The Influence of Selected Parameters on the Size and Shape of Alginate Beads Prepared by Ionotropic Gelation

Many bead biopharmaceutical characteristics are dependent on the bead shape. Furthermore, the shape is one of crucial parameters for incorporation of beads in more complex drug delivery system. Therefore, the aim of this study was to evaluate the influence of various processing parameters such as hardening time, temperature and concentration of calcium chloride solution and drying conditions on size, shape and morphology of alginate beads prepared by ionotropic gelation method. Theophylline was selected as a model drug. It was found that all studied parameters markedly affected bead form, resembling in most cases to ellipsoid spheres. Their sphericity was estimated three-dimensionally by measuring diameters of frontal and lateral side which were perpendicular to each other. Smaller and more spherical beads were obtained at longer hardening time and higher temperature of calcium chloride solution. The freeze-dried beads were the largest and the most spherical. It was demonstrated that optimization of bead shape as well as size and morphology could be achieved by altering processing parameters.


Introduction
Alginate is linear naturally occurring polysaccharide, consisted of Dmannuronic (M) and L-guluronic (G) acids [1]. The ability of alginate to form gel in the presence of multivalent ions has been applied to prepare beads by ionotropic gelation method where the dispersion of alginate and material to be encapsulated is added dropwise into multivalent ion solution. The contact of droplets with multivalent ions results in instantaneous formation of gel spheres containing uniformly dispersed material throughout the crosslinked alginate matrix. The size of wet beads is dependent on size of a droplet of polymer dispersion, which is influenced by diameter of a nozzle and viscosity of polymer dispersion. However, the drying may influence the size and shape of dry beads.
The scope of the present study was to investigate the influence of some processing parameters on size and shape of alginate beads containing theophylline as a model drug, prepared by ionotropic gelation.

Results and discussion
Wet beads were all spherical regardless of the studied processing parameters. However, the shape of wet bead was influenced by the distance between the top of a needle, through which the dispersion was dropped, and hardening solution as well as the stirring speed of the hardening solution. Namely, the beads flattened on contact with the hardening medium if the alginate dispersion was dropped from the height higher than 6 cm. Furthermore, elongation of the beads during hardening The Influence of Selected Parameters on the Size and Shape … 79 was observed at stirring speed higher than 600 rpm. Therefore, in the present work the stirring speed and the distance between the top of a needle and the hardening solution were fixed at 600 rpm and 6 cm, respectively.
Drying may influence bead size and shape. During drying, the beads shrank significantly and their shape changed. Dry beads had a typical shape with following characteristics: they resembled to flattened, ellipsoid spheres, some of them were completely flattened, some had a hollow in the center and some had regular spherical shape. Therefore, the method for their shape characterization was adapted to their characteristics. First, the shape of bead frontal sides was evaluated. From the mean values of aspect ratio (1.13 ±0.06) and shape factor (0.88 ± 0.06), which were close to 1, it can be seen that they were round-shaped.
However, beads are three-dimensional objects and estimation of their shape on the basis of only two-dimensional image is not sufficient. Therefore, the bead shape was additionally examined from two different perspectives (frontal and lateral) which were perpendicular to each other. The sphericity of beads was thus defined as the ratio between the mean values of the lateral and the frontal diameter.
It was found that bead size and shape was significantly affected by all the tested parameters. Increasing the hardening time from 1 to 30 minutes resulted in decrease of the mean values of bead frontal diameter from 1.91 to 1.33 (Fig. 1a). This could be due to lower drug content since a significant decline of theophylline content with prolongation of hardening time was observed (data not shown). The decrease in size as well as improvement of the sphericity of calcium alginate beads with decreasing theophylline content was also observed by Smrdel et al. [4].
Furthermore, more extensive alginate crosslinking after longer hardening time contributed to decrease of the bead diameter [20]. These results are in accordance with findings of other researchers [4,8,24]. On the other hand, the mean value of the lateral diameter and sphericity increased with increasing hardening time. Beads  At higher temperature of calcium chloride solution more spherical beads were obtained (Fig. 1b). As mentioned above, the bead shape could be affected by drug content (lower drug content at higher temperature was observed (data not shown)) as well as by degree of alginate crosslinking. Since in our case the hardening time was constant, it is assumed that the rate of alginate crosslinking was also affected by the temperature. Better sphericity at higher temperature of calcium chloride solution was therefore probably due to faster and thus more thorough alginate crosslinking and lower drug content. It was reported in the literature that by increasing the calcium chloride concentration smaller beads were obtained [6,18]. Our results are presented in Fig.   1c. The beads hardened in 0.1 and 0.27M calcium chloride solution had significantly lower mean values of frontal diameters than beads prepared using 0.5, 0.68 and 1.0M calcium chloride solution. However, the beads prepared with different calcium chloride concentrations differed significantly in their shape. Some of them were curved and some completely flattened. No correlation between drug content and bead size was observed. However, neither frontal nor lateral diameter neither sphericity alone represents a suitable approach for evaluation of actual size or shape of beads. Thus it is difficult to make any conclusions on the bases of presented parameters.
Drying technique could influence several bead characteristics such as size [11,16,21], shape, mechanical properties [11], swelling properties [11,16], drug release kinetics [24] and even drug migration to periphery of the beads along with water during drying [21,22]. Concerning the latter, accumulation of drug on the surface of the beads could be accountable for several difficulties, i.e. burst effect in drug release, ineffective taste masking of drugs with unpleasant taste, drug loss due to crumbling of the drug from the bead surface… Drug migration during bead drying can be avoided by freeze-drying which involves sublimation but not evaporation of water molecules [22].
In the present work it was demonstrated that drying conditions had great influence on shape and morphology of beads including surface as well as internal structure characteristics of the beads (Figs. 1d and 2). The freeze-dried beads were the largest as they remained almost of the same size as before drying. Furthermore, they were the most spherical proving by sphericity being approximately 1. Their surface was rough (Fig. 2a) and their internal structure was very porous (Fig. 2b)  On the other hand, fluidized-bed-dried and air-dried beads substantially shrank and became smaller during drying. The first ones were folded and irregularly shaped (Fig. 2c). It is supposed that fluidized-bed drying was very rapid resulting in immediate drying of the bead surface. As a consequence of liquid inner part, beads collapsed and particles of irregular shape were formed. The air-dried beads were the smallest (Fig. 1d). They were more spherical than fluidized-bed-dried beads ( Fig. 2e) probably due to slower water removal during air-drying resulting in slower and more uniform shrinkage of beads. Similar findings regarding the size of freezedried and air-dried beads were reported in the literature [11,16,21]. The surface of fluidized-bed-dried as well as of air-dried beads was, however, much smoother than that of freeze-dried. Furthermore, they were not as porous as freeze-dried ones.
From the cross-sections of fluidized-bed-dried and air-dried beads (Figs. 2d and 2f, respectively) a dense interior can be seen. This is in accordance with the results of George et al. [16] and Gal et al. [11]. The latter evaluated also mechanical properties of the freeze-dried and vacuum-dried beads and demonstrated that freeze-dried beads had significantly higher porosity and were mechanically weaker than vacuum-dried beads [11]. High porosity and consequently low density of freeze-dried beads provide a potential for development of multiunit gastroretentive drug delivery system due to buoyancy of the beads in gastric fluid [23].
Furthermore, it was reported that due to the porous structure and larger size of freeze-dried beads the water uptake into the beads was greater compared to vacuum-dried [11] and air-dried beads [16] with dense interior. Such difference in water absorption could have a tremendous impact on drug release kinetics. This hypothesis was confirmed by Sriamornasak [24] who demonstrated significantly faster release of bovine serum albumin from freeze-dried than from air-dried calcium pectinate beads. This indicates that owing to changes in bead morphology drying conditions affect drug release from the beads.
This study demonstrates that slight change in processing parameters can cause a significant influence on size, shape and morphology of alginate beads prepared by ionotropic gelation method. It was proven that altering various 84 P. Smrdel, M. Bogataj, and A. Mrhar: processing parameters enables optimization of bead shape which is one of crucial characteristics for further incorporation of beads into more complex drug delivery systems.

