and Public Health
Article
The Inorganic Component as a Possible Marker for
Quality and for Authentication of the
Hazelnut’s Origin
Paolo Inaudi1, Agnese Giacomino1,* , Mery Malandrino2 , Carmela La Gioia2, Eleonora Conca2, Tanmoy Karak3and Ornella Abollino1
1 Department of Drug Science and Technology, University of Torino, 10125 Torino, Italy;
paolo.inaudi718@edu.unito.it (P.I.); ornella.abollino@unito.it (O.A.)
2 Department of Chemistry, University of Torino, 10125 Torino, Italy; mery.malandrino@unito.it (M.M.);
carmela.lagioia@unito.it (C.L.G.); eleonora.conca@unito.it (E.C.)
3 Upper Assam Advisory Centre, Tea Research Association, Dikom 786101, Dibrugarh, Assam, India;
t.karak@tocklai.net
* Correspondence: agnese.giacomino@unito.it; Tel.:+39-0116705679
Received: 20 November 2019; Accepted: 7 January 2020; Published: 9 January 2020
Abstract:The inorganic component of hazelnuts was considered as a possible marker for geographical allocation and for the assessment of technological impact on their quality. The analyzed samples were Italian hazelnuts of the cultivar Tonda Gentile Romana and Turkish hazelnuts of the cultivars Tombul, Palaz and Çakildak. The hazelnuts were subjected to different drying procedures and different conservative methods. The concentration of 13 elements, namely Ba, Ca, Cu, Fe, K, Mg, Mn, Na, Ni, P, Sn, Sr and Zn, were quantified by inductively coupled plasma optical emission spectroscopy (ICP-OES). All the samples were previously digested in a microwave oven. Before proceeding with the analysis of the samples, the whole procedure was optimized and tested on a certified reference material. The results show that the inorganic component: (i) can represent a fingerprint, able to identify the geographical origin of hazelnuts, becoming an important quality marker for consumer protection; (ii) is strongly influenced by the treatments undergone by the investigated product during all the processing stages. A pilot study was also carried out on hazelnuts of the cultivar Tonda Gentile Trilobata Piemontese, directly harvested from the plant during early development to maturity and analyzed to monitor the element concentration over time.
Keywords: hazelnuts; metals; authentication; technological impact; marker; chemometric treatments
1. Introduction
The hazelnut is a fruit widely used in the confectionery industry. It is grown worldwide, but primarily in Turkey and Italy; Turkey produces 67% of the total, while Italy ranks second with 13% [1]. The remaining 20% of the production takes place in countries such as the USA, Spain, Azerbaijan, Georgia, and Iran. The United States produces about 90% of their total national amount in the Willamette Valley (Oregon), where they exclusively grow the Barcelona and Ennis varieties. In Spain, hazelnuts are grown in Catalonia, and the most widespread cultivar is Negret, which is mainly used by local industries.
Hazelnut cultivation occupies an important position in Italian agriculture, as Italy produces different quality cultivars such as the Tonda Gentile Trilobata from Piedmont that gained the Protected Geographical Indication (PGI) label under the name Nocciola Piemonte in 1996 from the European Union (EU) [2]. The cultivar Nocciola Romana was awarded the Protected Designation of Origin (PDO) mark in 2009, which refers in particular to the dried fruit, in shell or shelled forms, belonging to the
species Corylus avellana, variety Tonda Gentile Romana and Nocchione, grown in the provinces of Rome and Viterbo.
Nevertheless, in recent years, the production of hazelnuts has also quickly increased in other countries due to large investments for improving agricultural techniques and the selection of new cultivars. This situation has led to a competition between Italian hazelnuts and others, since the former are more expensive, due to their appreciated sensory properties.
Most (90%) of the hazelnuts produced in the world are used in the confectionery industry, so it is important to find a marker for quality authentication [3].
All aspects of the hazelnut have been studied for years; in particular, important studies were carried out in order to find a better method for shelling the fruits [4,5], to minimize the loss of material, for geographical characterization [6,7], and to evaluate the influence of fertilizers [8,9]. Besides, monitoring the concentrations of aflatoxins, highly toxic secondary metabolites produced by fungi present in hazelnuts [10], and analyzing the fatty acid composition [11] is essential. Significant works have been done to evaluate the volatile components present in the aroma of hazelnuts and the influence of roasting on the quality of the hazelnuts [12–14], while others were focused on authentication studies on the basis of the content of lanthanides [15].
Hazelnuts were completely ignored in some studies of trace element level [16,17] surveys. In the last two decades, the situation has gradually improved, with several reports focusing on essential and trace elements in nuts and seeds becoming available [18,19].
This may reflect the increasing recognition of nuts and seeds as important sources of some essential elements (e.g., B, Cu, Fe, Mn, Se and Zn) in human nutrition, especially for vegetarians [19,20], as well as the growing popularity of various healthy diets (including the Mediterranean diet) that rely on frequent use of these constituents.
Macroelement and microelement contents have been studied in hazelnuts [19,21] and similar studies have been carried out in pine nuts [22,23], almonds [24] and nuts [9,25–28].
The aim of this work is to monitor the inorganic components in hazelnuts in order to: (i) evaluate the possibility of using metal composition as a marker of product origin, and ii) to assess the impact that different treatments (drying, storing, roasting) could have on hazelnut composition.
Different techniques are commonly used to assess hazelnut metal composition (i.e., atomic absorption spectroscopy (AAS) [27–29], inductively coupled plasma mass spectrometry (ICP-MS) [30,31] or ICP optical emission spectroscopy (ICP-OES) [19]). Other studies have performed the analysis of hazelnuts for authentication purposes, using, for example, simple sequence repeats (SSR) [32,33], near infrared spectroscopy (NIR) [34], two-dimensional gas chromatography fast scanning quadrupole mass spectrometry [13] (GC–GC-QMS) or liquid chromatography [35] (LC-QqQ-MS/MS), in which
organic substances were determined, and ICP sector field mass spectrometry (ICP-SFMS) [28] and 1H-NMR [36].
In this study, attention was focused on the elements detectable using ICP-OES, namely Ba, Ca, Cu, Fe, K, Mg, Mn, Na, Ni, P, Sn, Sr and Zn, to identify if their concentration is sufficient to determine the origin of hazelnuts. Other techniques, such as ICP-MS, permit us to determine macro-and microelements, rare earth elements, macro-and isotopes, but require longer time, expert personnel macro-and/or higher cost with respect to ICP-OES.
Hazelnuts belonging to different cultivars, namely Italian hazelnuts from the Lazio region (cultivar: Tonda Gentile Romana) and Turkish hazelnuts from the Ordu region (cultivars: Tombul, Palaz and Çakildak) were considered in this work. The choice of these two types of samples was dictated by market needs: Italian confectionery companies have to supplement local hazelnut production by purchasing foreign hazelnuts, especially Turkish ones, to satisfy national needs. In these two countries, different drying procedures were adopted. Samples were provided in order to assess whether the preservation processes had the same impact on hazelnuts belonging to different cultivars and dried in different ways. This aspect is very important to guarantee the quality of the final product independently of the raw material used.
Moreover, some samples were roasted in a laboratory to determine the effect of this process on the inorganic component. Roasting is a crucial step in hazelnut processing, since it has positive effects for food safety (e.g., reduction of allergenicity and aflatoxins, water activity [37,38], improved flavor and color of kernels, and it leads to a desirable crunchy texture [39]). On the contrary, the roasting process affects the composition of the raw fruits. Several papers have been published previously about this aspect, but they are focused on sugars, organic acids, condensed tannins, free phenolic acids and fatty acid profiles [40,41]. No study about the impact of this step on the hazelnut elemental composition is available.
This study was made possible thanks to the access granted by a local farmer: Italian hazelnuts from the Piedmont region (cultivar: Tonda Gentile Trilobata Piemontese) were sampled directly from the plant at different times during the period of ripening of the fruit to study the evolution of the inorganic elements over time. Moreover, fruits harvested both in the middle of the plantation (Pm) and on the roadside (Pr) were compared to see if road traffic is able to influence the concentration of anthropogenic elements typically emitted by road traffic. The study on the Piedmont hazelnut samples was added here to give a complete picture of all the aspects related to the content of metals and other trace elements throughout the processing chain.
A chemometric analysis of the experimental results was performed by principal component analysis (PCA).
2. Materials and Methods
Sample dissolution was performed in polytetrafluoroethylene (PTFE) bombs, with a Milestone MLS-1200 Mega (Milestone, Sorisole, Italy) microwave laboratory unit.
Sample analyses were carried out with a Perkin Elmer Optima 7000 (Perkin Elmer, Norwalk, CT, USA) inductively coupled plasma optical emission spectrometer (ICP-OES).
Analytical grade reagents were used throughout the experiments. Standard metal solutions were prepared from concentrated Merck Titrisol stock solutions (Merck, Darmstadt, Germany).
High purity water (HPW) obtained from a Milli-Q apparatus (Millipore, Burlington, MA, USA) was used for the preparation of samples as well as standard solutions.
