Preparation and characterization of sodium hexameta phosphate cross-linked chitosan microspheres for controlled and sustained delivery of centchroman

International Journal of Biological Macromolecules 38 (2006) 272–283

Preparation and characterization of sodium hexameta phosphatecross-linked chitosan microspheres for controlled and sustained

delivery of centchroman

K.C. Gupta ∗, Fawzi Habeeb JabrailPolymer Research Laboratory, Department of Chemistry, Indian Institute of Technology, Roorkee 247 667, India

Received 14 October 2005; received in revised form 4 March 2006; accepted 8 March 2006Available online 14 March 2006

Abstract

The cross-linked microspheres using chitosan with different molecular weights and degree of deacetylation have been prepared in presenceof sodium hexameta polyphosphate (SHMP) as physical cross-linker. The degree of cross-linking through electrostatic interactions in chitosanmicrospheres has been evaluated by varying the charge density on chitosan and varying degree of dissociation of sodium hexameta polyphosphate bysaTmpfasido©

K

1

ahitoupsu

0d

olution pH. The degree of deacetylation and molecular weight of chitosan has controlled electrostatic interactions between hexameta polyphosphatenions and chitosan, which played significant role in swelling, loading and release characteristics of chitosan microspheres for centchroman.he microspheres prepared by hexameta polyphosphate anions cross-linker were compact and more hydrophobic than covalently cross-linkedicrospheres, which has been attributed to the participation of all amino groups of chitosan in physical cross-linking with added hexameta

olyphosphate anions. The microspheres prepared under different experimental conditions have shown an initial step of burst release, which wasollowed by a step of controlled release for centchroman. The extent of drug release in these steps has shown dependence on properties of chitosannd degree of cross-linking between chitosan and added polyanions. The degree of swelling and release characteristics of microspheres was alsotudied in presence of organic and inorganic salts, which shown significant effect on controlled characteristics of microspheres due to variations inonic strength of the medium. The initial step of drug release has followed first order kinetics and become zero order after attaining an equilibriumegree of swelling in these microspheres. The microspheres prepared using chitosan with 62% (w/w) degree of deacetylation and molecular weightf 1134 kg mol−1 have shown a sustained release for centchroman for 50 h at 4% (w/w) degree of cross-linking with SHMP.

2006 Elsevier B.V. All rights reserved.

eywords: Centchroman; Chitosan; Hexameta polyphosphate; Controlled drug release

. Introduction

Chitosan is a naturally occurring cationic polyelectrolyte [1]nd shown gel-forming properties [2] in presence of polyanions;ence it is in great demand in pharmaceutical [3–5] and foodndustries. The non-toxicity [6,7] and biodegradability [8] of chi-osan has further increased its importance for the development ofral and controlled drug delivery systems [9,10]. Chitosan is sol-ble in acidic conditions due to electrostatic repulsion betweenrotonated amino groups on chitosan [11,12] and also soluble inolution containing monovalent counter ions. However, the sol-bility of chitosan depends on the degree of deacetylation and

∗ Corresponding author. Tel.: +91 1332 285325; fax: +91 1332 273560.E-mail address: [email protected] (K.C. Gupta).

distribution of free amino groups [13] on chitosan backbone.The gelation in chitosan is often achieved by chemical cross-linking with glutaraldehyde [14,15], glyoxal [16,17] and otherreactive cross-linking agents [18] but these chemically cross-linked hydrogels were found to be hydrophobic and eroded withgreat difficulties in physiological conditions. The cross-linkingwith dialdehydes occurred in mild conditions without any aux-iliary additive. The main drawback of dialdehyde cross-linkersis that they are suspected to be toxic in human body [19]. Toavoid the possible toxicity of dialdehydes, the reversible cross-linking has been carried out using macro [20–22] and polyanions[23–26]. The complexation of chitosan in presence of physicalcross-linkers has been found to be simple and mild [27] anddepends on charge density and concentration of counter polyan-ions. The charge density on polyanions depends on solution pH,hence magnitude of electrostatic interactions between chitosan

141-8130/$ – see front matter © 2006 Elsevier B.V. All rights reserved.oi:10.1016/j.ijbiomac.2006.03.013

mailto:[email protected]

dx.doi.org/10.1016/j.ijbiomac.2006.03.013

K.C. Gupta, F.H. Jabrail / International Journal of Biological Macromolecules 38 (2006) 272–283 273

and polyanions vary with solution pH [24]. The application ofdifferent types of polyanions and variations in ionic strengthof medium [28] also provided opportunities to modify electro-static interactions to form network structure in chitosan gels.The behavior of chitosan largely shown dependence on solutionpH [24] and the presence of electrolytes [28]. The addition ofsalt affects cross-linking through variations in electrostatic inter-actions. However, the modification of electrostatic interactionsdepends on charge and size of ionic species added in the micro-spheres. Since the formation of cross-links takes place throughelectrostatic interactions between positively charged chitosanand added polyanions, hence the degree of physical cross-linkingwould also dependant on degree of deacetylation and molecularweight of chitosan [29,30]. The chitosan with different degreeof deacetylations has shown variations in electrostatic interac-tions due to the variations in charge density. To avoid toxiceffect of dialdehyde as cross-linkers and their unwanted sidereactions with drugs, the cross-linked beads have been reportedusing tripolyphosphate [31] as physical cross-linker. The phys-ical cross-linkers have shown poor mechanical strength due toweak electrostatic interactions between anions and chitosan. Toimprove the mechanical strength and other properties of micro-spheres, various ionic cross-linkers have been tried and out ofthem sodium hexameta polyphosphate (SHMP) has been foundto be useful cross-linker. It is interesting to mention that noreport is available in the literature for a physically cross-linkeddcpstasasa

2

2

pwrfiwarpTr

2

v

ing equation:

[η]25 ◦C = 1.81 × 10−3 cm3 g−1M̄0.93v (acetic acid) (1)

2.3. Deacetylation in chitosan

To obtain chitosan with different degree of deacetylations,the heterogeneous deacetylation was carried out refluxing 0.5 gchitosan sample (DDA, 48%, w/w) in 50 ml 40% (w/w) solu-tion of sodium hydroxide [32] at 80 ◦C for about 4 h. To obtainchitosan with high degree of deacetylation, the original chitosanwas also refluxed for 8 h. These samples were finally washedwith hot and cold water and dried at 30 ◦C in vacuum oven.

2.4. Elemental analysis of chitosan

To determine the degree of deacetylation in original and alkalitreated chitosan samples, the elemental analysis was carried outusing Haraeus Carlo Ebra 1108 Elemental Analyzer.

