Encapsulation of vanilin in agar beds for controlled release.

Chirilus, Alina Alexandra ; Stoica Guzun, Anicuta ; Stroescu, Marta 等

1. INTRODUCTION

Controlled release is a technology that can be used to increase the effectiveness of many ingredients. Despite the wide range of encapsulated products that have been developed, micro-encapsulation is still far from being fully developed in the food industry (Pothakamury et al., 1995). The active agent is released from the controlled release delivery systems by diffusion, biodegradation, swelling or osmotic pressure.

Two main types of microcapsules exist: reservoir systems and matrix systems. A reservoir system consists of an active agent contained within a rate-controlling barrier. The release rate from a reservoir system depends on the thickness, the area and the permeability of the barrier. In matrix systems, the active agent is homogeneously dissolved or dispersed throughout the polymer mass. The release pattern depends on the geometry of the system, the type of carrier material and the loading of the active agent (Arifin et al., 2006).

Agar is a gelatinous substance derived from red algae. Agar consists of a mixture of agarose and agaropectin. Vanilin is the major component of natural vanilla, which is one of the most widely used and important flavouring substances worldwide. Vanilin displays antioxidant and antimicrobial properties and has the potential for use as a food preservative (Walton et al., 2003).

The aim of this paper is to present experimental data concerning vanilin release from agar capsules.

2. MATERIALS AND METHODS

2.1 Preparation of encapsulated vanilin

The agar granules were obtained starting from a solution of agar in hot water. The solution was prepared by mechanical stirring. Another solution of synthetic vanilin of known concentration was poured in the solution. The mixture was poured with a syringe as droplets which were solidified by cooling in sunflower oil under vigorous stirring for ten minutes. The resultant agar particles have a particle size ranging between 2.5-7 mm. Their diameters were measured using a BX 51 Olympus microscope.

2.2 Analysis of vanilin in agar particles

A certain quantity of agar granules was treated with distillated water in the same ratio solid/liquid as in the release experiments in an ultrasonic bath for 10 minutes. The granules were completely disintegrated in order to release the entire vanilin content.

The mixture was then filtrated and the liquid was analyzed for the vanilin content using UV-VIS spectroscopy.

2.3 In vitro release study

In vitro release was performed in a vessel under magnetic agitation. A certain quantity of agar particles having the same diameter was put in contact with a known volume of alcoholic solution (80% vol/vol alcohol). The vanilin content was analyzed in the liquid phase using UV-VIS spectroscopy.

2.4. Treatment of data from release studies

The release data were analyzed by applying several release models proposed in literature.

Active agent released from the studied systems can be diffusion-controlled, swelling-controlled and erosion-controlled. Combinations between diffusion, swelling and erosion can be encountered for one particular system.

For diffusion-controlled micro-spheres, drug release profile is obtained by solving Fick's second law of diffusion subject to appropriate boundary conditions.

The simplest way to model the release from a matrix system is to consider an infinite, flat matrix system of radius R and with the active agent dissolved in the polymer at or below the saturation concentration. The amount of active agent released when the device is dissolved in a well-agitated, infinite medium is given by:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

For Sh>1, Baker and Lonsdale (Baker & Lonsdale, 1974) have proposed two simplified solutions by using early time and late-time approximations to explain the drug release from dissolved matrix system of a sphere during early and late-time periods respectively. The approximation results are expressed as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2a)

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (2b)

The polymer used as matrix may also have swelling properties. For swelling-controlled system, the hydrophilic polymer is susceptible to swelling, as water tends to penetrate and relax the polymer matrix. In this case, the composition of the hydrophilic polymer will determine the extent of swelling.

In swelling controlled systems, since both diffusion and dissolution occur altogether, they are quite indistinguishable. A very simple equation proposed to describe the release from these systems is:

[M.sub.t] / [M.sub.[infinity]] = k x [t.sup.n] (3)

The value of n indicates the type of active-agent transport. When n = 0.5, the active agent is released by simple Fickian diffusion. When n = 1.0, diffusion is described as 'case II diffusion'. In case II diffusion, the rate of the solvent uptaken by the polymer is largely determined by the rate of swelling and relaxation of the polymer chain. 'Super case II transport' occurs when n > 1.0. In the case 0.5 < n < 1, diffusion is a combination between Fickian and non-Fickian diffusion, and is known as anomalous diffusion.

