JFS C: Food Chemistry B. FALLICO, E. ARENA, AND M. ZAPPALA ABSTRACT: 5-Hydroxymethylfurfural (HMF) is the most important intermediate product of the acid-catalyzed dehydration reaction of hexoses and/or Maillard reaction; furthermore, it is the most used index to evaluate thermal damages or ageing in food products. Usually its degradation reactions, being very slow, are neglected. This study reports the findings concerning the degradation kinetics of HMF, in honeys of different floral origin at a temperature between 25 and 50 ◦ C. The results highlighted higher degradation rates (k HMF degradation ) compared to the corresponding formation rates (k HMF formation ) in chestnut and citrus samples. Similar k-values were found in multifloral honey. Moreover, the reaction of HMF degradation was characterized by lower activation energy (E a ) values compared to E a formation values. The final concentration of HMF in honey, during storage at room temperature, should be ascribed to high sugar concentration. The fluctuation of HMF in honeys could depend on the equilibrium between the accumulation and the degradation processes. This can affect the validity of HMF as storage index in some honeys, above all during the analysis of those honeys whose legislation is too restrictive (citrus) or in chestnut honey analysis where it does not accumulate. Keywords: browning, HMF kinetics, honey quality, shelf life Introduction H oney has been always considered as a healthy and natural product. The most important parameters to evaluate honey quality are: safety, the absence of contaminants (antibiotics, pesticides, and heavy metals), floral origins, the area of production, and freshness. The most used indexes to measure honey overheating and/or aging are: the 5-hydroxymethylfurfural (HMF) and diastase. The HMF is the most important intermediate product of 2 reactions: the acid catalysed degradation of hexose and the decomposition of 3-deoxyosone in Maillard reaction (Belitz and others 2004). Diastase is an enzyme naturally present in fresh honeys (Persano Oddo and others 1999), whose levels decrease during storage and/or heating (Sanchez and others 2001). A description of the step-bystep process using these indexes, as well as a critical essay of diastase as quality index of honeys, were reported by White (1992, 1994). Moreover, recently it has been shown that the addition of diastase levels to models to predict honey shelf life does not improve the results, but, on the contrary it increases the uncertainty of the models and gives, as results, shorter shelf life values (Fallico and others, 2009). The International Bodies (Codex Alimentarius Commission, CODEX STAN 12-1981 [rev. 2, 2001]) established that the highest allowed amount of HMF in honey should be 40 ppm, with the exception of honeys of tropical origin (80 ppm). At national level different approaches have been used; in the U.S.A. (Docket nr 2006P-0101), Australian and New Zealand (ANFS, Standard 2.8.2.) honey standards, the HMF level was not considered. Argentina (Cap. X, Codigo Alimentario Argentino) and the MERCOSUR area countries (Msys As. N. 003, 11.01.95), as well as Canada MS 20080305 Submitted 4/22/2008, Accepted 8/19/2008. Authors Fallico and Arena are with DOFATA sez. Tecnologie Agroalimentari, Univ. di Catania, Via S.Sofia 98, 95123 Catania, Italy. Author Zappala is with Quality Assurance Labs, Gruppo Zappalà, Via Ardichetto sn, 95019 Zafferana Etnea, Italy). Direct inquiries to author Fallico (E-mail: bfallico@unict.it). R Institute of Food Technologists doi: 10.1111/j.1750-3841.2008.00946.x  C 2008 Further reproduction without permission is prohibited (C.R.C., c. 287), have adopted the same HMF limits, but determined postblending or processing. The most restrictive HMF standards have been adopted by the European Union (Dir. 2001/110 EC), guaranteeing the limits up to the “Sell by” date, usually 36 mo. The EU legislation has adopted also a 3rd limit: 15 ppm HMF level for low natural enzyme honeys (3 < diastase < 8DU). The HMF level in honeys depends on the chemical properties and on the floral origin of honey (Singh and Bath 1997, 1998; Fallico and others 2004; Zappalà and others 2005), temperature, time of