram. I suspected at first that it was dirt on the mirrors or beam splitter, but after Azure wiped them all down, the dotty junk was still present. I then set off on a search for a cleaner beam splitter, but to no avail. One of the mirrors is also a little old and scratched, another possible reason for the less-than desirable pattern. I've been getting along nicely in terms of getting alignment down pat, so there shouldn't be much of a problem when I resume my fiddling tomorrow.

Taken from the back of a Fuze bottle: "Zinc: Known as a 'mind-sharpening' herb [sic] that enhances capillary circulation and increases the supply of oxygen to the brain."

Back to Top


August 2, 2004

Chris Wottowa's paper has been very helpful in instructing me how to build the set up that I will be looking at. I have yet to determine what I am using to translate the mirror, and whether or not I use a PZT depends on whether the PZT's expansion can be measured relative to the intensity of the interference measured by the photodetector. On another note, while browsing the websites in the Fourier transform spectroscopy section of my links page I found one source that continually showed up as a reference: R.J.Bell. Introductory Fuorier Transform Spectroscopy, Acad.Press, New York, 1972.

Immediately I went to the Stony Brook library online catalog and searched for it. Lo and behold, the book was available at our own Mathematics Physics and Astronomy library. Trekking upstairs, Azure and I began a futile search for call number QC.451.B46. The book wasn't in the physics library. Across the way at chemistry, however, where we met some resistance ("What are you doing on our turf? You're from physics. We don't like your kind around here"), the book was found. Right after returning I dived right into the book, finding immediately useful a section titled "Optical Alignment of a Michelson Interferometer." This was the solution to all my queries about how to align the mirrored branches of a Michelson.

On top of that, I read the first chapter, which contained a boatload of information on terms I had always wanted to know about (e.g. the multiplex advantage), as well as some anachronistic statements, like a line about access to computers being prohibitively expensive, another comment about using a sliderule to perform simple calculations, and the last about a $30,000 "minicomputer" with a "4000 word memory."

Tomorrow: chapter 2 and maybe a Michelson

Back to Top


July 30, 2004

Today, I found new sites on spectroscopy. One of them offered a new (to me) way to vary the path length of one of the mirrors - rather than physically moving the mirror, this group used a gas tube to which they added air and removed it to vary the density of the medium through that specific section of the Michelson, and thus vary the effective path length. To build a lab setup I just need:

  • a light source - could use a He-Ne to start out with, just for the hell of it, and test its spectrum)
  • a Michelson interferometer - easily built in half an hour (with mirrors of course, which might be hard to dig up)
  • some type of motor, PZT, or anything that can translate a mirror over wavelength (hundred nanometer scale) distances
  • photodetector or photomultiplier tube with computer interface - there are lots of them sitting around in the lab
From these ingredients I can build a simple test setup to analyze the spectrum of a He-Ne, and from there I can start analyzing some more exotic light sources. I could start first with an LED, and graduate to things like high intensity discharge headlamps and maybe even a superluminescent diode if we can get ahold of one. If, eventually, I do the Fourier transforms on the light sources I test, and am successful concerning spectroanalysis, perhaps I can move on to something that more closely resembles an optical coherence tomography layout.

Today we also went to the hospital to eat, which from experience we knew would be cheaper and better than SAC food. Despite the fifteen minute walk there (including getting lost), the food and the prices were worth it. After we came back, I began building a Michelson again, despite not having any materials except the mirrors and beam splitter. The kegger today had very disappointing salsa, but surprisingly good nachos. Despite the poor salsa quality, we had watermelons and nuts, two things missing in action since the first kegger.

Back to Top


July 29, 2004

Today, in the morning, I prepared for my little presentation for Sam Goldwasser by reading over all my resources on spectroscopy, optical coherence tomography, and Fourier transform spectroscopy. I took notes on Fourier transforms first, and then I looked over how exactly FTS works. When we finally sat down for a massive lunch of pizza and a presentation by Sam, we listened to various topics, including the Zeeman effect, an introduction to pumped green diode lasers (like the one in the lab's laser pointer), and a rundown of how he built a Fabry Perot interferometer. Afterward, we all gave our talks (and I didn't use any of my notes, idiotically) and watched a number of demonstrations that Sam showed us. We witnessed the beating of two lasers on the oscilloscope as a result of the Zeeman effect (which Matt claimed, and probably rightly, was the "coolest" thing we'd seen in our weeks here) and the revealed insides and construction of a green diode laser.

