This invention introduces a new paradigm for materials science and solid-state chemistry by moving away from an explorative experimental synthesis, followed by phenomenological modeling approach, towards a predictive process by identifying phase space where metastable states occur from machine learning and first principles.
Below is a comprehensive list of articles, events, projects, references and research related content that is specific to the organization described above. Use the filter to narrow the results further or please visit Science and Technology Partnerships and Outreach for more information.
- This invention comprises a predictive machine learning framework that leverages high performance computing and high-level quantum calculations to predict, determine, and validate phase diagrams in chemical systems without any recourse to experimental infoIntellectual Property Available to License
- This invention comprises the introduction of an active learning method that starts training a neural network with “a single” data point.Intellectual Property Available to License
Please contact us for additional information
- This invention addresses a critical issue of multiple objective optimization and is a general approach applicable to all types of model or property optimization.Intellectual Property Available to License
This invention comprises a solution to the problem of simultaneous optimization of multiple objectives by creating a “hierarchy of objectives with passing criteria” which allows an optimization algorithm to understand the relative importance of the objectives and also provides a pathway for it to tackle the list of objectives in a systematic baby-stepping manner that is based on well defined target values rather than arbitrary weights.
- Process and apparatus for mixed solid separations, including the separation of metal from plastic and metal from metal.Intellectual Property Available to License
Argonne’s full-scale mixed solid separation module. Available for collaborative demonstration projects. IP package includes a granted US patent and a set of copyrighted engineering manuals.
Opportunity & Solution
Electrical and electronic waste (E-waste) is a fast-growing segment of the global solid waste stream. Recycled materials such as metals, plastic and glass formed the largest market share. This technology has been demonstrated to significantly decrease contamination of low-grade E-waste plastic and metal recycling fractions.
▪ Provides new opportunities to increase E-waste value
▪ Low cost
▪ Easy to operate
▪ Modular design (basic sink/float; kinetic separation; froth flotation)
- Teaser: A general purpose technology for growing inorganic materials within polymeric templates.Intellectual Property Available to License
Argonne has developed a comprehensive technology portfolio for Sequential Infiltration Synthesis (SIS) technology, which includes creation of ordered nanoscale domains by infiltration of block copolymers, use of SIS for advanced lithography and for enhancing multi-pattern lithography, and the application of SIS for creation of advanced sorbents.
Benefits: SIS is a versatile platform for the precise growth of inorganic materials within polymer templates. The polymer can subsequently be removed using standard techniques or retained to maintain desirable properties of the initial template. This foundational technique has a wide range of industrial uses including:
- Robust methods for photolithography, electron-beam lithography, and block copolymer lithography patterning, including high-aspect-ratio lithography.
- Synthesis of high-performance dielectrics.
- Efficient fabrication of nanostructured surfaces for use in anti-reflective or anti-fouling/self-cleaning coatings.
- Creation of advanced sorbents through infiltration of polymer foams.
Figure 1: Dense (<20nm) lines etched into a Si structure using SIS demonstrate its potential for high aspect-ratio (greater than 6:1) semiconductor lithography.
Figure 2: SIS technology has been applied to create advanced sorbents for oil recovery, starting from polymer foams.
- This software uses trained machine learning algorithms to classify features in a scanning tunneling microscopy (STM) topography, and feeds the results to a custom automated atomic manipulation subroutine, thereby circumventing the requirement for operatorIntellectual Property Available to License
Invention Opportunity & Solution
- This program will allow for the building of complex structures that were previously unattainable.
- The automation will save the operator time.
- Scanning probe microscopy software development