Acta Poloniae Pharmaceutica ñ Drug Research, Vol. 63 No. 6 pp. 521ñ534, 2006 ISSN 0001-6837Polish Pharmaceutical Society

Ibuprofen, α-methyl-4-(2-methylpropyl)-ben-zene acetic acid is a non-steroidal anti-imflammato-ry, antipyretic and analgesic drug. This drug is indi-cated for the relief of mild to moderate pain andinflammation in conditions such as dysmenorrea,migraine, postoperative pain, dental pain, in whichdisorders an immediate available dose is requested.It is also used in chronic disorders as ankylosingspondylitis, osteoarthritis and rheumatoid arthritisfor all of which a sustained release is desirable (1).The usually daily dose by mouth is 1.2 to 1.8 gdivided in different administrations. The drug isreadily absorbed from the gastrointestinal tract, andpeak plasma concentrations occur about 1 to 2 hafter ingestion (2). Unfortunately, ibuprofen causesa certain irritation in the gastrointestinal mucousmembrane and possesses a bitter taste and aftertaste.The half-life in plasma is about 2 h. The short half-life and the low single administration dose makeibuprofen a very good candidate for the formulationof controlled release multiple-unit dosage forms (3-6). At the same time, great attention has been devot-ed on the possibility to prepare ibuprofen micros-

pheres in order to formulate oral controlled releasesystems, to protect the gastric mucous membranefrom drug irritation or to mask its unpleasant taste.

Microspheres are one of the microparticulatesystems and are prepared to obtain prolonged orcontrolled drug delivery, to improve bioavailabilityor stability and to target drug to specific sites.Microspheres can also offer advantages like limitingfluctuation within therapeutic range, reducing sideeffects, decreasing dosing frequency and improvingpatient compliance. They spread out more uniform-ly in the GI tract, thus avoiding exposure of themucosa to high concentration of drug and ensuringmore reproducible drug absorption. The risk of dosedumping also seems to be lower than with a single-unit dosage form (7-9). In this study, ibuprofenmicrospheres were prepared by a novel technique,called the quasi-emulsion solvent diffusion method,which was proposed by Kawashima et al. (10).Compared with the solvent evaporation method formicrosphere preparation, the solidification of theliquid droplet in this process was much faster.Further, the use of a harmful organic solvent could

PREPARATION AND EVALUATION OF MODIFIED RELEASE IBUPROFENMICROSPHERES WITH ACRYLIC POLYMERS (EUDRAGIT®) BY QUASI-EMULSION SOLVENT DIFFUSION METHOD: EFFECT OF VARIABLES

BURCU DEVRIM AND KANDEMIR CANEFE*

Ankara University, Faculty of Pharmacy, Department of Pharmaceutical Technology 06100-Tando�an ñ Ankara ñ TURKEY

Abstract: The aim of this study was to prepare and evaluate microspheres containing ibuprofen. Microsphereswere prepared by modified quasi-emulsion solvent diffusion method. The influence of formulation factors(drug-polymer ratio, volumes of solvent, polyvinyl alcohol concentration and type of polymer) on the mor-phology, particle size distribution, drug loading capacity, micromeritical properties and the in vitro release char-acteristics of the microspheres were investigated. Physical characterizations of ibuprofen microspheres werealso carried out using scanning electron microscopy, X-ray diffractometry and IR spectrophotometry. It wasfound that the yield of preparation was dependent on the initial temperature gradient between the emulsionphases. When there was an initial difference of temperature between the aqueous phase and dispersed emulsionphases, yield of preparation was increased distinctly. The drug loading capacities were very high for all for-mulations of the microspheres which were obtained. Mean particle size changed by changing the drug-polymerratio, volumes of solvent or polyvinyl alcohol concentration. The flow properties were much improved overthose of the original crystals. In vitro dissolution results showed that the release rate of ibuprofen was modifiedin all formulations. Although ibuprofen release rates from Eudragit® RS microspheres were very slow, theywere fast from Eudragit RL microspheres. These results observed that if Eudragit RS and Eudragit RL are usedin combination, optimum release profiles may be obtained.

Keywords: Ibuprofen; EudragitÆ; microspheres; quasi-emulsion solvent diffusion method; modified release

521

* Corresponding author: Tel.: +90-312-212 9151, Fax: +90-312-213 1081, E-mail: canefe@pharmacy.ankara.edu.tr.

522 BURCU DEVRIM and KANDEMIR CANEFE

be avoided, and the reduced pressure and/or theheating to evaporate the solvent would be unneces-sary in this method (11). The rare coalescence of amicrospheres observed in the process did not requirethe addition of antiadhesion agents, such as magne-sium stearate and talc, which are often used in theevaporation method.

