Blue Collar Scientist

Here’s a pretty clever scientific technique. Let’s say you’ve got a source of amber - lots of amber - and you want to check it for fossils. The problem, though - the amber is impure, or it has nonbiological inclusions, or it is badly scratched up - whatever the reason, you can’t see through it and can’t tell what is inside. It looks like this:

Researchers have come up with a pretty cool technique to get around this problem. They are x-raying it instead.

This may sound pretty obvious, but it isn’t. The surface texture of amber scatters x-rays like mad, and this makes it very difficult to get an image of what is inside. The researchers came up with a clever way around this. The density of amber ranges from 1.05 to 1.15 grams per cubic centimeter. That’s really close to water, at 0.998 grams per cubic centimeter. Since scattering occurs preferentially when electromagnetic radiation encounters a density variation, the researchers simply put their amber samples into water. The water flows into the scratches and fills them up, making the surface of the amber look a lot smoother to the x-rays than it otherwise would.

The researchers used a very specific imaging technique, with the winsome name propagation phase contrast microradiography. Someday, I hope to understand exactly how that works in x-rays, but I know about phase contrast micrography in optical wavelengths, and that’s a tangent I think is worth exploring. To do this in visible-light microscopes, you start with partially coherent light, and you shine it through a ring, and then focus it on your specimen with a condenser lens. Once it has gone through the specimen, it goes through the objective, and then passes through a ‘negative’ ring - instead of a piece of material that has a donut-shape cut out of it to let light through, this piece of material is a donut-shape that blocks light (but allows light through the donut hole as well as though the outside of the donut). Once the light passes through this apparatus, you focus it into an image, and you get a really weird effect. Diffraction converts small changes in the phase of the light into large differences in amplitude - or brightness. The result is that stuff you have a real hard time seeing any other way leaps out at you. It is such an impressive technique that its inventor, Frits Zernike, won the Nobel Prize for coming up with it. Nikon has an excellent discussion of the optical technique.

Presumably, the x-ray apparatus works on similar principles. In any case, they’ve come up with this:

The top image is a standard x-ray, while the bottom image utilizes the phase contrast technique. Amazing.

The scientists imaged 640 pieces of amber from the Charentes region in southwestern France. They discovered 356 fossil animals, going from wasps and flies, to ants or even spiders and acarians.

Acarians, by the way, are mites and ticks, and one of them that was found in the amber was only 0.8 millimeters in size. Quite impressive. And I really recommend you click over to the press release, which has lots of pictures well worth seeing. The researchers report they’ve been able to determine the families of 53% of the inclusions, in some cases from 3-dimensional images. I don’t know how that compares to optical techniques on transparent amber, but it is certainly good enough to be useful. The method could potentially shed a lot of light on the evolution of small arthropods.

There are two really cool things about this. The first is that most amber is opaque - 80% of it at Charentes, where the researchers took their samples. The second is that there’s probably lots of opaque amber sitting around in museums that has been kept because it is interesting for some reason or other. This amber can now be studied.

Want the original paper?

M. Lak, D. Néraudeau, A. Nel, P. Cloetens, V. Perrichot and P. Tafforeau, Phase contrast X-ray synchrotron imaging: opening access to fossil inclusions in opaque amber, Microscopy and Microanalysis, Forthcoming article, doi: 10.1017/S1431927608080264

For years, astronomers have tried to catch a supernova in the earliest stages of its explosion. They’ve gone to the extremes of building highly automated telescopes to search hundreds of galaxies each night for new supernova, often revisiting the same galaxy several times per night just so that if one did happen to be going off, they could catch it several times early in the process. Researchers in the highly competitive field have even collaborated with each other to insure they are getting most efficient coverage possible.

Catching a supernova early on in its detonation is of interest to astronomers for two reasons. The first reason is that certain supernova are useful standard candles for calibrating intergalactic distances. The better the measurement of supernova’s brightness over time, the better the distance measurement will be. So there’s an incentive to observe a supernova regularly over as much of its lifetime as possible. The other reason is that the very earliest stages of the explosion should tell us interesting things about how supernovae explode. We’ve got the broad strokes of supernova formation pretty well figured out, but watching a supernova in the early stages will allow us to figure out the details.

Astronomers studying supernovae have had some success with their strategies. They’ve caught quite a few supernovae on their rise to brightness, but they’ve never really achieved their holy grail - observing a supernova from the earliest stages of its detonation.

Back on January 9, that all changed. On that day, I got an e-mail from the ATEL service reporting that the Swift satellite had discovered a bright X-ray transient while it was observing a two-week old supernova in the galaxy NGC 2770:

Follow-up observations of SN2007uy with the Swift/XRT reveal a new transient source about 95 arcsec away from the SN position at RA=09:09:30.7, Dec=+33:08:19 (J2000). This position coincides with the outskirts of the host galaxy of SN2007uy (NGC 2770).

