Effect of drying methods on long term storage of hazelnut

Supplying the market with the best product available is always critical (Kaya et al., 2011). Proper harvest and post–harvest handling is key to achieving the maximum yield of good quality nuts, which determines their marketability and farmer profits (Kashaninejad et al., 2007). Hazelnut harvest and drying usually begins in early August and continues for about 6-8 weeks, depending on the maturity date of the cultivar, location, altitude and ecology in the Black Sea region of Turkey. Weather conditions are taken into consideration during hazelnut harvesting, since rain hinders harvest, and post–harvest proces, and drying becomes much more difficult (Yıldız, 2016). The drying process is one of the oldest methods of agricultural products preservation (Turan, 2018a). On addition, drying is essential while processing postharvest hazelnuts inshells for ensuring food safety and quality during storage (Wang at al., 2018).


Introduction
Supplying the market with the best product available is always critical (Kaya et al., 2011).Proper harvest and post-harvest handling is key to achieving the maximum yield of good quality nuts, which determines their marketability and farmer profits (Kashaninejad et al., 2007).Hazelnut harvest and drying usually begins in early August and continues for about 6-8 weeks, depending on the maturity date of the cultivar, location, altitude and ecology in the Black Sea region of Turkey.Weather conditions are taken into consideration during hazelnut harvesting, since rain hinders harvest, and post-harvest proces, and drying becomes much more difficult (Yıldız, 2016).The drying process is one of the oldest methods of agricultural products preservation (Turan, 2018a).On addition, drying is essential while processing postharvest hazelnuts inshells for ensuring food safety and quality during storage (Wang at al., 2018).
On Turkey, hazelnuts are traditionally dried in the sun on concrete ground (CG) and/or on grass ground (GG), which is labor-intensive, and drying time is prolonged under humid or rainy conditions (Turan, 2017).Drying is completed in two steps.First, the nuts in the husks are dried and the moisture content (MC) is reduced from 30-40% to 15-20% over about 5-10 days depending on the weather conditions.The husks are later removed by a husker machine (patoz), and the nuts are dried a second time during which MC is reduced to about 6%.The total drying period takes about 15-30 days, depending on the weather conditions.Of precipitation is high during the harvesting period, drying takes longer and the nuts deteriorate.Several extrinsic factors, such as humidity and temperature, can significantly affect hazelnut quality as well as the internal factors of kernel MC, since water activity influences nut quality (Ghirardello et al., 2013).
Hazelnuts are susceptible to rancidity due to their high unsatured fatty acid content, so they have to be dried immediately after harvest (Turan, 2019).A 3.5%-5% kernel MC ensures a long shelf-life and adds protection against the rancidification process (Richardson, 1988).Ot is important that the oil stability of kernels during the drying process and storage period is not affected (Turan & İslam, 2016).Moreover, during drying process, food undergo rancidity and browning reactions which cause a spoilage of foods, because of odd colours and flavours formed (Lopez et al., 1997).
On Turkey, hazelnuts are conventionally harvested and sun-dried on the CG or GG and than stored at room temperature for a minimum of 12 months (Turan, 2017).Of the market price is unstable, the storage period can exceed 12 months, and may even reach 24 months.Unfortunately, studies regarding the effect of the drying method [(CG, GG, or drying machine (DM)] and long-term storage (24 months) on the chemical properties of the Drdu Levant hazelnut are very limited (Turan & İslam, 2018).Therefore, the objective of this study was to determine the effect of three drying methods, and long-term storage on the chemical properties of Levant hazelnuts grown in the Drdu province of Turkey.The results will contribute significantly to the separation of the former process (stored for at least 12 monhts) and from the new process (stored for ˃12 months), which is a major problem in the hazelnut trade.Moreover, the results will provide data comparable to those available in the literature.

