Whether it is the oxygen that we breathe or animals that we eat, our lives are made possible by photosynthesis. Photosynthesis is the process that plants have used for billions of years to convert energy from the sun, water, and carbon dioxide into the life-sustaining biomolecules and oxygen gas (O2) which all other organisms need to survive. Even after years of investigation, questions concerning important details remain unanswered. One of the least understood steps in photosynthesis is the generation of oxygen gas from two molecules of water, which takes place at a small cluster of manganese (Mn), oxygen (O), and calcium (Ca) atoms known as the oxygen-evolving complex (OEC) Recently, a team led by Yuki Kurashige from the National Institutes of Natural Sciences in Okazaki, Japan is gleaning new information and rewriting what we know about this sensitive molecule with new computational techniques where traditional experiments have failed .
Energy Storage in Plants
Simply put, plants absorb light and give off oxygen gas, but the actual process is a complicated dance of electrons (negatively charged components of atoms) through a network of carrier molecules, ultimately being stored as in plant cells . This “dance” begins when light hits a chlorophyll molecule and removes an electron. This electron travels through a series of carrier molecules until it is trapped in a form that can be later utilized for energy. In order for the plant to continuously absorb and store energy during the day, the chlorophyll molecule needs to get another electron from somewhere to replace the one it lost. This is where the OEC comes into play, essentially taking electrons from water molecules, forming a bond between two oxygen atoms, and releasing O2 as the end product.
Video 1. Animation depicting the process of energy storage in plants .
It is notoriously difficult to remove electrons from water, and even more difficult to form bonds between oxygen atoms in order to generate O2, but the OEC performs these chemical reactions selectively and efficiently. Naturally there is great interest in studying the OEC from researchers who want to emulate the process and generate hydrogen and oxygen gas from water . Despite years of fastidious research, important details about the changes in the structure of, and arrangement of electrons within the OEC upon the removal of each electron remain unknown. Understanding these changes is key to uncovering the way in which the OEC forms O2 from water, but because the OEC is so delicate, it is difficult to study using conventional experiments. Computational chemistry is currently being used to arrive at a more complete picture of the OEC and providing new information where conventional experiments have fallen short.
Complicated Calculations Made Easy
Computational chemistry attempts to calculate the properties of molecules using standard chemical knowledge and complicated math. Researchers have used the results of these calculations not only to predict how molecules will react with each other, but also to uncover details that were unavailable before the introduction of modern computers. For instance, it is possible to creating simulations that allow one to “watch” how the entire structure of a molecule changes as it interacts with another molecule and forms a new bond. Computational chemistry allows researchers to understand chemical reactions without having to work with chemicals, and due to the recent advent of user-friendly computer programs and the decreasing cost of computer processors, it has experienced growth in popularity among researchers.
In order to begin to understand the properties of molecules, computational chemists perform quantum calculations to determine the total amount of energy stored in the arrangement of the electrons and atoms within a molecule. Knowing a molecule’s total energy, researchers can then predict molecular properties such as the way in which the electrons are arranged in the molecule, the likelihood of a reaction occurring between molecules, or the way that the molecule interacts with electromagnetic radiation. Although determining the total energy in a molecule is a complicated process in which interaction energies between all the electrons and atomic nuclei of the molecule must be calculated, scientists have facilitated the process by making a set of mathematical approximations and assumptions. These mathematical shortcuts have been crafted over decades to allow scientists to carry out simplified, yet meaningful calculations, despite modest computing resources. The results of these calculations have intrinsic errors in accuracy, but these errors are typically tolerable for most scientific calculations. In the case of in a system as sensitive and intricate as the OEC, however, a more rigorous method of calculation is needed, and rigorous calculations require can take weeks or even months as opposed to days.
Kurashige and coworkers took a method for simplifying quantum calculations known as the Density Matrix Normalization Group (DMNG), which is common in solid state physics, and applied it to the OEC.  This new approach allows for even the most complicated calculations to proceed in a timely manner. The DMNG allows scientists to calculate the simultaneous combination of all possible arrangements of electrons in a molecule by using powerful mathematical methods. Using even the most simplified picture of the OEC, the researchers had to calculate the interactions and possible configurations of 44 electrons in 35 orbitals. That’s an equation with nearly one quintillion (1,000,000,000,000,000,000) parameters!
Figure 2. The structural skeleton the OEC. The manganese and calcium atoms are also bound to water molecules and amino acids, but they have been removed for clarity. The depicted oxygen atoms most likely come from water molecules and are involved in the generation of O2.
