Application of qNMR:Quality control of organophosphorus pesticides

  • Summary

NM090010E

Quantitative NMR (qNMR) is widely applicable for determining purity, since the method does not require a target compound of known purity as a reference.
This Note introduces an application of qNMR to quality control, where purity of commercial organophosphorus pesticides (isoxathion oxon) is determined.

Target compound: Isoxathion oxon

Two standard samples for pesticide residue test, isoxathion oxon, were examined. Information of the products is listed in Table 1.

Isoxathion oxonIsoxathion oxon

Table 1. Compound information attached to the products
Sample Lot Purity(%)* Appearance m.p.(℃)
starndard
isoxathion
oxon
1 96.9 Yellow-brown crystalline powder 49.5
2 98.9 White crystalline powder 51.7
*Determined via GC analysis

qNMR reference

Hexamethyldisilane (HMD) was used as a qNMR reference. HMD gives an NMR signal at 0 ppm but is not SI-traceable, and so its concentration was calibrated with a certified reference material, diethyl phthalate. Thus, SI-treaceability of qNMR measurements using HMD is established.

qNMR measurement conditions

Table 2 shows NMR measurement conditions. Important point is setting a recycle delay to be more than five times of longitudinal relaxation time (T1), preventing from saturation. For the details of quantitative analysis using NMR, please refer to technical information of JEOL homepage.

Table 2. Measurement conditions
Parameter of JNM-ECA600
Observed nucleus 1H
Observed range -5ppm~15ppm
Data points 32K
Digital filter ON(8times)
Recycle delay 60s
Flip angle 90°
Scans 8 scan
Temperature 25℃
Spinning OFF

For the details of quantitative analysis using NMR, please refer to technical information of JEOL homepage.

Results and discussion

Fig. 1  qNMR spectra of two isoxathion oxon samples
Fig. 1 qNMR spectra of two isoxathion oxon samples

Figure 1 shows NMR spectra of isoxathion oxon Lot.1 and Lot.2. The former spectrum exhibits many impurity peaks (*), suggesting low grade of the product. The concentration deduced from qNMR analysis is listed in Table 3. The result for Lot.2, 98.5±1.3 %, agrees with that in the attached information, while 75.4±0.2 % for Lot.1 is different from 96.9 % in the attached information which is, therefore, considered to be inappropriate. These results are consis- tent with the concentration ratio determined by GC/MS, although GC/MS has uncertainty due to response factors. Thus, qNMR is promising in wide research fields including quality control.

Table 3.  Results of quantification analysis of qNMR and GC/MS
NMR  peak (ppm) Average GC/MS
Area ratio
a
7.85
b
7.49
c
6.81
d
4.25
e
1.32
Lot.1 75.3%
(0.8)a)
75.1%
(0.9)a)
75.5%
(0.7)a) 
75.8%
(0.9)a) 
75.4%
(0.9)a) 
75.4%
(0.2)a)
75.3%b)
Lot.2 99.0%
(0.3)a)
99.3%
(0.5)a)
96.3%
(0.2)a)
99.4%
(0.6)a)
98.6%
(1.1)a) 
98.5%
(1.3)a)
98.5%b)
a)Values in ( ) indicate standard deviation.
b) Value for Lot.2 is assumed to be the same as that from NMR, and value for Lot.1 is calculated using concentration ratio.

Reference

M. Tahara, N. Sugimoto,T. Suematsu et. al, Jpn.J.Food Chem.Safety, 2009, Vol.16(1), 28-33

SEARCH APPLICATIONS

Related Products