Speaker
Description
Serpentinization of ultramafic rocks offers a promising carbon-negative pathway for in-situ geologic hydrogen generation. By reacting water with magnesium-rich minerals like olivine, this process yields molecular hydrogen ($H_2$) and can simultaneously sequester carbon dioxide through mineral carbonation. However, the pore-scale mechanisms governing fluid–mineral interactions, mineral expansion, and gas phase evolution remain poorly understood due to the lack of high-resolution spatial and temporal data. This study presents a novel micromodel platform enabling real-time visualization of serpentinization and hydrogen evolution under controlled laboratory conditions.
The experimental platform utilizes a silicon wafer etched with a complex sandstone-inspired flow pattern through photolithography. To replicate the mineralogy of ultramafic reservoirs, the etched channels were functionalized with olivine sand, creating a reactive "fracture-on-a-chip." The micromodel was sealed with borosilicate glass and mounted in a custom-engineered aluminum holder designed to withstand temperatures up to 250°C and pressures up to 560 psi. The system was integrated with a high-precision ISCO pump and a back-pressure regulator to inject pre-degassed brine. A rigorous degassing protocol was implemented to ensure that any observed gas phases resulted from chemical reactions rather than liberated dissolved air.
Moving beyond initial 80°C proof-of-concept tests, experiments conducted at 110°C and 655 kPa (approx. 95 psi) provided critical insights into reaction kinetics and phase behavior. Under these elevated conditions, real-time reflected light microscopy captured the emergence of visible bubbles within the olivine-functionalized pores. These bubbles, a direct observation of gas evolution, were seen nucleating at mineral-fluid interfaces and coalescing within the flow channels. This phenomenon is vital for understanding how hydrogen gas might migrate or become trapped within the subsurface.
To quantify these dynamics, the machine learning tool Ilastik was employed for image segmentation. While the software was highly effective at identifying olivine grains and tracking the morphological evolution of the micromodel structure, it faced challenges in distinguishing gas bubbles from liquid brine in deeper or shadowed channels. Post-experimental characterization provided definitive evidence of the serpentinization reaction. Scanning Electron Microscopy (SEM) analysis of the olivine grains recovered from the 110°C tests revealed the development of secondary mineral phases. The SEM imagery showed the formation of wave-like structures consistent with proto-serpentine formation. These structures were found coating the original olivine surfaces, confirming that the micromodel platform successfully facilitates and captures the chemical transformation of ultramafic rock.
These results inform improvements for future tests, including the use of fluorescent imaging or micro-computed tomography (micro-CT) to map chemical changes and gas production in situ. By providing high-resolution data on mineral precipitation and gas evolution, this micromodel platform enables mechanistic investigations of natural hydrogen systems. Furthermore, the measured fluid and rock properties serve as critical inputs for pore- and reservoir-scale simulations, helping project collaborators identify the optimal conditions for carbon-negative hydrogen production in global ultramafic reservoirs.
| Country | United States |
|---|---|
| Student Awards | I would like to submit this presentation into both awards |
| Acceptance of the Terms & Conditions | Click here to agree |
