Partial Least Squares Factors Research Articles

Potential of a Newly Developed High-Speed Near-Infrared (NIR) Camera (Compovision) in Polymer Industrial Analyses: Monitoring Crystallinity and Crystal Evolution of Polylactic Acid (PLA) and Concentration of PLA in PLA/Poly-(R)-3-Hydroxybutyrate (PHB) Blends

This study was carried out to evaluate a new high-speed hyperspectral near-infrared (NIR) camera named Compovision. Quantitative analyses of the crystallinity and crystal evolution of biodegradable polymer, polylactic acid (PLA), and its concentration in PLA/poly-(R)-3-hydroxybutyrate (PHB) blends were investigated using near-infrared (NIR) imaging. This NIR camera can measure two-dimensional NIR spectral data in the 1000-2350 nm region obtaining images with wide field of view of 150 × 250 mm(2) (approximately 100 000 pixels) at high speeds (in less than 5 s). PLA with differing crystallinities between 0 and 50% blended samples with PHB in ratios of 80/20, 60/40, 40/60, 20/80, and pure films of 100% PLA and PHB were prepared. Compovision was used to collect respective NIR spectra in the 1000-2350 nm region and investigate the crystallinity of PLA and its concentration in the blends. The partial least squares (PLS) regression models for the crystallinity of PLA were developed using absorbance, second derivative, and standard normal variate (SNV) spectra from the most informative region of the spectra, between 1600 and 2000 nm. The predicted results of PLS models achieved using the absorbance and second derivative spectra were fairly good with a root mean square error (RMSE) of less than 6.1% and a determination of coefficient (R(2)) of more than 0.88 for PLS factor 1. The results obtained using the SNV spectra yielded the best prediction with the smallest RMSE of 2.93% and the highest R(2) of 0.976. Moreover, PLS models developed for estimating the concentration of PLA in the blend polymers using SNV spectra gave good predicted results where the RMSE was 4.94% and R(2) was 0.98. The SNV-based models provided the best-predicted results, since it can reduce the effects of the spectral changes induced by the inhomogeneity and the thickness of the samples. Wide area crystal evolution of PLA on a plate where a temperature slope of 70-105 °C had occurred was also monitored using NIR imaging. An SNV-based image gave an obvious contrast of the crystallinity around the crystal growth area according to slight temperature change. Moreover, it clarified the inhomogeneity of crystal evolution over the significant wide area. These results have proved that the newly developed hyperspectral NIR camera, Compovision, can be successfully used to study polymers for industrial processes, such as monitoring the crystallinity of PLA and the different composition of PLA/PHB blends.

A Partial Least Squares Algorithm for Microarray Data Analysis Using the VIP Statistic for Gene Selection and Binary Classification

An important application of microarray technology is the assignment of new subjects to known clinical groups (class prediction), but the huge number of screened genes and the small number of samples make this task difficult. To overcome this problem, the usual approach has been to extract a small subset of significant genes (gene selection) or to use the whole set of genes to build latent components (dimension reduction), then applying some usual multivariate classification procedure. Alternatively, both aims -gene selection and class prediction- can be achieved at the same time by using methods based on Partial Least Squares (PLS), as reported in the present work. We present an iterative PLS algorithm based on backward variable elimination through the “Variable Influence on Projection” (VIP) statistic, which finds an optimal PLS model through training and test sets. It simultaneously manages to reduce the number of selected genes by an iterative procedure and finds the best number of PLS factors to reach an optimal classification performance. It is a simple approach that uses only one mathematical method, maintains the identification of discriminatory genes, and builds an optimal predicting model with a fast computation. The algorithm runs as a module of the SIMFIT statistical package, where the optimal model and datasets can be re-run to further interpret the system through additional PLS options, such as scores and loadings plots, or class assignment of new samples. The proposed algorithm was tested under different scenarios occurring in microarray analysis using simulated data. The results are also compared against different classification methods such as KNN, PAM, SVM, RF and standard PLS. Keywords: Classification, gene selection, microarray, partial least squares, PLS, VIP statistic.