Back to Top


July 28, 2004

This morning I read the papers that Dr. Noé printed out for me yesterday, some of which I had already looked through on the computer. Most of them were on the topic of the function of optical coherence microscopes, or the techniques of optical coherence tomography. They're all quite interesting, but don't quite offer a simple setup for lab use. During lunch, we had a Simons meeting about writing abstracts which degenerated into talking about posters and what to wear to our farewell assembly on August 13.

This afternoon, after doing some more reading (and getting free bead keychains and ice cream from the SAC - which reminds me...today, during our brown bag lunch, Karen Kernan, the director for our program, mentioned free ice cream, and immediately looked at Danielle and me!), I was greeted by Dr. Noé with some good news: he had a long phone conversation with Richard Haskell (author of the Harvey Mudd College Optical Coherence Microscope page) about optical coherence tomography and possible projects that I could carry out within the laser teaching center. What he proposed was a Fourier Transform spectrometer, a concept that is key to optical coherence tomography.

Fourier Transform spectrometers use Michelson interferometer setups, which consists of two mirrors perpendicular to each other and a beam splitter between them turned at a 45° angle. A light source (whether it be a He-Ne, a superluminescent diode, or what I will most likely use, a simple LED) is directed into the beam splitter, and the light that is reflected back through it forms an interference pattern. However, in a FTS system, one of the arms of the Michelson is moved. As a result of this motion, a changing interference pattern is created (the light interferes constructively and destructively according to the path length, varied by the movable mirror).

Back to Top


July 27, 2004

This morning after helping with some multimeter problems and after searching for some superluminescent diodes online (with pretty decent success), I found again the original paper on OCT, written by D. Huang et al., published in Science in 1991. Although I did a filetype:pdf search I couldn't find the paper. I did find the summary, brief, and abstract, but not the paper itself. I'm thinking about building a spectrometer (just for the hell of it?) and figuring out how, if at all, an experiment might incorporate optical coherence tomography. Once I figure out how the setup works, and once I talk a little more with Dr. Noé, I might be able to carry out construction of a working one-dimensional OCT device. The problem with this is whether it is possible to demonstrate the concepts behind OCT while only scanning in one dimension.

This afternoon consisted of a torrent of reading concerning OCT and SLDs. I have found approximately five or six sources selling SLDs, although some of them are repeats (they sell what seems like the same model). Everything is in my bookmarks, and I will update the weblinks tomorrow with some of results from the SLD searches. At around 4 o'clock we also looked at some of Matt's recorded data, recorded from tinkering with his laser set up. His measurements of intensity with a photodiode almost exactly matched a Gaussian beam distribution.

Back to Top


July 26, 2004

Today was reading day. For the longest time since I can remember, we haven't had a day without events, either in the lab or out. Nevertheless, despite the lack of mandatory fun, this was one of the most productive days we've had. In the morning, I collected two articles on optical coherence tomography (an idea which has been abandoned but not forgotten) and read up on bioluminescence. So far, I've found basic chemical processes that make up bioluminescence, most of which, in animals anyway, are stimulated by a reaction that emits light due to luciferin, catalyzed by the agent luciferase. One idea floating around my head for the past week has been spectroanalysis of a bioluminescent light source. One example might be holding a firefly in a pipet tip, similar to the setup we saw during the CMM tour for the Drosophilia, having it shine light, and then analyzing the spectrum.

After lunch, my time was split between helping with and observing Ani and Danielle's improving setups, reading about coherent backscattering, diffuse reflectance, and optical coherence tomography, thinking about more project ideas, and helping (maybe?) to derive an equation that Dr. Noé put on the board. After dinner I talked with Dr. Noé about my current progress in terms of finding a feasible idea. I learned that a lot of the things I had on my mind are simply too difficult to do given the resources that we have and the time constraints. These ideas are currently in limbo while I determine how do-able they are:

