Sustainable Valorization of Whey by Electroactivation Technology for In Situ Isomerization of Lactose into Lactulose: Comparison between Electroactivation and Chemical Processes at Equivalent Solution Alkalinity.

The demand for production of prebiotics at a commercial scale is rising due to the consumers' growing health awareness. Whey, a coproduct of the dairy industries, is a suitable feed medium to produce a prebiotic lactulose through the isomerization of lactose under alkaline conditions. The aim of the present study was to compare the isomerization of lactose into lactulose in situ of whey by using electroactivation technology with the chemical isomerization method using KOH as catalysis under equivalent solution alkalinity. Electroisomerization of lactose into lactulose was performed by using whey solutions of 7, 14, and 21% (w/v) dry matter under current intensities of 300, 600, and 900 mA, respectively, during 60 min with a sampling interval of 5 min. The conventional chemical method was carried out using KOH powder as catalyst at the alkalinity that corresponded to that measured in the electroactivated whey at each 5 min interval. The results showed that lactulose production was dependent on the whey concentration, current intensity, and EA time. The highest lactulose yield of 32% was achieved under a 900 mA current intensity at 60 min for a 7% whey solution. Thereafter, the EA conditions were compared to those of a conventional chemical isomerization process by maintaining similar alkalinity in the feed solutions. However, no lactulose was produced by the chemical process for the equivalent solution alkalinity as in the EA technique. These results were correlated with the solution pH, which reached the required values in a 7% whey solution with values of up to pH 11.50, whereas the maximum pH values that were obtained at higher whey concentrations were around 10-10.50, which was not enough to initiate the lactose isomerization reaction. The outcomes of this study suggest that EA is an efficient technology to produce lactulose using whey lactose.

Inroduction

Whey is a coproduct of cheese or casein production, typically comprising 5–8% (w/w) of dry matter in which 60–80% is represented by lactose and 10–20% by proteins.[1,2] About 9 kg of whey can be generated to make 1 kg of cheese, and it mostly consists of water (93–94%), lactose (4–5%), proteins (0.6–0.8%), and minerals (0.5%).[3,4] The worldwide production of whey was estimated at around (180–190) × 106 tons/year, of which only 50% is being processed into different food and feed derivatives, and about 50% of total production is mainly discarded as dairy effluent with serious environmental concerns because of its high biochemical and chemical oxygen demands.[5,6] Therefore, innovative and sustainable approaches of managing the whey must be addressed for its valorization, more likely, in respect to its significantly high contents of potentially valuable ingredients like lactose.

Lactose is used for producing various value-added derivatives, like lactose, lactitol, lactobionic acid, lactosyl urea, lactosucrose, and galacto-oligosaccharides. Among these, the production of lactulose has received particular interest in recent years due to its proven bifidogenic functionality with many food and pharmaceutical applications.[2] Lactulose production is typically carried out by isomerization of lactose, following either a chemical or an enzymatic method.[7,8] However, recently, the isomerization of lactose into lactulose is successfully performed under autocatalytic conditions using electroactivation (EA) of pure lactose or whey solution by following self-generated alkaline conditions through water electrolysis at the solution/cathode interface.[9,10] Indeed, the hydroxyl (OH–) ions generated by water decomposition at the solution/cathode interface create a high alkaline condition needed for this isomerization reaction. The interference of the alkaline condition in the cathodic compartment and the acidic condition in the anodic compartment is usually avoided by using a suitable reactor configuration, in which a cation-exchange membrane (CEM) is placed between the cathodic and the central compartments and an anion-exchange membrane (AEM) between the anodic and the central compartments (Figure ). Besides lactulose production, EA can be effectively used in the food industry and biotechnology to enhance the antioxidant activity of whey following a formation of Schiff basis known to have a strong antioxidant capacity.[11]

Figure 1

Graphical representation of the electroactivation reactor used for the isomerization of lactose into lactulose in situ of whey.

In the case of whey valorization, EA could be employed to produce lactulose directly in situ of whey by electroisomerization of lactose into lactulose, which is a proven prebiotic. Consequently, a completely new product, lactulose-enriched whey, is produced, which could be used as a high-value-added prebiotic with antioxidant properties.[4,9] Furthermore, when whey is subjected to the EA, the amino groups of whey proteins or peptides interact with carbonyl functions of the reducing sugars existing in the medium to form intermediate Maillard reaction products, which enhance the antioxidant capacity of the final product.[12,13] Nevertheless, more recently, Djouab and Aïder[14] achieved a higher yield of lactulose formation (∼38%) under a 330 mA current intensity during only 14 min of EA using a 5% lactose solution. However, the type of feed solution (lactose/whey/whey permeate) and the concentration of lactose in the feed solution had a significant effect on the lactulose formation.[14,15] The influence of feed composition, especially for whey, would have a substantial impact on lactulose production since it has a broad variation in composition and may exhibit strong buffering capacity.[3] Aider and Gimenez-Vidal[15] observed that the formation of lactulose was different for whey permeate (8.84 ± 0.19%) compared to that of a pure lactose solution (25.47 ± 1.18%) even though similar experimental conditions (200 mA, 60 min, 23.1 ± 1 °C) and similar initial lactose concentration (∼5%, w/v) of feed solutions were used. Besides the types and lactose concentration of the feed solution, the activity of isomerization reaction in the EA can be affected by current intensity, time, temperature, volume of the feed solution, salt type and concentration used as electrolyte, electrode material, interelectrode–membrane distance, configuration, and geometry of the EA reactor.[10,14,16]