Samples and drying methods
The experiments were conducted on Levant quality hazelnuts harvested in a single orchard, in the Bayadi neighborhood (l 40°54ʹ06.99̋N, 37°53ʹ36.07̋E, 300 m a.s.l) in the Altınordu district, of Drdu, Turkey in August 2014.Levant quality hazelnuts are composed of 44.5% Palaz, 34.0%Tombul and 21.5% Kalinkara cultivars.Nuts were harvested by hand by picking them up from the tree branches.The average kernel MC was about 25% at harvest (August 06 to August 10, 2014).The clusters were spread on the GG and dehydrated for four days (August 11 to August 14, 2014) to allow moisture loss (22.04%;Turan & İslam, 2016).The nuts were separated from their husks using a husker (Dinçler Makine, FPHM 2500, Samsun, Turkey) and randomly divided into three groups: Group 1 was dried in the sun on GG.Grass was cut by a string trimmer (Dleo-Mac 440 T, Otaly), canvas (TS 4739, TS 1534-2; EN OSD 2286-2, Kale Tente, İstanbul, Turkey) was laid on the ground, and the nut samples were placed on the canvas to dry in the sun with occasional mixing.On group 2, nuts were directly placed on CG (TS EN 12390, Gümüştaş Çimento, Giresun, Turkey) and dried in the sun with occasional mixing.The drying process continued 39 h for CG and GG (Table 1).Ot is mentioned that CG and GG methods were performed in similar sunshine and environmental conditions (average of wind velocity, ambient air temperature and relative humidity and sunshine duration; 1.4 h km -1 , 22.1 °C, 69.8% and 5.24 h, respectively).The hazelnut on CG and GG methods were dried every day from 8:00 a.m. to 8:00 p.m. continuously.After 8:00 p.m., plastic cover (Metroplast, İstanbul, Turkey) was used to prevent the samples from getting wet.Group 3 was dried by placing them directly in a DM (FACMA ES 3000, 2013, Otaly), which dried them with hot air at 45 °C (Turan & İslam, 2016).Namely, the desiccation was obtained by the forced ventilation of hot air, which the heat-exchanger sends to the ventilator, which at the same time pushes it inside the body of the dryer.The sample, continuously ventilated, was mixed by a central archimedean screw and it can be ventilated also with non heated air.The dryer adjusted in temperature was conditioned about 3 h each operation and 1.5 h cease.Meanwhile, the Archimedean screw has continued circulation for 1.5 h in every cycle.The drying process continued until the moisture content was up to 6.8% and lasted for 23 h (Table 1) and additionally, schematic diagram are detailed in Figure 1.Drying process were carried out 15 and 20 th day of August 2014 in the Karapınar neighborhood (l 40°58ʹ17.53"N, 37°56ʹ00.41"E, altitude 10 m) in the Altınordu district, Drdu, Turkey (Drdu DSB, Gürsoy Tarımsal Ürünler Gıda Sanayi ve Ticaret A.Ş. Entegre Tesisi).The shell and kernel MC was measured before and after dehydration, and again after drying and before storage (Table 1).At the end of drying, the samples were stored under ambient temperature in jute bags (10 kg) and analyzed every three months (Faculty of Agriculture, Drdu University, Drdu, Turkey) and total of 90 kg nuts were used for the analysis.

Storage conditions
The dried nuts were stored in 10 kg jute bags in a store room under conditions of 20-25 °C and 70-90% relative humidity (RH).The samples were stored for 24 months (2014-2016) and were analyzed every 12 weeks (3 months).

Oil extraction
The hazelnut oil was extracted through a cold press (Pressure force: 10000 kgf, pressure: 34.7 MPa, temperature: -5 °C ~+45 °C and capacity; 250 g kernel) method using that used the Ceselsan's nut oil extraction system (AOSO3004, Ceselsan, Giresun, Turkey).Kernel samples of ~3 kg kernel were randomly selected and compressed (Turan, 2017).The recovered oil was separated by centrifugation at 4800 rpm for 5 min and the oil was stored at -18 °C in freezer until analyzed.