  • firefly spectroscopy - light from the bugs would be too weak to register in the inefficient spectrometer; Dr. Noé did suggest other forms of spectrometers, e.g. a Talbot effect spectrometer
  • bioluminescence with plants - by grinding up plant leaves and illuminating them with light, they luminesce, an effect which may be measurable, or at least detectable [note: this might go too deep into biology]
  • one-dimensional optical coherence tomography - while 3D OCT is too difficult (it requires computer-controlled motors to scan tissue), it might be possible to construct a 1D apparatus, which could detect tissue phantoms, like a dense object suspended in gelatin, or a bone in a chunk of chicken breast; the light source is another issue - only a directional, polychromic source would work, ruling out use of an LED or a laser [note: I did find superluminescent diodes for sale online]

These projecs still seem possible, though of course I can't say for sure they can be done, or are worth doing:

  • spectroanalysis of exotic light sources - sources include new high intensity discharge headlamps on cars or recently installed LED traffic lights; the problem is that this might be an interesting thing to examine but might not be suitable for a project
  • the GrIn tank - I have really pushed the possibility of a project involving gradient index of refraction into the back of my mind, but now that a lot of other things are being ruled out, it has surfaced again as something I'm also interested in examining

Regardless of what comes of my project ideas, I would still love to do an analysis of bioluminescence/chemiluminescence. Originally I was pretty much engrossed with optical coherence tomography, because just the concept of being able to peer into a solid with light (and more immediately, the idea that I could do such a thing) was overwhelming. However, figuring out what happens in a firefly that makes it glow, and why the light that is emitted has a specific composition has just captured me. It's a hard concept to carry out, and as much as I don't want it to it really delves into biology and chemistry (I'd like to stick to physics to a feasible extent). For now I will read the papers that I printed out today, and keep looking for ideas that might work.

Back to Top


July 23, 2004

Today is my birthday. It seems the same as any other day. Ah, well. This morning Dr. Metcalf dropped by the LTC again to talk about our projects, and we collaborated to write on the board a working table of topics and progress for each person. Azure wrote everything down on a sheet of paper, which we might put up on someone's webpages sometime soon. In the afternoon I went up to the library and finally read those journal articles, which were on confocal interference microscopal imaging, volume holographic imaging, and hiding images using a light-scattering medium for encryption. Sadly, a large part of the articles was of little use to me, partly because it was hard to understand and partly because some of the discussed topics simply did not apply to what I wanted to find.

Today was also the usual Friday kegger, which was jolly good fun, especially with the tubs of mango salsa, of which we stole a large portion of. I'm a little disappointed at the food that comes up at the kegger. The focus on the beer is too great, and the focus on getting decent chips and nuts too little. On the other hand, like they say, you can't have your cake and eat it too, so free loaders like us will just have to settle for what we can mooch.

Back to Top


July 22, 2004

Today we had a talk with Dr. Metcalf (our usual Pizza Thursday) during which we told him what we were interested in and talked about possible project ideas for all of us. After telling him about my interest in optical coherence tomography and spectroscopy, Dr. Metcalf suggested that instead I should look into backscattering as a way to analyze tissue (chicken breast?) or analyzing exotic light sources using spectroscopy (bioluminescence in fireflies perhaps, or LED traffic lights and xenon high intensity discharge automobile headlamps). By weekend's end I will have hopefully settled on a topic.

Later on today, José had a talk on laser diodes, which was quite informative. We also talked a little bit on the side about other topics. Having read about super luminescent diodes (SLDs) used in optical coherence tomography and other interesting optics applications, such as image encoding, the diodes talk was very relevant to the reading I've been doing.

Back to Top


July 21, 2004

This morning, after talking to Ani about yesterday's MOT talk and how the concepts of magneto optical traps work, our whole lab went to the Simons tour of the Center for Molecular Medicine (CMM). Most of us were just there to mooch food, but it was quite interesting, and surprisingly, optics came up more than once again! First, we went to a lab where the scientists were researching the aural mechanisms of Drosophilia melanogaster. For a demonstration, we spoke through a little tube and saw the vibrations of the fruit fly's antennae. The vibrations were connected to two wires which were linked to a very expensive and cool looking oscilloscope. Danielle and I were jealous. They also had hold of an impressive 20-watt pulsed laser which, when first turned on, burned a hole in the wall. From my reading, I recognized a confocal microscope and asked the scientist in the lab to explain how it worked. Although it was a little unclear, there are lasers involved (reminder - read those damn library articles!).

