Project Hologram

At the school my friend and I attended, we were given six weeks to work on a project of our choosing. Many students work as unpaid interns, doing data entry for hours on end (but gaining valuable work experience!) Our ambitions, however, were on a much different scale. We first decided that we would get sponsored for a human-powered aircraft, costing upwards of $10K and taking probably longer than six weeks to create. With a 20 meter wingspan but a maximum weight of around eighty pounds, it probably would not have been the easiest engineering feat to accomplish. After some thought, we settled on something much more reasonable: spending six weeks making cool holograms.
My father works at Johns Hopkins University, so we were given a surprisingly large amount of lab space to work with. Our 'supervisor' was a guy from MechE, who was luckily able to lend us a laser, lenses, and mirrors to work with (optical equipment they had used in the seventies for similar interferometry experiments). Another important aspect of our lab, besides its size, was a ~1 ton optical table sitting in the middle. Though we weren't sure at the time, that piece of equipment isolated our holograms more than adequately from all vibration.

Now, another item we were loaned was a special holographic camera that had reusable thermoplastic film, where development was in-place and instantaneous. Of course, after twenty or thirty years, the film was completely bust. Demonstrating our experience brilliantly, we had neglected to order regular, wet film until we were sufficiently sure the holographic camera had failed. So, the rest of the first week was spent fiddling with the equipment and twiddling our thumbs, waiting for the film to arrive. Eventually, it did, and it was the third week of the project by the time we finally succeeded in making a hologram; a grainy, dim hologram of dice viewable from a very narrow range of angles.

Now, I should mention that I won't explain how holograms work on this webpage. There are many others that can do it much better, and one of the ones we referred to from time to time was the HyperPhysics page, chock full o' content (holography under "Light and Vision"). Suffice to say it's like recording an entire wavefront; instead of recording a single view of an object, it's as if you capture all of the light passing through a window in every direction. Similarly, when you view a hologram, it's as if you are looking through that same window - the wavefront is reconstructed (nearly) exactly.

The first holograms we made are called Transmission Holograms, and the laser needs to reilluminate the film in order to view the hologram. After a few holograms of dice, we took a break and switched to another type of holography called Reflection Holography. Reflection holograms are much simpler to make and can be viewed in white light, but their depth is not as good and they are much darker. Anyway, through sheer luck we made two really good reflection holograms: one of a mini space shuttle, the other of a circuit board. Why luck? Well, we tried to make some more at around the fourth or fifth week and failed a great many times. There are a few variables at play - the exposure time, the color of the object, the angle of the film, the polarization angle of the laser. Suffice to say, we used the scientific process of controlled experiment on none of these, a lesson we learned many times over. So, if you're reading this, take it to heart: make sure what each variable does, and for the love of god, get a simple little light meter. More on that later.

After those reflection holograms, we returned to more transmission holograms. Some items of note:
  - There was a rubber dinosaur we tried numerous times to make holograms of. It failed continuously and consistently; the image simply would not appear. Our supervisor, when asked about this, told us that the polymer the dinosaur is made out of actually vibrates and flows/deforms enough all of the time to screw up the hologram (remember, the object cannot move more than one wavelength, or ~632.8 nm). So, that, along with some other plastic objects like a keyboard, failed miserably.
  - After some more transmission holograms, we experimented with some two-step holography. This is essentially making a hologram of a hologram, and it allows you to move the image of the object closer to or farther from the film. It wasn't until we returned to this method towards the end of the project that we had any real success.

Things we tried that did not succeed as hoped:
a) Interferometry. This is the process of measuring tiny deformations in a material by making two holograms of it. Live interferometry obviously didn't work because it would have required us to place the film back exactly to where we had exposed it after developing it. And double-exposure interferometry, where you expose the film twice before and after making some sort of deformation, didn't work because we're klutzes and kept bumping the objects being deformed. I mean, it kinda worked, but not reallly. Ironically, the best interferogram we have is unintentional, of the dinosaur: the skin shifting and the dinosaur tilting slowly caused a faint image of it to appear with large vertical lines running through it.
b) Rainbow holograms. This is a process of eliminating vertical parallax in order to make holograms visible in white light; the difference between these and reflection holograms is that despite not having vertical parallax, these holograms are much brighter than their reflective counterparts. We had a partial success, as you can see below. Partial, in that the image showed up but there was no 3d effect. Why? I think it's because everything was on the same plane, the slit and the recording laser and everything. I think that you need to shine the reference beam from above in order to make the rainbow hologram spread the different colors correctly. When you tilt the image on a vertical axis, the view is supposed to change; horizontal, and the color is supposed to change. Right now, the color and view both change on the same axis and when you tilt it in the other direction it all disappears. So, this leads me to believe we should've shone the laser from above. Unfortunately, we didn't have the time or means for such a setup - this is a shame, because one gets the impression that rainbow holograms could look really really cool.

