INFLUENCE OF ADDITIVES ON ETHYLCELLULOSE COATINGS
ONG KANG TENG (B.Sc. (Pharm.)(Hons.), NUS)
A THESIS SUBMITTED
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
DEPARTMENT OF PHARMACY
NATIONAL UNIVERSITY OF SINGAPORE
2006
ACKNOWLEDGEMENTS
With great gratitude, I wish to thank my respectful supervisors, Associate
Professor Paul Heng Wan Sia and Associate Professor Chan Lai Wah for devoting much time and effort in supervising and guiding me in my higher degree pursue. They have given me an education that is worth a lifetime.
I could not have carried out my study without the generous financial supports from the National University of Singapore and the use of research facilities in the
Department of Pharmacy and GEA-NUS Pharmaceutical Processing Research
Laboratory. I wish to thank the laboratory officers in the Department of Pharmacy, especially Teresa and Mei Yin, who have never failed to render me their technical assistance whenever needed. My heartfelt thanks go to all my friends and colleagues in the Department of Pharmacy and GEA-NUS, especially Celine, Tin Wui, Chit
Chiat, Liang Theng, Sze Nam, Gu Li, Wai See and Qiyun. They have not only shared with me their valuable experiences but also provided me with their friendship.
My parents and siblings have showered me with much love and supports which kept me going, especially through difficult times. This journey has been blessed with many prayerful and spiritual supports of saints both at home and aboard, for whom I thank God for. All thanks and glory be to God the Father and Lord Jesus
Christ, to whom I owe all things.
Kang Teng
Jan 2006
i TABLE OF CONTENTS
CONTENTS
ACKNOWLEDGEMENTS i
CONTENTS ii
SUMMARY vii
List of Tables xii
List of Figures xiv
I. INTRODUCTION 1
A. Film coating 1
1. Reasons for coating 1
2. Film coating polymers 1
3. Organic versus aqueous coating systems 4
4. Aqueous coating systems 5
a. Latex and pseudolatex 5
i. Emulsion polymerization 5
ii. Emulsion - solvent evaporation 5
iii. Phase inversion 6
iv. Solvent change 6
b. Suspension 7
B. Mechanism of film formation 7
C. Substrates for film coating 9
D. Drug release mechanisms of coated pellets 10
1. Diffusion - controlled systems 11
2. Swelling - controlled systems 14
3. Chemically - controlled systems 15
4. Osmotically - driven release systems 16
ii TABLE OF CONTENTS
E. Performance of film-coated products 17
1. Substrate 17
2. Coating formulation 19
a. Nature of the polymer 19
i. Cellulose derivatives 20
ii. Acrylic polymer 22
b. Additives 23
i. Plasticizers 23
ii. Colorants and opacifiers 29
iii. Surfactants 30
iv. Anti-tack agents 30
v. Hydrophilic additives 31
3. Coating process variables 31
F. Analysis and comparison of dissolution data 35
1. Zero order equation 36
2. First order equation 37
3. Hixson – Crowell equation 37
4. Higuchi equation 38
5. Baker and Lonsdale equation (Higuchi's model for spherical matrices) 40
6. Hopfenberg equation 40
II. OBJECTIVES 45
Part 1. 46
Part 2 47
III. EXPERIMENTAL 50
A. Materials 50
iii TABLE OF CONTENTS
1. Drugs, polymers and additives 50
B. Methodology 53
1. Preparation of film forming dispersions 53
a. EC and acrylic dispersions containing plasticizers 53
b. EC dispersions containing polymeric additives 53
2. Preparation of films 53
3. Evaluation of film properties 54
a. Surface morphology 54
b. Film transparency 55
c. Mechanical properties 55
d. Puncture test 56
e. Plasticizer content 57
f. Percent weight change of film 58
g. Moisture content 58
h. Water vapour permeability 59
i. Thermal properties 60
Glass transition temperature 60
Thermal mechanical spectra 61
j. Drug permeability 62
k. Swelling and leeaching of film 65
4. Preparation of pellets 66
5. Coating of pellets 67
a. Coating with Aquacoat 67
b. Coating with Surelease 68
6. In vitro dissolution studies 69
iv TABLE OF CONTENTS
7. Assay of drug content in coated pellets 69
IV. RESULTS AND DISCUSSION 70
Part 1. Influence of plasticizers and storage conditions on properties of films 70
A. Film morphology 70
B. Mechanical properties of films 73
1. Comparison between different polymeric films 73
2. Influence of plasticizers on properties of films 74
3. Influence of storage time on properties of films 75
4. Influence of storage humidity on properties of films 84
5. Water vapour permeability 87
C. Drug release 88
1. Influence of plasticizer on dissolution profiles of pellets coated with Aquacoat 88
2. Influence of storage conditions on dissolution profiles of pellets coated with plasticized Aquacoat 92
a. Citric acid esters (TEC and ATEC) 92
b. Phthalic acid esters (DEP) 101
c. Glycerol acid ester 104
3. Effect of storage temperature on drug release 107
Part 2. Influence of polymeric additives 109
A. Thermal and dynamic mechanical properties 109
B. Film morphology 121
C. Mechanical properties - tensile test 126
D. Mechanical properties - puncture test 131
E. Water vapour permeability 135
F. Drug permeability 137
v TABLE OF CONTENTS
1. Effect of polymer additives on drug permeability 138
2. Effect of drug properties on drug permeability 147
G. Correlation between physicomechanical properties and permeability of composite EC films 149
H. Release kinetics of pellets coated with EC and polymeric additives 151
1. Effect of coating levels on drug release from EC - coated pellets 151
2. Effect of curing on EC - coated pellets 161
3. Influence of PVP on EC - coated pellets 164
4. Effect of curing on drug release from EC-PVP coated pellets 181
5. Influence of PV/VA on drug release from EC - coated pellets 183
V. CONCLUSION 188
VI. REFERENCES 193
VII. APPENDICES 207
Symbols 207
Published communications and presentations 211
vi SUMMARY
SUMMARY
Many polymeric film coats applied onto dosage forms have been reported to undergo changes in mechanical properties upon storage. The extent of these changes is influenced by several factors, such as the amount of plasticizers added, type of plasticizers used, film forming conditions, film storage temperature and humidity.
Changes in film mechanical properties may ultimately influence the drug release, stability and other physicochemical properties of the coated dosage forms.
Ethylcellulose and acrylates are among the most commonly used polymers in the production of coated controlled-release dosage forms. Several researchers have studied the effects of different types of plasticizers on the mechanical properties of
Aquacoat films. Plasticizers, such as dibutyl sebacate, tributyl citrate, acetyl tributyl
citrate and oleyl alcohol were found to produce ethylcellulose films that showed greater elongation upon stretching, after the films had been stored under conditions of
elevated humidity. However, the actual mechanisms that caused the above change and
the extent of influence by the different types of plasticizers have not been reported.
The primary objective of this study was to investigate the effects of different types of
plasticizers on the stability and other properties of the films exposed to different
storage conditions. Attempts were made to elucidate the mechanisms responsible for the changes observed and correlate these changes in film properties to the release profiles of coated pellets.
This study demonstrated that ethylcellulose film stability is influenced by
several factors, including type and amount of plasticizers remaining in the film, as
well as the storage conditions. Plasticizers interact with ethylcellulose and affect its film properties primarily in the following three ways. Firstly, plasticizers with low permanence, such as glycerin triacetate and diethyl phthalate can cause formation of
vii SUMMARY brittle films as these plasticizers are volatile or degrade on extended storage.
Secondly, the extent of coalescence or ageing on ethylcellulose films varies with different plasticizers. In addition, the stability of ethylcellulose films under different storage conditions is also dependent on type of plasticizers used.
At a commonly applied concentration of 30 percent, the citrate ester plasticizers, particularly triethyl citrate, have been shown to interact well with
Aquacoat, while glycerin triacetate and diethyl phthalate were not able to plasticize
Aquacoat films as effectively. Exposing the film membrane to low humidity or high temperature accelerates loss of bound water, thus enhancing the coalescence or fusion of polymer molecules. Storage environments with high moisture content, on the other hand could delay and reduce the coalescence of films but not prevent it. With the exception of glycerin triacetate, chlorpheniramine release from pellets coated with plasticized Aquacoat films were affected to varying degrees by storage conditions.
Higher storage temperatures tend to cause greater changes for pellets coated with
Aquacoat containing citric acid class of plasticizers (triethyl citrate and acetyl triethyl citrate) compared to other plasticizers. This study demonstrated that the physicochemical properties of the plasticizer coupled with the storage conditions have an important influence on the stability and performance of the final film coat. Hence it is important to exercise caution in the selection of plasticizers for film coating in order to ensure good product stability and performance.
Ethylcellulose is used in controlled released preparations because of its good mechanical properties and poor permeability to water vapour. However, these properties strongly retard drug release and thereby limit the application of pure ethylcellulose coating as controlled release coating. Studies were undertaken to modify drug permeability through ethylcellulose coatings by reduction in thickness of
viii SUMMARY the coating layer, formation of pores using organic solvents and hydrophilic additives.
Polyvinylpyrrolidone is a water - soluble, physiologically inert synthetic polymer consisting essentially of linear 1-vinyl-2-pyrrolidinone groups, with varying degree of polymerization which results in polyvinylpyrrolidone of different molecular weights.
Viviprint 540 is a molecular – composite polyvinylpyrrolidone, which is formed by in situ incorporation of insoluble crosslinked polyvinylpyrrolidone nanoparticles into soluble, film–forming polyvinylpyrrolidone polymer. Hence molecular – composite polyvinylpyrrolidone has much larger molecular weight than polyvinylpyrrolidone and is less soluble in water. Plasdone S-630 copolyvidonum is a synthetic water- soluble copolymer consisting of N-vinyl-2-pyrrolidone and vinyl acetate in a random
60:40 ratio. All the water-soluble polymers discussed above are potential polymeric film modifiers for achieving improved drug release. However, to date, their potentials have not been explored.
In this study, the interaction between ethylcellulose and polyvinylpyrrolidone was found to be dependent on the molecular weight, concentration and chemical nature of the additives. When added to ethylcellulose, low molecular weight polyvinylpyrrolidone would be randomly distributed in the ethylcellulose matrix as a disperse phase. Increased concentration up to 30 %w/w did not alter the phase distribution. In contrast, greater interaction was exhibited between ethylcellulose and higher molecular weight polyvinylpyrrolidone, such as polyvinylpyrrolidone K60,
K90 and molecular – composite polyvinylpyrrolidone. At low concentration, higher molecular weight polyvinylpyrrolidone might exist as a disperse phase in the ethylcellulose matrix. However, as the concentration increased, the higher molecular weight additives tended to aggregate and formed a separate continuous phase.
Formation of separate continuous layers became more prominent with increasing
ix SUMMARY concentration. Increased molecular weight of additives also accelerated the formation of separate layers.
Addition of polyvinylpyrrolidone increased the glass transition temperature, water vapour and drug permeabilities, as well as strengthened the mechanical properties of composite ethylcellulose films. However, the magnitude of change was not consistent in all cases. For instances, the percent increases in glass transition temperature and tensile strength were relatively small (~ 27 percent), while greater increments (~ 76 – 80 percent) were observed for the elastic modulus and puncture strength of composite ethylcellulose films. Presence of polyvinylpyrrolidone further accentuates ethylcellulose film permeability to water vapour (~ 80 - 177 perecent) and brought about 1000 times or higher permeability to drugs. Moreover, while a linear relationship was found between the molecular weight of polyvinylpyrrolidones and permeability to chlorpheniramine, the thermal and mechanical properties of composite ethylcellulose films did not showed similar response to increasing molecular weight of polyvinylpyrrolidones.
Though addition of polyvinylpyrrolidones to ethylcellulose resulted in increase in glass transition temperature, tensile strength, elastic modulus and puncture strength of resultant film, these changes did not correlate with a reduction in permeabilities as suggested by other studies. This anomaly could be due to properties of additives present. Polyvinylpyrrolidone is a water-soluble and hygroscopic polymer which may increase the drug permeability through formation of pores resulting from rapid diffusion of polyvinylpyrrolidone into the surrounding diffusion medium.
Similar trends were observed for changes in physicomechanical and physicochemical and release rate from the composite ethylcellulose films regardless of the properties of
x SUMMARY drugs used. This suggested that influence of properties of adjuvants was the dominant factor that affects the permeability of composite ethylcellulose films.
xi LIST OF TABLES
LIST OF TABLES
Table 1. Polymers commonly used for film coating 3 Table 2. Physical properties of plasticizers used in film coating. 25 Table 3. Influence of various processing parameters in pellet coating 34 Table 4. Chemical structures, molecular weights, solubility in water and acidity/basicity of drugs used. 51 Table 5. Chemical structures, typical molecular weights and viscosity of polymeric additives. 52 Table 6. Comparison of mechanical properties, film transparency and water vapour permeability of EC and MA films. 74 Table 7. Comparison of plasticizer content in Aquacoat films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30 °C. 80 Table 8. Comparison of percent weight change of films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30°C. 81 Table 9. Pearson correlation for films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30 °C. 82 Table 10. Comparison of percent change in moisture content of films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30 °C. 85 Table 11. Fit factors of drug release models for pellets coated with Aquacoat containing 30 %w/w TEC, DEP, GTA or ATEC. 90 Table 12. Lag time, MDT50%, T50% and release rate constants of pellets coated with Aquacoat plasticized with 30 %w/w TEC, DEP, GTA and ATEC. 90 Table 13. Difference factor, f1 and similarity factor, f2 of pellets coated with Aquacoat. 96 Table 14. Glass transition temperature (Tg), film transparency and mechanical properties of Surelease (EC) and composite Surelease films. 117 Table 15. Mechanical properties of dry and wet Surelease (EC) and composite EC films. 132 Table 16. Permeability and correlation coefficients of Surelease (EC) and composite EC films to chlorpheniramine. 140 Table 17. Drug permeability coefficients of Surelease (EC) and composite EC films. 141 Table 18. Fit factors of drug release models for chlorpheniramine pellets coated with different level of EC. 153 Table 19. Fit factors of drug release models for theophylline pellets coated with different levels of Surelease. 157 Table 20. Fit factors of drug release models for paracetamol pellets coated with different levels of EC. 158 Table 21. Difference factors, f1, and similarity factors, f2, for chlorpheniramine (CPM), paracetamol (PA) and theophylline (THE) pellets coated with different levels of EC. 162 Table 22. r2 values of chlorpheniramine pellets with composite Surelease (EC) coat cured at 60 °C, 24 h. 168 Table 23. AIC values of chlorpheniramine pellets coated with composite Surelease (EC) coat cured at 60 °C, 24 h. 169
xii LIST OF TABLES
Table 24. r2 values of theophylline pellets coated with composite Surelease (EC) coat cured at 60 °C, 24 h. 170 Table 25. AIC values of theophylline pellets coated with composite Surelese (EC) coat cured at 60 °C, 24 h. 171 Table 26. Statistical analysis on influence of curing on chlorpheniramine release from pellets coated with composite Surelease (EC). 181 Table 27. Fit factors of drug release models for chlorpheniramine pellets coated with composite Surelease (EC) containing 30 %w/w of PV/VA cured at 60 °C, 24 h 186 Table 28. Hopfenberg release rate constants of chlorpheniramine (CPM) and theophylline (THE) pellets coated with composite Surelease (EC) containing 30 %w/w of PV/VA cured at 60 °C, 24 h 186 Table 29. Fit factors of theophylline pellets coated with composite Surelease (EC) containing 30 %w/w of PV/VA cured at 60 °C, 24 h 187
xiii LIST OF FIGURES
LIST OF FIGURES
Figure 1. Film coating of pellets. 2 Figure 2. Film formation from pseudolatex. 8 Figure 3. Coalescence of particles during the evaporative phase. 9 Figure 4. Drug release from coated pellets, (a) diffusion - controlled through porous coat, (b) diffusion – controlled through non – porous coat, (c) swelling – controlled, (d) chemically – controlled and (e) osmotically - driven 13 Figure 5. Chemical structures of cellulose derivatives. 21 Figure 6. Chemical structures of acrylic polymer 23 Figure 7. A typical load - strain profile of film undergoing tensile test. 56 Figure 8. SPM images of (a) MA, (b) Surelease, and Aquacoat films plasticized with (c) DEP and (d) DBP. 72 Figure 9. Effect of storage conditions on tensile strength of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. 76 Figure 10. Effect of storage conditions on % elongation at break of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. 77 Figure 11. Effect of storage conditions on elastic modulus of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. 78 Figure 12. Effect of storage conditions on work of failure of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH.) 79 Figure 13. Dissolution profiles of Aquacoat coated chlorpheniramine pellets dried at 22 °C, 70 %RH for 24 h. 89 Figure 14. Dissolution profiles of chlorpheniramine pellets coated with Aquacoat containing 30 %w/w TEC stored at (a) 75 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 10 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 953 Figure 15. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat containing 30 %w/w TEC 94 Figure 16. Dissolution profiles of chlorpheniramine pellets coated with Aquacoat containing 30 %w/w ATEC stored at (a) 10 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 75 %RH, 30 ºC and (d) 0%RH, 70 ºC 98 Figure 17. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat containing 30 %w/w ATEC. 99 Figure 18. Dissolution profiles of pellets coated with Aquacoat containing 30 %w/w DEP stored at (a) 75 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 10 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 102
xiv LIST OF FIGURES
Figure 19. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat containing 30 %w/w DEP. 103 Figure 20. Dissolution profiles of pellets coated with Aquacoat plasticized with 30 %w/w GTA stored at (a) 75 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 10 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 105 Figure 21. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat plasticized with 30 %w/w GTA. 106 Figure 22. DSC thermograms of (a) EC (b) EC-PVP K29 (9:1) (c) EC- PVP K29 (8:2) (d) EC-PVP K29 (7:3) (e) EC-PVP K29 (6:4) (f) EC-PVP K29 (5:5) (g) EC-PVP K29 (4:6) (h) EC-PVP K29 (3:7) (i) EC-PVP K29 (2:8) (j) EC-PVP K29 (1:9) (k) PVP K29 111 Figure 23. DSC thermograms of (a) EC (b) EC-MCPVP (9:1) (c) EC- MCPVP (8:2) (d) EC-MCPVP (7:3) (e) EC-MCPVP (6:4) (f) EC-MCPVP (5:5) (g) EC-MCPVP (4:6) (h) EC-MCPVP (3:7) (i) EC-MCPVP (2:8) (j) EC-MCPVP (1:9) (k) MCPVP 112 Figure 24. DSC thermograms of (a) EC (b) EC-PV/VA (9:1) (c) EC- PV/VA (8:2) (d) EC-PV/VA (7:3) (e) EC-PV/VA (6:4) (f) EC-PV/VA (5:5) (g) EC-PV/VA (4:6) (h) EC-PV/VA (3:7) (i) EC-PV/VA (2:8) (j) EC-PV/VA (1:9) (k) PV/VA. 112 Figure 25. DMA thermograms: (a) tan δ (b) loss modulus profiles of Surelease (EC) and EC-PVP K17 films. 116 Figure 26. Light microscope images of Surelease (EC) and composite Surelease films: (a) EC, (b) EC-PVP K17 (9:1), (c) EC-PVP K29 (9:1), (d) EC-PVP K90 (9:1), (e) EC-MCPVP (9:1) and (f) EC-PV/VA (7:3). 122 Figure 27. Effect of type and concentration of polymeric additives on the film transparency of Surelease films. 122 Figure 28. Models of interaction between PVP, MCPVP or PV/VA with Surelease (EC): (a) continuous phase of EC film without polymeric additives (b) minor component existing as random spheres/crevices in the continuous phase of EC (c) minor component existing as random spheres/crevices of larger size in the continuous phase of EC (d) minor component forming a separate continuous phase (e) overcoat of one continuous phase over another. 124 Figure 29. Water vapour permeability of Surelease (EC) and composite EC films 136 Figure 30. Release profiles of composite Surelease (EC) films to chlorpheniramine. 141 Figure 31. Relationship between permeability constant and molecular weight of PVP. 142 Figure 32. (a) Amount of extracted component and (b) swelling index of EC and composite EC films containing 30 %w/w of additives. 145
xv LIST OF FIGURES
Figure 33. Light microscope images of wetted composite EC films: (a) EC-PV/VA (7:3), (b) EC-PVP K29 (7:3), (c) EC-PVP K60 (7:3), (d) EC-PVP K90 (7:3) 146 Figure 34. Dissolution profiles of pellets coated with different levels of Surelease. 153 Figure 35. Effect of coating level of Surelease on (a) Baker-Lonsdale release rate constant (b) T50% (c) MDT50% of chlorpheniramine, theophylline and paracetamol pellets. 155 Figure 36. Dissolution profiles of theophylline pellets coated with different levels of EC cured at 60 °C for 24 h or at 22 °C for 48 h. 155 Figure 37. Dissolution profiles of paracetamol pellets coated with different levels of Surelease cured at 60 °C for 24 h or at 22 °C for 48 h. 156 Figure 38. Relationship between (a) Baker-Lonsadale release rate constant (b) MDT50% (c) T50% and drug solubility. 161 Figure 39. Dissolution profiles of chlorpheniramine pellets with Surelease (EC) and composite EC coats containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of PVP K17, K29, K60, K90, MCPVP and PV/VA cured at 60 °C, 24 h. 165 Figure 40. Dissolution profiles of theophylline pellets with Surelease (EC) and composite EC coats containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of PVP K17, K29, K60, K90, MCPVP and PV/VA cured at 60 °C, 24 h. 167 Figure 41. Influence of molecular weight of PVP on the (a) Hopefenberg release rate constant (b) T50% of chlorpheniramine pellets and (c) MDT50% of theophylline pellets coated with Surelease, cured at 60 °C. 24 h. 173 Figure 42. Extractable amount of Surelease (EC) and composite EC films containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of additives. 176 Figure 43. Swelling index of Surelease (EC) and composite EC films containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of additives. 177 Figure 44. Dissolution profiles of chlorpheniramine pellets coated with Surelease (EC) and composite EC containing 10 - 30 %w/w (a) PVP K29, (b) PVP K90 and (c) MCPVP, cured at 60 °C, 24 h or uncured. 183 Figure 45. Dissolution profiles of chlorpheniramine pellets coated with composite EC containing 30 %w/w PV/VA, cured at 60 °C for 24 h or uncured. 185 Figure 46. Dissolution profiles of theophylline pellets coated with composite Surelease (EC) coating containing 30 %w/w PV/VA, cured at 60 °C for 24 h. 185
xvi INTRODUCTION
I. INTRODUCTION
A. Film coating
Film coating is an important technique in the pharmaceutical industry for
preparation of oral dosage forms (Lachmann et al., 1986). It is a process involving the
application of a thin (20 to 200 µm) polymer-based coating on an appropriate
substrate, such as tablets, beads, granules, capsules, drug-powder and crystals (Figure
1).