Rapid Analysis of Zinc in Soil with Long-NIR Spectroscopy

Near-infrared (NIR) spectroscopy combined with the partial least squares (PLS) regression was successfully applied for the rapid quantitative analysis of Zn2+in soil. The models were established using an approach based on randomness and similarity to obtain objective and practical models. Sixty-three samples were randomly selected from a total of 148 samples as the validation set. The remaining 85 samples were used as the modeling set, and it was divided into similar calibration (50 samples) and prediction (35 samples) sets. The results show that the long-NIR region at 1100 nm to 2498 nm could be considered as the information waveband of Zn2+in soil. The optimal number of PLS factors was 10, and the validation root mean square error (V-SEP) and validation correlation coefficients of prediction (V-RP) were 21.817 mg kg-1and 0.849, respectively. The Zn2+prediction values of the validation samples are close to the measured values. The results provided valuable reference for designing the dedicated spectrometers.

Fourier Transform Near-Infrared Spectroscopy as a Reliable Method for Quick Real Time Analysis of Complex Samples in Industry

Near-infrared spectroscopy (NIRS) in conjunction with multivariate calibration have proved to be adequate techniques to predict water content as well as methanol concentration in a industrial multicomponent mixture. The calibration models were developed using real industrial complex samples of water, methanol, and several other components in variable but lower concentrations such as sulfonic acids, acetone, and methyl cyanide. These samples come from industrial distillation of hydrated methanol. Prior to the establishment of the two calibration models, wavelength selection, original spectral data compression by means of the principal components analysis technique, and preprocessing optimization were performed. Calibration was modeled according to partial least squares (PLS) regression which was developed using the mean centering and the leave-one-out cross validation (LOOCV) method for the optimization of the principal components (PC) number. Both models were developed in the 5369–5153 cm–1 spectral region. Karl Fischer titration and gas chromatography were used as reference methods for water and methanol quantification, respectively. The suitability of the models was evaluated through the calculation of the standard error of calibration and standard error of prediction. For the water content determination the variance scaling pretreatment was applied with 12 PLS factors, the calibration and prediction errors being 2.56% and 1.76% respectively. As for the methanol concentration variance scaling was applied in conjunction with Savitzky–Golay’s second order derivative with six PLS factors. In this case the calibration and prediction errors were 0.42% and 0.90%, respectively.

Improvement of near infrared spectroscopic (NIRS) analysis of caffeine in roasted Arabica coffee by variable selection method of stability competitive adaptive reweighted sampling (SCARS)

Coffee is the most heavily consumed beverage in the world after water, for which quality is a key consideration in commercial trade. Therefore, caffeine content which has a significant effect on the final quality of the coffee products requires to be determined fast and reliably by new analytical techniques. The main purpose of this work was to establish a powerful and practical analytical method based on near infrared spectroscopy (NIRS) and chemometrics for quantitative determination of caffeine content in roasted Arabica coffees. Ground coffee samples within a wide range of roasted levels were analyzed by NIR, meanwhile, in which the caffeine contents were quantitative determined by the most commonly used HPLC-UV method as the reference values. Then calibration models based on chemometric analyses of the NIR spectral data and reference concentrations of coffee samples were developed. Partial least squares (PLS) regression was used to construct the models. Furthermore, diverse spectra pretreatment and variable selection techniques were applied in order to obtain robust and reliable reduced-spectrum regression models. Comparing the respective quality of the different models constructed, the application of second derivative pretreatment and stability competitive adaptive reweighted sampling (SCARS) variable selection provided a notably improved regression model, with root mean square error of cross validation (RMSECV) of 0.375mg/g and correlation coefficient (R) of 0.918 at PLS factor of 7. An independent test set was used to assess the model, with the root mean square error of prediction (RMSEP) of 0.378mg/g, mean relative error of 1.976% and mean relative standard deviation (RSD) of 1.707%. Thus, the results provided by the high-quality calibration model revealed the feasibility of NIR spectroscopy for at-line application to predict the caffeine content of unknown roasted coffee samples, thanks to the short analysis time of a few seconds and non-destructive advantages of NIRS.

