91ÖÆƬ³§

An international team of scientists, including researchers at CU Boulder, has created a database of molecular dynamics models that simulate the properties of cement in all its varieties.

The database is called , for cement force fields. It gathers methods for simulating force-field parameters for the various types of inorganic minerals present in cement, which is used to bind concrete, the most-used construction material in the world. Cemff allows academic and industrial researchers to draw upon many types of force fields to make accurate simulations and predictions of purpose-built cement formulations.

Cemff could help industry design stronger, more durable construction materials that also curtail carbon dioxide emissions from the manufacture of more than 3 billion tons of cement and concrete a year. Concrete manufacturing contributes as much as 8 percent of the greenhouse gas to the atmosphere.

Fifteen scientists at 11 institutions worked on the project led by Ratan Mishra of ETH Zurich, Rouzbeh Shahsavari of Rice University and Paul Bowen of EPFL Lausanne. Details appear in the Elsevier journal .

In this research, the force field isn't an invisible barrier from a science-fiction story. It's the collection of parameters scientists use to build computer models of atomic interactions. These parameters include the intrinsic energy of the atoms in a simulation system. They are used to calculate how atoms interact individually and collectively with their neighbors to give the material its properties.

The models show how the component molecules in cement interact with each other. These microscopic interactions determine how well concrete performs in real-world applications and allow for fine-tuning the material to perform at its best for decades and in the most environmentally conscious way.

One of the models included in the database is the interface force field platform developed and maintained by Professor Hendrik Heinz’s research team at CU Boulder. The interface force field provides parameters for all-atom simulations of cement minerals and more than 50 other inorganic compounds such as metals, clays and oxides. According to the study, "IFF follows rigorous science for the validations of molecular models with the description of physically and chemically most consistent parameters. It quantitatively describes chemical bonding via reliable partial atomic charges and enables accurate simulations of both bulk properties and interfacial properties of cement minerals in the presence of different mineral phases, water and polymers." The atomistic simulations help understand and predict the assembly of cement materials from the 1 to 1000 nanometer scale, as well as to inform models for accurate predictions on the microscale and macroscale.

"Molecular modeling still requires multiple trade-offs," said Mishra, lead author of the paper and a materials scientist at ETH Zurich. "The typical example is time versus accuracy, but more importantly, it is essential to recognize what specific models are good at and what they may be challenged with. Cemff will allow researchers to have a more comprehensive view on this question and to select the best approach for the problem they are tackling."

Cement consists primarily of calcium silicates that react with water to produce the hardened material that confers mechanical properties and durability to concrete. Nearly 60 percent of carbon dioxide emissions from cement production come from the decomposition of limestone, the source of calcium in cement. To reduce the carbon footprint, manufacturers often supplement the mix with clays, waste materials like fly ash and recycled materials.

These all influence the mechanical characteristics and resilience of the product; that is why there is a need for simulations at nanoscale that let manufacturers test mixes for strength and durability even before making real cement.

"I hope the open format and international base of the cemff database will encourage both the modeling and experimental community to create solid benchmarks to help understand and predict more accurately the properties of the most-used material on Earth and help us build a more sustainable future," Bowen said.