Acrylic polymers are widely used as tabletcoatings and as retardants of drug release in sus-tained release formulations (12, 13). Methacrylatecopolymers (Eudragits) have recently receivedincreased attention for modified dosage formsbecause of their inertness, solubility in relativelynon-toxic solvents and availability of resins withdifferent properties (14-16). Eudragit® RS andEudragit® RL polymers are copolymers of poly(eth-ylacrylate, methyl-methacrylate and chlorotri-methyl-ammonioethyl methacrylate), containing anamount of quaternary ammonium groups between4.5-6.8% and 8.8-12% for RS and RL, respectlively(17). The copolymer Eudragit RS PM® differs fromEudragit® RS 100 in that it contains 0.5% talc.Eudragit® RS and Eudragit® RL are insoluble inwater and digestive juices, but they are permeableand both have pH-independent release profiles. Thepermeability of Eudragit® RS and RL in aqueousmedia is due to the presence of quaternary ammoni-um groups in their structure; Eudragit® RL has agreater proportion of these groups and as such ismore permeable than Eudragit® RS (18, 19).

In this study, ibuprofen loaded microsphereswere prepared by using quasi-emulsion solvent dif-fusion method. Firstly, we investigated some formu-lation variables (initial difference of temperature

between the emulsion phases, drug-polymer ratio,volumes of ethanol and stirring speed) to obtainspherical particles. Then the influences of formula-tion variables on the microsphere properties wereexamined and the microsphere formulations suitableto achieve our goal were determined.

EXPERIMENTAL

Reagents and equipment

Reagents: ibuprofen (Eczac�ba��), Eudragit®

RS PM, Eudragit® RS 100, Eudragit® RL 100(Rˆhm-Pharma), ethanol (Riedel de-HaÎn),polyvinyl alcohol (PVA 72 000), sodium hydroxide,potassium dihydrogen phosphate (Merck). The otherchemicals were of analytical grade and distilledwater was used for all experiments.

Equipment: UV spectrophotometer (ShimadzuUV-1202), pH meter (Meter Lab), laser diffrac-tometer (Sympatec HELOS-H0728), X-ray diffrac-tometer (Rigaku, D-Max/2200), IR-spectropho-tometer (Jasco, FT/IR 420), dissolution tester(Aymes), SEM analysis (JEOL JSM-840A).

METHODS

Preparation of microspheres

In order to produce the microspheres, modi-fied quasi-emulsion solvent diffusion method wasused (20). Weighed amounts of ibuprofen andacrylic polymer were dissolved in 5 mL of ethanolat 45OC. The formed ethanolic solution was pouredinto water containing polyvinyl alcohol (0.025-

Table 1. Formulations of the microspheres.

Formulation Ibuprofen Eudragit® PVA Ethanol (mL)code (%) (%) 72.000 (%)

F1 100 - 0.05 5

F2 80 20(RSPM) 0.05 5

F3 75 25(RSPM) 0.05 5

F4 71.43 28.57(RSPM) 0.05 5

F5 66.67 33.33(RSPM) 0.05 5

F6 66.67 33.33(RSPM) 0.05 3

F7 66.67 33.33(RSPM) 0.05 6

F8 66.67 33.33(RSPM) 0.025 5

F9 66.67 33.33(RSPM) 0.1 5

F10 66.67 33.33(RS100) 0.05 5

F11 66.67 33.33(RL100) 0.05 5

*Microspheres were produced with the agitation speed of 450 rpm.

Preparation and evaluation of modified release ibuprofen microspheres... 523

0.1% w/v) and stirring continuously with a pro-peller type agitator (Model RZR-2000; HeidolphElectro). The system was thermally controlled at20OC. Ethanol solution was finely dispersed in theaqueous phase as discrete droplets. The finely dis-persed droplets of the polymer solution of the drugwere solidified in the aqueous phase via counterdiffusions of ethanol and water out of and into thedroplets. After 30 min of stirring, the microsphereswere separated by filtration, washed twice with 50mL of water and then dried in oven at 37OC for 24h. Dried microspheres were stored in desiccatorcontaining CaCl2. Three independent batches wereprepared. In Figure 1, the processing of this tech-nique is illustrated. The representative formulationsfor the preparation of microspheres are tabulated inTable 1.

Microspheres dried at 37OC were then weighedand the yield of microsphere preparation was calcu-lated using the Eq. (1):

The amount of microspheres obtained (g)Percent yield = ñññññññññññññññññññññññññññññññññññ× 100 (1)

The theoretical amount (g)

CONTROLS ON MICROSPHERES

Powder X-ray diffractometry

Powder X-ray diffractometric analyses wereperformed using an X-ray diffractometer (Rigaku,D-Max/2200). Diffraction of raw crystals and

microspheres of ibuprofen were measured. Beforemeasuring the X-ray powder diffractogram, themicrospheres were crushed into powder becausethey were too big to measure. The data were collect-ed in the continuous scan mode using step size of0.01O 2θ. The scanned 2 range was 2θ to 35O.

