Update on Comparing Solvent Cleaning Treatments Using NMR

It is hard to believe that we have already completed two weeks here in the lab, but things are moving at a steady pace and a lot of work has been done in terms of experimenting with techniques which will be used to analyze paint samples that have been treated with solvents. While not directly moving forward with cleaning techniques on paint, determining the accuracy of these data gathering techniques is crucial because we need to be certain that we can retrieve the information that we need from the data we procure.

The first technique which was explored and improved was increasing the spatial resolution with which samples could be analyzed. This was done by running profiles on a sample of cream using our smaller magnet. The cream produces a very large signal which makes it easy to see compared to the noise, or the signal from the environment. The spatial resolution was found by determining the range of the interface between the cream and the glass slide, which produces no signal, on which it rested. The resolution was clear down to one millimeter. Knowing that this technique can provide a depth measurement accurate to 1 millimeter will be helpful when a sample of paint needs to be analyzed for how thick it is, or to what depth withing the paint a solvent has ingressed.

The fitting of T2 curves to the decaying signal of samples was explored to verify an interface between a sample with two signal giving components. A sample of isopropanol and Linseed Oil was used to create a profile and the data was processed to determine if it could be shown that there was a definite interface between the non-mixing compounds. It was clear to see that there was a significant change in signal indicating the existence and position of the interface. A T2 curve was then applied to the breadth of the data to see if the estimated values compared to what we expected to see based on earlier T2 experiments on the individual compounds. The comparable estimates confirmed that our methods of profiling provide accurate and useful information and that these methods will be applicable to the comparison of paint samples in the weeks to come.

We also ran diffusion tests on water to determine the actual gradient strength of the magnetic field produced by the magnet being used. Using an equation with certain standard constants for water and the variables pulled from the data after the diffusion test on water, the gradient was found to be 1016 MHz/m. Knowing this will allow us to make more accurate calculations in the future if we need to determine the diffusion constant of a sample whose constant is not already known.

The confirmation of the validity of these techniques is a great step forward in this project. Knowing that these techniques are viable will significantly cut down on experiment time when using the magnet to extract data on a sample. The testing of these methods also called for a large amount of data processing which, while time consuming at first, now means that we have pre-written codes for processing the same type of data which means we can do near real-time analysis on the samples we run experiments on. I look forward to applying what I have learned in the past two weeks to comparing the cleaning methods this project is interested in.