- Email: [email protected]
Using sensor and spectral analysis to classify botanical origin and determine adulteration of raw honey
Accepted Manuscript Using Sensor and Spectral Analysis to Classify Botanical Origin and Determine Adulteration of Raw Honey Zhilin Gan, Yang Yang, Jing Li, Xin Wen, Minghui Zhu, Yundong Jiang, Yuanying Ni PII:
S0260-8774(16)30016-4
DOI:
10.1016/j.jfoodeng.2016.01.016
Reference:
JFOE 8454
To appear in:
Journal of Food Engineering
Received Date: 10 September 2015 Revised Date:
18 January 2016
Accepted Date: 19 January 2016
Please cite this article as: Gan, Z., Yang, Y., Li, J., Wen, X., Zhu, M., Jiang, Y., Ni, Y., Using Sensor and Spectral Analysis to Classify Botanical Origin and Determine Adulteration of Raw Honey, Journal of Food Engineering (2016), doi: 10.1016/j.jfoodeng.2016.01.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Using Sensor and Spectral Analysis to Classify Botanical Origin and
2
Determine Adulteration of Raw Honey
3
Zhilin Gana,b, Yang Yangc, Jing Lid, Xin Wena,b, Minghui Zhua,b, Yundong Jianga,b,
4
Yuanying Ni*, a, b
5
a
6
Beijing 100083, China
7
b
8
University, Beijing 100083, China
9
c
10
d
11
100190, China
College of Food Science and Nutritional Engineering, China Agricultural
SC
Cloud Live Technology Group Co., Ltd., Beijing 100085, China
M AN U
Institute of Electrical Engineering of the Chinese Academy of Sciences, Beijing
12 13 14
18 19 20
EP
17
TE D
15 16
RI PT
National Engineering Research Center for Fruits and Vegetables Processing,
*Corresponding author. Professor of College of Food Science and Nutritional
22
Engineering, China Agricultural University; Vice Director of National
23
Engineering Research Center for Fruits and Vegetables Processing.
24
E-mail address: [email protected] (Yuanying Ni)
25 26 27 28
AC C
21
ACCEPTED MANUSCRIPT Abstract:
30
The feasibility of sensors (Electronic Nose, EN; Electronic Tongue, ET) and spectra (Near Infrared
31
spectrum, NIR; Mid Infrared spectrum, MIR) to evaluate raw honey samples (Vitex, Jujube and Acacia)
32
was explored. Partial least squares discriminant analysis (PLSDA) model, support vector machine (SVM)
33
algorithms model and Interval partial least squares (iPLS) model were used to classify the botanical
34
origin. The results indicate that spectra and sensors could classify the botanical origin of honey rapidly
35
and accurately, since total accuracy for calibration and prediction sets was all almost 100% in EN and ET
36
analysis by SVM model and in NIR and MIR analysis by iPLS model. Then principal components
37
analysis (PCA) and PLSDA model were used to determine the adulterants. Total accuracy for calibration
38
and prediction sets was all above 96% in NIR, MIR and ET by PLSDA model. The results indicate that
39
ET is more suitable for detecting honey adulteration.
40
Key words: Honey, Electronic nose, Electronic tongue, NIR, MIR
M AN U
41 42
AC C
EP
TE D
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
SC
RI PT
29
ACCEPTED MANUSCRIPT 1. Introduction Honey is a natural food product processed by honey bees by blending the sweetened sap collected from flowers with metabolic gastric enzymes (Anjos et al., 2015). Natural honey is a very nutritious food product, which contains water (17% in general), saccharides (main constituents in honey, 75% in general, mainly glucose and fructose), amino acids, minerals (Mg,
RI PT
Ca, K, Na, S, P, Fe, Mn, Co, Ni, etc.), vitamins, enzymes (invertase, catalase, amylase, etc.), phenols, organic acids, pigments, volatile oils and also over 100 varieties of aromatic substances (De la Fuente et al., 2011; Mateo et al., 1997; Ouchemoukh et al., 2010; Arvanitoyannis et al., 2005; Baroni et al., 2006). Honey needs to be rapidly evaluated and priced after collection from a bee-house, based on botanical
SC
origin classification and adulterant determination. Honey has customarily been distinguished as unifloral and multifloral depending on the botanical origin (Lenhardt et al., 2014). In China, over 20 kinds of unifloral honey are commonly produced such as jujube, vitex, acacia, rape, and linden. In general,
M AN U
unifloral honey types have higher market value due to their limited production and availability. Traditionally, botanical origin was determined by sensory analysis, pollen morphology, characteristic aroma analysis and characteristic interior components analysis. These methods have shown that botanical origin classification of honey is affected by color, aroma, content of pollen, method of processing and storage, content of trace substances and so on. However, these methods are expensive and time-consuming. Although the results of sensory analysis are accurate, it is objective and needs a well trained certified taste panel with a time investment. (Arvanitoyannis et al., 2005;Aparna et al., 1999;
TE D
Lawal et al., 2009; Bogdanov et al., 2004; Bianchi et al., 2005; Iglesias et al., 2004; Martos et al., 2000; Gonzlez-Miret et al., 2005; Conti et al., 2007).
For unethical economic gain, honey adulteration is an obvious problem in the market. Water, sucrose, inverted sugar, hydroxymethyl cellulose, dextrin and starch are adulterants which have been regularly
EP
identified by regular physicochemical analysis (Serrano et al., 2004). High performance liquid chromatography (HPLC), isotope mass spectrometry and capillary electrophoresis have also been used to evaluate these. However, these methods are complicated, time and labor consuming (Morales et al., 2008; Tu et al., 2011; Luo et al., 2012). C-4 botanical glycosides like corn syrup were added to honey
AC C
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
subsequently to avoid detection. Even though corn syrup can be detected by stable carbon isotope ratio analysis (SCIRA), it is too expensive and time consuming (Cotte et al., 2007). Moreover, SCIRA is not suitable for detection of C-3 botanical glycoside such as rice syrup. In recent years, rapid detection techniques such as spectral technology, sensor technology and chemical kits are widely used in testing of species, habitat, grades, freshness, nutrient quality, and drug residues in agricultural products since they are time-saving, more convenient and accurate than traditional methods. Electronic nose (EN) and electronic tongue (ET) are common methods for food analysis (Cozzolino et al., 2008; Wang et al., 2009; Ghasemi-Varnamkhasti et al., 2010; Vlasov et al., 2002; Lu et al., 2014; Pan et al., 2014; Haddi et al., 2013). A sample’s whole information so called fingerprint data, can be determined from EN and ET rather than qualification and quantification of some specific constituents. Then a
ACCEPTED MANUSCRIPT discriminant model will be established by compiling fingerprint data and unknown samples could be determined by the model. In previous studies, based on volatile components collected by solid phase micro-extraction, botanical origin of honey has been detected by Mass spectrum-electronic nose (MS-EN) (Ampuero et al., 2004). The bp-ANN model founded by EN consisting of ten MOSFET and twelve MOS sensors could discriminate the botanical origin and geographical honey production area (Benedetti et al., 2004). αAstree-ET could classify botanical origin of honey as well, in which the accuracy of the artificial neural network model was 100% (Major et al., 2011). Also PCA and ANN
RI PT
models were effective in classification of honey botanical origin by using ET with metallic compound electrode (Escriche et al., 2012).
NIR and MIR are extensively performed in agricultural products, food and medicine analysis (Magwaza et al., 2011; Cen et al., 2007; Balabin et al., 2011). Much research has revealed that NIR could classify
SC
the botanical origin of honey by using PCA analysis, canonical variate analysis, PLS discriminant analysis and linear discriminant analysis (Davies et al., 2002; Ruoff et al., 2007). Some reports showed that NIR was also an effective method to discriminate the adulterant of honey. Irish honey adulterated
M AN U
with beet syrup and high fructose syrup has been correctly discriminated and adulterant ratio could be predicted by using PLS analysis (Kelly et al., 2006a). Honey adulterated with different ratio of fructose/ glucose was also determined by PLSR, K-NN and SIMCA models (Downey et al., 2003). MIR could discriminate adulterant of honey as well. Honey with glucose, fructose, sucrose and corn syrup added was detected by MIR and the discriminant accuracy could reach 90% by using LDA model (Irudayaraj et al., 2003). In another study, Irish artificial honey adulterated with fully inverted beet syrup, high-fructose corn syrup, partially invert cane syrup, dextrose syrup and beet sucrose was evaluated by MIR and well
TE D
classified by using SIMCA and PLSAD models (Kelly et al., 2006b).
Compared with previous studies, this work focused on evaluating the performance of sensors (EN and ET) and spectra (NIR and MIR) in botanical origin classification and adulterant determination of raw honey. All honey samples used in this research were raw materials obtained directly from a bee-house
EP
and without any processing. Different types of discriminant analysis models were built. Sensors (ET and EN) and spectra (NIR and MIR) showed different accuracy and performance in botanical origin classification and adulterants determination.
AC C
106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
2. Materials and Methods 2.1. Sample preparation
Three types of honey samples, vitex, jujube and acacia, were provided by Beijing Baihua Apiculture Technology Development Company. Sample information is described in Table 1. Each sample, sourced from a different bee-house was filtered by 60 mesh screen in order to eliminate impurities. Samples were stored in the dark at 4 ℃ until analysis. In botanical origin classification, 105 samples were divided into two sets with calibration (79) and prediction (26) as shown in Table 2, respectively. Then 35 pure honey samples (including vitex, jujube
ACCEPTED MANUSCRIPT and acaica) were chosen randomly to prepare adulterant samples by adding two varieties of syrup (rice syrup and corn syrup). Both syrups were purchased from Cargill Investments (China) Co., Ltd. The physicochemical properties of syrups and honey samples were shown in Table 3. Adulterant samples were prepared by mixing pure honey solution with syrup in different concentrations (5%, 10%, 20% and 40%). The samples, 259 in total, included 105 pure honey samples and 154 adulterated samples. The division of sets was described as follow in Table 4.