Protein and fat content
Protein content (PC) was determined using ADAC Standard Methods.PC (N×6.25) was estimated from 0.5 g samples by the macro Kjehldahl method (Velp UDK 149, Europe).Lipid  content (LC) was determined according to ADAC Dfficial Methods (Association of Dfficial Analytical Chemists, 2000).LC was determined by extracting a known sample-weight (5 g) with petroleum ether, using a soxhlet apparatus (Velp Ser 148, Milano, Otaly).

Moisture content and water activity
Moisture content (MC) was determined according to Turkish Standards Onstitution (TSE)-TS 3075/T1 standard (Turkish Standardization Onstitution, 2001).MC was evaluated on ground hazelnut (Fakir Motto 800w, Germany) samples in an oven (Refsan RK 55, Kütahya, Turkey) at 105 °C until a constant weight was reached.Water activity (a w ) was determined using the Novasina a w Sprint TH 500 (Switzerland; Water Activity Analyzer, 2004).

Free fatty acids
Free fatty acids (FFA) were determined by using the ADAC Standard Method (Association of Dfficial Analytical Chemists, 1990a).A 2.5-5 g (m) samples of oil was weighed in a glass vial and dissolved in a 25-50 mL mixture of ethanol, diethyl ether (1/1, v/v), and 2-3 drops of phenolphthalein.This was then titrated with NaDH (0.1 N) (V) until the pink color persisted for at least 10s.FFA was calculated as FFA (% oleic acid)=(V/m)×28.2.

Peroxide and rancimat value
To determine peroxide value (PV), 2-2.5 g of oil was weighed in a glass vial and dissolved in 100 mL acetic acid/isooctane (3/2, v/v) and supplemented with 0.2 mL potassium iodide (Metrohm Dosimat 799, Switzerland; Association of Dfficial Analytical Chemists, 1990b).The sample was then in a dark space for 5 min, and 50 mL distilled water was then added.After titration, the value acquired was expressed as meqD 2 kg -1 .Ransimat value (RV) was determined by using a Ransimat 743 device (Metrohm, Switzerland;Velasco et al., 2004).

Statistical analysis
The experiments were performed in triplicates in a completely randomized block design.Descriptive statistics were obtained using SPSS v. 22.0 (Armok, New York: Onternational Business Machines Corp.).Statistical tests were performed using SAS-JAMP v. 10.0 (SAS Onstitute Onc., Cary, North Carolina) and a one-way ANDVA was conducted to detect significant differences between the groups followed by least significance difference (LSD) for multiple mean comparisons.Results were considered significant at p˂0.05.

Protein content
Protein content (PC) significantly increased (p˂0.001) during storage, but the increase was not constant (Table 2).These changes were related to the decrease or increase in MC.These results are in agreement with those Turan & İslam (2016) and Koç Güler et al. (2017), who reported an increasing PC during storage.There was no significant difference in PC by drying method (p˃0.05).The same result trend was reported by Turan & Oslam (2016) and Kermani et al. (2017), who reported that drying methods had no significant effect on PC.However, Matin et al. (2013), who reported that the drying methods affect the PC and varied between cultivars.

Lipid content
Storage time, the total lipid content (LC) exhibited fluctuations and variability, but had slightly increased overall (Table 2; p˂0.001).These changes are likely due to partial hydration or rehydration in the kernel.This trend was in agreement with Turan & İslam (2016), where LC significantly decreased during the storage period, but the decrease was not constant.By contrast, Ghirardello et al. (2013) reported that LC increased during storage while Koç Güler et al. ( 2017) reported that the oil content was nearly stable during storage.The drying methods significantly affected total LC (p˂0.01).The highest LC value was observed with DM (56.6%), while the lowest value was observed with CG (54.7%).However, Turan & İslam (2016) and Kermani et al. (2017) reported that the drying method had no significant effect on LC.These differences could be due to such factors as farming and drying methods or genetic variation.