On-line near infrared spectroscopy as a Process Analytical Technology (PAT) tool to control an industrial seeded API crystallization

The final step of an active pharmaceutical ingredient (API) manufacturing synthesis process consists of a crystallization during which the API and residual solvent contents have to be quantified precisely in order to reach a predefined seeding point. A feasibility study was conducted to demonstrate the suitability of on-line NIR spectroscopy to control this step in line with new version of the European Medicines Agency (EMA) guideline [1]. A quantitative method was developed at laboratory scale using statistical design of experiments (DOE) and multivariate data analysis such as principal component analysis (PCA) and partial least squares (PLS) regression. NIR models were built to quantify the API in the range of 9–12% (w/w) and to quantify the residual methanol in the range of 0–3% (w/w). To improve the predictive ability of the models, the development procedure encompassed: outliers elimination, optimum model rank definition, spectral range and spectral pre-treatment selection. Conventional criteria such as, number of PLS factors, R2, root mean square errors of calibration, cross-validation and prediction (RMSEC, RMSECV, RMSEP) enabled the selection of three model candidates. These models were tested in the industrial pilot plant during three technical campaigns. Results of the most suitable models were evaluated against to the chromatographic reference methods. Maximum relative bias of 2.88% was obtained about API target content. Absolute bias of 0.01 and 0.02% (w/w) respectively were achieved at methanol content levels of 0.10 and 0.13% (w/w). The repeatability was assessed as sufficient for the on-line monitoring of the 2 analytes. The present feasibility study confirmed the possibility to use on-line NIR spectroscopy as a PAT tool to monitor in real-time both the API and the residual methanol contents, in order to control the seeding of an API crystallization at industrial scale. Furthermore, the successful scale-up of the method proved its capability to be implemented in the manufacturing plant with the launch of the new API process.

Scatter-based quantitative spectroscopic analysis of milk fat and total protein in the region 400–1100 nm in the presence of fat globule size variability

Accurate prediction models for fat and total protein content in raw milk have been built on visible and short-wave near infrared spectra (400–1100nm) of a designed sample set with systematically varied nutrient composition and homogenization degree. Unlike the conventional approach to the spectroscopic milk analysis, exploiting the components' absorbance, the present method basically relies on the phenomenon of light scatter by fat globules and protein micelles.It has been shown that partial least-squares (PLS) regression on raw spectral data results in higher prediction accuracies of fat and total protein content compared to scatter-corrected spectra. Interpretation given to individual PLS factors confirms a dominating role of scatter in the modeling and aids in general understanding of quantitative multivariate analysis based on complex optical responses resulting from multiple scattering by different particles at significantly varying size distributions. This new approach can be used in the raw milk laboratory or field analysis as well as for in-line monitoring of various milk transfer and production stages.Practical modeling issues, i.e. experimental design of the sample set, model validation and refinement, have been also elaborated in this study. Variable selection by interval PLS (iPLS) regression essentially improved the model performance. Selected intervals can be utilized for technical simplification of suggested method of milk analysis.