RI PT
2.2. Physical and chemical analysis of honey and syrup
The conductivity was measured by a Orion 5-Star Benchtop Meter (Thermo Fisher Scientific, Waltham, USA). The pH values were tested by a PB-10 pH meter (Sartorius Corp., Bohemia, NY, USA). Proline was evaluated by spectrophotometry. Saccharides, fructose, gluctose, cane sugar and maltose were
SC
determined by HPLC equipped with a YMC-Pack Polyamine II (250 mm×4.6mm, 5µm) column. All of the physiochemical properties above were conducted by methods modified from previous studies
M AN U
(Bogdanovaet al., 2004; Bogdanova et al., 2002) 2.3. Electronic Nose
The EN system used for this study was a Heracles (Alpha Mos, Toulouse, France) which was equipped with a highly selective and sensitive specialty gas chromatograph. Volatile constituents of honey samples were collected by way of static headspace. Separation columns used in this work were DB5 and DB1701, respectively. Retention time and peak area were detected by hydrogen flame ionization detector
TE D
(FIDs).Three grams of honey sample was weighed into a septa-sealed screw-cap bottle and equilibrated for 10 min at 65 ℃ (500 rpm). The parameters of EN were as follows: (1) Injector: temperature 180 ℃, inject volume 3000 µL. (2) Temperature program: initial 40 ℃, final 200 ℃, heating rate 2.0 ℃/s. (3) Trap temperature: 250 ℃, purge time 15 s. (4) Temperature of FID 220 ℃. (5) Acquisition time: 84 s
EP
(Yüzay et al., 2007). All determinations reported were conducted in triplicate. 2.4. Electronic Tongue
The sensor array of the α-Astree ET (Alpha Mos, Toulouse, France) was comprised of seven
AC C
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189
potentiometric chemical sensors (ZZ2808, JE5292, BB2011, CA5292, GA2808, HA2808, JB2808) and an Ag/AgCl standard electrode. The cross-selectivity of sensors was first checked (Yang et al., 2013). The mass for each honey was 60 g, which was dissolved in an airtight glass jars with a volume of 150 mL (concentration chamber), 80 mL sample each (Wei et al., 2009). The acquisition time of one evaluation was 120 s. Data was recorded every second. Then the mean value recorded between 110 s and 120 s was determined. All determinations reported were conducted in triplicate. 2.5. Near Infrared Spectrum (NIR) NIR spectra of the honey samples were obtained by using a FT-NIR system (Bruker Optics, Inc, Germany). Samples were scanned in the region 10000-4000 cm-1 in transmittance mode by a Fiber Optic
ACCEPTED MANUSCRIPT Liquid Probe at room temperature. Each spectrum was obtained by averaging 64 scans. NIR spectral data were stored in absorbance format with a resolution of 8 cm-1 (Chen et al., 2012). 2.6. Mid-infrared spectrum (MIR) MIR spectra of honey samples were obtained by a FT-IR (Perkin Elmer, USA) equipped with an Attenuated Total Reflection (ATR). Samples were scanned in the region 4000-650 cm-1 at room absorbance format with a resolution of 4 cm-1 (Yusilawati et al., 2010). 2.7. Data Analysis
RI PT
temperature. Each spectrum was obtained by averaging 4 scans. MIR spectral data were stored in
SC
All the pure honey and adulterant honey samples were detected by EN, ET, NIR and MIR. Data analysis was performed by Matlab 7.8.0. Before building chemometrics models, raw data were firstly pretreated (sensor data: standard normal variate (SNV) and auto-scale; spectra data: SNV, smoothing, auto-scale
M AN U
and derivatives). Partial least squares discriminant analysis (PLSDA) models were built for botanical origin classification and adulterant determination. Support vector machine discriminant analysis (SVMDA) and Interval Partial Least Squares (iPLS) were applied to optimize the results of sensors (EN, ET) and spectra (NIR, MIR), respectively. Principal components analysis (PCA) was used to reduce the dimensions of data. Outliers were eliminated before the foundation of every model.
TE D
3. Results
3.1. Physicochemical properties of honey and syrup The mean values of the physicochemical parameters are shown in Table 3. Generally, the ratios and constituents of saccharides and pH were very similar in honey and syrup, while conductivity and proline
EP
were the more meaningful parameters to evaluate honey and syrup due to the big difference. The difference in conductivity is likely due to natural honey’s high mineral content. (Mateo et al., 1998) 3.2. Botanical origin classification
AC C
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
3.2.1. PLSDA model of EN, ET, NIR and MIR PLSDA model of EN was established based on discriminant analysis results of seven principal components by using 20 sensors. PLSDA model of ET was found according to the results from seven sensors’ evaluation. Five principal components were then chosen to optimize the results. PLSDA model of NIR was built by discriminant analysis results of ten principle components. PLSDA model of MIR was established by discriminant analysis results of five principle components. The PLSDA model results of these four methods were shown in Table 5. It was shown that the error rate
ACCEPTED MANUSCRIPT was less than one in every method, both in calibration set and prediction set. In the MIR study, the accuracy of calibration and prediction reached 100% for all samples. Compared with MIR, misjudgment and outliers were still present in the NIR, EN and ET studies. The results indicated that MIR performed best using the PLSDA model. 3.2.2. SVMDA model of sensors
RI PT
Type C of SVM was applied in this study and the kernel function chosen was RBF (Radial basis function). Cost and gamma, the main parameters of this function, were 100, 0.001 in EN and 100, 0.0031623 in ET. Also SVs was 28 and 20 in EN and ET, respectively. Although the results shown in Figure 1 confirmed that one Vitex sample and one Acacia sample which came from the calibration set misjudged Acacia and Jujube in EN and ET, respectively, yet no errors occurred in prediction set. This means the SVM model
SC
can raise the accuracy of botanical origin classification when compared with PLSDA model in EN and ET analysis. Traditionally, PCA, artificial neural network (ANN), linear discriminant analysis (LDA), discriminating factor analysis (DFA) were the best common models applied to EN and ET data for
M AN U
identification of honey botanical origin (Huang et al., 2014; Ampuero et al., 2004; Tiwari et al., 2013; Ulloa et al., 2013). Our work indicates that SVM is another potential model as well. 3.2.3. iPLSDA model of spectra
iPLSDA model was run to optimize the botanical origin classification results by using NIR and MIR. The spectra regions of NIR and MIR were divided into different sections. 6310-5847cm-1 for NIR, 3397-3298
TE D
cm-1，2893-2592 cm-1 and 1381-980 cm-1 for MIR were the optimized region. iPLSDA models of NIR and MIR were found by discriminant analysis results of six principle components. The accuracy was 100% for both the calibration and prediction sets. Compared with a previous study (Chen et al., 2012) which used the BP-ANN model for NIR analysis, iPLSDA model in this work more accurately distinguished the botanical origin of Acacia, Vitex and Jujube. Figure 2 is the Scatter plot for
EP
discriminating honey types by PLSDA based on NIR data ((a) Acacia (b) Jujue (c) Vitex). Only one error occurred in each type of honey. The results indicated that iPLS could optimize the spectra region which reduced computational cost. Moreover, the discrimination among the samples was improved due to the more obvious distribution of different samples. In the optimized region of NIR, O-H secondary multiple
AC C
232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
frequency, C-H secondary multiple frequency and combination band were the predominant characteristics. C-H, O-H, H2O, C-O and C-C stretching bands as well as O-C-H, C-C-H and C-O-H were the prominent information in MIR. 3.3. Adulteration determination 3.3.1. PCA analysis The PCA scatter plot of EN is shown in Fig. 3 (a). Adulterated samples and pure honey samples could not be well discriminated by EN due to the crossover although they were separated to some degree. Also it was inferred that obvious difference of volatile components among part of the samples existed because of
ACCEPTED MANUSCRIPT a few discrete points. There are two reasons for this difference: (1) constitution of volatile substance is different in three kinds of pure honey, even in different samples of the same botanical origin. Every sample was derived from different bee-house, so it could be influenced by the natural environment. (2) The fragrance characteristic has been changed after adding the syrup. So it is difficult for EN to classify adulterant samples since the volatile components are complicated. ET-PCA which is described in Fig. 3(b) shows that pure honey could be divided into three groups
RI PT
according to different botanical origin. The corresponding adulterants were discriminated as three groups as well. Additionally, pure honey and adulterant could be clearly distinguished by ET.
inability to classify adulteration of honey.
SC
As shown in Fig. 3(c), pure honey and adulterant were uniformly distributed leading to NIR-PCA’s
Both division and crossover existed in Fig. 3(d). This means pure honey and adulterant could not be
M AN U
accurately discriminated by MIR in PCA analysis.
It was concluded that honey samples of this study could be better discriminated by ET, compared using EN, NIR and MIR in PCA analysis, no matter whether among adulterant samples, or between pure honey and adulterant. Saccharide intermolecular bonds were the predominant bond type present in the spectra. Based on the result of 3.1, the ratio and constituent of saccharides and pH were very similar in honey and syrup. So adulterant could not be determined by PCA analysis in NIR and MIR. More pretreatment of data and foundation of discriminant models are needed. Combined with 3.1, difference of conductivity
TE D
between pure honey and syrup is mainly because more minerals exist in pure honey. ET is more sensitive to minerals. Furthermore, there is gustatory difference between pure honey and adulterants which can be exactly classified by the sensors of ET. In brief, ET is more suitable for detecting honey adulteration.
EP
3.3.2. PLSDA models
Fewer than 105 and 154 pure honey and adulterant samples, respectively, were usable due to outliers. The results from the PLSDA model for EN are shown in Table 6. Six principle components were
AC C
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315
collected in EN analysis and outliers were eliminated. Misjudgment still existed in calibration and prediction. Thus, EN was probably not suitable to determine the adulteration of honey based on the PLSDA model. The reason could be as follows: (1) Three different kinds of honey collected in this study were substantially different in terms of volatile components. (2) Even samples of same origin which were collected from different area were obviously different according to the EN footprints. (3) The “smell” affected by freshness of samples was different due to the collecting and processing time. ET-PLSDA model was established by five principle components after eliminating three abnormal adulterant samples. The total discriminant accuracy of calibration and prediction were 98.43% and 100%, respectively, as shown in Table 6. That means adulterant can be accurately determined by combining the ET-PLSDA model with the results from ET-PCA, which can be explained by syrup’s lack of trace
ACCEPTED MANUSCRIPT substances such as acids, phenols, enzymes, minerals and amino acids neither such physicochemical properties like pH, conductivity which all exist in honey and are easily detected by ET. However, a previous study (Zakaria et al., 2011), demonstrated that single modality assessment of EN and ET was ineffective to discriminate the adulterated honey by PCA and LDA. Table 7 shows PLSDA results for discriminating honey adulteration using NIR and MIR by pretreatment of SNV+2nd derivative+15point smooth+auto-scale. Except for one misjudgment that appeared in
RI PT
calibration set of MIR, the accuracy of other sets all reached 100% as shown in Table 7. SNV+2nd derivative+15point smooth+auto-scale performed best and was selected to build the models compared to SNV+auto-scale and SNV+1st derivative+15point smooth+auto-scale. It is probably because 2nd derivative can amplify the difference between adulterated honey and pure honey. Thus, PLSDA model is suitable for NIR and MIR. In a previous study (Zhu et al., 2010), NIR combined with chemometrics
SC
methods has been used to detect adulteration of honey samples. Least square support vector machine (LS-SVM) model was established and the total accuracy was 95.1%. Our model has more potential due
M AN U
to its higher accuracy. 4. Conclusions
Botanical origin of honey could be rapidly determined by EN, ET, NIR and MIR. Compared with PLSDA, iPLS and SVM could improve the accuracy of spectra and sensors. It was found that spectra are better than sensors due to higher accuracy. Additionally, samples need pre-treatment in sensors analysis. For instance, aroma of honey should be equilibrated in EN and samples need to be diluted in ET.
