Evaluation of the potential of FTIR and chemometrics for separation between defective and non-defective coffees

The objective of this work was to evaluate the potential of Fourier transform infrared spectroscopy (FTIR) for the discrimination of defective and non-defective coffee beans. Defective (black, immature and sour) and non-defective Arabica coffee beans were submitted to FTIR analysis by transmittance readings employing KBr discs and reflectance readings employing attenuated total reflectance (ATR) and diffuse reflectance (DR) accessories. Multivariate statistical analysis (PCA, clusters) was performed in order to verify the possibility of discrimination between defective and non-defective coffee samples. A clear separation between defective and non-defective coffee beans was observed, based on both PCA and cluster analysis of the reflectance spectra (ATR and DR accessories) and of the first derivatives of the transmittance spectra (KBr discs). Such results indicate that FTIR analysis has the potential for the development of a fast and reliable analytical methodology for the discrimination between defective and non-defective coffee beans.     

Discrimination between defective and non-defective roasted coffees by diffuse reflectance infrared Fourier transform spectroscopy

The objective of this work was to evaluate the feasibility of employing Diffuse Reflectance Infrared Fourier Transform Spectroscopy (DRIFTS) for discrimination between defective and non-defective coffees after roasting and grinding. Defective (black, immature and sour) and non-defective Arabica coffee beans were submitted to light, medium and dark roasts at 220, 235 and 250 °C. Principal Components Analysis of the DRIFTS spectra (normalized or not) and of the first derivatives of the spectra provided separation of the samples into four groups: non-defective, black, dark sour and light sour, with immature beans scattered among the sour samples. Classification models were developed based on Linear Discriminant Analysis and recognition and prediction abilities of these models ranged from 95 to 100%. Such results indicate that DRIFTS presents potential for the development of a fast and reliable analytical methodology for discrimination between defective and non-defective coffee after roasting and grinding.     

Chemical characterisation of non-defective and defective green arabica and robusta coffees by electrospray ionization-mass spectrometry (ESI-MS)

The coffee roasted in Brazil is considered to be of low quality, due to the presence of defective coffee beans that depreciate the beverage quality. These beans, although being separated from the non-defective ones prior to roasting, are still commercialized in the coffee trading market. Thus, it was the aim of this work to verify the feasibility of employing ESI-MS to identify chemical characteristics that will allow the discrimination of Arabica and Robusta species and also of defective and non-defective coffees. Aqueous extracts of green (raw) defective and non-defective coffee beans were analyzed by direct infusion electrospray ionization mass spectrometry (ESI-MS) and this technique provided characteristic fingerprinting mass spectra that not only allowed for discrimination of species but also between defective and non-defective coffee beans. ESI-MS profiles in the positive mode (ESI(+)-MS) provided separation between defective and non-defective coffees within a given species, whereas ESI-MS profiles in the negative mode (ESI(−)-MS) provided separation between Arabica and Robusta coffees.     

Simultaneous Detection of Multiple Adulterants in Ground Roasted Coffee by ATR-FTIR Spectroscopy and Data Fusion

This paper proposes a novel screening method for the simultaneous detection of four adulterants (spent coffee grounds, roasted coffee husks, roasted corn, and roasted barley) in ground roasted coffee using partial least squares discriminant analysis (PLS-DA) with mid-infrared spectroscopy. Two different acquisition modes (attenuated total reflectance, ATR, and diffuse reflectance, DR) are compared. Two recent chemometric approaches, hierarchical models (HM) and data fusion (DF), were employed in order to improve model performance. First level models provided discrimination between unadulterated and adulterated coffee samples, whereas second level models were able to identify the presence of each specific adulterant. The use of DF decreased the percentage of misclassified samples for the first level models from 19.6/14.7% (DR) and 7.5/14.5% (ATR) down to 2.5/4.5% considering the training/test sets. The percentage of misclassified samples in the second level models went as low as 0% (DF—spent coffee, training set). The proposed method is simple, fast, reliable for detecting adulteration in coffee samples, and capable of identifying these adulterants, even when in complex mixtures containing other adulterants.     