Predicting the heating value of solid manure with visible and near-infrared spectroscopy

Visible and near-infrared spectral data were used to predict the higher heating value (HHV) and dry, ash-free HHV (HHVdaf) of solid manure samples collected from cattle fed diets containing wet distillers grains plus solubles (WDGS) in 0%, 15%, 30%, 45%, and 60% dry matter concentrations. The HHV was determined by isoperibol bomb calorimetry and the HHVdaf was calculated from an equation based on the HHV and proximate analysis. Spectral models were developed in “The Unscrambler” software. The spectral models based on all treatments with random samples withheld for validation predicted the HHV with excellent reliability within 1.7%; RMSD=60.19calg−1 (108Btulb−1), RPD=2.29 (excellent), and bias=−15.29calg−1 (28Btulb−1), using five PLS factors and identifying 129 important wavebands. Accounting for estimated N and S content reduced the predictive accuracy of the spectral models by 0.1% with an RPD=2.28 (excellent). Spectral models based on all treatments with random samples withheld for validation predicted the HHVdaf with acceptable reliability within 2.0% with an RMSD=96.17calg−1 (173Btulb−1), RPD=1.17 (acceptable), and bias=−19.83calg−1 (−37Btulb−1), using five PLS (partial least squares) factors and identifying 29 important wavebands. Spectral models reliably predicted the HHV of feedlot manure with accuracy well under the 5% error margin tolerated in practical applications such as feedlot manure gasification.

Fast determination of trace dimethyl fumarate in milk with near infrared spectroscopy following fluidized bed enrichment

Near infrared spectroscopy (NIRS) has been proved to be a powerful analytical tool in different fields. However, because of the low sensitivity in near infrared region, it is a significant challenge to detect trace analytes with normal NIRS technique. A novel enrichment technique called fluidized bed enrichment has been developed recently to improve sensitivity of NIRS which allows a large volume solution to pass through within a short time. In this paper, fluidized bed enrichment method was applied in the determination of trace dimethyl fumarate in milk. Macroporous styrene resin HZ-816 was used as adsorbent material, and 1 L solution of dimethyl fumarate was run to pass through the material for concentration. The milk sample was pretreated to remove interference matters such as protein, fat, and then passed through the material for enrichment; after that, diffuse reflection NIR spectra were measured for the analyte concentrated on the material directly without any elution process. The enrichment and spectral measurement procedures were easy to operate. NIR spectra in 900-1,700 nm were collected for dimethyl fumarate solutions in the concentration range of 0.506-5.060 μg/mL and then used for multivariate calibration with partial least squares (PLS) regression. Spectral pretreatment methods such as multiplicative scatter correction, first derivative, second derivative, and their combinations were carried out to select the optimal PLS model. Root mean square error of cross-validation calculated by leave-one-out cross-validation is 0.430 μg/mL with ten PLS factors. Ten samples in an independent test set were predicted by the model with the mean relative error of 5.33%. From the results shown in this work, it can be concluded that the NIR technique coupled with on-line enrichment method can be expanded for the determination of trace analytes, and its applications in real liquid samples like milk and juice may also be feasible.

Design and development of topoisomerase inhibitors using molecular modelling studies

Topoisomerase inhibitors are used as anticancer and antibacterial agents. A series of novel 2,4,6-tri-substituted pyridine derivatives reported as topoisomerase inhibitors were used for quantitative structure-activity relationship (QSAR) study. In order to understand the structural requirement of these topoisomerase inhibitors, a ligand-based pharmacophore and atom-based 3D-QSAR model have been developed. A five-point pharmacophore with one hydrophobic group (H4), four aromatic rings (R5, R6, R7 and R8) was obtained. The pharmacophore hypothesis yielded a 3D-QSAR model with good partial least-square (PLS) statistic results. The training set correlation is characterized by PLS factors (r (2) = 0.7892, SD = 0.2948, F = 49.9, P = 1.379). The test set correlation is characterized by PLS factors (q (2) = 0.7776, root mean squared error = 0.2764, Pearson R = 0.8926). The docking study revealed the binding orientations of these inhibitors at active site amino acid residues of topoisomerases enzyme. The results of pharmacophore hypothesis and 3D-QSAR provided the detail structural insights as well as highlighted the important binding features of novel 2,4,6-tri-substituted pyridine derivatives and can be developed as potent topoisomerase inhibitors. FigureKey structural requirement for topoisomerase activity.