Infrared spectroscopy

Infrared spectra of the pure drug and of themicrospheres were determined using an IR-spec-trophotometer (Jasco, FT/IR 420), using the KBrdisk technique (about 10 mg sample for 100 mg ofdry KBr). The scanning range used was from 4000to 200 cm-1 at a scan period of 3 min.

Particle size analysis

The particle size of the microspheres wasdetermined by laser diffractometry using SympatecHELOS (H0728) particle size analyzer. In thisstudy, filtered and degassed purified water was usedas a carrier fluid. About 0.3-0.5 mg of microsphereswere dispersed in purified water in the sample unitand were circulated 2000 times per minute. Eachdetermination was carried out in triplicate.

Determination of the bulk density

Determination of the bulk density of themicrospheres was obtained by a tapping method andfound in g/mL (21, 22). 10 mL of microspheres wereweighed and were tapped 20 times. Upon tappingthe volume was measured again. Bulk density wasdetermined according to the ratio of the mass andvolume of the microspheres.

Determination of repose angle

Determination of the repose angle wasobtained for 10 g microspheres (23). For this study,microspheres were poured into a conical flask whichhad a 0.9 cm diameter and was placed 10 cm abovethe surface. Repose angle was calculated accordingto the tangent of the ratio of the height and diameterof the bulk.

Determination of drug loading capacity

Drug loading capacity of the microspheres wasdetermined by dissolving accurately weighed por-tions for each batch in ethanol which dissolves boththe active agent and the polymer (24). Filtration ofmicrospheres was done by using Whatman No. 42filter with a porosity of 2.5 µm. The dissolved drugamount was measured spectrophotometrically at264 nm. The polymer does not interfere with theassay at this wavelength. The drug loading capacitywas calculated according to Eq. (2).

Figure 1. Preparation procedure of microspheres.

524 BURCU DEVRIM and KANDEMIR CANEFE

Assayed drug contentDrug loading capacity = ññññññññññññññññññññññññ× 100 (2)

Theoretical drug content

SEM analysis

The shape and surface characteristics of micros-pheres were observed by a scanning electron micro-scope (JEOL JSM-840A). The samples were dustedonto double sided tape on an aluminum stub and sput-ter-coated with gold particles in an argon atmosphere.

In vitro dissolution studies

The dissolution rate of pure drug and the drugrelease rate from the microspheres were studiedusing USP 24 Type 2 (paddle) method (25). TheUSP paddle method was followed using a standard

1-L dissolution vessel and 7.5 cm diameter paddle,assembled as per the USP 24 Physical Test sectionon dissolution. The vessel was partially immersed ina suitable water bath that permitted holding the tem-perature inside the vessel at 37 ± 0.5OC during thetest and keeping the bath fluid in constant, smoothmotion. The paddle used as the stirring element wasformed a blade and a shaft. The shaft was positionedso that its axis was not more than 2 mm at any pointfrom the vertical axis of the vessel, and rotatedsmoothly without significant wobble. The distanceof 25 ± 2 mm between the blade and the inside bot-tom of the vessel was maintained during the test.The dissolution medium was agitated at 50 rpm forthe duration of the test. The samples per batch were

Figure 2. X-ray diffraction patterns of original crystals (A), F5 (B), F10 (C) and F11 (D).

Preparation and evaluation of modified release ibuprofen microspheres... 525

Figure 3. IR-spectra of original crystals (A), F5 (B), F10 (C) and F11 (D).

tested in 900 mL of pH 6.8 phosphate buffer as dis-solution medium. To improve wetting of the micros-pheres, a small amount of surfactant (Polysorbate80, 0.1% w/v) was added to the dissolution media.Accurately weighed samples of ibuprofen or micros-pheres were added to dissolution medium kept at 37± 0.5OC. In order to keep the total surface area of themicrosphere samples constant and thus to get com-parable results, the release studies all were carriedout with 350 µm size fractions of microspheres pre-pared. An aliquot of the release medium was with-drawn at predetermined time intervals and passed

through a 2.5 µm filter (Whatman No. 42). The ini-tial volume of the release medium was maintainedby adding equivalent amount of fresh medium aftereach sampling. The absorption of the samples wasrecorded at a wavelength of 264 nm spectrophoto-metrically. From the absorbance readings, cumula-tive percentage of ibuprofen dissolved was calculat-ed. Ibuprofen sink conditions were determined inphosphate buffer pH 6.8 (using an amount of drugequivalent to three times of the dose in the pharma-ceutical formulation in 900 mL of medium) andwere maintained during all measurements.

526 BURCU DEVRIM and KANDEMIR CANEFE

Figure 4. Mean particle sizes of microspheres (n = 3).