A Revised View of the OEC
Just last year, researchers were able to take the best picture to date of the structure of the OEC using a method called X-ray crystallography . When this structure was analyzed by Kurashige and coworkers, however, they obtained a result that is inconsistent with decades of experimental data. They assert that this widely-heralded structure of the OEC is incorrect because it was damaged by the X-rays that were used to determine its structure, changing the distances between atoms slightly. While the atomic connections are basically correct, there are two extra electrons in the OEC that would not be there in during its normal operation in nature.
Having obtained evidence that we actually do not know the structure of the active form of the OEC, the authors moved on to calculate what the structure of the OEC might actually be during normal operation in a plant. Using decades-old, undisputed data providing the number of electrons and their magnetic behavior , they calculated the possible structures and configurations of electrons within the OEC for each of the four stages of the water splitting process with remarkable results. The calculation suggested that two of the manganese atoms participate in a completely different type of bond with oxygen atoms. Two of the manganese atoms (pictured above in pink) participate in normal covalent bonds with each of the oxygen atoms to which they are bound. These covalent bonds can be thought of as an equal sharing two electrons.
Figure 3. Model of a covalent bond. One way to visualize a bond where two electrons are shared by two atoms and rarely separate from each other.
The other two manganese atoms (pictured above in blue) tend to participate in what could be described as “di-radical bonds,” where occasionally, one electron will find itself on the oxygen atom and the other will find itself on the manganese.
Many researchers who study the OEC believe that O-O bonds may form between two oxygen atoms in the OEC, each containing an unpaired electron just like the oxygen atoms participating in these “di-radical bonds”. This calculation identifies the manganese atoms that are most likely to have di-radical bonds, making their adjacent oxygen atoms likely candidates for form the all-important O-O bond, but further investigation is needed to identify these oxygen atoms conclusively.
Figure 5. Bond formation between adjacent, metal-bound oxygen atoms. One was to visualize a possible mechanism for the generation of O2 in the OEC.
Many scientists regard quantum calculations as nothing more than a compliment to physical experiments, but Kurashige and coworkers demonstrate that rigorous quantum calculations can be used to call even physical experiments into question. Along with predicting which oxygen atoms in the OEC are likely to be released as oxygen gas, the authors suggest that previous studies on the structure of the OEC should be reevaluated in light of their results. Applying their technique to another common type of calculation, Kurashige and coworkers may now watch how the total energy of the OEC changes while moving any two oxygen atoms together to form a bond. Such a calculation will allow the authors to not only predict which oxygen atoms are converted to oxygen gas, but also to suggest biological experiments to test their results. Moving beyond the OEC, the improvement to computational chemistry offered by the DMNG will enable any scientist with access to a computer to save time, money, and materials by rapidly and accurately identifying promising molecules and materials for consumer applications or research ventures.
Dan Graham is a Ph.D. candidate in the Department of Chemistry and Chemical Biology at Harvard University.
 Yuki Kurashige, Garnet Kin-Lic Chan, and Takeshi Yanai, “Entangled quantum electronic wavefunctions of the Mn4CaO5 cluster in photosystem II.” Nature Chemistry, 2013, 5, p660–666 DOI: 10.1038/nchem.1677
 Sunyoung Kim, et al. “Oxygen Evolution in Photosynthesis”. Virginia Tech College of Agriculture and Life Sciences. http://www.ahnrit.vt.edu/ProjectPortfolio/InstructionalModules/multimedia/oxygen_evo_photo_7-08.html.
 Virtual Cell Animation Collection created Dr. Phillip McClean and Christina Johnson at North Dakota State University. http://vcell.ndsu.nodak.edu/animations.
 Holger Dau and Michael Haumann, “The manganese complex of photosystem II in its reaction cycle—Basic framework and possible realization at the atomic level.” Coordination Chemistry Reviews, 2008, 252, p273-295. DOI: 10.1016/j.ccr.2007.09.001
 Steven R. White “Density Matrix Formulation for Quantum Renormalization Groups.” Physical Review Letters, 1992, 69, p2863-2866
 Yasufumi Umena, Keisuke Kawakami, Jian-Ren Shen, and Nobuo Kamiya, “Crystal structure of oxygen-evolving photosystem II at a resolution of 1.9 Å.” Nature, 2011, 473, p55–60. DOI: 10.1038/nature09913
 James P. McEvoy and Gary W. Brudvig, “Water-Splitting Chemistry of Photosystem II.” Chemical Reviews, 2006, 106, p4455-4483. DOI: 10.1021/cr0204294