Recently, Kareb et al.(9) described the effect of feed solution concentration and volume, temperature, and current intensity on whey lactose isomerization into lactulose. They obtained a maximum yield of ∼35% using 100 mL of a 7% whey solution during 40 min of EA under a 400 mA current intensity at a 10 °C temperature. However, the evolution of alkalinity, temperature, oxidation–reduction potential (ORP), and ion migration phenomena during the electroisomerization of whey has never been explained to date. In EA, the lactulose formation rate would be significantly influenced by the solution alkalinity in the cathodic compartment because an adequate level of alkalinity was needed to achieve the isomerization reaction.[9,14] Furthermore, the electrolyte concentration in the central compartment and the migration of ions toward the cathode can affect the solution alkalinity of the catholyte, which in turn may possibly influence the lactulose formation following lactose isomerization.[17] In addition, the feed solution could reach some metastable state due to the increased reactivity of the electroactivated solution while an external electric field is applied. Thus, the highly reactive solution probably intensifies the isomerization reaction of lactose.[11] This phenomenon could partly be explained by the evolution of ORP during the EA. Therefore, all of these aspects should be taken into consideration to understand the process mechanism of the action involved behind the isomerization of lactose into lactulose in situ of whey by using the EA technique.

Currently, conventional chemical isomerization is used to produce lactulose at the industrial scale following the Lobry de Bruyn–Alberda van Ekenstein (LA) transformation. Isomerization via LA rearrangement requires elevated temperature (50–130 °C) and addition of alkalinizing chemicals as catalysts with different reaction times.[2,18] The efficacy of such isomerization reaction is mostly affected by the pH, temperature, processing time, and concentration of catalysts.[2] Hashemi and Ashtiani[18] observed that the increase in temperature and pH could accelerate the conversion rate and shorten the reaction time to obtain the maximum yield. However, higher temperature or pH possibly led to the rapid degradation of lactose and lactulose into higher byproduct formation such as epilactose, galactose, glucose, or other acidic products. In a recent study, Seo et al.(2) observed that a higher amount of catalysts (0.51% Na2CO3) and temperature (90 °C) were required to achieve a high alkaline condition for efficient lactulose production. However, the yields of chemical isomerization were <30%.[2,18] In contrast to traditional chemical isomerization, the electroisomerization process could be performed without using any alkalinizing chemicals in the medium. In addition, the EA process can be carried out in the mild temperature range; thus, the formation of byproducts could be avoided.[15] Consequently, the purification step could be simplified, which is an economical advantage for industrial applications. Furthermore, the EA process may possibly reduce the energy consumption and use of chemicals, and thus it can be considered as an eco-friendly and sustainable technique to produce lactulose through the isomerization of lactose.[9,14] Regarding this context, it is necessary to compare the efficiency of the EA with a chemical isomerization process under equivalent solution alkalinity. However, no study was found to meet this disparity in the literature; thus, the efficiency of lactulose production must be compared by producing an equivalent alkaline condition in both processes as the solution alkalinity is one of the most critical parameters for the isomerization reaction. This important factor has usually been overlooked in the previous isomerization studies but can be formidable for the development of sustainable and economic processes.

In the present study, several EA process parameters such as solution alkalinity, pH, ion migration, temperature, and ORP during the EA of different whey solutions were studied to explain their impact on the isomerization of lactose contained in the whey into lactulose. In addition, a conventional chemical isomerization was carried out under equivalent solution alkalinity to compare with the EA process.

Results and Discussion

Evolution of Solution pH and Alkalinity

The development of pH in the cathodic compartment was studied during the 60 min of EA under different current intensities of 300, 600, and 900 mA for different whey solutions (7, 14, and 21%, respectively) and is depicted in Figure . It is apparent that the evolution of pH was significantly (p < 0.001) influenced by whey concentration, current intensity, and EA time. The higher pH was obtained for a 7% whey solution at 60 min compared to 14 and 21% whey solutions for all current intensities used. For 7% whey, the pH values were 10.60 ± 0.02, 11.47 ± 0.04, and 11.50 ± 0.06 after 60 min of EA under 300, 600, and 900 mA current intensities, respectively. The maximum pH was observed for a 900 mA current intensity whatever the solution concentrations. The highest pH values obtained for a 900 mA current intensity after 60 min of EA were 11.50 ± 0.06, 10.74 ± 0.04, and 8.90 ± 0.22 for 7, 14, and 21% whey solutions, respectively.