Table 2. Mean particle size of microspheres.

Formulation Mean particle size Standart deviation % 95 C.L.code (µm)

F1 131.85 0.14 0.08

F2 170.73 0.16 0.09

F3 180.84 0.76 0.44

F4 269.99 0.40 0.23

F5 316.65 0.34 0.20

F6 1595.65 0.27 0.16

F7 89.21 0.18 0.10

F8 618.12 0.20 0.12

F9 185.72 0.54 0.31

F10 272.96 0.20 0.12

F11 391.03 0.15 0.09

*C.L.: Confidence limits

Table 3. Bulk density of the microspheres.

Formulation Bulk Standart % 95 C.L*.code density deviation

(g/mL)

F1 0.544 0.004 0.003

F2 0.513 0.009 0.006

F3 0.524 0.008 0.006

F4 0.515 0.008 0.006

F5 0.532 0.004 0.003

F6 0.427 0.010 0.007

F7 0.501 0.004 0.003

F8 0.514 0.005 0.003

F9 0.515 0.007 0.005

F10 0.507 0.007 0.005

F11 0.450 0.006 0.004

* C.L.: Confidence limits

Table 4. Repose angle of the microspheres.

Formulation Repose Standart % 95 C.L.*code angle deviation

F1 23.68 0.510 0.426

F2 24.52 0.824 0.589

F3 23.23 0.168 0.120

F4 23.71 0.827 0.393

F5 18.06 0.566 0.180

F6 22.45 0.583 0.417

F7 23.27 0.619 0.443

F8 22.21 0.142 0.102

F9 20.58 0.687 0.492

F10 21.26 0.636 0.455

F11 24.61 0.110 0.092

* C.L.: Confidence limits

Preparation and evaluation of modified release ibuprofen microspheres... 527

Table 5. Results of the drug loading capacity.

Formulation Theoretical Assay Drug loading Standartcode drug content drug content capacity deviation % 95 C.L.*

(%) (%) (%)

F1 100 97.16 97.16 0.12 0.287

F2 80 79.29 99.11 0.49 0.777

F3 75 73.98 98.64 0.26 0.203

F4 71.43 70.43 98.60 0.25 0.266

F5 66.67 66.53 99.79 0.28 0.180

F6 66.67 65.74 98.61 0.11 0.273

F7 66.67 66.22 99.33 0.26 0.270

F8 66.67 64.13 96.19 0.30 0.192

F9 66.67 63.47 95.20 0.18 0.188

F10 66.67 66.10 99.15 0.15 0.154

F11 66.67 68.64 102.95 0.35 0.368

*C.L.: Confidence limits

Figure 5. Yield of preparation (n = 3).

Calculations of dissolution test results kinetically

Dissolution test results obtained from the paddlemethod in pH 6.8 phosphate buffer were studied byusing SPSS for Windows 11.0 according to zero order,first order, Hixson-Crowell and Higuchi kinetics.

Statistical analysis

The data obtained from the particle size, encap-sulation efficiency and release rate determinationstudies of ibuprofen microspheres were analyzedstatistically with ANOVA and t-test by using SPSSfor Windows 11.0.

DISCUSSION AND CONCLUSION

Modified quasi-emulsion solvent diffusionmethod was used to prepare ibuprofen micros-pheres. The reasons to choose this method as the

production method were its simplicity, low cost,success with poor aqueous solubility drugs and theproduction of microspheres of relatively high drugloading, as reported by Pignatello et al. (18). In thisprocess, the drug and polymer were dissolved in asolvent such as ethanol. The final solution was dis-persed into an aqueous phase with constant agita-tion, forming o/w emulsion droplets. The solventand water counter-diffused out of and into thedroplets, respectively. The diffused water within thedroplets may decrease the drug and polymer solubil-ities. Both components co-precipitated and contin-ued ethanol diffusion resulted in further solidifica-tion, producing matrix-type microspheres, as shownin Figure 6. First, the effect of initial difference oftemperature between the aqueous phase and dis-persed emulsion phases on the microsphere forma-tion was investigated. When there was no difference

528 BURCU DEVRIM and KANDEMIR CANEFE

Figure 6. Scanning electron micrographs of microspheres and their surfaces prepared from the formulations given in Table 1 (A) F2, (B)F5, (C) F6, (D) F7, (E) F10 and (F) F11.

