Interdisciplinarity in Physics

June 8, 2012

Much of my success in industrial research came from the diverse training that I obtained during my early years. As a materials scientist, I was necessarily skilled in a little chemistry, a little physics, and a little engineering. Somewhere along the way, I learned analog and digital electronics, which is great when you're a student trying to get antique laboratory equipment working; and, I learned computer programming, which was more fun than work.

The ability to bring many different perspectives to bear on a problem is very useful, especially when your problems involve something as technologically complex as a gas turbine engine. The formal term for bringing many scientific disciplines together is called interdisciplinarity. Fortunately, this practice of creating a whole that's greater than the sum of its parts is much easier than its spelling. How often do you use a nineteen-letter word? That's nearly five, four-letter words in a row.

A prime example of interdisciplinarity is the creation of integrated circuits. This field is the convergence of such disparate topics as quantum mechanics, chemistry, crystal growth, thermodynamics, optics, computer science and metrology. The basics of physics are nice as a start, but it takes a lot of technology to transpose such research in the behavior of electrons in solids into something that allows your child to video chat with grandma over the Internet.

I understand how it works, but I still can't understand how they're able to build it with any yield. Cross-section of a circa 2000 CMOS integrated circuit. The transistors are at the bottom, and the interconnections are just above them, with copper metalization in orange. The large bump at the top is a lead-free solder ball for attachment. (Original drawing by "Cepheiden," labels removed for clarity, via Wikimedia Commons).

I wrote about data mining in a recent article (Numb3rs, June 6, 2012). A team of biomedical engineers and computer scientists from the Aalto University School of Science in Finland have applied data mining techniques in a study of cooperation between the sub-fields of physics. They used the Physics and Astronomy Classification Scheme (PACS) codes of articles published from 1985-2009 as their dataset.

An alphabetical listing of the PACS codes can be found in ref. 1.[1] There you can find everything from Aberrations (42.15.Fr) to Z-pinch devices (52.58.Lq), with musical instruments (43.75.+a) in between. In the PACS scheme, which is a tree structure, the first digit specifies one of ten broad categories, while the second digit specifies a subfield of that category. The final characters specify a final subfield.

Unfortunately, professional associations try to monetize more than just their journals, and the American Institute of Physics, which created this Dewey Decimal System for physics, charges a licensing fee for the use of PACS.[2] I found this out many years ago, when, as a referee for an article in the pre-Internet days, I was asked to return a booklet of PACS codes to the editorial office so they wouldn't need to buy another.

The dataset for this study was limited to American Physical Society publications appearing in Physical Review (A, B, C, D and E), Physical Review Letters and Reviews of Modern Physics. If a paper contains two different PACS codes, then it's assumed that there's a connection between these subfields. The maximum spanning tree representing the 2009 connectivity of subdisciplines in physics is shown in the figure.

The maximum spanning tree representing the 2009 connectivity of subdisciplines in physics, as found from their PACS classifications. Click for larger image. (Fig. 5 of ref. 3, via the arXiv Preprint Server).[3]

The analysis shows that physics has been dominated by condensed matter and general physics in the 1985-2009 period under study. Interdisciplinary physics, however, has been steadily increasing in importance. So, when your biologist colleague asks you to participate in a study of inertial navigation in predatory birds, take him up on his offer.

References:

1. Alphabetic Index of the Physics and Astronomy Classification Scheme, American Institute of Physics.
2. AIP introduces a new licensing model for PACS, American Institute of Physics Press Release, July 14, 2009.
3. Raj Kumar Pan, Sitabhra Sinha, Kimmo Kaski and Jari Saramäki, "The evolution of interdisciplinarity in physics research," ArXiv Preprint Server, June 1, 2012.