Introduction

Honey is a natural sweet food produced by honey bees from different sources, such as plant nectar, secretions of plants, and excretion of plant-sucking insects [1]. Honey is known as the oldest natural sweetener since ancient times, and its consumption and popularity have been growing for centuries due to its therapeutic properties, high nutritional values as well as its uniquely pleasant aroma and sweetness. Although demand for authentic honey by consumers has been growing all around the world over the last years, the production and productivity of natural honey have been gradually decreased from 1.882.479 tons to 1.770.119 tons and from 20.5 kg/hive to 18.83 kg/hive, respectively, between 2017 and 2020 on a global scale [2]. As illustrated in Table 1, although Turkey is the world’s second-largest honey producer with a production of 104.077 tons, consumption of honey is very low due to high population: 3.38 g/day and 0.62 g/day in Turkey and worldwide, respectively.

Table 1 Honey production statistics
Full size table

It can be said that adulteration of honey is inevitable since the amount of honey production is very low and not enough when the world population is taken into consideration.

As a natural food product, honey is counted as one of the major sources of income and well-being for many people in the food sector. Honey adulterations are commonly made by beekeepers and sellers in two different approaches which are directly adding sugar syrups and indirectly feeding honeybees with sugar syrups. Low-cost commercially available syrups can be listed as high-fructose corn syrup (HFCS), inverted syrup, sugar cane syrup, sugar beet syrup, corn syrup, sucrose syrup inulin syrup, date syrup, and agave syrup [4,5,6].

Adulteration in honey can lead to customer misperceptions and make consumers think that authentic honey is expensively sold by producers. Furthermore, adulterated foods adversely affect public health as substances added to them may cause allergic and toxic reactions in human bodies. By providing an artificial market advantage with adulterated foods, especially bee and bee products producers, unfair competition in the food industry will arise and current stability will result in deterioration [7, 8]. Adulteration of honey made by sugar syrups has a negative impact on the chemical and physical properties of honey such as the enzymatic activity, electrical conductivity, HMF content, sugar profile and specific compound contents, etc. According to most national and international institutions and organizations, including the Turkish Food Codex, Council Directive Codex Alimentarius, FAO, and WHO, adulteration in honey is prohibited. No food ingredients or substances can be added to honey and honey can only be blended. Aims of these communiqués are to produce honey hygienically and preserve the unique properties of honey during the production processes (production, preparation, processing, storage, and transportation) until it reaches the final consumer [9, 10]. The criteria of the physico-chemical properties of blossom honey set by Turkish Honey Communiqué are given in Table 2 below.

Table 2 The criteria of the physico-chemical properties of blossom honey [10]
Full size table

In this study, key physico-chemical tests have been carried out to discriminate pure honeys from adulterated honeys.

Materials and methods

Materials

Sample collection and preparation

Pure blossom honey samples were directly collected on August 15–16, 2020 from beekeepers situated in Bingöl province, Turkey. In the laboratory, a 0.5 mm sieve (Retsch ASTM E11) was used to remove impurities from the honey samples. Adulteration agents, glucose–fructose corn starch syrup (SRF 30 Hüner), and maltose corn starch syrup (SM 40 Hüner) were purchased from Erseval candy factory located in Erzurum province, Turkey.

Sample production and design process

The pure honey samples were adulterated with the syrups at different levels (0%, 5%, 10%, 20%, 30%, 40% and 50% w/w). To prepare homogeneous adulterated samples, the mixtures were left in a controlled water bath at 35 ℃ for 1 h. Then these were stirred vigorously with a glass rod for 3 min. Finally, prepared samples were stored at room temperature until use. A total of 28 samples were designed including 24 honey–syrup samples, two pure honeys, and two syrup samples. In the sample ID column of the tables (see Tables 3, 4, and 5), the H1 and H2 represent pure honeys, GFS and MS stand for glucose–fructose and maltose syrups, respectively, and the following numbers refer to the ratio of the syrups. For instance, sample ID of H1GFS50 means H1 honey involving GFS syrup with a ratio of 50%.

Table 3 Moisture, pH, free acidity, diastase number, electrical conductivity, HMF, proline, and color parameters of the honey samples
Full size table
Table 4 Sugar profile of the samples
Full size table
Table 5 δ13honey, δ13protein, and C4 sugar values of the samples
Full size table

Methods

Moisture content, pH, free acidity, electrical conductivity, and proline content analyses were determined according to harmonized methods of the International Honey Commission (IHC) [36].