TE D
Although sensor technology is better than traditional methods, spectra will probably be more frequently applied in botanical origin classification for honey in real time and on-site. The performed tests show that ET, NIR and MIR could be used in adulterant determination of honey. EN is more sensitive to aroma substances and suitable for testing freshness of honey. Adulterated honey
EP
could not be determined by PCA in NIR and MIR analysis. After pretreatment of second derivative, NIR and MIR could accurately determine the adulteration in honey. ET performed best in PCA analysis. This method is easier and more rapid to test and analyze. Data processing is simpler when using ET as well. Additionally, ET is more sensitive to substances like amino acids, minerals, phenols, monosaccharides
AC C
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357
and disaccharides which exist in honey as shown with NIR and MIR. So it can be inferred that ET may lead to more extensive application for determination of honey adulterants. Further research is needed because few studies have published on this topic. Acknowledgement We thank the Beijing Baihua Apiculture Technology Development Company for providing the raw honey. Also, we are grateful to Prof. Han, Dr. Zhou, and Dr. Lv, who all work in College of Engineering, China Agricultural University, and Bruker Optics, Inc, Germany, for guiding us in the use of NIR and MIR. We acknowledge funding of Natural Science Foundation of China (NO. 31171675) and Special Fund for Agro-scientific Research in the Public Interest (No. 201503142-01) from the Ministry of
ACCEPTED MANUSCRIPT Agriculture of the People's Republic of China. References Ampuero, S., Bogdanov, S., Bosset, J.O., 2004. Classification of unifloral honeys with an MS-based electronic nose using different sampling modes: SHS, SPME and INDEX. Eur. Food Res. Technol. 218, 198–207. Anjos, O., Iglesias, C., Peres, F., Martinez, J., Garcia, A., Taboada, J., 2015. Neural networks applied to discriminate botanical origin of honeys. Food Chem. 175, 128–136.
RI PT
Aparna, A.R., Rajalakshmi, D., 1999. Honey—its characteristics, sensory aspects, and applications. Food Rev. Int. 15, 455–471.
Arvanitoyannis, I.S., Chalhoub, C., Gotsiou, P., Lydakis-Simantiris, N., Kefalas, P., 2005. Novel Quality Control Methods in Conjunction with Chemometrics (Multivariate Analysis) for Detecting Honey Authenticity. Crit. Rev. Food Sci. Nutr. 45, 193–203.
SC
Balabin, R.M., Smirnov, S. V, 2011. Melamine detection by mid- and near-infrared (MIR/NIR) spectroscopy: a quick and sensitive method for dairy products analysis including liquid milk, infant formula, and milk powder. Talanta. 85, 562–568.
M AN U
Baldwin, E.A., Bai, J., Plotto, A., Dea, S., 2011. Electronic noses and tongues: applications for the food and pharmaceutical industries. Sensors. 11, 4744–4766.
Baroni, M.V., Nores, M.L., Díaz, M.D.P., Chiabrando, G.A., Fassano, J.P., Costa, C., Wunderlin, D.A., 2006. Determination of volatile organic compound patterns characteristic of five unifloral honey by solid-phase microextraction-gas chromatography-mass spectrometry coupled to chemometrics. J. Agric. Food Chem. 54, 7235–7241.
Benedetti, S., Mannino, S., Sabatini, A.G., Marcazzan, G.L., 2004. Electronic nose and neural network use for the
TE D
classification of honey. Apidologie. 35, 397–402.
Bianchi, F., Careri, M., Musci, M., 2005. Volatile norisoprenoids as markers of botanical origin of Sardinian strawberry-tree (Arbutus unedo L.) honey: Characterisation of aroma compounds by dynamic headspace extraction and gas chromatography–mass spectrometry. Food Chem. 89, 527–532. Bogdanov, S., 2002. Harmonised methods of the international honey commission. Swiss Bee Research Centre,
EP
FAM, Liebefeld, CH-3003 Bern, Switzerland.
Bogdanov, S., Ruoff, K., Persano Oddo, L., 2004. Physico-chemical methods for the characterisation of unifloral honeys: a review. Apidologie.35, S4–S17. Cen, H., He, Y., 2007. Theory and application of near infrared reflectance spectroscopy in determination of food
AC C
358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400
quality. Trends Food Sci. Technol. 18, 72–83. Chen, L., Wang, J., Ye, Z., Zhao, J., Xue, X., Vander Heyden, Y., Sun, Q., 2012. Classification of Chinese honeys according to their floral origin by near infrared spectroscopy. Food Chem 135, 338–342. Conti, M.E., Stripeikis, J., Campanella, L., Cucina, D., Tudino, M.B., 2007. Characterization of Italian honeys (Marche Region) on the basis of their mineral content and some typical quality parameters. Chem Cent J. 1, 14. Cotte, J.F., Casabianca, H., Lhéritier, J., Perrucchietti, C., Sanglar, C., Waton, H., Grenier-Loustalot, M.F., 2007. Study and validity of 13C stable carbon isotopic ratio analysis by mass spectrometry and 2H site-specific natural isotopic fractionation by nuclear magnetic resonance isotopic measurements to characterize and control the authenticity of honey. Anal. Chim. Acta 582, 125–136. Cozzolino, D., Smyth, H.E., Cynkar, W., Janik, L., Dambergs, R.G., Gishen, M., 2008. Use of direct headspace-mass spectrometry coupled with chemometrics to predict aroma properties in Australian Riesling wine. Anal Chim Acta. 621, 2–7.
ACCEPTED MANUSCRIPT Davies, A. M. C., Radovic, B., Fearn, T., Anklam, E., 2002. A preliminary study on the characterisation of honey by near infrared spectroscopy. Journal of Near Infrared Spectroscopy. 10 (2), 121-135. De la Fuente, E., Ruiz-Matute, A.I., Valencia-Barrera, R.M., Sanz, J., Martínez Castro, I., 2011. Carbohydrate composition of Spanish unifloral honeys. Food Chem. 129, 1483–1489. Downey, G., Fouratier, V., Kelly, J. D., 2003.Detection of honey adulteration by addition of fructose and glucose using near infrared transflectance spectroscopy. Journal of Near Infrared Spectroscopy. 11, 447–456 Escriche, I., Kadar, M., Domenech, E., Gil-Sánchez, L., 2012. A potentiometric electronic tongue for the
Physicochemical parameters and volatile profile. J. Food Eng. 109, 449–456.
RI PT
discrimination of honey according to the botanical origin. Comparison with traditional methodologies:
Ghasemi-Varnamkhasti, M., Mohtasebi, S.S., Siadat, M., 2010. Biomimetic-based odor and taste sensing systems to food quality and safety characterization: An overview on basic principles and recent achievements. J. Food Eng. 100, 377–387.
SC
Gonzlez-Miret, M. L., Terrab, A., Hernanz, D., Fernandez-Recamales, M. A., Heredia, F. J., 2005. Multivariate
correlation between color and mineral composition of honeys and by their botanical origin. Journal of Agriculture and Food Chem. 53 (7), 2574-2580.
M AN U
Haddi, Z., Alami, H., El Bari, N., Tounsi, M., Barhoumi, H., Maaref, A., Jaffrezic-Renault, N., Bouchikhi, B., 2013. Electronic nose and tongue combination for improved classification of Moroccan virgin olive oil profiles. Food Res. Int. 54, 1488–1498.
Huang, L., Liu, H., Zhang, B., Wu, D., 2014. Application of Electronic Nose with Multivariate Analysis and Sensor Selection for Botanical Origin Identification and Quality Determination of Honey. Food Bioprocess Technol. 8, 359–370.
Iglesias, M.T., De Lorenzo, C., Polo, M.D.C., Martín-Álvarez, P.J., Pueyo, E., 2004. Usefulness of Amino Acid
TE D
Composition to Discriminate between Honeydew and Floral Honeys. Application to Honeys from a Small Geographic Area. J. Agric. Food Chem. 52, 84–89.
Irudayaraj, J., Xu, F., Tewari, J., 2003. Rapid determination of invert cane sugar adulteration in honey using FTIR spectroscopy and multivariate analysis. Food Engineering and Physical Properties. 68, 2040-2045. Kelly, J. D., Petisco, C., Downey, G., 2006a. Potential of near infrared transflectance spectroscopy to detect
EP
adulteration of Irish honey by beet invert syrup and high fructose corn syrup. Journal of Near Infrared Spectroscopy.14 (2), 139-146.
Kelly, J.D., Petisco, C., Downey, G., 2006b. Application of Fourier transform midinfrared spectroscopy to the discrimination between Irish artisanal honey and such honey adulterated with various sugar syrups. J. Agric. Food
AC C
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441
Chem. 54, 6166–6171.
Lawal, R. A., Lawal, A. K., Adekalu, J. B., 2009. Physico-chemical studies on adulteration of honey in Nigeria. Pakistan Journal of Biological Sciences. 12 (15), 1080-1084. Lenhardt, L., Zeković, I., Dramićanin, T., Tešić, Ž., Milojković-Opsenica, D.; Dramićanin, M. D., 2014. Authentication of the botanical origin of unifloral honey by infrared spectroscopy coupled with support vector machine algorithm. Physica Scripta. T162, 014042, 1-4. Lu, L., Deng, S., Zhu, Z., Tian, S., 2014. Classification of Rice by Combining Electronic Tongue and Nose. Food Anal. Methods 1893–1902. Luo, D., Luo. H., Xian Y., Wang, B., Wu, W., Chen, Y., Guo, X., Wu, Y., 2012. Identification the Authenticity of no Protein Honey by IRMS. Modern Food Science and Technology. 28, 862-866.
ACCEPTED MANUSCRIPT Magwaza, L.S., Opara, U.L., Nieuwoudt, H., Cronje, P.J.R., Saeys, W., Nicolaï, B., 2011. NIR Spectroscopy Applications for Internal and External Quality Analysis of Citrus Fruit—A Review. Food Bioprocess Technol. 5, 425–444. Major, N., Markovic, K., Krpan, M., Saric, G., Hruskar, M., Vahcic, N., 2011. Rapid honey characterization and botanical classification by an electronic tongue. Talanta. 85, 569–574. Martos, I., Ferreres, F., Tomás-Barberán, F. A., 2000. Identification of flavonoid markers for the botanical origin of Eucalyptus honey. J. Agric. Food Chem. 48, 1498–1502.
RI PT
Mateo, R., Boschreig F., 1997. Sugar profiles of Spanish unifloral honeys. Food Chem. 60 (1), 33-41.
Mateo, R., Boschreig, F., 1998. Classification of Spanish unifloral honeys by discriminant analysis of electrical conductivity, color, water content, sugars, and pH. J. Agric. Food Chem. 46 (2): 393-400
Morales, V., Corzo, N., Sanz, M.L., 2008. HPAEC-PAD oligosaccharide analysis to detect adulterations of honey with sugar syrups. Food Chem. 107, 922–928.
