Lifeboat Foundation

How are chemical elements produced in our Universe? Where do heavy elements like gold and uranium come from? Using computer simulations, a research team from the GSI Helmholtzzentrum für Schwerionenforschung in Darmstadt, together with colleagues from Belgium and Japan, shows that the synthesis of heavy elements is typical for certain black holes with orbiting matter accumulations, so-called accretion disks. The predicted abundance of the formed elements provides insight into which heavy elements need to be studied in future laboratories — such as the Facility for Antiproton and Ion Research (FAIR), which is currently under construction — to unravel the origin of heavy elements. The results are published in the journal Monthly Notices of the Royal Astronomical Society.

All heavy elements on Earth today were formed under extreme conditions in astrophysical environments: inside stars, in stellar explosions, and during the collision of neutron stars. Researchers are intrigued with the question in which of these astrophysical events the appropriate conditions for the formation of the heaviest elements, such as gold or uranium, exist. The spectacular first observation of gravitational waves and electromagnetic radiation originating from a neutron star merger in 2017 suggested that many heavy elements can be produced and released in these cosmic collisions. However, the question remains open as to when and why the material is ejected and whether there may be other scenarios in which heavy elements can be produced.

Promising candidates for heavy element production are black holes orbited by an accretion disk of dense and hot matter. Such a system is formed both after the merger of two massive neutron stars and during a so-called collapsar, the collapse and subsequent explosion of a rotating star. The internal composition of such accretion disks has so far not been well understood, particularly with respect to the conditions under which an excess of neutrons forms. A high number of neutrons is a basic requirement for the synthesis of heavy elements, as it enables the rapid neutron-capture process or r-process. Nearly massless neutrinos play a key role in this process, as they enable conversion between protons and neutrons.

Machine learning (ML) models are powerful tools to study multivariate correlations that exist within large datasets but are hard for humans to identify^16,23. Our aim is to build a model that captures the chemical interactions between the element combinations that afford reported crystalline inorganic materials, noting that the aim of such models is efficacy rather than interpretability, and that as such they can be complementary guides to human experts. The model should assist expert prioritization between the promising element combinations by ranking them quantitatively. Researchers have practically understood how to identify new chemistries based on element combinations for phase-field exploration, but not at significant scale. However, the prioritization of these attractive knowledge-based choices for experimental and computational investigation is critical as it determines substantial resource commitment. The collaborative ML workflow^24,25 developed here includes a ML tool trained across all available data at a scale beyond that, which humans can assimilate simultaneously to provide numerical ranking of the likelihood of identifying new phases in the selected chemistries. We illustrate the predictive power of ML in this workflow in the discovery of a new solid-state Li-ion conductor from unexplored quaternary phase fields with two anions. To train a model to assist prioritization of these candidate phase fields, we extracted 2021 M_xM ′_yA_zA ′_t phases reported in ICSD (Fig. 1, Step 1), and associated each phase with the phase fields M-M ′-A-A′ where M, M ′ span all cations, A, A ′ are anions {N³⁻, P³⁻, As³⁻, O²⁻, S²⁻, Se²⁻, Te²⁻, F⁻, Cl⁻, Br⁻, and I⁻} and x, y, z, t denote concentrations (Fig. 1, Step 2). Data were augmented by 24-fold elemental permutations to enhance learning and prevent overfitting (Supplementary Fig. 2).

ML models rely on using appropriate features (often called descriptors)²⁶ to describe the data presented, so feature selection is critical to the quality of the model. The challenge of selecting the best set of features among the multitude available for the chemical elements (e.g., atomic weight, valence, ionic radius, etc.)²⁶ lies in balancing competing considerations: a small number of features usually makes learning more robust, while limiting the predictive power of resulting models, large numbers of features tend to make models more descriptive and discriminating while increasing the risk of overfitting. We evaluated 40 individual features^26,27 (Supplementary Fig. 4, 5) that have reported values for all elements and identify a set of 37 elemental features that best balance these considerations. We thus describe each phase field of four elements as a vector in a 148-dimensional feature space (37 features × 4 elements = 148 dimensions).

To infer relationships between entries in such a high-dimensional feature space in which the training data are necessarily sparsely distributed²⁸, we employ the variational autoencoder (VAE), an unsupervised neural network-based dimensionality reduction method (Fig. 1, Step 3), which quantifies nonlinear similarities in high-dimensional unlabelled data²⁹ and, in addition to the conventional autoencoder, pays close attention to the distribution of the data features in multidimensional space. A VAE is a two-part neural network, where one part is used to compress (encode) the input vectors into a lower-dimensional (latent) space, and the other to decode vectors in latent space back into the original high-dimensional space. Here we choose to encode the 148-dimensional input feature space into a four-dimensional latent feature space (Supplementary Methods).