Discrimination between defective and non-defective Brazilian coffee beans by their volatile profile

The coffee roasted in Brazil is considered to be of low quality, due to the presence of defective coffee beans that depreciate the beverage quality. In view of the fact that coffee flavour is directly related to the volatile compounds produced during roasting, the objective of the present study was to perform a comparative evaluation of the volatile fraction of defective (black, immature, sour) and healthy coffee beans, in order to find possible chemical markers for detection of defective coffee beans in roasted coffee. Volatiles extraction and concentration was performed by solid phase micro-extraction (SPME) of the roasted coffee headspace, using a triple phase (divinylbenzene/carboxen/polydimethylsiloxane) fiber. Analysis of the volatile profiles was performed by GC–MS. The results obtained showed that the proposed methodology was adequate for extraction, concentration and analysis of the coffees volatile profile. Several substances were identified as possible markers for differentiating black, sour and immature beans from healthy coffee beans. Statistical analysis of the data by principal components (PCA) demonstrated that the volatile profile enables the differentiation of healthy and defective coffees. The data were separated into two major groups, one represented by immature and black beans and the other by healthy and sour coffee beans. Such results indicated that black and sour beans can be associated to fermentation of immature and of healthy beans, respectively.     

Approaches to Adulteration Detection in Instant Coffees using Infrared Spectroscopy and Chemometrics

Fourier transform infrared (FTIR) spectroscopy is examined as a rapid alternative to wet chemistry methods for the detection of adulteration of freeze-dried instant coffees. Spectra have been collected of pure coffees, and of samples adulterated with glucose, starch or chicory in the range 20–100 g kg⁻¹. Two different FTIR sampling methods have been employed: diffuse reflectance, and attenuated total reflectance. Three different statistical treatments of the spectra were carried out. Firstly, the spectra were compressed by principal component analysis and a linear discriminant analysis performed. With this approach, a 98% successful classification rate was achieved. Secondly, a simultaneous partial least square regression was carried out for the content of three added carbohydrates (xylose, glucose and fructose) in order to assess the potential of FTIR spectroscopy for determining the carbohydrate profile of instant coffee. Lastly, the discrimination of pure from adulterated coffee was performed using an artificial neural network (ANN). A perfect rate of assignment was obtained. The generalization ability of the ANN was tested on an independent validation data set; again, 100% correct classifications were achieved.     

Strontium and oxygen isotope fingerprinting of green coffee beans and its potential to proof authenticity of coffee

Multicollector inductively coupled plasma mass spectrometry (MC-ICP-MS) and Isotope Ratio Mass Spectrometry (IRMS) were applied to combine strontium and oxygen isotope abundance ratios of green coffees from 20 geographical origins in order to evaluate the suitability of these parameters as indicators of geographical origin, which is an important parameter of coffee quality. Results show that both isotopic systems of green coffee beans show a relation to environmental factors that influence processes occurring during the growth of the coffee bean. The final results allowed discrimination of local provenances investigated in this study by principal component analysis (PCA) and exhibit the potential to proof authenticity of world coffees. A good discrimination was obtained especially for coffee of South America region and coffees originating from islands.     

基于顶空气相色谱-离子迁移谱技术的生咖啡豆快速鉴别方法

采用顶空气相色谱-离子迁移谱（headspace-gas chromatography-ion mobility spectrometry,HS-GC-IMS）技术对云南不同品种生咖啡豆的挥发性有机物（volatile organic compounds,VOCs）进行无损分析。根据保留指数和迁移时间对其挥发性成分进行二维定性,创建生咖啡豆VOCs的差异谱图,并对其VOCs数据进行主成分分析（principal component analysis,PCA）。结果表明,HS-GC-IMS可有效分离生咖啡豆挥发性成分,对样品VOCs的信息采集及分析可在20 min内完成,并鉴定出42 种挥发性物质,主要为醛类、酯类、醇类、酮类、吡嗪类、酸类及含硫化合物。采用PCA分析HS-GC-IMS图谱,可准确区分不同品种生咖啡豆。该方法具有快速、灵敏、无损的特点,为生咖啡豆的品种识别、产地追溯、品质控制等提供了一定参考依据和理论基础。     

Discrimination between Arabica and Robusta green coffee using visible micro Raman spectroscopy and chemometric analysis

We have applied visible micro Raman spectroscopy combined with principal component analysis (PCA) as a powerful technique for the fast discrimination between the two coffee species, Arabica and Robusta, based on their chlorogenic acid (CGA) and lipid contents. The Raman spectra reveal different CGA and lipid compositions when comparing Arabica and Robusta green coffee. Analysing the whole Raman spectrum, the PCA yielded a clear separation between Arabica and Robusta with 93% of the total spectral variation. Here, the most significant spectral range lies between 1000 and 1750 cm⁻¹ and is dominated by the Raman bands of CGA. Also, by restricting the PCA analysis to the spectral range from 2700 to 3050 cm⁻¹, which is dominated by lipid bands, a reliable discrimination between the two coffee species could be achieved. In this case, the first two principal components of the PCA accounted for 85% of the explained total spectral variation.     