SC
Ouchemoukh, S., Schweitzer, P., Bachir Bey, M., Djoudad-Kadji, H., Louaileche, H., 2010. HPLC sugar profiles of Algerian honeys. Food Chem. 121, 561–568.
Pan, L., Zhang, W., Zhu, N., Mao, S., Tu, K., 2014. Early detection and classification of pathogenic fungal disease
162–168.
M AN U
in post-harvest strawberry fruit by electronic nose and gas chromatography-mass spectrometry. Food Res. Int. 62,
Ruge, C., Changzhong, R., Zaigui, L., 2011. The Effects of Different Inactivation Treatments on the Storage Properties and Sensory Quality of Naked Oat. Food Bioprocess Technol. 5, 1853–1859. Ruoff, K., Luginbühl, W., Bogdanov, S., Bosset, J.-O., Estermann, B., Ziolko, T., Kheradmandan, S., Amad, R., 2007. Quantitative determination of physical and chemical measurands in honey by near-infrared spectrometry. Eur. Food Res. Technol. 225, 415–423.
TE D
Serrano, S., Villarejo, M., Espejo, R., Jodral, M., 2004. Chemical and physical parameters of Andalusian honey: classification of Citrus and Eucalyptus honeys by discriminant analysis. Food Chem. 87, 619–625. Tiwari, K., Tudu, B., Bandyopadhyay, R., Chatterjee, A., 2013. Identification of monofloral honey using voltammetric electronic tongue. J. Food Eng. 117, 205–210. Tu, Z., Zhu, D., Ji, B., Chen, H., Qing, Z., 2011. Adulteration detection of honey based on near-infrared
EP
spectroscopy. Transactions of the Chinese Society of Agricultural Engineering. 27, 382-388. Ulloa, P.A., Guerra, R., Cavaco, A.M., Rosa da Costa, A.M., Figueira, A.C., Brigas, A.F., 2013. Determination of the botanical origin of honey by sensor fusion of impedance e-tongue and optical spectroscopy. Comput. Electron. Agric. 94, 1–11.
AC C
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483
Vlasov, Y., Legin, A., Rudnitskaya, A., 2002. Electronic tongues and their analytical application. Anal Bioanal Chem. 373, 136–146.
Wang, Y., Wang, J., Zhou, B., Lu, Q., 2009. Monitoring storage time and quality attribute of egg based on electronic nose. Anal Chim Acta. 650, 183–188. Wei, Z., Wang, J., Liao, W., 2009. Technique potential for classification of honey by electronic tongue. J. Food Eng. 94, 260–266. Yang, Y., Chen, Q., Shen, C., Zhang, S., Gan, Z., Hu, R., Zhao, J., Ni, Y., 2013. Evaluation of monosodium glutamate, disodium inosinate and guanylate umami taste by an electronic tongue. J. Food Eng. 116, 627–632. Yusilawati, A. N., Maizirwan, M., Hamzah, M.S., Ng, K.H., Wong, C.S., Islamic, I., Lumpur, K., 2010. Surface Modification of Polystyrene Beads by Ultraviolet / Ozone Treatment and its Effect on Gelatin Coating Department of Biotechnology Engineering. Am. J. Appl. Sci. 7, 724–731.
ACCEPTED MANUSCRIPT Yüzay, B.I.E., Selke, S., 2007. Development of Electronic Nose Method for Evaluation of Residual Solvents in Low-density Polyethylene Films #. Packing Technology and Science. 20, 99–112. Zakaria, A., Shakaff, A.Y., Masnan, M.J., Ahmad, M.N., Adom, A.H., Jaafar, M.N., Ghani, S.A., Abdullah, A.H., Aziz, A.H., Kamarudin, L.M., Subari, N., Fikri, N.A., 2011. A biomimetic sensor for the classification of honeys of different floral origin and the detection of adulteration. Sensors. 11, 7799–7822. Zhu, X., Li, S., Shan, Y., Zhang, Z., Li, G., Su, D., Liu, F., 2010. Detection of adulterants such as sweeteners
EP
TE D
M AN U
SC
RI PT
materials in honey using near-infrared spectroscopy and chemometrics. J. Food Eng. 101, 92–97.
AC C
484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499
ACCEPTED MANUSCRIPT Figure 1 Plot for discriminating honey types by SVMDA based on EN and ET data. (a) EN (b) ET
Figure 2 Scatter plot for discriminating honey types by PLSDA based on NIR data. (a) Acacia; (b) Jujube; (c) Vitex Figure 3 PCA score plot of pure and adulterant honeys based on EN, ET, NIR and MIR data. (a) EN (b) ET (c) NIR
AC C
EP
TE D
M AN U
SC
RI PT
(d) MIR
ACCEPTED MANUSCRIPT Table 1 Summaries of all the honey samples Variety
Place of origin (different province of China)
Quantity
Vitex
Liao Ning, Hei Bei, Bei jing, Shan Xi
41
Jujube
Liao Ning, Hei Bei, Shan Dong, Shan Xi
27
Acacia
Shan Xi
37
Table 2 Calibration and prediction sets of pure and adulterated honeys in botanical origin classification Calibration set
Prediction set
Vitex
31
10
Jujube
20
7
Acacia
28
9
Total
RI PT
Type of samples
41
27 41
pH
(μS/cm) Corn
Proline
Fructose
Glucose
Cane sugar
Maltose
(mg/kg)
(g/100mL)
(g/100mL)
(g/100mL)
(g/100mL)
32.66
37.32
nd
nd
42.60
33.32
0.21
nd
39.55
28.91
1.49
1.44
20.41
4.45
46
15.53
4.54
41
265
4.43
293
syrup Rice syrup Honey
M AN U
Conductivity
SC
Table 3 Physicochemical properties of syrups and honey
Table 4 Calibration and prediction sets of pure and adulterant honeys Calibration set
Prediction set
Total
Pure honey samples
79
26
105
Adulterated samples
115
39
154
Adulterated 5%
22
8
30
Adulterated 10%
35
12
47
Adulterated 20%
35
11
46
Adulterated 40%
23
8
31
AC C
EP
TE D
Type of samples
ACCEPTED MANUSCRIPT Table 5 Statistic results of discrimination using PLSDA based on NIR, MIR, EN and ET in botanical origin classification NIR Acacia
MIR
Jujube
Vitex
Acacia
Jujube
EN Vitex
Acacia
ET
Jujube
Vitex
Acacia
Jujube
Vitex
Calibration 28
20
31
28
20
31
28
20
30
28
20
30
0
0
0
0
0
0
0
0
1
0
0
1
CC%
100
100
100
100
100
100
100
100
Prediction CC
8
6
9
9
7
10
9
7
WC
1
1
1
0
0
0
0
0
CC%
88.88
85.71
90
100
100
100
100
100
96.77
100
100
96.77
9
8
6
10
1
1
1
0
90
88.88
85.71
100
SC
(C/ W)C: (correct/wrong) classifications
RI PT
CC WC
Table 6 PLSDA results for discriminating honey adulteration using EN and ET data
Type of samples
ET
Calibration
Prediction
M AN U
EN
Calibration
Prediction
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
ratio %
Total
ratio %
Total
ratio %
Total
ratio %
Total
Pure honey
95.95
3/74
98.72
1/25
97.47
2/79
100
0/26
Adulterated total
93.52
7/108
80.56
7/36
99.11
1/112
100
0/39
Adulterated 5%
95.24
1/21
71.42
2/7
95.45
1/22
100
0/8
Adulterated 10%
93.94
2/33
81.82
2/11
100
0/35
100
0/12
Adulterated 20%
87.88
4/33
80
2/10
100
0/35
100
0/11
Adulterated 40%
100
0/21
87.5
1/8
100
1/23
100
0/8
EP
TE D
Recognition
Table 7 PLSDA results for discriminating honey adulteration using NIR and MIR data MIR
Prediction
Adulterated total
AC C
NIR
100
0/115
100
0/39
99.13
1/115
100
0/39
Adulterated 5%
100
0/22
100
0/8
94.45
1/22
100
0/8
Adulterated 10%
100
0/35
100
0/12
100
0/35
100
0/12
Adulterated 20%
100
0/35
100
0/11
100
0/35
100
0/11
Adulterated 40%
100
0/23
100
0/8
100
0/23
100
0/8
Type of samples
Pure honey
Calibration
Calibration
Prediction
Recognition
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
ratio %
Total
ratio %
Total
ratio %
Total
ratio %
Total
100
0/79
100
0/26
100
1/79
96.15
1/26
ACCEPTED MANUSCRIPT
3
SC
2.2
M AN U
2.4
TE D
2 1.8
EP
1.6
AC C
Class Pred Most Probable
2.6
RI PT
EN data Acacia Jujube Vitex Acacia (Pred) Jujube (Pred) Vitex (Pred)
2.8
1.4 1.2 1
10
20
30
40
50 60 Sample
70
80
90
100
ACCEPTED MANUSCRIPT
3
M AN U
SC
2.4 2.2
TE D
2 1.8
EP
1.6
AC C
Class Pred Most Probable
2.6
RI PT
ET data Acacia Jujube Vitex
2.8
1.4 1.2 1
10
20
30
40
50 60 Sample
70
80
90
100
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
10 EN-PCA Adulterant Pure 192 190 236
4
209
49
EP
-2
128 248 154 255 153 246 166 247 129 239 254 227 245251 253 7 231 237 44240250 249 76256 46205 204 149 252 195 48206 200 219 17937 207 89 66 201 157 233170 148 40 230 39
51
M AN U
0
52
TE D
2
53 210
-4
AC C
Scores on PC 2 (19.64%)
6
RI PT
54
SC
8
-6 -8 -4
27 32 203 50
-2
0
2
4 6 8 Scores on PC 1 (32.55%)
10
12
14
16
0.6
0 -0.1
RI PT 37
211 246247 93 38 73249 31 49 124 64 250 258187 5132 259 164223 25170257253 219 197 155 54 230 254227 127 63 116 220 207
-0.2 -0.3 -0.4 -0.4
8 2367 5 235 4 3 2
1 17 239 1819238 240 12 142 144 233192 107 216 146 213 217 111 191232 145 224 212 194 113 110 97200 148 167171 189 234
EP
0.1
SC
0.2
M AN U
0.3
AC C
Scores on PC 2 (16.00%)
0.4
ET-PCA Adulterant Pure
243 27 30 242 25 23 241 22
TE D
0.5
ACCEPTED MANUSCRIPT
100 231225 166
112 147
168 -0.2
0
0.2 0.4 0.6 Scores on PC 1 (67.00%)
0.8
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights Evaluate the quality of raw honey by using sensors (Electronic Nose, EN; Electronic Tongue, ET) and spectra (Near Infrared, NIR; Mid Infrared, MIR);
RI PT
Classifying botanical origin and determining adulteration of raw honey based on several excellent prediction models; Spectra will probably be more frequently applied in botanical origin classification of honey;
AC C
EP
TE D
M AN U
SC
ET is considered as a more effective tool to determine honey adulterants;
S0260-8774(16)30016-4
DOI:
10.1016/j.jfoodeng.2016.01.016
Reference:
JFOE 8454
To appear in:
Journal of Food Engineering
Received Date: 10 September 2015 Revised Date:
18 January 2016
Accepted Date: 19 January 2016
Please cite this article as: Gan, Z., Yang, Y., Li, J., Wen, X., Zhu, M., Jiang, Y., Ni, Y., Using Sensor and Spectral Analysis to Classify Botanical Origin and Determine Adulteration of Raw Honey, Journal of Food Engineering (2016), doi: 10.1016/j.jfoodeng.2016.01.016. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Using Sensor and Spectral Analysis to Classify Botanical Origin and
2
Determine Adulteration of Raw Honey
3
Zhilin Gana,b, Yang Yangc, Jing Lid, Xin Wena,b, Minghui Zhua,b, Yundong Jianga,b,
4
Yuanying Ni*, a, b
5
a
6
Beijing 100083, China
7
b
8
University, Beijing 100083, China
9
c
10
d
11
100190, China
College of Food Science and Nutritional Engineering, China Agricultural
SC
Cloud Live Technology Group Co., Ltd., Beijing 100085, China
M AN U
Institute of Electrical Engineering of the Chinese Academy of Sciences, Beijing
12 13 14
18 19 20
EP
17
TE D
15 16
RI PT
National Engineering Research Center for Fruits and Vegetables Processing,
*Corresponding author. Professor of College of Food Science and Nutritional
22
Engineering, China Agricultural University; Vice Director of National
23
Engineering Research Center for Fruits and Vegetables Processing.