Application of FT-NIR spectroscopy on the determination of the fat and protein contents of lyophilized cheeses

Fourier transform near infrared spectroscopy (FT-NIR) has been developed for determining the fat and protein contents of hard, semi-hard and processed cheeses. Multivariate calibration models were carried out by partial least squares (PLS) regression. The diffuse reflection spectra of different type of lyophilized cheeses were measured by FT-NIR analyser in the 800–2500 nm spectral range. The calibration set of 62 samples were validated by leave-oneout cross-validation and by prediction set of 31 samples. The optimal result for fat content (root mean square error of cross-validation (RMSECV)=1.0; R2=99.1%, PLS factor= 6) was obtained when the spectra were preprocessed by first derivation (FD) combined with multiplicative scatter correction (MSC) and smoothing. The optimal result for protein content (RMSECV=1.4, R2=97.2%, PLS factor=6) was observed when the first derivation combined with straight line subtraction (SLS) and smoothing spectral preprocessing method was applied.

Determination of Hemoglobin in Human Whole Blood Using Near-Infrared Spectroscopy with Stability

Near-infrared (NIR) spectroscopy combined with the partial least-squares (PLS) regression was successfully applied for the rapid quantitative analysis of hemoglobin (HGB) based on human whole blood samples. Based on the varied divisions for the calibration and prediction sets, an appropriate waveband with stability was selected through a rigorous modeling process. Among 215 samples, 80 were randomly selected as the validation set. The remaining 135 samples were divided into the calibration set (80 samples) and the prediction set (55 samples) for a total of 200 times with certain similarities. The selected waveband was the combination of visible and short-wave NIR region at 400 nm to 1100 nm. The optimal PLS factor was 8, the modeling effect M-SEPAve, M-RP,Ave, M-SEPStd and M-RP,Std were 2.88gL-1, 0.986, 0.34gL-1 and 0.003, respectively, the validation effect V-SEP, V-RP and V-RSEP were 2.92gL-1, 0.986 and 2.26%, respectively. The results indicated that the method has high prediction precision and well stability. It is hopeful to be applied to clinic.

Rapid Determination of Chemical Oxygen Demand in Sugar Refinery Wastewater by Short-Wave Near-Infrared Spectroscopy

Near-infrared (NIR) spectroscopy combined with partial least squares (PLS) regression and Savitzky-Golay (SG) smoothing was successfully applied for the rapid no-reagent determination of chemical oxygen demand (COD) in sugar refinery wastewater. Among 141 samples, 60 were randomly selected as the validation set. The remaining 81 samples were divided into the calibration set (50 samples) and the prediction set (31 samples) for a total of 20 times with certain similarities. The results showed that the short-wave NIR region (780 nm to 1100 nm) can be used as information waveband of COD in sugar refinery wastewater. The optimal SG smoothing parameters and PLS factors OD, DP, NSP, and F were 5, 5, 7, and 6, respectively. The modeling effects M-SEPAve, M-RP,Ave, M-SEPStd, and M-RP,Stdwere 24.3 mg/L, 0.971, 2.7 mg/L, and 0.007, respectively. The validation effects V-SEP and V-RP were 25.9 mg/L and 0.963, respectively. The results indicated that the method has good prediction effect and stability, and provided a reliable analysis model and valuable references for designing specialized instruments.

FTIR/ATR Spectroscopy with Stability for Determination of HGB in Human Whole Blood

Fourier transform infrared spectroscopy (FTIR) and attenuated total reflection (ATR) combined with partial least-squares (PLS) and Savitzky-Golay (SG) smoothing were successfully applied for rapid no-reagent determination of hemoglobin (HGB) in human whole blood. The optimal SG smoothing parameters OD, DP, NSP and PLS factor were 1, 2, 49 and 14, respectively. The modeling effect M-SEPAve, M-RP,Ave, M-SEPStd, and M-RP,Std were 6.73g/L, 0.850, 0.306g/L, and 0.015, respectively. The validation effect V-SEP and V-RP were 6.83g/L and 0.864, respectively. The method provided a reliable model with stability for clinic application of FTIR/ATR.