Determination of protein concentration in raw milk by mid-infrared fourier transform infrared/attenuated total reflectance spectroscopy

This study investigates the potential use of attenuated total reflectance spectroscopy in the mid-infrared range for determining protein concentration in raw cow milk. The determination of protein concentration is based on the characteristic absorbance of milk proteins, which includes 2 absorbance bands in the 1500 to 1700 cm(-1) range, known as the amide I and amide II bands, and absorbance in the 1060 to 1100 cm(-1) range, which is associated with phosphate groups covalently bound to casein proteins. To minimize the influence of the strong water band (centered around 1640 cm(-1)) that overlaps with the amide I and amide II bands, an optimized automatic procedure for accurate water subtraction was applied. Following water subtraction, the spectra were analyzed by 3 methods, namely simple band integration, partial least squares (PLS) and neural networks. For the neural network models, the spectra were first decomposed by principal component analysis (PCA), and the neural network inputs were the spectra principal components scores. In addition, the concentrations of 2 constituents expected to interact with the protein (i.e., fat and lactose) were also used as inputs. These approaches were tested with 235 spectra of standardized raw milk samples, corresponding to 26 protein concentrations in the 2.47 to 3.90% (weight per volume) range. The simple integration method led to very poor results, whereas PLS resulted in prediction errors of about 0.22% protein. The neural network approach led to prediction errors of 0.20% protein when based on PCA scores only, and 0.08% protein when lactose and fat concentrations were also included in the model. These results indicate the potential usefulness of Fourier transform infrared/attenuated total reflectance spectroscopy for rapid, possibly online, determination of protein concentration in raw milk.     

ATR‐FTIR spectroscopy and chemometric analysis applied to discrimination of landrace maize flours produced in southern Brazil

This work aims at discriminating flours of 26 maize landraces from southern Brazil, by using the Attenuated Total Reflection Fourier Transform Infrared (ATR‐FTIR) spectroscopy and chemometrics (principal components analysis – PCA). PCA applied to the FTIR spectra in the 3‐600 (whole spectrum) and 1650–1500 cm^?1 (fingerprint region of proteins) spectral windows clearly discriminated the Amarelão landrace. Quantitative and semi‐qualitative analysis of proteins showed a wide range among the fractions, mainly of prolamine (13.47–28.43 g Kg^?1) and glutelin (5.57–30.98 g Kg^?1) contents. Pixurum 6, Pixurum 5, and MPA1 landraces are of superior nutritional value for their albumin, globulin, and glutelin contents. PCA of the spectral dataset in the fingerprint region to carbohydrates (1200–950 and 1065–950 cm^?1) also including commercial standards of amylose and amylopectin was able in separating the Moroti genotype, which grouped with the amylopectin standard. Thus, ATR‐FTIR and PCA showed to be useful tools for the quick screening and discrimination of maize with distinct chemical composition.     

Potential use of Fourier transform infrared spectroscopy for differentiation of bovine and porcine gelatins

In order to classify unknown gelatin into their species of origin, a simple and rapid method for the qualitative determination was developed using Fourier transform infrared (FTIR) in combination with attenuated total reflectance (ATR) and discriminant analysis. The spectra were analysed using a chemometric method, principal component analysis (PCA), to classify and characterise gelatin compounds using regions of the FTIR spectra in the range of 3290–3280 cm^?1 and 1660–1200 cm^?1 as calibration models. Results from PCA, which were subsequently represented by the Cooman’s plot showed a clear distinction between gelatin samples of bovine and porcine origins. This qualitative approach, besides providing a rapid determination of the source of gelatin, may also be established based on a second derivative study of the FTIR spectrum to alleviate any doubt of the gelatin source for applications in the food and pharmaceutical industries.     

Amino acid profiles in unripe Arabica coffee fruits processed using wet and dry methods

The quality of natural coffee produced in Brazil is quite variable. Fruits at different stages of maturation can be found on the same plant, and unripe fruits are naturally present during the harvest. The pulping of ripe fruits can be effectively used to improve the quality of the coffee, as the ripe fruits will be separated from the unripe fruits; however, the presence of a portion of unripe fruit (with lower quality) in the processing is unavoidable. The wet processing of immature coffee fruit appears to be a potential way of improving its quality. According to the coffee processing used post-harvest, changes were observed in the levels of free amino acids in immature coffee beans. Among the amino acids present, asparagine is the primary amino acid found in unripe coffee beans. Asparagine produces acrylamide, a potentially harmful substance generated during the course of the Maillard reaction. In this study, amino acids in immature coffee beans were analysed using reversed-phase chromatography and ultraviolet detection after the derivatisation with phenylisothiocyanate. The amino acid profiles of the immature coffee beans demonstrated that asparagine is present at more significant levels when immature coffee fruits were processed via dry processing, as compared to wet processing.     