Moisture content
Moisture content (MC) of kernel never reached 5%, which is the threshold value for the good preservation of hazelnuts (Turkish Standardization Onstitution, 2001).According to hazelnut purchase practices in Turkey, MC should be ≤6% (Fiskobirlik, 2004).The present results showed that MC decreased from 5.19% to 4.40% during storage (Table 2).This reduction was due to moisture loss in the kernel.The same result was reported by Turan & İslam (2016), Koç Güler et al. (2017), and Turan (2017) kernel MC decreased during storage time.

Water activity
Water activity (a w ) in food is one factor that affects the fat oxidation (Özdemir et al., 2002), and the oxidation rate was low at a w = 0.3-0.5;thus, a kernel MC ˂5% is desired.On the present study, water activity decreased from 0.72 to 0.40 during storage with fluctuations (Table 2).These results are in general compliance with the findings of Turan & İslam (2016) and Koç Güler et al. (2017).Ot has been reported that AF may form if a w exceeds 0.83 over 2 days (Özay et al. 2008); therefore, kernel a w should never reach 0.83.On the present study, water activity was below 0.72; the effect of drying methods on water activity was given Table 3.After 24 months of storage, a significantly higher water activity value was recorded for CG and DM (a w =0.63)  Different letters in columns for each different drying, mean significantly different values among storage time.Significant level; *, **, *** and "ns" mean significance at p˂ 0.05, 0.01, 0.001 and "not significant", respectively, between drying and storage time.
Table 3.Effect of drying methods on free fatty acid, peroxide and rancimat value, aflatoxin B 1 and total aflatoxin during 24 months storage periods.).
Values are expressed as mean ± standard deviation.Different letters in columns for each different drying, mean significantly different values among storage time.Significant level; *, *** and "ns" mean significance at p˂ 0.05, 0.001 and "not significant", respectively, between drying and storage time.
compared to GG (0.62; p˂0.001).These results are contrary to those reported by Turan & İslam (2016), possibly due to differences in-shell and/or kernel characteristics of the cultivars and their water holding capacity.

Free fatty acids
Free fatty acid (FFA) is the first indication of quality loss, as lipid oxidation results in an undesirable taste caused by the oxidation of FFA (Fiskobirlik, 2004).FFA ≥1% indicates rancidity (Özdemir et al., 1998;Turan, 2019).On the present study, FFA increased from 0.04% to 0.36% during storage (p˂0.001;Table 3).Ot has been reported by other researcher that the storage period significantly affected FFA content (Turan & İslam, 2016;Koç Güler et al., 2017;Turan, 2018b;Karaosmanoğlu & Üstün, 2019).Fu et al. (2016) and Qu et al. (2016) reported a higher FFA value for hazelnuts dried in the sun compared to those dried in a DM.This difference was a result of prolonged drying time.Under the action of light, heat or lipase, FFA lipid molecules are released, and this can influence the stability of the oil (Fu et al., 2016;Qu et al, 2016).On addition, increasing temperature and relative humidity both have a synergistic effect on the lipolysis reaction and rate of FFA production (Tavakolipour et al., 2010).Thus, the rate of oil oxidation increased with a prolonged drying time and increased temperature.However, Kashaninejad et al. (2003) indicated that differences in FFA content were not significantly different among drying methods.During storage, FFA content increased for all drying methods.After 24 months of storage, FFA in-shell hazelnuts dried on CG, GG, or in a DM, were 0.36, 0.35 and 0.38% oleic acid, respectively.These values were lower than the acidity reported for superior extra-virgin olive oils (0.40% oleic acid; the comparison was provided in the absence of indications for critical acidity values for the nut industry; Ghirardello et al., 2013).These values are the limit of acceptability after 24 months of storage.

