Computers excel in chemistry class
Machine learning models can rapidly and accurately estimate key chemical parameters related to molecular reactivity
© 2020 KAUST
KAUST researchers have developed a machine learning model that can analyze the structure of hydrocarbon molecules and accurately predict a property called enthalpy of formation. When it comes to estimating this property, the model already makes better predictions than conventional approaches, and its accuracy will only improve as more data is collected for the model to learn from.
"Data on molecular properties, such as enthalpy of formation, are essential for engineers modeling the kinetic mechanisms, or energy flows, of chemical reactions," says Kiran Yalamanchi, a Ph.D. student in the research group of Mani Sarathy, who led the research. "Kinetic mechanisms for hydrocarbon fuels are important for the development and optimization of engine designs and chemical reactors," Yalamanchi says.
Generating the large sets of thermodynamics data required for kinetic mechanism modeling typically uses an approach called group additivity, which has limited accuracy. "Group additivity was developed in the mid-20th century, and the field of data science has advanced a lot in the last few decades," Yalamanchi says.
So Yalamanchi and Sarathy approached KAUST computer scientist, Xin Gao, to apply machine learning to the problem. "Our initial study gave very promising results," Yalamanchi says. "This potential helped us to push toward converging machine learning with generating thermodynamic data."
Machine learning offers a way to take enthalpy of formation data--measured experimentally, or calculated for a small number of molecules using highly accurate but slow quantum chemistry computations--and then extrapolate to a much broader range of molecules.
The machine learning program analyzed a "training" dataset of molecule structures and their enthalpies of formation. It then used the patterns it detected to predict the enthalpy of formation of molecules it had not seen before.
Machine learning proved to be much more accurate than the traditional group additivity approach. "We got better estimates of enthalpy of formation of chemical species using machine learning methods compared to traditional methods," Yalamanchi says.
For example, although traditional group additivity can make relatively good predictions for simple molecules with linear structures, its accuracy decreases with more complex molecules, such as those that incorporate carbon rings in their structure. "The improvement we saw in estimates of enthalpy of formation, compared with traditional group additivity, was even more significant in the case of cyclic species," Yalamanchi adds.
"The results suggest that machine learning will become an increasingly important tool in the field," Sarathy says. "The ability to accurately predict important thermodynamic properties from molecular descriptors is an important step toward developing fully automated algorithms for predicting more complex chemical phenomenon," he adds.
The team is now running high accuracy quantum chemistry calculations to expand the machine learning models' training dataset. "In this way, we are developing a hybrid first-principles artificial intelligence framework for more accurate predictions of many physical-chemical properties," says Sarathy.
Other news from the department science
Blue-green algae sugar instead of glyphosate
Cooperation project develops environmentally friendly glyphosate alternative
Using clay to combat eternal toxins
TU Freiberg clarifies basis for innovative PFAS filter made of clay
Unveiling a new era of imaging: Boston University engineers lead breakthrough microscopy techniques
Researchers made significant advancements in the field of vibrational imaging
Phasing out fossil fuels could save millions of lives
The mortality burden attributable to air pollution from fossil fuel use is considerably higher than most previous estimates - a phaseout of fossil fuels would have tremendous, positive health outcomes
Replicating the structure of bird feathers
The new material could be used in batteries or filtration
Quantum tool opens door to uncharted phenomena
Method can contribute to a better understanding of quantum materials
Recovering instead of shredding: recycling batteries more efficiently
KIT researchers are working with industry to develop a more sustainable recycling process to recycle materials from lithium-ion batteries more effectively
Industry 4.0: No impact on energy consumption?
To what extent does the digitalisation of industrial and manufacturing processes (Industry 4.0) improve energy efficiency and thus reduce energy intensity?
