As for this particular position, the ideal candidate should have a background in any area of physics. As the job advert states, a broad scientific knowledge and training, excellent literary skills and a keen interest in the practice and communication of science are important, as are excellent communication and interpersonal skills.

To apply, please do so via our recruitment web site below. Closing date is July 9th.

http://jobs.macmillan.com/VacancyDetail.aspx?VacancyUID=000000003029

Please note that the position is based in London, and we do require a pre-existing work permit for the UK.

In case you have further questions, please feel free to contact me at any time.

The May 2013 issue of Nature Geoscience. (c) Nature Publishing Group

At Nature Geoscience, my colleagues are now trialling a double-blind peer review. In this voluntary scheme  author identities are hidden from the reviewers. Author names are removed from the manuscript, and identifying passages in the manuscript (“we have previously shown in ref. x”) are avoided. The idea is to remove any bias from the reviewers towards the authors that might potentially arise from a reviewer’s knowledge of nationality, gender, or identity of the authors. Also, reviewers sometimes do not declare an improper conflict of interest to the editors, and double-blind peer review nicely takes care of this problem, too.

An argument often heard against this scheme is that just as authors try to guess the identity of a peer reviewer – in my experience usually incorrectly – so could reviewers. But as the editors of Nature Geoscience write in their editorial:

But will they indeed be anonymous? Perhaps the most popular argument against a double-blind process is that referees will guess the authors’ identities, regardless of whether or not the names are actually sent along with the paper, because the content will give them away. We realise that this may be the case occasionally, and we will rely on participating authors to phrase their paper carefully so as to avoid easy identification. Nevertheless, we are convinced that for many papers the double-blind process will serve to remove unconscious biases.

This is not the first time that double-blind peer review has been trialled, the idea is not new. A reason why this scheme hasn’t seen wider implication yet might have been the perception that the effort required to hide author identities is not worth the potential benefits. However, in recent years the wider opinion towards improving the peer review process has changed. In a 2012 survey at Nature Geoscience, three-quarters of the respondents supported double-blind peer review, which has led to the present initiative.

There are also other experiments with peer review, and I like to highlight two here:

• In 2006, Nature trialled open peer review, where any registered user could leave comments on a participating paper under review. The uptake from these reviewers was not sufficient, and this trial ended unsuccessfully. But computer technology has changed a lot since then, and modern social network tools might improve such a scheme. The last word on this issue certainly hasn’t been spoken yet.
• In 2009 the EMBO journal started transparent peer review, where the editorial decision letters, the anonymous referee reports and the authors’ responses are all published along with the paper for increased transparency. Personally, as I argued before, I believe this is an important step that brings the peer review into the open and hopefully would make it less dogmatic but instead more interactive.

Either way, this great initiative hopefully will bring an improvement to peer review. After all, peer review is a modern invention that only in the past century became widespread practice. In that sense, it is a long-running trial of science publishing of its own, and as many scientists would agree on, could benefit from improvements.

References:
1. Editorial (2013). Double-blind peer review Nature Geoscience, 6 (6), 413-413 DOI: 10.1038/ngeo1853
2. Editorial (2012). Feedback received Nature Geoscience, 5 (9), 585-585 DOI: 10.1038/ngeo1575

Unlike optical microscopes that use light to image a sample, electron microscopes use, well, electrons. These electrons can be accelerated to very high energies to image objects on the atomic scale. In principle, it is also possible to obtain more information from these electrons than just that some atom is there in the image. As the electrons pass through a sample, they also bounce off these atoms, and lose energy in this process. How much energy is lost depends on the type of atom involved, so that by measuring the energy loss it is possible to figure out which chemical element is involved.

Chromatic aberration is an image distortion where colours are not mapped correctly on top of each other. Photo by Stan Zurek via Wikimedia.

To do the necessary energy filtering for all sorts of atoms, it is necessary to measure a broad range of energies of the scattered electrons. The problem doing this with an electron microscopes is that like in optical cameras there can be image artefacts. In particular chromatic aberration, see the figure on the right, is problematic here. Electron beams of different energies (colour) are not lined up, which distorts the image and reduces resolution. Although mapping on the atomic scale had been possible, only a limited number of elements could be resolved in that way.