This sort of thing is fairly routine. I get ATEL messages, and alerts from other services, primarily because some of the astronomers who run my software use those services to report and find targets of opportunity. It is nice to know what my clients and colleagues are up to - makes me feel like I’m still in the loop, even though I’m not doing the day-to-day thing at an observatory anymore.

After the initial report, several other astronomers chimed in, some with negative observations, and the early thinking was that the object was a flare from a soft gamma-ray repeater, or a weak X-ray flash - two classes of phenomena that are pretty interesting in their own right, but nothing that would keep me up nights.

But within a few hours, an optical counterpart had been discovered by J. Deng and Y. Zhu at the National Astronomical Observatory of China:

We took a 60s unfiltered CCD image and a 600s R-band image of NGC 2770 on 2008 Jan 10 18:19 UT and 18:22 UT with the BFOSC [Beijing Faint Object Spectrograph and Camera] instrument of the NAOC 2.16m telescope at the Xinglong Observatory, China. An optical source was clearly seen in both images which was conincident in position with the X-ray transient reported in GCN 7159. It was roughly 2 mag fainter than SN 2007uy in the same galaxy….. We suspect that this source is the SN counterpart of the X-ray transient (a XRF). Spectroscopic observations are in progress despite the cloudy weather condition.

Just fifteen hours later, Stefano Valenti reported:

Following the report of J. Deng and Y. Zhu (GCN7160) that the transient was roughly 2 magnitudes fainter than SN2007uy on Jan. 10.76, in R band, it seems that the luminosity of the transient is rising rapidly, supporting the very fortunate possibility that Swift has observed, in real time, the explosion of a massive star.

This was exciting - electrifying - because in the past, nothing like this had ever been done. For forty years, astronomers have theorized that when a supernova goes off, you should first get a burst of neutrinos, then some gravitational waves, and then some X-rays, all within seconds or minutes of each other. A couple days after that, you’ll start to get the visible light, so the best that all the astronomers using optical telescopes to hunt for supernova could possibly accomplish was to discover something that had exploded a day or two before.

It looked pretty good for this being the first discovery of a supernova at the very moment of the explosion. One of the things that astronomers were looking for was something called “shock breakout radiation” - a big burst of x-rays and ultraviolet light that results from the explosion’s shock wave emerging through the stellar surface. Shock breakout radiation had not, to the best of my knowledge, ever been observed before. Then several astronomers reported on ATEL that:

the optical transient was clearly detected in the U, V, and B bands of the UVOT. The source was marginally seen in the UVW1 image. The very rough B and V magnitude is 18.8 and 18.4, respectively. The Swift observations indicate that the optical transient becomes brighter in the past 24 hours. We note that the optical transient was not seen in the UVOT observations taken on 2008 Jan 9.

The “U” band referred to is ultraviolet light, so this was a good sign - they could not see the supernova at nearly the moment of detonation, but they could see it, and measured the brightness of the UV radiation from it, two days later. That’s almost exactly what theory predicted they should be able to do!

Over the next few days, a flood of messages about this newly-discovered object came down the pike, and after a while, the excitement died down. Almost a month ago, the observers submitted a paper to Nature, and the preprint appeared on the arXiv server. Everyone has been very responsible about not discussing it due to the embargo while the paper undergoes peer review, but now Science News has published a story, and given my small readership, I don’t see the point of being quiet about it any more.

The data suggests that this was a Type Ibc supernova discovered at the moment of detonation, that the progenitor was a Wolf-Rayet star, that shock breakout radiation was observed, that radio, optical, and long-lasting x-ray counterparts were observed, that fast ejecta was observed, and that all the observed phenomena fit very nicely into a model of how supernova explode.

Furthermore, the observations are inconsistent with one hypothesis about how supernova generate their radiation. The “relativistic outflow” model proposes that a lot of supernova emission is the result of particles that are accelerated to a substantial fraction of the speed of light, something that did not happen in this case.

And finally, it looks as though a wide-angle X-ray instrument similar to the one the Swift astronomers were using would allow scientists to detect several hundred core-collapse (Type Ib, Ic, and Type II) supernovae each year at the instant of explosion. This would not only allow the model to be tested with a lot of different supernovae, but scientists using neutrino and gravitational wave detectors might benefit from having some constraints on the position and time of the events these sensors are designed to detect.

The bottom line is, these results are very much what science is all about. Some of the hypotheses about supernovae made predictions that didn’t stand up to the observations of SN2008D very well at all, but others fit the observations just about perfectly. It opened up new lines of inquiry and research, and suggested some things that astronomers could do to do even better science in the future. And last but not least, it showed where we still have some things to learn about supernova that aren’t fully explained yet.

It is a good time to be an astronomer!