24
E-mail address: [email protected] (Yuanying Ni)
25 26 27 28
AC C
21
ACCEPTED MANUSCRIPT Abstract:
30
The feasibility of sensors (Electronic Nose, EN; Electronic Tongue, ET) and spectra (Near Infrared
31
spectrum, NIR; Mid Infrared spectrum, MIR) to evaluate raw honey samples (Vitex, Jujube and Acacia)
32
was explored. Partial least squares discriminant analysis (PLSDA) model, support vector machine (SVM)
33
algorithms model and Interval partial least squares (iPLS) model were used to classify the botanical
34
origin. The results indicate that spectra and sensors could classify the botanical origin of honey rapidly
35
and accurately, since total accuracy for calibration and prediction sets was all almost 100% in EN and ET
36
analysis by SVM model and in NIR and MIR analysis by iPLS model. Then principal components
37
analysis (PCA) and PLSDA model were used to determine the adulterants. Total accuracy for calibration
38
and prediction sets was all above 96% in NIR, MIR and ET by PLSDA model. The results indicate that
39
ET is more suitable for detecting honey adulteration.
40
Key words: Honey, Electronic nose, Electronic tongue, NIR, MIR
M AN U
41 42
AC C
EP
TE D
43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
SC
RI PT
29
ACCEPTED MANUSCRIPT 1. Introduction Honey is a natural food product processed by honey bees by blending the sweetened sap collected from flowers with metabolic gastric enzymes (Anjos et al., 2015). Natural honey is a very nutritious food product, which contains water (17% in general), saccharides (main constituents in honey, 75% in general, mainly glucose and fructose), amino acids, minerals (Mg,
RI PT
Ca, K, Na, S, P, Fe, Mn, Co, Ni, etc.), vitamins, enzymes (invertase, catalase, amylase, etc.), phenols, organic acids, pigments, volatile oils and also over 100 varieties of aromatic substances (De la Fuente et al., 2011; Mateo et al., 1997; Ouchemoukh et al., 2010; Arvanitoyannis et al., 2005; Baroni et al., 2006). Honey needs to be rapidly evaluated and priced after collection from a bee-house, based on botanical
SC
origin classification and adulterant determination. Honey has customarily been distinguished as unifloral and multifloral depending on the botanical origin (Lenhardt et al., 2014). In China, over 20 kinds of unifloral honey are commonly produced such as jujube, vitex, acacia, rape, and linden. In general,
M AN U
unifloral honey types have higher market value due to their limited production and availability. Traditionally, botanical origin was determined by sensory analysis, pollen morphology, characteristic aroma analysis and characteristic interior components analysis. These methods have shown that botanical origin classification of honey is affected by color, aroma, content of pollen, method of processing and storage, content of trace substances and so on. However, these methods are expensive and time-consuming. Although the results of sensory analysis are accurate, it is objective and needs a well trained certified taste panel with a time investment. (Arvanitoyannis et al., 2005;Aparna et al., 1999;
TE D
Lawal et al., 2009; Bogdanov et al., 2004; Bianchi et al., 2005; Iglesias et al., 2004; Martos et al., 2000; Gonzlez-Miret et al., 2005; Conti et al., 2007).
For unethical economic gain, honey adulteration is an obvious problem in the market. Water, sucrose, inverted sugar, hydroxymethyl cellulose, dextrin and starch are adulterants which have been regularly
EP
identified by regular physicochemical analysis (Serrano et al., 2004). High performance liquid chromatography (HPLC), isotope mass spectrometry and capillary electrophoresis have also been used to evaluate these. However, these methods are complicated, time and labor consuming (Morales et al., 2008; Tu et al., 2011; Luo et al., 2012). C-4 botanical glycosides like corn syrup were added to honey
AC C
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105
subsequently to avoid detection. Even though corn syrup can be detected by stable carbon isotope ratio analysis (SCIRA), it is too expensive and time consuming (Cotte et al., 2007). Moreover, SCIRA is not suitable for detection of C-3 botanical glycoside such as rice syrup. In recent years, rapid detection techniques such as spectral technology, sensor technology and chemical kits are widely used in testing of species, habitat, grades, freshness, nutrient quality, and drug residues in agricultural products since they are time-saving, more convenient and accurate than traditional methods. Electronic nose (EN) and electronic tongue (ET) are common methods for food analysis (Cozzolino et al., 2008; Wang et al., 2009; Ghasemi-Varnamkhasti et al., 2010; Vlasov et al., 2002; Lu et al., 2014; Pan et al., 2014; Haddi et al., 2013). A sample’s whole information so called fingerprint data, can be determined from EN and ET rather than qualification and quantification of some specific constituents. Then a
ACCEPTED MANUSCRIPT discriminant model will be established by compiling fingerprint data and unknown samples could be determined by the model. In previous studies, based on volatile components collected by solid phase micro-extraction, botanical origin of honey has been detected by Mass spectrum-electronic nose (MS-EN) (Ampuero et al., 2004). The bp-ANN model founded by EN consisting of ten MOSFET and twelve MOS sensors could discriminate the botanical origin and geographical honey production area (Benedetti et al., 2004). αAstree-ET could classify botanical origin of honey as well, in which the accuracy of the artificial neural network model was 100% (Major et al., 2011). Also PCA and ANN
RI PT
models were effective in classification of honey botanical origin by using ET with metallic compound electrode (Escriche et al., 2012).
NIR and MIR are extensively performed in agricultural products, food and medicine analysis (Magwaza et al., 2011; Cen et al., 2007; Balabin et al., 2011). Much research has revealed that NIR could classify
SC
the botanical origin of honey by using PCA analysis, canonical variate analysis, PLS discriminant analysis and linear discriminant analysis (Davies et al., 2002; Ruoff et al., 2007). Some reports showed that NIR was also an effective method to discriminate the adulterant of honey. Irish honey adulterated
M AN U
with beet syrup and high fructose syrup has been correctly discriminated and adulterant ratio could be predicted by using PLS analysis (Kelly et al., 2006a). Honey adulterated with different ratio of fructose/ glucose was also determined by PLSR, K-NN and SIMCA models (Downey et al., 2003). MIR could discriminate adulterant of honey as well. Honey with glucose, fructose, sucrose and corn syrup added was detected by MIR and the discriminant accuracy could reach 90% by using LDA model (Irudayaraj et al., 2003). In another study, Irish artificial honey adulterated with fully inverted beet syrup, high-fructose corn syrup, partially invert cane syrup, dextrose syrup and beet sucrose was evaluated by MIR and well
TE D
classified by using SIMCA and PLSAD models (Kelly et al., 2006b).
Compared with previous studies, this work focused on evaluating the performance of sensors (EN and ET) and spectra (NIR and MIR) in botanical origin classification and adulterant determination of raw honey. All honey samples used in this research were raw materials obtained directly from a bee-house
EP
and without any processing. Different types of discriminant analysis models were built. Sensors (ET and EN) and spectra (NIR and MIR) showed different accuracy and performance in botanical origin classification and adulterants determination.
AC C
106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147
2. Materials and Methods 2.1. Sample preparation
Three types of honey samples, vitex, jujube and acacia, were provided by Beijing Baihua Apiculture Technology Development Company. Sample information is described in Table 1. Each sample, sourced from a different bee-house was filtered by 60 mesh screen in order to eliminate impurities. Samples were stored in the dark at 4 ℃ until analysis. In botanical origin classification, 105 samples were divided into two sets with calibration (79) and prediction (26) as shown in Table 2, respectively. Then 35 pure honey samples (including vitex, jujube
ACCEPTED MANUSCRIPT and acaica) were chosen randomly to prepare adulterant samples by adding two varieties of syrup (rice syrup and corn syrup). Both syrups were purchased from Cargill Investments (China) Co., Ltd. The physicochemical properties of syrups and honey samples were shown in Table 3. Adulterant samples were prepared by mixing pure honey solution with syrup in different concentrations (5%, 10%, 20% and 40%). The samples, 259 in total, included 105 pure honey samples and 154 adulterated samples. The division of sets was described as follow in Table 4.