Figure 2

Evolution of pH as a function of EA time for (a) 7%, (b) 14%, and (c) 21% whey solutions under different current intensities.

It can be seen from Figure that the rate of pH progression was higher for greater current intensities for all solution concentrations. This could be attributed to the generation of more OH– ions by intensive water dissociation. In fact, a reduction reaction occurred in the cathodic compartment, which resulted in the production of H2 and OH–[2H2O(l) + 2e– → H2(g) + 2OH–(aq)] by water decomposition. The amount of water electrolysis is, indeed, directly proportional to the electric current applied. Figure a implies that the pH was profoundly increased during the first 30 min for both 600 and 900 mA. This was due to the production of more OH– ions in the medium followed by rigorous water splitting at the start of the reaction to allow the current transfer in the cathode–solution interface. Thereafter, the pH evolution showed a quasi-steady stage because the solution became saturated with enough OH– ions. In contrast to that, different phenomena were observed for 14 and 21% whey solutions, as depicted in Figure b,c. This might be due to the increased buffering capacity of the highly concentrated whey solutions, which was resisting the changes in pH by either absorbing or desorbing the OH– ions. Indeed, the higher concentration of buffering compounds like whey proteins and their respective degradation products, inorganic phosphate, and organic acids present in the highly concentrated whey solutions (14 and 21%) may increase the intensity of the buffering capacity.[20]

The solution alkalinity in the cathodic compartment was evaluated at 5 min intervals of the EA process under different current intensities (300, 600, and 900 mA) for different whey solutions (7, 14, and 21%), as presented in Figure . It was observed that the whey concentration, current intensity, and EA time had a significant impact (p < 0.001) on the solution alkalinity. For a 7% whey solution, the solution alkalinity was linearly rising with time and achieved a maximum of 14.67 and 34.00 mmol/L alkalinities under 300 and 600 mA current intensities, respectively. However, it drastically increased to 35.33 mmol/L during the first 40 min and reached a plateau at 45 min (36.00 mmol/L) for a 900 mA current intensity; thereafter, it gradually decreased to 33.33 mmol/L at 60 min. The difference of the alkalinity for the current intensities was correlated to the concentration of OH– ions formed in the medium. The decrease in the alkalinity after 45 min of EA time could be attributed to the fact that some H+ ions would have migrated to the cathodic compartment from the central compartment and caused acidification of the solution. In fact, H+ and OH– ions might be generated by water decomposition at the CEM interface facing the central compartment once the reaction reached a critical stage, to evade the ion deficiency in the central compartment.[19] Maximum alkalinities of 1.33, 16.00, and 25.33 mmol/L were obtained at 60 min of EA under 300, 600, and 900 mA current intensities, respectively, when a 14% whey solution was used. On the other hand, only 3.33 and 5.33 mmol/L alkalinities were obtained for 600 and 900 mA current intensities, respectively, and no alkalinity was created for 300 mA in the 21% whey solution. This difference of the alkalinity in the higher concentrations of whey could be correlated to the higher buffering capacity of the solution, as well as higher resistance for more concentrated whey solutions (Figure S1). Thus, the electric conductivity of the solutions (14 and 21% whey) was probably less than that of the 7% whey solution. Furthermore, a lesser amount of OH– ions were generated at the cathodic interface due to the higher concentration of whey (greater solid/water ratio).

Figure 3

Evolution of alkalinity during the EA process for (a) 7%, (b) 14%, and (c) 21% whey solutions under different current intensities.

Ion Migration

The concentration of K+ ions in the central compartment was evaluated during 60 min of EA under different current intensities of 300, 600, and 900 mA for different whey solutions (7, 14, and 21%, respectively), as demonstrated in Figure . It can be seen from Figure that the concentration of K+ ions in the central compartment was decreasing with EA time whatever the solution concentrations and current intensities used. The rate of decrease was relatively higher for greater current intensities and higher whey solution concentrations.

Figure 4

Concentration of potassium ions in the central compartment during the 60 min EA time for (a) 7%, (b) 14%, and (c) 21% whey solutions under different current intensities.

Figure a shows that the K+ ions were gradually decreasing during the EA process for a 7% whey solution; however, a quasi-static behavior was observed after 45 min under a 900 mA current intensity. This observation of K+ ion migration from the central compartment can be corroborated with the evolution of pH and alkalinity. The K+ ions were continuously migrating to the cathodic compartment and reacting with the OH– ions in the cathodic compartment to create alkalinity during the first 45 min. After 45 min of EA, the pH and solution alkalinity decreased due to the water splitting at the solution–CEM interface to reimburse the lack of current carriers toward the cathode. In fact, more H+ and OH– ions were produced through this water splitting, and newly generated H+ ions competed for electromigration with K+ ions.[19] The competition for electromigration from the central compartment toward the cathodic one was favorable for the H+ ions because H+ ion has higher electrophoretic mobility than K+ ion in solution. As a result, the K+ cation migration toward the catholyte also decreased at the same time. Likewise, as can be seen from Figure b,c, the K+ ions were also progressively decreasing during the EA process of 14 and 21% whey solutions, and a quasi-steady behavior was observed after 40 min at 600 and 900 mA, respectively. However, unlike the 7% whey solution, the pH and alkalinity were not found to be reduced for the 14 and 21% whey solutions. This difference in comparison with the 7% whey solution can be attributed to the absence of a very weak water decomposition at the solution–CEM interface.