A breakthrough has now come from a new generation of electron microscopes that have a built-in correction for chromatic aberration. These latest “TITAN” microscopes from the Dutch American company FEI (as used in this study) allow a larger energy scale to be sampled, so that elemental mapping has become more accurate.

In particular, Urban and colleagues applied this new technique to a material where crystals can be grown with unmatched precision: silicon. The image below of silicon atoms clearly shows the “dumbbell” arrangement of silicon atoms in pairs.

The scattering patterns observed in these images are rather complex, and studying these and comparing them to computer models will make it possible to gain more information on the chemical elements in the samples. In this case, the sample consists of only one type of atoms – silicon. It will now be interesting to see how good this approach is in resolving different elements. Still, it is amazing to see how much progress still can be made with a relatively established imaging technique such as electron microscopy, and how new sciences arises from improved measurement techniques.

Note – 11 June 2013: this blog post has been corrected to state that FEI is an American company. 

Mapping of silicon atoms in a high-resolution transmission electron microscope. The image shows a study of silicon atoms imaged for different microscope focus and sample thickness. Copyright (2013) by The American Physical Society.

Reference:
Urban, K., Mayer, J., Jinschek, J., Neish, M., Lugg, N., & Allen, L. (2013). Achromatic Elemental Mapping Beyond the Nanoscale in the Transmission Electron Microscope Physical Review Letters, 110 (18) DOI: 10.1103/PhysRevLett.110.185507

Rosalind Franklin’s x-ray diffraction image of DNA. (c) Nature Magazine. Franklin, R. & Gosling, R. G. Nature 171, 740-741 (1953) – doi:10.1038/171740a0

The way x-ray crystallography works is that a beam of x-rays is directed at a crystal, where the x-rays bounce off the atoms. Because the atoms in a crystal are periodically arranged, the x-rays form complex but regular patterns (such as the one seen for DNA). A detailed analysis of these patterns enables the precise determination of the crystal structure.

To this day such experiments aren’t easy. They require relatively large crystals and typically are done at major facilities such as electron synchrotrons. The synthesis of the crystals for these experiments can often be very difficult.

Yasuhide Inokuma, Makoto Fujita and colleagues from the University of Tokyo in Japan  and the University of Jyväskylä in Finland have now developed a clever method that does away with many limitations of x-ray crystallography. Their method works with tiny amounts of material, only about a half to 5 micrograms are enough. This is around a millionth of a gram – truly tiny. The difference between a microgram and a gram is the same as that between a gram and a metric ton. In addition, another major advance of their method is that the target molecules don’t even need to be in a crystalline state. How does this work?

A molecular framework that has absorbed absorbed target molecules (yellow) into a regular array. Reprinted by permission from Macmillan Publishers Ltd. Inokuma, Y. et al. Nature 495, 461–466 (28 March 2013). doi:10.1038/nature11990

The trick used here to avoid all cumbersome crystal synthesis is to simply embed the molecules into a larger host crystal. Metal-organic frameworks (MOF) are particularly suited for this. These are porous three-dimensional crystal structures whose pores and channels can be used to absorb other molecules. In a way, they are like sponges absorbing water. The difference to sponges is only that the MOFs are highly regular themselves. So any target molecule absorbed is held in very regular places as well.

This periodic array of target molecules within the MOF turns out to be just fine for x-ray crystallography. It doesn’t make a difference whether a perfect crystal of only target molecules is synthesised, as has been done in past experiments, or whether as done here they are held in place by embedding them into another crystal – the crystallography works just the same. It is no problem to separate the x-ray signature of the MOF from the target molecule.

This approach has a few advantages. As the authors write themselves in the paper: “Our method solves the real and intrinsic problems of X-ray crystallography and transforms it into a rapid and convenient method for the analysis of molecular structures using only a trace amount of sample.”

Despite this very nice demonstration, there still may be some issues that may still need to be solving. For once, it will be interesting to see how complex and large the measured target molecules can be. And then of course, does the arrangement of the molecules in the MOF really allow for a ultra-precise structure determination?  But such questions apart, this new method could fundamentally alter the way we do x-ray spectroscopy.