2.1. Samples
2.1.1. Reference Material
Unfortunately, no certified reference materials of nuts are available on the market. In order to check the accuracy of the experimental procedure and to assess the effect of vegetal matrices on element determination, the Standard Reference Material (SRM) 1573a of the National Institute of Standards and Technology (NIST), namely Tomato Leaves, was analyzed. The analyzed elements were Al (598 ± 12 mg·kg−1), As (0.112 ± 0.004 mg·kg−1), B (33.3 ± 0.7 mg·kg−1), Ba (63 mg·kg−1), Ca (5.05 ± 0.09%), Cd (1.52 ± 0.04 mg·kg−1), Co (0.57 ± 0.02 mg·kg−1), Cr (1.99 ± 0.06 mg·kg−1), Cu (0.57 ± 0.02 mg·kg−1), Fe (368 ± 7 mg·kg−1), Hg (0.034 ± 0.004 mg·kg−1), K (2.70 ± 0.05%), Mg (1.2%), Mn (246 ± 8 mg·kg−1), Na (136 ± 4 mg·kg−1), Ni (1.59 ± 0.07 mg·kg−1), P (0.216 ± 0.004%), Sb (0.063 ± 0.006 mg·kg−1), Se (0.054 ± 0.003 mg·kg−1), Sr (85 mg·kg−1), V (0.835 ± 0.010 mg·kg−1), and Zn (30.9 ± 0.7 mg·kg−1). 2.1.2. Main Study
Hazelnuts from Turkey, in particular from the Ordu region, belonging to the cultivars Tombul, Palaz and Çakildak (designated as O-samples in this work), and from Italy, in particular from the Lazio region, belonging to the cultivar Tonda Gentile Romana (designated as R-samples in this work), were considered for the main study of this work. All the samples were provided by a confectionery company. All samples had the same caliber, that is an average diameter of 13–15 mm.
The samples were subjected to a drying procedure. The hazelnuts from the Ordu region were purchased from Turkey, so they were dried following the local tradition, i.e., the fruits were sun-dried
in the field immediately after the harvest, typically at 30–35◦C in the local summer climate, for about three weeks. The hazelnuts from Italy were dried by the producer in artificial driers at different temperatures: one batch (R-D1) was dried at 35–40◦C (to simulate the drying conditions applied to the Turkish hazelnut), while a second batch (R-D2) was dried at 15–20◦C following the procedure commonly adopted by the Italian farmer.
By the time the hazelnuts arrive at the confectionery company, they have experienced five different types of storage methods in which atmospheric conditions and temperature were varied: (i) 3–8◦C/humidity: 60–70% (A-Method); (ii) 15–20◦C/humidity: 60–70% (B-Method); (iii) 3–8◦C/humidity: air (C-Method); (iv) 15–20◦C/vacuum (D-Method); (v) temperatures below zero (E-Method).
The samples were analyzed as received from the farmer (T0) and after one, two and four months in each type of storage from the company (T1, T2 and T4). The mass of the hazelnuts does not change during the storage period.
The whole list of considered samples is reported in Table1. Table 1.O-samples and R-samples.
R-Samples Drying 1 R-D1-T0 R-D1-AT1 R-D1-BT1 R-D1-CT1 R-D1-DT1 R-D1-ET1 R-D1-AT2 R-D1-BT2 R-D1-CT2 R-D1-DT2 R-D1-ET2 R-D1-AT4 R-D1-BT4 R-D1-CT4 R-D1-DT4 R-D1-ET4 Drying 2 R-D2-T0 R-D2-AT1 R-D2-BT1 R-D2-CT1 R-D2-DT1 R-D2-ET1 R-D2-AT2 R-D2-BT2 R-D2-CT2 R-D2-DT2 R-D2-ET2 R-D2-AT4 R-D2-BT4 R-D2-CT4 R-D2-DT4 R-D2-ET4 O-Samples O-T0
O-AT1 O-BT1 O-CT1 O-DT1 O-ET1
O-AT2 O-BT2 O-CT2 O-DT2 O-ET2
O-AT4 O-BT4 O-CT4 O-DT4 O-ET4
R: Romana; O: Ordu; T0, T1, T2, T4: 0, 1, 2, 4 months of storage; A-B-C-D-E: storage methods. 2.1.3. Evaluation of the Roasting Step
The roasting conditions generally used for hazelnuts range from 100◦
C to 160◦
C for 10 min to 60 min [42,43]. In this study, some R and O samples were roasted for 15 min at 105◦C to evaluate the effect of this process on the quality of the fruit (Table2).
Table 2.Samples of roasted hazelnuts. Roasted Samples rR-D1-T0 rR-D1-BT4 rR-D1-CT4 rR-D2-T0 rR-D2-BT4 rR-D2-CT4 rO-T0 rO-BT4 rO-CT4
All the hazelnuts were chopped before the digestion and analyzed in duplicate.
2.1.4. Pilot Study on the Trend of the Inorganic Content from Early Development to Maturity
A local company (located in the Piedmont region, Italy) allowed us to sample hazelnuts in their fields. In this case, the fruit belongs to the cultivar Tonda Gentile Trilobata (designated as P-samples in this study). The hazelnuts were collected monthly during the ripening of the fruit (from June to September 2017). These samples were collected separately from trees situated in the middle of the plantation (Pm) and on the roadside (Pr).
P-samples, with the exception of the hazelnuts collected in the first month which had no shell, were shelled immediately after harvesting. All samples were dried in a laboratory stove for one hour at 60◦C, chopped, and then subjected to the same digestion and analysis procedures of the other samples.
2.2. Procedures
For the digestion, aliquots of 0.5 g of each sample were treated with 5 mL of a reaction mixture composed of HNO3and H2O2(4:1), according to the literature [27].
Initially, the SRM sample was used to optimize the whole procedure. Regarding the digestion step, the addition of a known amount of HPW (1 mL or 3 mL) to each aliquot of SRM in the vessel before adding the digestion mixture [44] was tested. A known concentration of lutetium (Lu) was also added to the SRM samples as an internal standard to correct the concentration obtained.
After the heating program (2 min at 250 W; 2 min at 0 W, 6 min at 250 W, 5 min at 400 W, 5 min at 600 W) and ventilation step (25 min), the resulting solutions were filtered on Whatman Grade 5 cellulose filters and transferred into 50 mL Falcon tubes, where HPW was added to a final volume of 30 mL.
The solutions were analyzed by ICP-OES. The calibration was performed with standard solutions prepared in aliquots of sample blanks diluted at the same ratios as the sample solutions. Standard solutions were periodically analyzed and their signals were used to correct the drift of instrumental sensitivity. The limits of detection (LoD) were estimated as three times the standard deviation of the blank.
2.3. Chemometric Treatments
Principal component analysis (PCA) was carried out with the aid of the XLSTAT4.4 software package (Addinsoft, Paris, France), a Microsoft Excel plug-in. Unscrambler X 10.2 (Camo Analytics, Oslo, Norway) was employed for auto-scaling the dataset and for substituting values below LoDs with estimated values.
3. Results
3.1. Choice of Experimental Conditions
In order to optimize the procedures used in this study, 0.5 g of SRM 1573a were digested with and without wetting the sample with 1 mL or 3 mL of HPW before the addition of the HNO3/H2O2mixture. Moreover, the possible correction of the final element concentration with the recovery of the internal standard (Lu) was also tested. Table3shows the recoveries obtained in the different experiments.
It is possible to observe that the best results were obtained by wetting the sample with 3 mL of HPW; no improvement was observed by applying the internal standard correction. Therefore, each sample was wet with 3 mL of HPW before the digestion and the addition of Lu was not performed.
Table 3. Recoveries (%) obtained analyzing Standard Reference Material (SRM) 1573a with different procedures.
Element Without Internal Standard Correction With Internal Standard Correction
0 mL 1 mL 3 mL 0 mL 1 mL 3 mL Al 89.6 90.8 90 54.4 78.3 80.6 Ba 99.3 82.1 84.0 60.2 69.2 75.7 Ca 116 115 111 71.7 113 111 Cd 108 70.3 82.4 65.5 68.5 74.1 Co 122 75.4 79.1 74.1 60.9 71.0 Cr 107 68.3 80.8 64.7 65.6 72.6 Cu 114 90.8 91.2 112 75.8 82.1 Fe 110 86.5 88.4 66.6 72.7 79.6 K 116 112 111 90.1 119 119 Mg 128 97.7 101 69.8 96.4 99.6 Mn 116 90.2 92.3 70.1 75.6 83.1 Ni 109 56.2 66.3 65.9 47.3 59.7 P 118 99.6 101.9 77.6 84.7 91.7 Sr 114 74.8 88.3 70.0 72.8 79.6 V 104 56.1 58.4 63.3 49.0 52.0 Zn 132 78.6 94.5 80.2 75.8 85.1 3.2. Main Study
Table S1 (Supplementary Material) reports the concentrations of Ba, Ca, Cu, Fe, K, Mg, Mn, Na, Ni, P, Sn, Sr and Zn in the analyzed samples. All of the concentrations are related to the dry weight. The concentration of Co and Cr was below the LoD (0.08 mg·kg−1and 0.03 mg·kg−1, respectively) in all the samples. Firstly, the metal content measured for the O- and R-samples at T0, i.e., after the drying step and before the storage step, was compared. A significant difference (by ANOVA) between the two cultivars is observed: Turkish hazelnuts have a higher content of Ba and P, while Romana samples have higher concentrations of Fe, K, Mn, Na, Sn and Sr. The only element for which the concentrations are not significantly different is Cu.