Measurement of COD for Sugar Refinery Wastewater Using FTIR/ATR Spectroscopy with Stability

Fourier transform infrared spectroscopy (FTIR) and attenuated total reflection (ATR) technology combined with the partial least squares (PLS) regression were successfully applied for the rapid no-reagent determination of chemical oxygen demand (COD) for sugar refinery wastewater. Among 105 samples, 45 samples were randomly selected as the validation set. The remaining 60 samples were divided into the calibration set (40 samples) and the prediction set (20 samples) for a total of 30 times with certain similarities. The appropriate model with stability was established on the waveband of 4500 cm-1to 600 cm-1. The optimal PLS factor was 4, the modeling effect M-SEPAve, M-RP,Ave, M-SEPStd, and M-RP,Stdwere 38.1 mg/L, 0.944, 3.2 mg/L, and 0.013, respectively, the validation effect V-SEP and V-RPwere 38.2 mg/L and 0.949, respectively, which has good prediction effect and stability. The results provide a reliable analytical model for the monitoring of COD in sugar refinery wastewater.

Near-Infrared Spectroscopic Analysis of Hemoglobin with Stability Based on Human Hemolysates Samples

Near-infrared (NIR) spectroscopy combined with the partial least-squares (PLS) regression was successfully applied for the rapid quantitative analysis of hemoglobin (HGB) based on human hemolysates samples. Based on the varied divisions for the calibration and prediction sets, an effective modeling approach using stable model parameters was proposed. Among 255 samples, 80 were randomly selected as the validation set. The remaining 175 samples were divided into the calibration set (110 samples) and the prediction set (65 samples) for a total of 30 times with certain similarities based on partial least squares cross-validation predictive basis (PLSPB). The optimal PLS factor was 8, the modeling effects M-SEPAve, M-RP,Ave, M-SEPStd and M-RP,Std were 3.84g/L, 0.967, 0.16g/L and 0.006, respectively, the validation effects V-SEP, V-RP and V-RSEP were 3.59g/L, 0.980 and 2.7%, respectively. It indicated that the method has high prediction precision and well stability. The results show that NIR spectroscopy of hemolysates is accurate to HGB’s determination, and it is hopeful to be applied to clinic.

Crystallinity evaluation of tacrolimus solid dispersions by chemometric analysis

Different destructive and nondestructive analytical methods, namely powder X-ray diffractometry (PXRD), differential scanning calorimetry (DSC), Raman and near-infrared (NIR) spectroscopy and imaging, to detect and characterize tacrolimus trace crystallinity in an amorphous solid dispersion (SD) using chemometric analysis were developed. The SD was spiked with different percentages of the crystalline drug to construct an array of SDs with different crystallinity percentages. Partial least square (PLS) regression analysis was employed to compare the performance of the calibration models created using these analytical methods. The obtained results indicated a significant interaction between tacrolimus and the employed polymer and a drug dissolution dependency on the crystalline fraction within the SDs. Using two PLS factors, these analytical methods were ranked according to its specificity to detect the trace crystallinity of SDs as NIR>PXRD>Raman>DSC. Through the application of PLS, root-mean-squared error of calibration values of 2.91%, 5.36%, 7.07% and 11.58% were calculated for the calibration models constructed by NIR, PXRD, Raman and DSC, respectively. Having a prediction error of 2.1% and a correlation coefficient of 0.99, it is demonstrated that combined NIR imaging and chemometric analysis outperformed the other methods in detecting trace crystallinity in tacrolimus amorphous systems. The spatial distributions of amorphous and crystalline drug were also obtained in order to allow for studying the crystallization dissemination in the solid dispersions. Consequently, NIR and NIR imaging coupled with chemometry was shown to be a powerful tool for the prediction of drug crystallinity within SDs.

