研究目的
To investigate the feasibility of using NIR and FT-IR fingerprinting to detect economically motivated adulteration in black pepper, including adulterants such as papaya seeds, chili, black pepper husk, pinheads, and spent materials.
研究成果
FT-IR and NIR spectroscopy combined with chemometrics are capable of detecting adulteration in ground black pepper, including foreign plant materials and non-functional pepper materials. The method shows high potential for rapid screening, with AUROC values of 0.98 for both techniques. Further validation and system challenges are needed to implement this approach in routine testing.
研究不足
The study is a feasibility study; the model may need extension with more adulterants and black pepper samples from different regions. Long-term stability tests and robustness checks (e.g., different operators and instruments) are recommended for future work. The sample set represents only six of the twenty-six countries that cultivate black pepper, which may limit generalizability.
1：Experimental Design and Method Selection:
The study used a non-targeted fingerprinting approach with NIR and FT-IR spectroscopy combined with chemometrics (PCA and OPLS-DA) for classification.
2：Sample Selection and Data Sources:
A total of 115 black pepper samples from various countries (e.g., India, Vietnam) and 11 adulterant samples (husk, pinheads, spent material, papaya seeds, chili) were collected. Adulterated samples were prepared by spiking black pepper with adulterants at 10-40% levels.
3：List of Experimental Equipment and Materials:
Equipment included a Thermo iS50 spectrometer for FT-IR and NIR data acquisition, a ball mill (Planetary Ball Mill: PM-100 from Retsch) for sample grinding, and SIMCA 15 software for chemometric analysis. Materials included black pepper and adulterant powders.
4：Experimental Procedures and Operational Workflow:
Samples were ground to a homogeneous powder, spectra were acquired with specific parameters (e.g., 32 scans, 4 cm-1 resolution for FT-IR), and data pre-processing (e.g., SNV, derivatives) was applied. Models were developed and validated using internal and external validation sets.
5：Data Analysis Methods:
Chemometric analysis involved PCA for exploratory analysis and OPLS-DA for binary classification, with performance evaluated using R2, Q2, sensitivity, specificity, and ROC curves.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
