Removal of lactobionic acid by electrodialysis

SEPARATION PROCESSES

Removal of lactobionic acid by electrodialysis

J. B. Severo Júnior^*; T. L. M. Alves; H. C. Ferraz

Programa de Engenharia Química/COPPE, Universidade Federal do Rio de Janeiro, Cidade Universitária, CP 68502, 21941-972, Rio de Janeiro - RJ, Brasil. Phone: (55) (21) 25628343, Fax: (55) (21) 25628300. E-mail: [e-mail]

ABSTRACT

Lactobionic acid has a number of applications, such as in cosmetic formulations and detergents, as well as in the medical field, where it is used for the preservation of organs destined for transplantation. Previous studies have reported that a promising alternative procedure for the production of lactobionic acid is the biotechnological route, using permeabilized cells of Zymomonas mobilis to produce sorbitol and lactobionic acid from fructose and lactose. However, the acid produced during the process accumulates in the reaction medium, causing enzyme deactivation. It was found that this problem can be avoided by coupling an electrodialysis unit to the reaction vessel, resulting in efficient removal of the acid from the reaction medium and improved the stability of the enzyme. These tests employed a synthetic mixture containing lactobionic acid, sorbitol, lactose, and fructose, and a factorial design was performed to identify the most influential variables. The NaCl concentration in the concentrate stream, together with the potential difference, exerted the greatest effects on the rate of removal of lactobionic acid. In all experiments, the removal efficiency exceeded 95%. The best conditions for the system investigated were a potential of 60 V, and NaCl concentrations of 3 and 25 g L^-1 in the concentrate stream and the electrode compartment, respectively.

Keywords: Electrodialysis; Zymomonas mobilis; Lactobionic acid; Experimental design.

INTRODUCTION

The use of lactobionic acid in cosmetic formulations is already established. Another potential application based on its biodegradability and lack of toxicity is the use in the manufacture of substances with surfactant properties, such as detergents. Nevertheless, the greatest commercial application is in the medical field, as the main constituent of fluid for organ preservation during the procedure of transplantation. Lactobionic acid is obtained industrially by dehydrogenation of the lactose using a metallic catalyst (Splechtna et al., 2001; Dhariwal et al., 2006; Paul and Patrick, 2009; Pedruzzi et al., 2011; Severo Júnior et al., 2011; Malvessi et al., 2013).

Recent studies have shown some alternatives for the production of lactobionic acid, as for example, biotechnological and electrochemical processes, both still in development (Miyamoto et al., 2000; Splechtna et al., 2001; Dhariwal et al., 2006; Severo Júnior et al., 2011; Malvessi et al., 2013). According to Jonas and Silveira (2004), a promising route for production of lactobionic acid is from lactose and fructose by enzymatic catalysis using permeabilized cells of Zymomonas mobilis. The enzyme glucose fructose oxidoreductase (GFOR) produced only by the bacterium Zymomonas mobilis is capable of promoting simultaneously the oxidation of aldoses to their respective aldonic acids and the reduction of ketoses to their respective polyols (Zachariou and Scopes, 1986).

Natural substrates of GFOR are fructose and glucose. However, the aldonic acid resulting from this catalysis is gluconic acid, which is a low market value product. Lactobionic acid, on the other hand, is much more valuable, rendering its production by Zymomonas mobilis very attractive (Jonas and Silveira, 2004).

Just as important as the synthesis step is the downstream processing, that can have an enormous impact on the product final price. Conventional separation techniques include precipitation, ion exchange, among others. A more recent process is electrodialysis, which employs an electrical gradient to promote the separation of ionic species. In addition to accomplishing separation without adding chemicals to the medium, electrodialysis allows the continuous separation of the ionic product, which can contribute to increased conversion and avoid enzyme inactivation by accumulation of product in the reaction medium (Furlinger et al., 1998; Ferraz et al., 2001).

Ferraz et al. (2001) evaluated the production and simultaneous separation of sorbitol and gluconic acid by an electrodialysis unit coupled to a membrane bioreactor containing permeabilized and immobilized cells of Zymomonas mobilis. The results presented by these authors showed that the electrodialysis unit coupled to the bioreactor allowed an efficient removal of gluconic acid from the reaction medium, as well as an improvement in the stability of the enzyme, without reduction of the reaction rate, even after 60 hours of reaction. In addition, several studies reported in the literature show that this process offers considerable economic potential for the recovery and production of organic acids including citric acid, fumaric acid, and galacturonic acid (Novalic et al., 2000; Choi et al., 2002; Tongwen and Weihua, 2002; Bélafi-Bakó et al., 2004; Mólnar et al., 2009; Mólnar et al., 2010).