Later, we went to the basement, which housed the Nuclear Magnetic Resonance (NMR) equipment. From our tour of the LINAC and our talks, I answered questions about the magnetic field of the earth (1 Gauss) and the boiling point of Helium (I said 3 Kelvin - it was actually 4.2). One of the graduate students (lab assistants?) spoke about x-ray diffraction, a technique in which diffraction patterns from an x-ray being shined through a crystalline structure are analyzed to determine the chemical properties of the crystal (in this case, proteins). It was a little frustrating considering they didn't expand very much on how the equipment worked, except to say the particles are kinked and interact with the magnetic field, and that from the kinking/returning process the makeup of the particles can be determined.

After the tour of the CMM we scrounged a bunch of free food from the reception afterward and then played virtual reality games at the SAC (on top of getting more free nachos and popcorn) for a good fifteen minutes. At 3 o'clock, Lidiya began her Fourier talk, which we had been looking forward to for week(s). In the end, it was one huge mathematical proof, which I have to go over again to absorb.

Back to Top


July 20, 2004

This morning was chock full of fun. First, we played with the Michelson until we saw some fringes. Dr. Noé suggested we create a bullseye interference pattern by placing a lens in front of the laser source and moving one of the mirrors further away from the beam splitter. At first, all we saw was a quarter-bullseye pattern. All Dr. Noé did was turn the knobs on the mirrors and...voilà! The full bullseye appeared, centered on our image "plate" (a piece of oaktag). Then, on top of watching and helping Yiyi set up her optical tweezers, the lab entertained the REU fellows dropping by for a tour. Matt showed off his simple laser-polarizer setup to demonstrate cavity expansion in a HeNe laser as one of the many exhibits we put on in what the REU students called the "science museum" (the LTC).

After a late, late lunch, Dr. Metcalf gave us part three of the MOT talk, in which he further described magnetism and how the "optics" part of magneto optical trap actually works. The truth is, although I did take notes and participate, albeit minimally, in the talk, MOTs are much more complicated than I ever imagined. Tomorrow, I'll go over my notes again - especially the part about the three sublevels. That stuff really throws me.

Back to Top


July 19, 2004

This morning we had another long talk in Dr. Noé's office, where we talked about each other's projects and a lot of new, interesting stuff. I put some of it in my weblinks. Again, I remained focused on OCT, optical coherence tomography. However, seeing that this may not be that feasible, I did make note to return to the library and read the articles I jotted down from Applied Optics. One of them compares optical coherence tomography and coherence-probe microscopy, another explained volume holographic imaging, and the last described hiding and decoding an image with light-scattering media.

Before lunch, we decided to construct a Michelson interferometer, which was done in approximately 45 minutes (much shorter than I thought it would take) with José's help. After a late lunch, we observed the interference patterns resulting from a change of iris size in a certain part of Lidiya's set up, and asked Yiyi to demonstrate her laser tweezers. It wasn't till after dinner that the tweezer set up began to really get going, but before we ate, Ani and I went up to the library and I tried to take notes on the articles I took down. I was pretty far off in thinking that it would take 15 minutes to read three articles. It was more like fifteen minutes just to get through the abstract and introduction of each paper. So, we went back to the lab and left just in time for dinner. The rest of the day was spent observing how the laser tweezer set up was tested and how to align it so the beam actually goes through the microscope.

Back to Top


July 16, 2004

Today, after updating my journal, our lab group went to the Intel/ISEF meeting at Light Engineering, hosted by our friends (enemies) at the Garcia Center for Material Science Engineering. Unfortunately, we were unaware that the meeting was meant for Garcia students and that we were just guests. As such, we left after a couple of minutes after realizing much of it would be lectures relating to Garcia business and not Intel.

As soon as we got back to the lab we dived into a talk about random optics, led by Dr. Noé at the white board. From here I discovered the 105 Laser Experiments booklet, from which I asked different questions concerning various optics topics. Soon after I registered the website into my favorites and my weblinks page. From this long talk we suggested one easy to do project - building a GrIn (Gradient Index [of Refraction]) out of high fructose corn syrup and water. The GrIn would act like a reverse mirage by bending light in a frowny face rather than in a U shape like a mirage does. After lunch, Ani and I helped to fill the all-purpose fish tank with corn syrup and clean out the syrup bottles. Then, off to the kegger.