RI PT
2.2. Physical and chemical analysis of honey and syrup
The conductivity was measured by a Orion 5-Star Benchtop Meter (Thermo Fisher Scientific, Waltham, USA). The pH values were tested by a PB-10 pH meter (Sartorius Corp., Bohemia, NY, USA). Proline was evaluated by spectrophotometry. Saccharides, fructose, gluctose, cane sugar and maltose were
SC
determined by HPLC equipped with a YMC-Pack Polyamine II (250 mm×4.6mm, 5µm) column. All of the physiochemical properties above were conducted by methods modified from previous studies
M AN U
(Bogdanovaet al., 2004; Bogdanova et al., 2002) 2.3. Electronic Nose
The EN system used for this study was a Heracles (Alpha Mos, Toulouse, France) which was equipped with a highly selective and sensitive specialty gas chromatograph. Volatile constituents of honey samples were collected by way of static headspace. Separation columns used in this work were DB5 and DB1701, respectively. Retention time and peak area were detected by hydrogen flame ionization detector
TE D
(FIDs).Three grams of honey sample was weighed into a septa-sealed screw-cap bottle and equilibrated for 10 min at 65 ℃ (500 rpm). The parameters of EN were as follows: (1) Injector: temperature 180 ℃, inject volume 3000 µL. (2) Temperature program: initial 40 ℃, final 200 ℃, heating rate 2.0 ℃/s. (3) Trap temperature: 250 ℃, purge time 15 s. (4) Temperature of FID 220 ℃. (5) Acquisition time: 84 s
EP
(Yüzay et al., 2007). All determinations reported were conducted in triplicate. 2.4. Electronic Tongue
The sensor array of the α-Astree ET (Alpha Mos, Toulouse, France) was comprised of seven
AC C
148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189
potentiometric chemical sensors (ZZ2808, JE5292, BB2011, CA5292, GA2808, HA2808, JB2808) and an Ag/AgCl standard electrode. The cross-selectivity of sensors was first checked (Yang et al., 2013). The mass for each honey was 60 g, which was dissolved in an airtight glass jars with a volume of 150 mL (concentration chamber), 80 mL sample each (Wei et al., 2009). The acquisition time of one evaluation was 120 s. Data was recorded every second. Then the mean value recorded between 110 s and 120 s was determined. All determinations reported were conducted in triplicate. 2.5. Near Infrared Spectrum (NIR) NIR spectra of the honey samples were obtained by using a FT-NIR system (Bruker Optics, Inc, Germany). Samples were scanned in the region 10000-4000 cm-1 in transmittance mode by a Fiber Optic
ACCEPTED MANUSCRIPT Liquid Probe at room temperature. Each spectrum was obtained by averaging 64 scans. NIR spectral data were stored in absorbance format with a resolution of 8 cm-1 (Chen et al., 2012). 2.6. Mid-infrared spectrum (MIR) MIR spectra of honey samples were obtained by a FT-IR (Perkin Elmer, USA) equipped with an Attenuated Total Reflection (ATR). Samples were scanned in the region 4000-650 cm-1 at room absorbance format with a resolution of 4 cm-1 (Yusilawati et al., 2010). 2.7. Data Analysis
RI PT
temperature. Each spectrum was obtained by averaging 4 scans. MIR spectral data were stored in
SC
All the pure honey and adulterant honey samples were detected by EN, ET, NIR and MIR. Data analysis was performed by Matlab 7.8.0. Before building chemometrics models, raw data were firstly pretreated (sensor data: standard normal variate (SNV) and auto-scale; spectra data: SNV, smoothing, auto-scale
M AN U
and derivatives). Partial least squares discriminant analysis (PLSDA) models were built for botanical origin classification and adulterant determination. Support vector machine discriminant analysis (SVMDA) and Interval Partial Least Squares (iPLS) were applied to optimize the results of sensors (EN, ET) and spectra (NIR, MIR), respectively. Principal components analysis (PCA) was used to reduce the dimensions of data. Outliers were eliminated before the foundation of every model.
TE D
3. Results
3.1. Physicochemical properties of honey and syrup The mean values of the physicochemical parameters are shown in Table 3. Generally, the ratios and constituents of saccharides and pH were very similar in honey and syrup, while conductivity and proline
EP
were the more meaningful parameters to evaluate honey and syrup due to the big difference. The difference in conductivity is likely due to natural honey’s high mineral content. (Mateo et al., 1998) 3.2. Botanical origin classification
AC C
190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231
3.2.1. PLSDA model of EN, ET, NIR and MIR PLSDA model of EN was established based on discriminant analysis results of seven principal components by using 20 sensors. PLSDA model of ET was found according to the results from seven sensors’ evaluation. Five principal components were then chosen to optimize the results. PLSDA model of NIR was built by discriminant analysis results of ten principle components. PLSDA model of MIR was established by discriminant analysis results of five principle components. The PLSDA model results of these four methods were shown in Table 5. It was shown that the error rate
ACCEPTED MANUSCRIPT was less than one in every method, both in calibration set and prediction set. In the MIR study, the accuracy of calibration and prediction reached 100% for all samples. Compared with MIR, misjudgment and outliers were still present in the NIR, EN and ET studies. The results indicated that MIR performed best using the PLSDA model. 3.2.2. SVMDA model of sensors
RI PT
Type C of SVM was applied in this study and the kernel function chosen was RBF (Radial basis function). Cost and gamma, the main parameters of this function, were 100, 0.001 in EN and 100, 0.0031623 in ET. Also SVs was 28 and 20 in EN and ET, respectively. Although the results shown in Figure 1 confirmed that one Vitex sample and one Acacia sample which came from the calibration set misjudged Acacia and Jujube in EN and ET, respectively, yet no errors occurred in prediction set. This means the SVM model
SC
can raise the accuracy of botanical origin classification when compared with PLSDA model in EN and ET analysis. Traditionally, PCA, artificial neural network (ANN), linear discriminant analysis (LDA), discriminating factor analysis (DFA) were the best common models applied to EN and ET data for
M AN U
identification of honey botanical origin (Huang et al., 2014; Ampuero et al., 2004; Tiwari et al., 2013; Ulloa et al., 2013). Our work indicates that SVM is another potential model as well. 3.2.3. iPLSDA model of spectra
iPLSDA model was run to optimize the botanical origin classification results by using NIR and MIR. The spectra regions of NIR and MIR were divided into different sections. 6310-5847cm-1 for NIR, 3397-3298
TE D
cm-1，2893-2592 cm-1 and 1381-980 cm-1 for MIR were the optimized region. iPLSDA models of NIR and MIR were found by discriminant analysis results of six principle components. The accuracy was 100% for both the calibration and prediction sets. Compared with a previous study (Chen et al., 2012) which used the BP-ANN model for NIR analysis, iPLSDA model in this work more accurately distinguished the botanical origin of Acacia, Vitex and Jujube. Figure 2 is the Scatter plot for
EP
discriminating honey types by PLSDA based on NIR data ((a) Acacia (b) Jujue (c) Vitex). Only one error occurred in each type of honey. The results indicated that iPLS could optimize the spectra region which reduced computational cost. Moreover, the discrimination among the samples was improved due to the more obvious distribution of different samples. In the optimized region of NIR, O-H secondary multiple
AC C
232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273
frequency, C-H secondary multiple frequency and combination band were the predominant characteristics. C-H, O-H, H2O, C-O and C-C stretching bands as well as O-C-H, C-C-H and C-O-H were the prominent information in MIR. 3.3. Adulteration determination 3.3.1. PCA analysis The PCA scatter plot of EN is shown in Fig. 3 (a). Adulterated samples and pure honey samples could not be well discriminated by EN due to the crossover although they were separated to some degree. Also it was inferred that obvious difference of volatile components among part of the samples existed because of
ACCEPTED MANUSCRIPT a few discrete points. There are two reasons for this difference: (1) constitution of volatile substance is different in three kinds of pure honey, even in different samples of the same botanical origin. Every sample was derived from different bee-house, so it could be influenced by the natural environment. (2) The fragrance characteristic has been changed after adding the syrup. So it is difficult for EN to classify adulterant samples since the volatile components are complicated. ET-PCA which is described in Fig. 3(b) shows that pure honey could be divided into three groups
RI PT
according to different botanical origin. The corresponding adulterants were discriminated as three groups as well. Additionally, pure honey and adulterant could be clearly distinguished by ET.
inability to classify adulteration of honey.
SC
As shown in Fig. 3(c), pure honey and adulterant were uniformly distributed leading to NIR-PCA’s
Both division and crossover existed in Fig. 3(d). This means pure honey and adulterant could not be
M AN U
accurately discriminated by MIR in PCA analysis.
It was concluded that honey samples of this study could be better discriminated by ET, compared using EN, NIR and MIR in PCA analysis, no matter whether among adulterant samples, or between pure honey and adulterant. Saccharide intermolecular bonds were the predominant bond type present in the spectra. Based on the result of 3.1, the ratio and constituent of saccharides and pH were very similar in honey and syrup. So adulterant could not be determined by PCA analysis in NIR and MIR. More pretreatment of data and foundation of discriminant models are needed. Combined with 3.1, difference of conductivity
TE D
between pure honey and syrup is mainly because more minerals exist in pure honey. ET is more sensitive to minerals. Furthermore, there is gustatory difference between pure honey and adulterants which can be exactly classified by the sensors of ET. In brief, ET is more suitable for detecting honey adulteration.
EP
3.3.2. PLSDA models
Fewer than 105 and 154 pure honey and adulterant samples, respectively, were usable due to outliers. The results from the PLSDA model for EN are shown in Table 6. Six principle components were
AC C
274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315
collected in EN analysis and outliers were eliminated. Misjudgment still existed in calibration and prediction. Thus, EN was probably not suitable to determine the adulteration of honey based on the PLSDA model. The reason could be as follows: (1) Three different kinds of honey collected in this study were substantially different in terms of volatile components. (2) Even samples of same origin which were collected from different area were obviously different according to the EN footprints. (3) The “smell” affected by freshness of samples was different due to the collecting and processing time. ET-PLSDA model was established by five principle components after eliminating three abnormal adulterant samples. The total discriminant accuracy of calibration and prediction were 98.43% and 100%, respectively, as shown in Table 6. That means adulterant can be accurately determined by combining the ET-PLSDA model with the results from ET-PCA, which can be explained by syrup’s lack of trace
ACCEPTED MANUSCRIPT substances such as acids, phenols, enzymes, minerals and amino acids neither such physicochemical properties like pH, conductivity which all exist in honey and are easily detected by ET. However, a previous study (Zakaria et al., 2011), demonstrated that single modality assessment of EN and ET was ineffective to discriminate the adulterated honey by PCA and LDA. Table 7 shows PLSDA results for discriminating honey adulteration using NIR and MIR by pretreatment of SNV+2nd derivative+15point smooth+auto-scale. Except for one misjudgment that appeared in
RI PT
calibration set of MIR, the accuracy of other sets all reached 100% as shown in Table 7. SNV+2nd derivative+15point smooth+auto-scale performed best and was selected to build the models compared to SNV+auto-scale and SNV+1st derivative+15point smooth+auto-scale. It is probably because 2nd derivative can amplify the difference between adulterated honey and pure honey. Thus, PLSDA model is suitable for NIR and MIR. In a previous study (Zhu et al., 2010), NIR combined with chemometrics
SC
methods has been used to detect adulteration of honey samples. Least square support vector machine (LS-SVM) model was established and the total accuracy was 95.1%. Our model has more potential due
M AN U
to its higher accuracy. 4. Conclusions
Botanical origin of honey could be rapidly determined by EN, ET, NIR and MIR. Compared with PLSDA, iPLS and SVM could improve the accuracy of spectra and sensors. It was found that spectra are better than sensors due to higher accuracy. Additionally, samples need pre-treatment in sensors analysis. For instance, aroma of honey should be equilibrated in EN and samples need to be diluted in ET.