References:

1. WATSON, J., & CRICK, F. (1953). Molecular Structure of Nucleic Acids: A Structure for Deoxyribose Nucleic Acid Nature, 171 (4356), 737-738 DOI: 10.1038/171737a0

2. FRANKLIN, R., & GOSLING, R. (1953). Molecular Configuration in Sodium Thymonucleate Nature, 171 (4356), 740-741 DOI: 10.1038/171740a0

3. Inokuma, Y., Yoshioka, S., Ariyoshi, J., Arai, T., Hitora, Y., Takada, K., Matsunaga, S., Rissanen, K., & Fujita, M. (2013). X-ray analysis on the nanogram to microgram scale using porous complexes Nature, 495 (7442), 461-466 DOI: 10.1038/nature11990

I don’t want to comment on these specific cases, but focus on the broader issues here. To me these cases are a reminder that in some countries (such as Germany…) doctorate degrees mean far too much, and in some cases are sought after also because they give considerable social recognition. They are used to advance careers in completely unrelated careers, and in this way create entrance barriers based on unnecessary criteria. This is very wrong.

Too much of an academic club? Bestowing the doctorate degree at the Complutense University of Madrid. Photo via wikipedia

At its core, the doctorate is a qualification that allows to teach at universities. As a qualification for an academic career, in the context of scientific research it has its uses. Ideally, it shows that a candidate is able to conduct a research project independently, making their own independent scientific contribution to a research field.

However, beyond the world of academia it just doesn’t make sense to put too much emphasis on a doctorate. In Germany it is even common practice to include the doctorate in the passport, as part of the name. Why?

Sure, getting a doctorate is a great accomplishment that usually follows after several years of hard work and a lot of sweat and tears. But people in other careers also work hard. So why the desire for going through a doctorate without the intention of ever working in research? For example, neither Schavan nor Guttenberg, as far as I am aware, held any notable academic position at university after they obtained their PhDs. Indeed, Guttenberg then already was a member of parliament.

Sure, people always change their mind and leave academia, given the difficulty of pursuing a career for young researchers. However,obtaining a doctorate to advance a career outside research (whether in academia or industrial research) to my mind is not only unnecessary, but also creates a glass ceiling for those without. But why should a doctorate make any difference to someone’s skills as a politician?

The sooner we limit the career benefits of a PhD to academia and related careers, the better. It will remove unnecessary entrance barriers to other professions, and will allow people to focus on the actual qualifications needed for their careers.

Then, on January 16 the FAA grounded all 787, following a number of technical problems. Earlier, the Japanese airlines ANA and JAL had already suspended all 787 flights, which was a significant signal because combined these two airlines operate almost half of the 787 delivered to date. Aside from a number of other technical issues such as a fuel leak, a key reason to ground the entire fleet has been two incidences where the back-up batteries overcharged and overheated such that there was the danger of fire on board. But how big a deal are these battery incidents?

A Boeing 787 Dreamliner. Photo by Spaceaero2 via Wikimedia.

The batteries used in the plane are common ones, using lithium cobalt oxide. They are rechargeable batteries, where the lithium ions are used for charge storage and transportation. Lithium-ion batteries are widely used for all sorts of electronic devices such as laptops or mobile phones. In principle, this type of batteries is safe, although the 787 is the first large aircraft to heavily rely on such batteries.

Although the battery technology is well-known, designing an aircraft is complex, and if not done properly can lead to unexpected situations where the batteries are operated outside their specifications. In the case of the Dreamliner, overcharging the batteries has been suspected to be the cause of the plane’s battery problem.

That overcharged batteries are a hazard should not come as that much a surprise. In a lithium ion battery, the lithium atoms need to move through the crystal. Of course, this is no different from electrons moving through an electrical cable when you turn on the lights. But even though lithium is only the third element in the periodic table these ions are a lot larger than electrons, so that this movement is not that easy. To stay vaguely on topic, it is the same as squeezing an oversized passenger into a tiny air plane seat. It creates friction, and in case of the batteries this is the cause of heating – in particularly when charging the batteries, or when discharging them too fast.

For gadgets such as phones there will never be that much of a problem, the maximum power the batteries need to supply to the gadget is in no comparison to the power an aircraft needs for its back-up systems. Moreover, trying to push the lithium ions fast through the battery is far more problematic than taking it slow. That’s why the lithium-ion batteries on the Dreamliner are made of lithium cobalt oxide, which is more suited to heavy loads than lithium iron phosphate and other compounds used in consumer devices. The drawback is only that overcharging lithium cobalt oxide is more of an issue… as we have seen.

