Molecules, Models, and Magic: The Exciting World of Computational Chemistry

Vrinda Modi

I often spend time in a little cubicle in McLean in the freezing corporate cold wrapped in a blanket in the Hoboken summers sipping on free coffee from the office above. It almost always tastes like chalk, but the routine is so embedded in me. Hence, this is the perfect opportunity to talk a little more about my research in computational chemistry here at Stevens and the advances I expect in the next few years. This field excites me because it bridges my interest in chemistry with modern computing techniques, a powerful tool for advancing our understanding of chemical reactions and molecular interactions. This means we are a dry lab – running simulations on computers. No chemicals, no PPE, just models and algorithms.

But why does computational chemistry exist in the first place, especially when chemistry is so rooted in physical matter? The answer is efficiency. Modeling molecules on a computer saves us from ordering chemicals, waiting for deliveries (a delay in chemistry teaching labs), and waiting hours—or even days—for experiments to finish. Computational chemistry makes our lives easier by letting us bypass the physical hassle while still providing valuable insights.

We use a software called Gaussian, which runs computational calculations and generates an output geometric structure based on your input (your best guess). It then generates a file containing information about the molecule’s thermodynamic properties (energies, enthalpies, etc.), geometric parameters, and electronic charges. Then we use GaussView to view the 3D structure on our screens. This process allows us to compare molecular structures, determining which ones are more energetically favorable—lower energy configurations are generally more stable.

My research focuses on investigating catalysts and the goal is to see how efficiently they can lower activation energies, thereby accelerating chemical reactions. I’m particularly interested in biocatalysts derived from heme enzymes, which are present in biological systems (including humans). These biocatalysts feature a metal center, typically iron, and react with carbene species—highly reactive carbon atoms. One reaction we explore is cyclopropanation, which forms a cyclopropane, a triangular, three-membered carbon ring that is notoriously difficult to synthesize. Through computational models, we can simulate these reactions and assess their feasibility without needing to perform them in a physical lab.

Computational chemistry, however, is not a replacement for experimental chemistry. Instead, it supplements traditional methods by simplifying the research process. This brings us to a larger question in science: Are new discoveries really "new"? As John Horgan points out in his article, “Huge Study Confirms Science Ending! (Sort Of),” much of today’s research builds on past knowledge rather than introducing groundbreaking theories. Computational chemistry fits this model—it consolidates our understanding of biocatalysts and reactions. While the experimental work in this field dates back to the 1980s, tools like Gaussian have modernized how we approach these reactions. I do believe that despite these advancements, some aspects of this software are already becoming outdated, and there is plenty of room for improvement.

The publishing frenzy in science also makes it harder to challenge existing theories, as Horgan argues. Researchers are often incentivized to produce work that supports what’s already known, rather than taking risks on revolutionary ideas. This is another area where computational chemistry can contribute by providing a reliable, methodical approach to testing hypotheses before they are brought to the lab. It doesn’t always lead to breakthroughs, but it helps us build on solid foundations. It’s also simply easier to back up what’s already believable. 

One of the reasons I’m particularly drawn to computational chemistry is its connection to quantum mechanics and quantum computing. At its core, computational chemistry is an extension of theoretical chemistry, which in turn is based on theoretical physics. Thus, the basis of the tools we use is from unified theories of fundamental forces, like electromagnetic and nuclear forces, as Hawking defines in “Is the End in Sight for Theoretical Physics?” He addresses the challenge of finding such a unified theory for gravity and emphasizes the use of “computational methods that will enable us to make predictions.” So, we all agree that there is a growing need for computational methods that can unify our understanding of these forces and make accurate predictions about the behavior of particles at the quantum level. 

One of the most exciting prospects of computational chemistry lies in its potential to leverage AI/ML to predict molecular behavior more accurately (Lim, 2019). This could revolutionize fields like drug discovery, where computational models are already being used to screen potential pharmaceuticals (including those containing cyclopropanes, for instance) before they are synthesized in the lab, which is why it’s so exciting even though it doesn’t necessarily make way for breakthroughs. I also love how Lim writes in her article that as science progresses, we are looking for more ways to generalize rather than specialize our tools to be applicable in whatever circumstance we feed it. In a way, we are hungry for these unified theories and tools that may inadvertently make breakthroughs rarer. She also states that we are creatively working hard to be lazy and I could not agree more. Who doesn’t wanna be lazy now and then?

References

1. Horgan, J. (2019). Huge Study Confirms Science Ending! (Sort Of). John Horgan (The Science Writer). https://johnhorgan.org/cross-check/yrb9e7uefpeqrlkiasoc6octxtnm5g

2. Hawking, S. (1981). Is the End in Sight for Theoretical Physics? Physics Bulletin, 32(15). 10.1088/0031-9112/32/1/024

3. Lim, M. (2019). Machine Learning and Optimization in Chemistry. Bear With Us. https://medium.com/unicorns-in-hoodies/machine-learning-and-optimization-in-chemistry-b51efd3ea0a5