1. Reasons for coating
Over the years, there is increasing popularity for film coating. The reasons for film coating are many and varied (Porter et al., 1991). Generally, film coating can be categorized as those used for modified released and those used to improve product appearance, to mask unpleasant taste, as well as to protect against moisture and light.
In contrast, functional coating is used to modify the release of drug.
2. Film coating polymers
Polymers used for general or non release related coating are usually readily soluble in the gastrointestinal tract and do not influence drug release significantly.
They are usually the water soluble cellulose ethers such as methyl cellulose,
hydroxypropyl methyl cellulose, hydroxyethyl cellulose and hydroxypropyl cellulose
(Table 1).
Enteric coating and controlled released coating are examples of modified
released coating. Enteric coating polymers are usually soluble above a certain pH
such as pH 5.0. They remain insoluble in the stomach but start to dissolve once the dosage form reaches the small intestine. The most commonly used enteric coatings
1 INTRODUCTION
Droplets of coating liquid
Liquid penetration into core Spray nozzle
Coalescence of coating liquid to form film
Figure 1. Film coating of pellets. (Adapted from Porter, 1991)
are formulated with synthetic polymers that contain ionizable functional groups that render the polymer water soluble at specific pH values. Such polymers are often referred to as “poly-acids”. Examples of commonly used enteric coating polymers are listed in Table 1.
Polymers used for controlled release coating are usually insoluble or poorly soluble in water. Drug release is governed by drug diffusion, polymer dissolution, erosion and/or osmosis. For drug diffusion, the dosage form is coated with a water - insoluble polymer (e.g. ethylcellulose) alone or in combination with a water - soluble ingredient (e.g. hydroxypropyl methylcellulose) such that the gastrointestinal fluids can permeate through the film and dissolve the drug enabling it to diffuse out at a rate dependent on the physicochemical properties of both drug and film coating
(Sakellariou et al., 1995). For polymer dissolution or erosion, the dosage form is coated with either a sparingly soluble or pH dependent soluble film (e.g. cellulose acetate phthalate) such that the drug release will be dictated by the dissolution of the
2 INTRODUCTION polymer. For osmosis, the dosage form is coated with a semi- permeable membrane
(e.g. cellulose acetate) through which a small orifice is drilled. The gastrointestinal fluids pass across the membrane by osmosis, at a controlled rate determined by the
Table 1. Polymers commonly used for film coating
Category Class Polymer Form Application Methylcellulose (MC) Powder Hydroxypropyl methyl Powder cellulose (HPMC) Rapid Cellulose Hydroxypropyl ethyl release- derivatives Powder cellulose (HPEC) water Hydroxyethyl cellulose soluble Powder (HEC) Rapid drug release polymers Polyvinyl alcohol (PVA) Powder in gastrointestinal Vinyl Polyvinyl pyrrolidone tract derivatives Powder (PVP) Spray-dried Cellulose acetate Pseudolatex phthalate (CAP) powder Cellulose acetate Powder trimellitate (CAT) Cellulose Hydroxypropyl methyl derivatives cellulose phthalate Powder (HPMCP) Hydroxypropyl methyl cellulose acetate Powder Enteric succinate (HPMCAS) coating Polyvinyl acetate Spray-dried polymers Vinyl esters phthalate (PVAP) powder Drug release in Poly (methacrylic acid - Pseudolatex upper region of ethylacrylate) Spray-dried small intestine MA:EA = 1:1 latex Poly (methacrylic acid - Spray-dried Acrylates methylmethylacrylate) dispersible MA:MMA = 1:1 powder Poly (methacrylic acid - Spray-dried methylmethylacrylate) dispersible MA:MMA = 1:2 powder Cellulose Ethylcellulose (EC) Pseudolatex derivatives Poly (ethylacrylate - methylmethylacrylate) Latex EA:MMA = 2:1 Poly (ethylacrylate - Release of drug at Insoluble methylmethylacrylate – targeted sites of coating trimethylammonioethyl Latex gastrointestinal polymers Acrylates methacrylate chloride) tract. EA:MMA:TAMCl = 1:2:0.2 Poly (ethylacrylate - methylmethylacrylate – trimethylammonioethyl Latex methacrylate chloride) EA:MMA:TAMCl = 1:2:0.1
3 INTRODUCTION
permeability of the film, to dissolve the drug in the core. Hence, the core formulation
must include an osmotic agent. The dissolved drug is then released through the
orifice.
3. Organic versus aqueous coating systems
Generally, there are two types of coating systems – organic coating systems,
which involve the use of organic solvents and the aqueous type which uses water.
When film coating was first introduced, the polymers employed were only soluble in
organic solvents (Abbott laboratories 1953). Aqueous coating systems were less
preferred as the water contents of the coating solution were implicated as the cause of
stability problems and long processing time (Savage et al., 1995). The principal
benefits of using organic solvents were the considerable reduction in processing time
and absence of water from the process, thereby eliminating the loss of active
ingredient through hydrolysis. In addition, organic solvents can completely dissolve
the polymeric film formers, thereby allowing formation of smooth and continuous
coatings which were capable of protecting medicaments from environmental stresses
and making tablets more distinctive.
The three most common organo-soluble polymers (cellulose acetate,
ethylcellulose and methacrylic acid copolymer) used as sustained release membranes
were introduced to the industry before 1962. To date, there is no other organo-soluble
coating system that is as widely accepted by the industry. This is largely attributed to
the risks associated with organic solvent usage and improvement in aqueous coating systems. Stricter environmental legislation, in conjunction with the high cost of controlling organic solvent emission, has forced researchers to find alternative
‘environmentally friendly’ coating systems.
4 INTRODUCTION
The current USP lists only three sustained release coatings that function as a rate controlling membrane – cellulose acetate, ethylcellulose and methacrylic acid copolymer. Aqueous coating system is developed after the organo-soluble coating system. Hence, the current aqueous polymeric dispersion systems (e.g. Aquacoat®,
Surelease®, Eudragit®, Aquateric®) consist of mainly the 3 widely accepted polymers.
4. Aqueous coating systems
a. Latex and pseudolatex
These coating systems can be classified according to their manufacturing
processes. Generally, there are four methods of preparation, namely emulsion
polymerization, emulsion-solvent evaporation, phase inversion and solvent change
(Chang et al., 1987). Latexes have the disadvantage of being sensitive to several
stresses, such as electrolytes, pH change, storage temperature and high shear forces.
i. Emulsion polymerization
Latexes are colloidal polymer dispersions produced by emulsion
polymerization (Lehmann, 1997). After the monomers have been emulsified in water,
usually with the aid of anionic or nonionic surfactants, an initiator is added to induce
polymerization (Woods et al., 1968). The dispersions have a high solid content (up to
50 percent) but low viscosity. Examples are Eudragit NE 30D, Eudragit L100-55,
L100 and S 100 (Lehmann, 1997).
ii. Emulsion - solvent evaporation
Pseudolatex is prepared by dissolving a thermoplastic water-insoluble polymer in an organic solvent and emulsifying it with an aqueous solution of surfactant. The
5 INTRODUCTION organic phase is then removed by vacuum distillation, leaving an aqueous dispersion of polymer for coating application. If the polymer is prone to hydrolysis, the aqueous dispersion is spray dried (e.g. Aquateric, Aqoat, Sureteric). The spray - dried pseudolatex powder is redispersed in water containing surfactant, dye or other additives just before coating. Particle size of the emulsified polymer is an important factor affecting pseudolatex stability and subsequent film formation. Most pseudolatexes consist of spherical solid or semisolid particles less than 1 um in diameter, typically less than 0.1 µm (Wheatley and Steuernagel, 1997).
iii. Phase inversion
Surelease is an example of a pseudolatex prepared by phase inversion.
Production of this EC pseudolatex involved hot melt extrusion process, where the polymer is melted and the molten extrudate is then injected under pressure into ammoniated water. Plasticizers, such as dibutyl sebacate or fractionated coconut oil are added to reduce the melting temperature of EC, thereby preventing degradation of the polymer at high temperature. Phase inversion occurred when water in the ethylcellulose dispersion, formed upon extrusion of the plasticized hot melt mixture into water, inverted to EC in water dispersion (Porter, 1997).
iv. Solvent change
The solid polymers are synthesized by bulk polymerization. The polymer powder can be directly emulsified in hot water without further additives (emulsifiers), forming a stable aqueous dispersion. Examples of such systems are Eudragit RL 30D and Eudragit RS 30D.
6 INTRODUCTION
b. Suspension
Another aqueous coating system available is the suspension. It is prepared by
dispersing a micronized water - insoluble polymer in water containing plasticizer
(Keshikawa et al., 1994). An example of an aqueous suspension system is micronized
EC (N-10F, Shin Etsu Chemical Co., Ltd). There is little published literature on the
use of aqueous suspension systems, suggesting the lack of popularity of these
systems. Keshikawa et al. (1994) have shown that coating using an aqueous
ethylcelluose suspension eliminated the need for curing but a large amount of
plasticizer was usually required (up to 50%).
B. Mechanism of film formation
Film formation from aqueous polymeric dispersions requires the coalescence
of individual submicron-size polymer particles, each containing hundreds of polymer
chains, to form a continuous layer as the aqueous phase evaporates. Figure 2 shows a
pseudolatex consisting of polymer particles that are suspended and separated by
electrostatic repulsion. As water evaporates, interfacial tension between water vapour
and polymer pushes particles into point contact in a close-packed ordered array. A
strong driving force is necessary to overcome repulsive forces, to deform the particles,
and cause the particles to fuse, resulting in complete coalescence. Capillary forces,
generated by the high interfacial tension produced as water between polymer particles
evaporates, provide much of the energy required for film formation (Brown, 1956;
Bindschaedler et al., 1983). As illustrated in Figure 3, the polymer particles are pulled
closer together as the capillary forces build up with evaporation of water. The surface
tension of the spherical particle generated by the negative curvature of its surface can be described by Frenkel’s equation:
7 INTRODUCTION
3 γ t (1) θ 2 = N 2 π r
where θ is one-half the angle of coalescence (contact angle) at time t, γ is the surface tension, r is the radius of the particle, driving force (γ) necessary to fuse or coalesce the particles and N is the viscosity of the particles (Dillon et al., 1951). For this reason, a plasticizer is usually added to aid in the coalescence of the polymer particles. A plasticizer is a substantially nonvolatile, high-boiling and nonseparating substance that changes the physical and mechanical properties of the polymer to be plasticized (Banker et al., 1966). The addition of plasticizers is required for polymer dispersions that have a minimum film formation temperature above the
Aqueous dispersion deposited on surface
Water evaporation
Close packed particles with water filling voids
Water evaporation +
polymer deformation
Continuous polymer coating
Figure 2. Film formation from pseudolatex. (Adapted from Onions et al., 1986)
8 INTRODUCTION
Force exerted on three particles wet by water film
Force exerted on two particles wet water film
Water evaporation brings particles together
Fusion of deformable particles
Figure 3. Coalescence of particles during the evaporative phase (Adapted from Wheatley and Steuernagel, 1997).
coating temperature. During plasticization, the plasticizer will partition into and soften the colloidal polymer particles, thus promoting particle deformation and coalescence to form a homogeneous film (Bodmeier et al., 1997).
C. Substrates for film coating
Film coating is more commonly applied onto tablets and pellets. Pellets are multiparticulates, typically with size ranging between 0.5 - 1.5 mm, produced by an agglomeration process which converts fine powders or granules of bulk drugs and excipients into small, free-flowing, spherical or semi-spherical units (Ghebre-
9 INTRODUCTION
Sellassie, 1989). Pelletized products offer several advantages over single unit dosage forms such as maximizing drug absorption while reducing peak plasma fluctuations and potential side effects without compromising drug bioavailability (Bechgaard and
Nielson, 1978). Use of pellets can reduce intra- and inter-subject variability in gastro intestinal transit time, which is commonly observed for single unit dosage forms.
Pellets are drug delivery systems of choice, particularly when the active ingredients are inherently irritative or anesthetic, since they are less susceptible to dose dumping.
Combination therapy or delivery of two or more drugs in a single unit is also possible with pelletized products. Pellets are commonly prepared by depositing successive layers of drug from solution, suspension, or dry powder on preformed nuclei by a process known as layering (Iyer et al., 1993). Alternatively, they can be manufactured by extrusion-spheronisation (Nesbitt, 1994) where a wet mass of powder is extruded through a perforated screen to form cylindrical extrudates, which are then spheronised into pellets and dried.
D. Drug release mechanisms of coated pellets
For general coating, critical factors that may affect its performance include the morphology and opacity of the film and permeability of the film to moisture. In the case of modified released coating, factors that affect the release mechanism of the coated products should be controlled to ensure consistent product quality. Before considering what these factors may be, it is necessary to understand the mechanisms of release for coated dosage forms. Dosage forms with drug release that is mainly controlled by an applied film coat are commonly known as ‘membrane-controlled release systems’ or ‘reservoir systems’ (Porter, 1991). In a membrane-controlled system, the drug diffuses from the core through the rate-controlling membrane into
10 INTRODUCTION the surrounding environment. The rate of diffusion may depend on membrane porosity, tortuosity, geometry and thickness. In circumstances where film coat is insoluble in the dissolution media and in the absence of additives or where the influence of additives on drug release is negligible, transport of drug may occur through the following pathways:
• transport of drug through flaws in the membrane (as a result of stress-induced
cracks in film coat with poor mechanical properties), and
• transport of drug through pores that result from incomplete coalescence in
films prepared from aqueous polymeric dispersions.
Additives are commonly added to the polymer coating dispersions in order to improve the film coat quality as well as to modify the release of drugs. Under such conditions, drug release may be affected by the following phenomena:
• a network of capillaries (filled with dissolution media) created by leaching of a
water-soluble component from the film, and
• a hydrated, swollen film forms when the film coat contains a hydrophilic
component that cannot readily leach out.
Based on mechanisms of drug release, film-coated dosage forms can be broadly categorized as diffusion - controlled, swelling - controlled, osmotically - controlled and chemically - controlled systems.
1. Diffusion - controlled systems
In these systems, drug release is governed by molecular diffusion along a concentration gradient across the polymer film coat. The latter may be porous or nonporous. Porous controlled release systems contain pores that are large enough for diffusion of drug through water-filled pores of the coat (Figure 4a). The size of such
11 INTRODUCTION
pores is usually greater than 200-500 Å. In the nonporous systems, it is assumed that
the polymer forms a continuous phase (Figure 4b). The diffusion of a solute molecule
through a polymer phase is an activated process involving the cooperative movements
of drug and polymer chain segments around it. Thermal fluctuations of chain
segments allow sufficient local separation of adjacent chains to permit passage of a
drug. Another mechanism of release is configurational diffusion that involves the
movement of drug through the polymer chains. The rate of diffusion is therefore
dependent on polymer parameters such as degree of crystallinity, size of crystallites,
degree of cross-linking, swelling and molecular weight of the polymer.
The release rate of a non-porous system can be described by the following
equation (Ozturk et al., 1990):
P (C - C ) J = m s b (2) δ c
where J is the flux (release rate per unit surface area of coating), Cs and Cb are the
concentrations of drug at drug-coating interface and in the bulk, respectively, and δc is
the coating thickness. The permeability coefficient, Pm, of the coating polymer
membrane can be expressed as
Dε (3) P = = D ' K m τβ K
where D is the molecular diffusitivity of the drug, K is the distribution coefficient of
the drug between the polymer membrane and fluid in the core, ε the volume fraction
of the chain openings, β the chain immobilization factor, and τ the tortuosity factor.
The frequency with which a diffusion step occurs depends on (a) the size and shape of
the drug (b) the extent of packing and amount of attractive force between adjacent
12 INTRODUCTION
(a) (b)
K Cb Cs Drug Aqueous pore Drug
D’
(c) Swollen layer (d) Eroded coat
K K Drug
Drug
D’ D’
(e)
Drug
High π
Low π water influx
Figure 4. Drug release from coated pellets, (a) diffusion - controlled through porous coat, (b) diffusion – controlled through non – porous coat, (c) swelling – controlled, (d) chemically – controlled and (e) osmotically - driven
13 INTRODUCTION polymer chains, and (c) stiffness of polymer chains. The overall permeability of polymer membrane to drug will depend on the ability of the drug to partition into the polymer as well as its ability to diffuse through the polymer. This can be estimated from the solubility parameters of the drug and polymer. In general, the solution/diffusion mechanism is dominant in cases where the film coat is continuous
(lacks pores) and flexible, and where the drug has a high affinity for the polymer relative to water.
In the case of a porous system, the diffusion coefficient is often expressed as an effective diffusion coefficient, Deff, which is related to the diffusion coefficient of the drug through the pores filled with drug solution, Diw, as follows:
D ε (4) D = iw eff τ
As release occurs, more drug is lost, and the solution becomes more and more diluted.
Porosity is the fraction of pores in the system that is open (0 < ε < 1). The turtuosity,
τ, indicates the complexity of the pathways that constituent the pores. Diffusion in a porous system is dependent on the size of the drug and size of the pore.
2. Swelling - controlled systems
These systems consist of water-soluble drugs that are initially dispersed in solvent-free glassy polymers. When in contact with water, dynamic swelling of film coat leads to considerable volume expansion (Figure 4c). The swelling behaviour is characterized by two fronts:
• swelling interface that separates rubbery (swollen) state from glassy state and
moves inward with velocity, v
• polymer interface that separates rubbery state from water and moves outward.
14 INTRODUCTION
Swelling of glassy polymers is accompanied by macromolecular relaxation which affects drug diffusion through the polymer. The transport of drug through the polymer can be controlled by either the rate at which the macromolecular chains relax during transition from a glassy to rubbery state or by the diffusion of drug through the rubbery polymer. A dimensionless number known as Deborah, De, was employed by
Vrentas et al., (1984) to characterize this mode of drug transport
λ (5) De = θ d where λ is the relaxation time, and θ d is the diffusion time. When De >> 1 (i.e. when the transport is completely relaxed) or when De << 1 (i.e. when the transport is completely controlled by diffusion), Fickian behaviour is observed. When De ~ 1 (i.e. when the relaxation time is of the order of the diffusion time), anomalous diffusion behaviour is observed.
Water enters the film coat gradually, bringing about swelling in ‘layers’. The layers on the surface will swell first while the center is dry. The diffusional behaviour of water into the dosage form coated by a swelling film is described by Fujita equation:
Ddrug = Ddo exp{- α (β - cw)} (6)
where α and β are two characteristic constants of the polymeric system and cw is the water concentration in the system.
3. Chemically - controlled systems
Drug diffusion is controlled by the erosion of the polymer matrix (Figure 4d).
The polymer may undergo bioerosion and/or biodegradation. Biodegradation refers to
15 INTRODUCTION the breaking down of polymer due to chemical reaction. Bioerosion is a process whereby a phase of the carrier is lost, not by chemical reaction but by dissolution.
4. Osmotically - driven release systems
There is a possibility of release being driven by an osmotic pressure difference between core materials and release environment, when the coating is porous (Figure
4e). Sources of osmotic pressure in the core formulation include low molecular weight exicipients and the drug. However, for the drug to contribute significantly to osmotic pressure, it should be highly water soluble, of low molecular weight and present in a substantial dose (capable of achieving saturation concentration in the core). When the dosage form comes into contact with an aqueous environment, water is imbibed through the coating, creating a solution in the core. The osmoactive substances dissolve in the imbibed water and generate an interior osmotic pressure.
The osmotic pressure difference between core and external medium then provides a driving force for efflux, J, through pores in the coating. This transport phenomenon can be quantified by the following equation:
J = Kf σ ∆π (Ci - Cm) (7)
where Kf is the filtration coefficient, σ is the reflection coefficient of the coating, ∆π is the osmotic pressure difference across the coating, and Ci and Cm are the interior and media drug concentrations respectively.