On the other hand, lithium ion batteries have been successfully used in demanding applications, for example to power electric sports cars such as the Tesla Roadster. Indeed, Elon Musk, a co-founder of Tesla Motors, has offered their help to Boeing in a tweet:

Maybe already under control, but Tesla & SpaceX are happy to help with the 787 lithium ion batteries.—
Elon Musk (@elonmusk) January 18, 2013

To me, the problem also looks solvable. If indeed overcharging is the issue here, this might even point more towards a design flaw in the electrical system of the plane, rather than the battery design itself. But this is pure speculation on my part.

The bigger issue in the long-term might be how much rechargeable batteries might be able to replace crucial systems on air planes, to achieve more fuel-efficient aircraft designs. Storing a huge amount of energy in the confined space of a battery will always create safety issues, in particularly if the charging and release of the energy need to occur at very short time scales. There might always be a delicate balance between operational demands and safety, and further research in new batteries is crucially needed. But we should not forget that regardless of the use of such batteries, passenger jets always carry around materials storing high levels of energy. By this I mean the jet fuel. What happens in the jet engines is not much different nothing but a controlled explosion and burning of that fuel.

For my part, I am confident that Boeing’s engineers will eventually find a solution. Fuel-efficient, light-weight aircraft designs are urgently needed and planes such as the 787 do point towards the future in passenger transportation. For this reason, the lessons learned here for the battery systems will be very valuable in future.

Update – 20 January 2013: The US National Transportation Safety Board has now ruled out excessive voltage applied to the batteries as a cause for their failure.

It comes as no surprise that physicists are searching for more precise ways to measure weight, and the method now published in Science by Holger Müller and colleagues from Berkeley is one of the most elegant and beautiful ones that I have seen in a long time. It is based on a quantity that we know very well how to measure with very high precision – time. The question is how to measure the mass of something by telling the time.

Photo by bitzcelt via flickr.

This is where a concept from quantum physics comes into play, where any object can also be described as a wave. In mathematical terms it has a wavefunction. And one of the characteristic features of any wave is that it oscillates, like the ripples in a pond. For the wavefunctions of an object with mass, the heavier it is, the faster its wavefunction oscillates. And this gives us the direct link between mass and time. Measuring the oscillation speed of an object’s wavefunction gives us its mass. The only problem is that even for something tiny as an atom these waves oscillate far too fast to be measured by any technology available.

Müller and colleagues now used an ingenious trick to get some information on the wavefunction of an atom. They take a gas of atoms, caesium 133, and irradiate it with a laser beam. The beam intensity and pulses are chosen such that about half of the atoms of the gas are kicked out of their resting position and move away from the original gas. A little while later, another laser pulse pushes them back again towards their starting point, so that they meet the other half of atoms in rest again. This causes an interference of the waves similar to the wave patterns that form in a pond if you throw two stones into the water at the same time.

In case of the atom gas, the interference patterns also tell the difference between the wavefunctions of the moving atoms and those at rest. Einstein’s theory of relativity states that a particle that is moving has a higher mass than one that isn’t. In other words, the moving atoms have a different wave property than the ones at rest, their waves oscillate faster. Because the difference in speed between the atoms is not really that large, the difference between their wavefunction oscillations is small enough so that it can be measured experimentally. When measuring mass via time it is easier to measure the difference in weight than the actual weight itself.

The actual measurement of the interference patterns between the atoms in the gas is done with precisely calibrated lasers, using a technique called frequency comb. The precision achieved in measuring the mass of the caesium atoms already is ten times better than the existing mass standard, and comparable to some of the best alternative efforts to measure mass. Certainly, defining a new standard for mass still requires a lot more research, and it also isn’t that straightforward from measuring the mass of a single atom to defining a new standard for a macroscopic unit such as the kilogram. Regardless, this clever experiment represents a truly beautiful way of linking mass and time measurements.

Reference:
Lan, S., Kuan, P., Estey, B., English, D., Brown, J., Hohensee, M., & Muller, H. (2013). A Clock Directly Linking Time to a Particle’s Mass Science DOI: 10.1126/science.1230767

Key here is to understand what temperature actually means. It isn’t simply a uniform speed of atoms in a gas. Atoms at any temperature move randomly and at various speeds. So temperature is better defined by the energy distribution of atoms at that particular temperature. In normal systems with positive energies, it is very unlikely to find atoms at high energies. Indeed, the energy distribution, the likelihood of finding atoms at increasingly higher energies than the average, is decreasing exponentially. This is expected because atoms at very high energies quickly cool down by collisions with others.

