Materials Research Institute @ QMUL

[qmulmri-blog]

Death by PowerPoint: the laser pointer adds the final cut

We have all suffered the experience of "death by PowerPoint", the set of slides that accompany a talk that frankly act to make the presentation even less lively than it ever could have been. There is value to having slides in a talk – to add a helpful graphic, an interesting animation, some key points to aid the awakening listener or the members of the audience who are are only half listening in between reading email messages, and some pithy quotes – but how to extract this value is rarely understood by speakers. My pet peeve is the use of tables. Dear reader, promise me please you will NEVER put a table into a PowerPoint presentation. No audience can read tables. If you want to make a point with data, use bar charts. The point you want to make will become obvious. But this post is not about PowerPoint per se, it is about the evil friend of PowerPoint, namely the laser pointer. Designed specifically to make death by PowerPoint all that bit quicker. One's intuition is that the laser pointer should only be a good thing. It allows the speaker to highlight specific features on their slides, like a data point on a graph, a data value in a table (yes, you will be unable to resist the temptation to include a table in your PowerPoint slides), or a specific bullet point within a list of bullet points on a single slide. But I would argue that any use of the laser pointer simply reinforces all that PowerPoint allows the speaker to get wrong. If you need a laser pointer to highlight something on a slide, frankly your slide is too complicated. Or if your graphic needs you to point out something with a laser pointer, your graphic is unclear. Sure, sometimes graphics are complex, but actually PowerPoint and other graphics apps have tools to help you add the highlight, like a selection of colours perhaps. So there will be times when the speaker does need to point out something in the screen. I would argue that most lecture rooms allow the speaker to use a pointer that nature itself has given the speaker, namely his or her hands and arms. Walk over to the slide and just point to what you want to highlight. So what if that might look comical; looking comical will just add good entertainment to the talk. And this is the serious point. In a talk, the focus should be on the speaker, yet PowerPoint has allowed many speakers to slip into the background so that the sides become the focus. The speaker is merely adding a scripted commentary, rather than engaging in a conversation with the audience. The slides sound be an aid, but they are too easily allowed to take over. And the laser pointer merely exaggerates this. The speaker can remain in the shadow and direct the audience even more towards the slides. So I say "ban the laser pointer" until you know how to engage with your audience, and even then only use the laser pointer sparingly. Using your hand instead will ensure that the focus is on you. And knowing you can't use a laser pointer to add focus to cluttered slides will force you to create clear slides from the outset. It can be done!

[qmulmri-blog]

One hundred years of diffraction: "probably the most important discovery you have never heard of"

An interesting piece about the centenary celebration of the development of x-ray crystallography by the father and son team William and Lawrence Bragg published yesterday in the Metro.

Essentially 100 years ago the Braggs reported their understanding of what happens when a beam of x-rays – electromagnetic radiation like light and radio waves, but with a wavelength as small as the size of an atom – strikes and is reflected by a crystal (the original paper can be downloaded from this link). A crystal is a regular array of atoms, and viewed from different directions give the impression of there being layers of atoms. What the Braggs worked out was how the x-rays can reflect from these individual layers, and they gave a mathematic equation that links the angle of reflection to the spacing between layers in that particular view. So if you know the angles that the x-rays are reflected though, you can reconstruct the fundamental structure of the layers, namely what crystallographers call the crystal lattice.

The article has a rather nice graphic that I reproduce at the end. The point of the article is to say that 100 years ago there was no method to look at materials with a microscope that takes you down to the size of the atom, and the significance of what the Braggs did was to make that possible, both from a theory point of view but also from a practical point of view. The field of crystallography owes a debt to many more people, as seen by the fact that several Nobel prizes have been won by crystallographers, and converting the intensity of the reflected beams into information about the electron density and hence the atomic structure came from a number of works and is somewhat more complicated than the article suggests, but that would be too much for one newspaper article.

The nice graphic is this:

[qmulmri-blog]

Queen Mary students presenting posters at the 2013 UK Neutron and Muon User Meeting. Queen Mary had the largest contingent of PhD students attending. They are great!

[missleton]

First results from our PDF diffractometer

Whilst not our first measurement per se, we now have our first analysis from which we have produced our histogram of interatomic distances within a measured material.

The material is called orthoterphenyl. The molecule structure is shown in one of the photographs; essentially three rings of carbon and hydrogen atoms connected together, with formula C18H14. This material easily forms a glass, and we have an experiment to measure the structure of the glass scheduled at the ISIS neutron facility. Having complementary x-ray data will help enormously with analysis of the neutron total scattering data.

The histogram of interatomic distances is called the “Pair distribution function”. It is displayed here as a function of all possible distances, which are represented by the symbol r, and the histogram function is labelled D(r). The peaks in D(r) correspond to the most common interatomic distances.

The peaks mostly arise from carbon atoms. The first three peaks are the nearest-neighbour (1.4 Å; the unit of distance of 1 Å is 0.1 nm or one 10th of a billionth of a metre), second-neighbour (2.45 Å) and third-neighbour (3.05 Å) carbon distances within single rings. For larger distances we see an overlap of several types of interatomic distances within the molecule rather than distinct peaks. For example, the peak at 3.74 Å that forms part of a multiple-peak group is the fourth-neighbour distance involving a carbon atoms in different rings. However, there are other distances between carbon atoms in different rings that could be at distances of around 2.7 Å that will be hidden around the second and third neighbour peaks, so unravelling this structure will require some analysis.

It is encouraging that this has worked so well.

Credit to the team who have made this work so far: my colleague Anthony Phillips and our great team of students Ed Beake, Dominic Carter, Helen Duncan, Juan Du and Serena Maugeri, together with support from Rory Wilson and Isaac Abrahams.

[missleton]

We teach our students to never use the default graphs generated within an Excel spreadsheet, and this article from the BBC shows exactly why we do so. The article contains two graphs taken straight from a spreadsheet.

Here is the first one

It is legible, but the authors had to describe the axes in the caption rather than within the graph as you would normally expect.

The bad example though is the second graph:

This is simply hard to read. It has labelled the axes, but the fonts are too small to read (and when you magnify, you still have troubles because the characters are badly pixelated). The colours are hard to distinguish (well, I can see yellow and green clearly, but the dark colours don’t leap out at you). Note the default shadows on some of the curves! This is a prime example of a style of graph we would criticise our students for; with a few tweaks it could be made much easier to read. Shame on the BBC for not bothering.

Oh, and we would also complain about one other thing. Whoever produced the images used jpeg format. This is a great format for photographs, but not for text, because the compression method can easily lead to strange halo effects. They got away with it this time – other than the fact that it is badly pixelated – but surely someone within the BBC knows better about wise choice of graphics formats too.

[qmulmri-blog]

Our new PDF x-ray diffractometer

This arrived very recently, and is now running experiments alongside the tests we want to do.

It has a silver x-ray source, giving a wavelength of 0.05 nm. At this short wavelength, we can perform measurements to relatively high values of a quantity called the scattering vector, with symbol Q. The idea is that we can transform the data into a histogram of interatomic distances in the sample being measured, showing all the bonds between atoms and the local atomic environment (this is called the "Pair distribution function", aka PDF). The higher the value of Q in the measurement, the better the resolution in this histogram, and hence the value of a silver source and an instrument optimised for high-Q measurements.

The instrument was made by PANalytical, and some details are available from the company web site.

We plan to use this instrument for research into medical materials (bone and dental materials, both natural and synthetic, as well as pharmaceutical materials), energy materials (batteries, fuel cells, thermoelectric materials), disordered materials and glasses, and functional ceramics, with an emphasis on understanding how atomic structure determines the properties of materials.