Assuming that drug is released under sink conditions in the intestine, zero- order release is achieved only when (a) the materials responsible for maintaining the osmotic pressure difference are present in the core at concentrations above their solubilities (hence constant ∆π is maintained), and (b) the drug is present at a
16 INTRODUCTION concentration greater than its solubility. A semipermeable coating consists of presence of aqueous pores within the coating and core materials that are capable of generating sufficient osmotic pressure are necessary for attaining osmotically - driven release.
E. Performance of film-coated products
For ease of understanding, the factors that could influence the performance of film coatings, in particular the release properties of coated dosage forms. They are broadly classified into 3 areas: substrate, coating formulation, and coating process.
1. Substrate
In practice, while the release of drug from dosage forms may be membrane- controlled, the substrate is also critical in determining the behaviour of the final coated dosage form. Some of the substrate variables that need to be considered include the surface area, friability, porosity and chemical properties of the substrate.
The total surface area of the substrate cores largely determines the amount of coating material needed to produce coats with the required thickness. The total surface area is dependent on the size and size distribution, shape and surface roughness of the substrate cores (Porter, 1991; Ragnarsson and Johanson, 1988; Chen and Lee, 2002).
Smaller substrate cores, such as pellets, have higher surface area to volume ratios and hence would require more coating material to achieve similar coat thickness as compared to larger substrate cores, like tablets. Substrates, which are spherical in shape, have been found to be more evenly coated compared to needle-like substrates
(Chopra et al., 2002). The roughness of the substrate cores may affect the evenness of the film coat (Johansson and Alderborn, 1996). Great variation in thickness may result
17 INTRODUCTION in films with non-uniform mechanical strength and lower ability to withstand internal stress.
Coating processes are usually carried out in a highly stressed environment where there may be abrasion of the substrate. Consequently, some of the drug powder generated may be distributed in the coating and affect the release profiles of the final dosage form. The substrate porosity may affect drug release in two ways. Firstly, a highly porous substrate may permit water from the film coat to penetrate into the substrate, causing drug to be leached out. In addition, porous substrates tend to retain liquid by capillary forces and this may affect the coalescence of pseudolatex particles to form a coherent film (Tunon et al., 2003).
Besides the physical properties of substrates, the chemical nature of the drugs also plays a significant role in determining the release profiles of the final dosage form. These include drug solubility in the dissolution fluid and coating membrane, diffusivity in the film coat, molecular size and osmolarity, and the presence of osmoactive substances. Drugs of higher water soubilities show faster release rates
(Ragnarsson et al., 1992; Nesbitt et al., 1994). However, if chemical interaction occurs between the drug and coating, less soluble complexes may form and drug release will be reduced. Hydrophilic substrates have also been reported to aid in drug release by attracting water to the core, thereby facilitating dissolution of the drug
(Sousa et al., 2002). Drugs or excipients with high osmolarity can also result in a build up of osmotic pressure once the substrate core is wetted. The build up of osmotic pressure can increase drug release by aiding the transportation of drug through aqueous pores or channels (Schultz and Kleinebudde, 1997).
18 INTRODUCTION
2. Coating formulation
Film-coating formulations usually contain the following components, film forming polymers and additives such as plasticizers, colorants, surfactants, anti-tack agents and secondary polymers.
a. Nature of the polymer
The choice of polymer depends largely on its suitability for the intended functions as discussed previously. A good knowledge of the polymer chemistry would greatly help to understand the behaviour and interaction of the polymer with other substances. This includes the polymer chemistry, physical properties of the polymers such as the molecular weight, solubility, viscosity, permeability, mechanical and thermal properties (e.g. glass transition temperature and softening temperature). The molecular weight of the polymer has been reported to affect the mechanical properties of the film directly. For the film coat to perform its intended function, it must have sufficient mechanical strength so that it will not lose its intergrity when used.
Increasing the molecular weight of polymer will result in films of higher mechanical strength as well as elastic modulus. Rowe (1980) suggested a molecular value of 7 to
8 x 104 as the limiting value for tablet-coating polymers. The solubility of a polymer in the dissolution fluid will determine its suitability for various applications as discussed earlier. Hydroxypropyl methylcellulose and EC are usually available in a number of viscosity grades. Polymers of lower viscosity can be employed at higher concentrations, thus enabling reduction in solvent content of the polymer solution.
This also reduces the processing time as the amount of solvent needed to be removed during coating is less. However, very low viscosity polymers have low MW and tend to have poor film strength.
19 INTRODUCTION
Permeability of the polymer film to water vapour or other atmospheric gases, such as oxygen, is an important property of film coats that function as protective barriers to moisture or oxidation. The thermal properties of the polymer film are also important. The softening temperature, which is the temperature at which a film strip laid on a heated metal bar begins to soften corresponds to the degree of tackiness that may occur during high-temperature drying or during heat sealing process for strip or blister packing. Knowledge of the surface activity of the polymer solution is necessary for determining the degree of wetting of the substrate and spreading of the polymer spray droplet during film coating.
The more common polymers used for coating can be boardly classified as the cellulose derivatives, acrylic polymers and others. i. Cellulose derivatives
These include polymers, such as methylcellulose (MC), hydroxypropyl methylcellulose (HPMC), hydroxypropyl cellulose (HPC), ethylcellulose (EC), cellulose acetate phthalate (CAP), and hydroxypropyl methylcellulose phthalate
(HPMCP). These polymers possess the cellulose backbone made up of repeating units of anhydroglucose, with hydroxyl groups that can be substituted with other functional groups (Figure 5). The degree of substitution (DS) is expressed as the number or weight percentage of substituent groups. The term molar substitution (MS) takes into account the total number of moles of substituents. Both DS and MS affect the solubility and thermal gel point of the polymer. The polymer chain length, molecular size and extent of branching will determine the viscosity of the polymer in solution.
The cellulose derivatives can be further divided into water-soluble, water- insoluble and pH-dependent soluble polymers. The most commonly used water- soluble cellulose polymers include methylcellulose, hydroxypropyl cellulose and
20 INTRODUCTION hydroxypropyl methylcellulose. All these polymers are freely soluble in cold water but insoluble in hot water. Hydroxypropyl methylcellulose is popular because of its compatibility with organic solvents and water. Gelation temperature is the temperature at which the polymers separate from their aqueous solution. When this occurs, the viscosity of the solution will increase correspondingly. The approximate
Polymer Substituents groups (R) Methylcellulose -H –CH3 Hydroxypropyl methylcellulose -H –CH3 - CH2-CH-CH3
OH
Hydroxypropyl cellulose -H - CH2-CH-CH3
OH
Ethylcellulose -H –CH2CH3
Cellulose acetate phthalate -H O C –CH2CH3 O
O C
OH
Hydroxypropyl methylcellulose phthalate -H –CH3 - CH2-CH-CH3
OH
- CH2-CH-CH3
O
O C
O C
OH
Figure 5. Chemical structures of cellulose derivatives (R can be represented as –H, or otherwise specified).
21 INTRODUCTION
‘gelation’ temperature for methylcellulose, hydroxypropyl methylcellulose and hydroxypropyl cellulose are 50 °C, 60 °C and 45 °C respectively.
The most widely used water-insoluble cellulose derivative in film coating is ethylcellulose. It has a DS of between 2.27 and 2.62 corresponding to an ethoxyl content of between 44 and 51 %w/w. The ethoxyl groups are quite uniformly distributed on both the primary and secondary hydroxyl group of the anhydroglucose units. ethylcelullose is usually soluble in solvents that have nearly the same cohesive energy density or solubility parameter as itself. The amount of alcohol that is required to obtain minimum viscosity at a given concentration is proportional to the number of hydroxyl groups that remain unsubstituted.
The pH-dependent soluble cellulose polymers are usually the phthalyl (σ carboxybenzoyl) derivatives of cellulose or hydroxypropyl methylcellulose. They are insoluble at low pH but soluble at high pH. The pH at which they dissolve depends on the degree of acetyl and phthalyl substitution. The solubility of the polymers in organic solvents also depends on the degree and type of substitution.
ii. Acrylic polymer
Two main groups of acrylic polymers more commonly used in coating are methacrylate ester copolymers, and methacrylic acid copolymers (Figure 6)
(Lehmann, 1997). Methacrylate ester copolymers and methacrylic acid copolymers are structurally similar except that the former is totally esterified with no free carboxylic acid groups. Methacrylate ester copolymers are neutral in character and are insoluble over the entire physiological pH range. They can swell and become permeable to water and dissolved substances. Due to the presence of free carboxylic
22 INTRODUCTION acid groups, methacrylic acid copolymers are useful as enteric coating materials. They are capable of forming salts with alkalis and are relatively soluble at pH above 5.5.
H CH3 CH3
CH2 C CH2 C CH2 C O O O C C C OC2H5 OCH3 OR n1 n2 n3
+ - Methacrylate ester copolymers, where R = CH2-CH2N (CH3) 3Cl
CH3 R1
CH2 C CH2 C O O C C OH OR n1 2 n2
Methacrylic acid copolymers, where R1 = H or CH3 and R2 = C2H5 or CH3
Figure 6. Chemical structures of acrylic polymer
b. Additives i. Plasticizers
Plasticizers are low molecular weight materials which have the capacity to alter the physical properties of a polymer to render it more able to form a coherent film (Banker et al., 1966). The plasiticizer molecules interpose themselves between individual polymer strands, thus reducing polymer-polymer interactions and enabling the polymer strands to move past each other with greater ease. Plasticizers affect polymers which are either amorphous or have low crystallinity. Strongly crystalline polymers are difficult to plasticize as it is not easy to disrupt the intermolecular crystalline structure of the polymer.
23 INTRODUCTION
Plasticizers can be classified as polyols, organic esters or oils/glycerides
(Table 2). The polyols include glycerol, propylene glycol, polyethylene glycol (PEG) of molecular weights 200 - 6000. They are freely souble in water and hygroscopic in nature, except for higher molecular weight polyethylene glycol. The polyols are suitable for plasticizing water-soluble cellulose ethers containing a large proportion of hydroxyl groups. The organic esters include phthalate esters, dibutyl sebacate, citrate esters and triacetin (glycerol triacetate). This group of plasticizers is suitable for plasticizing the less polar cellulose esters, cellulose acetate phthalate and hydroxypropyl methylcellulose phthalate polymers. Most of the organic esters are miscible with organic solvents but have limited solubility in water. The commonly utilized vegetable oils or glycerides plasticizers include caster oil and acetylated monoglycerides. The vegetable oils themselves also contain a high proportion of glycerides. The distilled acetylated monoglycerides are modified fats of high purity and are available in various grades with degree of acetylation ranging from 50 to 98 percent. They are often used as plasticizers for ethylcellulose and polymers for the enteric coating. The choice of plasticizer depends on its compatibility with the polymer, plasticizer efficiency and permanence. Compatibility is demonstrated by the ability of the plasticizer to act as a solvent for the polymer to be plasticized.
Thermodynamically, in order for the polymer to dissolve in a solvent (plasticizer), the
Gibbs free energy of mixing, ∆G, must be negative,
∆G = ∆H - T∆S (8)
where ∆H is the heat of mixing, T is the absolute temperature and ∆S is the entropy of mixing. Okhamafe and York (1987) showed that ∆H could be calculated using the following equation (Hildebrand and Scott, 1950):
24 INTRODUCTION
2 ⎡ 1/2 1/2 ⎤ ⎢⎛ 1 ⎞ ⎛ 2 ⎞ ⎥ m ⎜ ∆E ⎟ − ⎜ ∆E ⎟ 1 2 ∆H = V ⎢⎜ ⎟ ⎜ ⎟ ⎥ φ φ (9) ⎣⎢⎝ V1 ⎠ ⎝ V2 ⎠ ⎦⎥
where Vm is the total volume of the mixture, ∆E1 and ∆E2 are the energies of components 1 and 2 respectively, V1 and V2 are the molar volumes of component 1 and 2 respectively and φ1 and φ2 are the volume fractions of components 1 and 2 respectively. The term ∆E/V is generally referred to as the cohesive energy density
Table 2. Physical properties of plasticizers used in film coating. (Adapted from Wick et al., 1999; Hoechst, 1997; Union Carbide, 1981; Chas. Ffizer, 1961; Scheflan and Jacobs, 1953; Eastman Chemical, 1981)
Plasticizer Molecular Specific Refractive Boiling Flash weight gravity index at point at pointa at 20°C 25°C 760 mm of (°C) mercury (°C) Polyols Glycerol 92 1.260 1.473 290 177 Propylenel glycol 76 1.035 1.431 188 99 Polyethylene glycol 400 400 1.128 1.452 > 300 245 Polyethylene glycol 4000 4000 1.20 1.455b Solid at 262 (solid) room temperature Organic esters Dimethyl phthalate 194 1.189 1.517 282 163 Diethyl phthalate 222 1.123 1.501 296 168 Dibutyl phthalate 278 1.051 1.490 340 171 Dibutyl sebecate 314 0.932 1.443 345 185 Triethyl citrate 276 1.136 1.440 127 155 Tributyl citrate 360 1.045 1.446 170 185 Triethyl acetyl citrate 318 1.135 1.438 132 188 Triethyl butyl citrate 402 1.048 1.441 173 204 Triacetin 218 1.156 1.431 258 132 Vegetable oils and glycerides Castor oil - 0.960 1.480 - - 50% acetylated - 0.94c - > 150 239 monoglyceride Fully acetylated - 0.98 - > 500 242 monoglyceride a Measured using Cleveland open cup; b Measured at 70 °C; a Measured at 50 °C.
25 INTRODUCTION
(CED) and its square root as the solubility parameter, δ.
Hence, equation 9 can also be expressed as,
2 ∆H = Vm()δ1 − δ 2 φ1φ 2 (10)
If δ1 and δ2 are the same, the heat of mixing will be zero, indicating a state of maximum compatibility between polymer and plasticizer. Hence, solubility parameter measurements can be used as a measure of platicizer compatibility. It has been shown from the solubility parameters of EC in several plasticizers that EC is most compatible with the phthalic acid esters and dibutyl sebacate, but incompatible with glycerol or propylene glycol (Burrell, 1975). However, the above approach for predicting compatibility between polymers and plasticizers is not widely adopted and hence many solubility parameters of polymers and plasticizers are unknown. The limitations of solubility parameter measurement can be overcome by using intrinsic viscosity measurement. The intrinsic viscosity of the polymer in the plasticizer will be high when there is a good interaction between the plasticizer and polymer.
Entwistle and Rowe (1979) had reported a correlation between the intrinsic viscosities of the polymer/plasticizer mixtures and the mechanical attributes of the resultant films formed. The mechanical strength of the film was found to be the lowest when the intrinsic viscosity of the polymer/plasticizer mixture was the highest.
This is attributed to the interaction between the plasticizer and polymer, giving rise to increased segmental movements. Thus, the resultant plasticized polymer film has a more porous, more flexible but less cohesive structure, as indicated by increased strain or film elongation and decreased elastic modulus and tensile strength. Changes in the mechanical properties of plasticized films have been widely used as a convenient means to evualate the efficiency of plasticizers (Aulton et al., 1981;
26 INTRODUCTION
Bodmeier and Paeratakul, 1994; Sinko and Amidon, 1989). Sinko and Amidon (1989) had further proposed a term to describe the efficiency of plasticizers based on their ability to induce changes on the rate of mechanical response as the solvent leaves the film and the polymer passes through a rubbery to glassy transition. In addition to the above mechanical methods, a thermal method that involves the measurement of glass transition temperature has also been used to evaluate the compatibility and effectiveness of plasticizers. Glass transition temperature is the temperature at which a polymer changes from a glassy rigid material to a viscous rubbery material. The glass- transition temperature of a plasticized polymer can be determined by studying the temperature dependence of properties such as specific heat, refractive index, elastic moduls, film hardness or mechanical damping. The most effective plasticizer is one that gives the greatest decrease in glass transition temperature per unit amount incorporated.
Permanence of the plasticizer is an important factor to consider when more volatile plasticizers, such as DEP and PG are used. Loss of such plasticizers may occur after coating, curing (Bindschaedler et al., 1987; Bodmeier and Paeratakul,
1994) and storage under stress conditions (Gutiérrez-Rocca and Mcginity, 1994;
Hutchings et al., 1994). The volatility of the plasticizer depends on the effective vapour pressure of the plasticizer and its rate of diffusion through the polymer matrix.
The effective vapour pressure is dependent on its relative proportion, interaction and compatibility with respect to the polymer and the molecular weight of the polymer.
Migration of plasticizer into the cores or into packaging materials can occur, resulting in brittle film coats and hence faster drug release (Hogan, 1995). Leaching of plasticizers from polymer film coat may also occur depending on the chemical and physical properties of the plasticizers (Lippold et al., 1990; Bodmeier and Paeratakul,
27 INTRODUCTION
1992, 1993, 1994).
It was mentioned earlier that one of the important endpoint requirements of a film coat is its ability to retard the entry of water vapour or other gases into the dosage form. The transport of a permeant across a barrier is defined by Crank’s equation
(Okhamafe and York, 1983):
P = D S (11)
where P, D and S are the permeability, diffusion and solubility coefficients of the film coat respectively. Transport of a permeant across a film involves dissolution of the permeant in the film material and diffusion of the permeant across of the film.
Diffusion of permeant across a film may occur through the polymer matrix itself and/or through the voids present. Since addition of plasticizers often results in alteration to the polymer structure, it may also change the permeability of the film coats. For instance, it was found that hydroxypropyl methylcellulose films plasticized with polyethylene glycol 400 and 1000 had higher diffusion coefficients for water than the non - plasticized film (Okhamafe and York, 1983).
Special considerations have to be taken when mixing water – insoluble plasticizers with aqueous polymeric dispersions. These plasticizers have to be dispersed efficiently in the aqueous medium and sufficient time should be allowed for plasticizer uptake by the colloidal polymer particles (Bodmeier et al., 1997).
Unabsorbed plasticizer will coalesce resulting in an inhomogeneous distribution of the plasticizer in the film formed.
28 INTRODUCTION ii. Colorants and opacifiers
Colorants and opacifiers can be boardly classfied as synthetic or natural.
Synthetic organic colorants include dyes which are usually water soluble and have restricted use. However, they can be complexed with hydrated alumina to form more useful water - insoluble complexes, known as pigments. Synthetic inorganic colorants, such as titanium dioxide have good opacifying capacity and provide good protection against light. The natural colorants are chemically and physically most diverse. They are not widely used because of their relatively poor stability to light and high price.
Pigments are generally considered superior to water - soluble dyes for several reasons. There has been concern with migration of water - soluble colorant over tablet surfaces during the drying process. This undesirable phenomenon is less likely to take place for water – insoluble pigments. The latter is also more opaque than dyes, thus offering better protection against light. In addition, low amount of pigments had been shown to decrease the permeability of films to water vapour and oxygen (Hogan,
1995). Pigments can contribute to the total solids of a coating suspension without increasing the viscosity significantly. However, increasing concentration of pigments would require more polymer particles to fill interparticle voids. The critical pigment concentration in a film is determined by the concentration beyond which marked changes in the film mechanical properties, appearance and permeability occur
(Gibson et al., 1988). The presence of pigments has been reported to reduce the tensile strength, decrease the percent elongation and increase the elastic modulus of a film (Aulton et al., 1984). Okhamafe and York (1985) found that polymer - pigment interactions also affected the mechanical properties of films. High pigment - polymer interaction decreased film elongation. Moreover, it was found that pigments used in
29 INTRODUCTION film coating of tablets can exert opposing effects on adhesion. In a study, the adhesion of hyrdoxypropyl methylcellulose film to tablet surface was increased when a small amount of talc was incorporated. However, when the talc concentration was increased to more than 10 percent, the internal stress of the film became more dominant and film – tablet adhesion strength decreased.
iii. Surfactants
Presence of surfactants in film coat has been known to cause variation in drug release (Lippold et al., 1989; Paeratakul and Bodmeier, 1989; Hutchings and Sakr,
1994; Gunder, et al., 1995; Dressman, et al., 1995; Goodhart et al., 1984). Cationic drugs, such as chlorpheniramine maleate, propranolol HCl, diltiazem HCl and quinidine sulphate, have also been reported to interact with the anionic surfactant, sodium lauryl sulphate and form water-insoluble ion-pair complexes (Wheatley and
Steuernagel, 1997).
iv. Anti-tack agents
Magnesium stearate, talc or kaolin as antitack or separating agents help to reduce agglomeration or sticking of particles during the coating process. Some soluble salts (Yuasa et al., 1997; Nakano and Yausa, 2000), plasticizers and film formers
(Heng et al., 1996; Wong et al., 2002) were also found to have antitack properties.
The lag time was also found to increase when talc was employed in the coating formula. Schultz (1997) suggested that increased diffusional resistance and hence reduced water permability of the film coat was responsible for the longer lag time. On the other hand, anti-tack agents such as magnesium stearate, stearic acid and silicon
30 INTRODUCTION dioxide were reported to increase the rate of drug release when their concentrations were higher than the critical pigment volume concentration (Wan and Lai, 1992).
v. Hydrophilic additives
Hydrophilic additives are often added to modify the barrier properties of film coats and membranes. These additives can be boarded divided into water – soluble, low molecular weight and water – soluble, high molecular weight. Low molecular weight additives are usually dispersed as discrete particles in the polymer matrix and dissolve readily in the dissolution media, resulting in formation of pores in the polymer matrix. Examples of such ‘pore-forming’ additives include micronised sucrose, lactose, sorbitol, sodium chloride and calcium phosphate (Lindholm et al.,
1987; Hennig and Kala, 1987; Kallstrand and Ekman, 1983; Bodmeier et al., 1997)
Water – soluble polymers such as hydroxypropyl methylcellulose and hydroxypropyl cellulose had also been added to EC to modify drug release properties (McAinsh and
Rowe, 1979; Rowe, 1986). These hydrophilic polymeric additives increase the permeability of hydrophobic films by several mechanisms. For example, polyethylene glycol dissolves or erodes in the dissolution medium to create pores in EC films
(Donbrow et al., 1975). In contrast, hydroxypropyl cellulose which dissolves partially forms a matrix with EC (Donbrow et al., 1980). Some additives modify drug release through EC films by acting as carriers for drugs, such as Span 20 and tetrabutylammonium bromide for salicylic acid (Lindholm et al., 1982).