Moisture content

The moisture content value was determined from the refractive index of the sample at 20 ℃ using an Abbe refractometer. The obtained refractive index was converted into moisture content (g/100 g) using the standard table.

pH

10 g sample was dissolved in C02-free 75 ml pure water and pH measurement was performed after calibrating the pH meter (Orion 3-Star, Thermo Fisher Scientific Inc.) with pH 4.00, 7.00 and 10.00 buffer solutions (Thermo Scientific Inc.).

Free acidity

Free acidity was performed by dissolving 10 g sample in 75 ml C02-free pure water and titrated with 0.05 M NaOH to pH 8.30 using a pH meter (Orion 3-Star, Thermo Fisher Scientific Inc.). Free acidity was calculated as follows;

Free acidity (mEq/kg) = 5 × volume of NaOH used in titration (ml).

Electrical conductivity

20 g of sample dry matter was dissolved in 100 ml ultrapure water (18.2 MΩ.cm resistivity, Sartorius H2O-I-1-UV-T Arium Comfort) and electrical conductivity of the sample solution was measured by a conductivity meter Consort C3010 (Consort bvba, Turnhout, Belgium) at 20 ℃. The result was expressed as µS/cm.

Proline

5 g of sample was dissolved in pure water and then diluted to a volume of 100 ml in a volumetric flask. 3% by volume ninhydrin (Sigma-Aldrich) in ethylene glycol monomethyl ether (Fisher Chemical) and 40 mg/50 ml aqueous L-proline (Merck) reference solution was prepared. 0.5 ml sample solution, 0.5 ml blank solution (pure water), and 0.5 ml prepared proline solution were added to each test tube. Then 1 ml of formic acid (Sigma-Aldrich) and 1 ml of ninhydrin solution were added to each tube. The tube caps were screwed firmly and shaken at 400 rpm in a shaker (IKA KS 4000 IC Control, Staufen, Germany) for 15 min at room temperature and then the tubes were subjected to heat treatments. The tubes were first heated in a boiling bath for 10 min and then immediately subjected to another heat treatment in a 70 ℃ water bath for 15 min and to each tube, 5 ml 2-propanol (Carlo Erba) was added. After the heat treatments, the tubes were left to cool down for 45 min at room temperature and then proline content was determined using a UV–Vis double beam spectrophotometer Jasco V-650 UV (JASCO, Tokyo, Japan) at 510 nm in a 1 cm quartz cuvette (Hellma Analytics, Müllheim, Germany). Proline concentration of samples was calculated using the following formula;

Proline (mg/kg) = (Es/Ea) × (E1/E2) × 80, where Es is the absorbance of the sample solution, Ea is the absorbance of the proline standard solution, E1 is the amount of proline used for the standard solution (mg), E2 is the weight of honey (gr), and 80 is the dilution factor [36].

Diastase number

The diastase activity was calculated as diastase number (DN) in Gothe units. DN is equal to % the amount of 1% starch solution (ml) that the diastase enzyme in 1 g of honey can completely hydrolyze in 1 h. The test was carried out according to the Turkish Standards Institution [37].

Color

To completely dissolve sugar crystals, the sample was heated to 50 ℃ in a water bath and then transferred into a spectrophotometer cuvette (Biosigma SpA, Italy). Color measurement was conducted using a chroma meter (Konica Minolta CR5, Japan) and L, a, and b values were obtained at 25 ℃. L, a, and b parameters were used to calculate the ΔE value which was calculated as follows;

$$\Delta E = {[{(L-L0)}^{2} + {(a-a0)}^{2} + {(b-b0)}^{2}]}^{1/2}$$

Hydroxyl methyl furfural (HMF) by HPLC–DAD

10 g of sample was diluted to 50 ml with ultrapure water (18.2 MΩ.cm resistivity, Sartorius H2O-I-1-UV-T Ariuim Comfort) and filtered through 0.45 μm Minisart filter (Sartorius Stedim Biotech GmbG, Goettingen, Germany). A stock solution of HMF (100 ppm) was prepared with a standard HMF (J&K, Haihang Industry Co., Ltd.) and then using the stock solution, six-point calibration curve of HMF in the range of 0.1 ppm–10 ppm was prepared. HMF content was determined using a HPLC instrument (1260 Infinity II, Agilent Technologies) equipped with a diode array detector (1260 Infinity II WR) and C18 column (ACE 5, 250 mm × 4.6 mm, 5 μm). Analysis was performed in isocratic mode (90:10 water: methanol at 1 ml/min flow rate) operated at 285 nm, at 30 ℃ column temperature.