The release of an active agent from a matrix type delivery system may be controlled also by erosion or by a combination between diffusion and erosion. Erosion controlled processes may be heterogeneous or homogeneous and depend on the hydrophobicity and morphology of the polymer. Heterogeneous erosion is more common with hydrophobic polymers, whereas homogeneous erosion is common with hydrophilic polymers (Arifin et al., 2006). The amount of active agent released in heterogeneous erosion is given by:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (4)

The values of n are different, depending on the matrix geometry: n = 3 for a spherical one, n = 2 for a cylindrical matrix and n = 1 for a slab shaped matrix.

3. RESULTS AND DISCUSSIONS

In order to understand the release kinetics the particles were selected with respect to their diameters. Figure 1 shows in vitro release profile of vanilin in alcoholic solutions for practically the same vanilin content in the particles 2.2 x [10.sup.-4] g vanilin/g capsules.

For all experiments, a burst effect is observed for the first 5-10 minutes from the beginning of the experiments. For the calculation of the proposed models parameters, only vanilin release after 5 or 10 minutes was considered, because in the first period of release a "non-steady state" behaviour occurs.

In their article (Chattopadhyaya et al., 1998), the authors suggested that there exists an important quantity of surface vanilin that can be recovered by washing agar beds with absolute alcohol since vanilin is soluble in absolute alcohol.

[FIGURE 1 OMITTED]

By our estimation, the surface vanilin varies between 30-55% from the total amount of the existing vanilin in the capsules. This surface vanilin is released in the first 5-10 minutes from the beginning of the experiment.

Table 1 summarizes the model parameters for the three models described by equations 2-4 and the correlation coefficients (R).

The diffusion model describes best the cumulative release for the particles with small diameters. For the particles with larger diameters, the phenomena of swelling and erosion are prevailing. Diffusion and dissolution occur altogether for the larger diameter particles.

4. CONCLUSIONS

The purpose of this paper was to present a study of food flavour encapsulation processes and methods, employing agar as matrix for vanilin encapsulation. The results obtained from release experiments were analyzed using different models and the parameters determined from these models suggested that, for small particles, the diffusion prevails, but for those with larger diameters, the swelling and erosion cannot be neglected.

Encapsulation has the potential to be used in the food industry, helping to protect flavours against oxidation or other reactions that occur in the presence of light, as well as undesired interactions with the food itself. This process could eventually be a viable solution to the problem of food flavour degradation with time. The research was funded by research grant type PN II, no. 61045/2007.

5. REFERENCES

Arifin D.Y., Lee L.Y. & Wang, C-H (2006). Mathematical modeling and simulation of drug release from micro-spheres: Implications to drug delivery systems, Advanced Drug Delivery Reviews, 581274-1325

Baker R.W. & Lonsdale H.K. (1974). Controlled release: mechanisms and rates, in: Tanquarry, A.C. & Lacey, R.E. (Eds.), Controlled Release of Biologically Active Agents, Plenum Press, New York, NY, pp. 15-71

Chattopadhyaya S., Singhal R.S. & Kulkarni P.R. (1998). Oxidised Starch as Gum Arabic Substitute for Encapsulation of Flavours, Carbohydrate Polymers 37, 143-144

Pothakamury, U.R. & Barbosa-Canovas, G.V. (1995). Review: Fundamental Aspects of Controlled Release in Foods, Trends in Food Science & Technology 6, 397-406

Walton, N.J., Mayuer, M.J. & Narbad, A. (2003). Molecules of Interest-Vanilin, Phytochemistry 63, 505-515.

Tab 1. Model parameters for vanilin release.

Particle Model 1 Model 2 Model 3
diameter Eq. 2a Eq. 3 Eq. 4
 (mm) and 2b [k.sub.ero]
 [D.sub.ef]
 ([m.sup.2]/s)

 2.5 5.9 x n=0.200 3.55 x
 [10.sup.-10] k=0.164 [10.sup.-11]
 R=0.989 R=0.985 R=0.983

 3.3 2.78 x n=0.282 2.26 x
 [10.sup.-10] k=0.054 [10.sup.-11]
 R=0.991 R=0.957 R=0.970

 5.0 1.50 x n=0.214 6.75 x
 [10.sup.-10] k=0.134 [10.sup.-11]
 R=0.945 R=0.969 R=0.927

 7.0 2.50 x n=0.371 6.55 x
 [10.sup.-10] k=0.033 [10.sup.-11]
 R=0.984 R=0.987 R=0.999