Dr. Yu-Qiao Song, Scientific Advisor at Schlumberger-Doll Research, presents "Magnetic Resonance for Chemical Analysis in Rocks" at the MIT Earth Resources Laboratory.

"Analysis of natural materials is often challenging, such as crude oils are consistent of a large number of different molecules of different sizes and functional moieties. Furthermore, unconventional reservoirs contain organic materials from solid form, extremely viscous, viscous to light oil and gas. The composition of complex mixtures is often analyzed using high-field magnetic resonance (MR) spectroscopy, which typically relies upon large and expensive superconducting magnets. Here we discuss methods based on low-field MR which can achieve portable and potentially low-cost applications. 

We will highlight the challenges in the practical implementation and analysis of data with limited signal-to-noise. We will discuss Bayesian approach to understand the uncertainty of the data analysis and the machine learning concept for realtime optimization of the data acquisition to achieve fast and robust measurements."