Preparation of alginate beads
Beads were prepared by dissolving sodium alginate (Protanal LF 120M, Selectchemie AG, Switzerland) in distilled water at concentration of 4% (w/v).
After stirring on a magnetic stirrer for half an hour homogeneous and bubble-free dispersion was filled into disposable syringe equipped with a 0.9 mm inner diameter needle. Dispersion was then manually added dropwise (80 drops) into 100 ml of 0.27M calcium chloride solution. The distance between the top of the needle and the surface of the hardening medium was 6 cm. Calcium ions crosslinked alginate and the beads were formed. Beads were allowed to harden in gently stirred calcium chloride solution (stirring rate of magnetic stirrer was 600 rpm) and after that they were isolated and washed twice with 800 ml of distilled water. Beads were dried by air-drying at room temperature, in fluidized-bed apparatus or by freeze-drying.
Fluidized-bed-drying was performed at 40°C and 0.6 bar of atmosphere using STREA1, Niro-Aeromatic, Switzerland. The freeze-drying was done according to the following procedure: immediately after bead isolation from calcium chloride solution and washing with distilled water the beads were poured over with liquid nitrogen and dried in a vacuum dryer VS -50 S Kambič, Laboratorial equipment, Slovenia at reduced pressure at room temperature for 4 hours. All beads were additionally dried at reduced pressure at room temperature overnight.
In general the beads were prepared by dropping the dispersion into 0.27M calcium chloride solution, allowed to harden for 5 minutes and air-dried. In order to examine the effect of hardening time (1, 3, 5, 10, 20 and 30 min), concentration

Characterization of beads
The bead size and shape was estimated using Stereomicroscope Olympus SZX12, Japan, equipped with digital camera (3CCD Color Video Camera, Power HAD, Sony, Japan). Five beads of each sample were examined. First, the bead frontal shape was characterized by aspect ratio and shape factor using image analysis system AnalySIS (AnalySIS® software program, SIS, Germany, 1999).
The aspect ratio was defined as ratio between the maximal and the minimal bead diameter, while shape factor, which provides the information about the roundness of the particles, was calculated by following equation (AnalySIS® software program The closer the values of the aspect ratio and the shape factor were to 1, the rounder the bead was.
Furthermore, the bead frontal diameters were measured in three directions.
After that, the bead was erected vertically so that the lateral diameter was feasible to measure at three different positions. The mean values of the lateral (Dl) and the frontal (Df) diameters were calculated. The parameter in the present work named as sphericity of the beads was defined as the ratio between the mean values of the lateral and the frontal diameter. The values closer to 1 indicated more spherical beads.
The morphological properties of bead surface and the internal structure characteristics of beads were evaluated using scanning electron microscope Supra TM 35 VP, Zeiss, Germany.

Statistical analysis
Statistical analysis was performed using SPSS software program (Version 12.0.1, SPSS Inc., Chicago, 2004). The one-way ANOVA followed by the LSD test for post hoc comparisons was applied for evaluation of the influences of hardening time, concentration and temperature of calcium chloride solution and drying conditions on bead's size and sphericity. Games-Howell test was used instead of the LSD test when homoscedasticity of variances was not proven. The significance level was set at 0.05 for all the tests.