Sugar profile by HPLC-RID

5 g of sample was weighed into a beaker and dissolved in ultrapure water (18.2 MΩ.cm resistivity, Sartorius H2O-I-1-UV-T Arium Comfort). 25 ml methanol (Chromasolv grade, Sigma-Aldrich) was pipetted into a 100 ml volumetric flask and then filled to the mark with sample solution and ultrapure water. The prepared methanol sample solution was filtered through a 0.45 μm filter (Sartorius Stedim Biotech GmbG, Goettingen, Germany). Determination of sugars was carried out using an HPLC instrument (1260 Infinity II, Agilent Technologies) equipped with a refractive index detector (RID, 1260 Infinity II) and Zorbax NH2 column (Agilent, 4,6 × 250 mm 5 μm). Analysis was performed in isocratic mode (80:20 acetonitrile (HPLC grade, Sigma-Aldrich)/water at 1.3 ml/min flow rate) at 30 ℃ column and detector temperature. Seven-point calibration curves were prepared for glucose (D-glucose anhydrous, Fluka, 300 ppm-25000 ppm), fructose (D(-)-Fruktoz, Sigma-Aldrich, 300 ppm-25,000 ppm), sucrose (Fluka, 10 ppm-2000 ppm), and maltose (D-( +)-maltose monohydrate, Fluka, 300 ppm-25000 ppm) sugar standards. An example of the individual sugar profile HPLC chromatogram is provided in the supplementary document (see Fig. S1).

C4 sugars by IR-MS

For the extraction of protein from honey, 12 g of the sample was put into a 50 ml centrifuge tube, diluted with 4 ml ultrapure water, and vortexed until the sample fully dissolved. Then 2 ml of 10% tungstic acid sodium salt solution and 2 ml of 0.335 M sulfuric acid was added to the sample solution, the tube was then placed in a water bath at 80 ℃ for 30 min. Following that, 1 ml of 10% tungstic acid sodium salt solution and 1 ml of 0.335 M sulfuric acid were added into the tube and this step was repeated three times. After this heat treatment, 25 ml ultrapure water was added to the tube and the solution was centrifuged at 1500 rpm for 5 min at room temperature. After centrifugation, the supernatant was removed and the tube was filled to the mark with ultrapure water. This centrifugation was repeated four times until the supernatant was clear. The supernatant was discarded, the pellet was transferred to a watch glass, and left to dry in an oven at 70 ℃ for 45 min. Between 150 and 200 µg dried protein, honey and syrup samples were weighed in each tin capsule and then were tightly closed. Packed samples were then introduced into the autosampler of the EA IR-MS system (DELTA V Plus, Thermo Fisher Scientific). The degree of adulteration of sample was calculated according to the AOAC method 998.12 [38] by the equation given below;

$$\% \; Adulteration= \left[\frac{\left(\delta \permil \;protein-\delta \permil \;honey\right)}{(\delta \permil\; protein- \delta \permil\; syrup)} \right]x100$$

Results and discussion

Physico-chemical parameters (moisture content, pH, free acidity, diastase number, electrical conductivity, HMF, proline, color) of blossom honey samples (H1 and H2), adulterants (MS and GFS syrups), and adulterated honey samples (H1GFS, H1MS, H2GFS, and H2MS) are presented in Table 3. HPLC-RID sugar profile and IR-MS C4 sugar analysis are given in Tables 4 and 5, respectively.

