Periodic Reporting for period 2 - Hafnium (Faster and smarter chemical R&D with accurate physical property predictions)
Reporting period: 2021-04-01 to 2022-09-30
Accurate knowledge of properties of chemical compounds is essential for all chemical research. Without it, cutting-edge research becomes very expensive or impossible. However, experts estimate that we have experimental physical property data for less than 0.2% of relevant chemical systems — and most is for petrochemicals. There is almost no data for biochemicals, carbon capture molecules, electrolytes, or other chemistries that we need for the green transition.
Experimentation is accurate but slow and very expensive. On the other hand, all computer models for chemical properties are inaccurate, because chemical interactions are too multi-layered and complex for us to model, even with the much-vaunted “power of AI”. Choosing the right model for your chemistry, fitting it correctly to the experimental data, and ensuring that the data is actually sufficient and reliable is difficult, even for a physical property expert.
That is why Hafnium Labs has developed Q-props, a pioneering software that sets a modern gold standard for obtaining reliable physical properties for any chemistry. Q-props uses all available data and models to always give the best possible property prediction and information on how reliable each prediction is.
Before the Horizon 2020 project was launched, Q-props was running internally on Hafnium Labs’ cloud servers. The main objective of the project was to enable customers to run Q-props on their cloud and in their process simulators. The second objective was to optimize and simplify how customers work with their own valuable data in Q-props, while the third objective was to ensure that the software can be used by all chemical scientists and engineers, including those who lack specialist knowledge of chemical modelling.
Experimentation is accurate but slow and very expensive. On the other hand, all computer models for chemical properties are inaccurate, because chemical interactions are too multi-layered and complex for us to model, even with the much-vaunted “power of AI”. Choosing the right model for your chemistry, fitting it correctly to the experimental data, and ensuring that the data is actually sufficient and reliable is difficult, even for a physical property expert.
That is why Hafnium Labs has developed Q-props, a pioneering software that sets a modern gold standard for obtaining reliable physical properties for any chemistry. Q-props uses all available data and models to always give the best possible property prediction and information on how reliable each prediction is.
Before the Horizon 2020 project was launched, Q-props was running internally on Hafnium Labs’ cloud servers. The main objective of the project was to enable customers to run Q-props on their cloud and in their process simulators. The second objective was to optimize and simplify how customers work with their own valuable data in Q-props, while the third objective was to ensure that the software can be used by all chemical scientists and engineers, including those who lack specialist knowledge of chemical modelling.
Work progressed according to plan.
In the first project period, the major focus was on developing a version of Q-props that could be deployed on customer side and, thus, run in full confidentiality. Further, Q-props was optimized to make it more robust and enhance security when accessing and updating the tool.
In the second project period, the major focus was on obtaining proof points and executing demonstration projects with industrial partners to show the potential of using Q-props to solve a wide range of problems encountered in the chemical and energy industries.
This work has included developing different means for easy data integration and developing Q-props interfaces for specific use cases relevant for demonstration users.
Commercially, Q-props has evolved from a single product with a somewhat complex deployment to a suite of tools with each their clear target audience and deployment:
- Q-predict: Predicts pure and pseudo-pure compound properties
- Q-model: Develop, tune, and validate thermodynamic models
- Q-package: Reliable physical property package for any chemical system, integrating with process simulation tools
- Q-twin and Q-utility: Custom web-apps for a specific purpose, e.g. a digital twin for a specific product
- Q-screen: Identify compounds or mixtures with a desired set of properties
Summary of results from the project:
1) A step change improvement in the functionality of Q-props
2) A clear definition of the Q-props suite of tools, their value proposition, and how each is deployed
2) The company's workforce grew by more than 100% over the course of the project by the addition of top scientific and tech talent from Australia to Central America
3) Following the successful completion of the project, Q-props is now ready for deployment and a strong commercial plan is in place for going to market
In the first project period, the major focus was on developing a version of Q-props that could be deployed on customer side and, thus, run in full confidentiality. Further, Q-props was optimized to make it more robust and enhance security when accessing and updating the tool.
In the second project period, the major focus was on obtaining proof points and executing demonstration projects with industrial partners to show the potential of using Q-props to solve a wide range of problems encountered in the chemical and energy industries.
This work has included developing different means for easy data integration and developing Q-props interfaces for specific use cases relevant for demonstration users.
Commercially, Q-props has evolved from a single product with a somewhat complex deployment to a suite of tools with each their clear target audience and deployment:
- Q-predict: Predicts pure and pseudo-pure compound properties
- Q-model: Develop, tune, and validate thermodynamic models
- Q-package: Reliable physical property package for any chemical system, integrating with process simulation tools
- Q-twin and Q-utility: Custom web-apps for a specific purpose, e.g. a digital twin for a specific product
- Q-screen: Identify compounds or mixtures with a desired set of properties
Summary of results from the project:
1) A step change improvement in the functionality of Q-props
2) A clear definition of the Q-props suite of tools, their value proposition, and how each is deployed
2) The company's workforce grew by more than 100% over the course of the project by the addition of top scientific and tech talent from Australia to Central America
3) Following the successful completion of the project, Q-props is now ready for deployment and a strong commercial plan is in place for going to market
Digital design is increasingly important to chemical R&D, enabling faster development of greener and safer products and processes. But digital chemical designs are only as reliable as the physical properties of the molecules and mixtures you put into them. It really is “garbage in, garbage out”.
With its groundbreaking approach to always obtain the best possible physical properties, Q-props can speed up chemical R&D and help researchers and engineers make safer and smarter design decisions. This will help us reach ambitious climate and environmental goals while also improving the bottom lines of our customers.
With €350B spent on chemical R&D and process plants annually and with everything - from drugs that save our lives to energy technologies that will save our tomorrow - depending on it: Supporting researchers and engineers working with chemistry creates a massive impact and a brighter future for all.
With its groundbreaking approach to always obtain the best possible physical properties, Q-props can speed up chemical R&D and help researchers and engineers make safer and smarter design decisions. This will help us reach ambitious climate and environmental goals while also improving the bottom lines of our customers.
With €350B spent on chemical R&D and process plants annually and with everything - from drugs that save our lives to energy technologies that will save our tomorrow - depending on it: Supporting researchers and engineers working with chemistry creates a massive impact and a brighter future for all.