By comparing R-samples after the two different drying procedures, it is possible to observe that R-D2-samples show a higher concentration of Mn and, to a lesser extent, of Fe, Sn and Zn than R-D1, while R-D1 presents a higher content of all the other elements examined.
By calculating the total amount of elements in the samples, it can also be noted that Romana hazelnuts contain a higher metal content than Turkish ones. Applying an ANOVA test, no significant differences were observed between the R-samples dried by the different procedures.
Tables S2–S6 and Tables S7–S16 report the composition of O- and R-samples at different storage times for each type of storage method. The metal content measured at the corresponding T0 is also reported to better observe possible variations during the storage time. A good storage procedure should maintain the characteristics of the fresh hazelnuts, so no variation in term of macro- and micronutrients is expected.
Figure1shows the results obtained for O-samples at the storage time T4, expressed as a percentage in comparison to the concentration found at T0 (horizontal red line: 100%) in order to compare the effect of each storage method.
Regarding O-samples, it is possible to see an increase of the metal content over the time for all the storage methods (with the exception of Fe, Na and Sn for A-Method (3–8◦C/humidity: 60–70%) and Zn for B-Method (15–20◦C/humidity: 60–70%). In the case of Ba, Ca, Cu, Mn, Ni, Sn and Sr, it is possible to observe a marked increase with respect to the initial composition. The explanation for this behavior is not obvious: a possible contamination of the samples from the packaging adopted during the storage time may have occurred. Among these elements, Cu, Mn and Ni are well known for their catalytic effect on the oxidation process, which can cause an oxidative rancidity of the hazelnuts.
Int. J. Environ. Res. Public Health 2020, 17, 447 7 of 14
Int. J. Environ. Res. Public Health 2020, 17, 447 7 of 15
Figure 1. Concentration (%) of each element in the O-samples after four months of storage following the different methods (A–E), expressed with respect to T0 (100%).
Regarding O-samples, it is possible to see an increase of the metal content over the time for all the storage methods (with the exception of Fe, Na and Sn for A-Method (3–8 °C/humidity: 60–70%) and Zn for B-Method (15–20 °C/humidity: 60–70%). In the case of Ba, Ca, Cu, Mn, Ni, Sn and Sr, it is possible to observe a marked increase with respect to the initial composition. The explanation for this behavior is not obvious: a possible contamination of the samples from the packaging adopted during the storage time may have occurred. Among these elements, Cu, Mn and Ni are well known for their catalytic effect on the oxidation process, which can cause an oxidative rancidity of the hazelnuts.
After applying an ANOVA test on the concentration found for O-T4 samples after the different storage methods, only the concentrations of K, Sn and Zn were not significantly different in comparison to O-T0, independently of the storage method adopted. The content of other elements varies randomly depending on the conservative method used.
Regarding the R-samples, Figure 2 shows the concentrations in the T4-specimens, reported as a percentage in comparison with the corresponding concentration in T0-samples.
The total metal content remains quite constant or seems to decrease over the time.
Figure 2. Concentration (%) of each element in the R-samples after four months of storage following the different methods (A–E), expressed with respect to T0 (100%).
0 50 100 150 200 250 300 350 400 450 Ba Ca Cu Fe K Mg Mn Na Ni P Sn Sr Zn % Element
O-samples
A B C D E 0 50 100 150 200 Ba Ca Cu Fe K Mg Mn Ni P Sn Sr Zn % ElementR-samples
A1 B1 C1 D1 E1 A2 B2 C2 D2 E2Figure 1.Concentration (%) of each element in the O-samples after four months of storage following the different methods (A–E), expressed with respect to T0 (100%).
After applying an ANOVA test on the concentration found for O-T4 samples after the different storage methods, only the concentrations of K, Sn and Zn were not significantly different in comparison to O-T0, independently of the storage method adopted. The content of other elements varies randomly depending on the conservative method used.
Regarding the R-samples, Figure2shows the concentrations in the T4-specimens, reported as a percentage in comparison with the corresponding concentration in T0-samples.
The total metal content remains quite constant or seems to decrease over the time.
Figure 1. Concentration (%) of each element in the O-samples after four months of storage following the different methods (A–E), expressed with respect to T0 (100%).
Regarding O-samples, it is possible to see an increase of the metal content over the time for all the storage methods (with the exception of Fe, Na and Sn for A-Method (3–8 °C/humidity: 60–70%) and Zn for B-Method (15–20 °C/humidity: 60–70%). In the case of Ba, Ca, Cu, Mn, Ni, Sn and Sr, it is possible to observe a marked increase with respect to the initial composition. The explanation for this behavior is not obvious: a possible contamination of the samples from the packaging adopted during the storage time may have occurred. Among these elements, Cu, Mn and Ni are well known for their catalytic effect on the oxidation process, which can cause an oxidative rancidity of the hazelnuts.
After applying an ANOVA test on the concentration found for O-T4 samples after the different storage methods, only the concentrations of K, Sn and Zn were not significantly different in comparison to O-T0, independently of the storage method adopted. The content of other elements varies randomly depending on the conservative method used.
Regarding the R-samples, Figure 2 shows the concentrations in the T4-specimens, reported as a percentage in comparison with the corresponding concentration in T0-samples.
The total metal content remains quite constant or seems to decrease over the time.
Figure 2. Concentration (%) of each element in the R-samples after four months of storage following the different methods (A–E), expressed with respect to T0 (100%).
0 50 100 150 200 250 300 350 400 450 Ba Ca Cu Fe K Mg Mn Na Ni P Sn Sr Zn % Element
O-samples
A B C D E 0 50 100 150 200 Ba Ca Cu Fe K Mg Mn Ni P Sn Sr Zn % ElementR-samples
A1 B1 C1 D1 E1 A2 B2 C2 D2 E2Figure 2.Concentration (%) of each element in the R-samples after four months of storage following the different methods (A–E), expressed with respect to T0 (100%).
In regards to the R-D1-samples, after applying an ANOVA: (i) the concentrations of K, Mg, Na, P and Sn T4 samples were not significantly different from T0 for all the conservative methods; (ii) the B-Method (15–20◦C/humidity: 60–70%) seems to be the best in order to preserve the inorganic composition during storage; and (iii) the D-Method appears to be the worst storage procedure, as it causes the greatest variation in the element concentrations.
By applying an ANOVA on R-D2-T4 samples, it is possible to see that the concentration of all elements, with the exception of Ca and Ni, were not significantly different independent of the conservative method adopted. This behavior could be due to a better stabilization of the hazelnut dried at a lower temperature compared to the other samples. On the contrary, it is possible to observe that Ni, a catalyst of the rancidity processes, increases with all methods, except the D-Method (15–20◦C/vacuum).
The A-Method (3–8 ◦C/humidity: 60–70%) seems to be the best, since the concentration of 12 elements in the T4 samples are not significantly different in comparison to D2-T0. It is not possible to define which is the worse storage procedure, since each element has a characteristic trend.
In any case, at the quantitative level, it is difficult to determine which drying or storage procedure should be taken as the optimum.
The experimental data obtained in this study were subjected to principal component analysis (PCA). The biplot obtained by PCA is shown in Figure3(PC1 vs. PC2). The first and the second PCs retain 55.45% and 16.95% of the total variance, respectively.
It is possible to observe a good separation of Italian and Turkish hazelnuts according to their different inorganic composition: O-samples contain the highest levels of Ba, Ni and P, while R-samples are characterized by high concentrations of Fe, K, Mg, Mn, Na, Sn and Sr.
Int. J. Environ. Res. Public Health 2020, 17, 447 9 of 15
Figure 3. Biplot obtained by principle component analysis (PCA) (PC1 vs. PC2) on the whole dataset.
3.3. Roasted Hazelnuts
Tables S17–S19 report the composition of roasted and not roasted samples, stored with method B and C at T0 and T4 for O- (S17), RD1- (S18) and RD2- (S19) samples. In regard to roasted hazelnuts, in the case of rO-samples, there is a significant discrepancy between the concentration of K and P found in roasted and non-roasted samples. It is possible to observe that the concentration of K and Mn doubles in the roasted samples, while the opposite behavior is observable for Ba and P. The decrease in the concentration of Ba and P after the roasting process could be explained by a greater affinity of these elements for the oily phase, which is lost during this treatment. The increase of K and Mn concentrations is difficult to explain [45], but the same behavior of K was observed by Adelakun et al. in roasted okra seeds [46]. No differences in the element concentrations were observed among the samples subjected to the different storage methods. This could indicate that the different conservative processes do not have any effect on the subsequent roasting. Obviously, this behavior should be verified by analyzing a higher number of samples. In the case of rR-(D1 and D2)-samples, pre- and post-roasting concentrations are not significantly different (by ANOVA test). The roasting process is a very important step in the shelf life of the hazelnuts, since it enhances and confers the organoleptic properties of the product, but it is also important that the nutritional value of the fresh nuts is not altered during this step. Romana hazelnuts seem to maintain their original composition better than the Turkish ones. This behavior seems to be different from that observed for other chemical parameters (proximate composition, fatty acids, total polyphenols, antioxidant activity, and protein fingerprint by SDS-PAGE) investigated in samples of Tonda Gentile Trilobata from Italy and from Chile, Tonda di Giffoni from Italy, and Tombul from Turkey by Locatelli et al. [11]. In their study, no differences were observed depending on the hazelnut origin.
