Characterization of Ion-Exchange Fibers for Controlled Drug Delivery

(1)

Division of Pharmaceutical Technology Faculty of Pharmacy

University of Helsinki Finland

Characterization of Ion-Exchange Fibers for Controlled Drug Delivery

Kaisa Hänninen

ACADEMIC DISSERTATION

To be presented, with the permission of the Faculty of Pharmacy of the University of Helsinki, for public examination in lecture room 1041,

(2)

Supervisors Professor Jouni Hirvonen

Division of Pharmaceutical Technology Faculty of Pharmacy

Docent Ann Marie Kaukonen

Drug Discovery and Development Technology Center Faculty of Pharmacy

Reviewers Professor Kyösti Kontturi

Laboratory of Physical Chemistry and Electrochemistry Helsinki University of Technology

Finland

Professor Annette Bauer-Brandl Department of Pharmacy

Faculty of Medicine University of Tromsø Norway

Opponent Docent Tomi Järvinen

Kuopio University Pharmacy Finland

ISBN 978-952-10-4631-5 (paperback)

ISBN 978-952-10-4632-2 (PDF, http://ethesis.helsinki.fi) ISSN 1795-7079

Helsinki University Printing House Helsinki, Finland, 2008

(3)

ABSTRACT

Hänninen, K., 2008. Characterization of Ion-Exchange Fibers for Controlled Drug Delivery.

Dissertationes bioscientiarum molecularium Universitatis Helsingiensis in Viikki, 16/2008, 65 pp., ISBN978-952-10-4631-5 (paperback), ISBN978-952-10-4632-2 (PDF), ISSN1795-7079 Increasing attention has been focused on methods that deliver pharmacologically active compounds (e.g. drugs, peptides and proteins) in a controlled fashion, so that constant, sustained, site-specific or pulsatile action can be attained. Ion-exchange resins have been widely studied in medical and pharmaceutical applications, including controlled drug delivery, leading to commercialisation of some resin based formulations. Ion-exchangers provide an efficient means to adjust and control drug delivery, as the electrostatic interactions enable precise control of the ion-exchange process and, thus, a more uniform and accurate control of drug release compared to systems that are based only on physical interactions. Unlike the resins, only few studies have been reported on ion-exchange fibers in drug delivery. However, the ion-exchange fibers have many advantageous properties compared to the conventional ion-exchange resins, such as more efficient compound loading into and release from the ion-exchanger, easier incorporation of drug-sized compounds, enhanced control of the ion-exchange process, better mechanical, chemical and thermal stability, and good formulation properties, which make the fibers attractive materials for controlled drug delivery systems.

In this study, the factors affecting the nature and strength of the binding/loading of drug-sized model compounds into the ion-exchange fibers was evaluated comprehensively and, moreover, the controllability of subsequent drug release/delivery from the fibers was assessed by modifying the conditions of external solutions. Also the feasibility of ion- exchange fibers for simultaneous delivery of two drugs in combination was studied by dual loading. Donnan theory and theoretical modelling were applied to gain mechanistic understanding on these factors. The experimental results imply that incorporation of model compounds into the ion-exchange fibers was attained mainly as a result of ionic bonding, with additional contribution of non-specific interactions. Increasing the ion-exchange capacity of the fiber or decreasing the valence of loaded compounds increased the molar loading, while more efficient release of the compounds was observed consistently at conditions where the valence or concentration of the extracting counter-ion was increased.

Donnan theory was capable of fully interpreting the ion-exchange equilibria and the theoretical modelling supported precisely the experimental observations. The physico- chemical characteristics (lipophilicity, hydrogen bonding ability) of the model compounds and the framework of the fibrous ion-exchanger influenced the affinity of the drugs towards the fibers and may, thus, affect both drug loading and release.

It was concluded that precisely controlled drug delivery may be tailored for each compound, in particularly, by choosing a suitable ion-exchange fiber and optimizing the

(4)

ACKNOWLEDGEMENTS

This study was carried out at the Division of Pharmaceutical Technology, Faculty of Pharmacy, University of Helsinki.

I am most grateful to my supervisor, Professor Jouni Hirvonen, for providing me this interesting and challenging subject, for excellent working facilities and for finding the funding also at the end of my work. His endless optimism and support during this study have helped me, especially on the ‘few’ moments of despair. My sincere gratitude is due to my other supervisor, Docent Ann Marie Kaukonen, for many valuable ideas, excellent advices and numerous scientific discussions. She was always available when help or encouragement was needed. Without her this study would have been a much harder.

I am most indepted to my co-author, Docent Lasse Murtomäki, whose contribution for this work has been indispensable. Our unforgettable and interesting discussions about physical pharmacy made my knowledge about ion-exchange possible. My other co-author Dr. Tarja Kankkunen is thanked for useful experimental hints at the beginning of this study.

My sincere thanks are due to Dr. Pasi Kauppinen and Dr. Mats Sundell for sharing me their knowledge about ion-exchange fibers.

Professor Kyösti Kontturi and Professor Annette Bauer-Brandl are acknowledged for carefully reviewing this thesis and for providing valuable comments and suggestions for its improvement.

I wish to thank Katri Lahtinen, Katja Suhonen, Maija Koljonen (M.Sc.) and Dr. Samuli Hirsjärvi for their excellent technical assistance.

I express my sincere thanks to my colleagues and the staff of the Division of Pharmaceutical Technology for creating a pleasant working atmosphere. I wish to thank especially Henna, Johanna, Sanna, Maija, Anna and Tarja for sharing my moments of excitement and despair during the study, for being an excellent travelling companion during the scientific meetings and for numerous memorizable and enjoyable events also at the free time. It was a pleasure to work with you.

The financial support from the Acadamy of Finland, the Finnish Cultural Foundation (Foundation of Elli Turunen), the Graduate School in Pharmaceutical Research and the University of Helsinki are gratefully acknowledged.

I want to thank my parents, Maaret and Markku, most warmly for their understanding and support in the course of my almost endless years of study. Markku will be thanked also for helping with IT issues.

Finally, my most sincere and warmest thanks belong to my dear husband Marko for his never ending care, support and understanding during these many years and to my children Pauliina and Joona for bringing joy and happiness to my life.