2.5. FT-IR spectra of chitosan and microspheres

The FT-IR spectra of chitosan samples and sodium hexam-eta polyphosphate cross-linked microspheres were recorded onKBr Pellets using Perkin-Elmer 1600 FT-IR Spectrophotome-ter. The spectra were also recorded for microspheres loaded withc

2s

m

D

whc

t3ama

D

2S

amtw4

elivery system with hexameta polyphosphate anions as physicalross-linker. The high charge density on hexameta polyphos-hate anions has been found to be responsible in providingtrong electrostatic interactions with chitosan in comparisono reported polyphosphate anions, hence in these investigationsn effort has been made to prepare SHMP cross-linked micro-pheres using chitosan with different degree of deacetylationnd molecular weights. Finally, the SHMP cross-linked micro-pheres were also characterized for degree of swelling, loadingnd controlled release characteristics of centchroman.

. Experimental

.1. Materials and methodology

Chitosan was obtained from Sigma Aldrich Chemical Com-any, USA and purified by dissolving in acetic acid (2%,/w). The impurities and undissolved fraction of chitosan were

emoved passing the solution through 0.8 �m cellulose nitratelter under pressure. The chitosan solution was precipitatedith NaOH (1.0 M) and washed with 1:1 mixture of water

nd ethanol. The sodium hexameta polyphosphate (SHMP) waseceived from Ranbaxy Ltd., India and used without furtherurification. The centchroman was received as gift sample fromorrin Pharmaceuticals Ltd., Ahmedabad, India and used afterecrystallization.

.2. Determination of molecular weight of chitosan

The molecular weights of chitosan samples were determinediscometrically using Ubbelohde viscometer and using follow-

entchroman.

.6. Determination degree of deacetylation in chitosanamples

The degree of deacetylation in chitosan samples was deter-ined potentiometrically [33] using following equation:

DA (%) = 203Q

1 + 42Q× 100, Q = N�V

m(2)

here m is the mass of chitosan sample, N the strength of sodiumydroxide used for titration, and �V is the volume of alkalionsumed in titration of amino groups present in the chitosan.

The degree deacetylation was further verified by elemen-al analysis [33] and by recording absorbance at 1655 and450 cm−1 in infrared spectra of chitosan samples [34] formide-I and OH groups. The following equation was used to esti-ate the degree of deacetylation in chitosan using absorbance

t 1655 and 3450 cm−1:

DA (%) =[

1 − A1655/A3450

1.33

]× 100 (3)

.7. Turbidimetric titration of mixture of chitosan andHMP

To determine the pH of maximum interactions between hex-meta polyphosphate anions and amino groups of chitosan, theixture of sodium hexameta polyphosphate (0.1 g l−1) and chi-

osan (0.2 g l−1) was dissolved in 20 ml 0.01 M HCl and titratedith 0.01 M NaOH using UV–vis spectrophotometer at λmax20 nm and change in solution pH was recorded. To determine

274 K.C. Gupta, F.H. Jabrail / International Journal of Biological Macromolecules 38 (2006) 272–283

the degree of dissociation, the sodium hexameta polyphosphate(SHMP) was dissolved in excess solution of 0.1 M HCl and backtitrated with alkali.

2.8. Preparation of chitosan microspheres andcross-linking by SHMP

To prepare chitosan microspheres, a calculated amount of chi-tosan (0.5 g) was dissolved in 200 ml 2% (w/w) acetic acid solu-tion under vigorous stirring for about 3 h at room temperature.The viscous solution of chitosan was blown through a nozzle asfine droplets into a vessel containing 250 ml methanolic solutionof NaOH (0.1 M) as precipitant and hexameta polyphosphatecross-linked chitosan microspheres were prepared by blow-ing chitosan solution in a vessel containing SHMP (4%, w/w)maintained at pH 5.4. Similar method was applied to preparemicrospheres using chitosan with different molecular weightsand degree of deacetylations. The degree of cross-linking inmicrospheres was controlled taking different concentrations ofsodium hexameta polyphosphate in the solution at pH 5.4.

2.9. Size and morphological studies of cross-linkedmicrospheres

To determine the size and morphology of chitosan micro-sws

2m

pmfBttmtopstctw

2

up1c

initial weight of chitosan (W0) was recorded:

Sw(%) =(

Wt − W0

W0

)× 100 (4)

2.12. Loading of centchroman on cross-linked microspheres

The loading of centchroman on chitosan–sodium hexametapolyphosphate cross-linked microspheres was carried out keep-ing 100 mg microspheres in 20 ml buffered solution (pH 4.0) ofcentchroman for about 24 h. The loading in microspheres wascarried out taking different amount of centchroman ranging from10 to 100 mg in the solution. The amount of centchroman loadedwas estimated by recording solution absorbance (λmax = 275 nm)using Shimadzu UV–vis 1601 PC Spectrophotometer.

2.13. Release of centchroman from microspheres

The drug release behavior of chitosan and sodium hexametapolyphosphate (SHMP) cross-linked microspheres was deter-mined keeping centchroman loaded microspheres in a releasemedia (pH 7) and the amount of centchroman released at differ-ent time intervals was recorded at 275 nm (λmax) with replace-ment using Shimadzu UV–vis 1601 PC Spectrophotometer. Theamount of drug released was reported as release ratio (%Rf) forfixed time interval (10 h) of drug release:

r

wv

re

b

c

wao

2

tav

3

(hv

pheres, the scanning electron micrographs of microspheresere recorded after mounting on metal studs with double adhe-

ive tape and vacuum coating with gold.

.10. Surface hydrophobicity of cross-linked chitosanicrospheres

The surface hydrophobicity of chitosan and sodium hexametaolyphosphate cross-linked chitosan microspheres was deter-ined keeping 100 mg microspheres of different surface area

or 2 h, in 20 ml solution (0.1 M) of hydrophobic dye (Roseengal). The volume of dye adsorbed per unit area of chi-

osan microspheres was taken from the slope drawn betweenhe ratio (Q) of volume of dye adsorbed (Vad) per unit area of

icrospheres to the volume of dye taken initially for adsorp-ion (Vin) versus area of microspheres (A) available per mlf dye taken in the vessel. The volume of hydrophobic dyeartitioned on these microspheres was used as a measure ofurface hydrophobicity of chitosan microspheres. To comparehe hydrophobicity of SHMP cross-linked microspheres withhemically cross-linked chitosan microspheres, the dye adsorp-ion was also carried out on chitosan microspheres cross-linkedith 4% (w/w) glutaraldehyde.