Back to Top


July 15, 2004

Last night, at 1 AM, my hallmates decided that my room was boring, and, Trading Spaces style, they turned my room into an entertainment center (minus the entertainment - i.e. TV, stereo, grill, etc.). The two twin beds in the room were smushed together to create a queen sized bed!

On another note, today was Pizza Thursday, and we talked about what we wanted to do for our projects (for now, I feel interested in holography and its applications in backscattering and tissue analysis - optical coherence tomography). Afterward, Dr. Noé showed us a phenomenon which involved "all of optics" - a HeNe beam that, when shined through a fish tank full of dissolved non dairy creamer, pulsated in intensity because of cavity expansion. On top of that, when more creamer was mixed into the tank, multiple scattering occurred, and the once polarized scattered light was now not polarized anymore. The same thing occurs when pointing a laser pointer at a wall - the light isn't polarized anymore, but rather it is scattered in different directions by the textured wall and when it is seen through a polarizer its intensity remains the same no matter which way the polarizer is turned.

Today's talk made me think about where I am going with my project. It is already the third week (soon to be the fourth), and I still have no exact direction. As such, I've finally touched the weblinks section of my website, and rather than doing general searching, as I have been, I've settled on a shortlist of topics:

  • optical coherence tomography (look up: superluminescent diodes)
  • spectroscopy
  • microscopy
  • holography (either in tissue imaging or cryptology)
  • chaos(??)

Yes, there is a lot more reading to be done, and of the topics that I have read about so far, despite being most interested in the above topics (for a project anyway), I've done little in terms of figuring out how each of these specific processes work. Tonight, after both a trip to the LINAC and dinner, I did reading in the lasers book (Lasers - Eberly and Milonni) that José has on his desk. It helped to clear up the lecture he gave and provide a number of equations to augment topics and ideas I've learned already, like how collimated lasers work, what goes on in the laser cavity, and how the Rayleigh range and aperture size of a laser are related.

Note: Weblinks will be updated first thing in the morning.

Back to Top


July 14, 2004

Today was the Brookhaven National Laboratory trip for all the Simons Fellows, which Danielle and I both attended. Suprisingly, despite how much the tours of the accelerator and the sensing equipment for the RHIC (relativistic heavy ion collider) reminded us of nuclear physics, we couldn't help but think optics was at work at every turn. The tour guides cited UV and infrared spectroscopy as the driving mechanisms behind the National Synchrotron Light Source, and on our tours there were plenty of warning signs telling of laser light. Optics is everywhere!

When we got back from BNL, I ate a huge lunch and retired to the lab, where I found that the other LTC students had taken a tour of the Van De Graaf accelerator and done interesting physics (including matrices!) on the board. Regardless, we had a 3 o'clock talk with José which lasted until about five o'clock. After dinner, I came back to make my journal links funkified, and I finally finished the optics chapter of the physics textbook and cemented all my basics in.

Back to Top


July 13, 2004

Last night at 3AM, just as I was entering deep sleep, the fire alarm rang in our dorm. At first I just waved it off, but then a few minutes later I heard people screaming. Thankfully, there was no fire, and the people screaming were doing it to get people out of their rooms. However, the entire residence hall had to get out, half of us having to walk (most of us in our underwear) in the freezing rain. On another note, this morning Dr. Metcalf gave part two of the MOT talk, which was essentially composed of a breakdown of characteristics of photons. Unlike other times when I entered all my information into this journal, I took notes in my lab journal today. As a result, most of the information on both of Professor Metcalf's recent talks is in the journal. We learned more about different phenomena that occur in atoms, specifically about energy levels.

In the afternoon Danielle and I went to the library, and I looked at more Applied Optics journals from 2004. Although I found a number of interesting articles and read through their abstracts just to get a good idea of what they were about (I wrote down the topics in my notebook - spectroscopy was one, and another was holographic tissue imaging), for the most part they did not help me establish basic concepts that I think I'm a little rusty in, having taken Physics B two years ago. I think Danielle, on the other hand, was going in the right direction, reading about different types of lasers and exactly how the light is amplified and how it escapes out of the laser cavity. There is one book that she is reading in the library, and another on optics in the lab, which I will begin to browse through soon.