OT0 OAT1 OBT1 OCT1 ODT1 OET1 OAT2OBT2 OCT2 ODT2 OET2 OAT4 OBT4 OCT4 ODT4 OET4 RD1T0 RD1AT1 RD1BT1 RD1CT1 RD1DT1 RD1ET1 RD1AT2 RD1BT2 RD1CT2 RD1DT2 RD1ET2 RD1AT4 RD1BT4 RD1CT4 RD1DT4 RD1ET4 RD2T0 RD2AT1 RD2BT1RD2CT1 RD2DT1 RD2ET1 RD2AT2 RD2BT2 RD2CT2 RD2DT2 RD2ET2 RD2AT4 RD2BT4 RD2CT4 RD2DT4 RD2ET4 Ba Ca Cu Fe K Mg Mn Na Ni P Sn Sr Zn -4 -3 -2 -1 0 1 2 3 4 5 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 P C 2 ( 16. 95% ) PC1 (55.45%) Biplot (PC1 and PC2: 72.40%)
Figure 3.Biplot obtained by principle component analysis (PCA) (PC1 vs. PC2) on the whole dataset.
A very interesting result is the good separation of hazelnuts belonging to the same cultivar (Romana), but subjected to different drying processes. This result shows the importance of this step
during the shell life of the product, since it can influence the final composition of the hazelnuts. In particular, using D1 (35–40◦C) results in an increase of the concentration of Ca, Mg and Zn and a decrease of the concentration of Mn, with respect to the use of D2 (15–20◦C).
3.3. Roasted Hazelnuts
Tables S17–S19 report the composition of roasted and not roasted samples, stored with method B and C at T0 and T4 for O- (S17), RD1- (S18) and RD2- (S19) samples. In regard to roasted hazelnuts, in the case of rO-samples, there is a significant discrepancy between the concentration of K and P found in roasted and non-roasted samples. It is possible to observe that the concentration of K and Mn doubles in the roasted samples, while the opposite behavior is observable for Ba and P. The decrease in the concentration of Ba and P after the roasting process could be explained by a greater affinity of these elements for the oily phase, which is lost during this treatment. The increase of K and Mn concentrations is difficult to explain [45], but the same behavior of K was observed by Adelakun et al. in roasted okra seeds [46]. No differences in the element concentrations were observed among the
samples subjected to the different storage methods. This could indicate that the different conservative processes do not have any effect on the subsequent roasting. Obviously, this behavior should be verified by analyzing a higher number of samples. In the case of rR-(D1 and D2)-samples, pre- and post-roasting concentrations are not significantly different (by ANOVA test). The roasting process is a very important step in the shelf life of the hazelnuts, since it enhances and confers the organoleptic properties of the product, but it is also important that the nutritional value of the fresh nuts is not altered during this step. Romana hazelnuts seem to maintain their original composition better than the Turkish ones. This behavior seems to be different from that observed for other chemical parameters (proximate composition, fatty acids, total polyphenols, antioxidant activity, and protein fingerprint by SDS-PAGE) investigated in samples of Tonda Gentile Trilobata from Italy and from Chile, Tonda di Giffoni from Italy, and Tombul from Turkey by Locatelli et al. [11]. In their study, no differences were observed
depending on the hazelnut origin.
In this study, the loss of lipid content was not considered: roasting slightly increased the oil extractability, but this increment was not significant in laboratory tests. As reported from other researchers [47], the impact of the industrial process would presumably be stronger than that of experimental laboratory roasting adopted in scientific studies, causing a higher damage of membranes and/or higher protein denaturation. In any case, the final concentrations refer to the weight of each sample after roasting to value the variation in the final product.
3.4. Pilot Study
The concentrations measured in P-samples collected monthly from the plant during their ripening are reported in Table S20 (hazelnuts sampled in the middle of plantation, Pm) and Table S21 (hazelnuts sampled on the roadside, Pr). No differences can be observed among the trends of the inorganic content obtained for Pm and Pr samples. We applied an ANOVA to our data that shows that the hazelnuts sampled on the roadside in the first month contain a significantly higher concentration of eight elements than those found in Pm samples. This difference was not observed in the hazelnuts sampled during the following months. This behavior reflects the fact that during the first month, the fruit does not possess a shell and is therefore more influenced by the dust composition. In any case, the proximity of the plant to the road does not seem to affect the inorganic composition of the final product. The road near the hazelnut plants is affected only by light traffic, so the input of metals from traffic is low. We conclude that the metals contained in hazelnuts are mainly derived from the soil.
Figure4a,b show the trends of macro- and microelements, respectively, for hazelnuts sampled on the roadside. From the first to the second sampling, the concentrations of all the analytes increase. After the second month, it is possible to observe a decrease in element concentrations, probably caused by a loss of water and the formation of the shell; then, the inorganic content remains quite constant.
Int. J. Environ. Res. Public Health 2020, 17, 447 10 of 14
This behavior is in agreement with the results reported by Seyan et al. [48]. The concentrations of Co and Cr remain unvaried throughout all time points.
In this study, the loss of lipid content was not considered: roasting slightly increased the oil extractability, but this increment was not significant in laboratory tests. As reported from other researchers [47], the impact of the industrial process would presumably be stronger than that of experimental laboratory roasting adopted in scientific studies, causing a higher damage of membranes and/or higher protein denaturation. In any case, the final concentrations refer to the weight of each sample after roasting to value the variation in the final product.
3.4. Pilot Study
The concentrations measured in P-samples collected monthly from the plant during their ripening are reported in Table S20 (hazelnuts sampled in the middle of plantation, Pm) and Table S21 (hazelnuts sampled on the roadside, Pr). No differences can be observed among the trends of the inorganic content obtained for Pm and Pr samples. We applied an ANOVA to our data that shows that the hazelnuts sampled on the roadside in the first month contain a significantly higher concentration of eight elements than those found in Pm samples. This difference was not observed in the hazelnuts sampled during the following months. This behavior reflects the fact that during the first month, the fruit does not possess a shell and is therefore more influenced by the dust composition. In any case, the proximity of the plant to the road does not seem to affect the inorganic composition of the final product. The road near the hazelnut plants is affected only by light traffic, so the input of metals from traffic is low. We conclude that the metals contained in hazelnuts are mainly derived from the soil.
Figure 4a,b show the trends of macro- and microelements, respectively, for hazelnuts sampled on the roadside. From the first to the second sampling, the concentrations of all the analytes increase. After the second month, it is possible to observe a decrease in element concentrations, probably caused by a loss of water and the formation of the shell; then, the inorganic content remains quite constant. This behavior is in agreement with the results reported by Seyan et al. [48]. The concentrations of Co and Cr remain unvaried throughout all time points.
0 5,000 10,000 15,000 20,000 1 2 3 4 mg kg -1 Month
a
Ca K Mg PInt. J. Environ. Res. Public Health 2020, 17, 447 11 of 15
Figure 4. Trends of the element concentrations (mg·kg−1) of (a) macro-nutrients and (b)
micro-nutrients during the development of plants in hazelnuts sampled monthly on the roadside.
A comparison among P-samples and the other cultivars is difficult, since the former were analyzed immediately after the harvest, while O- and R-samples were subjected to drying. The element concentrations found in P-samples harvested at four months was comparable with the concentrations of R-samples, since all Piedmont hazelnuts contain a high level of Fe, K, Mn, Na, Sn and Sr. P-samples are also characterized by high levels of Cu and Ni and by a detectable concentration of Co and Cr. These metals (in particular Ni and Co) probably have geogenic origin, since they derive from the composition of the ultramafic soils widespread in Piedmont [49]. Figure S1 shows the biplot obtained by PCA in which four-month P-samples were added to the dataset of the other cultivars. The first and the second PCs retain 49.64% and 22.29% of the total variance, respectively. Obviously, the number of P-samples should be increased to be representative, but from this preliminary treatment, is possible to observe a good separation of not only Italian and Turkish hazelnuts, but also of hazelnuts belonging to different Italian cultivars. Moreover, is possible to see that R-samples are grouped together (D1 and D2), showing that the elemental composition could be a good fingerprint of each cultivar even after industrial treatment. This confirmed the results reported by Locatelli et al. [11] for other analytes; even if the different processes modify the chemical profile of hazelnuts, this preliminary study suggests that the identification of their origin is still possible.
4. Conclusions
This study has shown how hazelnuts belonging to different cultivars have a well-defined composition from the point of view of inorganic elements. In particular, the samples belonging to
Romana cultivars and coming from the Ordu region differ mainly in the concentration of
macroelements. The Romana hazelnuts are richer in Ca, Fe, K, Mn, Na, Sn, Sr than the Turkish ones, which have higher concentrations of other elements (mainly P and Ba).
Moreover, the treatment and processes involved during the storage and shelf life of the hazelnuts influence the composition of the fruit. In regard to drying processes, it is difficult to assess their specific effect since the three processes (one for O-hazelnuts and two for R-hazelnuts) examined are different and they were applied to hazelnuts belonging to two different cultivars. In any case, the applied temperature and the duration of this step seem to cause a modification of the hazelnut composition during the storage time. Indeed, it is possible to observe that, regardless of the adopted storage method, the Romana hazelnuts dried using the D1 method retain their original concentration of K, Mg, Na, P and Sn over time independently of the method adopted. Drying at a lower temperature (D2) in combination with the A-Method seems to be the best, since the hazelnuts appear to better preserve their nutritional power and quality. Unfortunately, after storage, D2 samples show
0 50 100 150 200 250 1 2 3 4 mg kg -1 Month
b
Ba Co Cr Cu Fe Mn Na Ni Sn Sr ZnFigure 4.Trends of the element concentrations (mg·kg−1) of (a) macro-nutrients and (b) micro-nutrients
during the development of plants in hazelnuts sampled monthly on the roadside.