Can near‐infrared reflectance of green coffee be used to detect introgression in Coffea arabica cultivars?

Most new coffee cultivars disseminated over the last 15 years are derived from the Timor Hybrid (Coffea arabica × C canephora). Introgression of genes from the C canephora genome has been estimated at between 9 and 29% of the genome. It has been shown that introgression can have a negative impact on the cup quality of cultivars derived from the Timor Hybrid. Consequently, coffee buyers or roasters may wish to assess whether the coffee they are purchasing comes from introgressed varieties. The possibility of distinguishing between non‐introgressed Arabicas and genotypes carrying chromosome fragments introgressed from C canephora was investigated (i) using some classical chemical compounds (caffeine, chlorogenic acids, trigonelline, fat and sucrose) and (ii) using a new approach based on spectra acquired by near‐infrared reflectance of green coffee. Near‐infrared spectra were obtained for 129 samples from two collections (Nicaragua and Costa Rica) of introgressed and non‐introgressed coffee trees. The spectral collections were treated by principal component and factorial discrimination. When the introgressed coffee trees were compared with the non‐introgressed trees using the chemical compounds, small but significant differences were found in caffeine, trigonelline and chlorogenic acid contents. However, the small variations in those compounds are not enough to detect introgression. The spectral collections treated by principal component and factorial discrimination made it possible to class from 92.30 to 94.87% of the analysed samples correctly, while the percentages of correctly classified samples in the verification file varied from 88.23 to 94.11%. The NIRS method appears to be an efficient method for determining whether a green coffee comes from an introgressed variety. Copyright © 2005 Society of Chemical Industry     

Feasibility study of discriminating the manufacturing process and sampling zone in ripened soft cheeses using attenuated total reflectance MIR and fiber optic diffuse reflectance VIS–NIR spectroscopy

The use of visible–near infrared (VIS–NIR) and mid infrared (MIR) spectroscopies for rapid characterisation of 15 traditional and stabilised retail soft cheeses, manufactured with different cheese making procedures was described. A fiber-type, VIS–NIR spectrophotometer (Zeiss Corona 45 VIS–NIR) in a measurement range of 315–1700 nm and a Fourier transform spectrometer (IFS 66V/S, Bruker, Belgium) in a measurement range between 3000 and 900 cm⁻¹ were used to scan spectra in reflectance mode at the external (E) and central (C) zones of the investigated cheeses. The principal component analysis (PCA) applied to the normalised spectral data set (VIS–NIR and MIR) did not provide a good discrimination of cheeses. Therefore, the factorial discriminant analysis (FDA) was applied separately to the first 5 principal components (PCs) of the PCA performed on the VIS–NIR and MIR data sets. Regarding the MIR spectra, the percentage of samples correctly classified into six groups (three for the E and three for the C zones) by the FDA was 64.8% and 33.3% for the calibration and validation samples, respectively. Better classification was obtained from the VIS–NIR spectra since the percentage of samples correctly classified was 85.2% and 63.2% for the calibration and validation samples, respectively. Finally, a concatenation technique was applied on the first 5 PCs of the PCA performed on the VIS–NIR and MIR data sets. This technique allowed a quite satisfactory classification of the investigated cheeses according to their manufacturing process and their sampling zone. In this case, correct classifications (CC) of 90.7% and 80.6% were obtained for the calibration and the validation samples, respectively.     

Discrimination of organic coffee via Fourier transform infrared-photoacoustic spectroscopy

BACKGROUND: Procedures for the evaluation of the origin and quality of ground and roasted coffee are constantly needed for the associated industry due to complexity of the related market. Conventional Fourier transform infrared (FTIR) spectroscopy can be used for detecting changes in functional groups of compounds, such as coffee. However, dispersion, reflection and non‐homogeneity of the sample matrix can cause problems resulting in low spectral quality. On the other hand, sample preparation frequently takes place in a destructive way. To overcome these difficulties, in this work a photoacoustic cell has been adapted as a detector in a FTIR spectrophotometer to perform a study of roasted and ground coffee from three varieties of Coffea arabica grown by organic and conventional methods. RESULTS: Comparison between spectra of coffee recorded by FTIR‐photoacoustic spectrometry (PAS) and by FTIR spectrophotometry showed a better resolution of the former method, which, aided by principal components analysis, allowed the identification of some absorption bands that allow the discrimination between organic and conventional coffee. CONCLUSION: The results obtained provide information about the spectral behavior of coffee powder which can be useful for establishing discrimination criteria. It has been demonstrated that FTIR‐PAS can be a useful experimental tool for the characterization of coffee. Copyright © 2012 Society of Chemical Industry     

