University of Toronto
We build the labs that will discover tomorrow's formulations, today.
Automation, robotics, and machine learning applied to formulation science, electrochemistry, and advanced materials research.
Materials Science
Engineering matter at the molecular level for performance-driven applications.
Research
Our work sits at the intersection of automation, AI, and chemistry, building the infrastructure for formulation discovery at scale.
We built RAISE, a network of self-driving laboratories that uses an agentic system to predict and discover liquid-surface interfacial behaviour, such as wetting, spreading, and coverage. These properties are critical across agriculture, contact lenses, eye drop formulations, medical devices, personal care, and coatings.
We advance the discovery of polymeric soft materials and complex formulations by developing new formulation strategies and creating platforms that rapidly and autonomously characterize their properties, for applications in drug delivery, agriculture, and personal care.
We combine biomaterials engineering, high-throughput experimentation, and AI-guided formulation design to develop strategies that preserve the stability and function of isolated mitochondria, enabling more effective mitochondrial delivery and transplantation.
We develop electrochemical technologies for critical mineral recovery through Fe³⁺ regeneration, hydrogen and NaOH co-production, and industrial wastewater treatment to enhance biodegradability, biogas production, and resource recovery.
Tools such as polarized light microscopy use birefringence to visualize the microstructures of soft materials. Understanding and controlling these structures enables the engineering of desirable formulation properties, including controlled release, stability, and flow behavior.
We orchestrate self-driving laboratories into fully closed-loop systems, linking instruments, data, and decision-making through Bayesian optimisation and agentic AI to accelerate scientific discovery.
Selected Works
Peer-reviewed research from the Mixture Lab. Click any card to view the full publication.
RAISE: a self-driving laboratory for interfacial property formulation discovery
A robot that designs its own experiments: RAISE mixes surfactant formulations, photographs the droplets, measures how they wet a surface, and asks AI what to try next—automatically. It hits precise wettability targets for medical devices, coatings, or textiles while minimizing surfactant use and cost, running at roughly one measurement per minute.
M. Nazeri et al. · 2026
In situ formed Se–TiO2 as a highly reusable photocatalyst for selenium reduction and removal from industrial wastewater
During selenium removal from mining wastewater, TiO2 photocatalysts accumulate selenium on their surface, forming a heterojunction that actually improves performance. The catalyst maintained more than 99.3% selenium removal across 15 consecutive treatment cycles without cleaning or regeneration—demonstrating practical reusability that matters for industrial-scale water treatment deployment.
A. Ngan et al. · 2025
Leveraging flexible pipette-based tool changes to transform liquid handling systems into dual-function sample preparation and imaging platforms
Lab robots typically only move liquids—they can't watch what's happening as they work. This 3D-printed adapter clamps a USB camera alongside an Opentrons pipette, letting the same robot both dispense and image samples in real time. A single machine can now capture soft materials—like gelling hydrogels—forming during preparation.
M. Nazeri et al. · 2025
The Team
A multidisciplinary team of researchers, engineers, and scientists working at the frontier of automated discovery.
PhD Candidate
.png&w=3840&q=75)
PhD Candidate
MASc Student
Press & Updates
Lab updates, press coverage, and recent highlights from the Mixture Lab.
A pleasure hosting colleagues from AIST to discuss self-driving labs and formulation research
Hosted Don Futaba, Woosuck Shin, and Umi Shin from AIST in Toronto to discuss self-driving labs, automation, and the future of materials and formulation research.
Frank Gu · Oct 2026
Share
Acceleration Consortium hosts FedDev Ontario for a tour of self-driving lab research
Steve Masson, Acting VP at the Federal Economic Development Agency for Southern Ontario (FedDev Ontario), visited the Acceleration Consortium for a briefing and tour of Prof. Frank Gu's self-driving lab.
Frank Gu · Oct 2026
Share
Great to welcome Artem Mishchenko from Google DeepMind to the Acceleration Consortium
Artem Mishchenko from Google DeepMind visited the Acceleration Consortium for a seminar on autonomous experimentation and materials discovery.
Frank Gu · Sep 2026
Share
Blog
Automating the “thumb and forefinger test” to study how formulation impacts material viscoelasticity.
September 4, 2026 · Frank Gu
ShareJoin Us
The Mixture Lab is actively recruiting at all levels. We value curiosity, rigour, and the ability to build things that work.
To apply, send your CV and a brief note about your research interests to f.gu@utoronto.ca.
We are looking for motivated PhD candidates with backgrounds in chemical engineering, chemistry, materials science, or related fields. Experience with automation, programming, or machine learning is a plus.
Exceptional postdoctoral candidates with a strong publication record and interest in automated discovery are encouraged to reach out directly.
Undergraduate students interested in gaining research experience are welcome to apply. Positions are available through USRA and direct arrangements.
Contact
Whether you're a prospective student, collaborator, or industry partner. We'd love to hear from you.
Location
Department of Chemical Engineering & Applied Chemistry
University of Toronto
Toronto, Ontario, Canada