A comparison among P-samples and the other cultivars is difficult, since the former were analyzed immediately after the harvest, while O- and R-samples were subjected to drying. The element concentrations found in P-samples harvested at four months was comparable with the concentrations of R-samples, since all Piedmont hazelnuts contain a high level of Fe, K, Mn, Na, Sn and Sr. P-samples are also characterized by high levels of Cu and Ni and by a detectable concentration of Co and Cr. These metals (in particular Ni and Co) probably have geogenic origin, since they derive from the composition of the ultramafic soils widespread in Piedmont [49]. Figure S1 shows the biplot obtained by PCA in which four-month P-samples were added to the dataset of the other cultivars. The first and the second PCs retain 49.64% and 22.29% of the total variance, respectively. Obviously, the number of P-samples should be increased to be representative, but from this preliminary treatment, is possible to observe a good separation of not only Italian and Turkish hazelnuts, but also of hazelnuts belonging to different Italian cultivars. Moreover, is possible to see that R-samples are grouped together (D1 and D2), showing that the elemental composition could be a good fingerprint of each cultivar even after industrial treatment. This confirmed the results reported by Locatelli et al. [11] for other analytes; even if the different processes modify the chemical profile of hazelnuts, this preliminary study suggests that the identification of their origin is still possible.
4. Conclusions
This study has shown how hazelnuts belonging to different cultivars have a well-defined composition from the point of view of inorganic elements. In particular, the samples belonging to Romana cultivars and coming from the Ordu region differ mainly in the concentration of macroelements. The Romana hazelnuts are richer in Ca, Fe, K, Mn, Na, Sn, Sr than the Turkish ones, which have higher concentrations of other elements (mainly P and Ba).
Moreover, the treatment and processes involved during the storage and shelf life of the hazelnuts influence the composition of the fruit. In regard to drying processes, it is difficult to assess their specific effect since the three processes (one for O-hazelnuts and two for R-hazelnuts) examined are different and they were applied to hazelnuts belonging to two different cultivars. In any case, the applied temperature and the duration of this step seem to cause a modification of the hazelnut composition during the storage time. Indeed, it is possible to observe that, regardless of the adopted storage method, the Romana hazelnuts dried using the D1 method retain their original concentration of K, Mg, Na, P and Sn over time independently of the method adopted. Drying at a lower temperature (D2) in combination with the A-Method seems to be the best, since the hazelnuts appear to better preserve their nutritional power and quality. Unfortunately, after storage, D2 samples show generally higher concentrations of Ni, a catalyst of the self-oxidative process of the hazelnut, which causes rancidity.
Through the chemometric treatment of the data, a good separation between Ordu and Romana samples was obtained according to their different inorganic composition. The element concentrations depend on the type of cultivar and the environmental conditions (such as soil, water and temperature). Moreover, it is interesting to observe that Romana hazelnuts are grouped on the basis of the drying procedure adopted. No separation was observed among the samples subjected to different storage methods.
In regard to the effect of the roasting process, Romana hazelnuts seem to maintain their original composition better than Turkish ones. This behavior seems to be different from that observed for other chemical parameters, for which no differences were observed in samples pre- and post-roasting depending on the hazelnut’s origin.
Regarding the pilot study carried out on Piedmont hazelnuts, it was possible to appreciate a common trend of the concentration of the inorganic component during their growth: most elements increased during the first months of development, then decreased and finally remained approximately constant until complete maturation. Piedmont hazelnuts have a high K content and a composition similar to that of the other Italian cultivars considered in this study. No influence from vehicular traffic was observed, since no differences between the composition of hazelnuts sampled from plants located in the middle of the field and plants located near to the roadside were observed.
In conclusion, the inorganic component detectable with ICP-OES can be considered a good marker for the recognition of the origin and quality of hazelnuts. Moreover, even if the different processes modify the inorganic profile of hazelnuts, this preliminary study suggests that the statistical identification of the cultivar is still possible. The metal content has to be monitored to evaluate loss of nutrients and, on the contrary, possible contamination during all the steps of the shelf life of the product; that is, from the beginning (environment conditions in which the hazelnuts are grown), during the process (to monitor the effect of the different treatments and manipulations), and at the end (in the final product).
Future work will include: (i) the analysis of hazelnuts of the same cultivar grown in different areas to assess the impact of soil composition, the climate, proximity to the sea and various agricultural techniques on the inorganic composition; (ii) the analysis of hazelnuts of different cultivars grown in the same area and sampled in the same year to evaluate the different element absorption capacity of each cultivar; (iii) the study of the effect of roasting step conditions (time, temperature) on the element profile of hazelnuts belonging to different cultivars.
Supplementary Materials: The following are available online: http://www.mdpi.com/1660-4601/17/2/447/s1. Table S1: Concentration of elements (mg·kg−1) at T0 for O- and R-samples; Table S2: Concentration of elements (mg·kg−1) in O-samples stored with A-Method; Table S3: Concentration of elements (mg·kg−1) in O-samples stored
with B-Method; Table S4: Concentration of elements (mg·kg−1) in O-samples stored with C-Method; Table S5: Concentration of elements (mg·kg−1) in O-samples stored with D-Method; Table S6: Concentration of elements (mg·kg−1) in O-samples stored with E-Method; Table S7: Concentration of elements (mg·kg−1) in R-D1-samples stored with A-Method; Table S8: Concentration of elements (mg·kg−1) in R-D1-samples stored with B-Method; Table S9: Concentration of elements (mg·kg−1) in R-D1-samples stored with C-Method; Table S10: Concentration
of elements (mg·kg−1) in R-D1-samples stored with D-Method; Table S11: Concentration of elements (mg·kg−1) in R-D1-samples stored with E-Method; Table S12: Concentration of elements (mg·kg−1) in R-D2-samples stored with A-Method; Table S13: Concentration of elements (mg·kg−1) in R-D2-samples stored with B-Method; Table S14: Concentration of elements (mg·kg−1) in R-D2-samples stored with C-Method; Table S15: Concentration of elements (mg·kg−1) in R-D2-samples stored with D-Method; Table S16: Concentration of elements (mg·kg−1) in
R-D2-samples stored with E-Method; Table S17: Concentration of elements (mg·kg−1) in O-samples roasted and not roasted, stored with B-Method and C-Method at T0 and T4; Table S18: Concentration of elements (mg·kg−1) in R-D1-samples roasted and not roasted, stored with B-Method and C-Method at T0 and T4; Table S19: Concentration of elements (mg·kg−1) in R-D2-samples roasted and not roasted, stored with B-Method and C-Method at T0
and T4; Table S20: Concentration of elements (mg·kg−1) in Pm samples; Table S21: Concentration of elements (mg·kg−1) in Pr samples; Figure S1: Biplot obtained by PCA (PC1 vs. PC2) including hazelnuts from Piedmont in the main dataset.
Author Contributions:Conceptualization, A.G. and P.I. methodology, M.M.; software, M.M.; formal analysis, P.I. and C.L.G.; investigation, P.I.; resources, O.A.; data curation, A.G.; writing—original draft preparation, A.G. and P.I.; writing—review and editing, E.C. and T.K.; supervision, O.A. All authors have read and agreed to the published version of the manuscript.
Funding:This research was founded by the financial support of the University of Torino (Ricerca Locale). Acknowledgments:We thank Giordano Flavio for kindly allowing sampling in his field.
Conflicts of Interest:The authors declare no conflict of interest.
References
1. Food and Agriculture Organization (FAO). Available online:www.fao.org/faostat(accessed on 3 March 2019). 2. Commission Regulation (EC) No 1107/96 of 12 June 1996 on the registration of geographical indications and
designations of origin under the procedure laid down in Article 17 of Council Regulation (EEC) No 2081/92. Off. J. L 1996, 299, 31.
3. Klockmann, S.; Reiner, E.; Bachmann, R.; Hackl, T.; Fischer, M. Food Fingerprinting: Metabolomic Approaches for Geographical Origin Discrimination of Hazelnuts (Corylus avellana) by UPLC-QTOF-MS. J. Agric. Food Chem. 2016, 64, 9253–9262. [CrossRef] [PubMed]
4. Del Prete, C.; Sesana, R. Mechanical characterization of kernel and shell of hazelnuts: Proposal of an experimental procedure. J. Food Eng. 2014, 124, 28–34. [CrossRef]
5. Bonisoli, E.; Delprete, C.; Sesana, R.; Tamburro, A.; Tornincasa, S. Testing and simulation of the three point bending anisotropic behaviour of hazelnut shells. Biosyst. Eng. 2015, 129, 134–141. [CrossRef]
6. Caligiani, A.; Coisson, J.D.; Travaglia, F.; Acquotti, D.; Palla, G.; Palla, L.; Arlorio, M. Application of1H NMR for the characterisation and authentication of Tonda Gentile Trilobata hazelnuts from Piedmont (Italy). Food Chem. 2014, 148, 77–85. [CrossRef]
7. Dundar, M.S.; Bahçıvancı, E.; Muslu, C. Influence of variety and geographical region on mineral contents of hazelnuts (Corylus avellana L.) varieties. Acta Chim. Slov. 2002, 49, 537–544.