Moisture content

In terms of honey quality, moisture content is one of the most important parameters that affects the physico-chemical properties of honey and will help to determine the storage conditions and shelf life of the product. Moisture content in honey above 20% can trigger yeast growth which subsequently causes the fermentation process [11]. The initial moisture content of H1 and H2 honey samples and GFS and MS adulterants was found as 14.96%, 14.56%, 17.16%, and 13.76%, respectively (see Table 3). There were linear downward with the addition of GFS syrup and linear upward trends with the addition of MS syrup (see Fig. 1). Since both sugar syrups have lower than 20% moisture content, all adulterated samples ranged below 20% which do not exceed the maximum limit of moisture specified in the Turkish Food Codex Communiqué on honey [10].

Fig. 1
figure 1

Variation in moisture content of the samples at different syrup ratios (the red dashed lines represent the permitted maximum level)

Full size image

Among the adulterated samples, the highest moisture content was seen in the H1GFS50 sample at the amount of 15.96%.

In a study, different honey samples were adulterated with sucrose beet syrup at the ratio of 10–50% and the moisture content of honeys exceeded the 20% limit at 40% and 50% adulteration levels [12]. In another study [13], it was determined that the moisture content of honey adulterated with palm sugar syrup in the range of 5–30% decreased from 19.7% to 21.6% with the increase of the added syrup level. By examining the data obtained from this study and current literature, it can be said that it is not always possible to detect adulterated honey according to their moisture content.

pH

The value of pH is another parameter that has an impact on the characterization of honey. In the presence of low pH, microbial growth as well as microbial reproduction is inhibited [14]. The early pH values of H1 and H2 honey samples and GFS and MS syrups were observed as 3.70, 3.76, 5.42, and 5.19 respectively (see Table 3). In general, the pH values of the adulterated samples were increased with the increase in syrup addition levels. Since MS syrup has a higher pH value than GFS syrup, the increase in pH levels of adulterated samples in the samples prepared with MS syrup was higher than in samples prepared with GFS syrup. Among the adulterated samples, the highest pH value was examined in the B2MS50 sample as 3.80. Considering the pH values of the syrup and honey samples, changes in the pH values of adulterated samples were found lower than expected (Figure S2). This could be due to the buffering property of the honey matrix that keeps the pH values of the adulterated samples at lower levels.

(Transferred to Supplementary material as Fig. S2) Similar results were reported in a previous study [14], in which the addition of HFCS at different ratios (10%, 25%, 50% and, 75%) to pure honey samples increased honey’s pH value. Initially honey and HFCS had 3.10 and 4.70 pH values, respectively. An increase in the addition level of HFCS increased pure honey’s pH rate. In the adulterated honey samples, the pH rates ranged between 3.30 and 3.98.

Free acidity

The main source of free acidity is the presence of organic acids. Although organic acids are around 0.5% of the total honey composition, they have important roles in the organoleptic, physical, and chemical properties of honey [15]. Free acidity value is also seen as a fermentation indicator. Bad storage conditions, exposure to direct heat, microbial contamination, and/or components that are decomposed by the naturally occurring osmophilic yeasts in honey cause an increase in the free acidity value [16, 17]. Free acidity of H1 and H2 honey samples was measured as 27.00 mEq/kg and 18.50 mEq/kg, respectively, while the value for GFS and MS syrup was found as 1.63 mEq/kg and 1.40 mEq/kg, respectively (see Table 3). It was determined that there were high negative correlations between the free acidity values and the amount of syrup added to the samples. As the syrup addition level was increased, the free acidity values of the adulterated samples were decreased linearly (Fig. 2). Free acidity values of samples ranged between 24.17 mEq/kg and 17.17 mEq/kg depending on adulteration levels.

Fig. 2
figure 2

Variation in free acidity of the samples at different syrup ratios (the red dashed lines represent the permitted maximum level)

Full size image

In a study [18], glucose, hydrolyzed inulin syrup, malt wort, and inverted sugar were used for the adulteration of different authentic honeys at ratio of 5%, 10%, 20%, 30%, 40%, and 50% respectively. It was observed that glucose, hydrolyzed inulin syrup, malt wort, and inverted sugar syrups increased the amount of free acidity, although fructose did not alter the free acidity values of the samples. Average free acidity of authentic honey went up from 19.44 mEq/kg to 162.88 mEq/kg with the addition of 50% sugar syrups.

According to the Turkish Food Codex Communiqué on honey [10] and Council Directive (2001), honey should not have more than 50 mEq/kg free acidity; however, there is no minimum tolerance level for free acidity. In this study, since all the adulterated samples remained lower than 50 mEq/kg, no adulterations were detected among the samples (Fig. 2).

