The experiment described in the previous post was published in early 1998, but the work was done in 1997. This was the year when things really turned around for me in grad school– the optical control paper was done in the summer 0f ’94, and ’95 and ’96 were just a carnival of pain. Everything in the lab broke, was repaired, and then broke again.
The lead guy on the lattice collision experiment was a post-doc named John Lawall, who really took charge. He completely re-vamped the lock system for the laser, and spent a huge amount of time re-doing the laser alignment for the trap and the optical lattice. I learned a ton of stuff from working with him, and that really got things back on track in the lab. In mid-1997, everything started working, and continued working well for the next few years, and three more papers.
I don’t have many good stories about the lattice collision experiment itself– mostly what I remember is that it was a gigantic pain in the ass to keep the lattice laser tweaked up well enough to do the experiment. It was a Coherent 899 Ti:Sapph laser, and we had to strip it down completely once a week to get the power we needed. On the bright side, I got good at tweaking that laser up, but I was much happier to go back to experiments that were less demanding.
The best story I have regarding this work concerns the topic Steinn was talking about the other day: being scooped. This is one of three papers I was on where very similar work was done by Prof. Fujio Shimizu’s group in Japan. With the optical control paper, we arranged to send them in at about the same time, but in this case, they went to press well before we did.
Sadly, this wasn’t the sort of cutthroat kind of business that would make for a really colorful anecdote– everything was very collegial, possibly because none of the people involved had any real pressure. We knew they were working on similar problems– the cold-collision community was never very large– and had a good relationship with them. When they got results in their lattice experiment, they sent us word, including a copy of the key graph, and offered to hold off publishing a little while so we could get something ready to go.
At the time, we had only just gotten the first real signals out of the apparatus, and while the preliminary data looked good, there were still some puzzling aspects. We scrambled around a bit to see if we could get something we’d be willing to go to print with in a hurry, but we couldn’t quite make sense of what we were seeing, so we told them to go ahead and publish what they had.
John went back to first principles with the whole business, and really nailed down the dependence on a lot of factors. It turned out that the lattice alignment and stability were really critical, and once we got that sorted out, everything else fell into place. In the end, we ended up with a much stronger paper than we would’ve had earlier (arguably a stronger paper than Shimizu’s), and got it into Physical Review Letters despite having been “scooped,” so everything ends happily.
My role in this one was a little larger than for the first paper, but not much. I helped some with the data collection, but mostly did computer data analysis and made figures. I also participated in the ever-popular “paper torture” for the article, but this was primarily John’s paper, and he deserves the lion’s share of the credit.
The next two papers in the series, though, were primarily my stuff. I picked up some momentum from this work, and managed to carry it through to two articles that I’m really proud of, and that make much better stories…