The title of this post is taken from an (in)famous statement from Wolfgang Pauli:
God made solids, but surfaces were the work of the devil!
That diabolical nature of surfaces is, however, exactly what makes them so intriguing, so fascinating, and so rich in physics and chemistry. And it’s also why surface science plays such an integral and ubiquitous role in so many areas of condensed matter physics and nanoscience. That ubiquity is reflected in the name of a UK summer school for PhD students, nanoscience@Surfaces 2018, held at the famed Cavendish Laboratory at Cambridge last week, and at which I had the immense pleasure of speaking. More on that soon. Let’s first dig below the surface of surfaces just a little.
(In passing, it would be remiss of me not to note that the Cavendish houses a treasure trove of classic experimental “kit” and apparatus that underpinned many of the greatest discoveries in physics and chemistry. Make sure that you venture upstairs if you ever visit the lab. (Thanks for the advice to do just that, Giovanni!))
Although I could classify myself, in terms of research background, as a nanoscientist, a chemical physicist, or (whisper it) even a physical chemist at times, my first allegiance is, and always will be, with surface science. I’m fundamentally a surface scientist. For one thing, the title of my PhD thesis (from, gulp, 1994) nails my colours to the mast: A Scanning Tunnelling Microscopy Investigation of the Interaction of Sulphur with Semiconductor Surfaces. 
(There. I said it. For quite some time, surface science was targetted by the Engineering and Physical Sciences Research Council (EPSRC) as an area of funding whose slice of the public purse should be reduced, so not only was it unfashionable to admit to being a surface scientist, it could be downright damaging to one’s career. Thankfully we live in slightly more enlightened times. For now.)
Pauli’s damning indictment of surfaces stems fundamentally from the broken symmetry that the truncation of a solid represents. In the bulk, each atom is happily coordinated with its neighbours and, if we’re considering crystals (as we so very often do in condensed matter physics and chemistry), there’s a very well-defined periodicity and pattern established by the combination of the unit cell, the basis, and the lattice vectors. But all of that gets scrambled at the surface. Cut through a crystal to expose a particular surface — and not all surfaces are created equal by any means — and the symmetry of the bulk is broken; those atoms at the surface have lost their neighbours.
Atoms tend to be rather gregarious beasties so they end up in an agitated, high energy state when they lose their neighbours. Or, in slightly more technical (and rather less anthropomorphic) terms, creation of a surface is associated with a thermodynamic free energy cost; we have to put in work to break bonds. (If this wasn’t the case, objects all around us would spontaneously cleave to form surfaces. I’m writing (some of) this on a train back from London (after a fun evening at the LIYSF), having tremendous difficulty trying to drink coffee as the train rocks back and forth. A spontaneously cleaving cup would add to my difficulties quite substantially…)
In their drive to reduce that free energy, atoms and molecules at surfaces will form a bewildering array of different patterns and phases . The classic example is the (7×7) reconstruction of the Si(111) surface, one of the more complicated atomic rearrangements there is. I’ve already lapsed into the surface science vernacular there, but don’t let the nomenclature put you off if you’re not used to it. “Reconstruction” is the rearranging of atoms at a surface to reduce its free energy; the (111) defines the direction in which we cut through the bulk crystal to expose the surface; and the (7×7) simply refers to the size of the unit cell (i.e. the basic repeating unit or “tile”) of the reconstructed surface as compared to the arrangement on the unreconstructed (111) plane. Here’s a schematic of the (7×7) unit cell  to give you an idea of the complexity involved…
The arrangements and behaviour of atoms and molecules at surfaces are very tricky indeed to understand and predict. There has thus been a vast effort over many decades, using ever more precise techniques (both experimental and theoretical), to pin down just how adsorbed atoms and molecules bond, vibrate, move, and desorb. And although surface science is now a rather mature area, it certainly isn’t free of surprises and remains a vibrant field of study. One reason for this vibrancy is that as we make particles smaller and smaller — a core activity in nanoscience — their surface-to-volume ratio increases substantially. The devilish behaviour of surfaces is thus at the very heart of nanoscience, as reflected time and again in the presentations at the nanoscience@Surfaces 2018 summer school.
Unfortunately, I could only attend the Wednesday and Thursday morning of the summer school. It was an honour to be invited to talk and I’d like to take this opportunity to repeat my thanks to the organising committee including, in particular, Andy Jardine (Cambridge), Andrew (Tom) Thomas (Manchester), Karen Syres and Joe Smerdon (UCLAN) who were the frontline organisers in terms of organising my accomodation, providing the necessary A/V requirements, and sorting out the scheduling logistics. My lecture, Scanning Probes Under The Microscope, was on the Wednesday morning and, alongside the technical details of the science, covered themes I’ve previously ranted about at this blog, including the pitfalls of image interpretation and the limitations of the peer review process.
