Supercomputers and Atoms: The Career of a Computational Chemist
In the late seventeenth century, Isaac Newton laid down the foundation for classical physics with his laws of motion and gravity. His work remained the central pillar of the physical sciences until the early twentieth century, when an entirely new set of physical laws—quantum mechanics—was in its infancy. The development of quantum mechanics was motivated by the discovery of the atom and its constituent subatomic particles; although the concept of an atom—a fundamental unit of matter that cannot be split—dates back to ancient Greece, the first pieces of scientific evidence supporting its existence emerged in the nineteenth century.
“Since the discovery of the atom, chemistry has tried to explain phenomena at the atomic level. This motivated the development of theories that could predict molecular structure and behavior. Due to the immensely small size of atoms, however, experimental techniques have trouble seeing these details—thus computational chemistry stepped in to directly simulate the properties of very small things,” says Dr. Paul Zimmerman, a computational chemist at the University of Michigan.
Indeed, the properties of the subatomic realm are drastically different from our intuition of the macroscopic world. Under the rules of quantum mechanics, the states of particles are described in probabilities rather than definite states. Additionally, as the microscopic systems become more complex, from single atoms to large molecules, the math needed to describe them becomes completely intractable. These formidable challenges of the quantum world necessitated the use of a new tool—the digital computer—to make any progress on our understanding of this new frontier.
Although computational chemistry emerged from the singular development of quantum mechanics, its modern applications are quite diverse.
As Dr. Zimmerman explains, “the field extends from energy sciences and materials for solar cells to medicinal synthesis and biological systems and beyond. In my opinion, there are a huge number of applications to computational chemistry to be excited about.”
In one recent study, a team of biochemists at Duke University and the University of Hong Kong used computational methods to determine the chemical mechanism by which bacterial pathogens bind to human cells during the infection process, a task that would have proven tremendously difficult by experimental approaches. Although computer models are unlikely to replace traditional approaches in other fields, they have nonetheless shown to be useful new tools.
In addition to its novel utility in a wide variety of fields, computational chemistry also demonstrates the synergy between theory and experiment, an often-contentious topic within the scientific community.
Dr. Zimmerman notes, “Theory and experiment are highly complementary to one another. The most successful science employs both—experiment leads to hypotheses that can be modeled by theory, and theory returns the favor by allowing predictions of the outcome of new experiments.”
Experimental labs often collaborate with computational labs to explain interesting results or guide experimental procedures, and professionals often participate in both types of research settings to ensure the validity of their findings. One of Dr. Zimmerman’s current collaborations is with an experimental catalysis lab.
“Our group is working with the McNeil group at Michigan to develop catalysts that can switch their functionality. Normal catalysts perform a single task very well, for instance synthesis of a specific plastic from petroleum resources. The collaboration with McNeil's group is to design and test catalysts that can perform more than one task, which should yield not only new basic science for catalysis, but also abilities to synthesize new types of materials.”
The interdisciplinary nature of computational chemistry provides professionals with tremendous flexibility. An added advantage of using computational resources either to predict experimental results or confirm empirical findings is the tremendous computational power of modern supercomputers; these machines can run thousands of complex mathematical calculations and provide accurate simulations of chemical systems in a matter of hours, all without the need for advanced expertise.
The day-to-day routine of a computational chemist at a university is nothing out of the ordinary: “Typical days are spent meeting and advising students, preparing and teaching classes, as well as writing grant applications,” though Dr. Zimmerman also notes that “graduate students working in this area are able to spend the majority of their time preparing and analyzing simulations.”
Unlike many other fields however—where the scope of required knowledge is relatively narrow—computational chemistry demands proficiency in multiple fields: “Programming and mathematics [are] necessary tools for computational chemistry,” Dr. Zimmerman explains. “These skills, however, are most useful when applied using chemical ideas. I spend the majority of my time thinking chemically, and then apply these chemical principles to simulation using whatever aspects of math and programming that are needed.”
The central role of modern technology in computational chemistry makes the field an excellent career choice for students of the 21st century. Traditionally, science students with an interest in math would pursue physics or engineering; computational chemistry provides an additional opportunity for the mathematically minded.
Additionally, computer science students who are interested in the physical sciences may wish to consider the field. Dr. Zimmerman also suggests that “students who are interested in how physical properties lead to big picture outcomes might love computational chemistry.” Perhaps most significantly, the field’s diverse array of potential applications—from materials engineering to drug design—outfits students to become indispensable professionals in an economy where strong STEM skills are sorely needed.