Fig. 3
figure 3

Variation in diastase number of the samples at different syrup ratios (the blue dashed lines represent the permitted minimum level)

Full size image

Diastase number (DN)

Diastase is considered an indicator of the freshness and purity of honey [19]. The initial DN of the H1 and H2 honey samples was determined as 28.93 and 32.74 while diastase activity was not found in both GFS and MS syrups. There were negative correlations between the amount of syrup added and the diastase numbers of the adulterated samples. DN of the adulterated honey samples was generally gradually decreased with the increase of syrups added. The DN of the adulterated honey samples ranged between 31.25 and 17.89. DN number values of all adulterated samples were found over 8.00 (see Fig. 3) which is in the acceptable range stated in EC Council Directive [9] and Turkish Food Codex Communiqué on honey [10].

Fig. 4
figure 4

Variation in electrical conductivity of the samples at different syrup ratios (the red dashed lines represent the permitted maximum level)

Full size image

Pure honey samples were adulterated with sucrose syrup at the range of 10–50%. Due to the absence of diastase activity in sucrose syrup, the average DN of the pure honey samples gradually decreased from 14.60 to 7.50 with increasing adulteration levels [12]. Czipa et al. in their study [20] directly added a different type of sugar syrups (glucose, invert and fructose-glucose syrup) at a ratio of 30% and 40% to pure acacia honey samples. They reported that the average DN of samples was 28.80 and depending on the type of sugar syrup added, the value decreased down to 15.40. In a study conducted by Ozcan et al. [21], the diastase activities of honey obtained from bees fed with sucrose syrup and inverted syrup were compared with honey obtained from bees not fed with sugar. The highest diastase value was found in pure honey with 10.90 and the lowest diastase value was determined in honey obtained from bees fed with invert syrup. The diastase activity of honey produced by bees fed with sucrose syrup was found to be 8.30. It can be seen from our study and the literature that direct or indirect addition of sugar syrups decreases the diastase activity of honey but the DN value of the adulterated samples mostly remains within the safe limit which is the amount of 8.00 DN set by Turkish Honey Communiqué [10] and Council Directive [9] as a minimum requirement.

Electrical conductivity

The amount of electrical conductivity (EC) varies directly with the presence of mineral substances, organic acids, and other organic compounds [22]. The EC values of H1 and H2 honey were determined as 307.00 μS/cm and 242.00 μS/cm, respectively, while the EC values of GFS and MS syrups were found to be 5.20 μS/cm and 3.00 μS/cm, respectively (see Table 3). In the measurements of adulterated samples, high negative correlations were detected between the added syrup level and the EC of adulterated samples. It was determined that the amount of electrical conductivity decreased proportionally as the amount of added sugar syrups increased (see Fig. 4).

Oroian et al. [23] reported that EC values of different types of honey (acacia, tilia, and polyfloral) were altered by adulteration with fructose and hydrolyzed inulin syrups. While the addition of fructose syrup decreased, hydrolyzed inulin increased the EC value of the samples as the adulteration rate increased. EC values of the adulterated samples were found to be between 24.30 and 2920 μS/cm. According to Turkish Honey Communiqué regulation [10], EC value for blossom honey should be less than 800.00 μS/cm. In our study, blossom honey samples and all the adulterated honey samples showed complete conformity to this regulation. The highest and the lowest conductivity values in adulterated samples were determined as 295.00 μS /cm and 149.87 μS/cm in B1GFS5 and B2MS50 samples, respectively (see Table 3). These results show that according to the regulations, the addition of sugar syrups to blossom honey affected positively the electrical conductivity value so there should also be a minimum requirement for the electrical conductivity requirement of blossom honey in the regulation.

HMF

HMF is an indicator of honey freshness [24]. It is known that sugar syrups added to honey generally increase the HMF content due to the higher presence of HMF in sugar syrups [25]. Furthermore, for the added sugars to disperse homogeneously in the honey–syrup mixture and gain a uniform appearance, the mixture is heated after adding the syrup, and this process causes an increase in the amount of HMF. While the HMF content of H1 and H2 honey was determined as 7.12 mg/kg and 5.21 mg/kg, respectively, in GFS and MS syrups, HMF contents were found as 16.96 mg/kg and 10.53 mg/kg, respectively (see Table 3). As the syrup addition level increased in the adulterated samples, the amount of HMF increased linearly showing high positive correlations (see Fig. 5). HMF content of adulterated honey samples was ranged between 5.68 mg/kg and 11.94 mg/kg.