The optical properties of hand-woven carpet

Abstract

The reflectance spectra and color properties of yarn when used in carpet are different from its cross-sectional direction as the pile of hand-woven carpet. This paper presents a comprehensive study of reflectance spectra and color properties of the pile in longitudinal direction and hand-woven carpet. For this purpose, the wool yarns were dyed with madder as a natural dye in various concentrations from 10% to 100% on the weight of sample. In order to study of the optical properties of the pile of hand-woven carpet, the dyed yarns were used to weave the hand-woven carpet. The obtained results showed that the yarn in longitudinal direction is lighter than the pile of hand-woven carpet, and so, the color strength of hand-woven carpet is more than the color strength of yarn. The spectrophotometric properties of samples were also studied by Principal Component Analysis (PCA). The difference between reflectance spectra of yarn in longitudinal direction and reflectance spectra of the pile of hand-woven carpet was also shown by PCA results.     

Feasibility study on the use of attenuated total reflectance mid-infrared for analysis of compositional parameters in wine

A simple and fast method was developed for simultaneously determining ethanol, specific gravity (SG), volatile acidity (VA), glucose plus fructose (G + F), pH and titratable acidity (TA) in commercial Australian red and white wine samples using mid-infrared (MIR) spectroscopy and attenuated total reflectance (ATR). Wine samples (n = 130) were analysed using an MIR instrument equipped with a single bounce ATR cell. Results from this study demonstrated the capability of ATR-MIR coupled with partial least squares (PLS) regression to measure compositional parameters in wine. The standard errors of prediction (SEP) obtained were of 0.11 (%) for ethanol, 0.0007 for SG, 0.10 for pH, 0.53 g/L for TA, 1.35 g/L for G + F, and 0.12 g/L for VA. Both the sample preparation time of analysis and volume of sample required were considerably reduced compared to the transmission MIR measurements currently used by the wine industry.     

Characterization and Classification of Almond Cultivars by Using Spectroscopic and Thermal Techniques

Three different almond cultivars (Spanish Guara, Marcona, and Butte from U.S.A.) were characterized by using attenuated total reflectance Fourier transform infrared spectroscopy (ATR‐FTIR) and thermal analysis techniques (differential scanning calorimetry (DSC) and thermogravimetric analysis (TGA). All samples were directly analyzed without the need of a previous oil extraction. Similar FTIR bands were observed for all studied cultivars corresponding to specific functional groups characteristics of almond ingredients (water, fat, protein, and carbohydrates). Significant differences were observed between cultivars according to absorbance and maximum wave number values of specific bands observed by FTIR and melting and crystallization parameters obtained by DSC. TGA showed that samples were stable up to around 220 °C. Different stages of degradation were observed with increasing temperature corresponding to the degradation of the complex matrix of the samples. Successful discrimination was obtained for all samples by applying multivariate stepwise linear discriminant analysis (LDA) separately to data obtained from FTIR and DSC. A satisfactory multidisciplinary approach was also performed by inserting together all parameters obtained from the 3 techniques as predictors ensuring higher reliability of the obtained model. The obtained results proved the suitability of the studied analytical techniques combined with LDA for an easy and fast discrimination among different almond cultivars in food processing. Practical Application: The study of spectroscopic and thermal parameters could be used as a control tool for the direct and fast assessment of almond samples in food processing, particularly for protected designation of origin products.     

Classification of moroccan olive cultivars by linear discriminant analysis applied to ATR–FTIR spectra of endocarps

The potential of FTIR combined with chemometrics was studied to classify five Moroccan varieties of olives by analysis on the endocarps. Attenuated total reflectance (ATR) enabled the samples to be examined directly in the solid state. The spectral data were subjected to a preliminary derivative elaboration based on the Norris gap algorithm to reduce the noise and extract larger analytical information. Linear discriminant analysis (LDA) was adopted as classification method, and Principle component analysis (PCA) was employed to compress the original data set into a reduced new set of variables before LDA. The calibration set was built by using the IR data from seventy‐five samples scanned in reflectance mode, and the ranges 3000–2400 and 2300–600 cm^?1 were selected because furnishing the most useful analytical information. PCA allowed clustering the samples in five classes by using the first two principal components with an explained variance of 98.16%. Application of LDA on an external test set of twenty‐five samples enabled to classify them into five variety groups with a correct classification of 92.0%.