From Swiss Federal Institute of Technology in Lausanne [EPFL-École Polytechnique Fédérale de Lausanne] (CH): “Machine learning cracks the oxidation states of crystal structures”

From Swiss Federal Institute of Technology in Lausanne [EPFL-École Polytechnique Fédérale de Lausanne] (CH)

Nik Papageorgiou

Chemical engineers at EPFL have developed a machine-learning model that can predict a compound’s oxidation state, a property that is so essential that many chemists argue it must be included in the periodic table.


Chemical elements make up pretty much everything in the physical world. As of 2016 we know of 118 elements all of which can be found categorized in the famous periodic table that hangs in every chemistry lab and classroom.

Each element in the periodic table appears as a one-, two-letter abbreviation (e.g. O for oxygen, Al for aluminum) along with its atomic number, which shows how many protons there are in the element’s nucleus. The number of protons is enormously important, as it also determines how many electrons orbit the nucleus, which essentially makes the element what it is and gives it its chemical properties. In short, the atomic number is an element’s ID card.

The periodic table should include oxidation states.

Publishing in Nature Chemistry, chemical engineers at EPFL’s School of Basic Sciences investigate another number that must be reported for each element in the periodic table: the element’s oxidation state, also known as oxidation number. Simply put, the oxidation state describes how many electrons an atom must gain or lose in order to form a chemical bond with another atom.

“In chemistry, the oxidation state is always reported in the chemical name of a compound,” says Professor Berend Smit who led the research. “Oxidation states play such an important role in the fundamentals of chemistry that some have argued that they should be represented as the third dimension of the periodic table.” A good example is chromium: in oxidation state III it is essential to the human body; in oxidation state IV, it is extremely toxic.

Complex materials complicate things

But although figuring out the oxidation state of a single element is pretty straightforward, when it comes to compounds made up of multiple elements, things become complicated. “For complex materials, it is in practice impossible to predict the oxidation state from first principles,” says Smit. “In fact, most quantum programs require the oxidation state of the metal as input.”

The current state-of-the-art in predicting oxidation states is still based on a something called “bond valence theory” developed in the early 20th century, which estimates the oxidation state of a compound based on the distances between the atoms of its constituent elements. But this doesn’t always work, especially in materials with crystal structures. “It is well known that it is not only the distance that matters but also the geometry of a metal complex,” says Smit. “But attempts to take this into account have not been very successful.”

A machine-learning solution

Until now, that is. In the study, the researchers were able to train a machine-learning algorithm to categorize a famous group of materials, the metal-organic frameworks, by oxidation state.

The team used the Cambridge structural database, a repository of crystal structures in which the oxidation state in given in the name of the materials. “The database is very messy, with many errors and a mixture of experiments, expert guesses, and different variations of the bond valence theory are used to assign oxidation states,” says Smit. “We assume that chemistry is self-correcting,” he adds. “So while there are many errors on individual accounts, the community as a whole will get it right.”

“We basically made a machine-learning model that has captured the collective knowledge of the chemistry community,” says Kevin Jablonka, a PhD student in Smit’s group at EPFL. “Our machine learning is nothing more than the television game “Who Wants To Be A Millionaire?” If a chemist does not know the oxidation state, one of the lifelines is to ask the audience of chemistry what they think the oxidation state should be. By uploading a crystal structure and our machine-learned model is the audience of chemists that will tell them what the most likely oxidation state is.”

See the full article here .


Please help promote STEM in your local schools.

Stem Education Coalition

EPFL bloc

EPFL campus

The Swiss Federal Institute of Technology in Lausanne [EPFL-École polytechnique fédérale de Lausanne] (CH) is a research institute and university in Lausanne, Switzerland, that specializes in natural sciences and engineering. It is one of the two Swiss Federal Institutes of Technology, and it has three main missions: education, research and technology transfer.

The QS World University Rankings ranks EPFL(CH) 14th in the world across all fields in their 2020/2021 ranking, whereas Times Higher Education World University Rankings ranks EPFL(CH) as the world’s 19th best school for Engineering and Technology in 2020.

EPFL(CH) is located in the French-speaking part of Switzerland; the sister institution in the German-speaking part of Switzerland is the Swiss Federal Institute of Technology ETH Zürich [Eidgenössische Technische Hochschule Zürich)](CH). Associated with several specialized research institutes, the two universities form the Domain of the Swiss Federal Institutes of Technology (ETH Domain) [ETH-Bereich; Domaine des Écoles polytechniques fédérales](CH), which is directly dependent on the Federal Department of Economic Affairs, Education and Research. In connection with research and teaching activities, EPFL(CH) operates a nuclear reactor CROCUS; a Tokamak Fusion reactor; a Blue Gene/Q Supercomputer; and P3 bio-hazard facilities.