Fig. 5
figure 5

Variation in HMF content of the samples at different syrup ratios (the red dashed lines represent the permitted maximum level)

Full size image

In a study conducted by Craciun et al. [26], honey samples were adulterated by adding three different sugar syrups directly to authentic honey and indirectly feeding the bees with sugar syrups. While the average HMF content of authentic honeys was 1.21 mg/kg, the average HMF of directly syrup-added honeys was found to be 21.20 mg/kg. Furthermore, the average HMF value of adulterated honey obtained from bees fed with sugar syrups was found as 29.90 mg/kg. In another study, monofloral acacia, tilia, and sunflower honeys were adulterated with maple, inverted sugar, agave, rice and corn syrup in the concentration of 5%, 10%, and 20%. Initially, honey samples had an average of 3.50 mg/kg HMF content with the addition of sugar syrups and average HMF values of the samples increased gradually in the range of 10.10 to 35.10 mg/kg [5]. Although sugar syrups used for the adulteration of honeys had higher content of HMF, adulterated honeys mostly did not exceed the limit of 40 mg/kg set by the Turkish Food Codex Communiqué on honey [10].

Proline content

Salivary glands of honey bees and plants are the main sources of amino acids of honey counted as an indicator in determining honey fraud whether it has been imitated or adulterated. Proline is the dominant amino acid in honey and it is seen as an indicator of protein amount in honey since it constitutes 50–85% of the total amino acid content [27, 28]. In H1 and H2 honey samples, proline contents were found as 965.54 mg/kg and 587.37 mg/kg, respectively. Proline values of GFS and MS syrups were calculated as 53.13 mg/kg and 85.88 mg/kg, respectively (see Table 3). High linear negative correlations were determined between the addition rate of syrups and the proline content of adulterated samples (see Fig. 6).

Fig. 6
figure 6

Variation in proline content of the samples at different syrup ratios (the blue dashed lines represent the permitted minimum level)

Full size image

The highest and the lowest amounts of proline were found in H1MS5 and H2GFS50 at 902.53 mg/kg and 314.96 mg/kg, respectively. In the study conducted by Kropf et al. [20], a pure honey sample was adulterated with fructose, glucose, and two different inverted sugar syrups at the rates of 30% and 40%. While the proline value of pure honey was determined as 284.00 mg/kg, it was also determined that the proline values of the syrup-added honey samples decreased and the proline values of the samples ranged from 179.00 mg/kg to 274.00 mg/kg. In another study, proline values of honey obtained from bees fed with sucrose syrup at different rates were compared. Proline amounts of 416.40 mg/kg, 501.60 mg/kg, and 630.00 mg/kg were determined in honey obtained from bees fed with sugar syrup continuously, fed only with sugar syrup in spring, and not fed with sugar syrup, respectively [29].

Color

Color variability in honey depends on the nectar of plants and the plant origin because honeybees directly collect nectar and pollen from plants. It plays an important role in determining the market price of honey as it is one of the most important physical parameters affecting the preferences of consumers in many countries [30]. As a result of the color test, the L, a, and b values were obtained. Color differences, ΔE, were generated from the L, a, and b values of the samples. Raising the percentage of adulterants, GFS and MS, in the adulterated samples caused a gradually increase in the L and ΔE values (see Figures S3 and S4) while gradually causing decrease in a and b values (see Figures S5 and S6). The highest L and ΔE values were detected in the B2GFS50 and B1MS50 samples. The lowest a and b values were found in the samples of B2GFS50 and B2MS40, respectively (see Table 3).

In a study carried out by Ribeiro et al. [14], honey samples were adulterated with high-fructose corn syrup at a rate of 10–75%. It was found that, while the L value of honeys with syrup added increased and b value decreased proportionally with the syrup addition rate. Yılmaz et al. [31], in their study, detected the changes in L, a, and b values by adulterating a honey sample with fructose and sucrose syrups in the range of 5–50%. With the increase in the syrup ratio added to the honey sample, an increase in L value and a decrease in a and b values were determined. While the highest L value was detected in the sample with 50% sucrose syrup added, the lowest a and b values were determined in the samples including 50% sucrose syrup and 50% fructose syrup, respectively.