TE D
Although sensor technology is better than traditional methods, spectra will probably be more frequently applied in botanical origin classification for honey in real time and on-site. The performed tests show that ET, NIR and MIR could be used in adulterant determination of honey. EN is more sensitive to aroma substances and suitable for testing freshness of honey. Adulterated honey
EP
could not be determined by PCA in NIR and MIR analysis. After pretreatment of second derivative, NIR and MIR could accurately determine the adulteration in honey. ET performed best in PCA analysis. This method is easier and more rapid to test and analyze. Data processing is simpler when using ET as well. Additionally, ET is more sensitive to substances like amino acids, minerals, phenols, monosaccharides
AC C
316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357
and disaccharides which exist in honey as shown with NIR and MIR. So it can be inferred that ET may lead to more extensive application for determination of honey adulterants. Further research is needed because few studies have published on this topic. Acknowledgement We thank the Beijing Baihua Apiculture Technology Development Company for providing the raw honey. Also, we are grateful to Prof. Han, Dr. Zhou, and Dr. Lv, who all work in College of Engineering, China Agricultural University, and Bruker Optics, Inc, Germany, for guiding us in the use of NIR and MIR. We acknowledge funding of Natural Science Foundation of China (NO. 31171675) and Special Fund for Agro-scientific Research in the Public Interest (No. 201503142-01) from the Ministry of
ACCEPTED MANUSCRIPT Agriculture of the People's Republic of China. References Ampuero, S., Bogdanov, S., Bosset, J.O., 2004. Classification of unifloral honeys with an MS-based electronic nose using different sampling modes: SHS, SPME and INDEX. Eur. Food Res. Technol. 218, 198–207. Anjos, O., Iglesias, C., Peres, F., Martinez, J., Garcia, A., Taboada, J., 2015. Neural networks applied to discriminate botanical origin of honeys. Food Chem. 175, 128–136.
RI PT
Aparna, A.R., Rajalakshmi, D., 1999. Honey—its characteristics, sensory aspects, and applications. Food Rev. Int. 15, 455–471.
Arvanitoyannis, I.S., Chalhoub, C., Gotsiou, P., Lydakis-Simantiris, N., Kefalas, P., 2005. Novel Quality Control Methods in Conjunction with Chemometrics (Multivariate Analysis) for Detecting Honey Authenticity. Crit. Rev. Food Sci. Nutr. 45, 193–203.
SC
Balabin, R.M., Smirnov, S. V, 2011. Melamine detection by mid- and near-infrared (MIR/NIR) spectroscopy: a quick and sensitive method for dairy products analysis including liquid milk, infant formula, and milk powder. Talanta. 85, 562–568.
M AN U
Baldwin, E.A., Bai, J., Plotto, A., Dea, S., 2011. Electronic noses and tongues: applications for the food and pharmaceutical industries. Sensors. 11, 4744–4766.
Baroni, M.V., Nores, M.L., Díaz, M.D.P., Chiabrando, G.A., Fassano, J.P., Costa, C., Wunderlin, D.A., 2006. Determination of volatile organic compound patterns characteristic of five unifloral honey by solid-phase microextraction-gas chromatography-mass spectrometry coupled to chemometrics. J. Agric. Food Chem. 54, 7235–7241.
Benedetti, S., Mannino, S., Sabatini, A.G., Marcazzan, G.L., 2004. Electronic nose and neural network use for the
TE D
classification of honey. Apidologie. 35, 397–402.
Bianchi, F., Careri, M., Musci, M., 2005. Volatile norisoprenoids as markers of botanical origin of Sardinian strawberry-tree (Arbutus unedo L.) honey: Characterisation of aroma compounds by dynamic headspace extraction and gas chromatography–mass spectrometry. Food Chem. 89, 527–532. Bogdanov, S., 2002. Harmonised methods of the international honey commission. Swiss Bee Research Centre,
EP
FAM, Liebefeld, CH-3003 Bern, Switzerland.
Bogdanov, S., Ruoff, K., Persano Oddo, L., 2004. Physico-chemical methods for the characterisation of unifloral honeys: a review. Apidologie.35, S4–S17. Cen, H., He, Y., 2007. Theory and application of near infrared reflectance spectroscopy in determination of food
AC C
358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400
quality. Trends Food Sci. Technol. 18, 72–83. Chen, L., Wang, J., Ye, Z., Zhao, J., Xue, X., Vander Heyden, Y., Sun, Q., 2012. Classification of Chinese honeys according to their floral origin by near infrared spectroscopy. Food Chem 135, 338–342. Conti, M.E., Stripeikis, J., Campanella, L., Cucina, D., Tudino, M.B., 2007. Characterization of Italian honeys (Marche Region) on the basis of their mineral content and some typical quality parameters. Chem Cent J. 1, 14. Cotte, J.F., Casabianca, H., Lhéritier, J., Perrucchietti, C., Sanglar, C., Waton, H., Grenier-Loustalot, M.F., 2007. Study and validity of 13C stable carbon isotopic ratio analysis by mass spectrometry and 2H site-specific natural isotopic fractionation by nuclear magnetic resonance isotopic measurements to characterize and control the authenticity of honey. Anal. Chim. Acta 582, 125–136. Cozzolino, D., Smyth, H.E., Cynkar, W., Janik, L., Dambergs, R.G., Gishen, M., 2008. Use of direct headspace-mass spectrometry coupled with chemometrics to predict aroma properties in Australian Riesling wine. Anal Chim Acta. 621, 2–7.
ACCEPTED MANUSCRIPT Davies, A. M. C., Radovic, B., Fearn, T., Anklam, E., 2002. A preliminary study on the characterisation of honey by near infrared spectroscopy. Journal of Near Infrared Spectroscopy. 10 (2), 121-135. De la Fuente, E., Ruiz-Matute, A.I., Valencia-Barrera, R.M., Sanz, J., Martínez Castro, I., 2011. Carbohydrate composition of Spanish unifloral honeys. Food Chem. 129, 1483–1489. Downey, G., Fouratier, V., Kelly, J. D., 2003.Detection of honey adulteration by addition of fructose and glucose using near infrared transflectance spectroscopy. Journal of Near Infrared Spectroscopy. 11, 447–456 Escriche, I., Kadar, M., Domenech, E., Gil-Sánchez, L., 2012. A potentiometric electronic tongue for the
Physicochemical parameters and volatile profile. J. Food Eng. 109, 449–456.
RI PT
discrimination of honey according to the botanical origin. Comparison with traditional methodologies:
Ghasemi-Varnamkhasti, M., Mohtasebi, S.S., Siadat, M., 2010. Biomimetic-based odor and taste sensing systems to food quality and safety characterization: An overview on basic principles and recent achievements. J. Food Eng. 100, 377–387.
SC
Gonzlez-Miret, M. L., Terrab, A., Hernanz, D., Fernandez-Recamales, M. A., Heredia, F. J., 2005. Multivariate
correlation between color and mineral composition of honeys and by their botanical origin. Journal of Agriculture and Food Chem. 53 (7), 2574-2580.
M AN U
Haddi, Z., Alami, H., El Bari, N., Tounsi, M., Barhoumi, H., Maaref, A., Jaffrezic-Renault, N., Bouchikhi, B., 2013. Electronic nose and tongue combination for improved classification of Moroccan virgin olive oil profiles. Food Res. Int. 54, 1488–1498.
Huang, L., Liu, H., Zhang, B., Wu, D., 2014. Application of Electronic Nose with Multivariate Analysis and Sensor Selection for Botanical Origin Identification and Quality Determination of Honey. Food Bioprocess Technol. 8, 359–370.
Iglesias, M.T., De Lorenzo, C., Polo, M.D.C., Martín-Álvarez, P.J., Pueyo, E., 2004. Usefulness of Amino Acid
TE D
Composition to Discriminate between Honeydew and Floral Honeys. Application to Honeys from a Small Geographic Area. J. Agric. Food Chem. 52, 84–89.
Irudayaraj, J., Xu, F., Tewari, J., 2003. Rapid determination of invert cane sugar adulteration in honey using FTIR spectroscopy and multivariate analysis. Food Engineering and Physical Properties. 68, 2040-2045. Kelly, J. D., Petisco, C., Downey, G., 2006a. Potential of near infrared transflectance spectroscopy to detect
EP
adulteration of Irish honey by beet invert syrup and high fructose corn syrup. Journal of Near Infrared Spectroscopy.14 (2), 139-146.
Kelly, J.D., Petisco, C., Downey, G., 2006b. Application of Fourier transform midinfrared spectroscopy to the discrimination between Irish artisanal honey and such honey adulterated with various sugar syrups. J. Agric. Food
AC C
401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441
Chem. 54, 6166–6171.
Lawal, R. A., Lawal, A. K., Adekalu, J. B., 2009. Physico-chemical studies on adulteration of honey in Nigeria. Pakistan Journal of Biological Sciences. 12 (15), 1080-1084. Lenhardt, L., Zeković, I., Dramićanin, T., Tešić, Ž., Milojković-Opsenica, D.; Dramićanin, M. D., 2014. Authentication of the botanical origin of unifloral honey by infrared spectroscopy coupled with support vector machine algorithm. Physica Scripta. T162, 014042, 1-4. Lu, L., Deng, S., Zhu, Z., Tian, S., 2014. Classification of Rice by Combining Electronic Tongue and Nose. Food Anal. Methods 1893–1902. Luo, D., Luo. H., Xian Y., Wang, B., Wu, W., Chen, Y., Guo, X., Wu, Y., 2012. Identification the Authenticity of no Protein Honey by IRMS. Modern Food Science and Technology. 28, 862-866.
ACCEPTED MANUSCRIPT Magwaza, L.S., Opara, U.L., Nieuwoudt, H., Cronje, P.J.R., Saeys, W., Nicolaï, B., 2011. NIR Spectroscopy Applications for Internal and External Quality Analysis of Citrus Fruit—A Review. Food Bioprocess Technol. 5, 425–444. Major, N., Markovic, K., Krpan, M., Saric, G., Hruskar, M., Vahcic, N., 2011. Rapid honey characterization and botanical classification by an electronic tongue. Talanta. 85, 569–574. Martos, I., Ferreres, F., Tomás-Barberán, F. A., 2000. Identification of flavonoid markers for the botanical origin of Eucalyptus honey. J. Agric. Food Chem. 48, 1498–1502.
RI PT
Mateo, R., Boschreig F., 1997. Sugar profiles of Spanish unifloral honeys. Food Chem. 60 (1), 33-41.
Mateo, R., Boschreig, F., 1998. Classification of Spanish unifloral honeys by discriminant analysis of electrical conductivity, color, water content, sugars, and pH. J. Agric. Food Chem. 46 (2): 393-400
Morales, V., Corzo, N., Sanz, M.L., 2008. HPAEC-PAD oligosaccharide analysis to detect adulterations of honey with sugar syrups. Food Chem. 107, 922–928.
SC
Ouchemoukh, S., Schweitzer, P., Bachir Bey, M., Djoudad-Kadji, H., Louaileche, H., 2010. HPLC sugar profiles of Algerian honeys. Food Chem. 121, 561–568.
