Lab professionals are under increasing pressure to reduce waste, conserve energy, and adopt more sustainable workflows, driving urgent interest in green labs and lab sustainability. At Lab Manager’s 2025 Green Labs Digital Summit, running May 13–14, 2025, experts will share actionable strategies to help labs meet these goals through smarter design, procurement, and operations.
One featured session, Laboratory of the Future: Sustainable Practices and AI/ML-Driven Workflows, will be led by Rigoberto Advincula, researcher at Oak Ridge National Laboratory and professor in the department of chemical and biomolecular engineering at the University of Tennessee. His session explores how artificial intelligence (AI), machine learning (ML), automation, and greener chemistry can work in tandem to lower labs’ environmental footprint while improving safety and efficiency.
Get training in Lab Leadership Basics and earn CEUs.
One of over 25 IACET-accredited courses in the Academy.
Lab Manager spoke with Advincula about how these tools are reshaping lab operations—and what lab managers should be doing now to prepare for a more sustainable future.
Q: Greener chemistry methods can help labs reduce costs, minimize waste, and lower environmental impact. What are some of the most effective sustainable materials and practices that labs should adopt today?
A: The laboratory is a place to practice scientific methods; it is an ecosystem for doing experiments, characterization, and discovery. By also practicing sustainability and utilizing tools to reduce energy use, it becomes a place that combines safety and efficiency.
Greener chemistry methods can be practiced in the following ways:
Interested in lab design?
- Reduce the use of organic solvents, replacing them with more recoverable options or aiming for solventless reactions, methods, or purification.
- Minimize heat and water usage by implementing recirculating cooling baths.
- Conduct energy audits, including calculating and posting electricity consumption per unit of operation.
- Limit single-use plastic waste by reusing labware or selecting recyclable alternatives.
- Use machine learning tools to design experiments and minimize trial-and-error approaches.
- Maintain constant awareness of opportunities to reduce waste and conserve energy.
Q: AI and ML can potentially minimize the number of experiments needed for discovery. Can you explain how these technologies optimize formulation and manufacturing methods and what efficiency gains labs can expect?
A: The design of experiments and use of statistical methods have been carried out in laboratories that essentially target critical experiments that can be answered faster, for example, for optimization or property development. Matrix design optimization and principal component analysis are key. Various algorithms based on linear regressions, decision trees, and the use of neural networks are powerful tools for obtaining experimental efficiencies and, eventually, sustainable laboratory and greener protocols.
Lately, AI has been at the forefront of data analytics and generating correlational and causational knowledge. It can also point towards new research directions, including bio-inspired science. In the laboratory, ML, which many have been practicing as a subset of AI, will lead to better experiments and higher throughput experiments that eventually save time, effort, and resources.
Q: How do automation and digital tools complement sustainability efforts in the lab? Can you share an example of a workflow where AI/ML and automation have led to safer operations and improved sustainability?
A: Domain-specific AI and ML will lead to better efficiencies in formulation, manufacturing, and laboratory methods. This can be done first by developing digital twins of experiments that will be carried out and applying theory that establishes the basis for answering the hypothesis and guides the design of the experiment. This also targets the number of repetitions or feedback-loop guided experiments. Using automated and self-driven labs—for example, robotics, flow chemistry, sensors, and computational methods or edge computers that both control and monitor reactions or characterization experiments—makes it possible to have more efficient labs that carry fewer wasteful experiments, and thus saves time, resources, and minimizes disposable waste. We have demonstrated it in polymer synthesis and copolymerization, where we obtained desirable properties with fewer experiments and better results.
Q: Due to perceived complexity or cost, many lab managers might hesitate to integrate AI/ML-driven workflows. What are the most significant barriers to adoption, and how can labs overcome them?
A: Cost can be an issue, but a bigger barrier is the lack of talent or skills among scientists and engineers to put together protocols or teams to carry out ML-directed workflows and use automation and mechatronic tools. The concept has been proven, and steps or even simpler protocols can now be adopted to make a laboratory more AI/ML-driven. A theory or ML-optimized guided experiment saves time, resulting in significant cost savings and use of resources, or minimizing waste. However, the caution is that it may not be an ideal environment for serendipity or observational opportunities for a curious mind. The balance is that these steps are real advantages for most routine experiments for formulation, characterization, and well-designed synthesis protocols that need experimental or empirical validation.
Q: In your view, what does the laboratory of the future look like? How will sustainable materials, automation, and AI/ML-driven workflows redefine labs' operations in the next five to 10 years?
A: Many things that can happen in the next five to 10 years will affect how we do experiments and operate laboratories. Although cost and lack of talent are current issues, the more AI/ML is adopted in product instruments and, more importantly, the education and training of scientists and engineers, the more we will see this ecosystem grow. The growing skills among scientists, engineers, and teams to carry out ML-directed workflows will demand more software, automation, and mechatronic tools. A theory or ML-optimized guided experiment will save time and resources or minimize waste. With current green practices and waste minimization, including energy and heat audits, a modern lab or lab of the future will have real advantages in the laboratory and society.
To hear more from Advincula and explore additional strategies for advancing lab sustainability and safer, more efficient lab environments, register for Lab Manager’s 2025 Green Labs Digital Summit, taking place May 13–14. Advincula’s webinar, Laboratory of the Future: Sustainable Practices and AI/ML-Driven Workflows, will be presented live on May 14 from 12:30 to 1:30 p.m. ET.