3. Coating process variables
Coating process itself is an important factor that can affect the quality of film coat and uniformity of distribution of coat across all surfaces of the substrate
31 INTRODUCTION
(Eastman Chemical Products, 1981). For example, porosity of the coat may be affected by excessive ‘spray drying’ of the coat or incomplete coalescence of coat deposited. Vigorous drying conditions during the coating process may result in stress cracks on the coatings. Excessive spray rates with organosols or excessively high temperature with aqueous polymer dispersions can cause tackiness, thus giving rise to
‘picking’ of the coat.
A wide variety of coating equipment with different coating efficiencies is available (Yang et al., 1992; Wesdyk et al., 1993; Bertelsen et al., 1994). The first coating equipment consistent of conventional coating pans that have poor mixing and drying efficiencies. The design of the coating pans was subsequently improved with the addition of side or perforated vent, immersion swords and better exhaust systems
(Cole, 1995). The fluidized bed coater is recommended for smaller sized substrates which cannot be easily coated in coating pans. In the fluidized bed coating process, the coating materials are applied to a fluidized substrate bed using nozzles located either at the top, bottom or side of the coating chamber. Mehta (1997), Jones and
Percel (1994) studied the quality of film coatings produced by fluidized bed coater and coating pan. They found that the quality of coatings decreased in the following order: wruster = tangential spray > side-vented pan >> conventional pan
In the coating process, there are various processing parameters that can affect the quality of the coat. These parameters depend on the type of coating equipment employed and they include:
• product batch size,
• pellet size and size distribution,
• pellet density (and variability within the batch),
• coating liquid: organosol or aqueous polymer dispersion,
32 INTRODUCTION
• rheological characteristics of the coating liquid,
• solid content of coating liquid,
• quantity of coating liquid to be applied,
• spray application rate,
• drying air temperature, volume, and humidity,
• product temperature,
• atomizing air pressure/volume,
• partition height,
• pan speed,
• baffle design,
• spray nozzle design (fluid orifice size, fluid orifice design, air cap design),
• disk speed (tangential spray process), and/or
• split opening (tangential spray process).
The influence of various processing parameters in pellet coating listed in
Table 3. In practice, each processing parameter tends to be affected by the other. For instance, when determining the drying conditions (drying air volume, temperature and humidity), one has to consider them in the context of the spray application rate and solid content of the coating liquid. In addition, the impact of drying conditions on properties of the resultant film coat defers between organosols and aqueous polymer dispersion. Where organosols are used, inappropriate drying conditions may result in creation of unacceptable porous coatings, increased tackiness and ‘picking’ of film coat. In contrast, the inappropriate drying conditions may affect the coating applied from aqueous polymeric dispersions by incomplete coalescence of the coat, tackiness
33 INTRODUCTION due to exposure of the coating to temperature in excess of its glass transition temperature and leaching of drug from pellets (Yang and Ghebre-Sellassie, 1990).
After completion of the coating procees, additional time may be required for further drying and coalescence of the polymer coat (Gilligan and Wan Po, 1991). In order for complete coalescence of the film, the coated product should be stored at temperature above the glass transition temperature of the polymer film (Lippold et al.,
1989; Bodmeier and Paeratakul, 1994; Hutchings et al., 1994; Keshikawa and
Nakagami, 1994). It was found that curing of EC coated pellets at 60 °C for at least 1 h was necessary to prevent aging effects in drug release (Gilligan and Wan Po, 1991).
Curing beyond a particular duration did not contribute to further coalescence. Increase in drug release rate for curing periods in excess of 4 h was reported. This was
Table 3. Influence of various processing parameters in pellet coating (Adapted from Porter and Ghebre-Sellassie, 1994)
Effects Contributing processing parameters Film formation (structural Drying conditions (processing temperature, quality of coat) air volume and humidity) Spray application rate Atomizing air pressure/volume Tackiness of coat Drying conditions Spray application rate Solid content of coating liquid
Uniform distribution of coat Mixing of pellets in process Number of spray nozzles used Design of spray nozzles Atomizing air pressure/volume Spray application rate Solid content of coating liquid Quantity of coating applied
Leaching of drug from core Nature of coating liquid (organosol or aqueous polymer dispersion) Drying conditions Spray application rate Atomizing air pressure/volume Solid content of coating liquid
34 INTRODUCTION attributed to migration of drug from the interior to the surface of the pellets during curing. Curing at high temperature could cause excessive tackiness and agglomeration of the solid pellets.
F. Analysis and comparison of dissolution data
Drug release profiles of coated dosage forms are usually obtained by dissolution studies. The dissolution data are analyzed by various methods that provide different perspectives of the release profiles. These methods can be broadly classified as:
• explanatory data analysis (O’Hara et al., 1998),
• analysis of variance (ANOVA)-based method (Mauger et al., 1986; Polli et
al., 1997),
• model-dependent method (Sathe et al., 1996; Polli et al., 1997; Shah et al.,
1997), and
• model-independent method (Podczeck, 1993; Moore and Flanner, 1996; Polli
et al., 1997; Shah et al., 1997, 1998).
In the explanatory data analysis method, the dissolution profile is illustrated graphically by plotting each dissolution time point with error bars that usually extend to two standard errors. The dissolution profiles of two formulations may be considered significantly different from each other if the error bars of corresponding dissolution time points do not overlap. This method is usually performed as a first step to compare dissolution profiles. However, explanatory data analysis is not comprehensive and may be difficult to perform when there are too many formulations to be compared (O’Hara et al., 1998).
35 INTRODUCTION
The ANOVA - based method does not involve curve fitting. The dissolution data are used in their native form or as a simple transform. Information on the differences between profiles in level and shape can be obtained. However, while this method takes the variability in dissolution profiles into account in the comparison at each point, it ignores the correlation between the dissolution profiles, treating each point as if it were independent of the others (O’Hara et al., 1998). Hence, this method is deemed overly discriminating and not very useful (Polli et al., 1997).
The model - dependent and model - independent methods are the two most widely used analytical methods to compare dissolution profiles. Quantitative difference is then compared using both method dependent and independent methods in order to obtain a more accurate and complete understanding of the dissolution data.
In the model dependent method, the dissolution data are fitted into relevant selected mathematical models, characterized by suitable mathematical functions. The dissolution profiles are then evaluated in terms of the derived model parameters.
Mathematical models used in dissolution data analysis include zero-order, first-order exponential function (Gibaldi and Feldman, 1967), cubic root law (Hixson and
Crowell, 1931), square root of time (Higuchi, 1963), Baker and Lonsdale (Baker and
Lonsdale, 1974) and Hopfenberg equation (Hopfenberg, 1976; Katzhendler et al.,
1997).
1. Zero order equation
Drug release that occurs slowly from dosage forms that do not disintegrate
(assuming that area does not change) can be represented by:
W0 − Wt = K t (12)
36 INTRODUCTION
where W0 is the initial amount of drug in the dosage form, Wt is the remaining amount of drug in the dosage form at time t and K is a proportionality constant. The equation can also be expressed as:
Qt = Q0 + K0 t (13)
where Qt is the amount of drug dissolved in time t, Q0 is the initial amount of drug in the dissolution medium and K0 is the zero order release constant. Q0 is usually zero.
2. First order equation
This model was first proposed by Gibaldi and Feldman (1967) and can be expressed as:
-k Qt = Q0 e 1 t (14)
where K1 is the first order release rate constant. If drug release follows first order equation, the release is proportional to the amount of drug remaining in the dosage form.
3. Hixson – Crowell equation
Hixson and Crowell (1931) derived an equation that could be applied to pharmaceutical dosage forms, whose dimensions diminish proportionally as dissolution occurs:
1/3 1/3 W0 − Wt = Ks t (15)
37 INTRODUCTION
where W0 is the initial amount of drug in the dosage form, Wt is the remaining amount of drug in the dosage form at time t and Ks is a constant incorporating the surface – volume relationship. This equation can be rewritten as,
1/3 1/3 1/3 W0 − Wt = K′ N D Ce t / δd (16)
where K′ is a constant related to the surface, shape and density of the particle, N is the number of particles, D is the diffusion coefficient, Ce is the solubility in dissolution medium and δd is the thickness of the diffusion layer. The shape factor for cubic or spherical particles is constant if the particles dissolve in an equal manner from all sides. Equation 16 can be simplified as:
1/3 (1 − ft) = 1 − Kβ t (17)
where ft = 1 − (Wt/W0), respresenting the fraction of dissolved drug at time t and Kβ is a release constant. Hence, a plot of the cubic root of the unreleased fraction of drug versus time will be linear if the equilibrium conditions are not reached and if the geometrical shape of the dosage form diminishes proportionally over time. When this model is used, it is assumed that the release rate is limited by dissolution and not diffusion of drug that might occur through the polymeric matrix.
4. Higuchi equation
This model, which is also known as the Higuchi square root time equation, can be expressed as:
Q = D(2W − Cs)Cst (18)
38 INTRODUCTION where Q is the amount of drug released per unit area in time t, D is the diffusivity of the drug molecules (diffusion constant) in the matrix substance, W is the total amount of the drug per unit volume of the matrix, and Cs is the solubility of the drug in the matrix. When W >> Cs the above equation can be simplified to,
Q = 2WDCst (19)
This equation indicates that the amount of drug released is proportional to the square root of time, which is typical of a matrix-type system. Diffusion of drug can occur through insoluble matrices or film coats (Rao and Murthy, 2002). In the case of coated pellets, drug release involves penetration of dissolution fluid, dissolution of drug in the dissolution fluid and leaching of drug through interstitial channels or pores. The overall release rate is usually by the rate at which dissolved drug leaches out when the drug concentration in the matrix is lower than its solubility. The following equation can be used:
Dε 1/ 2 Q = 1/2 = kt (20) τ (2A - εC s )C s t)
where Q is the amount of drug released per unit surface area in time t, D denotes the drug diffusion coefficient in the matrix phase, Cs is the drug solubility in the matrix/dissolution medium, A represents the drug concentration in the matrix, ε is the porosity, τ denotes the tortuosity of the matrix and k is the dissolution rate constant of the drug. If the matrix is not coated and does not undergo significant alteration in the presence of water, the Higuchi model can be further simplified as:
1/2 Q = KH t (21)
39 INTRODUCTION
where KH is the Higuchi dissolution constant.
5. Baker and Lonsdale equation (Higuchi's model for spherical matrices)
The Baker and Lonsdale equation (Baker and Lonsdale, 1974), which was derived from Higuchi's model (Higuchi, 1963), describes drug release from sustained release microspheres (Lu et al., 1996):
3 2/3 []1− ()1− ft − ft = kt (22) 2
where ft is the fraction of drug released at time t. For controlled release of drug from a
2 spherical matrix, the constant k is equal to 3D Cs /ro C0, where, D is the diffusion coefficient, Cs is the drug solubility in the matrix, ro is the radius of the device and Co is the initial concentration of the drug in the polymer matrix.
If the matrix is not homogeneous and presents fractures or capillaries that may
2 contribute to the drug release, k is equal to 3DfCfsε/ro C0τ where Df is the diffusion coefficient, Cfs is the drug solubility in the liquid surrounding the matrix, τ is the tortuosity of the matrix and ε is the porosity of the matrix.
6. Hopfenberg equation
Hopfenberg proposed a general mathematical equation to describe drug release from slabs, spheres and infinite cylinders that display heterogeneous erosion
(Hopfenberg, 1976; Katzhendler et al., 1997):
n Q = 1- [1 - K t] 1 (23)
40 INTRODUCTION
where K = Ke/C0 a0, Ke is the erosion rate constant, C0 is the initial concentration of drug in the matrix and a0 is the initial radius for a sphere or cylinder or the half-
thickness for a slab. The value of n1 is 1, 2 and 3 for a slab, cylinder and sphere, respectively. This model assumes that the rate-limiting step of drug release is the erosion of the matrix itself and that time dependent diffusional resistance internal or external to the eroding matrix has no influence.
A number of mathematical models may be employed for characterizing drug release profiles and elucidating drug release mechanisms. The suitability of the model is indicated by the correlation coefficient (r2) and Akaike information criterion (AIC) values. The r2 value is a convenient and common metric used to measure model fit
(Costa and Lobo, 2001). The AIC value is a measure of goodness of fit based on maximum likelihood. When comparing several models for a given set of data, the model associated with the smallest value of AIC is regarded as giving the best fit. AIC is only appropriate for comparing models using the same weighting scheme.
AIC = n ln (WSSR) + 2p (24)
where n is the number of dissolution data points, p is the number of parameters in the model and WSSR is the weighed sum of square of residues. The latter is calculated as follows:
n 2 WSSR = ∑ [wi(yi-ŷi) ] (25) i=1
where wi is an optional weighing factor and ŷi denotes the predicted value of yi.
The model - independent method compares dissolution data directly, and does not rely on the choice of mathematical model which may be unrealistic at times. In
41 INTRODUCTION this method, graphical representation of the dissolution profiles is performed as a preliminary step to illustrate non-quantitative differences and their evolution along the profile. The dissolution data are subjected to further analysis, using time-point or pairwise approaches to determine similarity or dissimilarity between dissolution curves (Pillay and Fassihi, 1998). Dissolution studies are usually carried out in triplicate. The tx% parameter corresponds to the time taken for the release of a specific percentage of drug. In the time point approach, the t50% values as well as their mean
(MDT) were calculated for each formulation in each of the triplicate dissolution measurements. The MDT value can be calculated as follows:
n ˆ ∑ tj ∆M j j=1 MDT = n (26) ∑ ∆M j j=1
where j is the sample number, n is the number of dissolution sample times, ˆtj is the time at midpoint between tj and tj-1 and ∆Mj is the additional amount of drug dissolved between tj and tj-1.
In the pairwise approach, fit factors are employed. The "difference factor, f1" and "similarity factor, f2" are determined using equations 27 and 28 (Costa and Lobo,
2001). The difference factor (f1) calculates the percent difference between two dissolution curves at each time point and is a measure of the relative error between the two curves (Costa et al., 2003),
n ∑ Rj - Tj j =1 1 f = n ×100 (27) ∑ Rj j =1
42 INTRODUCTION
where n is the sampling number, Rj and Tj are the percent drug dissolved from reference and test products at each time point j. The percent error is zero when the dissolution profiles of the reference and test products are identical and increases proportionally with the dissimilarity between the two dissolution profiles. For curves to be considered similar, f1 values should be close to 0. Generally, f1 values up to 15 indicate an average difference of no more than 10 % and the two curves are deemed to be equivalent.
The similarity factor (f2) is a logarithmic transformation of the sum-squared error of differences between the test and reference products over all time points,
−0.5 ⎧ n ⎫ ⎪⎡ 2 ⎤ ⎪ f 2 = 50× log⎨⎢1+ (1/ n)∑ wj Rj - Tj ⎥ ×100⎬ (28) ⎩⎪⎣ j=1 ⎦ ⎭⎪
where wj is an optional weight factor.
When the two dissolution profiles are identical, f2 = 50 x log (100) = 100, and when the dissolution of one batch is completed before the other begins,
−0.5 ⎧ n ⎫ ⎪⎡ 2 ⎤ ⎪ f 2 = 50× log⎨⎢1+ (1/ n)∑(100) ⎥ ×100⎬ = -0.001 which can be rounded to 0. ⎩⎪⎣ j=1 ⎦ ⎭⎪
Hence the values of f2 range between 0 to 100 with a higher f2 value indicating greater similarity between the two profiles. In practice, a test batch is accepted as ‘similar’ to a reference batch if the dissolution profile difference between the two batches is no more than the dissolution profile difference between two reference batches. By limiting the average difference to no more than 10 percent at any sample point, f2 becomes:
−0.5 ⎧ n ⎫ ⎪⎡ 2 ⎤ ⎪ f 2 = 50× log⎨⎢1+ ∑(10) ⎥ ×100⎬ (29) ⎩⎪⎣ j=1 ⎦ ⎭⎪
43 INTRODUCTION
-0.5 f2 = 50 x log {[101] x 100} = 49.89 ~ 50 (30)
A test batch dissolution is considered similar to that of the reference batch if the f2 value of the two profiles is not less than 50.
Fit factors are adopted by the FDA Center for Drug Evaluation and Research
(CDER) and the European Medicines Evaluation Agency (EMEA) Committee for
Proprietary Medicinal Products (CPMP) in the assessment of similarity between two in-vitro dissolution profiles (Yuksel et al., 2000). However, it is noted that fit factors are incapable of indicating the sense of the deviation (Moore and Flanner, 1996).
They also do not directly take into consideration the shape of the curve nor allow for a variation in the spacing between sampling times (Costa, 2001; Costa and Lobo, 2001).
Moreover, they are sensitive to measurements obtained after either test or reference has released more than 85 percent of its drug content (Shah et al., 1998). The selection and determination of the dissolution end points therefore play a critical role in the calculation of fit factors and the subsequent decision as to whether the test and reference profiles resemble each other or not (Yuksel et al., 2000 and Pillay and
Fassihi, 1998). In spite of these limitations, fit factors are very useful for comparing dissolution curves by providing a single number to describe two curves that consist of several points (Moore and Flanner, 1996).
44 OBJECTIVES
II. OBJECTIVES
Ethylcellulose (Porter, 1989; Iyer et al., 1990) is one of the most commonly used polymers in the production of coated controlled-release dosage forms. Its popularity is largely owed to its good film forming ability and low permeability to moisture (Rekhi and Jambhekar, 1995). As a result of its being practically insoluble in water, coating preparations of ethylcellulose in the past were only possible when dissolved in organic solvents (Porter, 1989). However, over the past decade, water- based coating has gradually gained popularity and become the preferred coating system for new product development (Savage et al., 1995). Though both organic solutions and aqueous polymeric dispersions produce chemically similar polymeric film, the physical properties of the film produced by each system can be very different. This is essentially due to differences in film formation mechanism caused by differences in the molecular states of ethylcellulose in different types of solvents
(Savage et al., 1995; Wheatley et al., 1997). The existence of ethylcellulose as dispersed particles in water, in contrast to solvated molecules by organic solvent, can affect ethylcellulose film formation as well as interaction with other additives present in the coating formula. While the use of organic ethylcellulose coating system is better described in literature, there is still limited information on the newer aqueous ethylcellulose coating systems. In particular, knowledge pertaining to the type of interaction between ethylcellulose dispersed particles in water as well as in the presence of additives is still being sought (Porter and Ghebre-Sellassie 1994). Hence, it is the aim of this study to explore further the effects of interaction between ethylcellulose and additives that are commonly employed in aqueous coating systems, on the film coat properties. Two additives, plasticizers and hydrophilic polymeric additives, were specifically chosen for this study, as they are most likely to interfere
45 OBJECTIVES the film formation mechanism and consequently affect the film quality, product performance and stability.
Part 1.
A number of researchers have reported that polymeric film coats applied on dosage forms undergo changes on storage. The extent of these changes is influenced by several factors such as the amount of plasticizers added, type of plasticizers used, film forming conditions and storage temperature and humidity. It was found that films formed from aqueous ethylcellulose dispersions showed significant changes in their mechanical properties when they were dried at a higher temperature (Hutchings et al.,
1994). Exposure to a high relative humidity may cause a polymeric film coat to absorb moisture from the atmosphere. Findings of several researchers suggested that changes in mechanical properties of some polymeric films stored at high humidity are correlated with the plasticizing effect of the moisture present in the films (Aulton et al., 1981; Wu and McGinity, 2000). Changes in film’s mechanical properties may ultimately influence drug release, stability and other physicochemical properties of the coated dosage forms (Chowhan, 1982).
It has been reported that plasticizers such as dibutyl sebacate, tributyl citrate, acetyl tributyl citrate and oleyl alcohol produced ethylcellulose films that showed greater elongation upon stretching, after the films had been stored under conditions of elevated humidity (Hutchings et al., 1994). However, the actual mechanisms that caused the above change and the extent of influence by the different types of plasticizers have not been reported.
The objective of this study was to investigate the effects of different types of plasticizers on the properties and stability of ethylcellulose films exposed to different
46 OBJECTIVES storage conditions. Attempt were made to elucidate the mechanisms responsible for the changes observed and to correlate the micro and macro changes in the film structure with the release profiles of coated pellets.
Part 2
In the recent years, great interests in the modification of drug release from reservoir and matrix delivery systems is seen from the increased amount of research
(Verma and Garg 2004; Sangalli et al., 2001; Kim et al., 2000; Frohoff-Hulsmann et al., 1999; Bussemer et al., 2003). The goals of such innovations include developing systems capable of enhancing the release for poorly water-soluble drugs, providing precise control of drug release, as well as broad ranges of drug release patterns
(Lecomte et al., 2003, 2004). Due to its lack of permeability to water, ethylcellulose is one such polymer which has great practical impetus for seeking solution to overcome the undesirable slow release obtained for poorly soluble drugs (Frohoff-Hulsmann et al., 1999).