Evolution of Temperature

The temperature increase in the cathodic compartment was studied during the EA for different whey solutions (7, 14, and 21%) under current intensities of 300, 600, and 900 mA, respectively, and is presented in Figure . It appears that the evolution of temperature is mainly dependent on the current intensity and EA time. The higher temperature was achieved for the 7% whey solution for all current intensities, and the ultimate temperatures were 26.23 ± 0.12, 34.57 ± 0.42, and 43.13 ± 0.40 °C at 60 min of EA under 300, 600, and 900 mA current intensities, respectively. The maximum temperature was noticed at a 900 mA current intensity whatever the solution concentrations, and the highest temperatures were 43.13 ± 0.40, 40.63 ± 0.15, and 41.47 ± 1.31 °C for 7, 14, and 21% whey solutions, respectively.

Figure 5

Change in temperature during EA for (a) 7%, (b) 14%, and (c) 21% whey solutions under different current intensities.

The temperature increase during the EA was mainly due to the Joule effect in the electrode, and the produced heat was dispersed in the solution.[17] The initial resistance of the ion-exchange membranes and resistance of the feed solutions could also, to some extent, contribute to the temperature rise.[21] Another possible reason could be the decreased conductivity in the central compartment because the demineralization due to ion migration may have increased the resistance of the system.[19] The rate of temperature increase was higher for greater current intensities used because the heated energy dissipation is proportional to the increase in the electric tension and electric current (Joule’s law).[17] Nevertheless, the statistical analysis of the obtained data did not show any correlation between the temperature rise (from 22 to 42 °C) and the formed lactulose. This can be explained by the fact that this temperature rise was not enough to have any catalyzing effect on lactose isomerization into lactulose.

Evolution of Oxidation–Reduction Potential

The changes of oxidation–reduction potential (ORP) in the cathodic compartment were observed during the EA of different whey solutions (7, 14, and 21%) under the current intensities of 300, 600, and 900 mA, respectively, as presented in Figure . It was observed that the ORP values drastically deceased at the beginning of EA for all solution concentrations and current intensities. The ORP values reached −500 to −600 mV within the first 5 min of EA whatever the current intensities and solution concentrations used. Thereafter, it differently decreased depending on the current intensities and solution concentrations.

Figure 6

Variation of ORP as a function of time during EA for (a) 7%, (b) 14%, and (c) 21% whey solutions under different current intensities.

The negative ORP values of electroactivated solutions in the cathode may probably relate to the training effect of excess electrons and formation of various radicals and ionic species.[22] Similarly, Shironosov and Shironosov[23] explained the fact of highly negative ORPs in the electroactivated solution by the generation of the high-energy resonant water microclusters in the solution due to co-vibrating dipoles of water molecules and charged species near-electrode interfaces. Basically, EA caused a vigorous water splitting, which resulted in the enrichment of the negative-charge concentrations by accumulation of OH– groups in the cathodic compartment.[13] The drastic difference in ORP in the first 5 min of EA may be ascribed to the formation of excessive electrons and generation of other highly active reducers such as OH–, H–, H3O2–, •H, •HO, •O2–, •HO2–, and H2O2– caused by rigorous electrolysis.[15] Thereafter, it showed a quasi-steady fashion because the cathodic compartment might be saturated with the charged species. From the practical point of view, ORP is very important for the applications of the electroactivated whey. Indeed, apart from lactulose, which is a well-known prebiotic, the negative (reductive) ORP is very suitable because it characterizes a medium with reducing properties. This means that bacteria that can be grown in this medium will have oxidative protection against stressing factors.

Assessment of Lactulose Formation during the EA Process

The isomerization of whey lactose into lactulose was studied for different whey solutions (7, 14, and 21%) under 300, 600, and 900 mA current intensities, respectively. It was observed that the solution concentration, current intensity, and EA time had a significant effect (p < 0.001) on the conversion of whey lactose into lactulose. Lactulose was produced only for the 7% whey solution, as presented in Figure . However, no lactulose was noticed for the 14 and 21% whey solutions.

Figure 7

Formation of (a) lactulose and other byproducts like (b) galactose and (c) glucose during the EA for the 7% whey solution under different current intensities.

