Rapid quantitative detection of mineral oil contamination in vegetable oil by near-infrared spectroscopy

Mantong Zhao; Dawei Zhang; Lulu Zheng; Otto Condliffe; Yi Kang

doi:10.3788/COL202018.043001

Journals >Chinese Optics Letters >Volume 18 >Issue 4 >Page 043001 > Article

Chinese Optics Letters
Vol. 18, Issue 4, 043001 (2020)

Rapid quantitative detection of mineral oil contamination in vegetable oil by near-infrared spectroscopy

Mantong Zhao¹, Dawei Zhang^1、2、*, Lulu Zheng¹, Otto Condliffe³, and Yi Kang¹

Author Affiliations

¹Engineering Research Centre of Optical Instrument and System, Ministry of Education, Shanghai Key Laboratory of Modern Optical System, University of Shanghai for Science and Technology, Shanghai 200093, China

²Shanghai Institute of Intelligent Science and Technology, Tongji University, Shanghai 200092, China

³Sino-England International College, University of Shanghai for Science and Technology, Shanghai 200093, China

show less

DOI: 10.3788/COL202018.043001 Cite this Article Set citation alerts

Mantong Zhao, Dawei Zhang, Lulu Zheng, Otto Condliffe, Yi Kang. Rapid quantitative detection of mineral oil contamination in vegetable oil by near-infrared spectroscopy[J]. Chinese Optics Letters, 2020, 18(4): 043001 Copy Citation Text

EndNote(RIS)

BibTex

Plain Text

show less

Fig. 1. Transmittance spectra of rapeseed oil (B1 brand) and mineral oil.

Download full size | View in the Article

PC2 × PC1 score plot for the overall set of 120 rapeseed–mineral oil samples.

Fig. 2.

PC 2 \times PC 1

score plot for the overall set of 120 rapeseed–mineral oil samples.

Download full size | View in the Article

Fig. 3. Calibration results of (a) DOSC-SPA-PCR and (b) DOSC-SPA-PLS. Fifteen samples per concentration were used for modeling, and there were 120 samples in total.

Download full size | View in the Article

Fig. 4. Prediction results of (a) DOSC-SPA-PCR and (b) DOSC-SPA-PLS. Ten samples per concentration were used for prediction, and there were 80 samples in total.

Download full size | View in the Article

Fig. 5. Prediction results of (a) DOSC-SPA-PCR and (b) DOSC-SPA-PLS for B2 rapeseed oil.

Download full size | View in the Article

Set	Brand	Concentration ( $g \cdot {kg}^{- 1}$ )
Calibration	B1	1	0.5	0.2	0.1	0.08	0.05	0.03	0.01
Prediction	B1	1	0.5	0.2	0.1	0.08	0.05	0.03	0.01
Prediction	B2	0.8				0.1

Table 1. Sample Information in Calibration and Prediction Sets

View in the Article

No.	Wavelength (nm)	No.	Wavelength (nm)
1	910	19	1906
2	930	20	1922
3	1066	21	1930
4	1614	22	1970
5	1674	23	2026
6	1690	24	2030
7	1706	25	2050
8	1710	26	2066
9	1734	27	2078
10	1746	28	2090
11	1750	29	2098
12	1754	30	2106
13	1782	31	2110
14	1790	32	2114
15	1814	33	2122
16	1850	34	2126
17	1866	35	2134
18	1898	36	2150

Table 2. Effective Wavelengths Selected by SPA

View in the Article

Spectral Set	Model	Variable Number	R	RMSE
X	PCR	311	0.991	0.009
X	PLS	311	0.983	0.009
$X^{DOSC}$	PCR	311	0.995	0.009
$X^{DOSC}$	PLS	311	0.985	0.008
X (SPA)	SPA-PCR	36	0.996	0.006
X (SPA)	SPA-PLS	36	0.988	0.007
$X^{DOSC}$ (SPA)	DOSC-SPA-PCR	36	0.998	0.005
$X^{DOSC}$ (SPA)	DOSC-SPA-PLS	36	0.989	0.006