heating (Bath and Singh 1999), and storage conditions (Sancho and others 1992). The kinetics of HMF development in unifloral honeys was correlated to pH, total and free acidity, highlighting the nonequivalence among different honey types submitted to heating treatment (Fallico and others 2004). The level of HMF in food depends on the equilibrium between destruction by oxidation and formation from precursors (Morales and others 1997). Hidalgo and Pompei (2000), studying furosine and HMF kinetics in tomato products, found decreased HMF levels when such products were stored at room temperature. Ferrer and others (2005) highlighted an irregular evolution (increasing and decreasing) of HMF levels in 2 milk based infant formulas monitored for 2 y at 20 and 37 ◦ C storage temperature. The same HMF behavior was reported by Chàvez-Servı̀n and others (2006) investigating HMF evolution in a milk-based infant formula stored at 25 ◦ C, while the same formula stored at 37 ◦ C showed a regular increase of potential and free HMF. The degradation of sugars, as well as of HMF, to produce levulinic acid is well known (Girisuta and others 2006a, 2006b). Recently, the fluctuation of HMF of unifloral honeys, stored at room temperature, has been considered a very important parameter to put these honeys on the market or not and/or to estimate their shelf life (Fallico and others, 2009). It is well known that HMF is involved in a cascade of complex reactions (consecutive and parallel), that not easily conform to a “classical chemical reaction” (Yaylayan 1997), this creates difficulties in determining kinetics parameters as reported by Martins and Vol. 73, Nr. 9, 2008—JOURNAL OF FOOD SCIENCE C625 C: Food Chemistry Degradation of 5-Hydroxymethylfurfural in Honey Degradation of HMF in honey . . . others (2003, 2005). So, there are no data in the literature about HMF degradation kinetics in food. The aim of this article was to evaluate and measure the rate of HMF degradation in honeys of different floral origins. Materials and Methods calibration curve, measuring the signal at λ = 285 nm. Two aliquots were drawn from each sample of each honey type, and analyzed in double: the reported HMF concentration is therefore the average of 4 values. Kinetics C: Food Chemistry Honey samples and kinetics trials Samples of 2 kg of each honey type—citrus (Citrus aurantium L.), chestnut (Castanea sativa L.), and multifloral—were picked from relative stainless steel drums of 300 kg directly provided by local market leader in the production and packaging of honey (Zafferana Etnea, Catania, Italy). All honey samples were produced during the 2007 season and their floral origin was guaranteed by the producer. HMF degradation kinetics in each honey type was carried out using 1 reference and 3 fortified samples (500 g each) with standard of HMF. Three different amount of HMF, in very high concentration with respect to the normal levels naturally found in honey, were added to each honey sample. Samples were stirred up to the total dissolution of HMF (about 30 min at 30 ◦ C). The description of all fortified samples is reported in Table 1. For each honey type a control sample was heated, even the fortified samples were heated at 3 different temperatures 50 ± 1, 35 ± 1, and 25 ± 2 ◦ C for 8, 12, and 75 d, respectively. Samples heated at 50◦ and 35 ◦ C were heated in a water bath, while samples heated at 25 ◦ C were stored in a controlled temperature cabinet. A sample of each honey was withdrawn regularly and cooled in an ice water bath before analyses. The mathematical methods applied to HMF degradation kinetics should be those of the consecutive reactions (Erdoğdu and Şahmurat 2007), where the concentration of HMF at time t keeps in account both the acid-catalyzed hexoses dehydration (k 1 [sugars]n ) and the HMF degradation routes (–k 2 [HMF]m ). As the aim of this article was to measure the HMF degradation routes we tried to isolate the 2nd part of the reaction pattern as follows: 1. Increasing significantly the initial concentration of HMF; 2. Subtracting, at time t the amount of HMF developed during the heating of the corresponding reference samples as follows: [HMF]t = [HMF]fortified(t) − [HMF]ref.sample(t) Pseudo