Sugar profile by HPLC

To prevent imitation and adulteration, there are certain criteria set by national and international standards for the sugar content of honey. According to the Turkish Food Codex Communiqué on honey [10], maltose and sucrose can be found at a maximum of 4 (g/100 g) and 5(g/100 g) respectively, and the F + G value should be at least 60%, and the fructose/glucose ratio should be in the range of 0.9–1.4. Honey that does not meet these criteria is considered as fraudulent. In the adulterated honey samples, fructose + glucose (g/100 g) values and F/G ratios were decreased gradually (Figs. 7 and 8, respectively), while maltose (g/100 g) values progressively increased with increasing additions of GFS and MS syrups (Fig. 9).

Fig. 7
figure 7

Variation in F + G value of the samples at different syrup ratios (the blue dashed lines represent the permitted minimum level)

Full size image
Fig. 8
figure 8

Variation in F/G value of the samples at different syrup ratios (the red dashed lines represent permitted maximum level; the blue dashed lines represent the permitted minimum level)

Full size image
Fig. 9
figure 9

Variation in maltose value of the samples at different syrup ratios (the red dashed lines represent the permitted maximum level)

Full size image

Except for H1MS50 and H2MS50, all the adulterated samples contain more than 60% of F + G. Furthermore F/G ratio of all adulterated samples ranged between 0.9 and 1.4. Sucrose sugar was not found in H1 and H2 honey samples and in both GFS and MS syrups. Results of F + G (g/100 g), F/G ratio, and maltose (g/100 g) are exhibited in Table 4. These results indicate that adulterations are detected at the level of 5%—50% depending on the sugar syrup and honey type. In the samples of H1GFS, H1MS, H2GFS, and H2MS, adulteration was detected at ratio of starting from 20%, 10%, 20%, and 5% respectively.

In a study [32], honeydew honey was adulterated with glucose, fructose, inverted sugar, hydrolyzed inulin syrup, and malt wort at the rate of 5–50%. According to the EC Council Directive [9], adulterations could be detected in samples which were adulterated with glucose, fructose, and malt-worth additions at the rate of between 20–50%, 5–50%, and 20–50%, respectively. Adulterations were discovered by considering the F/G ratio and maltose ratio of the adulterated samples. The F/G ratio of adulterated samples ranged between 1.39–3.68 and 1.12–0.37, respectively. Furthermore, the maltose value of adulterated with malt worth ranged between 3.0 and 24.92 (g/100 g). In a similar study carried out by Tosun [33], three different blossom honey samples were adulterated with glucose, sucrose, and HFCS in the range of between 10% and 50%. According to the author’s results, the F + G ratio of three honey samples went below 0.9 with the addition of glucose syrup at a rate of 40% and 50%. The F/G ratio of samples ranged between 0.54 and 0.77. The F + G value of adulterated samples was found within safe limits which were more than 60 g/100 g.

C4 sugars by IR-MS

According to the Turkish Food Codex Communiqué on honey [10], the ratio of C4 should not be greater than 7%, δ13 protein—δ13 honey (‰) value should not be greater than ‰1, and the honey δ13 C value should be—23 (‰) or more negative (See Table 2). The δ13 C honey (‰) values of H1 and H2 pure honey samples were found as − 25.80 and − 26.13, respectively, while the values for GFS and MS syrups were detected as -15.86 and − 16.14, respectively. C4 sugar adulteration levels (%) of the samples were increased with increasing ratio of sugar syrups in the samples (See Fig. 10).

Fig. 10
figure 10

Variation in C4 sugar value of the samples at different syrup ratios (the red dashed lines represent the permitted maximum level)

Full size image

As can be seen from Table 5, adulterations were detected at the level ranged between 5 and 50% depending on honey and sugar syrup type using EA-IRMS.