Thus, this work aims at evaluating the removal of lactobionic acid from synthetic mixtures containing sorbitol, lactose and fructose by electrodialysis, based on factorial design, to verify the feasibility of the integration of this step with a bioreactor containing permeabilized cells of Zymomonas mobilis.

MATERIALS AND METHODS

The electrodialysis unit comprises 9 compartments of acrylic, separated by anionic and cationic exchange membranes alternately, with a total effective area of 226.2 cm², as shown in Figure 1; this unit was constructed in-house. The electrodialysis unit is composed of 4 pairs of cells, with 4 cationic exchange membranes (CR 67 HMR-412, Ionics) and 4 anionic exchange membranes (AR 204 SZRA-412, Ionics), placed alternately and 7 mm apart from each other.

According to Figure 1, the feed (1) containing the synthetic solution of fructose, lactose, sorbitol and lactobionic acid (total volume of 150 ml) was pumped into channels 5, 7 and 9. Because fructose, lactose and sorbitol have no electric charge, they are not attracted by the electrodes, coming out in the dilute stream (2), while the lactobionic ion and the hydrogen ion permeate through the anion exchange membrane and through the cation exchange membrane, respectively, towards compartments 4, 6, 8 and 10, coming out in the concentrated stream (12) and being collected in the tank (15). This separation occurs due to the electrical potential difference applied to the electrodialysis unit by carbon electrodes (13 and 14) using a continuous current power supply (16) from Instrutherm, model FA-3050.

To evaluate the removal of lactobionic acid, a factorial design was employed having as independent variables the potential difference (or voltage), the NaCl concentration in the electrode compartment and in the concentrated stream. Table 1 shows the levels of the variables used. The voltage ranged from 20 V to 60 V. The use of voltages lower than 20 V is not efficient to remove the acid in this system. The limits of the others variables were chosen based on results obtained in preliminary tests.

A factorial design type 2³ with 3 central points and 6 axial points was chosen, showed in Table 2, and the dependent variable was the specific removal rate of the lactobionic acid (g.min^-1.m^-2). Equations (1) to (3) represent the normalized variables, (voltage), (NaCl concentration in the electrodes compartment), (NaCl concentration in the concentrated stream).

All experiments were carried out at 30°C and at a flow rate of 20 l.h^-1 using 150 ml of synthetic solutions of lactose (40 g L^-1), fructose (20 g L^-1), sorbitol (5 g L^-1) and lactobionic acid (10 g L^-1) (Sigma Aldrich, 99%, all the reagents). The lactobionic acid concentration was monitored indirectly by the feed conductivity (Quimis conductivimeter).

RESULTS AND DISCUSSION

Figures 2 to 5 show the removal curves of lactobionic acid as a function of the time for all experiments performed. Table 2 shows the values of the specific removal rate of lactobionic acid, calculated from the derivative of the curves at 90% removal. The observed removal rates varied from 0.58 to 1.48 g_acid.min^-1.m^-2, and in all experiments it was possible to remove more than 95% of lactobionic acid in less than 2 hours. Figures 2 and 3 show experiments 1-4 and 5-8, respectively. In each set of experiments, the NaCl concentration in the concentrate stream was the same. The voltage and the NaCl concentration in the electrode compartment were changed. In both figures it can be observed that, as the voltage increases, the removal rate increases as well, indicating that the NaCl concentration in the electrode compartment does not exert a significant influence on the removal of lactobionic acid. Figure 4 shows the removal curves for the replicates at the central point, experiments 9-11, pointing out the good reproducibility of the data.

Figure 5

Based on the results of the experimental design and using Equation (4), it is possible to analyze the effect of each variable on the removal rate of lactobionic acid, as well as to verify the significance of these effects. According to the theory of factorial design, this equation means that the effects can be represented by a sum of linear contributions. Model parameters and parameter variances can be obtained with the aid of maximum likelihood estimation procedures, as described in the literature (Schwaab and Pinto, 2007).

The confidence regions of the parameter estimates can be written in the form:

where represents the estimated parameter value; represents the "true" (and unknown) parameter value and represents the parameter uncertainty, obtained with the help of the standard t-distribution, in the form:

with the level specified by the user. is the standard deviation of the parameter estimate. Thus, for a parameter to be significant, should be smaller than (Schwaab and Pinto, 2007). Taking this into account, Equation (7) was obtained after a parameter estimation procedure, performed with the software Statistica 6.0 (Statsoft, 2001).