zero-order models, belonging both to HMF formation (50 ◦ C) and degradation (at the 3 studied temperatures) were used. The rates of HMF degradation were obtained through the following equation: [HMF]t − [HMF]0 = kt where k is the kinetics constant (ppm/d) extrapolated by plotting [HMF]t – [HMF] 0 and t the heating time (days). The activation energy (E a kJ/mol or kcal/mol) values of HMF Analytical determinations degradation routes were calculated from the rate coefficients at difMoisture, water activity, electrical conductivity, ash, free acids, ferent temperatures applying the Arrhenius equation (Robertson lactones, total acidity, and pH were determined on each sample 1993). according to Fallico and others (2004). Sugars (D-glucose and Dfructose) were determined by an enzimatic-spectrophotometric Results and Discussion method, using a kit of Boheringer Mannheim Enzymatic Bio Analn this study, we have chosen 3 honey types having different ysis, R-biopharm (Germany). chemical properties (Table 2), and different capacity of HMF forHMF analysis. Five grams of honey were diluted with distilled mation (Fallico and others 2004, 2009). All samples had a high moiswater up to 50 mL, filtered with a 0.45-µm filter and immediately inture value, about 20%. Multifloral and chestnut honeys are charjected in a HPLC (Varian 9012Q) equipped with a diode array detecacterized by high levels of fructose, only chestnut for low levels of tor (Varian, Star 330). The HPLC column was a Merck Lichrospher, glucose. The multifloral honey showed the lowest pH and the highRP-18, 5 µm, 125 × 4 mm, fitted with a guard cartridge packed est total acidity and HMF level. On the other hand, the highest pH with the same stationary phase (Merck, Milan, Italy). The HPLC value and the lowest acidities and HMF levels were found in chestconditions were the following: isocratic mobile phase, 90% water nut honey (Table 2). The chemical characteristics of the 3 studied at 1% of acetic acid and 10% methanol; flow rate, 0.7 mL/min; inhoneys are typical of their botanical origin (Piazza and others 1991; jection volume, 20 µL. All the solvents were HPLC grade (Merck). Persano Oddo and others 1995). The wavelength range was 220 to 660 nm and the chromatograms The accumulation of HMF, in honey, should be ascribed to the were monitored at 285 nm. The HMF was identified by spiking the very high amount of initial sugar concentration but their converpeak in honey with a standard of HMF (Sigma-Aldrich, Milan, Italy), sion into HMF is lower than 2%. So the initial concentration of and comparing the spectra of HMF standard with that of honey samples. The amount of HMF was determined using an external Table 2 --- Chemical characteristics of honeys. I Table 1 --- HMF levels (mg/kg) in fortified honeys. Samples 50 ◦ C Fortified Citrus 1 123.0 Fortified Citrus 2 173.8 Fortified Citrus 3 232.8 Fortified Chestnut 1 73.4(145.6)a Fortified Chestnut 2 94.5 (194.4)a Fortified Chestnut 3 138.9 (223.2)a Fortified Multifloral 1 145.9 Fortified Multifloral 2 207.6 Fortified Multifloral 3 233.5 a 35 ◦ C 25 ◦ C 132.6 158.0 238.4 79.2 (143.2)a 127.3 (198.0)a 190.6 (239.2)a 135.2 193.8 238.4 121.0 157.7 200.7 66.6 (107.5)a 97.2 (166.7)a 104.3 (207.3)a 134.8 180.1 251.0 Measured value, in parentheses added value. C626 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 9, 2008 Moisture (g/100 g) Aw Conductivity (mScm−1 ) Ash (g%) Fructose (g/100 g) Glucose (g/100 g) pH Free acidity (meq/kg) Lactones (meq/kg) Total acidity (meq/kg) HMF (mg/kg) Citrus honey Chestnut honey Multifloral honey 19 0.63 0.23 0.05 36.2 31.5 3.6 13.5 9.5 23 11.8 18 0.62 1.22 0.6 39.2 27.4 6.5 11.3 2.5 13.8 0.83 20 0.65 0.33 0.12 39.5 32.3 3.2 28 7.5 35.5 32.2 sugars in honey did not influence the formation of HMF. To evaluate the degradation rate of HMF in honey, each honey sample was fortified with HMF in very high concentration with respect to the normal levels naturally found in honeys. This method was applied to reduce the effect of HMF formation and to highlight the HMF degradation. For each honey the real HMF level and the recovery were measured (94.6% ± 4.6). While for citrus and multifloral honeys the measured HMF levels (by HPLC) were in