Järvenpää, May 2008

Kaisa Hänninen

(5)

SYMBOLS AND ABBREVIATIONS

3-IPSA 3-isopropylsalicylic acid

5-Br 5-bromosalicylic acid

5-CH3 5-methylsalicylic acid

5-Cl 5-chlorosalicylic acid

5-F 5-fluorosalicylic acid

5-NH2 5-aminosalicylic acid

5-OH 5-hydroxysalicylic acid

ACN acetonitrile

ai activity of ion

ASA acetylsalicylic acid

c_i concentration of ion

c_Clw concentration of chloride in aqueous external solution

c_Clf concentration of chloride in ion-exchange fiber

c_Naw concentration of sodium in aqueous external solution c_SAw concentration of salicylic acid in aqueous external solution

c_SAf concentration of salicylic acid in ion-exchange fiber

0 , Cl

cw initial concentration of chloride in aqueous external solution

0 , SA

cf initial concentration of salicylic acid in ion-exchange fiber

di-COOH 5-hydroxyisophthalic acid

DMSO dimethylsulfoxide

DVB divinylbenzene

F Faraday constant (96,487 C mol^–1)

f ion-exchange fiber

GI gastrointestinal

HPLC high performance liquid chromatography

i ion

ISA ionic strength adjusted

K equilibrium constant of ion-exchange,i.e. K = KeKc

Kc orK^ch_p_,_i chemical partition coefficient of ion-exchange Ke electrical partition coefficient of ion-exchange Kexp experimental equilibrium constant of ion-exchange

Kexp,j experimental equilibrium constant of ion-exchange after j^th extraction stage

log Doct octanol/water distribution coefficient log Poct octanol/water partition coefficient

n molar amount of ions

nCl molar amount of extracting chloride ions which is based on the capacity of ion-exchange fiber

(8)

0

nCl initial molar amount of chloride in ion-exchange fiber

0

nD initial molar amount of drug in ion-exchange fiber

0

ns initial molar amount of extracting salt in aqueous external solution n₀w initial molar amount of model compound in aqueous external solution

0 , SA

nf initial molar amount of salicylic acid in ion-exchange fiber

n_SAw molar amount of salicylic acid in aqueous external solution

pKa dissociation constant

R gas constant (8.314 J K^–1mol^–1)

r volume ratio of fiber phase (V^f) and aqueous external solution phase (V^w)

RDS rate-determining step of the ion-exchange

SA salicylic acid

SEM scanning electron microscopy

Smopex DS-218v viscose-g-trimethylammoniumpropylmethacrylamide chloride fiber Smopex^®-103pe poly(ethylene)-g-vinylbenzyltrimethylammonium chloride fiber Smopex^®-105pe poly(ethylene)-g-vinylpyridine chloride fiber

Smopex^®-105v viscose-g-vinylpyridine chloride fiber

T temperature

TFA trifluoroacetic acid

V^f volume of ion-exchange fiber phase

V^w volume of aqueous external solution phase

w aqueous external solution

x experimentally determined amount of drug/compound released from fiber (mmol)

X amount of fixed ion-exchange groups in fiber (binding sites)

z experimentally determined fraction of drug/compound released from fiber

zf charge number of ion-exchange group in fiber

zi charge number (valence) of ion

0

i standard chemical potential of ion in external solution phase

0

i standard chemical potential of ion in ion-exchange fiber phase

~i equal electrochemical potential of ion

i activity coefficient of ion,i.e. ai = ici

-1 ratio of model compound in loading solution (n₀^w)and fiber capacity (X) galvani potential

D or D Donnan potential

(9)

LIST OF ORIGINAL PUBLICATIONS

This doctoral dissertation is based on the following original publications, which are referred to in the text by Roman numerals asI-IV:

I Hänninen, K., Kaukonen, A.M., Kankkunen, T., Hirvonen, J., 2003. Rate and extent of ion-exchange process: the effect of physico-chemical characteristics of salicylate anions. Journal of Controlled Release 91, 449- 463.

II Hänninen, K., Kaukonen, A.M., Murtomäki, L., Hirvonen, J., 2005. Effect of ion-exchange fiber structure on the binding and release of model salicylates.

Journal of Pharmaceutical Sciences 94, 1772-1781.

III Hänninen, K., Murtomäki, L., Kaukonen, A.M., Hirvonen, J., 2007. The effect of valence on the ion-exchange process: theoretical and experimental aspects on compound binding/release. Journal of Pharmaceutical Sciences 96, 117-131.

IV Hänninen, K., Kaukonen, A.M., Murtomäki, L., Hirvonen, J., 2007.

Mechanistic evaluation of factors affecting compound loading into ion- exchange fibers.European Journal of Pharmaceutical Sciences31, 306-317.

(10)

(11)

1 INTRODUCTION

Increasing attention has been focused on methods that deliver pharmacologically active compounds (e.g. drugs, peptides and proteins) in a controlled fashion, as numerous drugs are required to be released at a controlled rate so that constant, sustained, site-specific or pulsatile action is obtained. Ion-exchange resins have been studied in pharmaceutical applications since the early 1950s, leading to patenting and commercialisation of some resin based formulations (Anand et al., 2001; Elder, 2005). Resins have found use as pharmacologically-active ingredients (Iversen et al., 1995; Elder, 2005) and as pharmaceutical excipients that improve drug stability (Elder, 2005), mask the taste of a drug (Borodkin and Sundberg, 1971; Lu et al., 1991), enhance the dissolution of poorly soluble drugs (Irwin et al., 1990), or achieve sustained or controlled drug delivery (Burke et al., 1986; Irwin et al., 1990; Jani et al., 1994; Conaghey et al., 1998a,b). Advantageous properties, such as a high capacity for drug loading, an easily executed loading procedure, good drug-retaining properties, and more uniform drug release makes the ion-exchangers, as such, attractive for drug delivery systems. Compared to resins, there are only a few studies in the literature about the pharmaceutical use of ion-exchange fibers. Typically, the ion-exchange kinetics of the fibers has been studied with small inorganic anions and cations, as their prevalent applications include, e.g., analytical technologies, air and water purification (removal of toxins, pesticides, heavy metals), and extraction of useful/rare ions (e.g. uranium and gold) from water (Soldatov et al., 1988; Lin et al., 1992; Lin and Hsieh, 1996; Chen et al., 1996; Economy et al., 2002). Overall, the Donnan theory and mathematical modelling of ion-exchange are mainly discussed in the literature on the basis of inorganic ions. However, the size of a drug molecule is usually larger than that of an inorganic ion, and the molecule contains various functional groups capable of non-specific interactions with the fibers, e.g. hydrophobic interactions, which may affect the release kinetics.

