A discussion of recent FAH work on ab initio nanoreactor

What’s an ab initio nanoreactor for?

In an ab initio nanoreactor, molecules are allowed to react freely with each other over the course of the molecular dynamics simulation, and then we observe what products come out of it and how the products were formed. Besides obeying the fundamental laws of physics, no additional assumptions were imposed to the system, hence ab initio.

The number of reactant molecules used to seed the simulations was small (50-100 molecules) compared to the number of molecules typically used in experimental methods, but is nonetheless very large from the standpoint of quantum chemistry calculations. To make the reactions occur more rapidly, we periodically push the molecules to the center of the ab initio nanoreactor with a virtual piston. What this does is to make the molecules bump into one another more frequently, and also provide the energy required for certain reactions to take place.

The significance of ab initio nanoreator

Traditionally, experimental methods are heavily relied on to discover new molecules and reaction pathways, and computational methods mainly played a supportive role to complement experimental methods. The results of this study prove that computational methods can also play the leading role in discovery, and can help guide experimental methods by posing new hypotheses and suggesting which experiments to do. It’s especially useful for detecting complex chemical reactions where several things happen at the same step during the reaction process that’s hard to detect via experiments.

The potential applications of ab initio nanoreactors are broad. Because of the ab initio approach coupled with some refinement methods and automatic analysis, we can achieve the goal of discovering new molecules, new reaction pathways and mechanisms in many different settings and environments. For instance, it could contribute to out future understanding of the origin of life, birth of stars, means to increase the rate of chemical reactions, earth’s atmosphere, etc.

Results of the study

We carried out two ab initio nanoreactor simulations. The first simulation started with purely acetylene molecules, and we call it acetylene nanoreactor. The second simulation started with a mixture of chemicals postulated to exist in the early earth atmosphere. The second simulation is the computational version of a famous experiment conducted in 1952 (Urey-Miller experiment) that showed complex building blocks of life could form from simple inorganic molecules (1). We call the second simulation Urey-Miller nanoreactor.

For the acetylene nanoreactor, nearly 100 distinct products were formed after ~500 picoseconds simulation time. Many of these product molecules are large (up to over 70 atoms) due to the tendency of acetylene molecules to form long chains and 3D networks. These products are also diverse, for example some have rings some don’t; some are linear some are branched. After comparing our results with those of previous experiments, we found that the acetylene nanoreactor produced not only similar products, but also new products (2, 3).

For the Urey-Miller nanoreactor, the products were relatively small (up to 16 atoms). Among the discovered products, we have amino acids (which are what proteins consist of), urea (participating in metabolism, and the first byproduct of life to be synthesized in the lab) and a bunch of other molecules, all of which have also been detected in meteorites that may have delivered organic molecules to the early earth (4). Many of these molecules are also found in interstellar clouds (5). In addition to the high diversity of products, the Urey-Miller nanoreactor also identified a complex network of reactions (more than 700 distinct reactions). A significant fraction of these reactions are viable in the common environment we live in. Moreover, we found out that water and ammonia allow reactions to proceed faster with less energy for many of these reactions. Last but not least, hydrogen was found rarely involved in the synthesis of a naturally occurring amino acid, glycine, which supports previous proposals that molecules that tend to lose electrons (including hydrogen) don’t participate in biomolecule formations (4).

Method of analysis

To derive insight from a complex network of reactions, we focus on a particular molecule in the network and investigate the reactions it’s involved in, either it’s the product or the reactant. In this way, it allows us to trace the synthetic pathways that lead from the starting molecules. There could be several different pathways to get from the starting material to our molecule of interest. Some intermediate molecules are more common than the others among these distinct pathways.


(1) Miller, S. L. & Urey, H.C. Organic Compound Synthesis on the Primitive Earth. Science 130, 245-251 (1959). Doi: 10.1126/science.130.3370.245

(2) Trout, C.C. & Badding, J.V. Solid State Polymerization of Acetylene at High Pressure and Low Temperature. J. Phys. Chem. A 104, 8142-8145 (2000).

(3) Sakashita, M., Yamawaki H. & Aoki, K. FT-IR Study of the Solid State Polymerization of Acetylene Under Pressure. J. Phys. Chem. 100, 9943-9947 (1996).

(4) Danger, G., Plasson, R. & Pascal R. Pathways For the Formation and Evolution of Peptides in Prebiotic Environments. Chem. Soc. Rev. 41, 5416-5429 (2012).

(5) Menten, K. M. & Wyrowski, F. in Sterstellar Molecules: Their Laboratory and Interstellar Habitat (eds Yamada, K. M. T. & Winnewisser, G.) 27-42 (Springer Tracts in Modern Physics 241, Springer, 2011).

Everything else described here is from Wang, L. P., Titov, A. McGibbon, R., Liu, F., Pande, V. S. & Martinez, T. J. Discovering Chemistry With An ab initio Nanoreactor. Nature Chemistry. 2014. Doi: 10.1038/nchem.2099. The article can also be read about on Nov 10th issue of C&E News: http://cen.acs.org/articles/92/i45/Simulation-Technique-Finds-Reaction-Products.html