.11. Degree of swelling in cross-linked microspheres

The loading and release behavior of microspheres dependspon the degree of swelling, hence degree of swelling in thesehysically cross-linked microspheres was determined keeping00 mg microspheres in 20 ml solution (pH 3) for 24 h and per-ent increase in weight of microspheres (Wt) with respect to

elease ratio (%Rf) = Wt

W0× 100 (5)

here Wt is the amount of drug released within fixed time inter-al (10 h) and W0 as amount of drug loaded on microspheres.

The drug released from microspheres is expressed as burstelease (%RB) and controlled release (% RC) using followingquations:

urst release (%RB) =∑ (

Wt

W0× 100

)t

(6)

ontrolled release (%RC) =∑ (

Wt

W0× 100

)t

(7)

here ((Wt/W0) × 100)t is the variable for burst release stepnd constant for controlled release step for a fixed time intervalf drug release.

.14. Effect of solution pH

The degree of ionization of physical cross-linker shows varia-ion with solution pH, hence the effect of solution pH on loadingnd release characteristics of microspheres was determined byarying solution pH from 1 to 8.

. Results and discussion

The degree of deacetylation (DDA) and molecular weightM̄v) of chitosan influence the solution property of chitosan,ence these characteristics of chitosan have been consideredery significant in formulation of controlled delivery systems


using chemical and physical cross-linkers. The intermolecularinteractions in chitosan have shown dependence on molecu-lar weight [36]. The degree of deacetylation in chitosan [35].The degree of deacetylation has shown a significant control onphysico-chemical properties of chitosan due to variable numberof amino groups on chitosan back bone. The number and distri-butions of amino groups on chitosan backbone has ultimatelycontrolled the physical cross-linking in the chitosan micro-spheres. To study the effect of molecular weight of chitosanon these interactions, the molecular weights of chitosan sam-ples were determined viscometrically (Eq. (1)) and found to be760 kg mol−1, 1134 kg mol−1 and 2224 kg mol−1, which werecategorized as low (L MWt), medium (M MWt) and high molec-ular weight (H MWt) chitosans, respectively. These chitosansamples were further characterized for degree of deacetylationand finally used in preparation of microspheres for controlleddelivery of non-steroidal contraceptive drug in a controlled andsustained manner.

To obtained chitosan samples with different degree ofdeacetylation (DDA), the medium molecular weight chitosansample (1134 kg mol−1) was deacetylated for 4 and 8 h in pres-ence of alkali (40%, w/w) at 80 ◦C so that depolymerization inchitosan was avoided. The molecular weight of deacetylated chi-tosan found to be constant (1134 kg mol−1), which has given anindication that no depolymerization and degradation of chitosanwas taken place during treatment of chitosan with alkali.

3

tm(tcim

sm

Fig. 1. FT-IR spectra of pure chitosan (A), SHMP cross-linked chitosan (B) andcentchroman loaded SHMP cross-linked (C) microspheres.

obtained by infrared spectrum was considered more reliable thanother methods. The degree of deacetylation in these chitosansamples was found to be 48%, 62% and 75%, w/w and thesesamples were categorized as low (L DDA), medium (M DDA)and high degree of deacetylated chitosan (H DDA) and subse-quently used in the preparation of microspheres for loading andrelease of centchroman.

3.2. Solution pH and electrostatic interactions

In fabrication of physically cross-linked chitosan micro-spheres, the electrostatic interactions between chitosan andcounter ions should be maximum. The chitosan and sodium

TD

C method (%, w/w) DDA by elemental method of analysis

C (%, w/w) N (%, w/w) DDA (%, w/w)

S 48.36 8.03 48.80S 35.35 6.22 61.98S 28.56 5.13 75.03

M 0% (w/w) alkali for 4 h (Sample-2) and 8 h (Sample-3).

.1. Characterization of sample for degree of deacetylation

The degree of deacetylation in original and alkali treated chi-osan samples was determined by potentiometric [33] titration

ethod (Eq. (2)) and by using the ratio of absorbance of amide-I1655 cm−1) to hydroxyl group (3450 cm−1) in infrared spec-rum [34] of chitosan samples (Fig. 1A) and using Eq. (3). Thealculated degree of deacetylation in chitosan samples is givenn Table 1. The degree of deacetylation determined by these

ethods was further authenticated by elemental method [33] as

hown in Table 1. The degree of deacetylation of chitosan deter-ined by these methods was almost same. However, the value

able 1etermination of degree of deacetylation (DDA) in chitosan

hitosan samples DDA by potentiometric (%, w/w) DDA by IR

ample-1 48.0 48.0ample-2 61.8 62.0ample-3 74.6 75.0

¯ v = 1134 kg mol−1 of original chitosan (Sample-1) and chitosan treated with 4


Fig. 2. Turbiditymetric titration and degree of ionization curves for chitosan andsodium hexameta polyphosphate.

hexameta polyphosphate (SHMP) have shown variations indegree of ionization with solution pH. The chitosan is a weakbase and ionized below pH 5 (Fig. 2), whereas, sodium hexametapolyphosphate (SHMP) has shown a decreasing trend in degreeof ionization below pH 5 (Fig. 2). These variations in degree ofionization with solution pH have controlled electrostatic inter-actions between chitosan and hexameta polyphosphate anionsdue to the variations in charge number of hexameta polyphos-phate anions and chitosan. These interactions were verifiedby turbiditymetric titration of hexameta polyphosphate andchitosan mixture (Fig. 2). The solution was optically clear belowpH 3 (Fig. 2) as electrostatic interactions between chitosan andsodium hexameta polyphosphate were minimum. The chitosanwas ionized below pH 3 but hexameta polyphosphate was poorlyionized but as the solution pH increased (Fig. 2), the turbidityhas started to increase slowly. But above pH 5, the solutionturbidity was increased greatly, which was due to the significant

increase in electrostatic interactions between chitosan andhexameta polyphosphate anions. The turbidity has decreasedabove pH 6 due to the decrease in electrostatic interactionsbetween chitosan and hexameta polyphosphate anions becausethe degree of ionization of chitosan was decreased (Fig. 2).This variation in electrostatic interactions between chitosan andhexameta polyphosphate anions with solution pH has clearlyindicated that chitosan microspheres would be cross-linkedstrongly, if they are prepared within a pH range of 4.5–6.0,otherwise they would be weakly cross-linked. The microspheresobtained at high pH were having low amount of hexametapolyphosphate anions and the same was washed out with water.