conformity with the added ones, for chestnut honeys the measured HMF level was much lower than the HMF added amount (Table 1, in parenthesis). During kinetics trials moisture, a w , conductivity, ash, sugars, free acidity, and lactones values did not change (data not shown). Figure 1 reports HMF level in citrus honeys at the 3 heating temperatures. At 50 ◦ C in the reference sample, the HMF level varied from 11.8 to 31.6 ppm, confirming literature data (Fallico and others 2004). At the same time, both in the reference sample and in fortified samples, there was a slight increase of pH and a decrease of total acidity (data not shown). All fortified citrus samples, showed a decrease in the starting HMF value. The value decrease registered in the lowest fortified sample (fortified 1) showed HMF levels falling from 123 to 102.5 ppm (Figure 1). The same behavior was seen in citrus fortified 2, where, after 8 d of heating, HMF level passed from 173.8 to 144 ppm. In this sample, the higher reduction was measured after the first 2 d of heating (134 ppm). This behavior was even more evident in the highest fortified sample (fortified 3), where there was a very strong HMF decrease, up to 158 ppm after 3 d of heating, showing an increase up to 205.4 ppm during the following days (Figure 1). At 35 ◦ C the increase of HMF in citrus reference samples was not important. In these samples the amount of HMF, after 12 d of heating passed from 11.8 to 13.7 ppm, while the other chemical parameters remained constant. At the same temperature, HMF decreased in fortified citrus samples (Figure 1), above all in the first 3 d of heating. In fact, at lower concentrations, after 12 d of heating, HMF level passed from 132.6 to 106.4 ppm. The same behavior was observed in the fortified 2, where HMF decreased from 158 to 134.6 ppm after 12 d of heating (Figure 1). The sample fortified 3 showed a decrease of HMF, reaching the minimum level of 214 ppm after 12 d of heating. In reference citrus honey stored at the lowest temperature, 25 ◦ C, the HMF level increased from 11.8 to 13.3 ppm after 75 d of storage (Figure 1). In samples fortified 1, HMF reached 101.8 ppm at the end of storage. At the same time, in fortified 2, the HMF level decreased from 157.7 to 140.7 ppm, while in the fortified 3 sample it decreased from 200.7 to 110.3 ppm. The highest degradation rate of HMF was measured, after 15 d of storage, in fortified 3, reaching 124 ppm. Figure 2 shows the changes of HMF level, at the 3 studied temperatures, in chestnut honey. All the other chemical parameters remained constants. In the reference samples, only at the highest temperature there was a measurable increase of HMF (from 0.8 to 10.8 ppm). As for the chestnut fortified samples, the main difference with all the other fortified samples, as discussed previously, was a deep discrepancy between the added and measured amount of HMF (Table 1). The heating treatment at 50 ◦ C caused significant reduction of HMF in all fortified samples. In fact, in fortified 1 it decreased from 73.4 to 39.1 ppm, in fortified 2 from 94.5 to 39 and in fortified 3 from 138.9 to 72 ppm (Figure 2). The decrease of HMF in fortified samples occurred also at lower temperatures. In fact, at 35 ◦ C, the decrease of HMF was less evident in sample fortified 1, where it passed from 79.2 to 64.3 ppm, the decrease was even more evident in fortified 2 and fortified 3 where HMF level passed from 127.3 to 92.7 ppm and from 190.6 to 143.6 ppm, respectively. At the lowest temperature, 25 ◦ C, the decrease of HMF in fortified chestnut honeys was really evident in all samples. It declined from 66.6 to 18.4 ppm after 75 ds of storage in fortified 1, from 97.2 to 29.4 ppm in fortified 2 and from 104.3 to 34.9 ppm in fortified 3, respectively. In all samples, the decrease was even higher after 110 d of storage. In fact, in these samples the HMF level was: 8.2, 14.4, and 17.8 ppm in fortified 1, 2, and 3, respectively (data not shown). Figure 3 shows the behaviors of HMF during heating in multifloral honeys. In the reference sample, heated at 50 ◦ C, the HMF level increased from 32.2 to 69.3 ppm after 8 d. In this case, unlike citrus honey, the pH value, as well as acidity, decreased (data not shown). The amount of HMF in fortified multifloral honeys seemed not to change significantly. In fact, in fortified 1 the HMF level passed from 145.9 to 136.8 ppm, in fortified 2 from 207.6 to 198.5 ppm and in fortified 3 from 233.5 to 221.5 ppm (Figure 3). At 35 ◦ C, in the reference multifloral honey the HMF level remained constant (Figure 3), as well as pH and acidity (data not shown). All multifloral fortified samples showed a decrease of HMF during heating. At this temperature, sample fortified 1, showed a significant HMF reduction during the 1st days of heating (104.2 ppm), then, during the subsequent 7 d of heating it increased to finally fall down reaching a final level of 121 ppm. The HMF level, in fortified 2, changed from 193.8 to 187.8 ppm, showing the lowest value of 161 ppm, after 24 h of heating. The samples with the highest HMF concentration passed from 238.4 to 214.4 ppm after 12 d of heating. The HMF level in multifloral reference sample at room temperature remained constant during the 75 d of storage (Figure 3). But, it showed a small decrease both in pH and in total acidity (data not shown). The fortified samples did not show any significant variation of pH or acidity. In fortified 1, after the 75 d of storage, the initial amount of HMF (134.8 ppm), remained 109.2 ppm. While in fortified 2 HMF decreased from 180.8 to 156 ppm. At the highest concentration, there was the highest reduction of HMF. In fact, after 75 d of storage, it declined from 251 to 79.1 ppm. At this temperature, citrus honey, as well as multifloral samples, showed a gradual and constant decrease in HMF. Data shown above highlight that in reference samples HMF immediately increases, according to literature data (Fallico and others 2004), only during a heating treatment at 50 ◦ C. At lower temperatures, in the studied intervals of time, it remained almost constant. Longer intervals of storage time were necessary to have a significant increase of HMF (Sancho and others 1992; Fallico and others, 2009). On the other hand, data reported that in all fortified samples but one (multifloral fortified 2, 35◦ C Figure 3) there a significant HMF degradation, partially compensated, at 50 ◦ C, by the HMF formation (Figure 1, 2, and 3). The heaviest HMF degradation was measured, on the average, in fortified chestnut samples, and above all in all samples stored at room temperature. Kinetics The HMF degradation occurred through a lot of reactions influenced by the composition of food products (Belitz and others 2004; Girisuta and others 2006a, 2006b). Pseudo-zero order reactions in food systems, does not imply a monomolecular mechanism of reaction, but, always indicate that a complex reaction is occurring, involving a number of steps and it is typical of deteriorative reactions (for example, nonezymic browning) (Robertson 1993). Table 3 reports k and the activation energy values of the studied honeys. The high standard deviation associated to k values, Vol. 73, Nr. 9, 2008—JOURNAL OF FOOD SCIENCE C627 C: Food Chemistry Degradation of HMF in honey . . . Degradation of HMF in honey . . . confirms the difficulties of kinetics in heated systems (Martins and others 2003, 2005). At 50 ◦ C, and at the established time intervals, the k degradation of HMF in citrus honey was almost the double of k formation , 4.40 and 2.63 ppm/d, respectively. In chestnut honey, the k degradation of HMF was 7.43 ppm/d, while the k formation was 1.35 ppm/d. At this temperature, in multifloral honeys, the k-values had almost the same values, 4.6 and 5.2 ppm/d, respectively (Table 3). At 35 ◦ C, the k degradation of HMF were 1.95, 3.25, and 1.35 ppm/d for citrus, chestnut, and multifloral honeys, respectively. At 25 ◦ C, the k degradation of C: Food Chemistry 50°C 250 HMF ppm 200 150 100 50 0 0 2 4 6 8 days of heating Fortified 1 Fortified 2 Reference Sample Fortified 3 35°C 300 250 ppm HMF 200 150 100 50 0 0 2 4 6 8 10 12 days of heating Reference Sample Fortified 1 Fortified 2 Fortified 3 250 25°C ppm HMF 200 150 100 50 0 0 10 20 Reference Sample C628 30 40 50 days of heating Fortified 1 Fortified 2 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 9, 2008 60 70 Fortified 3 Figure 1 --- Evolution of HMF concentration in citrus honeys during kinetics trials. Degradation of HMF in honey . . . 