It can be seen from Figures a and 8 that the lactulose formation was started at 40 min (17.51%) and then gradually increased until the end (32.13% at 60 min) of EA under 900 mA for the 7% whey solution. The pH and alkalinity significantly influenced (p < 0.001) the production of lactulose. Lactulose formation started at 40 min while the pH and alkalinity reached 11.44 ± 0.04 and 35.33 ± 1.15 mmol/L, respectively, and thereafter, the yield increased until the end of EA under a 900 mA current intensity. On the other hand, only 19.58% lactulose was produced at the end of reaction under a 600 mA current intensity while pH and alkalinity attained 11.47 ± 0.04 and 34.00 ± 2.00 mmol/L, respectively. Higher alkalinity is the sinequanon for lactulose formation because the molecular rearrangement of lactose isomerization into lactose needs proton acceptors. The higher current intensities (600 and 900 mA) might produce enough OH– ions by intensive water splitting, which ensured the high alkaline condition in the cathodic compartment of the EA reactor.[9,15] However, no lactulose was detected for a 300 mA current intensity. This might be due to the lack of adequate alkalinity. It is generally believed that a high pH (>10.00) is required for lactose isomerization.[9,14,18] Kareb et al.(9) stated that a solution pH of 11 ± 0.3 is more suitable for lactulose synthesis from whey using EA. In this study, it was observed that the lactulose could not be formed even if the pH reached around 11 in several cases. Thus, it can be argued that not only pH but also the adequate alkalinity of the feed solution is crucial to achieve isomerization of lactose. Moreover, the required alkalinity could be different depending on the type and concentration of the feed solutions.

Figure 8

High-performance liquid chromatography (HPLC) chromatograms of lactulose formation for the 7% whey solution during EA. (a) Initial feed solution, (b) at 60 min under 300 mA (glucose: 1.08%, galactose: 2.77%), (c) at 60 min under 600 mA (lactulose: 19.58%, galactose: 3.76%), and (d) at 60 min under 900 mA (lactulose: 32.13%, galactose: 8.12%).

As it can be seen from Figures b and 9, the formation of galactose significantly increased with running time during lactulose production using the EA process, because some part of the lactose and lactulose was later hydrolyzed into galactose with increased isomerization time. Moreover, galactose formation was intensified by higher current intensity. This could be due to the higher temperature rise, pushing the reaction on the other side pathways due to higher activation energy.[9,18] However, the maximum 8.12% of galactose was formed at 60 min of EA under 900 mA, while the acceptable limit for the commercial lactulose syrup can be up to 16% (according to the United States Pharmacopeia).[24] It is worth noting that only galactose was produced as a byproduct, and no other impurities such as glucose, tagatose, epilactose, etc. were detected in the EA reactor (Figures and 9). Therefore, it can be postulated that the glucose and fructose moieties from lactose and lactulose hydrolysis might be isomerized to galactose under the EA conditions. This observation was well corroborated with those previously reported.[9,14] In contrast to EA, the lactulose production via chemical isomerization was usually followed by a rapid degradation of lactulose into tagatose, epilactose, galactose, and other acidic byproducts such as isosaccharinic acids and formic acids. This is because an isomerization reaction via LA transformation was typically performed with a higher concentration of alkalinizing chemicals at elevated temperature (50–130 °C) combined with a long reaction time.[18,25−27] Thus, the purification steps could be more complex and costly.[14,18,28] However, in this study using EA, the temperature has never exceeded 43 °C and no alkalinizing chemical was required.

Figure 9

HPLC chromatograms of galactose formation as a function of EA time for 7% whey under a 900 mA current intensity: (a) at 45 min (lactulose: 21.59%, galactose: 4.57%), (b) at 50 min (lactulose: 26.24%, galactose: 5.54%), (c) at 55 min (lactulose: 30.47%, galactose: 6.75%), and (d) at 60 min (lactulose: 32.13%, galactose: 8.12%).

Regarding 14 and 21% whey solutions, no lactulose was produced but only galactose (Figure S2). This is probably because of the retarding pH and low alkalinity development in these two feed solutions due to the higher buffering capacity and resistance of the solutions. For 14% whey, the pH reached 10.20 ± 0.10 and 10.74 ± 0.04 at the end of the EA under the current intensities of 600 and 900 mA, respectively, but the solution alkalinity was probably inadequate (16.00 ± 2.00 and 25.33 ± 1.55 mmol/L) for the isomerization reaction to occur. It seems that the findings in the present study are perhaps contrary to what has been reported in the earlier literature by Kareb et al.(9) They reported that the 7 and 14% whey solutions were suitable, whereas the 28% whey solution was less effective for lactulose production using EA. However, the geometrical parameters in their study were quite different from those used in the present study, mainly the volume of the cathodic compartment and the distance between the CEM and the cathode. A volume of 100 mL feed solution was used in their study, whereas it was 350 mL in the present study. In fact, the volume of the feed solution significantly influenced the formation of lactulose. Kareb et al.(9) observed that the lactulose yield decreased from 34.57 ± 0.79 to 7.08 ± 1.07% when the volume was increased from 100 to 300 mL under similar EA conditions (400 mA, 40 min, 10 °C). This is because an increase in the volume of feed solution under a constant current intensity and for the same electrode surface results in lower pH evolution due to the similar amount of OH– ions produced at the solution/cathode interface. Nevertheless, the lactulose yield is directly related to the amount of OH– ions formed in the medium because they act as proton acceptors and ensure the optimum alkalinity for lactose isomerization into lactulose.[9,16] The lower the solution volume, the higher was the alkalinity in the cathodic compartment.