3.3. Preparation of microspheres and their characterization

Since the solution pH has influenced the degree of ionizationof sodium hexameta polyphosphate and chitosan, hence chitosanmicrospheres were physically cross-linked in slightly acidicsolution (pH 5.4) of hexameta polyphosphate so that hexametapolyphosphate anions diffused in microspheres and producedstrong electrostatic attractions between positively charged chi-tosan and hexameta polyphosphate anions. The microsphereswere kept for 6 h in solution of sodium hexameta polyphosphateso that sufficient amount of hexameta polyphosphate anions wasdiffused into chitosan microspheres to form physical cross-linksaIttasiudhaw(6a

Table 2Physical characteristics of SHMP cross-linked chitosan microspheres

Types and composition of microspheres

Type of microspheres M̄v of chitosan(kg mol−1)

DDA of chitosan(%, w/w)

C(%

CHP-1 760 62CHP-2 1134 62CCCCCC 1C

S osan (g

HP-3 2224 62HP-4 1134 48HP-5 1134 75HP-6 1134 62HP-7 1134 62HP-8 1134 62H-9 1134 62

odium hexameta polyphosphate (SHMP), degree of deacetylation (DDA), chitlutaraldehyde cross-linked chitosan microspheres (0.036 ml �m−2).

nd to maintain electro-neutrality within these microspheres.n physical cross-linking, the maximum amino groups of chi-osan were consumed in formation of cross-links in comparisono covalently cross-linked chitosan microspheres, hence freemino groups were absent in physically cross-linked micro-pheres. Since physical cross-linking depends on electrostaticnteractions, hence chitosan microspheres were also preparedsing chitosan of different molecular weights and degree ofeacetylation (Table 2) at constant concentration of sodiumexameta polyphosphate (4%, w/w) and solution pH (5.4). Tonalyze the concentration effect of polyanions, the microspheresere prepared using chitosan with constant molecular weight

1134 kg mol−1), degree of deacetylation (62%, w/w) at pHand varying the concentration of hexameta polyphosphate

nions. To control the electrostatic interactions of hexameta

Size of microspheres(�m)

Degree of hydrophobicity(ml/�m2)

onc. of SHMP, w/w)

4 99.35 0.0584 86.20 0.0594 80.88 0.0604 109.60 0.0524 71.42 0.0752 125.60 0.0466 62.50 0.0662 41.70 0.0700 164.60 0.029

CH) and SHMP cross-linked chitosan microspheres (CHP). Hydrophobicity of


polyphosphate anions, the microspheres were also prepared inpresence of other anions and cations, which were mixed withsolution of chitosan before spraying as droplets in a vesselcontaining solution of hexameta polyphosphate anions main-tained at pH 5.4. The chitosan microspheres in neutral (pH 7)and alkaline solution (pH > 7) were having low charge density;hence electrostatic interactions of added anions and cations withchitosan were minimum. The sodium hexameta polyphosphateanions cross-linked microspheres with added salts were havingdifferent properties than obtained without adding these salts.

3.4. FT-IR and SEM characterization of chitosanmicrospheres

The infrared spectra of chitosan and hexameta polyphosphatecross-linked chitosan microspheres were compared (Fig. 1A andB), which clearly indicated that absorption band at 3454 cm−1

attributed to –NH2 group in chitosan (Fig. 1A) was broadeneddue to physical interactions with hexameta polyphosphate anda shoulder was appeared at 1655 cm−1 (Fig. 1A) due to chi-tosan amide at same position (Fig. 1B) after cross-linking withSHMP, which indicated for interactions of chitosan amide withadded polyanions. The observed absorption bands at 1278 and1103 cm−1 have been assigned to –P O groups of polyphos-phate anion. In cross-linked microspheres, the –NH2 bendingv −1 −1

d–

msci

chitosan (Table 2 and Fig. 3A). The morphological investiga-tions have indicated that microspheres after cross-linking withhexameta polyphosphate anions were smooth (Fig. 3D) in com-parison to pure chitosan microspheres (Fig. 3B). The size andmorphology of microspheres varied (Table 2) on varying the con-centration of hexameta polyphosphate anions due to variationsin electrostatic interactions between hexameta polyphosphateanions and chitosan. The variation in molecular weight anddegree of deacetylation of chitosan has also shown variationin size and morphology of the microspheres.

3.5. Hydrophobicity in chitosan microspheres

The hexameta polyphosphate anion cross-linking hasincreased hydrophobicity in microspheres (0.05 ml �m−2) incomparison to pure chitosan (0.029 ml �m−2) microspheres(Table 2 and Fig. 4). The hydrophobicity of chitosan micro-spheres was expressed as volume of hydrophobic dye (Rose Ben-gal) adsorbed per unit area of the microspheres. The hydropho-bicity of microspheres depends on the value of slope of the plotdrawn (Fig. 4) between dye adsorption quotient (Q) versus dyetaken per unit area for adsorption in the vessel, hence higher wasthe slope, higher was the hydrophobicity in the microspheres.The hydrophobicity has increased on increasing the degree ofdeacetylation in chitosan microspheres from 48% (w/w) to 75%(tCiptra

inked

ibration was observed at 1630 cm in place of 1590 cmue to the interactions of hexameta polyphosphate anions withNH3

+ ions of chitosan [37].The size and morphology of chitosan microspheres was deter-

ined with scanning electron micrographs (Fig. 3). The micro-pheres prepared with hexameta polyphosphate anions wereompact and smaller in size (86.2 �m) (Table 2 and Fig. 3C)n comparison to microspheres (164.6 �m) prepared with pure

Fig. 3. Scanning electrons micrographs of pure (A) and cross-l

w/w) due to the increase in degree of physical cross-linkinghrough electrostatic interactions in the microspheres (CHP-4,HP-2, and CHP-5 in Table 2). However, it remained constant

n microspheres (CHP-1, CHP-2 and CHP-3 in Table 2) pre-ared with different molecular weights of chitosan. On varyinghe molecular weight of chitosan, the charge density on chitosanemained constant; hence the magnitude of electrostatic inter-ctions between hexameta polyphosphate anions and chitosan

chitosan (C) microspheres and their morphologies (B and D).


Fig. 4. Degree of hydrophobicity in pure chitosan and SHMP cross-linked chi-tosan microspheres.

remained constant without varying the surface hydrophobic-ity of microspheres. The degree of hydrophobicity has showndependence on cross-linking, which is due to the variationin electrostatic interactions between chitosan chains (CHP-6,CHP-2, CHP-7, and CHP-8 in Table 2). The hydrophobicityof glutaraldehyde cross-linked chitosan was found to be low(0.036 ml �m−2) in comparison to SHMP cross-linked micro-spheres (Table 2).