Methods employed to improve drug permeability through ethylcellulose coatings included reduction in coat thickness (Osterwald et al., 1985), as well as formation of pores in ethylcellulose films using organic solvents (Narisawa et al.,
1993; Narisawa et al., 1994) and hydrophilic additives (Frohoff-Hulsmann et al.,
1999a; Frohoff-Hulsmann et al., 1999b). A number of studies were directed at employing additives to modify the permeability of ethylcellulose films: hydroxypropyl methylcellulose (Gilligan et al., 1991 and Hjärtstam et al., 1998), hydroxypropyl cellulose (Umprayn, et al., 1999), carboxy methylcellulose (Yamada et al., 2001), acrylates (Sánchez-Lafuente et al., 2002, Fan et al., 2001, Phuapradit et. al., 1995), pectin (Macleod et al., 1997), maltodextrins (Rohera et al., 2002), xylitol
47 OBJECTIVES
(Rohera et al., 2002) and poly ethylene glycol (Sakellariou et al., 1986 and Rohera et al., 2002). Hydroxypropyl methylcellulose is the most popular secondary polymer used to for pore formation of ethylcellulose films. However, it was found to have limited degree of miscibility in an organic mixture of ethylcellulose and hydroxypropyl methylcellulose (Sakellariou et al., 1986), with occurrence of phase separation, especially at high concentration (Sakellariou and Rowe, 1995).
Complicated release phenomenon has also been reported for theophylline pellets coated with ethylcellulose containing hydroxypropyl methylcellulose (Gunder et al.,
1995). It was believed that the two-phase release profiles portrayed was partly attributed to the closure of pores which were initially created by dissolution of hydroxypropyl methylcellulose in the films. Consequently, the subsequent release was controlled by the distribution and diffusion of dissociated drug through the coating membrane.
Polyvinylpyrrolidone (PVP) is a water-soluble, physiologically inert synthetic polymer consisting essentially of linear 1-vinyl-2-pyrrolidinone groups, with varying degree of polymerization which results in polymers of different molecular weights
(Wade and Weller, 1994; Blecher et al, 1990; BASF, 1986; Aldeyeye and Barabas,
1993; Kumar et al., 1999). Polyvinylpyrrolidone has been reported to increase the solubility of active substances by forming water-soluble complexes. Viviprint 540 is a molecular – composite polyvinylpyrrolidone (MCPVP) which is formed in situ by incorporation of insoluble crosslinked poly(Polyvinylpyrrolidone) nano particles into soluble, film–forming polyvinylpyrrolidone (Hood et al., 2003). It is therefore of much higher molecular weight than polyvinylpyrrolidone and less soluble in water. It can be used as a binder, as well as a top coat in multi-layer coating systems. The swelling property of cross-linked polyvinylpyrrolidone was employed to enhance
48 OBJECTIVES drug release (Fan et al., 2001). Plasdone S-630 copolyvidonum (PV/VA) is a synthetic water-soluble copolymer consisting of N-vinyl-2-pyrrolidone and vinyl acetate in a random 60:40 ratio. It can be used as a binder in dry and wet granulation methods. Zingone et al. (1994) reported that the solubility and dissolution rate of carbamazepine were enhanced by mixing with PV/VA copolymer, possibly due to decrease in crystallinity and increase in wettability of the drug.
All the water-soluble polymers discussed above are potential polymeric film modifiers for enhancing drug release. Their good solubility in water also has the added advantage of convenient mixing with ethylcellulose aqueous dispersion. To date, studies on composite ethylcellulose films containing polyvinylpyrrolidone and its derivatives for controlled release have not been reported. Hence, one of the aims of this study was to explore the potential of polyvinylpyrrolidone and its derivatives as ethylcellulose film modifier, through systematic investigations of their compatibility and impact on the functionality of the composite ethylcellulose films and coated products.
49 EXPERIMENTAL
III. EXPERIMENTAL
A. Materials
1. Drugs, polymers and additives
The model drugs used were propranolol USP, sulphanilamide BP, sulphaguanidine BP (Luen Wah Med Co., Singapore), metformin hydrochloride BP
(Auro Lab, India), theophylline USP (Fluka Buchs, Switzerland), chlorpheniramine maleate BP (Merck, Singapore), verapamil hydrochloride USP (Nicholas Piramal,
India), and paracetamol B.P. (China). Their properties are shown in Table 4.
Lactose monohydrate (Pharmatose, 200 Mesh, De Melkindustrie Veghel, The
Netherlands) and microcrystalline cellulose (MCC; Avicel PH-102, Asahi Chemical,
Japan) were used to prepare core spheroids.
The main film formers and coating polymers were methacrylic acid-ethyl acrylate copolymer (MA; Eudragit® L 30D, Röhm Pharma, Darmstadt, Germany) and ethyl cellulose (EC; Aquacoat®, type ECD-30, FMC Corp., Newark, US and
Surelease®, Colorcon, West Point, US).
The additives consist of plasticizers: dibutyl phthalate (DBP), triethyl citrate
(TEC), glycerin triacetate (GTA) from Merck-Schuchardt (Darmstadt, Germany), diethyl phthalate (DEP) from The British Drug House (Poole, UK), acetyltriethyl citrate (ATEC) and acetyltributyl citrate (ATBC) were donated by Morflex
(Greensboro, US); and soluble polymers: Plasdone® K-17 (PVP K17), Plasdone® K-
29/32 (PVP K29), Plasdone® K-60 (PVP K60), Plasdone® K-90/D (PVP K90),
Plasdone® S-630 (PV/VA) and Viviprint® 540 (MCPVP) from ISP (New Jersey, US).
The properties of the polymeric additives are shown in Table 5.
50 EXPERIMENTAL
Table 4. Chemical structures, molecular weights, solubility in water and acidity/basicity of drugs used.
Acidity/ Solubility in Drug Chemical structure Mw Basicity water at 20ºC
Metformin 165.6 Basic 1 in 2 parts hydrochloride
Weak Theophylline 180.2 1 in 120 parts base
Propranolol 295.8 Basic 1 in 20 parts hydrochloride . HCl
1 in 1000 Sulphaguanidine 232.3 Basic parts
Sulphanilamide 172.2 Basic 1 in 170 parts
491.1
Verapamil Basic 1 in 20 parts hydrochloride
Chlorpheniramine 390.9 Basic 1 in 4 parts maleate
Paracetamol 151.2 Neutral 1 in 70
51 EXPERIMENTAL
Table 5. Chemical structures, typical molecular weights and viscosity of polymeric additives.
Typical molecular Typical viscosity Polymeric additive Chemical Structure K-Value e weight (mPa.s)
N-Vinyl-2-pyrrolidone and 25.4-34.2 58,000a 2.5b vinyl acetate co-polymer PV/VA
Polyvinylpyrrolidone 16-18 8,000a 1.5b PVP K17 PVP K29 29-32 58,000a 2.5b
PVP K60 50-62 400,000 147
PVP K90 85-95 1,300,000a 55.0b
Molecular-composite polyvinylpyrrolidone - 1,500, 000 – 5, 000 - MCPVP 2,000, 000d 20,000c a Weight average, determined by light scattering; b 5% solution in deionized water. Brookfield LVT viscometer (60rpm at 25°C); c Brookfield LV, Spindle 4, 12 rpm at 25°C; d weight range determined by GPC/MALLS; e K-value represents a viscosity index which is calculated by the 2 1/2 2 Fikentscher’s formula: K= (1.5 logηrel -1)/ (0.15 + 0.003c) + (300c logηrel + (c + 1.5clogηrel ) ) / (0.15c + 0.003c ), where ηrel is relative viscosity of aqueous PVP solution to water and c is the concentration of PVP in the aqueous polymer solution. 52
EXPERIMENTAL
B. Methodology
1. Preparation of film forming dispersions a. EC and acrylic dispersions containing plasticizers
Three types of aqueous polymeric dispersions were studied: Surelease,
Aquacoat and Eudragit L 30D. Except for Surelease which already contained medium chain triglycerides as a plasticizer, 30 %w/w (based on polymer weight) of plasticizers were added to the polymeric dispersions which were then diluted with water to obtain final polymer concentration of 10 %w/w. The plasticizers were used at level suggested by the manufacturer where appropriate. The water-soluble plasticizers
(TEC and GTA) were stirred in the polymeric dispersions for 5 h using magnetic stirrers while the water-insoluble plasticizers (DEP, DBP, ATEC and ATBC) were stirred for 24 h.
b. EC dispersions containing polymeric additives
10 to 30 %w/w of polymeric additives (based on total solid content of film) were added to an appropriate amount of EC dispersion. The mixture was then diluted with distilled water to 10 %w/w of EC and was stirred for 5 h using a magnetic stirrer before it was used to cast films. 10 %w/w EC dispersion without any polymeric additives was used as a control. This was prepared by diluting an appropriate amount of EC dispersion with distilled water and the mixture was stirred for 1 h before use.
2. Preparation of films
Films of 200 µm thick were prepared by casting pre-determined amounts of film forming dispersions on leveled polytetrafluoroethylene (PTFE)-coated glass
53 EXPERIMENTAL plates (casting area = 17 cm × 17 cm). Films prepared from Eudragit L 30D were dried at 40 °C and the rest at 55 °C. After 48 h of drying, the films were removed from the glass plates using a sharp knife. Only films that were free from imperfections, such as cracks and presence of air cavities, were used for subsequent tests.
3. Evaluation of film properties
Films were aged at room temperature in desiccators containing silica gel for at least 5 days prior to testing, unless otherwise stated. They were cut into strips of specific sizes for determination of various properties. For evaluation of mechanical properties and puncture strength, the thickness of each piece of cut film was determined at five locations on the film using a digital micrometer (Mitutoyo, Japan).
Only films with thickness deviation less than 10 % of the mean thickness were used for the respective tests.
In part I of the study, film samples used for mechanical testing, assay of plasticizer and moisture content and determination of percent weight change when stored in controlled environment chambers of 30 °C and 50 or 75 %RH. Samples were removed after 0, 3, 7, 14, 28 days of storage for evaluation.
a. Surface morphology
Surface morphology of the films was examined using a light microscope (BX
61, Olympus, Japan). Film surface roughness was evaluated using a scanning probe microscope (SPM – 9500J, Shimadzu, Japan) over an area of 25 µm by 25 µm in the dynamic mode at a frequency of 1 Hz and Z range of 5 µm.
54 EXPERIMENTAL b. Film transparency
The film sample (40 mm x 25 mm, n = 3) was mounted on the cell holder of a spectrophotometer (UV-1201, Shimadzu, Japan) and light transmittance at 600 nm through the film was determined. The mean percent transmittance for each film formulation was calculated.
c. Mechanical properties
The mechanical properties of the films (70 mm x 10 mm) were evaluated using a tensile testing instrument (EZ Test-100N, Shimadzu, Japan) mounted with a
100 N capacity load cell. The test procedure was based on the ASTM D 882 - 75d method using flat-faced metal grips with surfaces laminated with sand paper for better hold. The initial gauge length was set at 50 mm and the extension speed was 5 mm/min. The films were equilibrated for 1 h ambient conditions of prior to test. The ambient conditions for Part I and Part II of the study were 55 ± 2 %RH, 22 ± 2 °C and 47 ± 2 %RH, 25 ± 2°C respectively.
Four mechanical properties, namely tensile strength, percentage elongation at break, elastic modulus and work of failure were computed from the load - strain profile (Figure 7) and film dimensions as shown below.
Lmax Ts = (31) Ai
∆lb E L = (32) li
dL dm EM = (33) Ai
AUC. δ ω = s (34) Ai
55 EXPERIMENTAL
where, Ts is the tensile strength, Lmax, the maximum load, Ai, the initial cross- sectional area of the sample, EL, the percent elongation at break, ∆lb, the increase in length at break point, li, the initial gauge length, EM, the elastic modulus, dL/dm, the slope of the linear potion of the elastic deformation, ω, the work of failure, AUC, the area under the curve and δs, the cross-head speed. Five measurements were taken and the average calculated for each film formulation.
Slope (≅ Elastic modulus) 25
Load at Failure (≅ Tensile strength) 20
Plastic Deformation 15
10
Elastic Applied Load (N) Deformation 5 Area Under the Curve (≅ Work done) (≅ Toughness) 0
0246810 Strain (%)
Figure 7. A typical load - strain profile of film undergoing tensile test.
d. Puncture test
The puncture test (n = 4) was carried out on both dry and wet circular films of diameter 25.2 mm, using a tensile testing instrument (EZ Test-500N, Shimadzu,
Japan) with a 500 N capacity load cell. A conical-shaped puncture probe with an
56 EXPERIMENTAL angle of 60° and lateral area of 5 cm2 was attached to the load cell. The film, sandwiched between rubber and polytetrafluoroethylene gaskets, was placed over the mouth of a stainless steel cup and secured by an open screw cap. The wet film, immersed in distilled water for 24 h, was blotted dry prior to mounting. The puncture probe was driven downward through the center of the mounted film at a crosshead speed of 5 mm/min, with the load-displacement data recorded from the point of contact of the probe with the film until the film was pierced. Puncture strength and percentage elongation were derived from the load-displacement profile based on the following equations (Bodmeier and Paeratakul, 1994):
F γ p = (35) Acs
{(R 2 + D 2 )1/2 - R} ∂ = 100[ p ] (36) R
2 where γp is the puncture strength (N/mm ), F, the load required for puncture, Acs, the cross-sectional area of the film, ∂ , the % elongation, R, radius of the film exposed across the open screw cap and Dp, the displacement of the probe from point of contact to point of puncture.
e. Plasticizer content
The amount of plasticizer in the Aquacoat film was determined using gas chromatography (Model 5890 series II, Hewlett Packard, US) with a a 23.5 m by 0.32 mm fused silica-polyethylene glycol capillary column (HP-FFAP X-linked polyethylene glycol) and a flame ionization detector. About 200 mg of film sample were accurately weighed and dissolved in methanol. An internal standard of 1 ml was then added to the mixture. TEC (5 mM) was employed as the internal standard for
57 EXPERIMENTAL
GTA while GTA (10 mM) was the internal standard for the rest of the plasticizers.
Using methanol, the mixture was made up to a final volume of 20 ml for DBP and 10 ml for the rest. Nitrogen was used as the carrier gas at a flow rate of 28.5 ml/min, with injector temperature at 240 °C and detector temperature at 270 °C. For the assay of
DEP, TEC, ATEC and GTA, the column temperature was increased from 150 °C to
230 °C at a rate of 10 °C/min and held at 230 °C for 3 min. For the assay of DBP and
ATBC, the column was heated up from 150 °C to 240 °C at a rate of 10 °C/min and held at 240 °C for 5 min.
f. Percent weight change of film
Film samples of 2.5 cm by 4.5 cm were accurately weighed and stored in the controlled environment chambers at 30°C and 50 or 75 %RH. At specified time intervals, the films were removed and weighed immediately. The percent change in weight was computed from the difference between the final and initial weights, with respect to the initial weight of the film.
g. Moisture content
The moisture content of the film was determined by Karl Fischer analysis (701
KF Titrino, Metrohm, Switzerland). About 0.2 g of film sample, were accurately weighed and dissolved in methanol before titration. Four sets of measurements were obtained for each film formulation. The moisture content of film was expressed as percent weight of film moisture with respect to weight of film. The percent change in moisture was represented by the difference between the final and initial moisture contents, with respect to the initial moisture content of the film.
58 EXPERIMENTAL h. Water vapour permeability
For Part I of the study
Water vapour permeability of the films was determined using the ASTM water vapour transmission test method (Dry cup method, ASTM E 96 - 95). The film samples (circular pieces with diameter of 7.46 cm) were conditioned by storing at ambient conditions of 22 ± 2°C and 55 ± 2 %RH for at least 5 days. The drying agent, silica gel beads, was activated by heating at 200°C. A sample of 15 g of activated silica gel beads was placed in each aluminium permeability cup (Paul Garner, US), which was sealed with film sample sandwiched between rubber and polytetrafluoroethylene gaskets. The cup was then tightly closed with a screw cap with an opening exposing an effective film area of 24.5 cm2 for water vapour permeation. The whole assembly was weighed and placed in a controlled environment chamber set at 30°C, 50 % RH or 30°C, 75 % RH for 30 h. At specified time intervals, the cup was briefly removed and weighed. At least 3 sets of measurements were obtained for each film formulation. Plots of weight gain against time were constructed.
The water permeation rate was calculated using the following formula:
∆W R = (37) wvp A × ∆p
where Rwvp is the water vapour permeation rate, ∆W, the amount of water vapour permeated through the film, A, the area of exposed film, and ∆p, the vapour pressure difference.
Assuming that the air on the desiccant side was dry (i.e. 0 mmHg of water), the water vapour pressure difference across the film at 30 °C would be 15.5 mmHg
59 EXPERIMENTAL and 23.8 mmHg for cups stored at 50 % RH and 75 % RH respectively (Wiederhold,
1997). The permeability, P was calculated as follows,
P = R wvp x t (38) where t is the thickness of the film.
For Part II of the study
The same ASTM water vapour transmission test method was employed, with different film size, permeability cup and test conditions. The circular film sample
(diameter of 25.2 mm) was placed over the mouth of a stainless steel cup containing
10 g of activated silica gel and held in place with a screw lid with an opening of 5 cm2. The cup was then placed in a controlled environment chamber set at 38 °C and
90 %RH. The amount of moisture transmitted through the film was determined by weighing the cups at specific time intervals of 0, 2, 4, 6, 8, 10, 12, 24, 48, 72 h. A plot of weight gain against time was constructed. The permeability of the film was determined using equations 37 and 38.
i. Thermal properties
Glass transition temperature
The thermal properties of EC and the additives were determined using differential scanning calorimeter (DSC-50, Shimadzu, Japan) which was coupled with a thermal analyzer (TA-60W, Shimadzu, Japan) to the computer. Films of pure EC,
PVP, PV/VA, MCPVP and polymer blends were prepared by casting 5 %w/w of polymer(s) in solvent mixture of methylene chloride and acetone in the ratio of 5:1 onto a polytetrafluoroethylene dish. The film forming solutions were dried by evaporation at about 30 °C for 24 h and the films were then kept in a dessiccator
60 EXPERIMENTAL containing silica gel for 5 days before analysis. About 5 mg of film samples were accurately weighed for determination of thermal properties of the polymers and polymer blends under nitrogen. The samples were first heated to 200 °C at a rate of 20
°C/min to erase the previous thermal history and remove any residual moisture. After cooling, a second scan was carried out at a heating rate of 10 °C/min to 250 °C. The glass transition temperature (Tg) was taken as the temperature corresponding to 50 % of the transition, i.e. the midpoint of the discontinuity in heat flow in the second heating cycle.
Thermal mechanical spectra
The thermal mechanical spectra of the films were obtained using a dynamic mechanical analyzer (DMA 2980, TA Instruments, USA). This technique involves measurement of a sinusoidal strain induced by application of a sinusoidal mechanical force on the film. The response of the film consists of 2 components: one in phase with the applied stress (elastic response) and the other out of phase with the applied stress (viscous response) (Lafferty et al., 2002). Changes in these responses are studied as a function of temperature and frequency. For a viscoelastic material, the elastic response (recoverable energy) is represented by the storage modulus E’ while the viscous response (lost energy) is represented by the loss modulus E”. The tangent of the loss angle, tan δ, is equal to the ratio of the energy lost (dissipated as heat) to energy stored per cycle:
loss modulus E" tan δ = = (39) storage modulus E'
61 EXPERIMENTAL
Plots of elastic response (E’) and mechanical loss (tan δ) with respect to temperature at fixed frequency are constructed. The glass transition temperature corresponds to a sharp drop in E’ and a peak for tan δ.
The instrument employed consisted of a clamp assembly with a plate affixed to the drive clamp. The film sample (10 mm by 5 mm) with thickness varying from
0.225 to 0.348 mm was sandwiched between the plate and stud mounted on the fixed clamps. The clamp assembly was surrounded by a furnace which was used to heat and cool the sample. The samples were tested in tension mode and heated from 20°C to
120°C at 3 °C/min with a frequency of 1 Hz. The maximum tan δ or a sharp drop in modulus was taken as the glass transition temperature. The upper temperature was limited due to the rapidly decreasing mechanical stability of the plasticized sample above its glass transition temperature. All experiments were performed under a dry nitrogen atmosphere.
j. Drug permeability
Drug permeability tests were conducted using acrylate diffusion cells. The film sample (circular piece with diameter of 32.2 mm) was mounted between two compartments, which were then filled with saturated chlorpheniramine maleate solution (20 g/250 ml) and distilled water respectively. The diffusion cells were placed in a water bath at 37 ± 1°C and shaken at a rate of 35 shakes/min. Five ml samples were collected at specified time intervals from the distilled water compartment and assayed for the drug spectrophotometrically (UV-1201, Shimadzu,
Japan).
Calculation of permeability coefficient
62 EXPERIMENTAL a. Simple steady state model
Several mathematical models have been used to determine permeability coefficient (Grassi et al., 1999). Among them, the steady state model is most commonly used. Generally, diffusion in an isotropic medium can be characterized by
Fick’s second law which states that the rate of change in concentration in a volume element within the diffusional field is proportional to the rate of change in concentration gradient at that point on the field. The proportionality “constant” is referred as the diffusivity, D.