A group of scientists at the U.S. Department of Energy’s Ames Laboratory has developed computational quantum algorithms that are capable of efficient and highly accurate simulations of static and dynamic properties of quantum systems. The algorithms are valuable tools to gain greater insight into the physics and chemistry of complex materials, and they are specifically designed to work on existing and near-future quantum computers.

Scientist Yong-Xin Yao and his research partners at Ames Lab use the power of advanced computers to speed discovery in condensed matter physics, modeling incredibly complex quantum mechanics and how they change over ultra-fast timescales. Current high performance computers can model the properties of very simple, small quantum systems, but larger or more complex systems rapidly expand the number of calculations a computer must perform to arrive at an accurate model, slowing the pace not only of computation, but also discovery.

“This is a real challenge given the current early-stage of existing quantum computing capabilities,” said Yao, “but it is also a very promising opportunity, since these calculations overwhelm classical computer systems, or take far too long to provide timely answers.”

Scientists found an easier way to produce a chemical reaction important in making biofuels, possibly lowering the cost of biofuel production.

Artificial Intelligence’s biggest Problems is their inability to keep on learning after they’ve completed their training. But now, Google’s Deepmind has created a Meta-Learning AI which keeps on learning and improving indefinitely without any Human supervision. Deepmind created the AI Game: Alchemy, which is a chemistry-based game for AI Agents to play and improve in. But Artificial Intelligence improving without limits also puts some concerns into AI researchers focused on deep learning.

There has been rapidly growing interest in meta-learning as a method for increasing the flexibility and sample efficiency of reinforcement learning.
–

Continue reading “The Only Artificial Intelligence that can Learn — Deepmind Meta-Learning” »

And it can be hacked.

The authors of a new study in Nature Catalysis reprogrammed these blobs—called exosomes—into an army of living nanobioreactors. It’s a seemingly simple process of mix and match: each blob is filled with a different chemical that’s involved in a biological reaction. By bringing two together, the blobs merge into a single squishy container, allowing the two chemicals to react.

The results were explosive. The tiny bioreactors pumped out energy molecules, called ATP, inside living cells. The burst of energy saved injured cells, providing them with a boost of power to fight back against dangerous molecules that otherwise lead to cell death.

To produce a cost-effective redox flow battery, researchers based at the South China University of Technology have synthesized a molecular compound that serves as a low-cost electrolyte, enabling a stable flow battery that retains 99.98% capacity per cycle. They published their approach on August 14 in the Energy Material Advances.

Comprising two tanks of opposing liquid electrolytes, the battery pumps the positive and negative liquids along a membrane separator sandwiched between electrodes, facilitating ion exchanges to produce energy. Significant work has been dedicated to developing the negative electrolyte liquid, while the positive electrolyte liquid has received less attention, according to corresponding author Zhenxing Liang, professor in the Key Laboratory of Fuel Cell Technology of Guangdong Province, School of Chemistry and Chemical Engineering, South China University of Technology.

“Aqueous redox flow batteries can realize the stable electrical output for using unsteady solar and wind energy, and they have been recognized as a promising large-scale energy storage technology,” Liang said. “Electroactive organic merit of element abundance, low cost and flexible molecular control over the electrochemical features for both positive and negative electrolytes are regarded as key to developing next-generation redox flow batteries.”

The nutrient content of our vegetables is down 40% over the last two decades and our soil health is suffering due to increasingly harsh herbicide use, according to Carbon Robotics founder Paul Mikesell. And farmers are increasingly concerned about the long-term health impacts of continually spraying chemicals on their fields.

But not weeding will cost half your crop, killing profitability.

The solution?

To understand why Mark Zuckerberg thinks “the metaverse” is the next frontier, consider the case of Sam Peurifoy. The 27-year-old chemistry PhD from Columbia University left his job at Goldman Sachs at the height of the pandemic and is now seeking out his fortune in crypto by playing video games.

He has recruited dozens of people from Mexico to the Philippines to a “Guild” that plays under the command of “Captain” Peurifoy. In exchange, he ponies up the funds needed to enter Axie Infinity, a game where players collect Smooth Love Potion — a digital token that can potentially be converted into real money.