3.6. Thermal studies of cross-linked microspheres

The degree of swelling, loading and drug release from chi-tosan microspheres shown dependence on types of water [38,39]determined by thermo gravimetric analysis (TGA) of micro-spheres. The chitosan microspheres were having 13.75% (w/w)free water and about 4.25% (w/w) bound water, which was evap-orated at 200 ◦C. In comparison to chitosan microspheres, thehexameta polyphosphate cross-linked microspheres were hav-ing 1.5% (w/w) free water and 1.95% (w/w) bound water. This

has given an indication that hexameta polyphosphate cross-linked microspheres were more hydrophobic in comparisonto pure chitosan and chemically cross-linked chitosan micro-spheres [39].

3.7. Swelling behavior of cross-linked microspheres

In physically cross-linked microspheres, the swelling ratio (%Sw) has shown dependence on charge number of counter ionsand chitosan, which varied with solution pH. However, chargenumber depends on types of anions and types of chitosan usedin fabrication of microspheres. To investigate the effect of thesefactors, the swelling ratio (% Sw) of microspheres was evaluatedin solution of different pH ranging from 1 to 8. The swelling ratioof microspheres was decreased from 485% (w/w) to 70% (w/w)on varying solution pH from 1 to 8. At low pH, the strength ofelectrostatic interactions between added physical cross-linkerand chitosan was low due to the substantial decrease in degreeof dissociation of hexameta polyphosphate, hence at low pH,the high degree of swelling was due to the decrease in chargenumber of physical cross-linker (Table 3). On increasing thesolution pH beyond 6, the degree of dissociation of chitosanwas minimum (Fig. 2) and hexameta polyphosphate was fullydissociated. At high pH (>6), the strength of electrostatic inter-actions was low due to the decrease in solubility of chitosan,wa(hcotitppClcpdtm

Table 3Effect of solution pH on swelling, loading and centchroman release from SHMP cros

pH of solution Degree of swelling(%, w/w)

Loading of centchroman(mg/100 mg)

Centc

Burst

1 68.002 56.003 44.004 39.005 42.807 83.308 80.25

M icros

485 30395 34250 58215 62150 42184 36

70 28

¯ v (chitosan) = 1134 kg mol−1, DDA (62%, w/w), SHMP (4%, w/w), type of m

hich has decreased the degree of swelling at high pH (>6)nd has minimized the amount of free water in the microspheresTable 3). The physically cross-linked microspheres have shownigh degree of swelling on varying solution pH in comparison tohemically cross-linked microspheres. The variation in degreef swelling in chemically cross-linked microspheres was dueo the surface charge density on chitosan molecules, whereasn physically cross-linked microspheres, the variations was dueo surface charge on chitosan and degree of hydration of addedhysical cross-linkers. The degree of swelling in microspheresrepared with chitosan of different molecular weights (CHP-1,HP-2, and CHP-3, Table 4) and different degree of deacety-

ation (CHP-4, CHP-7, and CHP-5, Table 4) at constant con-entration of physical cross-linker (4%, w/w) was evaluated atH 3, which has clearly indicated that degree of swelling wasecreased (CHP-1, CHP-2, and CHP-3, Table 4) on increasinghe molecular weight (Table 4) and degree of deacetylation in

icrospheres (CHP-4, CHP-1 and CHP-5, Table 4). The vari-

s-linked microspheres

hroman release Diffusion constant(D)/10−12 cm2 s−1

step (% RB, w/w) Controlled step (% RC, w/w)

32.00 7.66944.00 2.84654.00 1.03061.00 0.91257.20 0.44716.70 0.32819.75 1.463

pheres (CHP-2).


Table 4Effect of molecular weight (M̄v), degree of deacetylation (DDA) and cross-linking on swelling, loading and release of centchroman from microspheres

Type of microspheresa Degree ofswelling (%, w/w)


Centchroman release Diffusion constant(D)/10−12 cm2 s−1

Burst step(% RB, w/w)

Controlled step(% RC, w/w)

CHP-1 285.0 48.0 46.3 53.7 0.912CHP-2 250.0 62.0 39.0 61.0 0.557CHP-3 140.0 26.0 64.0 36.0 6.500CHP-4 316.0 53.0 52.0 48.0 0.507CHP-5 97.0 21.0 84.0 16.0 6.978CHP-6 280.0 56.0 42.0 52.0 0.448CHP-7 183.5 42.0 82.5 17.5 0.186CHP-8 108.0 28.0 84.0 16.0 15.790CH-9 300.0 15.0 67.3 32.7 2.974

Sodium hexameta polyphosphate (SHMP), degree of deacetylation (DDA), chitosan (CH) and SHMP cross-linked chitosan microspheres (CHP).a Composition as in Table 3.

Table 5Effect of addition of salt and acids on swelling, loading and release of centchroman from microspheres

Microspheres with andwithout additives

Degree ofswelling (%, w/w)


Centchroman release Diffusion constant(D)/10−12 cm2 s−1

Burst step(% RB, w/w)

Controlled step(% RC, w/w)

Reference 250.0 62.0 39.0 61.0 0.912FeCl3 324.0 70.0 60.0 40.0 0.301Na3PO4 645.0 29.0 50.0 50.0 115.240Na2C2O4 571.0 32.7 33.3 66.6 10.210CH3COONa 263.0 60.0 57.0 43.0 0.230

M̄v (chitosan) = 1134 kg mol−1, DDA (62%, w/w), SHMP (4%, w/w), additives (0.2%, w/w).

ation in degree of swelling on increasing the molecular weightand degree of deacetylation was due to the variations in electro-static interactions between chitosan and added physical cross-linker (SHMP). The degree of swelling has also been affectedin presence of added anions and cations, which have modi-fied electrostatic interactions between chitosan and hexametapolyphosphate by shielding and repulsion effect of added ions(Table 5).

3.8. Loading of centchroman on chitosan cross-linkedmicrospheres

The effect of degree of cross-linking and property of chi-tosan on loading capacity of microspheres for centchroman hasbeen evaluated. Centchroman is a non-steroidal drug withoutany ionizable group; hence centchroman was not having sig-nificant electrostatic interactions with cross-linker or chitosan.The size of centchroman might be responsible for its diffusionin chitosan microspheres.