50 ◦ C (Table 3) and 70 ◦ C (Fallico and others 2004) with respect to the k-values of the other types of honey. This means that chestnut honey produces less HMF in comparison to its own degradation. Moreover, citrus honey, with respect to chestnut honey, produces double quantities of HMF and degrades the half (Table 3). 50°C 160 140 C: Food Chemistry HMF were 0.48, 0.7, and 0.78 ppm/d for citrus, chestnut, and multifloral honeys, respectively. At 50 and 35 ◦ C chestnut honey showed the highest k degradation values and at 25 ◦ C the k-value was almost the same of multifloral honey. k formation of HMF in chestnut honey was always lower at Figure 2 --- Evolution of HMF concentration in chestnut honeys during kinetics trials. HMF ppm 120 100 80 60 40 20 0 0 1 2 3 4 5 days of heating Reference Sample Fortified 1 6 7 Fortified 2 8 Fortified 3 35°C 250 ppm HMF 200 150 100 50 0 0 2 4 6 days of heating Reference Sample Fortified 1 8 10 Fortified 2 12 Fortified 3 120 25°C 100 80 ppm HMF 60 40 20 0 0 10 20 30 40 50 60 70 days of heating Reference Sample Fortified 1 Fortified 2 Fortified 3 Vol. 73, Nr. 9, 2008—JOURNAL OF FOOD SCIENCE C629 Degradation of HMF in honey . . . respectively (Table 3). These values are almost a half or even less than HMF E a formation values, for instance, 136.1 KJ/mol for citrus honey (Fallico and others 2004). Moreover, although the reported formation activation energy value in chestnut honey, 182.3 KJ/mol (Fallico and others 2004), was an underestimated value, it is three fold higher than the activation energy of HMF degradation. 50°C 250 ppm HMF 200 150 100 50 0 0 1 2 3 4 5 6 7 8 days of heating Reference Sample Fortified 1 Fortified 2 Fortified 3 35°C 300 250 ppm HMF 200 150 100 50 0 0 2 4 6 days of heating Reference Sample 8 Fortified 1 10 Fortified 2 12 Fortified 3 300 25°C 250 200 ppm HMF C: Food Chemistry Thus, the legislative requirements concerning HMF should take into account this behavior, but up to today legal limits seem too restrictive for some honeys (for example, citrus) and too permissive for others types (for example, chestnut) (Fallico and others 2004, 2006). The activation energies of HMF degradation in citrus, chestnut, and multifloral honeys are: 69.2, 74, and 47.3 KJ/mol, 150 100 50 0 0 10 20 30 40 50 60 70 80 days of heating Reference Sample C630 Fortified 1 Fortified 2 JOURNAL OF FOOD SCIENCE—Vol. 73, Nr. 9, 2008 Fortified 3 Figure 3 --- Evolution of HMF concentration in multifloral honeys during kinetics trials. Degradation of HMF in honey . . . Temperature (◦ C) Honey K Degradation (ppm/d) K Formation (ppm/d) E aDegradation 2.63 (±1.29) 1.35 (±0.19) 5.22 (±1.54) Citrus: 69.2 KJ/mol (16.5 Kcal/mol) 50 Citrus Chestnut Multifloral 4.40 (±0.74) 7.43 (±2.24) 4.62 (±0.71) 35 Citrus Chestnut Multifloral 1.95 (±0.14) 3.25 (±1.19) 1.35 (±0.61) Chestnut: 74.0 KJ/mol (17.7 Kcal/mol) 25 Citrus Chestnut Multifloral 0.48 (±0.28) 0.70 (±0.17) 0.78 (±0.83) Multifloral: 47.3 KJ/mol (11.3 Kcal/mol) Data shown above highlight that the reaction routes leading to HMF formation were kinetically favored, besides the reaction routes leading to its degradation were thermodynamically favored. In fact, in all cases but one, at 50 ◦ C, the k degradation values of multifloral honey (Table 3) were much higher than the k formation values. On the other hand, E a values concerning HMF degradation were significantly lower than formation. This means that an increase of temperature promotes formation routes much more than the HMF degradation pathways. Conclusions T his preliminary study has highlighted for the 1st time HMF degradation kinetics in honey treated at different temperatures. At room temperature HMF degradation kinetics was significant in comparison to the kinetics of formation, thus the evaluation of honey shelf life or ageing with levels of HMF could not be so accurate. Kinetics trials could be extended to more honey samples of the same type and to other unifloral honey samples to verify the behavior of the degradation route and to establish, wherever necessary, new legal limits of HMF in honey. References [ANFS] Australian and New Zealand Food Standard Code. Standard 2.8.2 Honey Issue, 53:1–2. Bath PK, Singh N. 1999. A comparison between helianthus annuus and eucalyptus lanceolatus hoeny. Food Chem 67:389–7. Belitz HD, Grosch W, Schieberle P. 2004. Food chemistry. 