In a study [34], honey samples were adulterated with high-fructose corn syrup, glucose syrup, and sucrose syrup by the addition level of 10%, 20%, 30%, 40%, and 50%. In this study, although adulterations could not be detected in samples adulterated with glucose syrup and sucrose syrup, adulterations made with high-fructose corn syrup were detected at the rate of 20%, 30%, 40%, 50%, as 14.30%, 32.80%, and 41.65%, respectively. In a similar study, honey samples adulterated with high-fructose corn syrup at rates of 20%, 60%, 90% were analyzed, revealing adulteration levels of 11.20%, 30.60%, and 48.20%, respectively [35].

Statistical analyses

SPSS analysis

IBM SPSS version 22 was used to generate statistical data through general linear model (GLM) at significance level (p value) of 0.05. Syrup type and syrup ratio were established as independent variables whereas moisture, pH, free acidity, diastase number, electrical conductivity, HMF, proline, C4 sugar, F + G, F/G, maltose, L, a, b, and ΔE values were chosen as dependent variables. All the experimental results are presented in Tables 6 and 7. In Table 6, the results of samples prepared with H1 honey, and in Table 7, results of samples prepared with H2 are listed. For the results of samples prepared with H1 and H2 honey samples, independent variables, syrup type and syrup ratio, had statistically significant effects on moisture, free acidity, electrical conductivity, HMF, proline, F + G, F/G, maltose, L and b values, while syrup type was not significant for pH diastase number and C4 sugar, while syrup type was not significant (p > 0.05) for diastase number, C4 sugar, ΔE values of the samples. Furthermore, syrup type did not significantly impact on pH and value of the samples prepared with H1 honey.

Table 6 Statistical analysis of the test results for samples prepared with H1 honey (H1-GFS, H1-MS)
Full size table
Table 7 Statistical analysis of the test results for samples prepared with H2 honey (H2-GFS, H2-MS)
Full size table

Principle component analysis (PCA)

According to results obtained from moisture, pH, free acidity, diastase number, electrical conductivity, HMF, and proline content analyses, we were not able to determine adulteration by these routine laboratory methods. Because, PCA software (XLSTAT, 2021, Addinsoft, New York, NY) was used to detect adulteration based on the physico-chemical parameters of pure honey and adulterated honey samples. By utilizing PCA, adulterations were detected at the range of 5–50% levels. It can be seen from the data obtained from PCA results, as shown in Table 8 and Fig. 11, that two different PC values F1 and F2 were enough to classify the blossom honey samples and the adulterated honey samples.

Table 8 Results of the PCA analysis using data obtained from moisture, pH, free acidity, diastase number, electrical conductivity, HMF, and proline analyses of the samples
Full size table
Fig. 11
figure 11

Principal component scores of the authentic honey samples and honey samples adulterated with GFS and MS syrup

Full size image

The resulting data matrix included 18 variables which are moisture, pH, free acidity, diastase number, electrical conductivity, HMF, color (L, a, b, and ΔE), and proline values of the samples. The first two PCAs scores, F1 + F2 axes were able to explain 95.82% and 97.17% total variance of the samples prepared with H1 and H2 honey samples, respectively. These high percentages of total variances indicate the success of the PCA for the classification of the honey samples. It can be observed from Fig. 11, the PCA score plot for H1 and H2 samples, both pure honey samples were settled at the same place on the left side of the plot near to origin point. Furthermore, both H1 and H2 samples adulterated with GFS and MS syrups at the range of 5%, 10%, 20%, 30%, 40%, and 50% were also clustered between each other based on adulteration levels in the same quadrant.

Conclusion

According to the results of the experimental tests, several conclusions can be drawn regarding the adulterated honey samples. First, the free acidity, proline, diastase number, electrical conductivity, a and b values of the adulterated samples demonstrated a linear decrease with an increase in the addition level of syrups. In addition, all the adulterated honey samples displayed linear increases in HMF, C4 sugar, F+G, F/G, maltose, L, and ΔE values with the increasing addition of syrup. Moreover, the moisture content of the adulterated samples exhibited a decrease with the addition of GFS syrup but an increase with the addition of MS syrup. Adulterations were detected at a certain level by HPLC sugar profile and IR-MS C4 sugar analyses. The outcomes from assessing moisture, pH, free acidity, proline, diastase number, color, electrical conductivity, and HMF individually did not reveal any evidence of sugar syrup adulteration. Nevertheless, employing principal component analysis (PCA) to examine the data collected from these evaluations effectively identified adulterations across all-syrup ratios (5–50%).