A paper in Nature and a rumoured but still statistically insignificant result prompt me to write something today about what it’s like to be on the front line of science.
The usual public perception of science is that we provide definitive results that are the unassailable truth. The reality is rather more complex. Firstly, there isn’t really any such thing as truth – well established models of the way the world works can always be challenged when you test their predictions with experiments that try things that have never been tried before. Einstein’s theory of gravity, with the inverse square law, is hugely accurate in describing most of the things that involve gravity. NASA engineers can navigate around the solar system using Newton’s laws to slingshot spacecraft around planets with very impressive accuracy. But start to do things that approach the speed of light, and things go wrong. That’s when you need to replace Newton’s inverse square law with Einstein’s general relativity. It doesn’t mean Newton was wrong, just incomplete.
This means that every observer or experimentalist, like myself, is always wanting to look places nobody has looked before, or examine little niggling issues that might be nothing, or that might, with further work, blow open an entirely new area of physics. This week we saw one of these niggles close, and, just maybe, another one open up.
The issue that seems to have been solved is the FE XVII problem (or, to translate from astronomer, the Iron 17 problem). This relates to the X-ray spectrum of 16-times ionised Iron, Fe 16+ to a chemist of Fe XVII to an astronomer. The strength of this line as observed in astrophysical sources was always too weak. This resulted in various ideas about the environments in which the line was found, and other, more exotic possible solutions. The more prosaic solution, that we didn’t have the basic parameters of this particular ionisation state of Iron, was also always a possibility, but there was no way to test this. Until now. A paper in Nature this week announces the solution to the FeXVII problem. Bernitt et al. used a free electron X-ray laser to excite FeXVI ions with ultrashort intense pulses. This eliminated many of the complicating factors found in astrophysical factors and allowed the quantum mechanical oscillator strengths to be directly measured.
And lo and behold, it turns out the oscillator strength is surprisingly low! This explains the astrophysical problem, and means that the source of the problem is likely in the approximations made in the calculation of the basic parameters. So nothing to shake the foundations of physics or astrophysics, and, interestingly enough, a solution that could not have been achieved observationally using any astrophysical methods – big fusion or laser labs with free electron lasers were necessary.
Meanwhile, at the LHC, a hint of what might be new physics. This comes from preliminary results from the ATLAS experiment at CERN, discussed in a blog post from Scientific American. This relates to the Higgs boson, whose discovery by both the ATLAS and CMS experiments was announced back in June. One of the interesting points about that discovery was that both experiments had found that one of the two decay modes used in finding the Higgs, the one that involved two photons rather than the other involving two Z particles, seemed to be occurring more frequently than predicted.
ATLAS now have enough data to calculate the Higgs mass for each of these decay modes separately. Their plot is shown here:
The latest results from the Atlas experiment indicate that there may be two different Higgs bosons—one that weighs 123.5 GeV (in blue) and another that’s 126.6 GeV (in red).
The two decay modes suggest (slightly) different masses for the Higgs particles involved in the different decay modes. While there are some versions of the standard model of particle physics that can have several different Higgs bosons, they would not have masses as close as is hinted at by this data, so this could be a sign of new physics.
Of course there are many other possible explanations, including random fluctuations that will go away – the difference is at best only 2 sigma -, slight differences in calibration of the detectors, in the way the data is treated etc. etc. As usual in these things more data is needed, and this will hopefully arrive in March. The persistent disagreement between model and defy rate for the two photon mode, though, is interesting, as that has not, as originally expected, gone away, though it is still a less than 3 sigma effect.
The dog that isn’t barking here is the other general purpose detector experiment at CERN, CMS. They haven’t made any announcements on these issues yet, and, if I understand it correctly, won’t be saying anything until March.
So there you are – a peek inside the box of science, where small problems lurk until they’re squashed, but others emerge all the time.
It all gives me hope for future job security.