A NIR Photometer Prototype With Integrating Sphere for the Detection of Added Water in Raw Milk
| Autor: | Lucas da Silva Dias, José Cláudio da Silva Júnior, Ana Lucia de Souza Maudeira Felicio, José Alexandre de França |
|---|---|
| Rok vydání: | 2018 |
| Předmět: |
Sample (material)
020208 electrical & electronic engineering 010401 analytical chemistry Near-infrared spectroscopy 02 engineering and technology Photometer 01 natural sciences 0104 chemical sciences law.invention Photodiode Freezing point Integrating sphere law 0202 electrical engineering electronic engineering information engineering Calibration Environmental science Sample preparation Electrical and Electronic Engineering Instrumentation Remote sensing |
| Zdroj: | IEEE Transactions on Instrumentation and Measurement. 67:2812-2819 |
| ISSN: | 1557-9662 0018-9456 |
| DOI: | 10.1109/tim.2018.2829398 |
| Popis: | The freezing point is the main test for the detection of added water in raw milk. However, other contaminants can be used to mask fraud in the cryoscope. The near infrared spectroscopy has proved to be a workaround in the determination of water addition, since it is a fast, nondestructive, and widespread method for compounds identification, which reduces the risk of being deceived. This paper presents a prototype for raw milk analysis that identifies the added water. A sample preparation methodology was elaborated in order to avoid a high scattering of the infrared light by fat globules. The sampling method is based on diffuse reflectance associated with a low-cost integrating sphere, which avoids the expensive commercial solutions. The developed sphere presents a reflectance index of 88% in the near infrared response region. The prototype uses LEDs as infrared light sources and an In-Ga-As-Sb photodiode for detection. The calibration was performed from a set of samples with different adulterations, later a new set was tested to validate the model created by the estimator. A coefficient of determination (R2) equals to 0.9562 was obtained. In the validation step, the root-mean-squared error of prediction was 0.01794. Therefore, the prototype showed that it could determine the concentration of water in the sample. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|