8. Ozec, N.; Ozenc, D.M. Nut traits and nutritional composition of hazelnut (Corylus avellana L.) as influenced by zinc fertilization. J. Sci. Food Agric. 2015, 95, 1956–1962. [CrossRef]
9. Kafao ˘glu, B.; Fisher, A.; Hill, S.; Kara, J. Chemometric evaluation of trace metal concentrations in some nuts and seeds. Food Addit. Contam. Part A 2014, 31, 1529–1538. [CrossRef]
10. Prelle, A.; Spadaro, D.; Garibaldi, A.; Gullino, M.L. Aflatoxin monitoring in Italian hazelnut products by LC-MS. Food Addit. Contam. Part B 2012, 5, 279–285. [CrossRef]
11. Locatelli, M.; Coisson, J.D.; Travaglia, F.; Bordiga, M.; Arlorio, M. Impact of Roasting on Identification of Hazelnut (Corylus avellana L.) Origin: A Chemometric Approach. J. Agric. Food Chem. 2015, 63, 7294–7303. [CrossRef]
12. Burdack-Freitag, A.; Schieberle, P. Changes in the Key Odorants of Italian Hazelnuts (Coryllus avellana L. Var. Tonda Romana) Induced by Roasting. J. Agric. Food Chem. 2010, 58, 6351–6359. [CrossRef] [PubMed] 13. Cordero, C.; Liberto, E.; Bicchi, C.; Rubiolo, P.; Schieberleb, P.; Reichenbachc, S.E.; Tao, Q. Profiling
food volatiles by comprehensive two-dimensional gas chromatography coupled with mass spectrometry: Advanced fingerprinting approaches for comparative analysis of the volatile fraction of roasted hazelnuts (Corylus avellana L.) from different origins. J. Chromatogr. A 2010, 1217, 5848–5858. [PubMed]
14. Marzocchi, S.; Pasini, F.; Verardo, V.; Ciemniewska-Zytkiewicz, H.; Caboni, M.F.; Romani, S. Effects of different roasting conditions on physical-chemical properties of Polish hazelnuts (Corylus avellana L. Var. Katalonski). LWT Food Sci. Technol. 2017, 77, 440–448. [CrossRef]
15. Oddone, M.; Aceto, M.; Baldizzone, M.; Musso, D.; Osella, D. Authentication and Traceability Study of Hazelnuts from Piedmont, Italy. J. Agric. Food Chem. 2009, 57, 3404–3408. [CrossRef] [PubMed]
16. Santos, E.E.; Lauria, D.C.; Porto da Silveira, C.L. Assessment of daily intake of trace elements due to consumption of foodstuffs by adult inhabitants of Rio de Janeiro city. Sci. Total Environ. 2004, 327, 69–79. [CrossRef] [PubMed]
17. Lombardi-Bocca, G.; Aguzzi, A.; Cappelloni, M.; Di Lullo, G.; Lucarini, M. Total-diet study: Dietary intakes of macro elements and trace elements in Italy. Br. J. Nutr. 2003, 90, 1117–1121. [CrossRef]
18. Pereira-Lorenzo, S.; Ramos-Cabrer, A.M.; Diaz-Hernandez, M.B.; Ciordia-Ara, M.; Rioz-Mesa, D. Chemical composition of chestnut cultivars from Spain. Sci. Hortic. 2006, 107, 306–314. [CrossRef]
19. Momen, A.A.; Zachariadis, G.A.; Anthemidis, A.N.; Stratis, J.A. Use of fractional factorial design for optimization of digestion procedures followed by multi-element determination of essential and non-essential elements in nuts using ICP-OES technique. Talanta 2007, 71, 443–451. [CrossRef]
20. Bond, B.; Fernandez, D.R.; VanderJagt, D.J.; Williams, M.; Huang, Y.S.; Ghuang, L.T.; Millson, M.; Andrews, R.; Glew, R.H. Fatty acid, amino acid and trace mineral analysis of three complimentary foods from Jos, Nigeria. J. Food Comp. Anal. 2005, 18, 675–690. [CrossRef]
21. Cevik, U.; Celik, N.; Celik, A.; Damla, N.; Coskuncelebi, K. Radioactivity and heavy metal levels in hazelnut growing in the Eastern Black Sea Region of Turkey. Food Chem. Toxicol. 2009, 47, 2351–2355. [CrossRef] 22. Gomez-Ariza, J.L.; Arias-Borrego, A.; Garcıa-Barrera, T. Multielemental fractionation in pine nuts (Pinus
pinea) from different geographic origins by size-exclusion chromatography with UV and inductively coupled plasma mass spectrometry detection. J. Chromatogr. A 2006, 1121, 191–199. [CrossRef]
23. Gomez-Ariza, J.L.; Arias-Borrego, A.; Garcıa-Barrera, T. Combined use of total metal content and size fractionation of metal biomolecules to determine the provenance of pine nuts (Pinus pinea). Anal. Bioanal. Chem. 2007, 388, 1295–1302. [CrossRef] [PubMed]
24. Prats-Moya, S.; Grane’-Teruel, N.; Berenguer-Navarro, V.; Luisa Martin-Carratala, M. Inductively Coupled Plasma application for the classification of 19 almond cultivars using inorganic element composition. J. Agric. Food Chem. 1997, 45, 2093–2097. [CrossRef]
25. Rodushkina, I.; Engströma, E.; Sörlinb, D.; Baxterb, D. Levels of inorganic constituents in raw nuts and seeds on the Swedish market. Sci. Total Environ. 2008, 392, 290–304. [CrossRef] [PubMed]
26. Köksala, A.I.; Artik, N.; ¸Sim¸sekc, A.; Güne¸s, N. Nutrient composition of hazelnut (Corylus avellana L.) varieties cultivated in Turkey. Food Chem. 2006, 99, 509–515.
27. Plessi, M.; Bertelli, D.; Monzani, A.; Simonetti, M.S.; Neri, A.; Damiani, P. Dietary fiber and some elements in nuts and wheat brans. J. Food Compos. Anal. 1999, 12, 91–96. [CrossRef]
28. Ozdemir, F.; Akinci, I. Physical and nutritional properties of four major commercial Turkish hazelnut varieties. J. Food Eng. 2004, 63, 341–347. [CrossRef]
29. Ackurt, F.; Ozdemir, M.; Biringen, G.; Loker, M. Effects of geographical origin and variety on vitamin and mineral composition of hazelnut (Corylus avellana L.) varieties cultivated in Turkey. Food Chem. 1999, 65, 309–313. [CrossRef]
30. Yin, L.L.; Tian, Q.; Shao, X.Z.; Kong, X.; Yin, X.Y.; Ji, Y.Q. Determination of Trace Elements in Edible Nuts in the Beijing Market by ICP-MS. Biomed. Environ. Sci. 2015, 28, 449–454.
31. Simsek, A.; Aykut, O. Evaluation of the microelement profile of Turkish hazelnut (Corylus avellana L.) varieties for human nutrition and health. Int. J. Food Sci. Nutr. 2007, 58, 677–688. [CrossRef]
32. Mehlenbacher, S.A.; Brown, R.N.; Nouhra, E.R.; Gokirmak, T.; Bassil, N.V.; Kubisiak, T.L. A genetic linkage map for hazelnut (Corylus avellana L.) based on RAPD and SSR markers. Genome 2006, 49, 122–133. [CrossRef] [PubMed]
33. Gokirmak, T.; Mehlenbacher, S.A.; Bassil, N.V. Characterization of European hazelnut (Corylus avellana) cultivars using SSR markers. Genet. Resour. Crop Evol. 2009, 56, 147–172. [CrossRef]
34. Moscetti, R.; Radicetti, E.; Monarca, D.; Cecchini, M.; Massantini, R. Near infrared spectroscopy is suitable for the classification of hazelnuts according to Protected Designation of Origin. J. Sci. Food Agric. 2015, 95, 2619–2625. [CrossRef] [PubMed]
35. Klockmann, S.; Reiner, E.; Cain, N.; Fischer, M. Food Targeting: Geographical Origin Determination of Hazelnuts (Corylus avellana) by LC-QqQ-MS/MS-Based Targeted Metabolomics Application. J. Agric. Food Chem. 2017, 65, 1456–1465. [CrossRef] [PubMed]
36. Bachmann, R.; Klockmann, S.; Haerdter, J.; Fischer, M.; Hackl, T.1H NMR Spectroscopy for Determination of the Geographical Origin of Hazelnuts. J. Agric. Food Chem. 2018, 66, 11873–11879. [CrossRef] [PubMed] 37. Masthoff, L.J.; Hoff, R.; Verhoeckx, K.C.M.; van OsMedendorp, H.; Michelsen-Huisman, A.; Baumert, J.L.;
Pasmans, S.G.; Meijer, Y.; Knulst, A.C. A systematic review of the effect of thermal processing on the allergenicity of tree nuts. Allergy 2013, 68, 983–993. [CrossRef]
38. Arzandeh, S.; Jinap, S. Effect of initial aflatoxin concentration, heating time and roasting temperature on aflatoxin reduction in contaminated peanuts and process optimization using response surface modelling. Int. J. Food Sci. Technol. 2011, 46, 485–491. [CrossRef]
39. Saklar, S.; Ungan, S.; Katnas, S. Microstructural changes in hazelnuts during roasting. Food Res. Int. 2003, 36, 19–23. [CrossRef]
40. Alasalvar, C.; Pelvan, E.; Amarowicz, R. Effects of roasting on taste-active compounds of Turkish hazelnut varieties (Corylus avellana L.). J. Agric. Food Chem. 2010, 58, 8674–8679. [CrossRef]
41. Alasalvar, C.; Pelvan, E.; Topal, B. Effects of roasting on oil and fatty acid composition of Turkish hazelnut varieties (Corylus avellana L.). Int. J. Food Sci. Nutr. 2010, 61, 630–642. [CrossRef]
42. Donno, D.; Beccaro, G.L.; Mellano, G.M.; Di Prima, S.; Cavicchioli, M.; Cerutti, A.K. Setting a protocol for hazelnut roasting using sensory and colorimetric analysis: Influence of the roasting temperature on the quality of Tonda Gentile delle Langhe cv. hazelnut. Czech J. Food Sci. 2013, 31, 390–400. [CrossRef]
43. Ciemniewska-Zytkiewicz, H.; Brys, J.; Brys, A.; Sujka, K.; Koczon, P. Effect of roasting process on moisture content and colour of polish in shell hazelnuts. Acad. Food J. 2014, 12, 6–10.