of initial temperature gradient between the emulsionphases, as reported by Kawashima et al. (10),microspheres coalesced together and the resultantyield of microspheres was relatively low.Conversely, no coalescence was observed by initialtemperature gradient between the continuous anddispersed emulsion phases as in our study. Theinhibitory effect on the coalescence of dropletsobserved with increase in the gradient temperaturebetween the emulsion phases may be explained byinfluence of this factor on the microsphere harden-

ing sequence. Due to the heat transfer, the rapidsolidification of droplets by precipitation of polymerat the droplet surface prevented the coalescence andfusion of droplets. As a consequence, the yield ofmicrospheres varied from 65 to 89.9%. Then, vari-ous formulations with different drug-polymer ratiosand ethanol volume were tried and stirring speedwas changed to obtain spherical particles. Whenamount of drug and polymer in dispersed phase wastoo high (drug-polymer ratio was 1:1, w/w) nospherical particles were obtained. These results

Preparation and evaluation of modified release ibuprofen microspheres... 529

Figure 6 (cont.). Scanning electron micrographs of microspheres and their surfaces prepared from the formulations given in Table 1 (A)F2, (B) F5, (C) F6, (D) F7, (E) F10 and (F) F11.

show that the amount of solid, thus the viscosity ofthe inner phase is an important factor for the prepa-ration of microspheres. Keeping the drug amountand the solvent volume constant, spherical particleswere obtained as the amount of polymer wasdecreased to give a drug-polymer ratio of 2:1, 3:1 or4:1. The ethanol volume of the inner phase alsoinfluenced microsphere formation in the variousbatches. The microsphere yield declined sharply aslarger volumes of ethanol or less viscous dispersedphase solutions were used. When a solvent volume

of 10 mL was used, the resulting ethanol phase wasvery dilute. The addition of the two phases togetherproduced rapid intermixing of the phases, resultingin immediate precipitation of drug and polymerbefore droplets could form. Hence, no microsphereswere produced. With increasing concentration ofibuprofen, the recoveries of spherical matricesincreased rapidly. At a concentration of 1.25 g/mL,the loss of microspheres due to adhesion to the pro-peller and vessel wall became relatively larger thanthe recovered amount, since the volume of ethanol

530 BURCU DEVRIM and KANDEMIR CANEFE

Figure 6 (cont.). Scanning electron micrographs of microspheres and their surfaces prepared from the formulations given in Table 1 (A)F2, (B) F5, (C) F6, (D) F7, (E) F10 and (F) F11.

solution was small (2 mL). Three different stirringspeeds (400, 450 and 500 rpm) were selected and itwas observed that particle size of microspheresdecreased with the increasing of the stirring speed,but at 500 rpm, because of the turbulence of aqueousphase, the polymer stuck around the paddle of themixer and a great loss of the polymer was indicated.Therefore, as a stirring speed of 450 rpm for thepreparation of microspheres was suitable.

According to the particle size analysis, it wasfound that particle size was dependent on drug-poly-

mer ratio, volume of ethanol, type of polymer andconcentration of polyvinyl alcohol (Table 2). It wasobserved that when polymer amount increased, par-ticle size of the microspheres increased (p < 0.05).F5 formulation produced with 33.33% polymer con-centration had bigger particle size than F2 formula-tion produced with 20% polymer concentration.Increasing the polymer load led to a more viscoussolution. When the viscous polymeric solution waspoured into the aqueous phase, larger droplets andthus larger microspheres, were formed.

Preparation and evaluation of modified release ibuprofen microspheres... 531

Figure 7. Release profiles of ibuprofen from microspheres pre-pared with different drug-polymer ratio in pH 6.8 phosphate bufferat 37 ± 0.5OC (n = 3).

Figure 9. Release profiles of ibuprofen from microspheres pre-pared at different PVA 72 000 concentration in pH 6.8 phosphatebuffer at 37 ± 0.5OC (n = 3).

Figure 8. Release profiles of ibuprofen from microspheres pre-pared with different ethanol volume in pH 6.8 phosphate buffer at37 ± 0.5OC (n = 3).

Figure 10. Release profiles of ibuprofen from microspheres pre-pared different polymer in pH 6.8 phosphate buffer at 37 ± 0.5OC(n = 3).

When the amount of solvent was decreased,polymer and drug concentration increased. As a resultof the increase in the polymer concentration, micros-pheres particle size increased (p < 0.05) as shown inTable 2. The biggest particle size was obtained fromF6 which was produced with 3 mL of ethanol.

In this study, polyvinyl alcohol was preferredas an emulsifying agent rather than sucrose fattyacid ester used in the studies of Kawashima et al.(10). The presence of polyvinyl alcohol significant-ly prevented aggregation of the droplets with solidi-fied outer shells during the process. The dispersionof the ethanolic solutions of the drug and polymerinto droplets in the medium depended on the con-centration of polyvinyl alcohol in the medium.

Hence, when polyvinyl alcohol concentrationincreased in disperse phase, particle size of micros-pheres decreased (p < 0.05). As polyvinyl alcoholconcentration was increased from 0.025 to 0.1%,particle size of microspheres decreased from 618 µmto 186 µm (Table 2).