Besides the volume of the cathodic compartment, the distance between the cathode and the CEM could have a significant effect on lactulose formation. In the present study, the distance (7.5 cm) was about 3 times higher than that of Kareb et al.,(9) which they used in their study. In a study, Aït-Aissa and Aïder[16] demonstrated that the interelectrode–membrane distance in the cathodic compartment significantly (p < 0.001) influenced the production of lactulose, as well as galactose using lactose as feed solution. The highest lactulose formation (32.50%) was obtained by using the shortest interelectrode–membrane distance (∼2.5 cm). They observed that lactulose production was decreased when the distance was increased from 2.5 to 5 cm. This is because the internal resistance of the cell, which depends on the electrode surface area and the distance between the electrodes, could be increased if the distance is extended. Obviously, with the higher distance, the moving electrical charges/ions encounter more collisions and, therefore, the resistance might be increased.[16] Consequently, the current intensities used for higher concentrations (14 and 21%) and volume (350 mL) of whey were not adequate to create enough alkalinity for isomerization in the present study. In addition, the concentration of the electrolyte (0.5 M Na2SO4) in the central compartment used by Kareb et al.(9) was higher compared to that of the present study (0.1 M K2SO4), which could be another reason for anomalies in lactulose production. Therefore, it can be deduced that the higher volume, higher distance between the CEM and cathode, and other geometrical factors affected the lactulose production by using EA for 14 and 21% whey as feed solutions.

Conventional Chemical Isomerization Using KOH as Catalyst

The conventional chemical isomerization reactions were performed at an ambient temperature by producing equivalent alkalinity as those obtained in the EA using similar whey concentrations (7, 14, and 21%). The evolutions of pH and ORP were studied during the chemical isomerization and are presented in Figures and 11. A significant difference between chemical isomerization and electroisomerization was observed regarding the pH and ORP evolution. In chemical isomerization, the pH never reached 8 (pH < 8), except for the equivalent alkalinity as for EA under the current intensities of 600 and 900 mA for the 7% whey solution. For equivalent alkalinities of 600 and 900 mA, the maximum pH values of 10.26 ± 0.02 and 10.48 ± 0.01 were achieved after 50 and 40 min, respectively. However, it is noteworthy that no lactulose but only some galactose and glucose (in some cases) was produced by chemical isomerization reactions (Table S1 and Figure S3).

Figure 10

Evolution of pH as a function of reaction time in the chemical isomerization for (a) 7%, (b) 14%, and (c) 21% whey solutions.

Figure 11

Variation of ORP as a function of reaction time in the chemical isomerization process for (a) 7%, (b) 14%, and (c) 21% whey solutions.

In chemical isomerization, a higher temperature and time were typically required to achieve lactulose. Hashemi and Ashtiani[18] reported that a maximum lactulose of 11.8 ± 5.1% could be obtained with a pH of 10 and 50 °C temperature using a 10% lactose solution and 1.0 M NaOH as catalyst; however, a long reaction time of 300 min was required. They observed that by increasing the pH and temperature to 11 and 70 °C, an optimum production of 25.4 ± 0.4% could be achieved in 60 min; then again, the higher pH and temperature led to the degradation of lactose and lactulose into many byproducts like epilactose, glucose, galactose, and other acidic products. In the present study, the chemical isomerization was carried out in the ambient temperature for a short duration, which seems to be a reason why lactulose was not produced. However, in our previous study, it was found that lactulose can be produced without external heating with a high pH (>10.00).[19] It is worth mentioning that all of the above-mentioned reports used lactose as feed solution, whereas whey was used in our study. Thus, it is obvious that higher activation energy was required for the isomerization reaction for whey than for lactose because it contains proteins, peptides, and other components with high buffering capacity. Consequently, higher alkalinity was needed to induce an isomerization reaction. Moreover, a higher amount of catalysts was required to produce an adequate level of alkalinity when whey was used as feed solution. In a recent study, Seo et al.(2) produced a high alkalinity of 48.11 mmol/L in the cheese whey solution by using high amounts of 5.1 g/L or 0.51% Na2CO3 as catalyst to obtain only 3–4% lactulose at 60 °C. However, they achieved an optimum lactulose yield of 29.6% by increasing the temperature up to 90 °C. In another study, Seo et al.(8) used 7.6 g/L or 0.76% (NH4)2CO3 as catalyst to produce an alkalinity of 79.09 mmol/L, using cheese whey and obtained a maximum yield of 29.6% lactulose at 97 °C and 28 min. In contrast, in the present study, only 1.91 and 2.02 g/L KOHs were required to produce the maximum alkalinities of 34 and 36 mmol/L as corresponding to EA under 600 and 900 mA for the 7% whey solution. This could be one of the main reasons why lactulose was not formed in this study using chemical isomerization.

