Everyday Scientist

Global Climate and Energy Project

The latest Global Climate and Energy Project (GCEP) brochure recently came out.  It’s a $225 million Stanford Program, funded primarily by General Electric, Toyota, ExxonMobil, and Schlumberger, whose purpose is to fund environmentally-motivated research.  This involves everything from carbon dioxide sequestration to alternative fuel synthesis to hydrogen storage to fuel cell design.  It’s also what funds my research in part.

By the way…check out page 26 (28 in the pdf).

Source 

bubbling nitrogen

The extent of chemistry chemistry that I do is making solutions of fluorescent dyes. But once in a while, I get to bubble N₂ through a solution or something like that. I’m checking to see if this solution will bleach when de-oxygenated. But look how cool I am:

Isn’t that so cool? Here’s what my whole operation looks like (note the awesome poster in our lab):

lasers treating me a little better today

Although the OPO is still not working after several more attempts at re-aligning (and the Millenia 532-nm laser across the hall is out for repair/replacement) I have some good news in the laser department. I got (at least for the time being) a 532-nm source: a little 12-mW diode laser that another lab member used to use. I made a neat stand for it out of a bunch of random aluminum pieces I found around lab (with some minor drilling required). Now I have two colors directly on my table, look:

Now I can switch between 488 nm and 532 nm when I look at different samples, simply by flipping a little mirror:

The best part is that it only took me 20 minutes to align into the fiber, which has taken me days in the past. My trick is to first make sure the spot goes evenly through the coupler (by taking off the fiber), then aligning the beam path using a multimode fiber. That’s pretty easy, but it gives you a coarse alignment. Then I replace the multimode with my fiber and maximize the output. Today, it worked like a charm.

Now here are my lasers:

In this picture, the flippy mirror is up, so 532 goes into the fiber coupler and the 488 beam scatters off the back of the flippy. With a flip of that mirror, the 488 goes perfectly into the fiber and will only require minor retweaking downstream! The best part is that both prefiber alignments are orthogonal, so I don’t have to worry about screwing up one color if I want to mess with the other (as long as I don’t touch the coupler or the mirror immediately before the coupler, of course). I was pretty thrilled with this, especially because all the rest of the lasers I’m trying to use are the suck right now.

caution: annoying lasers

Man, I had bad luck with lasers on Friday. First, the guy from Spectra Physics showed up two hours late then to check out our Millenia (532 nm, 5 W diode) laser. It sounds like the power supply ($1200) is dead; also, one or both of the diodes ($11000 each) might need replacing. We asked him about trade-in possibilities for a 532-nm, 250 mW laser.

Then Jim, Frank, and I tried to get the OPO working. We re-aligned it three times to no avail. Here’s a pic of the setup:

The Ar⁺ pumps the Ti-Sapph, which pumps the OPO. We can get the Ar⁺ to be pretty stable after several hours of tweaking, and the Ti-Sapph stays mode-locked pretty well. But the OPO won’t … well … it won’t OPO, I guess. We got to open up the OPO, which was pretty neat:

Actually, it’s kinda fun tweaking all these scary knobs and mirrors and crystals. But it would be a lot funner if it worked after all that tweaking. Oh well, there’s always Monday…

I just measured F_opt for my ‘scope

So here’s the deal: the detection efficiency for the stream of photons coming from one single fluorophore is kinda crappy (~10%). But that’s the state of the art; it’s part of the reason single-molecule measurements are so tough … and fun!

The detection efficiency D is:

D = n_QF_collF_optF_filter

where n_Q is the camera quantum efficiency, F_coll is the percetage of light collected from the objective, F_opt is the efficiency through all the optics in the microscope, and F_filter is the transmission through the various filters. Today, I measured F_opt to be 50% at 488 nm. Here’s a page from my notebook that shows my setup:

orbecular candy?

Doesn’t this look like some kind of molecular lolly-pop material candy?

So I don’t really get any interesting info from this result, but Gaussian can really make some pretty pictures … that take a long time to make.

This is DCDHF-2V optimized using BLYP/6-31G(d). I’m displaying the HOMO orbitals.

scisearch

I just found that SciSearch v2.4 has a cool feature that you can plot the number of citations of someone’s papers. Here’s an example of W.E.’s plot of number of citations versus publication year:

Cool, eh? Go play with it and try your favorite scientist.

h value

I presented a paper at my Chemical Physics Journal Club a while back. Here’s a citation: Hirsch, J. E. An index to quantify an individual’s scientific research output. Proc. Nat. Acad. Sci. 2005, 102(46), 16569–16572.

The basic idea is that h measures your (career) scientific output, the higher the value, the greater the output (convolved with impact). h is the number of papers that you have authored, each with h or more citations. In other words, if you have 50 papers, but only 30 of them have been cited 30 or more times, your h = 30. This method tries to avoid giving a lot of credit to people who write oft-cited review articles or people who write millions of papers that no one ever reads.

It’s mostly sorta bullshit, but I enjoyed calculating the “h value” (or “h index” or “h number” or “Hirsch number” or…) of several of the profs at Stanford:

But there are some things that are unfair: Pande is a younger scientist, so he hasn’t had as much time to publish or have his works cited as, say, Zare or Fayer. So we can normalize to the number of years since earning PhD (where m = h/#years):

Comparing m values is probably a little more fair. But now there are some more problems. Pande is in his steep period, which will presumably level off like this:

So you get your PhD, then you die. The value of m also doesn’t take into account that Andersen is a theorist or that W.E. spent until the mid-90s in industry, where people normally don’t publish as much. Oh well, it’s still pretty fun.

By the way, h_sam = 1

UPDATE: Check out my new measure, the Lord h-bar index.