44. Naozuka, J.; Carvalho Vieira, E.; Nogueira Nascimento, A.; Oliveira, P.V. Elemental analysis of nuts and seeds by axially viewed ICP OES. Food Chem. 2011, 124, 1667–1672. [CrossRef]
45. Gonçalves, A.M.; Fernandes, K.G.; Ramos, L.A.; Cavalheiro, É.T.G.; Nóbrega, J.A. Determination and fractionation of barium in Brazil nuts. J. Braz. Chem. Soc. 2019, 20, 760–769. [CrossRef]
46. Adelakun, O.E.; Ade-Omowaye, B.I.O.; Adeyemi, I.A.; Van de Venter, M. Mineral composition and the functional attributes of Nigerian okra seed (Abelmoschus esculentus Moench) flour. Food Res. Int. 2012, 47, 348–352. [CrossRef]
47. Locatelli, M.; Amaral, J.S.; Casal, S.; Seabra, R.M.; Oliveira, B.P.P. Effects of roasting on hazelnut lipids. J. Agric. Food Chem. 2006, 54, 1315–1321.
48. Seyhan, F.; Ozay, G.; Saklar, S.; Ertas, E.; Satır, G.; Alasalvar, C. Chemical changes of three native Turkish hazelnut varieties (Corylus avellana L.) during fruit development. Food Chem. 2007, 105, 590–596. [CrossRef] 49. Marescotti, P.; Comodi, P.; Crispini, L.; Gigli, L.; Zucchini, A.; Fornasaro, S. Potentially Toxic Elements in Ultramafic Soils: A Study from Metamorphic Ophiolites of the Voltri Massif (Western Alps, Italy). Minerals 2019, 9, 502. [CrossRef]
© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
quality and for authentication of the hazelnut’s
origin
Paolo Inaudi1, Agnese Giacomino1*, Mery Malandrino2, Carmela La Gioia2, Eleonora Conca2,
Tanmoy Karak3 and Ornella Abollino1 1Department of Drug Science and Technology, University of Torino, 10125 Torino, Italy 2Department of Chemistry, University of Torino, 10125 Torino, Italy 3 Upper Assam Advisory Centre, 786101 Dikom, Assam, India
Supplementary materials Table S1. Concentration of elements (mg∙kg−1) at T0 for O‐ and R‐samples.
Element Ordu T0 Romana D1 T0 Romana D2 T0
Ba 23.4 ± 0.2 2.21 ± 0.15 2.77 ± 0.06 Ca 1097 ± 7 1351 ± 27 1200 ± 13 Cu 12.3 ± 0.2 13.3 ± 0.8 13.5 ± 0.1 Fe 26.1 ± 0.2 38.8 ± 1.1 33.4 ± 0.7 K 3472 ± 56 6588 ± 202 6290 ± 57 Mg 1342 ± 40 1569 ± 22 1524 ± 9 Mn 27.5 ± 1.3 48.6 ± 2.1 60.2 ± 0.1 Na 0.06 ± 1 × 10−4 4.91 ± 0.18 4.68 ± 0.27 Ni 0.98 ± 0.09 0.68 ± 1 × 10−4 0.62 ± 0.07 P 2863 ± 79 1439 ± 21 1435 ± 19 Sn 2.77 ± 0.01 6.32 ± 0.86 5.38 ± 0.31 Sr 3.95 ± 0.14 23.2 ± 2.1 22.4 ± 0.6 Zn 17.2 ± 0.2 18.0 ± 0.2 16.5 ± 0.2 Sum 8865 ± 106 11,103 ± 206 10,608 ± 62 Table S2. Concentration of elements (mg∙kg−1) in O‐samples stored with method A. Element T0 T1 T2 T4 Mean Ba 23.4 ± 0.2 44.2 ± 1.5 63.3 ± 1.0 42.5 ± 0.4 50.0 ± 11.6 Ca 1097 ± 7 1289 ± 26 1563 ± 10 1423 ± 1 1425 ± 137 Cu 12.3 ± 0.2 11.3 ± 0.7 12.9 ± 0.2 14.6 ± 0.2 12.9 ± 1.6 Fe 26.1 ± 0.2 26.8 ± 0.1 24.7 ± 0.3 22.3 ± 0.2 24.6 ± 2.3 K 3472 ± 56 4007 ± 216 3786 ± 214 3880 ± 122 3891 ± 111 Mg 1342 ± 40 1326 ± 2 1496 ± 54 1470 ± 1 1431 ± 91 Mn 27.5 ± 1.3 45.8 ± 0.9 43.0 ± 0.3 28.6 ± 1.8 39.1 ± 9.2 Na 0.06 ± 1 × 10−4 0.06 ± 1 × 10−4 0.12 ± 1 × 10−4 0.06 ± 1 × 10−4 0.08 ± 0.03 Ni 0.98 ± 0.09 1.57 ± 0.02 1.63 ± 0.09 1.23 ± 1 × 10−4 1.48 ± 0.21 P 2863 ± 79 2774 ± 60 2885 ± 100 3061 ± 80 2906 ± 145 Sn 2.77 ± 0.01 2.78 ± 0.22 3.01 ± 0.05 2.34 ± 0.21 2.71 ± 0.34 Sr 3.95 ± 0.14 9.43 ± 0.18 15.8 ± 1.6 16.3 ± 0.4 13.8 ± 3.8 Zn 17.2 ± 0.2 16.9 ± 0.1 16.6 ± 0.8 17.3 ± 0.2 16.9 ± 0.3 Sum 8865 ± 106 9555 ± 226 9854 ± 243 9979 ± 146
Element T0 T1 T2 T4 Mean Ba 23.4 ± 0.2 63.6 ± 1.5 50.6 ± 3.3 48.5 ± 2.4 54.2 ± 8.18 Ca 1097 ± 7 1316 ± 26 1420 ± 30 1171 ± 9 1302 ± 125 Cu 12.3 ± 0.2 11.7 ± 0.7 13.1 ± 0.1 14.8 ± 0.2 13.2 ± 1.53 Fe 26.1 ± 0.2 25.4 ± 0.1 27.6 ± 1.1 28.9 ± 0.7 27.3 ± 1.77 K 3472 ± 56 3552 ± 216 3279 ± 104 3824 ± 149 3551 ± 272 Mg 1342 ± 40 1347 ± 2 1466 ± 32 1507 ± 14 1440 ± 83 Mn 27.5 ± 1.3 44.3 ± 0.9 38.6 ± 1.6 30.7 ± 0.4 37.9 ± 6.79 Na 0.06 ± 1 × 10−4 0.06 ± 1 × 10−4 0.12 ± 1 × 10−4 0.12 ± 1 × 10−4 0.10 ± 0.03 Ni 0.98 ± 0.09 1.10 ± 0.02 1.54 ± 0.07 1.54 ± 0.01 1.39 ± 0.25 P 2863 ± 79 2924 ± 60 2860 ± 66 3086 ± 37 2957 ± 116 Sn 2.77 ± 0.01 2.83 ± 0.22 3.55 ± 0.15 3.52 ± 0.11 3.30 ± 0.41 Sr 3.95 ± 0.14 10.1 ± 0.2 12.4 ± 0.9 14.2 ± 0.8 12.2 ± 2.08 Zn 17.2 ± 0.2 15.7 ± 0.1 17.8 ± 0.6 16.4 ± 0.1 16.6 ± 1.10 Sum 8865 ± 106 9314 ± 226 9190 ± 130 9747 ± 154 Table S4. Concentration of elements (mg∙kg−1) in O‐samples stored with method C. Element T0 T1 T2 T4 Mean Ba 23.4 ± 0.2 40.7 ± 26.6 29.8 ± 1.6 60.0 ± 0.3 43.5 ± 15.3 Ca 1097 ± 7 1274 ± 234 1425 ± 38 1520 ± 84 1406 ± 124 Cu 12.3 ± 0.2 13.7 ± 0.7 