Recently, it has been shown that the ion-exchange fibers enable controlled transdermal drug delivery both in vitro andin vivo, and improve the stability of drugs when used as a drug reservoir in an iontophoretic patch (Kankkunen et al., 2002a, b; Vuorio et al., 2004;

Yu et al., 2006). Compared to the conventional ion-exchange resins, enhanced ion- exchange of drugs has been observed with the ion-exchange fibers due to the more open and uniform structure of the fibers (Vuorio et al., 2003). The ion-exchange is more rapid and efficient in the case of the fibers (Soldatov et al., 1988, 1999; Chen et al., 1996;

Vuorio et al., 2003), resulting in more efficient drug loading into and drug release from the ion-exchangers, easier incorporation of drug-sized molecules, and a more direct control of the loading/release.

The objective of this study was to evaluate comprehensively the factors affecting the binding/loading of drug-sized model compounds into the ion-exchange fibers and the controllability of subsequent drug release/delivery from the fibers by modifying the

(12)

2 REVIEW OF THE LITERATURE

2.1 STRUCTURE AND PROPERTIES OF ION-EXCHANGE MATERIALS

Ion-exchange materials, such as macroreticular resins, gels, membranes and fibers, contain two components: a water insoluble structural component consisting of a polymer framework, and a functional component consisting of fixed acidic or basic ion-exchange groups (Kunin and Myers, 1952; Helfferich, 1995; Anand et al., 2001; Elder, 2005). Ion- exchangers can be classified according to the nature of these structural and functional components.

The ion-exchangers can hence be divided into cation-exchangers and anion-exchangers on the basis of the functional groups (Kunin and Myers, 1952; Helfferich, 1995). The ionisable acidic/basic groups, such as -SO3-

, -COO^- and -PO32-

(cation-exchangers) or -N⁺(CH3)3, -NH3+

, -NH2+

and -NH⁺ (anion-exchangers), are covalently attached to the framework of the ion-exchanger, and their charge compensated for by mobile counter-ions (Figure 1). Thus, the exchangeable mobile counter-ions are cations in the case of cation- exchangers and anions in the case of anion-exchangers. For example, an anion-exchanger containing exchangeable Cl^--ions is said to be in Cl^--form. Conversion of the anion- exchanger’s ionic form (originally in Cl^--form) to e.g. SO42-

-form can be achieved by treating the anion-exchanger with a sufficient excess of a solution of a sulphate. A small amount of co-ions, i.e. mobile ions possessing the same charge as the ion-exchanger, can also be present in the vicinity of the fixed ion-exchange groups. Ion-exchange materials may also contain different types of ionic groups of the same charge (bifunctional/polyfunctional exchangers) or opposite charges (amphoteric exchangers).

Ion-exchangers can be characterized in a quantitative manner by their capacity, which is defined as the number of ion-exchange groups in a specified amount of ion-exchanger (e.g. mmol/g) (Kunin and Myers, 1952; Helfferich, 1995). Instead, the effective capacity reveals the amount of counter-ions (i.e. ionic groups capable of ion exchange) per particular amount of ion-exchanger. Depending on the acid or base character of the ionic groups, the ion-exchangers can be classified into strong and weak ion-exchangers. The weakly basic groups (-NH3+

, -NH2+

, -NH⁺) are ionised only under low pH conditions, and the effective capacity of compound binding by these groups is progressively decreased by an increase in pH. Similarly, the weakly acidic groups (-COO^-) are ionised only under high pH conditions. In contrast, strongly basic/acidic groups (-N⁺(CH₃)₃/-SO₃^-, -PO₃^2-) remain ionised over the entire pH-scale and, thus, the effective capacity of the ion- exchanger is practically pH independent.

Ion-exchange resins

The nature of the structural component (framework) determines the ion-exchangers’

physical form, whereby the ion-exchange materials can be classified as macroreticular resins, gels, membranes or fibers (Kunin and Myers, 1952; Helfferich, 1995). The most

(13)

important class of the ion-exchangers is the organic ion-exchange resins (Figures 1A-B).

The first ion-exchange resins were synthesized in 1935 by Adams and Holmes (Adams and Holmes, 1935) and, nowadays, innumerable types of resins with different properties can be prepared for various applications. Resins are prepared as spherical beads, which consist of a three-dimensional network of hydrocarbon chains carrying covalently bound ion-exchange groups. The various hydrocarbon chains of the resins are covalently interconnected by cross-linking, which can be achieved using cross-linking agents such as divinylbenzene (DVB). The resins are, thus, cross-linked polyelectrolytes. The currently used ion-exchange resins are mostly based on polystyrene or polymethacrylic polymers (Elder, 2005).

The degree of cross-linking determines the mesh width of the resins (porosity and pore size) and, thereby, their physical properties (e.g. moisture content), their behaviour towards biological substances and, to a certain extent, their capacity (Ranade, 1990).

Although the resins are insoluble in all biological media, they usually have the ability to strong swelling (Kunin and Myers, 1952; Helfferich, 1995; Anand et al., 2001; Elder, 2005). By lowering the degree of cross-linking, a resin with higher porosity, lower density, and higher moisture content is formed. These modifications result in (1) a higher rate of ionic diffusion within the resin bead and, therefore, a higher rate of ion-exchange, and (2) a higher capacity for ions of large molecular size. Instead, highly cross-linked resins are harder and more resistant to mechanical breakdown, but the ion-exchange may be very ineffective due to the sterical hindrance.

Ion-exchange fibers

Unlike other ion-exchange materials, ion-exchange fibers consist of non-cross-linked polymeric frameworks (Ekman, 1994) (Figures 1C-D). Typically the framework is composed of hydrophobic polymer chains (e.g. poly(propylene), poly(ethylene)) carrying a positive (anion-exchanger) or a negative (cation-exchanger) fixed electric charge. Unlike the conventional resins, the ion-exchange groups are located on the surface of the fibers which enables an easier access of ions to the ion-exchange groups (Soldatov et al., 1988;

Chen et al., 1996; Economy et al., 2002; Vuorio et al., 2003). The surface area (to unit volume ratio) of the non-cross-linked ion-exchange framework is higher and the thickness of the fiber shell (where the ion-exchange process occurs) is smaller (~ 5 µm) and more uniform in comparison to the radius of the conventional spherical ion-exchange bead (~ 0.1-0.5 mm). Therefore, the ion-exchange process is more rapid and efficient in the case of fibers, achieving more efficient drug loading and release into/from the ion- exchanger, easier incorporation of large sized molecules and more straightforward controllability of the loading/release. Other beneficial properties of the fibers are good mechanical and thermal strength, chemical inertness of the framework and the possibility to achieve a very high capacity by high extents of grafting (Sundell and Näsman, 1993).

Extremely high osmotic stability of ion-exchange fibers allows their exposure to multiple drying-moistening cycles and conversion from one ionic form to the other without

(14)

the forms of filaments, staples, cloths and non-woven materials, opening up many possibilities for new technological processes (Soldatov et al., 1988; Economy et al., 2002).

