Development of a Nuclear Magnetic Resonance

Method and a Near Infrared Calibration Model

for the Rapid Determination of Lipid Content in
the Field Pea (Pisum sativum)

Presented By: -
Guided By: -
Ashish T S (322003)
Bhimanshu Kurre (322018) Dr. Bhaswati Bhattacharya
Assistant Professor, Department
Mukesh Dewangan (322007)
of Basic & Applied Science
M. Tech (FSQM)
Content
• Introduction
• HR-MAS NMR Spectroscopy
• Instrumental setup
• 1H NMR spectrum of the pea oil sample
• Quantification of Total Lipid Content Using 1H HR-MAS NMR
• NIR Spectroscopy
• Calibration and Validation Model Design
• Conclusions
Objective
• To develop a low-cost, rapid and non-destructive method for analyzing
pea lipids with no chemical modifications.
 Introduction
• Fatty acid content of field pea ranges between 0.9–5.0% of dry matter.
• Traditional method of analysis involves triacylglycerol hydrolysis,
esterification and gas chromatography.
• Modern methods include nuclear magnetic resonance (NMR) and near
infrared (NIR) spectroscopies
• Calibration methods have been developed using NIR spectroscopy for rapid
determination of seed contents, including lipids, starch, carbohydrates,
proteins, moisture, fatty acid composition, phytonutrients, and many other
components.
HR-MAS NMR Spectroscopy
• 1H high-resolution magic angle spinning.
• The 1H NMR spectrum of the pea oil contains 11 distinct signals of variable
intensity.
• The HR-MAS probe was highly efficient at averaging residual dipolar interactions
and variations in the bulk magnetic susceptibility of the samples.
Fatty acid composition
• The linoleic acid content of the tested pea seed varieties ranged between 13.89–
43.98%
• The highest linolenic fatty acid content measured was 26.62% (Sample F1) and
the lowest content measured was 20.81% (Sample 45760).
Instrumental setup
• High-resolution magic angle spinning (HR-MAS) NMR equipment
• Spectrometer - Bruker AVANCE III HD 600, operating at 600.17 MHz for 1H.
• 4-mm triple-resonance 1H/31P/13C high-resolution MAS probe with a Z-
gradient directed along the magic angle axis.
• 15 µL deuterium oxide containing 5 mM 3-(trimethylsilyl) propionic-2, 2, 3, 3-
d4 acid used as a standard.
• 7 mg of each sample were placed in insert and placed in rotor
• Total 128 scans were done with a 6 µs 90◦ pulse length, 8417.5 Hz spectral
width, 16k data points, a 0.97 s acquisition time, and a 1 second relaxation delay
 1H NMR spectrum of the pea oil sample
Signal Represents
A Bis-allylic hydrogen atoms
B β positions of methylene hydrogen atoms
D α positions of methylene hydrogen atoms
C α methylene hydrogen atoms in single double
bond
F the overlap of triplet signals, oleic acyl and
linoleic acyl groups
E methyl protons of ω-3 acyl groups.
G hydrogen atoms of olefinic groups
H olefinic hydrogen
atoms of the different acyl groups

1H NMR spectrum of the pea oil sample.

Quantification of Total Lipid Content Using 1H HR-MAS NMR
• Correlation was observed between the NMR and traditional methods
• The correlation coefficients for each method increased to 0.97 (R2 of 0.99) when
NMR data were corrected ( earlier 0.77).
• Correlation coefficients of 0.77 (R2 = 0.60), 0.56 (R2 = 0.47), and 0.78
(R2 = 0.62) were observed
• The correlation coefficients for each is increased to 0.97 (R2 of 0.99)
when NMR is used
• Statistical analyses showed no significant differences (p > 0.05)
between the corrected NMR data for NMR and the traditional method
data.
 NIR Spectroscopy
• Spectroradiometer – Field Spec 4 standard-Res
• ASD contact probe – High intensity contact probe
• 10mm scan spot size
• Light source – Halogen bulb
• 350 nm to 2500 nm
• Samples were rotated at 22 rpm
• For each samples 100 scans over a 17 seconds
• The spectra was analysed using UCal 4tm Custom
calibration software.
Contact probe
NIR spectra from six pea cultivars.
Graph of NIR versus reference values.
Calibration and Validation Model Design
• Partial least squares (PLS) regression analyses were run for datasets
that were preprocessed in the preceding section using UCal 4TM
Custom Calibration Software.
• Reflectance spectra were converted to absorbance by taking the
logarithm of the inverse of the reflectance values
• A leave-one-out cross-validation (LOOCV) was performed to generate
the standard error of cross-validation (SECV) statistics, and each
sample was used as a singular test set against the model built from the
remaining training set.
Conclusions
• The results showed that NMR and NIR are low-cost, rapid, and non-
destructive methods for analyzing lipids in pea.
• NMR technology can be used for quantitative and qualitative analyses
of the fatty acids present in oilseeds.
• These data show a high correlation between the NMR method and the
traditional method in predicting the lipid content present in seeds.
• PLS regression analyses confirm the possible application of NIR
analysis for the rapid determination of pea lipid content.
Reference
• Addo, P. W., Ossowski, P., Macpherson, S., Gravel, A. E., Kaur, R.,
Hoyos-Villegas, V., Singh, J., Orsat, V., Dumont, M. J., & Lefsrud, M.
(2022). Development of a Nuclear Magnetic Resonance Method and a
Near Infrared Calibration Model for the Rapid Determination of Lipid
Content in the Field Pea (Pisum sativum). Molecules, 27(5).
https://doi.org/10.3390/molecules27051642
THANK YOU