After dinner, I picked up Hecht's Physics: Algebra / Trig to start reading some basic optics concepts like how an electromagnetic wave actually propagates through a vacuum. Tomorrow I hope to finish out the optics chapter (it's hard going back to elementary physics) and begin reading. Then again, tomorrow is BNL and we're having a talk with José, so...yeah.


July 9, 2004

Dr. Metcalf gave us a talk on the "magnet" part of MOTs (Magneto Optical Traps), which helped clear up interactions between particles in space. Dr. Metcalf explained velocity space and phase space density, the combination of velocity space and position space, defined by six coordinates: Vx, Vy, Vz, Sx, Sy, Sz.

Back to Top


July 8, 2004

Today started with welcoming Professor Kiko Galvez from Colgate University, our speaker and guest for today. For quite a long time, we croweded into the little room (the laser room? the tweezing room?) and talked about Yiyi's laser tweezing set up. Mainly it was a discussion between Yiyi and Professor Galvez, but we watched them talk about different laser tweezer set ups, and use of new things that we need to examine like donut shaped laser modes and, among other things, chaos, like Rita spoke about yesterday. We also talked about Professor Galvez's current work with higher-powered Argon lasers and the Stony Brook University Hospital's work with lasers with power on the order of one watt.

Lunch consisted of pizza in the conference room and a presentation by Professor Galvez on converting laser modes into different shapes (including turning a Hermite Gaussian mode into a Laguerre Gaussian mode). Afterward, we entertained Professor Galvez in the LTC for a while, and then, realizing much of the talk had escaped our grasp of laser modes, Danielle, Ani and I had a long talk with the famous Alex Ellis and José, both of whom taught us about laser modes (try typing in "exotic laser modes" into Google - the first hit is Alex Ellis's site) and the concept of polarization. Essentially, we knew about horizontal and vertical polarization, but from José we learned that light can also be circularly polarized, and we learned how exactly it happens. Azure also helped explain how polarized light "forms itself into twisties," a statement that eventually made sense. We also figured out how HG and LG modes work, and we learned their denotations, how to modify the modes, and what happens when operating paramaters are changed.

Back to Top


July 7, 2004

I spent this morning updating yesterday's journal and reading an issue of Physics Today which reminded me greatly of a discussion we had with Dr. Noe about metastable helium. The morning ended with Danielle asking for help on Linux. Lunch consisted of a loong brainwashing session led by the Stony Brook Admissions Office. On the upside, there was free food outside the SAC, and free spray painted t-shirts (Yiyi and Azure both got equations on theirs). A bunch of us were watching them being painted for a while (it was quite impressive).

Afterward, I remembered the article that I was reading the other day in the library on phased arrays and wanted to know exactly how the term was being used in that particular piece. THe article, called Detection Performance of a Diffusive Wave Phased Array, shows that phased arrays are much more effective at identifying inhomogenous objects in tissue than one single system. After the short library trip we returned to the LTC and listened to a talk given by Rita on spatiotemporal chaos. It took me a while and some talks with Azure and Ani to (almost) fully understand it, but it was an interesting topic, especially because Danielle is interested in chaos. In short, she used a laser-SLM (spacial light modulator) combination to create chaos on a liquid crystal surface. She carried out experiments with feedback from a computer and without feedback, and found that with the open circuit (without feedback), the chaos created by the liquid crystal surface turned out the same every time.

Back to Top


July 6, 2004

Today, the lab was being waxed so we went to the conference room to have some short talks about our projects and about lasers in general. Part of the discussion focused on doing some short problems concerning the installation of windmills off the south shore of Long Island, which we figured out using the degrees of arc calculation we did with the sun. Afterward, we blew into a half-full coke bottle to review basic wave concepts like resonance and standing wave patterns in closed-ended containers. We also talked about metastable helium molecules which are used as mini atomic bombs, which can hold great use in lithography and nano-circuit design. The last thing we did today was to go to the oscilloscope and try to determine whether our estimates about the water bottle's frequency were good. Unfortunately, we lost our other water bottle and used another that Danielle had on hand. With some calculations with wavelength and frequency we finally recognized the fact that shape (or volume) as well as length affects the tone of a Helmholtz resonator (any container of gas with an open top).