Quantifying TiO2 abundance of lunar soils: Partial least squares and stepwise multiple regression analysis for determining causal effect

Partial least squares (PLS) regression was applied to the Lunar Soil Characterization Consortium (LSCC) dataset for spectral estimation of TiO2. The LSCC dataset was split into a number of subsets including the low-Ti, high-Ti, total mare soils, total highland, Apollo 16, and Apollo 14 soils to investigate the effects of interfering minerals and nonlinearity on the PLS performance. The PLS weight loading vectors were analyzed through stepwise multiple regression analysis (SMRA) to identify mineral species driving and interfering the PLS performance. PLS exhibits high performance for esti- mating TiO2 for the LSCC low-Ti and high-Ti mare samples and both groups analyzed together. The results suggest that while the dominant TiO2-bearing minerals are few, additional PLS factors are re- quired to compensate the effects on the important PLS factors of minerals that are not highly corrected to TiO2, to accommodate nonlinear relationships between reflectance and TiO2, and to correct incon- sistent mineral-TiO2 correlations between the high-Ti and low-Ti mare samples. Analysis of the LSCC highland soil samples indicates that the Apollo 16 soils are responsible for the large errors of TiO2 es- timates when the soils are modeled with other subgroups. For the LSCC Apollo 16 samples, the domi- nant spectral effects of plagioclase over other dark minerals are primarily responsible for large errors of estimated TiO2. For the Apollo 14 soils, more accurate estimation for TiO2 is attributed to the posi- tive correlation between a major TiO2-bearing component and TiO2, explaining why the Apollo 14 soils

FT‐NIR spectroscopy: a rapid method for estimation of moisture content in bael pulp

PurposeThe purpose of this paper is to develop FT‐NIR technique for determination of moisture content in bael pulp.Design/methodology/approachCalibration and validation sets were designed for the conception and evaluation of the method adequacy in the range of moisture content 70 to 95 per cent (wb). The prediction models based on partial least squares (PLS) regression, were developed in the near‐infrared region (4,000‐2,500cm‐1). Conventional criteria such as the R2, the root mean square errors of cross validation (RMSECV), root mean square errors of estimation (RMSEE) as well as the number of PLS factors were considered for the selection of three pre‐processing (vector normalization, minimum‐maximum normalization and multiplicative scatter correction) methods.FindingsThe best calibration model was developed with min‐max normalization (MMN) spectral pre‐processing (R2=99.3). The MMN pre‐processing method was found most suitable and the maximum coefficient of determination (R2) value of 0.993 was obtained for the calibration model developed. The developed results indicated that FTNIR spectroscopy could be used for rapid detection of moisture content in bael pulp samples without any sample destruction.Originality/valueThe research in this paper is useful for the quick detection of moisture content of bael fruit pulp during processing.

FTIR/ATR Spectroscopy Analysis of COD in Wastewater Based on Combination Optimization of SG Smoothing Modes and PLS Factor

A directly rapid quantitative analysis method for chemical oxygen demand (COD) of wastewater was established by Fourier transform infrared (FTIR) spectroscopy and attenuated total reflection (ATR) technology. The combination selection of Savitzky-Golay (SG) smoothing modes combined with partial least squares (PLS) factor was applied to the model optimization here. The optimal smoothing parameters were 2ndorder derivative smoothing, 6thdegree polynomial and 51 smoothing points. The corresponding optimal PLS factor, root mean squared error of predication (RMSEP) and correlation coefficient of predication (RP) were 11, 24.4 (mg/L) and 0.968 respectively, which was obviously superior to the direct PLS model without SG smoothing. This demonstrated that the combination selection of SG smoothing parameters and PLS factor can be effectively applied to the model optimization for FTIR/ATR analysis of COD in wastewater.