After fiddling around some more, we finally got to ordering some holographic filmplates, which means the emulsion is on glass instead of crummy celluloid (more expensive, but way nicer). In fact, if you have the sponsored funds for it, I would highly recommend using plates instead of film. Saves you some trouble at the beginning (with setup) and looks much better. We then went through making final holograms, among them transmission, two-step transmission, and some more reflection (the reflection holograms we actually did after the project was over and we had given our presentation, because we had some materials left over). As subject material, we mostly used little objects like lego men, model toys, screws, or some shells and coral that my mother provided us. In fact, our clearest, sharpest hologram was on a small (2.5") plate, and is of the shell pictured below. It is worth noting that a significant amount of quality is often lost when photographing or filming holograms due to the high contrast present in the image (at least, it's that way for amateurs like us).


Technical details

The lab we were in was not light-tight (aka dark), by any stretch of the imagination. There were giant bright gaps underneath the doors, holes in the wall for electrical and network cables, and there were even gaps where walls met in a perpendicular manner. So, during the first week, we spent a day sealing the place up with black plastic and cloth to the point where there were only a few extremely faint dots of light visible around the door; sometimes, none at all. It was extremely dark, and we didn't have a safelight (see below).
Another aspect of the room we dealt with was vibration. Luckily, the optical table provided was sufficient; nevertheless, we set up a Michelson Interferometer to check for vibrations. What we found was that tapping or hitting the table caused the entire assembly to wobble for about ten seconds, so it settled down quickly. Walking and talking didn't really do much, but that was one area where we were overly cautious before and during exposures. Also, we noticed that occasionally (about every minute or so) the entire system would shift by about half a wavelength, which is enough to destroy a hologram being exposed for that long. Eventually, we traced the probable cause of the problem to a ~100 lb optical slab we had put on the same table, off to the side, resting on innertubes. When that thing started to vibrate, it kept going for a good long time; ironically, that system is supposed to be stable. The problem was that we didn't use bicycle innertubes but truck tire tubes, and we overinflated them by a ton. Oops.

We got our developer chemicals and film from one company, Integraf, who sell supplies for amateur holographers. If you want to learn how to make holograms, that's a good place to start. If you have the money, I cannot recommend plates highly enough. They're so much easier to work with, both while making the hologram and viewing them afterwards. Film tends to curl, and must be sandwiched in between pieces of glass for recording and viewing. We used three or four batches (10 pc.) of 4x5 in. celluloid film (which we typically cut into smaller squares, an advantage of film), one batch (6 pc.) of 2.5x2.5 in. glass plates, and two batches (6 pc. each) of 4x5 in. plates. The film was all PFG-01, the developer we used was JD-2, and we also ordered some Form-A-Flo (which helped).

Our mistakes that you should learn from: If you want to do any real holography, heed the following advice: first, use a safelight. The film we used is red-sensitive only, so if you can find a specially designed green bulb (they're around) that won't record on the film, use it. We worked in the pitch dark, and everything takes much longer in addition to being much more prone to errors (we, especially I, bumped things often). The amount of money you save on film is probably far more than how much a safelight bulb actually costs. Next? Do some proper tests to measure exposure times. We were never sure if they were right or not, or how close we were. I believe there's a 2-3x margin of error with the times, but too low or too high and your holograms will look bad. If you have the ability, I would suggest acquiring or building a lightmeter. It allows you to measure the beam ratios properly in addition to giving you an approximate exposure time to try. There are some instructions on the internet - I just settled for a single-chip solution from TAOS, which manufactures a variety of opto-electronic devices. The devices in question were the TSL12S and TSL13S sensors, found here. My plan was to simply measure them using a voltmeter, since they only need a positive power supply to operate. When I tried this, however, the chip lit up and fried - apparently, the voltmeter drew a little more power than the chip could handle. So I ended up having to use an op-amp to buffer the voltage, which was not a big deal. I also put both the 12S and 13S in the same project box, so I could use one for low-sensitivity and one for high-sensitivity measurements, which was convenient. All that I had to do then was figure out the formula for voltage <-> exposure time conversion, which I calculated from the fact that the ideal exposure for the PFG-01 film is 100 microJoules/cm^2 (found on some technical website). Anyway, I totally forgot the values since then so you'll have to recalculate them, no biggie.

Our lab

Us in our lab

Laser, shutter, spatial filter

Michelson Interferometer

Transmission hologram

Rainbow hologram

Our final reflection hologram

Chemical baths

Shutter controller

Video of our best transmission hologram (10 MB)