Back to Top


July 2, 2004

*DANGER - LONG ENTRY AHEAD*

Today, after reading some OPN, I followed Professor Metcalf into a talk on laser cooling, continued from our talk the other day. First, we spoke about the motion of particles, and then about excitation of specific atoms. Dr. Metcalf explained that only certain frequencies of light could excite atoms into higher energy levels, and that these same frequencies of light would be released when the atom decayed back to its ground state. The approximation of frequency accuracy turned out to be about two parts in a billion, which Dr. Metcalf described would be like lining up ticks from here (Stony Brook) to the Empire State Building and choosing a single leg from one of the fleas as the correct frequency. Then, as the presentation went on we learned that if there was a frequency (ω) and time (t), ωt = ~1. Therefore, with a given frequency &omega and error &delta, the original graph of the energy released from the atom would be similar to the graph of energy released from the atom using the &omega ± δ frequency (i.e. even with error, the energy absorbed and released by the atom in question would be similar). However, this is only for a short time period. So, if the graph of energy absorbed were viewed over a longer period of time, the error would become much more apparent. From this we learned the basis of the Heisenberg uncertainty principle.

On to laser cooling then. The other day we learned that Δp =0. which represents the idea of conservation of momentum. We also learned that light, which is composed of photons (even though today it is Friday - light is a wave today), has momentum, and when it strikes objects it transfers its momentum. Another concept besides momentum central to laser cooling is the Doppler Effect, which states that if a point of reference moves in the same axis of motion of a wave, the perceived wave frequency will be greater or smaller, depending on direction of motion. So, if we have for instance an atom that is going to be cooled, and we shine laser light in two opposing directions at it, it will be trapped in what is called "optical molasses" - the slowing of moving atoms. If the atom is not moving, the net force on it due to the lasers will be zero (equal and opposite forces cancel out). However, if the particle begins to move right, the positive frequency of the laser moving in that direction will seem smaller due to the Doppler Effect, whereas the negative frequency of the laser moving in the opposite direction will increase force on the particle to the left.

In effect, this represents velocity-dependent force (that is, as velocity increases, the effect on the particle due to the Doppler Effect increases, and as the particle's velocity is slowed by the optical molasses, the effect of the lasers slowing down the particle is decreased). We can see therefore how optical molasses, and laser cooling, works: the forces on a specific particle will always contradict the direction of motion (not unlike friction, the only other velocity-dependent force), and slow the particle, and thereby lower its energy, cooling it. It is a deceptively simple concept.

The last thing today was a little talk by José on CCD (Charge Coupled Detector) cameras, which are basic digital cameras that detect photons by transferring them into electrons. The surface of the CCD is made of a material that releases electrons according to a quantum efficiency - a 40% efficiency would release 4 electrons for every 10 photons that strike the surface. The released electrons fall into little wells that catchthem. One phenomenon that happens is "blooming," when electrons overflow out of their respective wells, and light up lines of pixels in the camera's display. Another thing we learned is the effect of thermal noise in the CCD's display. All the pixels automatically have a certain level of noise, and infrared light also may trigger release of electrons, which results in a pretty but noisy image which looks somewhat like a galaxy. Finally we played with diffraction gratings and lighters and a green laser to figure out how the CCD really works, how it records images, and how to play around with gain and bias, the two light-enhancing imaging tools which use multiplicative and additive factors to respectively alter contrast.

And today is Friday, which means I will be home for the long July 4th weekend.

Back to Top


July 1, 2004

Today, early on, we determined that Matt's new nickname would be "Penguin," because he is a Linux expert, and the Linux mascot is a penguin. Last night, Danielle taught me how to play lacrosse, except with a girl's stick. I also learned how to catch on my left side, which is pretty tough. On another note, the lab had its first Pizza Thursday today, when we eat lunch, and, according to Dr. Metcalf, talk about "random crap." After a quick lunch and some optics discussion, Yi-yi, an REU fellow and winner of massive amounts of money and awards in ISEF and Intel, along with Azure, another member of the Laser Teaching Center lab, led us to the SAC to mooch food off a presentation being made to incoming freshmen.