The steady state model represents the analytical solution of Fick’s second law assuming sink condition in the receiver volume, a constant drug concentration in the donor volume and neglecting the presence of the stagnant layers. It also assumes that the solute is neither irreversibly bound within the film nor degraded during transport.
Mathematically, Fick’s second law can be represented as
∂C ∂ 2C = D (40) ∂t m ∂ 2 x 2
where Dm is the diffusion coefficient across the membrane and C is the concentration of drug within the membrane, t is time, and x is the space coefficient normal to the surface of the membrane (Shah, 1993). The membrane concentrations at the surfaces can be related to the solution concentrations by the partition coefficient provided that equilibrium exists at the interface of the membrane and solution. The flux can be determined from the following equation (Shah, 1993):
∂C D m K (C d − C r ) J = − D m = (41) ∂x hm
63 EXPERIMENTAL
where K is the partition coefficient, Cr and Cd are the concentrations of the drug in the receptor and donor fluids respectively, and h is the thickness of the membrane.
From equation 41, transport in side-by-side membrane cells can be described by the following equation (Tarvainen et al., 2002):
dC h P = V r [ ] (42) dt A (Cd - C r )
where V is the volume of the receptor fluid (cm3), A is the effective area for mass transfer (cm2) and P is the permeability coefficient (cm2/s). The rate of drug diffusion can obtained from the slope of cumulative amount of drug diffused versus time plot after 3 lag times.
b. Solutions involving quasi-steady state
Equation (42) is based on the assumption that sink condition prevails, and the drug concentration gradient is kept constant during the experiment. However in some cases the donor and receptor concentrations may continue to change even after a long time. In such cases mathematical analysis becomes more complicated. After some finite time, an instantaneous or quasi-steady state will develop and the time course for the diffusional process is initiated and followed after the onset of the quasi-steady state. Assumptions are made that the gradient within the membrane must instantaneously adjust to the external conditions (continuous phase concentrations) and the amount of diffusant in the membrane must be negligible. When the conditions are such that there is a linear fall of concentration within the barrier, the instantaneous concentration gradient may be expressed by (Mooter et al., 1994):
∂C (C − C ) − = o h (43) ∂x h
64 EXPERIMENTAL
where Co and Ch are the respective surface concentrations of the membrane. Co and Ch can be related to the concentrations in the donor and acceptor compartments by incorporation of the partition coefficient, K. A concise mathematical derivation is given by Mooter et al. (1994), for quasi-steady state:
2 PAt ⎡ (C o − 2C a ) ⎤ − = − In ⎢ ⎥ (44) hV ⎣ C o ⎦
where Ca denotes the solute concentration in the receptor P can be calculated from the plot of – In [(Co - 2Ca)/Co] vs time. In this study, the drug permeability coefficients of composite EC films were determined from both simple steady state and quasi-steady models using equations (42) and (44).
k. Swelling and leeaching of film
The initial weight of the completely dried film sample (40 mm × 25 mm) was measured before it was introduced into a bottle containing 50 ml of distilled water.
The bottle was placed in a water bath at 37 ± 1 ºC and shaken at a rate of 35 shakes/min. The film was removed at predetermined time intervals, blotted gently to remove excess water and weighed immediately. The hydrated films were then oven dried for 2 days and weighed. Swelling index was calculated according to the following equation (Blanchon et al., 1991),
100 (Ws - Wd ) Is = (45) Wd
65 EXPERIMENTAL
where Is is the swelling index at specific time, Ws, the weight of the swollen film and
Wd, the initial weight of the film. The amount of extracted components, EXC, was calculated as follows (Lecomte et al., 2003):
100 (Wd - We ) EXC = (46) Wd
where EXC is the amount of extracted components and We, the final weight of the film dried to constant weight. Three measurements were taken for each film formation and the average reported.
4. Preparation of pellets
A blend of 1 kg, consisting of 2 %w/w drug, 25 %w/w MCC and 73 %w/w lactose was mixed using a double cone tumbler mixer (Erweka AR 401, Germany) for
30 min at 6 rpm. The dry powder mixture was transferred to a planetary mixer
(Kenwood Major, UK) and an appropriate amount of water (37.5 to 41 %w/w) was gradually added and mixed over a period of 5 min. The resultant wet powder mass was extruded using a radial screw extruder (Niro E 140, UK) fitted with a screen of 1 mm thickness with 1 mm circular dies. One kg of the extrudate formed was transferred to a spheronizer (Niro S320, UK) with a 300 mm diameter friction plate and rotated for 10 min at 600 rpm. The pellets produced were dried first in a fluid bed
(Niro-Aeromatic, Strea-1, Switzerland) for 30 min at 60 °C inlet temperature and then further dried in an oven at 40 °C overnight. The dried pellets were sieved to obtain size fraction of 710 µm to 1100 µm for coating.
66 EXPERIMENTAL
5. Coating of pellets a. Coating with Aquacoat
The drug-loaded pellets were coated with Aquacoat dispersions plasticized with 30 %w/w (based on polymer weight) of TEC, ATEC, DEP or GTA. The plasticizers were added to the polymer dispersions which were then diluted with water to 15 %w/w polymer. The water-soluble plasticizers (TEC and GTA) were stirred in the polymer dispersions for 5 h using magnetic stirrers while the water-insoluble plasticizers (DEP, and ATEC) were stirred for 24 h. The coating dispersions thus prepared were also stirred throughout the coating process.
The drug-loaded pellets (250 g) were transferred into a fluid bed coater (Strea
I Aeromatic, Muttenz, Switzerland), equipped with a bottom-spray nozzle and operating with an air inlet temperature of 50 ± 5 °C. The pellets were preheated for 10 min before coating. A slow spray rate of 2 g/min was employed for the first 10 min to avoid overwetting of the pellets and drug migration into the film coat. The spray rate was gradually increased to 3 - 5 g/min with atomizing pressure ranging from 1 - 1.2 bar and air volume 70 - 80 m3/h. The total spraying time was 1 to 2 h. The amount of
EC employed was about 10 % of the weight of drug-loaded pellets. On completion of coating, the pellets were fluidized in the coating chamber for a further 15 min to evaporate residual solvent in the film coatings. The coated pellets dried were further at room temperature of 22 ± 2 °C for 24 h before transferring them to controlled environment chambers set at different temperature and humidity levels.
67 EXPERIMENTAL b. Coating with Surelease
The drug-loaded pellets were coated with Surelease dispersion with and without additives (PVP K17, K29, K60, K90, PV/VA and MCPVP). The amount of additives employed varied from 10 to 30 %w/w, based on total solid content of
Surelease dispersion. The required amount of additive was first dissolved in distilled water before adding to Surelease dispersion. The mixture was then diluted with distilled water to obtain 15 %w/w of Surelease solid content. The mixtures containing additives were mechanically stirred for 5 h while those without additives were stirred for 1 h. The coating dispersions thus prepared were also stirred throughout the coating process.
Known weights of drug-loaded pellets (250 g) were transferred into a fluid bed coating apparatus (Strea I Aeromatic, Muttenz, Switzerland), equipped with a bottom- spray nozzle. The inlet temperature of drying air was 50 ± 5 °C. The pellets were preheated for 5 min before coating. Initially, the pellets were coated at a slow spray rate (2 g/min for the first 10 min) in order to avoid overwetting of the pellets and avoid drug migration into the film coat. Subsequently, the coating dispersions was delivered at a flow rate of 3 - 5 g/min and an atomizing pressure ranging from 1 - 1.2 bar was used. Air volume used ranged from 50 - 70 m3/h and the total spraying time was 1 to 2 h. After coating, the beads were dried for a further 15 min in order to evaporate residual solvent in the coating prior to the curing step. A portion of the pellets was cured at 60 °C for 24 h while the rest was stored at room temperature. The film coating consisting of EC with or without additives amounted to about 12 percent of the weight of the drug-loaded pellets. For pellets coated with EC and EC-PV/VA
(7:3), samples of 120 g of pellets were removed from the coating pan when the
68 EXPERIMENTAL coating had reached 4 %, 6 %, 8 %, 10 % 12 % w/w. The coating level was calculated by weight of the sprayed dispersion without additives.
6. In vitro dissolution studies
Drug release from coated pellets was investigated by dissolution study (USP
XXII, method II, Optimal DT-1 dissolution tester, Optimal Instruments, USA). 900 ml of 0.1 M pH 7.4 phosphate buffer or 0.1 M HCl, as dissolution medium, was maintained at 37 ± 1 °C with paddle speed 50 rpm. At predetermined time intervals, 5 ml samples were collected through Pyrex- G3 filter stick (Pyrex, UK) and assayed spectrophotometrically (UV-1201, Shimadzu, Japan) for the drug: 261.6 nm and
264.2 nm for chlorpheniramine maleate in phosphate buffer and 0.1 M HCl respectively; 272.4 nm and 268.6 nm for theophylline in phosphate buffer and 0.1 M
HCl respectively; and 244.6 nm for paracetamol in both media. The dissolution data was obtained from the average of three determinations.
7. Assay of drug content in coated pellets
Predetermined quantities (3g, 1g, and 0.8g of pellets containing chlorpheniramine malaete, theophylline and paracetamol respectively) of each batch of coated pellets were pulverized using a pestle and mortar. A specific amount of the resultant powder was accurately weighed and dispersed in 900 ml of pH 7.4 buffer.
The dispersion was stirred at 50 rpm, 37 °C for 4 h. Aliquots of the dispersion were filtered through Pyrex- G3 filter stick (Pyrex, UK) and assayed spectrophotometrically (UV-1201, Shimadzu, Japan) at appropriate wavelength stated previously.
69 RESULTS AND DISCUSSION
IV. RESULTS AND DISCUSSION
Part 1. Influence of plasticizers and storage conditions on properties of films
A. Film morphology
Both Eudragit L 30D and Surelease films appeared homogeneous and continuous. The texture of Aquacoat films varied with the type of plasticizers used.
Aquacoat films plasticized with DBP appeared to be more flexible, smoother and homogeneous while those plasticized with DEP and ATEC had raised spots and undulating surfaces. Orange peel appearance was also observed for some of the
Aquacoat films. SPM images showed that Eudragit L 30D and Surelease films were relatively smooth with small perturbations and peak heights of less than 0.5 µm. In contrast, Aquacoat films had rougher surfaces with prominent protrusions (Figure 8).
The average peak heights for the plasticized Aquacoat films were greater than 0.5 µm, with DEP plasticized Aquacoat films having the highest mean peak height of more than 2 µm.
All the freshly dried Eudragit L 30D, Surelease and Aquacoat films, except those plasticized with ATEC, appeared transparent initially. Upon exposure to ambient conditions of 22 ± 2°C and 55 ± 2 %RH for about 5 h, Eudragit L 30D and
Surelease films showed little change in transparency but Aquacoat films became translucent as shown by their markedly lower transmittance values (Table 6). Similar findings for Aquacoat films had been reported (Hutchings et al., 1994). Surfactants were employed to form a fine and stable aqueous dispersion of the practically insoluble EC. It was suggested that the translucency of Aquacoat films might be due to the migration of the surfactants to the minuscule ‘islets’ of the film (Binschaedler et al., 1987, 1989).
70 RESULTS AND DISCUSSION
(a) (b)
(c) (d)
Figure 8. SPM images of (a) MA, (b) Surelease, and Aquacoat films plasticized with (c) DEP and (d) DBP.
71 RESULTS AND DISCUSSION
Table 6. Comparison of mechanical propertiesa, film transparencyb and water vapour permeabilityb of EC and MA filmsc.
Films Tensile % Elastic Work of Tensile Film transparency at Water vapour permeability strength elongation modulus failure strength/ 600 nm (% (g h-1 cm-3 mmHg-1) x 104 (N/mm2) at break (N/mm2) (KJ/m2) Elastic transmittance) modulus
0 h 5 h 50 %RH, 75 %RH, 30 °C 30 °C Surelease 4.47 ± 0.12 11.1 ± 0.82 204 ± 21 20.9 ± 1.81 0.0219 66.0 ± 2.53 65.7 ± 2.70 1.23 ± 0.159 1.24 ± 0.169
Aquacoat + DEP 2.55 ± 0.60 2.43 ± 0.47 162 ± 43 1.99 ± 0.69 0.0158 53.9 ± 3.89 18.8 ± 6.01 1.30 ± 0.200 1.41 ± 0.169
Aquacoat + DBP 2.17 ± 0.22 4.15 ± 3.03 140 ± 14 3.50 ± 3.12 0.0156 43.1 ± 0.87 9.03 ± 4.43 1.66 ± 0.141 1.40 ± 0.035
Aquacaot + TEC 1.73 ± 0.18 6.57 ± 1.08 101 ± 19 4.31 ± 1.15 0.0172 40.7 ± 9.77 7.32 ± 0.68 1.49 ± 0.130 1.41 ± 0.236
Aquacoat + ATEC 2.76 ± 0.41 2.89 ± 0.37 157 ± 29 2.62 ± 0.68 0.0176 28.8 ± 5.77 4.20 ± 0.31 1.30 ± 0.272 1.58 ± 0.090
Aquacoat + ATBC 3.04 ± 0.63 4.20 ± 0.18 144 ± 9.1 4.87 ± 0.31 0.0211 44.3 ± 2.21 1.93 ± 0.08 1.36 ± 0.142 1.30 ± 0.127
Aquacoat + GTA 2.57 ± 0.71 1.17 ± 0.70 319 ± 122 0.91 ± 0.59 0.0081 57.0 ± 3.75 3.28 ± 1.59 0.127 ± 0.013 0.242 ± 0.006
Eudragit L 30D 4.42 ± 0.35 221 ± 61.2 180 ± 30 399 ± 129 0.0235 66.8 ± 0.68 70.0 ± 1.4 0.304 ± 0.021 0.399 ± 0.077
a Freshly dried films stored under ambient conditions for 1 h before test. b Freshly dried films evaluated at 0 h and 5 h. c Freshly dried films stored in desicator at ambient temperature for 5 days before test under different conditions.
72 Ambient conditions: 55 ± 2%RH and 22 ± 2 °C
RESULTS AND DISCUSSION
B. Mechanical properties of films 1. Comparison between different polymeric films
An ideal film coat should be hard and tough without being brittle. These properties can be defined in terms of elastic modulus, tensile strength, percentage elongation at break
(Bilmeyer, 1984) as well as work of failure (Parikh, 1993). Elastic modulus is a key indicator of stiffness or rigidity of polymer films (Wu and McGinity, 2000) while percentage elongation at break can be used to predict the ductility of the film. Hence, a hard and tough polymer film has high elastic modulus, high tensile strength and high percentage elongation at break while a soft but tough polymer film has low elastic modulus, moderate tensile strength and high percentage elongation at break. Tough films are also characterized by high work of failure. The ratio of tensile strength to elastic modulus could be related to in situ performance of the film (Rowe, 1983). It is desirable to have films with higher ratio of tensile strength to elastic modulus as coating defects were observed to increase with lower values of this ratio.
The mechanical properties of the films studied are summarized in Table 6. Eudragit
L 30D plasticized with TEC formed a tough film as shown by the high percentage elongation at break and work of failure values, at least 19 times higher compared with the
EC films. Surelease and Aquacoat films had relatively low percentage elongation at break and work of failure values, indicating that they were brittle. Bodmeier et al., (1994) suggested that the interchain hydrogen bonding and bulkiness of glucose subunits of the polymer might have given rise to the brittle nature of the film. Compared to Aquacoat films, Surelease films were tougher as they had higher work of failure, percentage elongation at break values, as well as higher tensile strength to elastic modulus ratio. It was suggested that the presence of sodium lauryl sulfate in Aquacoat might have been responsible for the lower tensile strength (Bodmeier et al., 1994). The higher percentage elongation at break also showed that Surelease films were better plasticized. Surelease is
73 RESULTS AND DISCUSSION stabilized by an anionic surfactant, ammonium oleate in combination with a plasticizer composed of medium chain triglycerides (Colorcon). During film drying, the less stable ammonium oleate would be converted to oleic acid that could act as an additional plasticizer for EC, resulting in a better plasticized film.
Despite varying markedly in the values of percentage elongation at break and work of failure, the films prepared from Surelease, Eudragit L 30D plasticized with TEC and
Aquacoat plasticized with ATBC showed comparable ratios of tensile strength to elastic modulus (0.0211 to 0.0235). Compared to the rest, these 3 films had the highest ratio value and were expected to produce coats of the least defects. Marked variation in the ratio values (0.0081 to 0.0211) was observed for Aquacoat containing different types of plasticizers, indicating the important effects of plasticizers.
2. Influence of plasticizers on properties of films
The effects of plasticizers on mechanical properties of EC films were also compared. The plasticizers used were classified into 3 types: phthalic acid esters (DEP,
DBP), citric acid esters (TEC, ATEC, ATBC) and glycerol acid ester (GTA). DBP and
ATBC increased the percentage elongation at break and work of failure of the films to a greater extent than the corresponding DEP and ATEC (Table 6), suggesting that the plasticizers with larger alkyl substituents produced tougher EC films. This phenomenon was probably associated with higher hydrophobicity of the plasticizers, which was more compatible with water - insoluble EC. It should be recalled that DBP-plasticized films were smoother and homogeneous while DEP- and ATEC-plasticized films had raised spots and undulating surfaces. No clear trends between the effects of the citric acid esters and phthalic acid esters on the percentage elongation at break and work of failure of the films were observed. However, the citric acid esters were found to produce films with relatively
74 RESULTS AND DISCUSSION higher tensile strength to elastic modulus ratios than phthalate acid esters. The glycerol ester, GTA, produced films with the lowest ratio value. The influence of the class of plasticizers on the ratio value was noted. EC consists of ether and hydroxyl groups which are capable of interacting with plasticizer molecules via hydrogen bonds. The 3 classes of plasticizers have different molecular structures with a number of functional groups for hydrogen bonding and are expected to show different extents of interaction with EC.
Hence, the extent of interaction between the plasticizer and the polymer has an important influence on the tensile strength to elastic modulus ratio of the film produced. In this respect, phthalic acid esters were found to be generally better plasticizers than citric acid esters and glycerol esters.
3. Influence of storage time on properties of films
Mechanical properties of a film are critical as they affect the intended function of the film on a dosage form. These properties were found to vary to different extents with time when the plasticized EC films were stored at 30°C, 50 %RH for over 4 weeks
(Figures 9 - 12). The ATBC- and DBP-plasticized films were the most stable as they showed little change in the four mechanical properties studied while TEC- and DEP- plasticized films were the most unstable. With the exception of Surelease films, the tensile strength and elastic modulus of the plasticized films were generally more affected than their % elongation at break and work of failure. Increase in tensile strength and decrease in elastic modulus were observed in most cases, indicating that the films were gradually becoming brittle and rigid on storage.
75 RESULTS AND DISCUSSION
Figure 9. Effect of storage conditions on tensile strength of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. (n = 5, Mean ± SD)
76 RESULTS AND DISCUSSION
Figure 10. Effect of storage conditions on % elongation at break of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. (n = 5, Mean ± SD)
77 RESULTS AND DISCUSSION
Figure 11. Effect of storage conditions on elastic modulus of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. (n = 5, Mean ± SD)
78 RESULTS AND DISCUSSION
Figure 12. Effect of storage conditions on work of failure of EC films: Surelease and Aquacoat films plasticized with DBP, DEP, TEC, ATEC, ATBC, GTA stored at 30 °C, 50 %RH and 30 °C, 75 %RH. (n = 5, Mean ± SD)
Assays of DEP, DBP, GTA, TEC, ATEC and ATBC remaining in Aquacoat films at different time intervals were carried out. The amounts of DEP and GTA were found to decrease significantly after a week of storage at 50 %RH (Table 7). The amount of GTA was not detectable after 3 weeks of storage. More than 30 % of ATEC was also lost during storage at 50 %RH though it was not statistically significant. The loss of plasticizers could be attributed to volatization or degradation of the plasticizers during storage. The percent weight changes of films stored at 50 %RH and 75 %RH were also determined (Table 8).
All Aquacoat films, except for those plasticized with ATBC, generally showed decline in
79 RESULTS AND DISCUSSION
Table 7. Comparison of plasticizer content in Aquacoat films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30 °C.
Plasticizers Percentage of plasticizer content remaining in the film 0 day 3 days 1 week 2 weeks 3 weeks 4 weeks Storage condition: 50 %RH, 30 °C DEP 100 81.83 57.41* 49.59* 51.26* 41.34* DBP 100 94.81 84.20 84.20 84.05 82.01 TEC 100 83.73 75.64 75.00 73.74 72.64 ATEC 100 95.90 90.74 74.26 66.47 66.51 ATBC 100 91.25 89.55 86.96 86.18 84.96 GTA 100 99.30 60.31* 22.13* - - Storage condition: 75 %RH, 30 °C DEP 100 69.73* 49.93* 32.68* 27.84* 27.21* DBP 100 83.03 93.99 90.07 89.07 86.53 TEC 100 80.10 79.68 80.54 79.46 79.13 ATEC 100 92.07 92.48 88.75 78.41* 76.52* ATBC 100 85.73 85.26 84.90 82.97 82.21 GTA 100 97.35 33.63* 18.34* - - *values significant at p< 0.05 compared to initial content in film
weight during storage. The percent weight loss was greatest for Aquacoat plasticized with
GTA, ATEC, TEC and DEP. A two tailed correlation analysis was conducted for percent weight change, plasticizer content, moisture content and mechanical parameters of
Aquacoat films stored at 50 %RH. Only Pearson correlations for percent weight change and content of Aquacoat films plasticized with DEP, GTA or ATEC were found to be statistically significant at 0.05 level (Table 9). This suggested that the decline in percent weight change of Aquacoat films plasticized with DEP or GTA stored for more than 3 days was primary due to loss of plasticizers. Other researchers have also reported loss of plasticizer, such as propylene glycol, during storage (Pickard, 1979; Skultety and Sims,
1987). Statistical analysis also showed significant correlations between tensile strength or elastic modulus and DEP content of films stored at 50 %RH. Pearson correlation between tensile strength and GTA content of films was also found to be significant at 0.05 level.