Equation (7) shows that the significant parameters with 95% confidence were the voltage and the NaCl concentration in the concentrated stream, while the NaCl concentration in the electrodes compartment did not exert an influence on the specific removal rate of lactobionic acid, within the experimental range used.

Table 3 shows the statistical analysis of the model performance, where:

In these equations, and are the predicted and experimental values, respectively; and are the predicted and experimental average values, respectively; NE and NP are the number of experiments and the number of model parameters, respectively; and are the experimental and model variances, respectively; R is the correlation coefficient; F_calculated is the value of the calculated F.

According to Table 3, the F Test employed between the model and the experimental variances, resulting in F_calculated is within the tabulated interval, meaning that the prediction errors of the model are similar to the experimental errors, as can be observed in Figure 6. It is important to note that an assumption of the factorial design is that the error is constant for all the experimental range. This figure shows that the predicted values are within the experimental errors, represented by the average value of the central point, with its confidence interval, calculated by the t Test with 95% confidence.

Equation (7) also shows that the voltage is the most influential variable, being a linear and positive parameter. In other words, as the voltage increases so does the removal rate. Besides, there is a combination effect between the voltage and the salt concentration in the concentrate stream, indicating the existence of non-linearity in the investigated system. According to Equation (7), the effect of the NaCl concentration in the concentrated stream goes through a maximum. With support of the software Mathcad, an optimization routine (based on the Levenberg-Marquard algorithm) was used to calculate the values of the variables that maximize the specific removal rate in the studied range. Thus, the condition of 60 V and 3.0 g L^-1 for the NaCl concentration in the concentrated stream is the best condition found, as it is evident from Figure 7.

Figure 8 shows the apparent resistance of the electrodialysis unit calculated for experiments 12 and 13, in order to evaluate the voltage effect. Experiments 16 and 17 show the influence of the salt concentration in the concentrated stream. The apparent resistance of the electrodialysis unit was calculated based on Equation (12), Ohm's law, as follows:

where V and i are the applied tension and the electrical current of the system, respectively.

It is possible to verify in Figure 8 that, for experiments 12 and 13, as the voltage increased from 20 V to 60 V, the electrical resistance of the system decreased because the electrical current increased. The effect of the salt concentration in the concentrated stream is evident in Figure 8, because at the very beginning of the experiment (time = 0), the apparent resistance of the unit without addition of salt is about 500 Ohms, while with 3 g L^-1 salt solution it is 60 Ohms; since the ions in the solution increased the conductivity in the channels between the membranes, thus the transport resistance decreased.

For all experiments of the factorial design, a removal of lactobionic acid superior to 95% was obtained in less than 100 min. In a similar work (Peretti et al., 2009), a removal of 38.7% of lactobionic acid from synthetic mixtures was obtained in about 250 min and at 15 V.

The results presented in this work show that this membrane separation process can be successfully integrated into the step of production of sorbitol and lactobionic acid by permeabilized cells of Zymomonas mobilis from lactose and fructose, with a possible positive effect of avoiding the enzyme inhibition (Furlinger et al., 1998; Ferraz et al., 2001). For a lactobionic acid production rate of 1 g_acid.g_cell^-1.h^-1 (the highest obtained in our previous studies), using 1 g of cells, a membrane area inferior to 100 cm² would be needed.

CONCLUSIONS

The results of the proposed experimental design showed that the variables with the greatest influence on the lactobionic acid removal rate by electrodialysis are the NaCl concentration in the concentrated stream and the voltage. The increase in the value of these variables results in an increase of the driving force, thus facilitating the removal. The best conditions for the separation of the acid within the studied range were 60 V, 3 g L^-1 of NaCl in the concentrated stream and 25 g L^-1 of NaCl in the electrode compartment.

In all experiments it was possible to remove more than 95% of lactobionic acid in less than 2 hours. The increase of the number of cell pairs of the electrodialysis stack, and thus in the membrane area, is easily accomplished due to its modular design, allowing the prompt scale-up of this process. This work opens new perspectives for the industrial production of sorbitol and lactobionic acid, which is currently still based on conventional processes.

