A field-invariant method for quantitative analysis with benchtop NMR

Improving the Accuracy of Model-based Quantitative NMR

“…On the other hand, relative intensities b p of peaks pertaining to the same chemical necessarily remain constant, as they are defined by the atomic composition of the molecule. In the present work, we use the quantum mechanical approach for modelling the signatures of chemical species (Matviychuk et al, 2019). It allows us to minimize the number of free parameters and produce relevant model spectra at any field strength of the spectrometer.…”

“…Higher-order coupling between neighboring and distant protons often cause different 1 H peaks to show different asymmetric distortions due to separation of transition resonances (Kuprov et al, 2007). Quantum mechanical models were found useful for describing such data but also can not guarantee the perfect fit of complex spectra (Tiainen et al, 2014;Matviychuk et al, 2019).…”

“…Using these datasets, we estimate the mole fractions of each chemical species in the mixtures and compare them with the gravimetric values. To define the models for all chemical species, we use their complete quantum mechanical formulations and find the corresponding chemical shifts and J-couplings using the high-field data (Matviychuk et al, 2019). We use these parameters to initialise corresponding models at the lower field strength of the benchtop instrument and then refine them by minimizing Eq.…”

“…1). This allows us to test various aspects of our model, including the quantum mechanical formulation of coupled spin systems (see Matviychuk et al (2019) for more detail). Since the true ratios of peak intensities are known and fixed, by estimating them separately -as if the peaks belonged to several chemical species in unknown concentrations -we can unequivocally compare different qNMR methods in terms of their accuracy.…”

“…The reduction of a spectrum to a frequency-intensity table of peaks (Krishnamurthy, 2013) allows for easier automation of post-processing tasks and simplifies the analysis of large arrayed datasets (Kriesten et al, 2008;Alsmeyer et al, 2004). Finally, quantum mechanical formulations minimize the number of free model parameters and are inherently invariant with respect to the spectrometer field strength (Kuprov et al, 2007;Tiainen et al, 2014;Dashti et al, 2017); they enable the analysis of highly complex low-resolution spectra acquired on medium-field benchtop instruments and are found successful in modern practical applications (Matviychuk et al, 2019).…”