How to visually represent an idea

Felice Frankel in American Scientist:

I thought Sightings readers would be interested in an example of what has become for me an obvious, but too-often ignored, transformative exercise: clarifying and learning science by thinking about how to visually represent an idea, a process or a structure in science, for the purpose of explaining it.

My Harvard colleague and coauthor George Whitesides, with whom I am working on the book No Small Matter, forthcoming in 2009 from Harvard University Press, asked that I make an interesting representation of nanotubes. I am a science photographer, not an illustrator, so my first course of action is usually to think photographically. The obvious, making a scanning electron micrograph of a nanotube, was not an option. Others have done that, probably much better than I would have. I decided to photographically simulate a nanotube structure.

Here’s what I did. First I printed a black hexagonal pattern, representing a standard carbon lattice, on an 8×10 piece of transparent acetate (a). I then began to roll the acetate to make a tube. Immediately, something wonderful happened: I couldn’t make a decision about how to longitudinally connect the edges of the paper. I was faced with a few choices. The literature informed me that there were indeed various possible configurations for carbon nanotubes, and that the ultimate configuration was significant in determining the electrical properties of the nanotube.


For this image, I decided (for no particular reason other than aesthetic) to adopt what’s called the “zigzag” configuration and not to attempt to show the endcaps that tie up the dangling carbon bonds in a nanotube. I secured the edges of the acetate with a couple of pieces of tape and placed the tube on my flatbed scanner.  [Below left] is the image I came up with (b). Nothing terribly compelling. I then “inverted” the nanotube in Adobe Photoshop and combined a few “layers” of the same image to make multiple layers with varying degrees of transparency, resulting in (c). For the final composite image at left, I went a little further in nudging the image using various filters and additional inversions.

More here.