It is important to mention here that the ORPs in the chemical isomerization were reduced only to a value of around +150 to +100 mV within the first 5 min and reached a maximum value of around +50 to −100 mV at the end of reactions depending on different reaction conditions (Figure ). Contrary to chemical isomerization, the ORP values reached at around −500 to −600 mV whatever the solution concentration and current intensity used; thereafter, they achieved a value of −650 to −850 mV depending on the solution concentration and current intensities and remained almost steady during the EA process (Figure ). The highly reduced ORPs in the EA process rendered the whey solutions highly reactive because the electric field triggered the feed solutions to a metastable state.[11,19] Consequently, the isomerization reactions in the EA were probably intensified by a higher internal potential energy of the activated solution. This could be another reason why an isomerization reaction could not occur in the chemical isomerization, although the same alkaline conditions were generated by KOH as catalyst. Besides ORP, other physical and chemical factors such as solution alkalinity, pH, temperature, and ion migration in the EA technique may possibly facilitate the feed solution to achieve such conditions, in which the activation energy required for the isomerization of whey lactose into lactulose would significantly be reduced. Thus, in contrast to the conventional chemical method, the EA technique offers a high potential to produce lactulose by using whey as a lactose source. Moreover, the difference between the ORP values of the chemical method versus the EA method can be explained as follows: during the electroactivation (EA) process, water electrolysis at the cathode interface generates two main components: OH-, which is responsible for the medium alkalinization, and hydrogen (H2) gas, which is a strong reducing agent. Thus, the ORP in the EA process was highly reduced. In the case of the chemical method, the addition of KOH had only an effect on solution pH. Thus, even at equivalent solution alkalinity, the ORP in the EA process was very much higher than that of the chemically alkalinized whey solutions.

Conclusions

To sum up, the EA process could be an environmentally friendly and sustainable method for producing prebiotic lactulose through the isomerization of whey lactose because it can be performed under complete autocatalytic conditions, meaning that alkalinizing chemicals and external heating are not required. In this present study, a maximum lactulose yield of ∼32% was achieved under a 900 mA current intensity at 60 min of EA for the 7% whey solution. The results suggest that the formation of lactulose was dependent on the whey concentration, current intensity, and EA time. Furthermore, no lactulose was produced in the chemical isomerization although equivalent alkalinity was created as in the EA. Thus, it is obvious that other process mechanisms of action were involved with the EA technique to achieve the required alkaline conditions. However, despite some meaningful achievements of EA, an ideal optimized condition must be developed that is more economical in terms of the geometry and configuration of the reactor and the type and concentration of the feed solutions. Moreover, the electrical conductivity of the used whey was 5.27 ± 0.12 mS/cm, which is enough to allow the passage of the electric current in the EA reactor. This means that whey can be used in all of the compartments of the EA reactor instead of electrolytic solutions. However, because whey contains some chlorine-containing salts, supplementary protecting conditions must be considered to avoid chlorine accumulation in the working environment. This can be ensured by connecting the EA reactor to an adequate ventilation system.

Materials and Methods

Chemicals and Reagents

The high-performance liquid chromatography (HPLC)-grade sugars such as lactose, lactulose, galactose, glucose, and fructose were obtained from Sigma-Aldrich (Ottawa, Ontario, Canada), whereas, the high-purity (purity ≥95%) chemicals and reagents of analytical grade were procured from different suppliers. The potassium hydroxide (KOH), potassium sulfate (K2SO4), hydrochloric acid (HCl), and phenolphthalein (C20H14O4) were purchased from Fisher Chemical (Fair Lawn, NJ), Sigma-Aldrich Co. (St. Louis, MO), Fisher Chemical (Geel, Belgium), and MAT Laboratory Inc. (Laboratoire MAT Inc., Quebec, Canada), respectively. The food-grade whey powder (lactose, 75%; total proteins, 12%; ash, 7%; and moisture content, <5%) was obtained from Agropur Co-operative (St-Hubert, Quebec, Canada). The cation-exchange membrane (CMI 7000S) and anion-exchange membrane (AMI 7001S) were bought from Membrane International Inc. (Ringwood, NJ) and were directly used in the EA reactor without any pretreatment.