This showed that the changes in mechanical properties observed were largely due
80 RESULTS AND DISCUSSION
Table 8. Comparison of percent weight change of films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30°C.
Films Storage Percent weight change of films (n = 5) conditions 3 days 1 week 2 weeks 3 weeks 4 weeks Surelease 50 %RH, 30°C 0.474 -0.129 -2.28 -3.59 -1.88 ± 0.04 ± 0.10 ± 0.18 ± 0.80 ± 0.48 75 %RH, 30°C -5.14 -0.282 -1.69 -2.74 -2.15 ± 1.52 ± 0.03 ± 1.00 ± 0.96 ± 1.28 Aquacoat 50 %RH, 30°C -0.428 -0.849 -2.03 -3.59 -4.02 + DEP ± 0.40 ± 0.35 ± 1.23 ± 0.41 ± 0.84 75 %RH, 30°C -0.889 -1.324 -3.86 -7.52 -7.55 ± 0.26 ± 0.29 ± 1.47 ± 1.87 ± 0.44 Aquacoat 50 %RH, 30°C 0.217 -0.123 -0.522 -2.48 -2.74 + DBP ± 0.03 ± 0.28 ± 0.40 ± 0.51 ± 0.32 75 %RH, 30°C 0.628 0.620 -0.032 -1.64 -1.24 ± 0.16 ± 0.12 ± 0.18 ± 0.76 ± 0.53 Aquacaot 50 %RH, 30°C -0.889 -1.324 -3.86 -7.52 -7.55 + TEC ± 0.26 ± 0.29 ± 1.47 ± 1.87 ± 0.44 75 %RH, 30°C 0.968 -0.832 -1.57 -7.38 -10.0 ± 0.12 ± 0.07 ± 0.18 ± 1.40 ± 1.20 Aquacoat 50 %RH, 30°C -0.548 -1.67 -3.54 -5.63 -6.33 + ATEC ± 0.12 ± 0.45 ± 0.36 ± 1.40 ± 1.20 75 %RH, 30°C -0.839 -1.40 -1.87 -3.18 -5.88 ± 0.23 ± 0.25 ± 0.49 ± 1.00 ± 1.10 Aquacoat 50 %RH, 30°C -0.992 0.996 1.51 1.50 1.34 + ATBC ± 0.09 ± 0.20 ± 0.20 ± 0.13 ± 0.18 75 %RH, 30°C -0.601 1.24 1.76 2.11 2.62 ± 0.11 ± 0.31 ± 0.30 ± 1.1 ± 0.39 Aquacoat 50 %RH, 30°C -2.53 -2.85 -6.79 -8.62 -6.87 + GTA ± 0.63 ± 1.40 ± 0.38 ± 0.46 ± 0.91 75 %RH, 30°C -3.43 -5.46 -3.63 -11.3 -7.65 ± 0.69 ± 1.7 ± 2.2 ± 2.20 ± 2.70 Eudragit 50 %RH, 30°C 0.922 -0.129 -0.289 -0.634 -2.30 L 30D ± 0.09 ± 0.18 ± 0.15 ± 0.07 ± 0.58 75 %RH, 30°C -0.107 -2.94 -3.53 -5.48 -3.78 ± 0.09 ± 0.23 ± 0.58 ± 1.6 ± 0.20
81 RESULTS AND DISCUSSION
Table 9. Pearson correlation for films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30 °C.
Films Pearson correlation (significance values) % weight Tensile % elongation at Elastic Work of change strength break modulus failure 50 %RH, 30 °C Aquacoat % weight 1 -0.752 -0.013 -0.717 -0.727 + DEP change (0.085) (0.981) (0.109) (.0.101) Plasticizer 0.823 -0.929 0.133 -0.943 -0.791 content (0.044)* (0.007)** (0.802) (0.005)** (0.061)* Moisture 0.363 -0.400 -0.213 -0.386 -0.372 content (0.48) (0.432) (0.686) (0.449) (0.467) Aquacoat % weight 1 -0.841 0.353 -0.802 -0.241 + TEC change (0.036)* (0.493) (0.055) (0.646) Plasticizer 0.534 -0.469 0.917 -0.854 0.57 content (0.275) (0.348) (0.01)* (0.03)* (0.238) Moisture -0.91 0.943 -0.23 0.694 0.472 content (0.012)* (0.005)** (0.661) (0.126) (0.345) Aquacoat % weight 1 -0.908 -0.721 -0.876 -0.894 + ATEC change (0.033)* (0.169) (0.051) (0.041)* Plasticizer 0.964 -0.821 -0.009 -0.781 -0.629 content (0.008)** (0.045)* (0.987) (0.067) (0.181) Moisture 0.785 -0.365 -0.379 -0.313 -0.466 content (0.116) (0.477) (0.458) (0.546) (0.351) Aquacoat % weight 1 -0.816 -0.078 -0.098 -0.653 + GTA change (0.048)* (0.883) (0.854) (0.16) Plasticizer 0.945 -0.852 -0.344 0.145 -0.787 content (0.005)* (0.031)* (0.504) (0.784) (0.063) Moisture -0.475 0.434 -0.229 0.652 0.208 content (0.341) (0.39) (0.662) (0.161) (0.693) 75 %RH, 30 °C Aquacoat % weight 1 -0.811 -0.093 -0.726 -0.598 + DEP change (0.05)* (0.861) (0.102) (0.21) Plasticizer 0.788 -0.889 -0.044 -0.96 -0.67 content (0.063) (0.018)* (0.935) (0.002)** (0.146) Moisture 0.293 -0.665 -0.324 -0.725 -0.627 content (0.573) (0.149) (0.532) (0.103) (0.183) Aquacoat % weight 1 -0.818 0.837 -0.873 0.804 + TEC change (0.047)* (0.037)* (0.023)* (0.054) Plasticizer 0.368 -0.562 0.464 -0.451 0.349 content (0.472) (0.246) (0.353) (0.37) (0.498) Moisture -0.948 0.802 -0.745 0.854 -0.718 content (0.004)** (0.055) (0.089) (0.03)* (0.108) Aquacoat % weight 1 -0.444 0.948 -0.948 0.961 + ATEC change (0.378) (0.004)** (0.004)** (0.002)** Plasticizer 0.787 -0.501 0.794 -0.659 0.805 content (0.063) (0.311) (0.059) (0.154) (0.053) Moisture 0.186 -0.885 0.455 -0.448 -0.059 content (0.724) (0.019)* (0.365) (0.372) (0.912) Aquacoat % weight 1 -0.058 0.342 -0.787 0.066 + GTA change (0.914) (0.507) (0.063) (0.901) Plasticizer 0.811 -0.589 -0.211 -0.865 -0.491 content (0.05)* (0.218) (0.689) (0.026)* (0.322) Moisture 0.271 -0.54 -0.684 -0.232 -0.657 content (0.603) (0.269) (0.134) (0.658) (0.156)
82 RESULTS AND DISCUSSION to loss of DEP or GTA from films. Similar findings were reported in the study of Eudrgait L
30D films plasticized with triacetin (Gutierrez-Rocca and McGinity, 1994).
Changes in mechanical properties of Aquacoat films plasticized with TEC and ATEC also indicated a hardening of film resulting in increased brittleness. The determination of TEC content by gas chromatography in Aquacoat films showed a loss of approximately 25 percent after a week of storage at 50 %RH. However, no further loss in TEC content was observed on prolonged storage. In contrast, the elastic modulus and % elongation at break values continued to show changes beyond a week of storage. Statistical analysis showed a significant correlation
(p < 0.05, Table 9) between TEC content and percentage elongation at break or elastic modulus of films. This suggests that the initial changes of Aquacoat films were due to TEC loss. After prolonged storage, other factors also contributed to the continual hardening of the films.
Generally, the film forming process depends on the manufacturing method, coating conditions and formulation variables (Onions, 1986; Keshikawa and Nakagami, 1994; Sun, et al., 1999). It was also reported that the degree of coalescence of latex particles increased with storage time (Guo et al., 1993). In most of the studies, the degree of coalescence was found to be greatly affected by storage temperature (Hutchings et al., 1994; Guo et al., 1991, 1993). It was likely that the changes in mechanical properties and loss of weight of Aquacoat films plasticized with TEC and ATEC on prolonged storage was mainly due to the phenomenon of coalescence. On the other hand, changes in mechanical properties of Aquacoat films plasticized with DEP and GTA were probably influenced by two factors, loss of plasticizers and increasing coalescence of polymer particles during storage. This explained the greater magnitude of change in tensile strength and elastic modulus, as well as percent weight change observed for
Aquacoat films plasticized with DEP and GTA.
83 RESULTS AND DISCUSSION
4. Influence of storage humidity on properties of films
The mechanical properties of EC films stored at medium and high humidity levels of 50
%RH and 75 %RH respectively were compared Figures 9 - 12. A humidity of 75 % represents a high humidity associated with outdoor conditions whilst 50 %RH is medium level, reflecting indoor controlled environment conditions commonly seen in local pharmaceutical manufacturing facilities. Changes in the mechanical properties were not always greater for films stored under higher RH of 75 %. For example, the tensile strength of DBP-plasticized films stored at 50 %RH increased significantly with time while those stored at 75 %RH had relatively constant tensile strength. The magnitudes of change in tensile strength and elastic modulus of Aquacoat plasticized with DEP after 3 days of storage at 50 %RH were also greater compared to those stored at 75 %RH for the same number of days. On the other hand,
Aquacoat films plasticized with citrate esters, TEC and ATEC, showed similar differences in magnitude of change in tensile strength, elastic modulus, as well as, percentage elongation at break when stored at different humidity levels. Nevertheless, storage at high or medium humidity levels affected the mechanical properties of the plasticized Aquacoat films to a significant extent.
The moisture content of the films stored at 50 %RH and 75 %RH were determined
(Table 10). Most of the plasticized Aquacoat films stored at both humidities showed an initial increase in moisture content, ranging from 5 to 124 percent after 3 days of storage. On prolonged storage, the moisture content of Aquacoat films plasticized with phthalates (DEP and
DBP) and citrate esters (TEC and ATEC) decreased to lower than or almost similar to initial values. However, there was no significant correlation between the moisture content and mechanical parameters of the Aquacoat films, except those plasticized with TEC (Table 9). The
84 RESULTS AND DISCUSSION influence of moisture on Aquacoat films was temporal and on prolonged storage, other prominent factors such as loss of plasticizers and further coalescence resulted in a greater
Table 10. Comparison of percent change in moisture content of films exposed to storage conditions of 50 %RH, 30 °C and 75 %RH, 30 °C.
Films Storage Percent change in moisture content of films (n = 4) conditions 3 days 1 week 2 weeks 3 weeks 4 weeks
Surelease 50 %RH, 30°C 181 ± 34 54.3 ± 6.9 202 ± 23 207 ± 49 193 ± 25
75 %RH, 30°C 90 ± 18 17 ± 25 219 ± 49 105 ± 46 103 ± 32
Aquacoat 50 %RH, 30°C 14.8 ± 57 -52.0 ± 6.3 -46.5 ± 2.5 -10.2 ± 14 -38.1 ± 4.6 + DEP 75 %RH, 30°C 59.0 ± 57 -28.4 ± 3.4 22.3 ± 5.2 -1.1 ± 3.7 -16.3 ± 8.9
Aquacoat 50 %RH, 30°C 95.0 ± 100 -22.5 ± 2.5 -21.1 ± 2.4 19.9 ± 3.8 -12.4 ± 2.8 + DBP 75 %RH, 30°C 6.81 ± 71 -46.0 ± 3.6 -44.8 ± 7.0 -7.96 ± 13 -40.6 ± 4.6
Aquacaot 50 %RH, 30°C -33.5 ± 2.1 -16.2 ± 6.7 -4.88 ± 11 42.1 ± 24 122 ± 29 + TEC 75 %RH, 30°C 4.57 ± 4.6 23.5 ± 6.2 21.9 ± 6.5 23.2 ± 13 160 ± 64
Aquacoat 50 %RH, 30°C 119 ± 61 45.4 ± 19 -4.89 ± 5.2 7.61 ± 5.2 13.3 ± 3.2 + ATEC 75 %RH, 30°C 124 ± 120 4.89 ± 6.0 -37.5 ± 13 -12 ± 22 16.8 ± 29
Aquacoat 50 %RH, 30°C 50.4 ± 9.4 277 ± 47 50.4 ± 36 37.3 ± 6.7 39.7 ± 10 + ATBC 75 %RH, 30°C -7.54 ± 10 145 ± 110 7.54 ± 5.1 21.8 ± 24 11.5 ± 13
Aquacoat 50 %RH, 30°C 70.1 ± 15 41.8 ± 2.6 53.1 ± 33 41.2 ± 6.4 46 ± 12 + GTA 75 %RH, 30°C 91 ± 130 1.61 ± 13 -6.43 ± 6.9 -8.36 ± 4.9 0.96 ± 3.9
Eudragit L 50 %RH, 30°C 34.2 ± 17 52.7 ± 16 91.1 ± 16 14.5 ± 18 33.2 ± 6.4 30D 75 %RH, 30°C -36.3 ± 8.5 -34.6 ± 6.4 -22.8 ± 7.3 -35.2 ± 14 -42.6 ± 6.6
change in film structure.
For example, Aquacoat films plasticized with ATBC did not show significant change in mechanical properties except for a small rise in elastic modulus value. This slight increase
85 RESULTS AND DISCUSSION might be due to the small amount of ATBC lost from Aquacoat films during storage. It is interesting to note that the moisture content of Aquacoat plasticized with ATBC showed the greatest increase after 1 week, but decreased markedly to near initial value after 2 weeks of storage at both humidities (Table 10). The percent weight of these films reflected a small but sustained increase in weight of the film though this increase was found to be insignificant
(Table 8). The variation in moisture content however did not affect the mechanical properties of Aquacoat films plasticized with ATBC stored at a high humidity. This suggested that
Aquacoat plasticized with ATBC was less influenced by moisture in the environment than those plasticized by other citrate esters.
Films containing GTA which is soluble in water, showed a marked increase in moisture content when stored at both 50 %RH and 75 %RH (Table 10). While the moisture content of the films stored at 50 %RH remained almost constant after 1 week of storage, those stored at 75
%RH showed a significant decline to near initial value. This latter corresponded to a lower
GTA content in the films stored at 75 %RH (Table 7), indicating affinity for GTA for moisture.
Neverthess, the high loss in GTA content of the films stored at both humidities caused modifications in film structure, resulting in similar change in mechanical properties of the films.
Compared to Aquacoat films, the differences in mechanical properties of Surelease films stored at both 50 %RH and 75 %RH were less marked. A small but significant (p < 0.05) increase in tensile strength was observed for Surelease films stored at both 50 %RH and 75
%RH. Independent sample T-tests were conducted and the differences between the tensile strength values for Surelease stored at 50 %RH and 75 %RH for 1, 2 and 4 weeks, were found to be statistically significant at 95 % confidence interval. The percent weight change of
86 RESULTS AND DISCUSSION
Surelease films showed a small decline over 4 weeks of storage at both 50 %RH and 75 %RH
(Table 8). On the contrary, the moisture content showed a significant increase after 3 days of storage. This indicated that other more prominent factors, such as greater coalescence, were responsible for the weight loss of Surelease films. Unlike EC films, there was no prominent change in mechanical properties of MA films stored at high humidity for up to 4 weeks. A small but significant decrease in film weight was observed after 4 weeks of storage (Table 8).
This could be attributed to loss of moisture from films to atmosphere at 75 %RH and 30 °C
(Table 10).
5. Water vapour permeability
The water vapour permeabilities of EC and MA films are shown in Table 6. With the exception of Aquacoat film plasticized with GTA, the rest of the EC films were at least thrice more permeable to water vapor than the plasticized Eudragit film and only slightly more permeable than Surelease films. This implies that the plasticized MA films were denser in structure compared to the plasticized EC films. There was no significant difference in the permeabilities of EC films plasticized with DBP, DEP, TEC, ATEC and ATBC. However, when EC films were plasticized with GTA, the permeability decreased by at least 5 folds, to
1.27 x 10-5 g h-1 cm-3 mmHg-1at 50 %RH and to 2.42 x 10-5g h-1 cm-3 mmHg-1 at 75 %RH.
Among the plasticizers used, GTA showed the greatest loss from films (Table 7). It was therefore likely that during the conditioning of EC films for permeability study, a large amount of GTA was lost by degradation or through volatization. This would result in a re-alignment of the polymer chains and formation of a more compact structure which became less permeable to
87 RESULTS AND DISCUSSION water vapour. Both plasticized Eudragit L 30D and EC films did not show any significant difference (p < 0.05) in the permeability measured at 30°C, 75 %RH and 30°C, 50 %RH.
C. Drug release
It was clearly seen in this study above that properties of plasticized Aquacoat films could be substantially affected by the humdity and temperature of the environment. Further study was conducted to examine the consequential effects that these changes in film properties might have on drug release. Chlorpherniramine pellets were coated with Aquacoat plasticized with DEP, TEC, ATEC and GTA representively. The coated pellets were exposed to environmental conditions of 30 °C and 10, 40 or 75 %RH for over a period of 4 weeks. At predetermined time intervals of 3, 7, 14, 21 and 28 days, samples of pellets were removed for dissolution test. In addition, the effect of storage temperature was also studied by storing the pellets in the oven at 30 °C or 70 °C for 4 weeks.
1. Influence of plasticizer on dissolution profiles of pellets coated with Aquacoat
The release profiles of pellets coated with Aquacoat plasticized with DEP, ATEC and
GTA were similar and closely resembled that of a rapid release formulation with 90 percent of chlorpheniramine released within 30 min (Figure 13). In contrast, pellets coated with Aquacoat plasticized with TEC showed greater retarding effect on the release of chlorpheniramine, as only 80 percent of the drug was released after 10 h. The dissolution data were fitted into zero- order, first-order, Hixson-Crowell, Higuchi equations and analyzed
88 RESULTS AND DISCUSSION
120 Aquacoat + TEC Aquacoat + DEP Aquacoat + GTA Aquacoat + ATEC
100
80
60
40
20 Chlorpheniramine released (%)
0 0 60 120 180 240 300 360 420 480 540 600 Time (min)
Figure 13. Dissolution profiles of Aquacoat coated chlorpheniramine pellets dried at 22 °C, 70 %RH for 24 h.
using linear regression. The fit factors, represented by r2 and AIC values, showed that chlorpheniramine release from the coated pellets was best described by Higuchi or zero-order equations (Table 11). The Higuchi release rate constant of pellets coated with Aquacoat + TEC was at least 3 folds lower and their MDT50% and T50% at least 6 folds higher, compared to other coated pellets (p < 0.05, Table 12). The release of chlorpheniramine from Aquacoat + TEC coated pellets was also characterized by a significantly longer lag time of 48 min in constrast with 3 – 9 min for the rest.
The above findings indicate a wide variation in the effect of plasticizers on drug release through Aquacoat films. Plasticizers are indispensable for aiding coalescence of EC polymer to
89 RESULTS AND DISCUSSION
Table 11. Fit factors of drug release models for pellets coated with Aquacoat containing 30 %w/w TEC, DEP, GTA or ATEC.
Film coat Mathematical models Zero order First order Hixson-Crowell Higuchi
r2 Aquacoat + TEC 0.919 0.763 0.823 0.955 Aquacoat + DEP 0.998 0.959 0.979 0.994 Aquacoat + GTA 0.953 0.807 0.868 0.987 Aquacoat + ATEC 0.985 0.879 0.934 0.983 AIC Aquacoat + TEC -21.9 -1.79 -21.6 -24.9 Aquacoat + DEP -47.1 -16.0 -38.0 -42.6 Aquacoat + GTA -24.1 -0.730 -21.7 -32.1 Aquacoat + ATEC -30.6 -1.70 -24.6 -29.4
Table 12. Lag time, MDT50%, T50% and release rate constants of pellets coated with Aquacoat plasticized with 30 %w/w TEC, DEP, GTA and ATEC.