Particle size analysis of microspheres preparedby using different types of Eudragit® observed thatmean particle size of the microspheres depended onthe type of polymer (p < 0.05) in contrast to the ear-lier report by DortunÁ and Haznedar (19). Smallerparticles were obtained with Eudragit RS as shownin Table 2.

The X-ray powder diffraction patterns and IRspectra of the microspheres, as well as those of orig-

532 BURCU DEVRIM and KANDEMIR CANEFE

inal drug crystals are shown in Figures 2 and 3,respectively. The drug was dispersed in the poly-meric matrices in a microcrystalline form, withoutpolymorph change or transition phenomena into anamorphous form. Comparison of the X-ray diffac-tion patterns of ibuprofen (pattern A) and micros-pheres prepared with Eudragit® RS PM (F5),Eudragit® RS 100 (F10) and Eudragit® RL 100 (F11)(patterns B, C and D), showed no significant reduc-tion in the characteristic peak intensities, suggestingthat the extent of ibuprofen crystallinity was notreduced by the polymer.

The spectral observations indicated that theprincipal IR absorption peaks observed in the spec-tra of ibuprofen were close to those in the spectra ofthe microspheres (Figure 3). IR spectra of the drugand microspheres showed characteristic bands ofC=O vibrations of the esterified carboxylic acidgroups at 1730 cm-1 and of the carboxylic acidgroups at 1705 cm-1. For all the microspheres wereindications that there is no strong interactionbetween the drug and the polymers.

According to bulk density experiments, thesmallest value was obtained from the microspheresprepared with 3 mL of ethanol (F6) (Table 3). It wasseen that as the particle size increased, the bulk den-sity decreased. That was because, the bulk densitydecreased due to the increase in the particle size andintraparticular space.

Repose angle results showed that all formula-tions had suitable flow properties as shown in Table4. All repose angles were under 30O. The flow prop-erties were much improved over those of the origi-nal crystals. All formulations presented similar flowproperties while the particle sizes were differentbecause all formulations were spherical and theirsurface was smooth. It seems that the presence of

polymer contributes to improved sphericity of themicrospheres.

According to Figure 5, the yields of prepara-tion were very high for all microspheres prepared(80.1-89.9%) with initial temperature gradientbetween emulsion phases.

Assays generally showed that the drug contentof the microspheres in the various batches showedgood correlation with the theoretical drug loadings.The drug was uniformly encapsulated into themicrospheres irrespective of initial drug concentra-tions. As seen in Table 5, drug loading capacitieswere very high for all microspheres and were notaffected by the type of polymer (p > 0.05), the drug-polymer ratio (p > 0.05), volumes of ethanol (p >0.05) and polyvinyl alcohol concentration (p >0.05). The high content of ibuprofen in micros-pheres was believed to be due to the poor solubilityof drug in poor solvent.

Figure 6 shows the surface morphology andcross-section of the resultant microspheres whichwere investigated by SEM. It was observed thatmany micropores were formed on the surface of themicrospheres. It was assumed that the coprecipita-tion of drug and polymers firstly occurred on thesurface of the quasi-emulsion droplets and formedthe film like a shell on the outer surface of droplets.Further diffusion of the ethanol out of the dropletsresulted in the cavity and micropores were left in theinside of the microspheres. The surface morphologyof microspheres was dependent on the polymer con-centration in the ethanol droplet. With a decrease inthe polymer concentration, the surface becamerough and porous. Conversely, at higher concentra-tions, a smooth shell-like structure with small poreswas formed on the microspheres. While F2 formula-tion prepared with 4:1 drug-polymer ratio had a very

Table 6. Kinetic evaluation of dissolution rate tests of microsphere formulations in pH 6.8 phosphate buffer.

Formulation code F1 F2 F3 F4 F5 F6 F7 F8 F9 F10 F11

Kinetics

Zero k0 0.937 0.234 0.197 0.140 0.117 0.106 0.121 0.127 0.103 0.1100 0.184

order r2 0.689 0.760 0.825 0.853 0.895 0.927 0.898 0.913 0.883 0.868 0.809

First k1 0.050 0.020 0.007 0.004 0.002 0.002 0.002 0.002 0.002 0.002 0.006

order r2 0.863 0.969 0.952 0.962 0.965 0.976 0.968 0.946 0.942 0.939 0.940

Hixson- k4 0.040 0.010 0.007 0.004 0.003 0.002 0.003 0.003 0.002 0.002 0.006

Crowell r2 0.813 0.949 0.938 0.934 0.946 0.963 0.948 0.946 0.924 0.918 0.924

Q→t1/2 k 10.5 4.99 4.50 3.62 2.99 2.67 3.07 3.19 2.63 2.82 4.52

r2 0.821 0.911 0.976 0.967 0.986 0.995 0.986 0.980 0.978 0.974 0.948

k0 ñ Zero order dissolution rate constant (mg/minute); k1 ñ First order dissolution rate constant (minute-1);k4 ñ Hixson-Crowell rate constant (minute-1); k ñ Higuchi rate constant (minute-1/2)

Preparation and evaluation of modified release ibuprofen microspheres... 533

porous surface, it was obtained smooth shell-likestructure with 2:1 drug-polymer ratio at F5 formula-tion, as shown in Figure 6. Additionally, it wasobserved that the surface of microspheres preparedwith Eudragit RS was smoother than that preparedwith Eudragit RL.

