Automated Data Processing of Nanalysis NMR Pro 60 MHz NMR spectrometer for pseudo-2D fast-acquisition protocol

The described here acquisition and data processing approach relies on the use of MatLab software. It enables one to acquire NMR spectra in an automated fashion dynamically as fast as every 0.2 seconds, which is important in the context of dynamic studies with fast relaxing species. The processing code was developed by Dr. Jonathan Birchall and validated by Dr. Nuwandi Ariyasingha for studies of hyperpolarized propane gas using 1.4 T Nanalysis NMR Pro spectrometer for the following chemical reaction:

 

 

 

 

 

 

 

Zipped code with instructions and example data set can be found here.

The instructions manual alone can be found here​.

In the nutshell, this acquisition mode perform ultra-fast acquisition of n NMR spectra (e.g. n =64) with specified delay between each acquisition. The data is recorded in n files, where each n-th file is a sum of all acquisitions to that time point. For example, 1st file has the data of the first scan, the 2nd file has the sum of the data of the first and the second file, the n-th file has the sum of all the data acquired. These recorded files cannot be used directly and therefore need to be de-accumulated. 

The automated MatLab code first opens the data file ( .dx) using jcampread function of the Matlab software. Next, the data is de-accumulated as described above. Each NMR spectrum is processed from FID data to frequency spectrum (using a set of functions kindly provided by Dr. Zhehong Gan). Each de-accumulated NMR spectrum is saved as raw FID data in the form of .csv file. Then, the software prints each processed spectrum into a PDF file. Examples below show the examples of processed NMR spectrum of hyperpolarized propane (left) and thermally polarized spectrum (right). The thermally polarized spectrum is employed as a signal reference to compute the signal enhancement values (SE) for Ha and Hb. Note: individual NMR spectrum of hyperpolarized propane is shown on the left, and 64-scan-spectrum of the thermal signal reference is shown on the right (device is in averaging mode).

 

 

 

 

 

 

 

The intensity of Ha and Hb hyperpolarized resonances is next integrated and plotted as a function of time in the following plot below (left). A specified range of the intensities is also selected for relaxation analysis in the following plot below (right)

This Matlab code can also be fine-tuned for many other uses for automated data processing. Use at your own risk. You are welcome to re-use and adapt this code, but please, consider citing the following paper, where the core of this code has been developed for data processing of low-field NMR spectroscopy of hyperpolarized Xe-129: Birchall, J. R.; Irwin, R. K.; Chowdhury, M. R. H.; Nikolaou, P.; Goodson, B. M.; Barlow, M. J.; Shcherbakov, A.; Chekmenev, E. Y., Automated Low-Cost In Situ IR and NMR Spectroscopy Characterization of Clinical-Scale 129Xe Spin-Exchange Optical Pumping. Anal. Chem. 2021, 93 (8), 3883-3888: https://pubs.acs.org/doi/abs/10.1021/acs.analchem.0c04545

The following software is required (academic pricing is provided ca. 2021):

Matlab license (ML): perpetual $500

Bioinformatics Toolbox (BI): $200

Curve Fitting Toolbox (CF): $200

Statistics and Machine Learning Toolbox (ST): $200

heterogeneous pairwise parahydrogen addition to propylene
Processed NMR spectra
Build-Decay-T1-01.jpg