New approach to the sensible utilisation of carbon dioxide from car exhaust gases
"A method has been discovered that uses impure CO2 streams and enables a breakthrough in the synthesis of valuable chemicals and pharmaceuticals"
Boosting PET recycling with higher standards for laboratory experiments
New study shows how enzymatic plastic degradation could be brought one step closer to commercialisation
Innovating Optoelectronic Components with Phosphorus
Significant breakthrough: phosphorus chemists develop new method to selectively introduce phosphorus and nitrogen atoms into polyaromatic systems
Artificial intelligence finds ways to develop new drugs
The chemists tested the process using borylation – a reaction that activates hydrocarbon scaffolds
X-rays reveal how glasses lose their stability
PETRA III experiment shows how atoms in glass behave as weaknesses appear
Most read news
Plastic-eating bacteria turn waste into useful starting materials for other products
Microbial Upcycling of Waste PET
Microbes could help reduce the need for chemical fertilizers
A coating protects nitrogen-fixing bacteria: Start-up to commercialise coated bacteria for large-scale use in regenerative agriculture
New designs for solid-state electrolytes may soon revolutionize the battery industry
Scientists achieve monumental improvements in lithium-metal-chloride solid-state electrolytes
Dow and Evonik announce startup of hydrogen peroxide to propylene glycol (HPPG) pilot plant
Innovative technology offer flexibility, lower costs, and a smaller environmental footprint
Inauguration of the world’s first pilot plant for the cost-efficient production of green methanol
Start-up C1 Green Chemicals AG and research partners develop fundamentally new production process
This is a battery
Two colored liquids bubbling through tubes: Is this what the battery of the future looks like?
Converting PFAS “forever chemicals” into valuable compounds
Scientists develop a new method to incorporate harmful perfluoroalkenes into N-heterocyclic carbene ligands
Not so silver lining: Microplastics found in clouds could affect the weather
Low-altitude and denser clouds contained greater amounts of microplastics
Graphene's proton permeability: A switch for future energy technologies
This discovery could lead to the development of more efficient hydrogen fuel cells and solar water-splitting devices
Lithium-ion batteries are no longer the gold standard in battery tech
On the way to safer and more powerful energy sources
CO2-free hydrogen: BASF receives funding approval for 54-megawatt water electrolysis plant
Proton exchange membrane (PEM) electrolyzer expected to produce up to 8,000 metric tons of hydrogen per year
More news from our other portals
Bowel cancer: aspirin activates protective genes
Researchers have identified a signaling pathway by which aspirin can inhibit colorectal cancer.
Dunning-Kruger effect with muesli bars
Those who know the least consider themselves highly competent
Autonomous measuring instruments systematically detect new materials
A new algorithm measures materials libraries up to four times faster than before: It’s based on machine learning
New drug delivery system could reduce daily diabetes shots to just three a year
Dietary management drugs have transformed Type 2 diabetes care, but daily injection routines are challenging for some patients
Naked Clams: The New Superfood Sensation Emerging from the Depths
Researchers found Naked Clams contain almost twice the amount of Vitamin B12 as blue mussels and have developed an efficient way to farm them
Researchers discover new ultra strong material for microchip sensors
A material that doesn't just rival the strength of diamonds and graphene, but boasts a yield strength 10 times greater than Kevlar
Pushers, overcrowded trains and phone zombies
Sprite presents the world's first vending machine that responds to the things that bother Generation Z the most
Aston University technology to combat the not-so sweet practice of honey fraud
Light technology to be used to detect if honey is blended with cheap additions
How stem cells and immune cells communicate
Lisec Artz Award for Simon Haas: Groundbreaking discovery of an unknown protective mechanism against blood cancer from stem cells
Scientists use quantum biology, AI to sharpen genome editing tool
"This study represents an exciting advancement toward, understanding how we can avoid making costly ‘typos’ in an organism’s genetic code"
From the trough to the plate - digitally calculated
Computer program "ConTrans" estimates how much of an undesirable substance is transferred from animal feed to food
Viral Impostors: Breakthrough for Virus Research
The penetration of viruses into cells can now be tracked with unprecedented accuracy thanks to innovative design for pseudoviruses
Fatty acid factory filmed at work
High-resolution images provide new insights into cellular fatty acid production: Potential for medicine and biotechnology
Textbook knowledge turned on its head: 3-in-1 microorganism discovered
Newly multifunctional bacterial species
Tönnies Group launches first nationwide "Meat Climate Platform"
100 guests at the Future Forum for Agriculture
Tracking down Environmental Toxins
Detection of per- and polyfluoroalkyl substances (PFAS) by interrupted energy transfer
Could eating turkey ease colitis?
According to data in mice, extra tryptophan could reduce the risk of future colitis flares
The weight of pollution: exposure linked to obesity
Chronic exposure to environmental pollutants found to increase risk of cardiovascular disease
Are healthy foods automatically sustainable, too?
Perceptions about sustainability and healthy food choices are closely linked