No abstract

No abstract

On-site material inspection and quality analysis of food and agricultural produce require portable sensing systems. We report the development of a miniaturized spectrometer with an integrated light source operating in the visible and near-infrared range, for chemometrics based material-sensing applications. The proposed system uses off-the-shelf light source and detector. The electronic circuit is designed, developed, and tested in-house. To validate the system’s usability, a set of classification experiments are carried out with measured spectra from culinary white powders and medicinal pills. Several classification algorithms are used to build predictive models and the best-suited ones give prediction accuracies of 80% and 92.6% respectively. A regression model built to estimate the curcumin content in turmeric shows a coefficient-of-determination of 0.97 for prediction. With more than 90% repeatability in the measured reflectance spectra, robustness of the device is demonstrated. Realization of a portable spectrometer, along with a framework for building appropriate prediction models, is expected to spur the development of point-of-use material sensing in the Vis-NIR range.

For quantification of curcumin content in turmeric, a low-cost multivariate-analysis-based sensing system is desired. It can be realized by exploiting the spectra in the visible region, which enables the use of off-the-shelf, relatively inexpensive light sources and detectors. To address this, we propose a novel decision-tree method for improved prediction accuracy. Two sets of models with PLSR algorithm are developed with the measured reflectance spectra from 66 turmeric samples in the range of 360–750 nm, and their respective curcuminoids content are quantified by HPLC. A suite of a coarse-model for initial prediction of turmeric samples in the broad range of 1%–4%, and five finer-models for subsequent prediction (in the ranges 1%–2%, 2%–3%, 3%–4%, 1.5%–2.5%, and 2.5%–3.5%) constitute the proposed decision-tree approach. The method’s efficacy is substantiated from an improved coefficient of determination (R 2) for the finer models (0.90–0.96) as compared to the coarse-model’s 0.92. This is further corroborated with lower RMSECV of 0.06–0.13 and an RMSEP of 0.15–0.25 for finer models, as compared to 0.219 and 0.45 for the coarse model, respectively. Testing reveals that the method results in 46% reduction in prediction error. Realization of a robust prediction approach in the visible range sets the stage for the development of cost-effective field-deployable devices for on-site measurement of curcumin.

We experimentally study test structures of more than 100 microelectromechanical systems (MEMS) beam resonators with clamped ends and residual stresses varying from highly tensile to compressive loads beyond buckling and provide experimental verification of some key theoretical results reported in the literature. We compare the theoretically predicted natural frequencies over a large range of residual stresses that make the one-dimensional micro-mechanical resonators behave like beams or strings, depending on the relative magnitude of the effective axial load and the flexural stiffness. In particular, we measure the natural frequencies of the first four modes of buckled beams to show the drastically different behavior of beams under post critical buckling load from those under tension and, for the first time, present experimental evidence of invariance of even modes to compressive residual stresses in microscale beams. We then derive the sensitivity of these modes to residual stresses and discuss the consequences of such sensitivity on sensing applications along with recommendations on how to engineer the required level of residual stresses.