Much of the afternoon was devoted to writing in my journal about new topics and learning a new contradiction dealing with a strange new diffraction gradient. When we shined light through the gradient, which had holes in it, it did not diffract normally. Rather than being a bright dot in the center and less visible dots moving outward from center, the dots switched between bright and dim. Another member of the lab who has been looking at the problem, Lidiya, helped point out the mystery: the strange image was probably due to two interference patterns interfering. While a single-slit interference pattern has a bright dot in the center, double-slit interference patterns switch between bright and dim. Therefore, the diffraction grating's single slit combined both multiple-slit and single-slit diffraction patterns. I will look into this further tomorrow.

Ani and I also spent a little time trying to figure out an 8-layer Towers of Hanoi puzzle, which turned out pretty difficult. After we gave up and I started updating my journal, Dr. Noé gave us a quick question to solve: when light is focused through a perfect converging lens and crosses the principle axis at one defined point, what is the relationship between the light's original distance from the axis and the angle at which it crosses the axis, keeping in mind of course that focal length remains the same. Even though I was right, Danielle and Ani continued to doubt me. Eventually they'll realize that my inverse radius-tangent relationship is correct.

Back to Top


June 30, 2004

Yesterday I woke up early (7:30) and got to the lab to continue the experiment relating focal length and image size. Finally, at the end of the day, after making measurements up the wazoo and lighting sheets and sheets of paper on fire, we all graphed our results, which were way off.

Today, I woke up at 8:30 thinking that everyone would do the same. Wrong. When I got to the lab everyone just looked at me like, where were you. However, I made up for my tardiness by making my journal and starting my bio, with help from Matt, the resident Linux expert. I also used my glasses as an example of a cylindrical lens (I have an astigmatism!) as part of a demonstration to some seventh graders.

After a quality "brown bag lunch" (aka bring your own food - I had a slice of pizza and a bacon cheeseburger from the SAC) with Karen Kernan, our director, about lab safety, Danielle and I went back to have a talk with Professor Metcalf. He gave us a short introduction to the physics of electromagnetism and the mechanics of atoms (including but not limited to the principles of conservation of momentum and a synopsis of Maxwell's work in relating electromagnetism with optics), and then delivered a short introduction to laser cooling.

To understand laser cooling, we only have to understand one basic concept: cooling occurs when energy is removed from a system; less movement and lower speeds is equivalent to less energy; and therefore, slowing things down is the key to cooling them. In this way, scientists use lasers, whose photons collide with objects and slow them down by transferring their momentum, to slow down particles and thereby lower their energy and temperature. Part of this talk also consisted of particle-wave duality, and we learned that in the Physics building, on Mondays, Wednesdays, and Fridays light is a wave, on Tuesdays, Thursdays, and Saturdays, it's a particle, and on Sundays it's anything. After the talk Dr. Noé had a question-and-answer session with us and showed us which journals to read in the lab (also known as "the den" because it has couches and is a comfy little area) and which not to read. Tomorrow I will begin hardcore browsing of OPN, Optics and Photonics News.

Back to Top


June 28, 2004

Today, to summarize, I arrived at Stony Brook University for my first day as a Simons Fellow. After meeting in the Wang Center for Asian Cultural Studies (AKA the Schlong Center) and after getting acquainted with my two mentors, Professor Metcalf and Dr. Noé, I moved onto the Laser Teaching Center Laboratory. With my fellow Fellow Danielle Bourget, and two other high school students, Ani and Matt, and under the guidance of Dr. Noé, I burned holes in pieces of paper with magnifying glasses.This kind of grew into an experiment relating focal length and image size.

In the middle of the day, at around 2:00 after lunch at the SAC with Dr. Noé and Louise, an REU fellow, I tried to get my ID card but couldn't (the woman said I needed to have my old card), and then went on a campus tour. Afterward we went to a seminar on Proposed Electric Dipole Moment Measurement Using Radon Isotopes. 'Nuff said. Finally, after Matt left, Danielle Ani and I went to the SAC to finish up our journals. By the time we were finished it was past 6 and the SAC was closed. I bought a roast beef sandwich and got ditched by Danielle.

Back to Top


j j w u m a s t e r ( a t ) g m a i l ( d o t ) c o m