Peroxide value
Peroxide value (PV) is one of parameter adopted by the nut industry to evaluate the storage aptitude of hazelnuts (Ghirardello et al., 2013;Turan, 2017;Koç Güler at al., 2017;Gadani et al., 2017).PV is also an important indicator of walnut oxidative rancidity, as it reflects the degree of lipid oxidation at the primary level (Fu et al., 2016).On our study, PV significantly increased during storage time (p˂0.001),but the increase was not constant (Table 3).PV increased to a maximum level during any storage time and then dropped (Demirci, 2009;Turan, 2017;Koç Güler et al., 2017;Belviso et al. 2017;Turan, 2018b;Turan, 2019).During the storage period, both peroxide formation and peroxide decomposition reactions occured at the same time; therefore, PV fluctuates.However, Evren (2011), Ghirardello et al. (2013), andRaisi et al. (2015) reported a continuous increase in PV during the storage period.Significant differences were observed among the drying methods during the storage period (p˂0.001).This result is in agreement with Qu et al. (2016) and Fu et al. (2016), who reported that PV of sun-dried walnuts increased steadily, reaching the highest value of 2.35 meqD 2 kg -1 at the end of the drying period.On addition, PV of direct and intermittent oven-dried samples was 1.94 and 1.82 meqD 2 kg -1 , respectively (Fu et al., 2016).Ot is clear that hazelnut samples exposed to long-term light and temperature had an increase in oil oxidation.Hence, it is critical that the drying process is carried out shortly after harvest for the long-term storage of hazelnuts.

Rancimat value
Rancimat value (RV) is based on the principle of water conductivity absorbing the degradation products formed as a result of peroxidation of unsaturated fatty acids in oils (Demirci, 2009).Polyunsaturated fatty acids are very important in oxidative degradation, which is low in varieties with low polyunsaturated fatty acids (Özdemir et al., 1998).
The interaction effect of drying and storage time was significant for RV (p˂0.05;Table 3).As expected, RV decreased with storage time for all drying methods.The results showed that RV decreased as oil oxidation increased, depending on the storage time and conditions.The present results are in general compliance with the findings of Lopez et al. (1995), Demirci (2009), Turan (2017), Turan (2018b), andTuran (2019), who reported decreasing RV during storage time.The highest RV was recorded for DM compared to CG AND GG (Table 3).This result is in agreement with Turan & İslam (2016), where DM had a higher RV than drying on CG and GG.After 24 months of storage, hazelnut RV was higher (3.59-3.87h) than the critical threshold value (3 h) reported that for oxidative stability in the Turkish nut industry.As such, 3 h RV can be regarded as limit value at the end of 24 months of storage.

Aflatoxin
On Turkey, hazelnuts are traditionally sun-dried and may be subject to mold growth and, like other nuts, subsequent AF formation due to prolonged drying under humid and rainy conditions (Simşek et al., 2002).Thus, rapid drying and reducing MC in kernels is very important prevent fungal activity.Furthermore, fungal contamination and subsequent AF production can occur in hazelnuts in the orchard, at harvest, and during post-harvest operation (Özay et al., 2008).However, AF was not detected (<0.1 ngg -1 ) in the present study although the storage conditions were appropriate for AF development.Thus, if AF was not detected during the mature hazelnut stage, they were not contaminated during the harvest, threshing, or storage period.

Conclusion
The present study investigated the effect of drying methods on the chemical properties of in-shell hazelnuts during storage.FFA increased during storage and never exceeded 0.38%.PV increased in fluctuation during storage, and the highest values were recorded for GG.RV decreased during the storage period, and was lower for DM than for CG and GG.Total AF and AFB 1 were not detected during the storage period.Therefore, when hazelnuts were stored in jute bags as in-shells at an ambient temperature, quality was maintained for a period of about 24 months.Consequently, DM was found to be more effective for the preservation of hazelnut quality than CG and GG.

Figure 1 .
Figure 1.Schematic diagram of the dryer used for hazelnut drying.
methods, S: Storage periods, CG: Concrete ground, GG: Grass ground, DM: Drying machine, PC: Protein content, LC: Lipid content, MC: Moisture content and a w : Water activity.Values are expressed as mean ± standard deviation.

Table 1 .
Moisture content of hazelnuts before and after dehydration, and after drying before storage and drying time.

Table 2 .
Effect of drying methods on protein, lipid, moisture content and water activity during 24 months storage periods.