The degree of swelling in microspheres has shown variationswith solution pH, hence loading of centchroman was carried outat different pH and shown significant variations in loading ofcentchroman per 100 mg of microspheres in solution of differentpH (Table 3). The loading was increased on increasing the solu-tion pH below 4, which was due to the increase in electrostaticinteractions between ionized chitosan and hexameta polyphos-phate anions. These interactions were able to provide optimumintermolecular separation, which allowed the retention of drugdiffused from solution phase to the microspheres and these inter-actions were optimum at pH 4 (Table 4). At this pH, the loadingof centchroman was found to be maximum (62%, w/w). The fur-ther increase in solution pH (>4) there was a decreasing trend incentchroman loading till pH 8 (28%, w/w). This decreasing trendwas due to the decrease in degree of ionization of chitosan, whichhas decreased the surface positive charge on chitosan molecules,which resulted in precipitation of chitosan at high pH and showna substantial decrease in drug loading (Table 3). The loading ofcentchroman on microspheres prepared with chitosan of differ-ent molecular weights and degree of deacetylation has also beenstudied at constant concentration of hexameta phosphate (4%,w/w). The loading of centchroman has increased initially from48% (w/w) to 62% (w/w) on increasing the molecular weightof chitosan from 760 to 1134 kg mol−1 but on further increas-ing the molecular weight of chitosan beyond 1134 kg mol−1, theloading capacity of microspheres was decreased (Table 4). Thevw
ariation in loading capacity of microspheres with moleculareight of chitosan might be due to the increased compaction in


the microspheres. The variation in electrostatic interactions onvarying the molecular weight of chitosan, was minimum due tothe fixed concentration of hexameta polyphosphate (4%, w/w)and amino groups (62%, w/w) on the chitosan, hence varia-tion in loading of centchroman on varying the molecular weightof centchroman was attributed to the variation in intermolec-ular space in the microspheres. The variation in centchromanloading in microspheres with different degree of deacetylationsand different concentrations of hexameta polyphosphate wasattributed to the variation in electrostatic interactions betweenchitosan and hexameta polyphosphate. These variations in sur-face charge on chitosan and hexameta polyphosphate anionshave controlled the degree of cross-linking through electro-static interactions. The microspheres prepared with 62% (w/w)degree of deacetylation in chitosan (Table 4) at constant con-centration of hexameta polyphosphate (4%, w/w) and molecularweight (1134 kg mol−1), were loaded optimally (62%, w/w) thanmicrospheres prepared with other experimental conditions. Sim-ilarly the microspheres prepared with 4% (w/w) concentrationof hexameta polyphosphate were highly loaded in comparisonto microspheres prepared at other concentrations of hexametapolyphosphate (Table 4). These investigations have clearly indi-cated that loading in these microspheres was affected on varyingcharge density on chitosan with degree of deacetylation or withconcentration of hexameta polyphosphate anions in the micro-spheres (Table 4).

i(tdsmcaa

3m

tbotmaemmRiraomw

Fig. 5. Effect of solution pH on release ratio of centchroman from SHMP cross-linked chitosan microspheres. Degree of cross-linking = 4% (w/w). Releasemedia = 100 mg microspheres in 20 ml buffered solution (pH 7), temperature25 ◦C.

release of drug in the system and this release is called as burstrelease (%RB). After attainment of structural equilibrium, micro-spheres have started releasing a constant amount of centchromanwithin fixed internal of time. This step of drug release has beencalled as controlled step of drug release. The duration and theamount of drug released (% RC) have shown dependence on theresponse of microspheres with release media and the specifi-cations of the microspheres. To investigate these responses ofmicrospheres for the release of centchroman, the microspheres(100 mg) were kept in solution of different pH (Table 3 andFig. 5) and the amount of drug released from these microsphereswas estimated. The drug release data (Table 3 and Fig. 5) haveclearly indicated that the percent of centchroman released inburst release step was decreased from 68% (w/w) to 39% (w/w)upto pH 4 and at high pH (>4), the amount of centchromanreleased in burst release step was increased. The centchromanreleased in controlled step of drug release has shown a oppo-site trend to the burst release step on increasing the solutionpH (Table 3 and Fig. 5). The microspheres at pH 4 have shownmaximum release of centchroman (61%, w/w) within a period of50 h in controlled step of drug release, which is due to the opti-mal degree of swelling and electrostatic interactions betweenhexameta polyphosphate anions and chitosan, whereas at lowpH (<4), the high degree of swelling in microspheres has facili-tated the maximum release of centchroman in burst step of drugrelease (Fig. 5). The effect of molecular weight and degree ofdslttet

The effect of electrostatic interactions on centchroman load-ng was also evaluated in presence of organic and inorganic saltsTable 5). In presence of these salts, the loading was increasedo 70% (w/w) but with trivalent anion (PO4

−3), the loading wasecreased to 29% (w/w). The addition of sodium oxalate andodium acetate has also shown variations on loading of centchro-an (Table 5) but the decrease in loading capacity was low in

omparison to sodium triphosphate (PO43−), which has been

ttributed to enhanced repulsion between hexameta phosphatenions and added phosphate anion (Table 5).

.9. Release behavior of cross-linked chitosanicrospheres

The effect of molecular weight and degree of deacetyla-ion of chitosan on percent release of centchroman was studiedecause these factors have played a significant role in degreef swelling and loading behavior of microspheres. Similarlyhe pH of release media has controlled the release pattern of

icrospheres by modifying the degree of swelling through vari-tions in electrostatic interactions between chitosan and hexam-ta phosphate anions. The release pattern of centchroman fromicrospheres was studied at different pH (Table 3) and centchro-an was released in two steps. In first step of drug release (%B), there was an increasing trend in the amount of drug released

n a fixed interval of release time (10 h) and in this step of drugelease, the chitosan chains were relaxed and degree of swellingnd drug release continued to increase. Therefore, in this stepf drug release, the structural changes have taken place in theicrospheres. Since the amount of drug released in this stepas not constant, hence not useful for sustained and controlled

eacetylation of chitosan on release of centchroman was alsotudied and shown in Table 4. The microspheres prepared withow (>760 kg mol−1, CHP-1) and high molecular weight chi-osan (2224 kg mol−1, CHP-3) have released a maximum frac-ion of loaded centchroman in burst step of drug release (% RB)xcept medium molecular weight (1134 kg mol−1, CHP-2) chi-osan microspheres (Table 4 and Fig. 6). In these microspheres


Fig. 6. Effect of molecular weight of chitosan on release ratio of centchromanfrom SHMP cross-linked microspheres in 20 ml buffered solution (pH 7), tem-perature 25 ◦C.