Electroactivation Protocol

An EA reactor made of Plexiglas, comprising three compartments (anodic, central, and cathodic compartments), was used in this study (Figure ). The dimensions of the cathodic compartment, in which the EA process for lactose isomerization into lactulose was targeted, have the following geometrical dimensions: 6.5 (L) × 5.5 (W) × 10 (D) cm3 for a total volume of 357 cm3. The anodic and central compartments are similar to the following geometrical dimensions: 5.5 (L) × 1.8 (W) × 10 (D) cm3. The anodic compartment was connected to the positive side of a DC-regulated power generator (model: CSI12001X, CircuitSpecialists.com) by a titanium electrode coated with ruthenium–iridium, whereas the cathodic compartment was linked to the negative side through a food-grade stainless steel electrode. However, the cathodic and anodic compartments were separated by the central compartment and were communicating with the cathodic and anodic compartments via a CEM and an AEM, respectively. A freshly prepared whey solution (350 mL) of different concentrations (7, 14, and 21%) was placed in the cathodic compartment, while the anodic and central compartments were filled with 0.1 M K2SO4 solution (Figure ). The selected whey concentrations are based on the following considerations: the 7% concentration was selected because it corresponds to the whey that is generally obtained as a coproduct of the cheese-making industry. Thus, the EA process can be applied to whey directly after it is generated without further concentrating step. The 21% concentration was chosen on the basis of whey solubility in water so as to avoid any precipitation. The 14% concentration was selected as an intermediate between the whey as obtained following a cheese-making process and a concentrated whey without any precipitation. The potassium sulfate was used as an electrolyte in the anodic side of the EA reactor to avoid chlorine generation following the oxidation reaction that takes place at the anode surface. The experiments were carried out under three current intensities of 300, 600, and 900 mA for a 60 min duration. The samples were obtained from the central and cathodic compartments at regular intervals of 5 min and were stored at 4 °C until further analysis. All experiments were operated at an ambient temperature (22 ± 2 °C). The EA reactor was properly cleaned with DI water prior to each experiment and remained filled with DI water after each batch to maintain high membrane hydration.

Figure 12

Schematic representation and geometrical parameters of the electroactivation reactor used for lactose electroisomerization into lactulose in situ of whey.

Determination of pH, Alkalinity, Temperature, and Oxidation–Reduction Potential

The pH, alkalinity, temperature, and ORP of whey solution in the cathodic compartment were measured at 5 min intervals during the 60 min EA process. The pH was evaluated using a digital pH meter (Oakton pH 700) outfitted with a pH probe (Oakton, Vernon Hills, IL). The temperature and ORP were assessed by using an ORP meter (Ultrapen, Myron L Company, Carlsbad, CA). A standard titration method was used to determine the total alkalinity of the electroactivated whey solutions (catholytes), as described by Karim and Aider.[19] Finally, the total alkalinity was calculated using eq and was expressed in mmol/LVtitrant is the total volume of the titrant (0.1 M HCl) needed for titration in mL, Ctitrant is the concentration of the titrant in mol/L, Vsample is the volume of sample in mL, and total alkalinityeq is the equivalent concentration of NaOH/KOH in the electroactivated solutions in mmol/L (equiv).

Evaluation of Potassium Concentration

The atomic absorption spectrometry was used to evaluate the concentration of potassium (K+) ions in the central compartment. The samples (that were collected from the central compartment) were analyzed following a standard protocol for the atomic absorption spectrometer (PerkinElmer Instruments, model AAnalysts 200, Boston, MA).

Chemical Isomerization of Lactose in Whey

The conventional chemical isomerization was performed by adding the equivalent (to the total alkalinity in the EA) amounts of potassium hydroxide (KOH) to the feed solutions (7, 14, and 21% whey). In brief, the total alkalinity (mmol/L) obtained in the EA–whey solutions was converted to the equivalent amounts of KOH in mg/L. Thereafter, the equivalent amounts of KOH were added to the feed solution at regular intervals of 5 min to maintain a similar alkaline condition as in the EA. The mixture was continuously stirred at an ambient temperature (22 ± 2 °C). The pH and ORP of the medium were measured at 5 min intervals during the reaction. Finally, the samples were obtained from the reaction medium at regular intervals of 5 min and were preserved at 4 °C for further analysis.

Evaluation of Sugar Composition

A HPLC system (Water, Millipore Corp., Milford, MA) was used to determine the sugar contents in the samples (that were collected from the cathodic compartment and reaction medium of whey solutions during the EA and chemical isomerization, respectively). The HPLC system was equipped with a carbohydrate analysis column (Waters Sugar Pak-I, 300 × 6.5 mm2, Waters Co.) and a refractive index detector (Hitachi, model: L-7490). The column temperature was set at 90 °C, and an isocratic mobile phase (a solution of 50 mg/L Ca-ethylenediamine tetraacetic acid) was used at a flow rate of 0.5 mL/min. The analysis was then performed by injecting 50 μL of sample and setting the operating time to 30 min/sample. Finally, the identification and quantification of different sugars (lactose, lactulose, glucose, galactose, and fructose) were achieved by matching their retention times with the standard solutions.

Statistical Analysis

A complete randomized factorial design with repeated measurements was used for statistical analysis. The factors for the analysis were whey concentration, current intensity, and reaction time. The pH, alkalinity, K+ ion migration, temperature, ORP, lactulose yield, and the yield of byproducts (galactose, glucose, and fructose) were considered as dependent variables. Each experiment was performed in triplicate, and the mean values ± standard deviation was used for the analysis. Differences at p < 0.05 were considered as significant. Analysis of variance (ANOVA) of the data was carried out using SAS software (V9.3, SAS Institution Inc., Car, NC).