Figure 1. Schematic structures and SEM micrographs of strongly basic anion-exchange resin (A- B) and anion-exchange fiber (C-D) (Figures A and C are modified from Helfferich, 1995 and Ekman, 1994). Note the different scaling in SEM pictures.

2.2 THEORY AND MECHANISMS OF ION-EXCHANGE

The ion-exchange reaction is a reversible, selective and stoichiometric interchange of mobile ions of like charges between the ion-exchanger and the external liquid phases (Helfferich, 1995). Each counter-ion that is released from the ion-exchanger is replaced by an equivalent amount of another ionic species of same sign and valence due to the electroneutrality requirement. Based on the nature of the ionic species being exchanged, the ion-exchange process is either anionic or cationic.

Ion-exchanger⁺ A^- + B^- Ion-exchanger⁺ B^- + A^-

CH CH₂ CH CH₂ CH CH₂ CH CH₂

CH CH₂

CH CH₂ CH CH₂ CH

CH CH₂ CH₂

CH

A

R1 R1

R1

R1 R1

B

CH₂ CH CH₂ CH CH₂

CH CH₂ CH

CH₂ CH CH₂ CH CH₂ CH

O CH₂

N⁺(CH₃)₃Cl^-

C

R2

R2 R2

R2

R2 : R1 :

N⁺(CH₃)₃Cl^-

D

(15)

When the ion-exchanger is placed in an electrolyte solution containing counter-ions which are different from those bound to the ion-exchanger, the migration of the first few external ions into the ion-exchanger and bound ions into the surrounding external solution creates an electrical potential difference (Donnan potential) between the ion-exchanger and the external solution phases (Donnan, 1911; Helfferich, 1995). The created Donnan potential accomplishes the interchange of counter-ions between the two phases until an equilibrium stage (Donnan equilibrium) is reached, that is, the equality of electrochemical potentials for each mobile-ion between the phases (Figure 2) (Donnan, 1911; Helfferich, 1995; Ramírez et al., 2002). The higher the Donnan potential, the stronger is the co-ion exclusion from the ion-exchanger and, on the other hand, the stronger is the attraction of counter-ions towards the ion-exchanger. In a concentrated external solution the Donnan potential is low and, thus, the interaction between the mobile counter-ion and the ion- exchanger is weak, achieving high rates of ion-exchange. In addition to the concentration of the surrounding solution, the Donnan potential is dependent on the selectivity and capacity of the ion-exchanger, the charge of the ions present, and the pressure (Donnan, 1911; Boyd et al., 1961; Helfferich, 1995).

Figure 2. Schematic representation of the ion-exchange process between the strong base (A) or weak base (B) anion-exchangers and the external solution, respectively (modified from Ramírez et al., 2002). The model systems contain five mobile ions: the anionic drug (D ^-), the salt anion (Cl ^-), the salt cation (Na ⁺), the hydrogen ion (H⁺) and the hydroxide ion (OH^-). In addition to these mobile ions, the ion-exchanger phases contain also fixed species covalently attached to the polymer framework (circumscribed): strongly basic trimethylammonium groups (-N(CH3)3

+) (A) or weakly basic vinylpyridine groups in ionised (-NH⁺) and neutral (-N) forms (B) depending on

D

N

D

H⁺ H⁺

H⁺

Ion-exchanger phase at potential External solution phase at potential

Na⁺

Cl^-

D^- N(CH₃)₃⁺

OH^-

Na⁺

Cl^- D^-

OH^-

H⁺

Na⁺ Cl^-

D^-

OH^-

Na⁺

D^-

Cl^-

OH^- A)

B)

(16)

The ion-exchange is essentially a diffusion process, but is also related to chemical reaction kinetics (Helfferich, 1995; Abdekhodaie and Wu, 2006). It can be described as a series of consecutive reaction and mass transfer processes (1-5). First, (1) the exchangeable counter-ion must diffuse through the adherent external solution to the surface of the ion-exchanger (film diffusion) and then, (2) within the ion-exchange material, to the ionised functional groups (particle diffusion). The actual ion-exchange reaction between the mobile counter-ions occurs at the fixed ionic binding site (3). Finally, the released counter-ion diffuses from the ionic binding site into the surrounding solution by particle and film diffusion (4, 5). The rate of the ion-exchange process is determined by the slowest of these five steps. In most of the cases, the rate-determining step (RDS) of the ion-exchange is the diffusion of the larger ion (e.g. drug-ion) within the polymer framework, as (a) the transport resistance outside the polymer framework is negligible, when sufficient stirring is applied, (b) the diffusion coefficient of organic drugs/compounds with molecular weights of a few hundreds is smaller than that of the small counter-ions like Na⁺ and Cl^- and, (c) the actual chemical exchange reaction at the fixed ion-exchange site is, at least in the majority of cases, found to be faster compared to the diffusion of the counter-ions (Abdekhodaie and Wu, 2006). In this situation, the kinetics of the ion-exchange is said to be ‘particle diffusion controlled’ (RDS is step 2 in the case of drug loading and step 4 in the case of drug release). Due to the electroneutrality requirement, the slower counter-ion (drug) reduces the diffusion rate of the faster one so that no net transfer of charge occurs between the phases, i.e. the fluxes of the exchanging counter-ions are equivalent. In the case of steps 1 or 5 being the RDS, the kinetics of the ion-exchange reaction is said to be ‘liquid-film diffusion controlled’, and if step 3 is the RDS, the reaction is ‘chemical reaction controlled’. This might be the case, when the ion- exchanger is exceedingly tightly cross-linked or the exchangeable ions are very large, preventing the diffusion of ions inside the ion-exchanger; the ion-exchange reaction is thus limited to the surface of the ion-exchanger (Kunin and Myers, 1952).

Due to the particle diffusion controlled kinetics of the ion-exchange, an interfacial loading mechanism of drugs into the resin beads/fiber filaments is assumed (Figure 3) (Lin and Hsieh, 1996; Abdekhodaie and Wu, 2006). According to this ‘moving boundary’

model, the drug loading process creates a sharp boundary in the ion-exchanger that separates the reacted surface (exterior) from the unreacted interior. The created boundary advances from the surface towards the interior part of the ion-exchange material during the loading process. Thus, the ion-exchange reaction between the drug in the solution and the counter-ion in the ion-exchanger takes place mainly at the interphase of these two phases. Thereafter, the drug-ion diffuses further inwards the resin bead/fiber filament due to a concentration gradient and exchanges with the counter-ion bound at the next ionic binding site located at the inner part of the ion-exchanger. The released counter-ions diffuse rapidly to the surface of the resin/fiber owing to the concentration gradient, weaker affinity to the ion-exchanger and larger diffusion coefficient compared to the drug. At the surface, the counter-ions exchange with the drug-ions in the solution phase. This exchange-diffusion-removal process proceeds until the Donnan equilibrium is reached.