(CHP-2), the interactions between hexameta polyphosphateanions and chitosan molecules were suitable to create optimumswelling in the microspheres, whereas, low molecular weightmicrospheres (CHP-1) have shown high degree of swelling,hence released a sufficient amount of loaded centchroman (46%,w/w) in burst step of drug release. In contrast to these micro-spheres, the high molecular weigh microspheres (CHP-3) werecompact, hence equilibrium swelling was prolonged (70 h) andmaximum centchroman (64%, w/w) was released in burst stepof drug release (Table 4 and Fig. 6). These investigations haveclearly indicated that the increase in molecular weight of chi-tosan beyond 1134 kg mol−1, a negative effect on controlled stepof drug release was observed, whereas, microspheres with lowmolecular weight of chitosan (760 kg mol−1) have released max-imum loaded drug in burst release manner due to high degree ofswelling in these microspheres.

The microspheres prepared, using chitosan with differentdegree of deacetylation (CHP-4, CHP-2 and CHP-5) and con-stant concentration of hexameta polyphosphate (4%, w/w) andmolecular weight of chitosan (1134 kg mol−1) were used to ana-lyze the effect of charge density of chitosan on drug release atpH 4. The data shown in Table 4 and Fig. 7 have clearly indicatedthat the degree of deacetylation in chitosan has influenced therelease pattern of centchroman from these microspheres. Themicrospheres prepared with chitosan of low degree of deacety-lation (CHP-4) have released 52% (w/w) of loaded drug in burstsiot

Fig. 7. Effect of degree of deacetylation in chitosan on release ratio of centchro-man from SHMP cross-linked microspheres in 20 ml buffered solution (pH 7),temperature 25 ◦C.

microspheres prepared using chitosan with 75% DDA (CHP-5) were more compact due to the strong electrostatic interac-tions between hexameta polyphosphate anions and chitosan.These microspheres (Table 4, CHP-5) have released centchro-man (Fig. 7) in a similar fashion as found with microspheres(CHP-3) prepared with high molecular weight chitosan (Fig. 6).Since in both cases, the initial step of drug release was prolongedand have released maximum drug in this step of burst release,hence found insignificant for controlled delivery of centchro-man. The microspheres (CHP-2) prepared using chitosan with62% (w/w) DDA have released a significant amount of centchro-man (61%, w/w) within a period of 50 h in controlled manner incomparison to microspheres prepared with low (CHP-4, 48%,w/w) and high (CHP-6, 75%, w/w) degree of deacetylation inchitosan (Table 4 and Fig. 7).

The effect of electrostatic interactions on centchromanrelease was studied using microspheres prepared with differ-ent concentrations of hexameta polyphosphate anions (CHP-6,CHP-4, CHP-7 and CHP-8) using chitosan with constant molec-ular weight (1134 kg mol−1) and constant degree of deacetyla-tion (62%, w/w). The centchroman released from these micro-spheres was studied at pH 4. The data shown in Table 4 and Fig. 8have clearly indicated that the drug release steps were influ-enced significantly with concentration of hexameta polyphos-phate anions. The microspheres prepared at low (<4%, w/w)and high (>4%, w/w) concentration of hexameta polyphosphatehsrd4(C

tep of drug release of 30 h and 48% (w/w) drug was releasedn controlled step of drug release of 20 h. Although, the amountf centchroman released in this step was sufficient (48%) buthe duration of controlled release was very short (20 h). The

ave released a significant amount of centchroman in initialtep of burst release and relatively a small amount of drug waseleased in controlled step of drug release. The burst releaseuration was short in microspheres (Table 4 and CHP-2, CHP-) prepared at low concentration of hexameta phosphate anions<4%, w/w) and it was longer in microspheres (Table 4 andHP-7, CHP-8) prepared with high concentration of cross-linker


Fig. 8. Effect of degree of deacetylation on release ratio of centchroman fromSHMP cross-linked microspheres in 20 ml buffered solution (pH 7), temperature25 ◦C.

(>4%, w/w), which was due to the difference in degree ofswelling in these microspheres (Table 4 and Fig. 8). The micro-spheres prepared with 4%, w/w of hexameta polyphosphate(CHP-2) have released a significant amount of centchroman(61%, w/w) in a controlled step of 50 h, hence found suitablefor controlled and sustained delivery system. These variationshave clearly indicated that the release pattern of centchromanwas dependant on electrostatic interactions, which ultimatelywere influenced by concentration of hexameta polyphosphateanions and characteristics of chitosan. The effect of addition ofsalts was studied on release pattern of centchroman (Table 5and Fig. 9), which clearly indicated that added cations (Fe3+)and anions (PO4

3−, C2O42− and Ch3COO−) have significantly

influenced the steps of drug release due to the modificationsin electrostatic interactions between hexameta polyphosphateanions and chitosan. The higher the negative charge on addedanions, the faster was the burst release step and reverse was truewith added cations of higher valency (Fe3+).

To determine the mechanism of drug release from micro-spheres prepared using chitosan with different degree of deacety-lation, molecular weight and different concentrations of hex-ameta polyphosphate anions, the plots were drawn betweenfractional drug release (Mt/M�) versus square root of releasetime (

√t), which shown a complete agreement with drug release

model [40], indicating a non-Fickinan behavior of drug releasefrom these microspheres. The deviation from linear trend atlvsbwm

Fig. 9. Effect of added salt (0.1%, w/w) on release ratio of centchroman fromSHMP cross-linked microspheres in 20 ml buffered solution (pH 7).

Fig. 10. Effect of degree of deacetylation on fractional release of centchromanfrom cross-linked microspheres. pH 7, temperature = 25 ◦C.

released from these microspheres has initially followed firstorder kinetics and become zero order after attainment of equi-librium swelling in the microspheres [42].

4. Conclusion

The sodium hexameta polyphosphate has been used suc-cessfully to prepare physically cross-linked microspheres usingchitosan with different degree of deacetylation and molecu-lar weights. On varying the solution pH and using chitosanof different degree of deacetylation, the strength of electro-

ater stage of drug release (Fig. 10) was attributed to structuralariations in the microspheres after attainment of equilibriumwelling. The similar behavior of drug release was also reportedy other workers [41]. The slope of linear portion of these plotsas used to determine the diffusion constant (D) of centchro-an from the microspheres as reported in Tables 3–5. The drug


static interactions was controlled significantly. The micro-spheres prepared with hexameta polyphosphate were strongerthan sodium tripolyphosphate reported in the literature. Thedegree of swelling and controlled characteristics of micro-spheres has shown dependence on solution pH. The hexametapolyphosphate anions have shown strong electrostatic interac-tions with ionized chitosan, which provided sufficient oppor-tunities to control the drug release pattern of centchroman ina much better way than reported with other phosphate cross-linkers. The addition of inorganic and organic salts has providedfurther opportunities to control the release behavior of chitosanmicrospheres by modifying electrostatic interactions betweenhexameta polyphosphate anions and chitosan.