Comparison of the batches made with acrylicpolymer and the batch made without polymer showsthat microencapsulation of ibuprofen with polymerresulted in a marked decrease in the drug release rateof the microspheres (Figure 7). The difference wassignificant at each time point (p < 0.01). This wasconsistent with the localization of crystals in thecore of the microspheres. Eudragit® RS and RL arenot biodegradable and the solvent has to diffuse inpolymer to dissolve the drug. Thereafter, the solu-tion of ibuprofen can leave the core of the micros-pheres. Further, Figure 7 clearly illustrates that therate of drug release from the microspheres dependedon the polymer concentration in the system. Whenthe concentration of the polymer in the systemincreased the release rate of ibuprofen decreased.The difference was also significant (p < 0.01) for 8h. It is suggested that a reduced diffusion path andincreased tortuosity may retarded the drug releaserate from the matrix at higher concentrations ofpolymer. An inverse relationship was observedbetween polymer content and drug release rate fromprepared microspheres. Microspheres containing20% Eudragit® RS PM released the drug more rap-idly, while those with 33.3% Eudragit® RS PMexhibited a relatively slower drug release profile.

The effect of ethanol volume on drug releasefrom microspheres is presented in Figure 8. Varyingvolumes of ethanol were used in the preparation ofthe dispersed phase. The quantity of ibuprofen andEudragit® added was constant for all batches, there-fore varying the ethanol volume affected drug andpolymer concentration in ethanol (Table 1). Figure 8illustrates that the drug release profiles differedaccording to the volume of ethanol used duringmicroencapsulation. This difference was significant(p < 0.01) for 8 h. The highest release rate wasachieved from F7 prepared with 6 mL of ethanol.The use of increasing volumes of ethanol producedmicrospheres with increasing drug release, despitethe final composition of the batches being the same.The surface morphology of microsphere wasdependent on the concentration of drug and polymerin the ethanol droplet as demonstrated in Figure 6.With a decrease in concentration, the surfacebecame rough and porous. These findings suggestedthat these pores provided a channel for release ofdrug from the microspheres.

As shown in Figure 9, dissolution rate was notaffected by polyvinyl alcohol concentration (p > 0.01).

The release profiles of ibuprofen from micros-pheres prepared with different Eudragit® types areillustrated in Figure 10. The effect of retardation onthe dissolution rate depended on the type ofEudragit®. As drug release rates were very slow andincomplete from Eudragit® RS microspheres, thesame formulation was prepared using Eudragit® RLas polymer and different drug release profile wasobserved. The release of ibuprofen from F11 pre-pared with Eudragit® RL was remarkably faster thanfrom F5 and F10 prepared with Eudragit® RS andthe difference was significant (p < 0.01) for 8 h. Itcould be explained considering the chemical struc-ture of Eudragit®. The Eudragit® RL and RS are syn-thesized from acrylic and methacrylic esters withhigh and low content of quaternary ammoniumgroups (8.8-12% and 4.5-6.8%, respectively) andresult in microspheres with different water perme-ability. Due to the content of the quaternary ammo-nium groups, Eudragit® RS is only slightly perme-able; hence drug release is relatively retarded,whereas Eudragit® RL is freely permeable, so thatrelease is less retarded. These results observed thatif Eudragit RS and Eudragit RL are used in combi-nation, optimum release profiles may be obtained.

In addition, the release mechanism of ibupro-fen from these Eudragit® microspheres was evaluat-ed on the basis of theoretical dissolution equationsincluding zero order, first order, Hixson-Crowelland Higuchi kinetic models, since different releasekinetics are assumed to reflect different releasemechanisms. The results are shown in Table 6. Itshows that the release pattern of ibuprofen fromEudragit® microspheres corresponded best to theHiguchi equation, which shows a linear relationshipbetween the dissolved percent of drug and thesquare root of residence time. It was demonstratedthat the microspheres possessed matrix structure,from which the dissolved drug diffused into the dis-solution medium. Furthermore, when polymer con-centration was increased in the formulations, corre-spondence was increased to the Higuchi equation.