Photo by tlindenbaum via flickr.

Not so for systems with negative absolute temperature. There, looking at the physical formulas negative temperatures mean that the energy distribution of the atoms is reversed. Close to absolute zero the atoms prefer to be at high energy states such that there are more atoms moving at higher speeds than at lower speeds. So negative temperatures work in the opposite way to positive ones – higher energies are preferred to lower ones. In that sense, a system with a negative absolute temperature is ‘hotter’ than systems with positive temperatures. If brought together, energy would flow from the negative to the positive system, as all those many atoms at high energies in the negative system would give their energy to the slow ones in the positive system.

It comes as no surprise that realizing systems with negative absolute temperatures is very challenging, as there the energy distribution is against the long-term tendency found in nature for atoms and other systems to relax to lower energies over time. But it is not impossible. In the past, negative temperatures have been realized with magnetic spins, which are easier to control by magnetic fields and where such reversed energy distributions can be better achieved.

In the case of atomic movements of gases the situation is more difficult. The system studied now by Schneider and colleagues is an ultracold atomic gas, which is made of atoms at very low temperatures. In these atomic gases not only the interaction between the individual atoms can be tuned experimentally, but also the energetic barrier that separates the atoms from each other is adjustable.

At normal positive temperatures, the atoms repel each other, and this reduces their energy, just as an expanding gas would cool down by itself. To realize negative temperatures, the researchers then swapped the conditions in the atomic gas around by using magnetic fields and adjusting the laser beams that holds the atoms together. Now, the interaction between the atoms was set such that the atoms attract each other. This attractive force between the atoms increases their energy. However, to avoid the atoms smashing against each other, which again would bring their energy distribution down to that for positive temperatures, the energy barrier between the atoms has to be set to keep them separate from each other. In this way the energy distribution between the atoms is stabilized. More atoms are in a higher energy state than a lower one – negative absolute temperatures are achieved.

Of course, this realisation of negative absolute temperatures doesn’t mean that absolute zero can be reached. It is a completely different challenge to slow down atomic motions towards zero than it is to change the overall energy distribution of the ensemble. Still, these are intriguing thermodynamical systems, and there is plenty to study about the implications of such reversed energy distributions.

Reference:

Braun, S., Ronzheimer, J., Schreiber, M., Hodgman, S., Rom, T., Bloch, I., & Schneider, U. (2013). Negative Absolute Temperature for Motional Degrees of Freedom Science, 339 (6115), 52-55 DOI: 10.1126/science.1227831

Topological insulators are electrically insulating in their interior, but at the surface they do conduct current. Moreover, the surface currents are topologically protected (hence the name), which means that the electrons that carry those currents don’t veer off the track easily and maintain their properties over long distance. Although a number of topological insulator compounds are known, the problem so far has been that it has been difficult to fabricate these with sufficient purity such that the interior was indeed insulating. This has been a problem, as the electrical current inside the materials just overwhelms the surface properties.

Samples of samarium hexaboride crystals. (c) Johnpierre Paglione, reprinted by permission from Macmillan Publishers Ltd. Nature 492, 165 (2012) doi:10.1038/492165a.

Not so in samarium hexaboride. A number of papers put on the arXiv preprint server by Zachary Fisk and colleagues from the University of California at Irvine and the University of Maryland have now shown that this material could have all the desired properties of a topological insulators. It is insulating in the interior, but on the surface it shows very good conductivity. Indeed, at low temperatures the crystals, several millimetres in size,  show very high surface mobilities of 72,000 cm²/Vs.

So far, however, many of the actual properties that make topological insulators so unique haven’t been demonstrated yet. And of course, none of the papers has appears to have been published in a peer-reviewed journal yet.  But they certainly have made headlines already. The first of the three arxiv papers referenced below was placed on arXiv on November 21st. On the 24th Ross McKenzie wrote about it in Condensed Concepts, which is from where I heard first about these new results. And then today Eugenie Samuel Reich wrote a story about them for Nature News & Comment. I am sure at the upcoming APS March Meeting 2013 in Baltimore these materials will feature prominently, too. For sure we haven’t heard the last of samarium hexaboride yet.