(17)

Figure 3. Schematic illustration of drug loading mechanism proposed in the ‘moving boundary’

model (modified from Abdekhodaie and Wu, 2006). indicates the available binding sites in the ion-exchanger that are occupied either by drug-ions (D^-) or counter-ions (Cl^-).

The ion-exchange reaction includes both the electrostatic interactions (K_e) and the chemical contributions (Kc, K^ch_p_,_i) (Helfferich, 1995; Jaskari et al., 2000, 2001; Ramírez et al., 2002). Thus, the ion-exchange equilibrium constant (K) is a product of these two

c eK K K

The electrostatic interactions are determined by the electrical partition coefficient RT

F z K_e exp _i _D

where zi, F, R and T are the charge number of the ion, the Faraday constant, the gas constant and the thermodynamic temperature, respectively. D is the Donnan potential (i.e. the difference of the electrical potentials between the ion-exchanger and external solution phases). In the case of anionic compounds in anion-exchangers, zi is < 0, D is positive and, thus, Ke > 1. When the charge number (valence) of the anion increases, the Donnan potential will decrease and the value of Ke gets smaller.

o o o

o

Cl^- Cl^- Cl^- D^-

Cl^- Cl^- D^-

o o

Cl^- Cl^- Cl^- D^-

Cl^- Cl^- D^- D^-

o o o

o

Cl^- Cl^-

D^-

Cl^- Cl^- D^- D^-

D^-

o o

Cl^-

D^-

Cl^- Cl^- D^- D^-

D^- D^-

Surface of ion-exchanger Surface of ion-exchanger

(18)

The chemical partition coefficient can be expressed as RT

K_c exp _i⁰ _i⁰

where _i⁰ and _i⁰ are the standard chemical potentials of the ions in the external and the ion-exchanger phases, respectively. The chemical partition coefficient expresses the tendency of the counter-ion (e.g. drug) to get into the ion-exchanger due to non-specific adsorption (e.g. hydrophobic interactions) (Jaskari et al., 2001). The more hydrophobic the counter-ion, the larger is the value of the chemical partition coefficient (Kc or K^ch_p_,_i > 1), which explains the differences between the ion-exchange affinities of ions with like charges. In the case of hydrophilic drug-ions, Kc can be < 1, as the water content of the ion-exchanger also affects the chemical partition coefficient. Without the chemical contributions of the ionic partition coefficients, the ion-exchange affinities of ions with like charge would theoretically be equal (Kc or K^ch_p_,_i 1). This would assumably be the case for small inorganic ions, like Na⁺ and Cl^- (Ramírez et al., 2002). However, according to experimental data, the relative affinity of various inorganic ions towards the ion- exchange resins is different (see the selectivity sequences illustrated in next chapter) (Kunin and Myers, 1952; Helfferich, 1995). This may be attributed to the differences in the radius of the hydrated ions.

2.3 FACTORS AFFECTING ION-EXCHANGE

Drugs can be bound to the ion-exchangers either by ionic bonds formed between the ionised drugs and the fixed ion-exchange groups of the ion-exchangers (via ion-exchange reaction) or by adsorbing more loosely on the ion-exchanger via hydrophobic interactions (Borodkin and Yunker, 1970; Mosquera et al., 1999; Marchal-Heussler et al., 2000).

Hydrophobic interactions can be formed between the hydrophobic parts of the drug and the ion-exchanger or between the hydrophobic drugs bound to the ion-exchanger. The prevalent binding mechanism of the ionised drugs into the ion-exchanger is usually ionic bonding, although hydrophobic interactions may contribute to variable extents.

The ion-exchange process is affected by the physico-chemical properties of the ion- exchanger and the counter-ion (e.g. drug) and, by the surrounding external conditions (Helfferich, 1995; Ramírez et al., 2002), which will be discussed under the following sub- sections (Table 1).

(19)

Table 1. Factors affecting the loading of drugs/compounds into and release from the ion-exchange materials.

Factor Mechanism of effect Reference

Ion-exchanger dependent

Ion-exchange capacity Donnan potential,

number of ionic binding sites

Helfferich, 1995; Kriwet and Kissel, 1996; Jaskari et al., 2001;

Ramírez et al., 2002; Uchida et al., 2003

Nature of fixed ionic groups ionisation, selectivity

Kunin and Myers, 1952; Sprockel and Price, 1989; Helfferich, 1995;

Conaghey et al., 1998a; Jaskari et al., 2001;

Kankkunen et al., 2002b; Vuorio et al., 2003

Preloaded counter-ion pH, selectivity Kunin and Myers, 1952; Chen et al., 1992; Helfferich, 1995;

Sriwongjanya and Bodmeier, 1998

Particle size surface area, particle diffusion Raghunathan et al., 1981; Irwin et al., 1990;

Ranade, 1990; Conaghey et al., 1998b; Abdekhodaie and Wu, 2006

Degree of cross-linking pore size of ion-exchanger, particle diffusion

Kunin and Myers, 1952; Sawaya et al., 1987, 1988;

Irwin et al., 1987, 1990; Ranade, 1990; Conaghey et al., 1998b;

Åkerman et al., 1999

Drug dependent

Lipophilicity binding affinity

Kril and Fung, 1990; Åkerman et al., 1998; Liu et al., 1999, 2001;

Jaskari et al., 2000, 2001; Ramírez et al., 2002;

Vuorio et al., 2003, 2004

pK_a ionisation Åkerman et al., 1998; Kankkunen et al., 2002b

Sterical properties binding accessibility Borodkin and Yunker, 1970; Kril and Fung, 1990

Molecular size diffusion coefficient, binding affinity, binding accessibility

Kunin, 1949; Kunin and Myers, 1949; Irwin et al., 1987;

Farag and Nairn, 1988; Åkerman et al., 1998, 1999

External conditions

Concentration of solution Donnan potential

Helfferich, 1995; Irwin et al., 1987; Sawaya et al., 1987, 1988;

Conaghey et al., 1998a; Jaskari et al., 2001;

Kankkunen et al., 2002b; Abdekhodaie and Wu, 2006

Valence of surrounding ions Donnan potential, electroselectivity Sawaya et al., 1987, 1988; Bhandari et al., 1993; Helfferich, 1995;

Jaskari et al., 2001; Liu et al., 2001; Kankkunen et al., 2002b

pH ionisation of drug and ion-exchanger Borodkin and Yunker, 1970; Åkerman et al., 1998;