The roots of modern-day EPFL(CH) can be traced back to the foundation of a private school under the name École spéciale de Lausanne in 1853 at the initiative of Lois Rivier, a graduate of the École Centrale Paris (FR) and John Gay the then professor and rector of the Académie de Lausanne. At its inception it had only 11 students and the offices was located at Rue du Valentin in Lausanne. In 1869, it became the technical department of the public Académie de Lausanne. When the Académie was reorganised and acquired the status of a university in 1890, the technical faculty changed its name to École d’ingénieurs de l’Université de Lausanne. In 1946, it was renamed the École polytechnique de l’Université de Lausanne (EPUL). In 1969, the EPUL was separated from the rest of the University of Lausanne and became a federal institute under its current name. EPFL(CH), like ETH Zürich(CH), is thus directly controlled by the Swiss federal government. In contrast, all other universities in Switzerland are controlled by their respective cantonal governments. Following the nomination of Patrick Aebischer as president in 2000, EPFL(CH) has started to develop into the field of life sciences. It absorbed the Swiss Institute for Experimental Cancer Research (ISREC) in 2008.

In 1946, there were 360 students. In 1969, EPFL(CH) had 1,400 students and 55 professors. In the past two decades the university has grown rapidly and as of 2012 roughly 14,000 people study or work on campus, about 9,300 of these being Bachelor, Master or PhD students. The environment at modern day EPFL(CH) is highly international with the school attracting students and researchers from all over the world. More than 125 countries are represented on the campus and the university has two official languages, French and English.


EPFL is organised into eight schools, themselves formed of institutes that group research units (laboratories or chairs) around common themes:

School of Basic Sciences (SB, Jan S. Hesthaven)

Institute of Mathematics (MATH, Victor Panaretos)
Institute of Chemical Sciences and Engineering (ISIC, Emsley Lyndon)
Institute of Physics (IPHYS, Harald Brune)
European Centre of Atomic and Molecular Computations (CECAM, Ignacio Pagonabarraga Mora)
Bernoulli Center (CIB, Nicolas Monod)
Biomedical Imaging Research Center (CIBM, Rolf Gruetter)
Interdisciplinary Center for Electron Microscopy (CIME, Cécile Hébert)
Max Planck-EPFL Centre for Molecular Nanosciences and Technology (CMNT, Thomas Rizzo)
Swiss Plasma Center (SPC, Ambrogio Fasoli)
Laboratory of Astrophysics (LASTRO, Jean-Paul Kneib)

School of Engineering (STI, Ali Sayed)

Institute of Electrical Engineering (IEL, Giovanni De Micheli)
Institute of Mechanical Engineering (IGM, Thomas Gmür)
Institute of Materials (IMX, Michaud Véronique)
Institute of Microengineering (IMT, Olivier Martin)
Institute of Bioengineering (IBI, Matthias Lütolf)

School of Architecture, Civil and Environmental Engineering (ENAC, Claudia R. Binder)

Institute of Architecture (IA, Luca Ortelli)
Civil Engineering Institute (IIC, Eugen Brühwiler)
Institute of Urban and Regional Sciences (INTER, Philippe Thalmann)
Environmental Engineering Institute (IIE, David Andrew Barry)

School of Computer and Communication Sciences (IC, James Larus)

Algorithms & Theoretical Computer Science
Artificial Intelligence & Machine Learning
Computational Biology
Computer Architecture & Integrated Systems
Data Management & Information Retrieval
Graphics & Vision
Human-Computer Interaction
Information & Communication Theory
Programming Languages & Formal Methods
Security & Cryptography
Signal & Image Processing

School of Life Sciences (SV, Gisou van der Goot)

Bachelor-Master Teaching Section in Life Sciences and Technologies (SSV)
Brain Mind Institute (BMI, Carmen Sandi)
Institute of Bioengineering (IBI, Melody Swartz)
Swiss Institute for Experimental Cancer Research (ISREC, Douglas Hanahan)
Global Health Institute (GHI, Bruno Lemaitre)
Ten Technology Platforms & Core Facilities (PTECH)
Center for Phenogenomics (CPG)
NCCR Synaptic Bases of Mental Diseases (NCCR-SYNAPSY)

College of Management of Technology (CDM)

Swiss Finance Institute at EPFL (CDM-SFI, Damir Filipovic)
Section of Management of Technology and Entrepreneurship (CDM-PMTE, Daniel Kuhn)
Institute of Technology and Public Policy (CDM-ITPP, Matthias Finger)
Institute of Management of Technology and Entrepreneurship (CDM-MTEI, Ralf Seifert)
Section of Financial Engineering (CDM-IF, Julien Hugonnier)

College of Humanities (CDH, Thomas David)

Human and social sciences teaching program (CDH-SHS, Thomas David)

EPFL Middle East (EME, Dr. Franco Vigliotti)[62]

Section of Energy Management and Sustainability (MES, Prof. Maher Kayal)

In addition to the eight schools there are seven closely related institutions

Swiss Cancer Centre
Center for Biomedical Imaging (CIBM)
Centre for Advanced Modelling Science (CADMOS)
École cantonale d’art de Lausanne (ECAL)
Campus Biotech
Wyss Center for Bio- and Neuro-engineering
Swiss National Supercomputing Centre