Consequently, ibuprofen microspheres withacrylic polymers (Eudragit® RS and RL) were pre-pared successfully by the modified quasi-emulsionsolvent diffusion method. It was observed that theinitial difference of temperature between the aque-ous phase and dispersed emulsion phases influencedthe microsphere formation. When there was an ini-tial temperature gradient between the continuousand dispersed emulsion phases no coalescence wasobserved in the formulations. The formulation vari-

534 BURCU DEVRIM and KANDEMIR CANEFE

ables, i.e. drug-polymer ratio, volume of ethanol,concentration of polyvinyl alcohol and type of poly-mer were found to have an effect on the particle size,micromeritic and in vitro drug release characteristicof the prepared microspheres. The resultant micros-pheres have the desired micromeritic properties. Thedissolution test results shown that the release rate ofibuprofen decreased when polymer concentrationwas increased in the system. Ibuprofen release ratesfrom Eudragit RS microspheres were very slowwhereas release rates from Eudragit RL micros-pheres were faster. The drug release profile aimedfor peroral administration may be obtained byadding Eudragit RS and RL and changing the ratioof these polymers. In vitro dissolution findingsshowed that drug release appeared to fit the Higuchimatrix model.

Acknowledgments

The authors would like to thank Eczac�ba��(Turkey) for providing ibuprofen.

REFERENCES

1. Adams S.S., Buckler J.W.: Ibuprofen andFlurbiprofen. in: E.C. Huskisson (Ed.), Anti-Rheumatic Drugs, p. 244, Praeger, West Port,CT 1983.

2. Sweetman S.C.: Martindale, The ExtraPharmacopeia, 34th ed., Royal PharmaceuticalSociety, London 2005.

3. Gallardo A., Eguiburu J.L., Berridi M.J.F.,Roman J.S.: J. Controlled Rel. 55, 171 (1998).

4. Jbilou M., Ettabia A., Guyot-Hermann A.-M.,Guyot J.-C.: Drug Dev. Ind. Pharm. 25(3), 297(1999).

5. Leo E., Forni F., Bernabei M.T.: Int. J. Pharm.196, 1 (2000).

6. Maheshwari M., Ketkar A.R., Chauhan B., PatilV.B., Paradkar A.R.: Int. J. Pharm. 261, 57(2003).

7. Davis S.S., Illum L.: Biomaterials 9, 111(1988).

8. Ritschel W.A.: Drug Dev. Ind. Pharm. 15, 1073(1989).

9. Follonier N., Doelker E.: S.T.P. PharmaSciences 2, 141 (1992).

10. Kawashima Y., Niwa T., Handa T., TakeuchiH., Iwamoto T., Itoh K.: J. Pharm. Sci. 78, 68(1989).

11. RÈ M.I., Biscans B.: Powder Technology 101,120 (1999).

12. Khanfar M., Salem M., Najib N., Pillai G.: ActaPharm. Hung. 67, 235 (1997)

13. Jovanovic M., Jovicic G., Djuric Z., Agbaba D.,Karljikovic-Rajic K., Nikolic L.: Acta Pharm.Hung. 67, 229 (1997).

14. Zaghloul A.A., Vaithiyalingam S.R., FaltinekJ., Reddy I.K., Khan M.A.: J. Microencapsul.18, 651 (2001).

15. Yang M., Cui F., You B., Fan Y., Wang L., YueP., Yang H.: Int. J. Pharm. 259, 103 (2003).

16. Mateovic-Rojinik T., Frlan R., Bogataj M.,Bukovec P., Mrhar A.: Chem. Pharm. Bull. 53,143 (2005).

17. Eudragit RS and Eudragit RL data sheets, RˆhmPharma GmbH, Darmstadt 1991.

18. Pignatello R., Bucolo C., Ferrara P., Maltese A.,Puleo A., Puglisi G.: Eur. J. Pharm. Sci. 16, 53(2002).

19. DortunÁ B., Haznedar S.: Int. J. Pharm. 269,131 (2004).

20. Devrim B., Canefe K.: Proceedings of the 8th

European Congress of PharmaceuticalSciences, Brussels, Belgium; Eur. J. Pharm. Sci.23 (Suppl. 1), S40 (2004).

21. Canefe K., Ate�o�lu-GˆÁmez S.: Boll. Chim.Farmac. 141, 423 (2002).

22. You J., Cui F., Han X., Wang Y., Yang L., YuY., Li Q.: Colloids and Surfaces B:Biointerfaces 48, 35 (2006).

23. Khan M.A., Bolton S., K�slal�o�lu M.S.: Int. J.Pharm. 102, 185 (1994).

24. Kachrimanis K., Ktistis G., Malamataris S.: Int.J. Pharm. 173, 61 (1998).

25. The United States Pharmacopoeia, 24th ed.,United States Pharmacopoeial Convention Inc.,National Publishing Philadelphia, PA 2000.

Received: 31.05.2006