Kankkunen et al., 2002b

Temperature porosity of ion-exchanger, diffusion Borodkin and Yunker, 1970; Irwin et al., 1990; Chen et al., 1996

Agitation film diffusion Sawaya et al., 1988; Irwin et al., 1987, 1990; Chen et al., 1996

(20)

2.3.1 Ion-exchanger dependent factors

Ion-exchange capacity

Theoretically, the fixed ionic groups largely determine the ion-exchange behaviour of the ion-exchange materials (Helfferich, 1995). The number of the groups determines the ion- exchange capacity, while the chemical nature of the groups greatly affects the equilibrium of ion-exchange. The number of the ionisable groups per specified amount of ion- exchanger expresses the theoretical ion-exchange capacity of the material and, it can be used to characterize the ion-exchangers. However, from a practical point of view, the effective capacity of ion-exchangers, which is defined as the number of exchangeable counter-ions in the material, is more important. Only ionised ion-exchange groups are able to act as fixed charges in the ion-exchanger and will, thus, be balanced with counter-ions.

Groups, which are not ionised, do not contribute to the effective capacity. Strong acid/base groups are practically completely ionised under any pH conditions and, therefore, the effective capacity of ion-exchangers containing such groups is essentially constant and equivalent to the theoretical ion-exchange capacity. Instead, the ionisation degree of weak acid/base groups depends on the surrounding pH conditions and, thus, the effective capacity of such ion-exchangers will vary accordingly. A high degree of cross-linking may also reduce the effective capacity of ion-exchangers (resins) relative to their theoretical ion-exchange capacity, by hiding a portion of fixed ionic groups (Helfferich, 1995; Kriwet and Kissel, 1996; Liu et al., 2001).

The effective ion-exchange capacity of the ionic polymer affects the Donnan potential created between the ion-exchanger and the external solution phases (Helfferich, 1995;

Ramírez et al., 2002). The Donnan potential is higher when the difference in ionic concentrations between the ion-exchanger and the external solution is larger; the absolute value of the Donnan potential increases with decreasing external and/or increasing internal (i.e. effective ion-exchange capacity) counter-ion concentrations. Therefore, the higher the effective capacity of the ion-exchanger, the stronger is the affinity/interaction between the fixed ion-exchange groups and the mobile counter-ions (e.g. ionic drugs) and, thus, the smaller is the drug release from the ion-exchanger. Accordingly, the loading of drugs into the ion-exchanger is higher/more efficient with ion-exchangers of higher effective capacity (Jaskari et al., 2001; Ramírez et al., 2002; Uchida et al., 2003).

Nature of fixed ion-exchange groups

The strength of interactions formed between the mobile counter-ions and the fixed ion- exchange groups may also depend on the chemical nature of the functional groups (e.g.

–SO3H vs. –COOH) (Kunin and Myers, 1952). For example, Na⁺-ions are more difficult to replace by Ca²⁺- or Mg²⁺-ions from the sulfonic acid resin compared to the carboxylic acid resin. The strength of specific interactions between the ion-exchanger and the drug has also been found to differ with these strong and weak cation-exchange materials in a drug and fiber/resin specific manner (Sprockel and Price, 1989; Jaskari et al., 2001;

Vuorio et al., 2003). Lipophilic drugs were bound stronger into the strong sulfonic acid groups compared to the weak carboxylic acid groups, while the opposite was true with a hydrophilic drug. Additionally, the acid/base strengths of the fixed ionic groups affect the

(21)

ion-exchange in the fiber/resin via the ionisability of the ion-exchange material and, thus, the Donnan potential (see the above chapter).

Preloaded counter-ion

The nature of the preloaded counter-ions may affect the binding of drugs and other compounds into the ion-exchanger (Chen et al., 1992; Sriwongjanya and Bodmeier, 1998).

This may be attributed primarily to two factors. Firstly, ion-exchangers in the base/acid form may change the solubility of the drug in the solution phase and simultaneously the dissociation degree of the ion-exchanger, as the pH of the surrounding solution changes by the release of OH^--/H⁺-ions from the ion-exchange material during the drug loading process. In contrast, ion-exchangers in salt form (i.e. containing e.g. Na⁺- or Cl^--ions as counter-ions) sustain a constant environmental pH. Secondly, dissimilar selectivity of the ion-exchangers (i.e. the selection by the ion-exchanger of one counter-ion in preference to the other) towards the different counter-ions may induce differences in the drug loading/release behaviour, although the binding selectivity of the organic drug-ions is normally higher than that of smaller inorganic ions (Kunin and Myers, 1952; Helfferich, 1995; Abdekhodaie and Wu, 2006). The selectivity of ion-exchangers is affected at least by the nature of the fixed ionic groups in the ion-exchange materials and by the valence, lipophilicity and ionic radii of the mobile counter-ions. Generally, the selectivity sequence of the most common cations is (Kunin and Myers, 1952; Helfferich, 1995):

Ba²⁺ > Pb²⁺ > Sr²⁺ > Ca²⁺ > Ni²⁺ > Cd²⁺ > Cu²⁺ > Co²⁺ > Zn²⁺ > Mg²⁺

Tl⁺ >Ag⁺ > Cs⁺ > Rb⁺ > K⁺ > NH4+

> Na⁺ > Li⁺

The sequences of the monovalent and divalent cations may overlap in ion-exchangers of high capacity and high degree of cross-linking. For strong acid ion-exchangers, H⁺ usually falls between Na⁺ and Li⁺. For weak acid ion-exchangers, the position of H⁺ depends on the acid strength of the ion-exchange groups.

For the most common anions, the following selectivity sequence is given:

Citrate > SO42-

> oxalate > I^- > NO3-

> CrO42-

> Br^- > SCN^- > Cl^- > formate >

acetate > F^-

With strong base ion-exchangers, OH^- usually falls between acetate and fluoride. For weak base ion-exchangers, the position of OH^- is farther to the left and depends on the base strength of the fixed ionic groups. However, these series may not be constant over wide ranges of concentration and pH, due to the possible differences in the activity coefficient- concentration relationships of the ions. At high concentrations, the differences in the exchange selectivities of ions with different valences diminish and in some cases the ion of lower valence may even have a higher selectivity (Kunin, 1949).

