The Snowman, The NPL Christmas Card

Having got a request to make an image for the NPL Christmas card I decided to show-case the techniques I use on a daily basis when making devices for the Quantum Detection group. The following tells the story of the materials used, the methods used, images of the process, and the aftermath...


There are two main items used in the making of the snowman. Atomic force microscope (AFM) probes and tin balls. Think of AFM cantilevers as very small beams with a sharp tip at the end. The sharp tip “feels” the surface, and from this we can reconstruct an image of what it looks like, sometimes even resolving the atoms. For a much better description and more detail please see the excellent FAQ’s at the website of
MikroMasch one of the leading suppliers of AFM probes. For our purpose the basic description of an AFM probe would be a probe body consisting of a piece of silicon approximately 3mm long, 1.5mm wide and 0.3mm thick. On one end of the probe body is a very thin “diving board” beam. This beam is approximately 200 micrometers long, 30 micrometers wide, 0.5 micrometers thick, and on the underside at the end of the beam is the sharp tip.

Tin balls are exactly as you would imagine. They are balls made from the metal tin. These are used to help correct an abberation in electron optical lenses called astigmatism. Over an area of a few square mm a huge number of these tin balls is deposited. Depending on the resolution of the instrument you are using, these are typically from a few tens of nm in diameter, up to as much as 100 micrometers or more. Quite simply if the microscope is astigmatic the balls will not appear round but will appear elongated. We can correct for this until the balls appear round again and we then know the microscope is free from astigmatism.

The Method

Starting with our tin balls I first designed a simple smiley face. I can control the path of an ion beam to physically carve anything under the beam. For examples see my
YouTube page. Using this method I made the head of the snowman. I also deposited a small blob of platinum for a nose. Here is a couple of images of some of the test heads I made on the tin ball sample.

Having tested out the pattern for the face I next assembled the body and head on the AFM cantilever. The reason I used an AFM cantilever was to at least have an element of metrology in the finished thing. Here are two images. One showing me putting the head in place using the nanomanipulation system and the second showing a low-magnification image of the AFM probe.

Finally here are some images of the finished thing at different magnifications. It is a little over 10 micrometers wide (about 25% the width of a human hair), the nose is 700nm wide. The last image is as it appeared on the Christmas card, having been false coloured and some photoshop snow added.

The Aftermath...

Well what can I say. Shortly after I made this and captured the images, NPL decided to make a video to accompany it, as they decided it would make good publicity for the lab at Christmas time. About three hours after I agreed the accompanying text for the press release I was told it had been picked up by a national UK newspaper. By the next morning it had been picked up by five national newspapers, either in print or in the online versions. By the next weekend it had been picked up all over the world, was on 70,000 websites, countless newspapers, and my inbox had gone into meltdown. It has now appeared in many publications, often to help get science across to children, and has featured on no less than CNN where it might have had an audience of zillions. For those in the science community I wish I could trade this publicity for cites. For those few not in the science community who raised questions about the cost, or asked why I was making snowmen and not curing cancer or the common cold, the answers to those questions is this. The cost of a tin ball sample is about £60. The particular sample I used I already owned, and now has two tin balls missing from hundreds of thousands, so effectively no cost there. The cost of an AFM cantilever can be a few £s up to several hundreds of pounds, I made use of a pre-used one, you can see the tip is damaged in the last image, so cost there is also effectively zero. The making of the snowman took about three hours. This of course has a cost, as my time and the instrument is paid for. What I will say in my defense if that I need to practice my skills on something, as I often work with devices costing many thousands of pounds, and also the publicity it has raised has lead to potential project partners who did not know of such techniques. Therefore it may lead to further funding and more than pay for itself. I do not, and have never worked in medical science. I do however work on projects that make devices that might be used in medicine, and I use the techniques described on these pages to make them. Finally, the feedback I received was overwhelmingly positive with only a few dissenting voices, so I feel it was justified.