NOMENCLATURE

Latin Letters

Greek Letters

Submitted: February 11, 2013

Revised: September 16, 2013

Accepted: December 2, 2013

• Bélafi-Bakó, K., Nemestóthy, N., Gubicza, L., Study on application of membrane techniques in bioconversion of fumaric acid to L-malic acid. Desalination, 162, 301 (2004).
• Choi, J-H., Kim, S-H., Moon, S-H., Recovery of lactic acid from sodium lactate by ion substitution using ion-exchange membrane. Separation and Purification Technology, 28, 69 (2002).
• Dhariwal, A., Mavrov, V., Schroeder, I., Production of lactobionic acid with process integrated electrochemical enzyme regeneration and optimisation of process variables using response surface methods (RSM). Journal of Molecular Catalysis B: Enzymatic, 42, 64 (2006).
• Ferraz, H. C., Alves, T. L. M., Borges, C. P., Coupling of an electrodialysis unit to a hollow fiber bioreactor for separation of gluconic acid from sorbitol produced by Zymomonas mobilis permeabilized cells. Journal of Membrane Science, 191, 43 (2001).
• Furlinger, M., Haltrich, D., Kulbe, K. D., Nidetzky, B., A multistep process is responsible for product-induced inactivation of glucose-fructose oxidoreductase from Zymomonas mobilis European Journal of Biochemistry, 251, 955 (1998).
• Jonas, R., Silveira, M. M., Sorbitol can be produced not only chemically but also biotechnologically. Applied Biochemistry and Biotechnology, 118, 321 (2004).
• Malvessi, E., Carra, S., Pasquali, F. C., Kern, D. B., Silveira, M. M., Ayub, M. A. Z., Production of organic acids by periplasmic enzymes present in free and immobilized cells of Zymomonas mobilis Journal of Industrial Microbiology Biotechnology, 40, 1 (2013).
• Miyamoto, Y., Ooi, T., Kinoshita, S., Production of lactobionic acid from whey by Pseudomonas sp LS13-1. Biotechnology Letters, 22, 427 (2000).
• Molnár, E., Eszterle, M., Kiss, K., Nemestóthy, N., Fekete, J., Bélafi-Bakó, K., Utilization of electrodialysis for galacturonic acid recovery. Desalination, 241, 81 (2009).
• Molnár, E., Nemestóthy, N., Bélafi-Bakó, K., Utilisation of bipolar electrodialysis for recovery of galacturonic acid. Desalination, 250, 1128-1131 (2010).
• Novalic, S., Kongbangrend, T., Kulbe, K. D., Recovery of organic acids with high molecular weight using a combined electrodialytic process. Journal of Membrane Science, 166, 99 (2000).
• Paul, L. H. M., Patrick, F. F., Advanced Dairy Chemistry: Lactose, Water, Salts and Minor Constituents. Springer, New York, v. 3 (2009).
• Pedruzzi, I., Silva, E. A. B., Rodrigues, A. E., Production of lactobionic acid and sorbitol from lactose/fructose substrate using GFOR/GL enzymes from Zymomonas mobilis cells: A kinetic study. Enzyme and Microbial Technology, 49, 183 (2011).
• Peretti, F. A., Silveira, M. M., Zeni, M., Use of electrodialysis technique for the separation of lactobionic acid produced by Zymomonas mobilis Desalination, 245, 626 (2009).
• Schwaab, M., Pinto, J. C., Análise de dados experimentais I - fundamentos de estatística e estimação de parâmetros. E-papers, Brazil, v. 1 (2007). (In portuguese).
• Severo Júnior, J. B., Pinto, J. C., Ferraz, H. C., Alves, T. L. M., Production of lactobionic acid and sorbitol using the GFOR (glucose-fructose oxidoreductase) enzyme from permeabilized cells of Zymomonas mobilis Journal of Industrial Microbiology Biotechnology, 38, 1575 (2011).
• Splechtna, S., Petzelbauer, I., Bamingerm, U., Haltrich, D., Kulbe, K. D., Nidetzky, B., Production of a lactose-free galacto-oligosaccharide mixture by using selective enzymatic oxidation of lactose into lactobionic acid. Enzyme and Microbial Technology, 29, 434 (2001).
• StatSoft, Inc., STATISTICA (Data Analysis Software System), version 6. www.statsoft.com (2001).
• Tongwen, X., Weihua, Y., Citric acid production by electrodialysis with bipolar membranes. Chemical Engineering and Processing, 41, 519 (2002)
• Zachariou, M., Scopes, R. K., Glucose-fructose oxidoreductase, a new enzyme isolated from Zymomonas mobilis that is responsible for sorbitol production. Journal of Bacteriology, 167, 863 (1986).

*

To whom correspondence should be addressed

Publication Dates

• Publication in this collection
  14 Nov 2014
• Date of issue
  Dec 2014

History

• Received
  11 Feb 2013
• Accepted
  02 Dec 2013
• Reviewed
  16 Sept 2013