References:
1. Steven Wolgast, Cagliyan Kurdak, Kai Sun, J. W. Allen, Dae-Jeong Kim, & Zachary Fisk (2012). Discovery of the First Topological Kondo Insulator: Samarium Hexaboride - arXiv: 1211.5104v2
2. Xiaohang Zhang, N. P. Butch, P. Syers, S. Ziemak, Richard L. Greene, & J. Paglione (2012). Hybridization, Correlation, and In-Gap States in the Kondo Insulator SmB6 - arXiv: 1211.5532v1
3. J. Botimer, D. J. Kim, S. Thomas, T. Grant, Z. Fisk, & Jing Xia (2012). Robust Surface Hall Effect and Nonlocal Transport in SmB6: Indication for an Ideal Topological Insulator - arXiv: 1211.6769v1

Plasmons in graphene can be created by illuminating the tip of an atomic force microscope (grey) with an infrared laser beam (red). Reprinted by permission from Macmillan Publishers Ltd. Fei Z. et al. Nature 487, 82–85 (2012). doi:10.1038/nature11253

But absorption of light is not where the true potential of graphene lies, namely on the nanoscale. On the same scale as electronic applications, because ultimately the aim is to achieve photonic functionality on a chip.

However, the control of light on the nanoscale typically requires surface plasmons. These are collective movements of electrons at the surface of metals. So in a sense surface plasmons function a bit like antenna that can focus light into tiny spots.

Graphene itself is not a metal, or at least not in the same way as the noble metals gold or silver that are usually used for plasmonic applications. Still, graphene’s electrical conductivity is not too bad and its electrons can move pretty fast. For this reason, graphene does also show plasmonic effects. This is good news, because in comparison to gold or silver, graphene has a significant advantage: its plasmonic properties can be easily influenced by external means such as an electrical voltage, which either adds or removes the free electrical charges needed for plasmonics.

Already, good plasmonic properties have been demonstrated in graphene, with plasmon propagation distances of several micrometers (see figure). This suggests that graphene could in principle be useful in guiding light on a chip. To me, however, the potential of graphene is not necessarily that it can transmit light. The promise lies perhaps more in the fact that plasmonic effects can be controlled. And all this based on a single atomic layer only.

For example, it is well known that graphene’s electrical properties are quite sensitive to its environment. Whether it is an applied electric voltage or other factors such as molecules that are bound to graphene’s surface and that change the electronic properties this way. Combined with a strong sensitivity of plasmonic effects this would make for very efficient detectors.

And then of course there are all the ongoing attempts that make use of graphene’s electronic properties. Optical devices could benefit a lot from coupling such electronic effects with optical ones. The recent demonstration of electrical switching of terahertz radiation using graphene devices is just a first step amongst a number of very promising strategies investigated, many of which have been reviewed nicely in this month’s issue of Nature Photonics. So watch out for new plasmonic applications using graphene.

References:

Nair, R., Blake, P., Grigorenko, A., Novoselov, K., Booth, T., Stauber, T., Peres, N., & Geim, A. (2008). Fine Structure Constant Defines Visual Transparency of Graphene Science, 320 (5881), 1308-1308 DOI: 10.1126/science.1156965

Fei, Z., Rodin, A., Andreev, G., Bao, W., McLeod, A., Wagner, M., Zhang, L., Zhao, Z., Thiemens, M., Dominguez, G., Fogler, M., Neto, A., Lau, C., Keilmann, F., & Basov, D. (2012). Gate-tuning of graphene plasmons revealed by infrared nano-imaging Nature DOI: 10.1038/nature11253

Lee, S., Choi, M., Kim, T., Lee, S., Liu, M., Yin, X., Choi, H., Lee, S., Choi, C., Choi, S., Zhang, X., & Min, B. (2012). Switching terahertz waves with gate-controlled active graphene metamaterials Nature Materials, 11 (11), 936-941 DOI: 10.1038/nmat3433

Grigorenko, A., Polini, M., & Novoselov, K. (2012). Graphene plasmonics Nature Photonics, 6 (11), 749-758 DOI: 10.1038/nphoton.2012.262