3rd ed. Berlin, Germany: Springer Verlag. p 260–3. Canada Agricultural Product Act. Honey Regulations. C.R.C., c. 287 p. Chàvez-Servı̀n JL, Castellote AI, López-Sabater MC. 2006. Evolution of potential and free furfural compounds in milk based infant formula during storage. Food Res Int 39:536–43. CODEX STAN 12-1981 (Rev. 2 (2001)) – Revised Codex Standard for Honey. Adopted by the 24th session of the Codex Alimentarius Commission. Codigo Alimentario Argentino, Capitulo X – Alimentos Azucarados, art. 782 (res. 2256, 16 12. 85). Directive 2001/110/EC of 20 December, Relating to honey. L 10/47. Docket#2006P-0101, Petition for review of Codex standard for honey under 21 CFR 130.6. www.cfsan.fda.gov. Erdoğdu F, Şahmurat F. 2007. Mathematical fundamentals to determine the kinetic constants of first-order consecutive reactions. J Food Process Eng 30:407–20. Fallico B, Zappala M, Arena E, Verzera A. 2004. Effects of conditioning on HMF content in unifloral honeys. Food Chem 85(2):305–13. Fallico B, Arena E, Verzera A, Zappala M. 2006. The European Food Legislation and its impact on honey sector. Accredit Qual Assur 11(1–2):49–54. Fallico B, Arena E, Zappala M. Prediction of honey shelf life. 2009. J Food Qual. Forthcoming. Ferrer E, Alegrı́a A, Farré R, Abellán P, Romero F. 2005. High-performance liquid chromatographic determination of furfural compounds in infant formulas during full shelf-life. Food Chem 89:639–45. Girisuta B, Janssen L, Heeres HJ. 2006a. A kinetic study on the conversion of glucose to levulinic acid. Chem Eng Res Des 84(A5):339–49. Girisuta B, Janssen L, Heeres HJ. 2006b. A kinetic study on the decomposition of 5hydroxymethylfurfural into levulinic acid. Green Chem 8(8):701–9. Hidalgo A, Pompei C. 2000. Hydroxymethylfurfural and furosine reaction kinetics in tomato products. J Agric Food Chem 48(1):78–82. Martins S, Van Boekel M. 2005. A kinetic model for the glucose/glycine Maillard reaction pathways. Food Chem 90(1–2):257–69. Martins S, Marcelis ATM, Van Boekel M. 2003. Kinetic modelling of Amadori N-(1deoxy-D-fructos-1-yl)-glycine degradation pathways. Part I - Reaction mechanism. Carbohydr Res 338(16):1651–63. Morales FJ, Romero C, Jimenez-Perez S. 1997. A kinetic studyon 5hydroxymethylfurfural formation in Spanish UHT milk stored at different temperatures. J Food Sci Tech Mys 34(1):28–32. Msys As. N. 003, 11.01.95. Reglamento Tecnico MERCOSUR de identidad y calidad de la miel (Res. GMC n. 015/95). Persano Oddo L, Piazza MG, Sabatini AG, Accorti M. 1995. Characterization of unifloral honeys. Apidologie 26:453–65. Persano Oddo L, Piazza MG, Pulcini P. 1999. Invertase activity in honey. Apidologie 30:57–66. Piazza MG, Accorti M, Persano Oddo L. 1991. Electrical conductivity, ash, colour & specific rotatory power in italian unifloral honeys. Apicoltura 7:51–63. Robertson GL. 1993. Food packaging: principles and practice. New York: Marcel Dekker, Inc. p 273–79. Sanchez MP, Huidobro JF, Mato I, Muniategui S, Sancho MT. 2001. Evolution of invertase activity in honey over two years. J Agric Food Chem 49:416–22. Sancho MT, Muniategui S, Huidobro JF, Lozano JS. 1992. Aging of honey. J Agric Food Chem 40:134–8. Singh N, Bath PK. 1997. Quality evaluation of different types of Indian honey. Food Chem 58:129–33. Singh N, Bath PK. 1998. Relationship between heating and hydroxymethylfurfural formation in different honey types. J Food Sci Tech 35:154–6. Yaylayan VA. 1997. Classification of the Maillard reaction: a conceptual approach. Trends Food Sci Tech 8:13–8. White JW. 1992. Quality evaluation of honey—role of HMF and diastase assays. 1. Am Bee J 132:737–42. White JW. 1994. The role of HMF and diastase assays in honey quality evaluation. Bee World 75:104–17. Zappalà M, Fallico B, Arena E, Verzera A. 2005. Methods for the determination of HMF in honey: a comparison. Food Control 16(3):273–7. Vol. 73, Nr. 9, 2008—JOURNAL OF FOOD SCIENCE C631 C: Food Chemistry Table 3 --- Kinetics parameters of HMF degradation and formation in honey.