14.9 ± 0.1 13.5 ± 0.3 14.0 ± 0.73 Fe 26.1 ± 0.2 25.7 ± 0.4 27.7± 0.3 26.4 ± 0.4 26.6 ± 0.98 K 3472 ± 56 3296 ± 140 4060 ± 162 3932 ± 225 3763 ± 409 Mg 1342 ± 40 1390 ± 24 1436 ± 10 1638 ± 41 1488 ± 132 Mn 27.5 ± 1.3 31.3 ± 4.8 39.94 ± 1.55 40.41 ± 3.40 37.2 ± 5.10 Na 0.06 ± 1 × 10−4 0.06 ± 1 × 10−4 0.12 ± 1 × 10−4 0.12 ± 1 × 10−4 0.10 ± 0.03 Ni 0.98 ± 0.09 1.00 ± 0.08 1.22 ± 0.04 1.70 ± 0.39 1.30 ± 0.36 P 2863 ± 79 2971 ± 5 3191 ± 9 3179 ± 111 3113 ± 124 Sn 2.77 ± 0.01 2.79 ± 0.10 3.88 ± 0.06 3.76 ± 0.85 3.48 ± 0.60 Sr 3.95 ± 0.14 7.18 ± 5.05 8.26 ± 0.74 16.44 ± 2.54 10.63 ± 5.06 Zn 17.2 ± 0.2 16.9 ± 0.1 17.92 ± 0.04 17.97 ± 0.32 17.61 ± 0.58 Sum 8865 ± 106 9070 ± 275 10,256 ± 167 10,449 ± 268 Table S5. Concentration of elements (mg∙kg−1) in O‐samples stored with method D. Element T0 T1 T2 T4 Mean Ba 23.4 ± 0.2 22.7 ± 0.02 58.9 ± 1.5 42.6 ± 3.1 41.4 ± 18.1 Ca 1097 ± 7 1397 ± 3 1260 ± 13 1504 ± 87 1387 ± 122 Cu 12.3 ± 0.2 13.0 ± 0.7 13.9 ± 0.3 14.6 ± 1 × 10−4 13.8 ± 0.81 Fe 26.1 ± 0.2 25.0 ± 1.0 26.7 ± 0.8 28.6 ± 0.7 26.8 ± 1.80 K 3472 ± 56 3731 ± 48 3355 ± 42 3942 ± 610 3676 ± 297 Mg 1342 ± 40 1470 ± 44 1454 ± 31 1485 ± 48 1470 ± 15 Mn 27.5 ± 1.3 25.9 ± 0.6 38.5 ± 0.5 38.1 ± 0.8 34.2 ± 7.18 Na 0.06 ± 1 × 10−4 0.12 ± 1 × 10−4 0.12 ± 1 × 10−4 0.12 ± 1 × 10−4 0.12 ± 1 × 10−4 Ni 0.98 ± 0.09 0.85 ± 0.09 1.19 ± 0.01 1.42 ± 0.08 1.15 ± 0.29 P 2863 ± 79 2965 ± 216 2844 ± 46 3029 ± 93 2946 ± 94 Sn 2.77 ± 0.01 2.10 ± 1.40 3.18 ± 0.05 3.77 ± 0.06 3.02 ± 0.85 Sr 3.95 ± 0.14 19.05 ± 1.25 11.7 ± 0.7 6.97 ± 0.19 12.6 ± 6.09 Zn 17.2 ± 0.2 16.9 ± 0.4 16.01 ± 0.02 17.7 ± 1.2 16.8 ± 0.84 Sum 8865 ± 106 9689 ± 226 9083 ± 71 10,114 ± 625
Element T0 T1 T2 T4 Mean Ba 23.4 ± 0.2 57.1 ± 0.8 48.8 ± 0.6 47.3 ± 1.8 51.1 ± 5.28 Ca 1097 ± 7 1485 ± 19 1342 ± 24 1307 ± 22 1378 ± 94 Cu 12.3 ± 0.2 15.0 ± 0.1 13.4 ± 0.1 13.2 ± 0.1 13.9 ± 1.03 Fe 26.1 ± 0.2 29.7 ± 1.1 26.8 ± 0.7 26.11 ± 0.93 27.5 ± 1.88 K 3472 ± 56 3603 ± 158 3331 ± 30 3528 ± 99 3487 ± 140 Mg 1342 ± 40 1428 ± 14 1427 ± 7 1353 ± 29 1403 ± 43 Mn 27.5 ± 1.3 28.8 ± 0.2 35.6 ± 1.6 40.0 ± 2.4 34.8 ± 5.63 Na 0.06 ± 1 × 10−4 0.12 ± 1 × 10−4 0.09 ± 0.04 0.12 ± 1 × 10−4 0.11 ± 0.02 Ni 0.98 ± 0.09 0.99 ± 0.04 1.10 ± 0.04 1.29 ± 0.13 1.13 ± 0.15 P 2863 ± 79 3026 ± 77 3035 ± 14 2775 ± 22 2945 ± 148 Sn 2.77 ± 0.01 4.23 ± 0.16 3.21 ± 0.17 2.79 ± 0.73 3.41 ± 0.74 Sr 3.95 ± 0.14 15.1 ± 0.4 8.36 ± 0.06 10.6± 0.6 11.4 ± 3.42 Zn 17.2 ± 0.2 17.3 ± 0.1 16.7 ± 0.5 17.6 ± 0.5 17.2 ± 0.47 Sum 8865 ± 106 9710 ± 177 9289 ± 42 9,122 ± 108 Table S7. Concentration of elements (mg∙kg−1) in R‐D1‐samples stored with method A. Element T0 T1 T2 T4 Mean Ba 2.21 ± 0.15 1.60 ± 0.13 1.64 ± 0.01 1.88 ± 0.05 1.71 ± 0.15 Ca 1351 ± 27 1300 ± 4 1426 ± 62 1208 ± 5 1311 ± 110 Cu 13.3 ± 0.8 14.1 ± 1.3 13.6 ± 0.7 14.9 ± 0.11 14.2 ± 0.65 Fe 38.8 ± 1.1 35.3 ± 1.5 35.1 ± 3.1 37.9 ± 0.6 36.1 ± 1.54 K 6588 ± 202 6447 ± 1404 5914 ± 270 7424 ± 251 6595 ± 766 Mg 1569 ± 22 1617 ± 136 1636 ± 7 1626 ± 18 1626 ± 9 Mn 48.6 ± 2.1 37.3 ± 2.2 35.4 ± 0.9 64.15 ± 0.94 45.6 ± 16.1 Na 4.91 ± 0.18 3.89 ± 0.38 4.28 ± 0.59 6.42 ± 0.28 4.87 ± 1.36 Ni 0.68 ± 1 × 10−4 0.66 ± 0.20 0.77 ± 1 × 10−4 0.77 ± 0.04 0.74 ± 0.07 P 1439 ± 21 1666 ± 167 1627 ± 31 1502 ± 27 1599 ± 86 Sn 6.32 ± 0.86 6.72 ± 0.15 7.20 ± 0.52 5.73 ± 0.58 6.55 ± 0.75 Sr 23.2 ± 2.1 16.7 ± 0.6 20.5 ± 0.7 39.3 ± 3.5 25.5 ± 12.09 Zn 18.0 ± 0.2 17.2 ± 0.9 17.0 ± 1.3 17.5 ± 0.1 17.24 ± 0.24 Sum 11,103 ± 206 11,163 ± 1420 10,738 ± 279 11,949 ± 253 Table S8. Concentration of elements (mg∙kg−1) in R‐D1‐samples stored with method B. Element T0 T1 T2 T4 Mean Ba 2.21 ± 0.15 1.52 ± 0.05 1.25 ± 0.10 1.45 ± 0.06 1.41 ± 0.14 Ca 1351 ± 27 1320 ± 72 1147 ± 62 1349 ± 46 1272 ± 109 Cu 13.3 ± 0.8 13.8 ± 0.9 14.5 ± 1.1 12.6 ± 0.1 13.6 ± 0.97 Fe 38.8 ± 1.1 41.9 ± 0.8 38.5 ± 5.3 41.3 ± 0.8 40.5 ± 1.82 K 6588 ± 202 6035 ± 132 6057 ± 435 6601 ± 226 6231 ± 321 Mg 1569 ± 22 1614 ± 40 1550 ± 32 1652 ± 2 1605 ± 51 Mn 48.6 ± 2.1 47.5 ± 0.1 52.8 ± 3.5 37.4 ± 0.1 45.9 ± 7.85 Na 4.91 ± 0.18 6.29 ± 0.88 5.45 ± 1.31 6.69 ± 0.05 6.14 ± 0.63 Ni 0.68 ± 1 × 10−4 1.08 ± 0.33 1.00 ± 0.09 0.74 ± 0.08 0.94 ± 0.18 P 1439 ± 21 1623 ± 57 1434 ± 51 1429 ± 34 1492 ± 114 Sn 6.32 ± 0.86 8.09 ± 0.07 7.39 ± 0.45 6.53 ± 0.33 7.34 ± 0.79 Sr 23.2 ± 2.1 37.3 ± 1.6 17.5 ± 2.2 20.7 ± 0.3 25.2 ± 10.6 Zn 18.0 ± 0.2 19.1 ± 0.4 17.3 ± 1.9 18.9 ± 0.4 18.4 ± 0.94 Sum 11,103 ± 206 10,769 ± 166 10,344 ± 444 11,177 ± 233