(22)

Particle size and degree of cross-linking

While the ion-exchange equilibrium is primarily dependent on the chemical structure of the ion-exchanger, the kinetics of the exchange reaction depends strongly on the physical structure of the ionic polymer (Kunin and Myers, 1952; Ranade, 1990; Abdekhodaie and Wu, 2006). A number of physical (and also chemical) properties of ion-exchange resins can be varied by modifying their particle size and degree of cross-linking. The diffusional path of exchangeable ions is longer in larger sized resin beads, reducing therefore the rate of ion-exchange as the particle size of the resin is increased (Raghunathan et al., 1981;

Irwin et al., 1990; Conaghey et al., 1998b; Abdekhodaie and Wu, 2006). Cross-linking modifies the porosity of the resin framework and, thus, affects the diffusion of ions within the ion-exchange material (Kunin and Myers, 1952; Sawaya et al., 1987, 1988; Irwin et al., 1987, 1990; Ranade, 1990). As the degree of cross-linking is increased, the rate of drug/compound release from the resin is generally slower due to the decreased moisture content of the resin framework. In addition, the resins possess ion-exchange groups inside the cross-linked resin beads that are accessible only through pores of varying diameter.

Reduction of the pore diameter by increased cross-linking may ‘block out’ ions having an effective diameter larger than the diameter of the pore and, therefore, hinder their uptake and release (i.e. the ability of the ion-exchanger to accommodate large ions is reduced) (Kunin, 1949; Sawaya et al., 1987).

2.3.2 Drug dependent factors

Lipophilicity

Physicochemical characteristics of drugs have often an influence on the selectivity of ion- exchange. The effect of drug lipophilicity on the binding and release from different ion- exchange materials has been studied widely with a variety of drugs (Kril and Fung, 1990;

Åkerman et al., 1998; Liu et al., 1999; Jaskari et al., 2000, 2001; Vuorio et al., 2003, 2004). Generally, drugs possessing higher lipophilicity have higher affinity towards the ion-exchangers. Their loading into the ion-exchangers is more effective and the release is reduced compared to the more hydrophilic drugs. This is due to the contribution of the chemical partition coefficient (K^ch_p_,_i or Kc) to the equilibrium distribution of the drugs between the ion-exchanger and external solution phases (see 2.2) (Jaskari et al., 2000, 2001; Ramírez et al., 2002).

pKa and sterical properties

Drug pKa is essential for the ion-exchange process, as only ionised drugs can be bound into the ion-exchange groups via ionic bonding (Helfferich, 1995; Åkerman et al., 1998;

Kankkunen et al., 2002b). Many drugs are either weak bases or weak acids (Newton and Kluza, 1978) and, therefore, their ionisation and ion-exchange ability is often dependent on the environmental pH. Alteration of the pH may increase the amount of unionised drug and, thus, cause decreasing interaction between the ion-exchanger and the drug.

Ampholytic drugs, possessing simultaneously a positive and a negative charge (zwitterions), carry inherently a net charge of zero but may also be able to form an

(23)

intrinsic molecular salt, which may reduce their loading into the ion-exchangers (Kankkunen et al., 2002b). Overall, release of zwitterionic drugs from the ion-exchange material is very difficult to control.

Some researchers have also found that the relative affinity (i.e. selectivity) of ion- exchangers for different drugs may depend on the steric structure of the charged functional groups of the drugs (Borodkin and Yunker, 1970; Kril and Fung, 1990). This could presumably be attributed to the different accessibility of the drugs to the charged location of the ion-exchanger. Drugs possessing lesser sterical hindrance have higher selectivity towards the ion-exchanger.

Molecular size

Analogously to the increase of the cross-linking in the ion-exchangers (described above), the increase of molecular size may achieve lower drug loading into the ion-exchangers, especially in the case of highly cross-linked resins, as the larger drugs are unable to diffuse into the inner part of the ion-exchanger (‘sieve effect’) (Kunin, 1949; Kunin and Myers, 1949; Irwin et al., 1987; Farag and Nairn, 1988). Additionally, the diffusion rate of the larger drug-ions is slower compared to the smaller ones, retarding the ion-exchange (Abdekhodaie and Wu, 2006). It has been observed that as the molecular weight of the drug-ion increases, more water is displaced from the resin beads during the drug loading process, leading to a decrease in the moisture content within the resin and a smaller self- diffusion coefficient of the drug (Farag and Nairn, 1988). On the other hand, the strength of the formed interactions between the drugs and the ion-exchangers may increase with increasing the size of the organic counter-ions due to the increased possibility of additional attractive forces (e.g. hydrophobic interactions).

2.3.3 External conditions

Concentration and valence of external ions

The Donnan potential between the ion-exchanger and external solution phases depends strongly on the concentration of the surrounding ions, which then affects the rate and extent of ion-exchange (see 2.3.1) (Helfferich, 1995; Jaskari et al., 2001; Kankkunen et al., 2002b; Abdekhodaie and Wu, 2006). In concentrated external solutions, the Donnan potential between the phases is smaller and, therefore, the interactions between the counter-ions (e.g. drug) and the fixed ion-exchange groups of the ion-exchanger are weaker, facilitating the drug release. Similarly, the Donnan potential is smaller with counter-ions of high valence and co-ions of low valence; the sequence of decreasing Donnan potential in the case of anion-exchange is, for example, Na2SO4 < NaCl < CaCl2

(Helfferich, 1995). At higher valence, the extracting counter-ions are more strongly attracted to the ion-exchanger (electroselectivity), enabling more efficient drug release (Sawaya et al., 1988; Bhandari et al., 1993; Helfferich, 1995; Jaskari et al., 2001;

(24)

separate functional groups, the multivalent ions may partly cross-link the ion-exchanger and, therefore, prevent the drug release. Strong association of the multivalent counter-ion with the ion-exchange groups can also reduce or even reverse the sign of the effective charge of the ion-exchanger, which will naturally affect the Donnan potential and, thus, the drug release (Jaskari et al., 2001).

pH of surrounding solution

The pH of the external solution may have a strong effect on the ionisation of both the drug and the ion-exchanger, which in turn affects the ion-exchange (Borodkin and Yunker, 1970; Åkerman et al., 1998; Kankkunen et al., 2002b). A sharp decrease of drug binding into the weak carboxylic acid (pKa ~ 4) ion-exchangers was observed below the pH of 4.5, being negligible at pH 2, as the number of anionic binding sites on the resin/fiber was diminished (Borodkin and Yunker, 1970; Kankkunen et al., 2002b). Moreover, decreasing interaction between the resin and some amine drugs was observed above pH 5.5, despite the greater ionisation degree of the –COOH groups. This can be attributed to the increasing portion of unionised form of the drug. On the other hand, drug loading into the strong sulphonic acid (pKa < 1) fiber was efficient at both pH 2 and 7.4, as the fiber remained in ionised form also at the acidic pH (Kankkunen et al., 2002b).