Pan, L., Zhang, W., Zhu, N., Mao, S., Tu, K., 2014. Early detection and classification of pathogenic fungal disease
162–168.
M AN U
in post-harvest strawberry fruit by electronic nose and gas chromatography-mass spectrometry. Food Res. Int. 62,
Ruge, C., Changzhong, R., Zaigui, L., 2011. The Effects of Different Inactivation Treatments on the Storage Properties and Sensory Quality of Naked Oat. Food Bioprocess Technol. 5, 1853–1859. Ruoff, K., Luginbühl, W., Bogdanov, S., Bosset, J.-O., Estermann, B., Ziolko, T., Kheradmandan, S., Amad, R., 2007. Quantitative determination of physical and chemical measurands in honey by near-infrared spectrometry. Eur. Food Res. Technol. 225, 415–423.
TE D
Serrano, S., Villarejo, M., Espejo, R., Jodral, M., 2004. Chemical and physical parameters of Andalusian honey: classification of Citrus and Eucalyptus honeys by discriminant analysis. Food Chem. 87, 619–625. Tiwari, K., Tudu, B., Bandyopadhyay, R., Chatterjee, A., 2013. Identification of monofloral honey using voltammetric electronic tongue. J. Food Eng. 117, 205–210. Tu, Z., Zhu, D., Ji, B., Chen, H., Qing, Z., 2011. Adulteration detection of honey based on near-infrared
EP
spectroscopy. Transactions of the Chinese Society of Agricultural Engineering. 27, 382-388. Ulloa, P.A., Guerra, R., Cavaco, A.M., Rosa da Costa, A.M., Figueira, A.C., Brigas, A.F., 2013. Determination of the botanical origin of honey by sensor fusion of impedance e-tongue and optical spectroscopy. Comput. Electron. Agric. 94, 1–11.
AC C
442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483
Vlasov, Y., Legin, A., Rudnitskaya, A., 2002. Electronic tongues and their analytical application. Anal Bioanal Chem. 373, 136–146.
Wang, Y., Wang, J., Zhou, B., Lu, Q., 2009. Monitoring storage time and quality attribute of egg based on electronic nose. Anal Chim Acta. 650, 183–188. Wei, Z., Wang, J., Liao, W., 2009. Technique potential for classification of honey by electronic tongue. J. Food Eng. 94, 260–266. Yang, Y., Chen, Q., Shen, C., Zhang, S., Gan, Z., Hu, R., Zhao, J., Ni, Y., 2013. Evaluation of monosodium glutamate, disodium inosinate and guanylate umami taste by an electronic tongue. J. Food Eng. 116, 627–632. Yusilawati, A. N., Maizirwan, M., Hamzah, M.S., Ng, K.H., Wong, C.S., Islamic, I., Lumpur, K., 2010. Surface Modification of Polystyrene Beads by Ultraviolet / Ozone Treatment and its Effect on Gelatin Coating Department of Biotechnology Engineering. Am. J. Appl. Sci. 7, 724–731.
ACCEPTED MANUSCRIPT Yüzay, B.I.E., Selke, S., 2007. Development of Electronic Nose Method for Evaluation of Residual Solvents in Low-density Polyethylene Films #. Packing Technology and Science. 20, 99–112. Zakaria, A., Shakaff, A.Y., Masnan, M.J., Ahmad, M.N., Adom, A.H., Jaafar, M.N., Ghani, S.A., Abdullah, A.H., Aziz, A.H., Kamarudin, L.M., Subari, N., Fikri, N.A., 2011. A biomimetic sensor for the classification of honeys of different floral origin and the detection of adulteration. Sensors. 11, 7799–7822. Zhu, X., Li, S., Shan, Y., Zhang, Z., Li, G., Su, D., Liu, F., 2010. Detection of adulterants such as sweeteners
EP
TE D
M AN U
SC
RI PT
materials in honey using near-infrared spectroscopy and chemometrics. J. Food Eng. 101, 92–97.
AC C
484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499
ACCEPTED MANUSCRIPT Figure 1 Plot for discriminating honey types by SVMDA based on EN and ET data. (a) EN (b) ET
Figure 2 Scatter plot for discriminating honey types by PLSDA based on NIR data. (a) Acacia; (b) Jujube; (c) Vitex Figure 3 PCA score plot of pure and adulterant honeys based on EN, ET, NIR and MIR data. (a) EN (b) ET (c) NIR
AC C
EP
TE D
M AN U
SC
RI PT
(d) MIR
ACCEPTED MANUSCRIPT Table 1 Summaries of all the honey samples Variety
Place of origin (different province of China)
Quantity
Vitex
Liao Ning, Hei Bei, Bei jing, Shan Xi
41
Jujube
Liao Ning, Hei Bei, Shan Dong, Shan Xi
27
Acacia
Shan Xi
37
Table 2 Calibration and prediction sets of pure and adulterated honeys in botanical origin classification Calibration set
Prediction set
Vitex
31
10
Jujube
20
7
Acacia
28
9
Total
RI PT
Type of samples
41
27 41
pH
(μS/cm) Corn
Proline
Fructose
Glucose
Cane sugar
Maltose
(mg/kg)
(g/100mL)
(g/100mL)
(g/100mL)
(g/100mL)
32.66
37.32
nd
nd
42.60
33.32
0.21
nd
39.55
28.91
1.49
1.44
20.41
4.45
46
15.53
4.54
41
265
4.43
293
syrup Rice syrup Honey
M AN U
Conductivity
SC
Table 3 Physicochemical properties of syrups and honey
Table 4 Calibration and prediction sets of pure and adulterant honeys Calibration set
Prediction set
Total
Pure honey samples
79
26
105
Adulterated samples
115
39
154
Adulterated 5%
22
8
30
Adulterated 10%
35
12
47
Adulterated 20%
35
11
46
Adulterated 40%
23
8
31
AC C
EP
TE D
Type of samples
ACCEPTED MANUSCRIPT Table 5 Statistic results of discrimination using PLSDA based on NIR, MIR, EN and ET in botanical origin classification NIR Acacia
MIR
Jujube
Vitex
Acacia
Jujube
EN Vitex
Acacia
ET
Jujube
Vitex
Acacia
Jujube
Vitex
Calibration 28
20
31
28
20
31
28
20
30
28
20
30
0
0
0
0
0
0
0
0
1
0
0
1
CC%
100
100
100
100
100
100
100
100
Prediction CC
8
6
9
9
7
10
9
7
WC
1
1
1
0
0
0
0
0
CC%
88.88
85.71
90
100
100
100
100
100
96.77
100
100
96.77
9
8
6
10
1
1
1
0
90
88.88
85.71
100
SC
(C/ W)C: (correct/wrong) classifications
RI PT
CC WC
Table 6 PLSDA results for discriminating honey adulteration using EN and ET data
Type of samples
ET
Calibration
Prediction
M AN U
EN
Calibration
Prediction
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
ratio %
Total
ratio %
Total
ratio %
Total
ratio %
Total
Pure honey
95.95
3/74
98.72
1/25
97.47
2/79
100
0/26
Adulterated total
93.52
7/108
80.56
7/36
99.11
1/112
100
0/39
Adulterated 5%
95.24
1/21
71.42
2/7
95.45
1/22
100
0/8
Adulterated 10%
93.94
2/33
81.82
2/11
100
0/35
100
0/12
Adulterated 20%
87.88
4/33
80
2/10
100
0/35
100
0/11
Adulterated 40%
100
0/21
87.5
1/8
100
1/23
100
0/8
EP
TE D
Recognition
Table 7 PLSDA results for discriminating honey adulteration using NIR and MIR data MIR
Prediction
Adulterated total
AC C
NIR
100
0/115
100
0/39
99.13
1/115
100
0/39
Adulterated 5%
100
0/22
100
0/8
94.45
1/22
100
0/8
Adulterated 10%
100
0/35
100
0/12
100
0/35
100
0/12
Adulterated 20%
100
0/35
100
0/11
100
0/35
100
0/11
Adulterated 40%
100
0/23
100
0/8
100
0/23
100
0/8
Type of samples
Pure honey
Calibration
Calibration
Prediction
Recognition
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
Recognition
Misjudged/
ratio %
Total
ratio %
Total
ratio %
Total
ratio %
Total
100
0/79
100
0/26
100
1/79
96.15
1/26
ACCEPTED MANUSCRIPT
3
SC
2.2
M AN U
2.4
TE D
2 1.8
EP
1.6
AC C
Class Pred Most Probable
2.6
RI PT
EN data Acacia Jujube Vitex Acacia (Pred) Jujube (Pred) Vitex (Pred)
2.8
1.4 1.2 1
10
20
30
40
50 60 Sample
70
80
90
100
ACCEPTED MANUSCRIPT
3
M AN U
SC
2.4 2.2
TE D
2 1.8
EP
1.6
AC C
Class Pred Most Probable
2.6
RI PT
ET data Acacia Jujube Vitex
2.8
1.4 1.2 1
10
20
30
40
50 60 Sample
70
80
90
100
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
10 EN-PCA Adulterant Pure 192 190 236
4
209
49
EP
-2
128 248 154 255 153 246 166 247 129 239 254 227 245251 253 7 231 237 44240250 249 76256 46205 204 149 252 195 48206 200 219 17937 207 89 66 201 157 233170 148 40 230 39
51
M AN U
0
52
TE D
2
53 210
-4
AC C
Scores on PC 2 (19.64%)
6
RI PT
54
SC
8
-6 -8 -4
27 32 203 50
-2
0
2
4 6 8 Scores on PC 1 (32.55%)
10
12
14
16
0.6
0 -0.1
RI PT 37
211 246247 93 38 73249 31 49 124 64 250 258187 5132 259 164223 25170257253 219 197 155 54 230 254227 127 63 116 220 207
-0.2 -0.3 -0.4 -0.4
8 2367 5 235 4 3 2
1 17 239 1819238 240 12 142 144 233192 107 216 146 213 217 111 191232 145 224 212 194 113 110 97200 148 167171 189 234
EP
0.1
SC
0.2
M AN U
0.3
AC C
Scores on PC 2 (16.00%)
0.4
ET-PCA Adulterant Pure
243 27 30 242 25 23 241 22
TE D
0.5
ACCEPTED MANUSCRIPT
100 231225 166
112 147
168 -0.2
0
0.2 0.4 0.6 Scores on PC 1 (67.00%)
0.8
1
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights Evaluate the quality of raw honey by using sensors (Electronic Nose, EN; Electronic Tongue, ET) and spectra (Near Infrared, NIR; Mid Infrared, MIR);
RI PT
Classifying botanical origin and determining adulteration of raw honey based on several excellent prediction models; Spectra will probably be more frequently applied in botanical origin classification of honey;
AC C
EP
TE D
M AN U
SC
ET is considered as a more effective tool to determine honey adulterants;