Much more important, however, were the other talks during the school. I regretfully missed Monday’s and Tuesday’s presentations (including my Nottingham colleague Rob Jones’ intriguingly named “Getting it off and getting it on“) which had a theory and photoemission flavour, respectively. Wednesday, however, was devoted to my first love in research: scanning probe microscopy, and it was great to catch up on recent developments in the field from the perspective of colleagues who work on different materials systems to those we tend to study at Nottingham.
Thursday morning’s plenary lecture/tutorial was from Phil Woodruff (Warwick), one of not only the UK’s, but the world’s, foremost (surface) scientists and someone who has pioneered a number of elegant techniques and tools for surface analysis (including, along with Rob Jones and other co-workers, the X-ray standing wave method described in the video at the foot of this post.)
Following Phil’s talk, there was a session dedicated to careers. Although I was not quite in the target demographic for this session, I nonetheless hung around for the introductions from those involved because I was keen to get an insight into just how the “careers outside academia” issue would be addressed. Academia is of course not the be-all-and-end-all when it comes to careers. Of the 48 PhD researchers I counted — an impressive turn-out given that 50 were registered for the summer school — only 10 raised their hand when asked if they were planning on pursuing a career in academia.
Thirteen years ago, I was a member of the organising committee for an EPSRC-funded summer school in surface science held at the University of Nottingham. We also held a careers-related session during the school and, if memory serves (…and that’s definitely not a given), when a similar question was asked of the PhD researchers in attendance, a slightly higher percentage (maybe ~ 33%) were keen on the academic pathway. While academia certainly does not want to lose the brightest and the best, it’s encouraging that there’s a movement away from the archaic notion that to not secure a permanent academic post/tenure somehow represents failure.
It was also fun for me to compare and contrast the Nottingham and Cambridge summer schools from the comfortable perspective of a delegate rather than an organiser. Here’s the poster for the Nottingham school thirteen years ago…
…and here’s an overview of the talks and sessions that were held back in 2005:
A key advance in probe microscopy in the intervening thirteen year period has been the ultrahigh resolution force microscopy pioneered by the IBM Zurich research team (Leo Gross et al), as described here. This has revolutionised imaging, spectroscopy, and manipulation of matter at the atomic and (sub)molecular levels.
Another key difference between UK surface science back in 2005 and its 2018 counterpart is that the Diamond synchrotron produced “first light” (well, first user beam) in 2007. The Diamond Light Source is an exceptionally impressive facility. (The decision to construct DLS at the Harwell Campus outside Oxford was underscored by a great deal of bitter political debate back in the late nineties, but that’s a story for a whole other blog post. Or, indeed, series of blog posts.) The UK surface science (and nanoscience, and magnetism, and protein crystallography, and X-ray scattering, and…) community is rightly extremely proud of the facility. Chris Nicklin (DLS), Georg Held (Reading), Wendy Flavell (Manchester) and the aforementioned Prof. Woodruff (among others) each focussed on the exciting surface science that is made possible only via access to tunable synchrotron radiation of the type provided by DLS.
I was gutted to have missed Stephen Jenkins‘ review and tutorial on the application of density functional theory to surfaces. DFT is another area that has progressed quite considerably over the last thirteen years, with a particular evolution of methods to treat dispersion interactions (i.e. van der Waals/London forces). It’s not always the case that DFT calculations/predictions are treated with the type of healthy skepticism that is befitting a computational technique whereby the choice of functional makes all the difference but, again, that’s a topic for another day…
Having helped organise a PhD summer school myself, I know just how much effort is involved in running a successful event. I hope that all members of the organising committee — Tom, Joe, Andy, Karen, Neil, Holly, Kieran, and Giovanni — can now have a relaxing summer break, safe in the knowledge that they have helped to foster links and, indeed, friendships, among the next generation of surface scientists and nanoscientists.
(a) Sulphur. S.u.l.p.h.u.r. Not the frankly offensive sulfur that I had to use in the papers submitted to US journals. That made for painful proof-reading. (b) I have no idea why I didn’t include mention of photoemission in the title of the thesis, given that it forms the guts of Chapter 5. I have very fond memories of carrying out those experiments at the (now defunct) Daresbury Synchrotron Radiation Source (SRS) just outside Warrington in the UK. Daresbury was superseded by the Diamond Light Source (DLS), discussed in this Sixty Symbols video.
 Assuming that there’s enough thermal energy to go around and that they’re not kinetically trapped in a particular state.
 Schematic taken from the PhD thesis of Mick Phillips, University of Nottingham (2004).