Temperature and agitation

Higher rates of drug incorporation into and release from the ion-exchangers have been observed with increasing temperature and/or agitation (Borodkin and Yunker, 1970;

Sawaya et al., 1988; Irwin et al., 1990; Chen et al., 1996). The diffusion rate of ions into the ionic binding sites of the ion-exchanger gets higher as the temperature is increased.

Additionally, higher temperatures may increase the pore-size of cross-linked resins (Borodkin and Yunker, 1970; Irwin et al., 1990). When the drug was loaded into the ion- exchange resin at elevated temperature and thereafter released at lower temperature, the rate of drug release was observed to be lower compared to the situation where the drug was both loaded and released at the lower temperature, despite higher initial drug loading content in the former case (Irwin et al., 1990). The authors suggest that lowering the temperature after the loading process causes a reduction in the pore diameter of the resin and leads, thus, to the entrapment of large ions in the resin framework. An increase of the stirring rate will decrease the thickness of the unstirred liquid diffusion layer adherent to the surface of the ion-exchanger, achieving a higher rate of ion-exchange (Sawaya et al., 1988; Irwin et al., 1990).

(25)

2.4 PHARMACEUTICAL APPLICATIONS OF ION-EXCHANGERS

2.4.1 Medicinal use as pharmacologically-active ingredient

Certain synthetic ion-exchange resins can be utilised as therapeutic agents in the treatment of various diseases, such as hypercholesterolemia, hyperkalemia, hypertension, cardiac edema and toxemia of pregnancy (Anand et al., 2001; Elder, 2005). In addition, they may be used in drug overdoses and other poisoning conditions. The pharmacological activity of the ion-exchangers is attributed simply to their ability to bind oppositely charged ions. For example, anion-exchange resins, cholestyramine (Questran^®) and cholestipol (Lestid^®), which are used to treat hypercholesterolemia, bind anionic bile acids in the gastrointestinal (GI) tract. Due to the large size of the formed polymer-bile acid complexes, the reabsorption of bound bile acids from the gut into the bloodstream is prevented and, thus, the cholesterol levels of the body are decreased due to the increased use of cholesterol in the synthesis of bile acids.

When using the ion-exchange materials as therapeutic agents, it should be noted that the administration of large quantities of ion-exchangers via the oral route can disturb the composition of body fluids, reduce the absorption of essential nutrients (e.g. vitamins) and drugs into the body and cause, thus, even severe side effects (Ranade, 1990). It should also be recognised, that the duration of action of the orally administered ion-exchangers may vary considerably from one patient to another, as the transport of solid dosage forms through the GI tract is not a standardized process.

2.4.2 Drug stabilization

To obtain a therapeutically effective drug delivery system, it is important to produce a system, which will not only achieve an optimal drug loading and release rate, but also maintains the drug activity within the system (Chen et al., 1992). Ion-exchangers may provide a good choice in maintaining/enhancing the stability of drugs and other medicinal products, such as vitamin B12, omeprazole, nicotine and levodopa, by protecting them from hydrolysis, oxidation or degradative enzymes during the storage and compound delivery (Ranade, 1991; Anand et al., 2001; Kankkunen et al., 2002b; Hughes, 2004;

Elder, 2005). Depending on the application, the drug/vitamin loaded ion-exchange material may be kept in a wet state or dried after loading. In the case of external drug delivery systems like transdermal patches, exposure to the extracting salt might occur immediately prior to the application of the medication to a patient (Kankkunen et al., 2002b).

The use of ion-exchange resins to stabilize vitamin B12 (cyanocobalamin) was discovered as early as 1958 and such products are still sold commercially (Hughes, 2004;

(26)

the resin is shown to be protected from the action of acidic gastric juices after oral delivery and will, therefore, pass unchanged through the stomach into the intestinal tract for absorption. Thus, the biological activity of the vitamin remains unaffected until the vitamin is releasedin vivo.

The buffer-like behaviour of the ion-exchange materials may also improve drug stability in aqueous solutions by holding the pH of the surrounding solution constant (Chen et al., 1992; Kankkunen et al., 2002b). Levodopa, the primary drug used in the treatment of Parkinson’s disease, is easily oxidized in the presence of water, especially under high pH conditions. When the skin permeation of levodopa was studied from the ion-exchange fiber (Na⁺-form) and solution formulations, significant oxidation of levodopa was observed in the solution formulation, whereas in the ion-exchange fiber formulation levodopa remained stable due to the fiber-buffered release environment (Kankkunen et al., 2002b). Chen et al. (1992) observed that the counter-ion of the ion- exchanger might have an influence on the buffering effect of the resin and, thus, affect the stability of the loaded drug. Doxorubicin loaded into the the ion-exchange resin, containing H⁺-ions as counter-ions (H⁺-form), was degraded by acid hydrolysis due to the resin’s strong acidic nature, while loading the drug into the Na⁺-form of the same resin maintained the stability of doxorubicin.

Recently, it has been shown that drug-resin complexes can reduce excess water uptake of highly hygroscopic drugs during the manufacture and storage, even after exposure to ambient air, which improves the drugs’ flowing properties and permits their formulation into typical dosage forms (e.g. capsules and tablets) with standard equipments (Hughes, 2004; Elder, 2005).

2.4.3 Taste masking

Many therapeutically useful drugs have quite a bitter taste, which limits their utility in oral formulations. The fact that drug release from ion-exchange materials is highly dependent on the physiological pH and electrolyte concentration within the GI tract can be applied for taste masking of drugs (Borodkin and Sundberg, 1971; Lu et al., 1991; Agarwal et al., 2000; Pisal et al., 2004). Typically, the ionised drug and the ion-exchanger form a stable complex under buccal conditions (pH 6.7, low ion concentration) for the relatively short period of exposure, making the drug unavailable for taste sensation. As the formulation passes to the further parts of the GI tract, that is, under gastric and intestinal conditions, the drug is released from the ion-exchanger into the surrounding media due to the (1) decreased pH in the stomach, (2) increased ionic concentration of the GI tract, (3) larger volume of the surrounding media and/or (4) increased residence time in the stomach and intestine and is, thus, available for absorption. Efficient taste protection by ion-exchange resins has been demonstrated with a variety of drugs, such as dextromethorphan, ephedrine, pseudoephedrine, ranitidine, ciprofloxacin, erythromycin and clarithromycin (Borodkin and Sundberg, 1971; Lu et al., 1991; Anand et al., 2001; Pisal et al., 2004;

Elder, 2005). Unlike with many other approaches used in taste masking, ion-exchange

Characterization of Ion-Exchange Fibers for Controlled Drug Delivery