Interview with Dr. John Bechhoefer

Professor, Department of Physics

Thermodynamics and Statistical Mechanics of Small Systems

With interests ranging from the physics of liquid crystals to biological physics, Dr. Bechhoefer is one of SFU’s most versatile scientists. Recently, his research has taken a new direction: thermodynamics and statistical mechanics of small systems. Capitalizing on the control that can be exercised over small systems, Dr. Bechhoefer is testing principles that are central to traditional thermodynamics and statistical physics. In the 19th century, people thought about machines and developed thermodynamics to understand large systems, such as gasoline engines. With today’s technology, scientists can probe scales that are appropriate to biology (e.g., cellular level systems), necessitating a new kind of thermodynamics and theoretical physics to capitalize on what can be done in small systems.

What early life experiences influenced you to pursue a career in science?
As a five-year old, I wanted to be an astronaut, just like all the other kids who watched the Gemini and Apollo space missions; that dream morphed into an interest in astronomy. In high school I joined a summer science program attended by kids from all over the country. We had lectures during the day and in the evening we were tasked with calculating the orbit of an asteroid. That exciting experience drew me into science.

How did you go from studying astronomy to studying snowflakes?
I was an astronomy major as an undergrad, but I was also interested in history, philosophy, and science. In graduate school in physics, I became interested in nonlinear dynamics and chaos, a topic that is big now but was essentially unheard of in the early 1980s. My thesis looked at the equivalent of a snowflake in liquid crystals, such as those in the LCD laptop display, which are somewhere between a crystal and a liquid.

I gained a deeper understanding of why the crystals all look so different. Very small differences in their environment or trajectory will give differences in shape that are magnified as they grow; that's fundamentally the origin of why there are so many different types of snowflakes. And that sensitivity to conditions is exactly the mechanism of chaos.

How has your research program evolved since you started at SFU?
When I first came to SFU I worked on new dynamics and pattern formation, as well as problems of solidification and crystallization in liquid crystals and polymers. I continued with liquid crystals research but explored deeper questions about the nature of phase transitions.

 Later, I learned about a new technique developed in the lab of a biologist who studied DNA replication. They put genomic DNA on a cover slide and looked at it under the microscope, stretching it in a way that allowed them to do optical mapping. They labelled the DNA to see when it was replicated, but they couldn't tell the difference between replications with one origin versus two origins or ones that started at different times. It occurred to me that the replicative process is analogous to what happens in crystallization. Guided by the calculations and mathematics used for analysing crystallization, we applied mathematics to DNA replication. One outcome of this very theoretical work was to develop great research tools for biologists.

An attractive aspect of my main focus now—thermodynamics and statistical mechanics of small systems—is the chance to do more lab experiments. Changing focus every few years can be a good thing because you're asking fresh questions.

Your new research direction lies between theoretical physics and biology. Which topics are you studying right now?
Part of my work on thermodynamics and statistical mechanics of small systems looks at the interface between information and dynamics. In the early 1960s Rolf Landauer at IBM started thinking about the efficiency of computers, i.e., what is the minimum amount of energy required for the computer to function? Landauer realized that, in principle, much of what a computer does in terms of computations would not require drawing electricity; however, the memory of a computer must eventually be erased, and that process dissipates energy.

As part of my new research direction, we confirmed a theory that sets the minimum amount of power required to run a computer. We used a physical model to carry out memory erasures in a setting where you can keep track of the very small levels of energy required. Going forward, we want to understand how the information we acquire can be converted to other possibilities. For example, looking at a biological cell’s ability to sense its environment, what trade-offs must the cell make to do this? What information is being taken in, and how much out of equilibrium is the system as a result – i.e. how much ATP is required to carry out functions at a given rate? Compromises are made between the energy supply required, the power dissipated, and the accuracy of the information you're obtaining. In essence, we are trying to do for cells what 19th century physicists did for engines.

How does your approach to a biological problem differ from that of a biologist?
A lot of our work involves physical models, e.g. using a small colloidal particle in fluid and subjecting it to forces that we control very accurately. This positions us between the theorists who come up with the idea on paper and experimentalists who work with the biological systems, which due to their complexity are difficult to control and observe accurately. Physicists are trained to understand simple settings and then increase the level of complexity by adding elements, whereas biologists often study systems (e.g., the working cell) with a top down approach.

What research obstacle keeps you awake at night?
What keeps me up at night is the evolution of the research funding system in Canada over the last number of years – I see a shift within Canadian science that makes funding difficult to obtain, particularly for people whose interest is basic science. Some of the changes made during the past ten years to the Natural Sciences and Engineering Council of Canada (NSERC) have been good, but many have not. I think the NSERC system has been changed in a way that does not favour places like SFU, or people like me. I feel like there is a paradox where on one hand, my work is going as well as it has at any point in my career and yet this is the point where I am most concerned about what's going to happen with my NSERC funding.

What topics are you excited about that you don't study?
Big data. I see right now that all of a sudden there is real data available where there wasn't before. Now you can construct a real science around new real data; for instance, if you want to study transportation and you sit at an intersection and count cars there's only so much you can do, but once all the cars have GPS and you know what everyone is doing in the entire city, then you have a more global picture. There are many examples where suddenly there is access to large amounts of data, allowing us to answer deeper scientific questions. There's an enormous opportunity to be at the start of a scientific approach to questions that previously could only be examined qualitatively.

What other interest or occupation fascinates you so much that you might choose it were you to start all over again?
Among other things, I'm fascinated by the ability to use sequencing and genomics of historical evolutionary things to reconstruct history on the basis of genomic evolution. For example, people are sequencing Neanderthal and ancient human DNA to reconstruct what happened 45,000 years ago. Using DNA sequencing as an independent way of reconstructing history is pretty cool – it would be fun to be a historian who reconstructs events using genomic information.


Read more: Dr. Bechhoefer’s profile on the Department of Physics website, his personal website and the Featured Researchers page

Interview by Jacqueline Watson with Theresa Kitos