Acknowledgements

Authors are thankful to I.I.T Roorkee, India, for providingresearch facilities to carryout these investigations. One of theauthors Mr. Fawzi Habeeb Jabrail is thankful to ICCR NewDelhi, Government of India for awarding a fellowship undercultural exchange programme.

References

[1] T. Yoshizawa, Y. Shin-Ya, H. Kyung-Jin, T. Kajiuchi, Eur. J. Pharmaceut.Biopharmaceut. 59 (2005) 307–313.

[

[11] T.W.G. Solomon, Organic Chemistry, 4th ed., Wiley, New York, 1988,p. 887.

[12] M. Rinaudo, G. Pavlov, J. Desbrieres, Polymer 40 (1999) 7029–7032.[13] R.A.A. Muzzarelli, Chitin, Pergamon Press, New York, 1977.[14] A.A. Al-Helw, A.A. Al-Angary, G.M. Mohrous, M.M. Al-Dardari, J.

Microencapsul. 15 (1998) 373–382.[15] W. Arguelles-Monal, F.M. Goycoolea, C. Peniche, I. Higuera-Ciapra,

Polym. Gels Networks 6 (1998) 429–440.[16] S. Nakatsuka, A.L. Andrady, J. Appl. Polym. Sci. 44 (1992) 17–28.[17] M.N. Khalid, L. Ho, J.L. Agnely, J.L. Grossiord, G. Courraze, STP

Pharm. Sci. 9 (1999) 359–364.[18] A.A. De Agnelis, D. Capitani, V. Crescenzi, Macromolecules 31 (1998)

1595–1601.[19] H.W. Leung, Ecotoxicol. Environ. Saf. 49 (2001) 26–39.[20] M.L. Gonzalez-Rodriguez, M.A. Holgado, C. Sanchez-Lafuente, A.M.

Rabasco, A. Fini, Int. J. Pharm. 3 (2002) 225–234.[21] T.H. Kim, Y.H. Park, K.J. Kim, C.S. Cho, Int. J. Pharm. 250 (2003)

371–383.[22] P. Calvo, C. Remunan-Lopec, J.L. Vella-Jata, M.J. Alonso, J. Appl.

Polym. Sci. 63 (1997) 125–132.[23] X.Z. Shu, K.J. Zhu, Int. J. Pharm. 201 (2000) 51–58.[24] X.Z. Shu, K.J. Zhu, J. Microencapsul. 18 (2001) 237–245.[25] C. Aral, J. Akbuga, Int. J. Pharm. 168 (1998) 9–15.[26] C. Valenta, B. Christen, A. Bernkop-Schurch, J. Pharm. Pharmacol. 50

(1998) 445–452.[27] S. Dumitriu, E. Chornet, Adv. Drug Del. Rev. 31 (1998) 223–246.[28] M.L. Tsaih, R.H. Chen, J. Appl. Polym. Sci. 73 (1999) 2041–2051.[29] X. Yongmei, D. Yumi, Int. J. Pharm. 250 (2003) 215–226.[30] S.S. Sabnis, L.H. Block, Int. J. Biomacromol. 27 (2000) 181–186.[31] E. Esposito, R. Cortesi, C. Mastnizzi, Biomaterial 17 (1996) 2004–2020.[32] S. Takanori, K. Keisuke, I. Yoshio, Die Makromol. Chem. 1 (1976)

[[

[

[[

[[[[[

[2] J. Berger, M. Reist, J.M. Mayer, O. Felt, R. Gurmy, Eur. J. Pharm.Biopharm. 5 (2004) 35–42.

[3] V. Dodane, V.D. Vilivalam, Pharm. Sci. Technol. Today 1 (1998)246–258.

[4] C. Peniche, W.A. Monal, H. Peniche, N. Acasta, Macromol. Biosci. 3(2003) 511–520.

[5] R. Hizazi, M. Amiji, J. Control. Release 89 (2003) 151–165.[6] J. Kristi, J. Smid-Korbar, E. Strue, M. Schera, H. Rupprecht, Int. J.

Pharm. 99 (1993) 13–19.[7] T.J. Aspden, J.D. Mason, N.S. Jones, J. Pharm. Sci. 86 (1999) 509–513.[8] D. Koga, in: R. Chen, H.C. Chen (Eds.) Adv. Chitin Sci. 3 (1998) 16–23.[9] P. Aksungur, A. Sungur, S. Unal, A.B. Iskit, C.A. Squier, S. Senel, J.

Control. Release 98 (2004) 269–279.10] K.C. Gupta, M.N.V. Ravi Kumar, Biomaterial 1 (2000) 1115–1119.

3589–3600.33] K.C. Gupta, F.H. Jabrail, Carbohydr. Res. 341 (6) (2006) 744–756.34] A.C. Oyrton, C.A. Monteiro Jr., Int. J. Biol. Macromol. 26 (1999)

119–128.35] J. Li, J.F. Revol, R.H. Marchessault, J. Appl. Polym. Sci. 65 (2) (1997)

373–380.36] T. Aspden, L. Illum, O. Skaugrud, Eur. J. Pharm. Sci. 4 (1996) 13–31.37] T.Z. Knaul, S.M. Udson, K.A.M. Creber, J. Appl. Polym. Sci. 72 (1999)

1721–1731.38] X. Qu, A. Lursen, A.C. Albertsson, Polymer 41 (2000) 4589–4598.39] K.D. Yao, W.G. Liu, J. Liu, J. Appl. Polym. Sci. 71 (1999) 449–453.40] S.W. Kim, Y.H. Bae, T. Okano, Pharm. Res. 9 (3) (1992) 283–290.41] J. Li, Z. Xu, Pharm. Sci. 9 (7) (2002) 1669.42] J. Siepmann, M.A. Peppas, Adv. Drug Del. Rev. 48 (2004) 139–157.

Preparation and characterization of sodium hexameta phosphate cross-linked chitosan microspheres for controlled and sustained delivery of centchroman

Documents