Film coat Lag time MDT50% T50% Zero order Higuchi (min) (min) (min) (min-1) (min-1/2)
Aquacoat + TEC 47.7 115.5 202.2 0.0026 0.069
Aquacoat + DEP 6.3 14.2 22.5 0.0264 0.225
Aquacoat + GTA 3.1 10.6 18.9 0.0244 0.193
Aquacoat + ATEC 9.1 19.1 29.7 0.0214 0.206
form a homogeneous film. Incomplete coalescence will result in film coat with flaws, cracks or pores through which drug and water can diffuse through easily. The plasticizing effect depends partly on the interaction between the polymer and plasticizer. Greater interaction between the polymer and plasticizer will lead to more uniform and complete films. Hence, TEC was the most effective plasticizer for Aquacoat, as it resulted in the longest lag time and lowest release rate constant. The dissolution profiles showed that drug was released from Aquacoat + TEC
90 RESULTS AND DISCUSSION coated pellets in 3 phases: an initial lag period, which was then followed by a phase of constant release and a slower release in the final phase. The initial lag period was attributed to a combination of factors, such as the time required for wetting of the EC coat, dissolution of TEC and other water-soluble components, as well as diffusion of water molecules through the coat to wet the pellet matrix. Once the film coat and outer matrix were wetted, the water - soluble components began to dissolve and leach out. This created a more permeable coat to drug diffusion. At the final stage, the drug particles in the outer layers were depleted and longer time was needed for the remaining drug to diffuse from the inner core, thus resulting in a slower release.
Conversely, the rapid drug release and short lag time obtained for pellets coated with
Aquacoat + ATEC, DEP and GTA implied that these plasticizers were relatively less efficient in aiding EC film coalescence. The poor plasticizing efficiency of ATEC, DEP and GTA for
Aquacoat was further illustrated by the formation of hard and brittle films in contrast with the comparatively more flexible TEC - plasticized film (Table 6). The brittle films were less able to withstand the physical stress encountered during dissolution and hence ruptured easily, causing mass exodus of drug within 30 min. The release parameters obtained from both model- dependent and model-independent analysis showed a unanimous pattern of increasing release rate for EC coated pellets plasticizied with ATEC, DEP and GTA. This trend correlates with the mechanical properties of the respective films (Tables 6 and 12). As indicated by the percentage elongation at break, elastic modulus and work of failure values, EC films plasticized with GTA were most brittle while those plasticized with ATEC possessed greater flexibility.
91 RESULTS AND DISCUSSION
2. Influence of storage conditions on dissolution profiles of pellets coated with plasticized
Aquacoat a. Citric acid esters (TEC and ATEC)
Generally, a notable right shift trend was observed for release profiles of pellets stored at all level of humidity. In particular, pellets kept at high RH of 75 % showed a distinct right shift in correspondence to the storage time, except at 3 days and 1 week (Figure 14a). On the other hand, dissolution profiles of pelltes stored at medium RH of 40 % rebound toward the left after 2 weeks (Figure 14b). Similar left shift was observed for pellets exposed to 10 %RH for more than 2 weeks (Figure 14c). Pellets kept in low RH of 10 % showed the greatest change in release profile, reducing the total amount of chlorpheniramine released to only 50 percent after
10 h.
The release profiles of the coated pellets were best described by Higuchi equation. The release rate constants, lag time, MDT50% and T50% values of the pellets exposed to different storage conditions for varying period of time were compared (Figures 15a - d). A significant drop (p < 0.05) in Higuchi release rate constant was noted for all pellets after 3 days, irregardless of the storage conditions they were exposed to. The storage humidity however played a role in dictating the degree of change in the release rate. For instance, the release rate constant for 10 %RH decreased by 4 folds compared to 1 fold for 40 %RH.
In general, there was no further decline in release rate for pellets exposed to 10 and 40
%RH after 2 weeks, whereas pellets kept at 75 %RH showed further retardation of drug release even at 4 weeks. Interestingly, the T50% curve of pellets kept at 10 %RH indicated a drop from
900 min at 2 weeks to about 720 min at 3 and 4 weeks, while MDT50%, curve showed only small fluctuation after 1 week (Figures 15c and d). The small dip in T50% was attributed to the
92 RESULTS AND DISCUSSION
(a) (b) 100 100
80 80
60 60
40 40
20 20 0 0 0 100 200 300 400 500 600 0 100 200 300 400 500 600
(c) (d) 100 100
80 80
60 60
40 40 Chlorpheniramine released (%) 20 20
0 0 0 100 200 300 400 500 600 0 100 200 300 400 500 600
Time (min) Figure 14. Dissolution profiles of chlorpheniramine pellets coated with Aquacoat containing 30 %w/w TEC stored at (a) 75 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 10 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 0 day; 3 days; 1 week; 2 weeks; 3 weeks; 4 weeks. 93
RESULTS AND DISCUSSION
0.08 (a) (b) 180
0.07 160
0.06 140
120 0.05 100 0.04 80 0.03 60 Lag time (min) 0.02 Higuchi rate constant 40
0.01 20
0 0 0 7 14 21 28 0 7 14 21 28
(c) 400 (d) 1000
350 900
300 800 700 250 600 (min) 200 500 (min) 50%
150 50% 400 T MDT 300 100 200 50 100
0 0 0 7 14 21 28 0 7 14 21 28
Storage time (days) Figure 15. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat containing 30 %w/w TEC. 75 %RH, 30 °C; 40 %RH, 30 °C; 10 %RH, 30 °C ; 0 %RH, 70 °C.
declining lag time even after 2 weeks of storage (Figure 15b). In order to determine if
dissolution profiles obtained after 1 week of storage at 10 %RH were significantly
different from those kept at 2 weeks or more, the release data were compared using
the pairwise analysis with pellets stored for 1 week as the reference batch. In all cases,
94 RESULTS AND DISCUSSION
the f2 values were greater than 50 and f1 values were smaller than 15, except for pellets stored for 2 weeks. Hence, it may be concluded from the pairwise analysis that there was generally not much difference in the dissolution profiles of pellets stored at
10 %RH after 2 weeks.
The f1 and f2 values for dissolution profiles of pellets stored at different humidity levels were also obtained using control pellets (not subject to storage) as the reference (Table 13). As shown in Table 14, the f1 and f2 values of pellets stored at 10
%RH were greater than 15 and smaller than 50 respectively, indicating less than 10 percent dissimilarity in the dissolution profiles upon storage for 3 days and more. This result supports the trend observed from model - dependent and model - independent analysis. Overall, both the model - dependent and model - independent analysis showed the storage of Aquacoat + TEC pellets at high humidity (75 %RH) or low humdity (10 %RH) resulted in significant changes to the release profiles. The pellets stored at 40 %RH were stable and showed little change in release profile.
As discussed previously, Aquacoat + TEC films showed a gradual weight loss throughout the storage period of 4 weeks at 75 %RH (Table 8). Plasticizer loss and and polymer coalescence were believed to be responsible for the weight loss. Loss of
TEC only affected the changes in the film properties during the initial 3 days since its content remained relatively constant after 3 days (Table 7). Exposure to RH of 75 % caused the film to absorb moisture from the environment which modulated the effect of TEC loss on mechanical properties of the film (Table 9 and Figures 9 - 12). Similar trend was observed for films kept at 50 %RH despite more rapid change in weight.
The percent weight change reached equilibrium faster (within 3 weeks) for films stored at lower RH (Table 8).
95 RESULTS AND DISCUSSION
Table 13. Difference factor, f1 and similarity factor, f2 of pellets coated with Aquacoat.
Storage Peroid Aquacoat Aquacoat Aquacoat Aquacoat conditions + TEC + DEP + GTA + ATEC f1 75 %RH, 30°C 3 days *15.3 *15.8 6.6 *39.9 1 week 14.3 12.2 4.9 *48.4 2 weeks *24.0 11.7 4.8 *50.1 3 weeks *28.7 *16.6 3.1 *51.1 4 weeks *33.0 *15.7 7.6 *29.9 40 %RH, 30°C 3 days 8.1 9.6 3.3 7.2 1 week 7.0 6.5 3.0 13.3 2 weeks *24.3 11.4 2.4 10.1 3 weeks 8.4 15.2 7.5 13.9 4 weeks 7.0 *16.7 6.9 12.4 10 %RH, 30°C 3 days *48.7 *35.0 6.5 7.5 1 week *61.8 *28.4 8.7 9.8 2 weeks *68.9 *24.8 8.8 5.2 3 weeks *62.5 *24.9 7.4 6.3 4 weeks *59.4 *24.4 7.1 8.8 0 %RH, 70°C 3 days *34.4 *16.5 *22.0 *63.1 1 week *43.1 11.4 *20.1 *55.9 2 weeks *42.1 7.4 *15.8 *30.4 3 weeks *21.4 12.0 *16.3 *60.1 4 weeks *27.2 11.5 14.8 *79.9
f2 75 %RH, 30°C 3 days 53.3 *49.2 63.5 *27.7 1 week 53.9 53.9 73.9 *24.2 2 weeks *42.1 53.7 71.2 *21.0 3 weeks *38.3 *46.4 74.3 *23.2 4 weeks *34.7 *47.1 63.0 *35.1 40 %RH, 30°C 3 days 66.5 59.5 76.2 60.9 1 week 68.4 67.3 80.9 49.5 2 weeks *41.9 56.1 82.2 51.6 3 weeks 63.0 49.9 51.5 *48.1 4 weeks 65.2 *48.1 55.7 51.2 10 %RH, 30°C 3 days *26.5 *31.4 65.9 59.0 1 week *21.3 *36.3 61.2 54.5 2 weeks *18.9 *39.1 60.2 67.1 3 weeks *28.0 *39.1 64.7 64.4 4 weeks *22.2 *38.9 64.4 58.1 0 %RH, 70°C 3 days *34.2 *48.1 *43.6 *12.5 1 week *28.8 54.3 *46.1 *14.8 2 weeks *29.6 63.2 *49.2 *28.2 3 weeks *44.9 52.1 *48.7 *17.4 4 weeks *40.0 53.0 50.3 *11.2 * f1 > 15 or f2 > 50 indicates significant difference
96 RESULTS AND DISCUSSION
The decline in release rate of Aquacoat + TEC coated pellets kept at medium to high RH for 3 days was due to an overall increase in moisture content. The plasticizing effect of absorbed moisture caused the film to be more flexible and less vulnerable to impact encountered during dissolution. At the same time, the polymer film underwent further coalescence, which greatly reduced the number of pores or cracks through which the drug escaped. The dissolution profiles of the pellets kept at
75 %RH continued to change throughout the study period of 4 weeks. In contrast, pellets stored at 40 %RH showed only slight fluctuation after 2 weeks, indicating that the coalescence of polymer had reached equilibrium. The greater retardation in drug release achieved for pellets stored at 10 %RH showed that the Aquacoat films had undergone greater extent of coalescence, forming harder films. Consequently, much longer time was necessary for wetting of the coating and outer pellet core before diffusion of the drug could occur. This explained the longer lag time and slower release rate observed (Figures 15a and b). The above results suggested that lower humidity facilitated coalescence of Aquacoat to form less permeable films.
Compared with Aquacoat + TEC coated pellets, those coated with Aquacoat +
ATEC showed less changes in dissolution profile when stored at 10 and 40 %RH. As seen from Figures 16a and b, the dissolution profiles of pellets stored at 10 and 40
%RH were similar to those of pellets before storage. In contrast, a right shift in the dissolution profiles of pellets was observed after 3 days of storage at high relative humidity of 75 %RH (Figure 16c). Drug release rate decreased further for storage up to 2 weeks, after which no further change was observed. However, the dissolution profile of the pellets stored for more than 4 weeks shifted back to the left, suggesting faster release. The shift in dissolution profiles of Aquacoat + ATEC coated pellets may partly be due to the increase in lag time as seen in Figure 17b. The greatest
97 RESULTS AND DISCUSSION
(a) (b) 100 100 80 80 60 60
40 40
20 20
0 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360
(c) (d)
100 100
80 80
60 60
40 40 Chlorpheniramine released (%) 20 20 0 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360 420 480
Time (min)
98 Figure 16. Dissolution profiles of chlorpheniramine pellets coated with Aquacoat containing 30 %w/w ATEC stored at (a) 10 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 75 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 0 day; 3 days; 1 week; 2 weeks; 3 weeks; 4 weeks.
RESULTS AND DISCUSSION
(a) (b) 0.35 100 90 0.3 80
0.25 70
0.2 60 50 0.15 40
0.1 Lag time (min) 30
Higuchi rate constant Higuchi rate 20 0.05 10 0 0 0 7 14 21 28 0 7 14 21 28
(d) (c) 200 350
180 300 160
140 250
120 200 (min)
100 (min) 50% 150 80 50% T MDT 60 100 40 50 20
0 0 0 7 14 21 28 0 7 14 21 28 Storage time (days) Figure 17. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat containing 30 %w/w ATEC. 75 %RH, 30 °C; 40 %RH, 30 °C; 10 %RH, 30 °C; 0 %RH, 70 °C.
increase in lag time was observed for pellets after 3 days of storage. Beyond this, the lag time of pellets stored at 10 and 40 %RH remained almost constant while that of pellets stored at 75
%RH decreased slightly.
Unlike Aquacoat + TEC coated pellets, those coated with Aquacoat + ATEC showed an increase in Huguchi rate constant after 3 days of storage, irrespective of the humidity level
99 RESULTS AND DISCUSSION
(Figure 17a and b). The rate constant of pellets stored at 40 %RH remained fairly constant on prolonged storage while that for stored at 10 %RH decreased after 1 week, increased again after 2 weeks and remained constant thereafter. Interestingly, pellets stored at 75 %RH also showed a gradual and more pronounced decrease after the initial rise at day 3.
According to model-independent analysis, Aquacoat + ATEC coated pellets stored at
75 %RH showed the greatest changes in both T50% and MDT50% values (Figures 17c and d).
In both cases, there was a significant increase after 3 days of storage, which then leveled off for MDT50%. Though independent t-test analysis showed significant increase in T50% and
MDT50% values of Aquacoat + ATEC pellets stored at 10 and 40 %RH, the magnitude of increase was considerably small compared to those stored at 75 %RH. Moreover, pairwise analysis also demonstrated that the dissolution profiles of pellets stored at 75 %RH were more than 10 % dissimilar to those prior to storage. Though the f1 and f2 values for all the batches stored throughout 4 weeks were greater than 15 and smaller than 50 respectively, greater difference and dissimilarity factors were observed for those stored for 2 to 3 weeks
(Table 13). In contrast, no difference or dissimilarity was observed for all the batches of pellets stored at 10 and 40 %RH, except for one batch of pellets stored at 40 %RH for 3 weeks. Taken together, it was clear that storage of Aquacoat + ATEC pellets at high humidity had significant effect on the release of chlorpheniramine while storage in an environment of low or medium humidity did not affect the release profiles.
Aquacoat films plasticized with ATEC or TEC responded similarly to storage humidity. Particularly, Aquacoat films plasticized with ATEC showed an increase in moisture content when exposed to 75 %RH for 3 days (Table 10). The absorbed moisture exerted a plasticizing effect on the film and brought about a decrease in chain stiffness (Wu and Mcginity, 2000; Bicerano, 1996). The combinational influence of film strengthening as a result of coalescence and increased flexibility due plasticizing action of absorbed moisture,
100 RESULTS AND DISCUSSION gives rise to a tougher film that is able to withstand the physical stress of dissolution. The weight of Aquacoat films plasticized with ATEC declined gradually over 4 weeks of storage at both 50 and 75 %RH. This decrease in weight was attributed to ATEC loss and further coalescence of polymer which occurred at a slower rate at higher humidity. Hence pellets kept at 75 %RH had lower release rate than those stored at lower humidity (Figure 17c).
b. Phthalic acid esters (DEP)
The influence of storage humidity on Aquacoat + DEP coated pellets is shown in
Figures 18a - c. For the initial 30 min of dissolution study, drug release profiles of the pellets exposed to 40 and 75 %RH closely resembled that of pellets before storage. More prominent differentiation between dissolution curves for pellets subject to different storage time was observed for 75 %RH and 10 %RH. For those pellets, the rate and amount of drug release generally decreased with increasing storage time.
Analysis of the dissolution data confirmed that Aquacoat + DEP coated pellets were relatively stable to storage at 40 %RH as the release parameters remained fairly constant throughout the test period (Figures 18a - d). Keeping the pellets at low (10 %RH) and high
(75 %RH) humidity, however brought about an overall reduction in both the release rate and lag time (Figures 19a and c). Although the lag time profile portrayed a general declining trend, an initial increase was seen on day 3 (Figure 19b). All the release parameters affirmed that greater changes in chlorpheniramine release were obtained for Aquacoat + DEP coated pellets when stored at low humidity (Figure 19a – d). This trend was further attested to be significant at p < 0.01 from pairwise analysis (Table 13). Pellets exposed to higher RH of 75
% had small variation in MDT50% and T50% and declining lag time and release rate. The
101 RESULTS AND DISCUSSION
(a) (b)
100 100
80 80
60 60
40 40
20 20
0 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360
(c) (d)
100 100
80 80
60 60
Chlorpheniramine released (%) 40 40
20 20
0 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360
102 Time (min) Figure 18. Dissolution profiles of pellets coated with Aquacoat containing 30 %w/w DEP stored at (a) 75 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 10 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 0 day; 3 days; 1 week; 2 weeks; 3 weeks; 4 weeks.
RESULTS AND DISCUSSION
(a) (b) 0.3 12
0.25 10
0.2 8
0.15 6
0.1 4 Lag time (min) time Lag
Higuchi rate constant rate Higuchi 0.05 2
0 0 0 7 14 21 28 0 7 14 21 28
(c) (d) 30 45
40 25 35
20 30
25 (min)
15 (min)
50% 20 50% T 10 15 MDT 10 5 5
0 0 0 7 14 21 28 0 7 14 21 28
Storage time (days) Figure 19. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat containing 30 %w/w DEP. 75 %RH, 30 °C; 40 %RH, 30 °C; 10 %RH, 30 °C; 0 %RH, 70 °C.
opposing effect of reduced lag time substantially negated the effect of decreasing release rate.
DEP was found to be a relatively unstable plasticizer for EC films. When
Aquacoat films plasticized with DEP were exposed to high humidity of 75 %, a great
103 RESULTS AND DISCUSSION amount of DEP was lost over the test (Table 7). DEP loss eventually caused an increase in film hardness characterized by higher elastic modulus and tensile strength
(Figures 9 and 11). The hardened films, having a lower permeability to drug and dissolution media, contributed to decreased release rate. At the same time, the lag time for chlorpheniramine release also decreased, especially after 2 weeks, as a consequence of increased film brittleness arising due to DEP depletion. Nevetheless, once wetted by the dissolution media, the release rate remain fairly constant and similar. DEP loss was significantly less and slower when the films were kept at medium RH of 50 % (Table 7), thus explaining the reduction in lag time of pellets stored for 4 weeks at 75 %RH and the accelerated release during the initial period observed. Storage at low RH of 10 % had a great impact on chlorpheniramine release which closely resembled the trend seen for pellets stored at 75 %RH. Further coalescence of the film was likely the major attributing factor to the changes.
Although coalescence, manifested by increased lag time and decreased release rate, was clearly occurring for coated pellets kept at all humidty levels, pellets stored at 10
%RH experienced a greater degree of coalescence. Low moisture content at 10 %RH was more conducive for coalescence and hence brought about significantly higher magnitude of changes in lag time and release rate.
c. Glycerol acid ester
Unlike other plasticizers, the dissolution profiles of pellets coated with
Aquacoat + GTA were very similar to one another irrespective of the storage conditions and time. Pairwise analysis revealed that there was no significant dissimilarity between the release profiles. Likewise, the realease parameters derived from model-dependent and model - independent analysis showed little variation with
104 RESULTS AND DISCUSSION
(a) (b) 120 100 100 80 80 60 60
40 40 20 20
0 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360
(c) 100 (d) 100
80 80
60 60
40 40 Chlorpheniramine released (%)
20 20
0 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360
105 Time (min) Figure 20. Dissolution profiles of pellets coated with Aquacoat plasticized with 30 %w/w GTA stored at (a) 75 %RH, 30 ºC, (b) 40 %RH, 30 ºC, (c) 10 %RH, 30 ºC and (d) 0 %RH, 70 ºC. 0 day; 3 days; 1 week; 2 weeks; 3 weeks; 4 weeks.
RESULTS AND DISCUSSION
(a) (b)
0.35 6
0.3 5 0.25 4 0.2 3 0.15 2 0.1 Lag time (min) time Lag
Higuchi rate constant rate Higuchi 0.05 1
0 0 0 7 14 21 28 0 7 14 21 28
(c) 14 (d) 25
12 20 10
8 15 (min) (min) 50% 6
50% 10 T
MDT 4 5 2
0 0 0 7 14 21 28 0 7 14 21 28 Storage time (days)
Figure 21. Effect of storage humidity and temperature on (a) Higuchi rate constant (b) lag time (c) MDT50% and (d) T50% of pellets coated with Aquacoat plasticized with 30 %w/w GTA. 75 %RH, 30 °C; 40 %RH, 30 °C; 10 %RH, 30 °C; 0 %RH, 70 °C.
time (Figures 20 a – d).
In the earlier discussion, GTA was shown to have poor plasticizing ability for aqueous EC films. Due to its low stability on storage, virtually no GTA remained in the films even for those kept at less humid environment of 50 %RH (Table 7).
Consequently, Aquacoat films plasticized with GTA were mechanically weakest and
106 RESULTS AND DISCUSSION most brittle compared to other plasticizers (Table 6). Hence, it was not unexpected to see poorly sustained chlorpheniramine release for pellets coated with Aquacoat +
GTA. Nevetheless, the release study proved that Aquacoat membrane plasticized with
GTA were so weak that neither storage at low or high humidity could exert any substantial effect on the film and release properties.
3. Effect of storage temperature on drug release
Storage of EC films at high temperature is known to be capable of either strengthening or weakening the film, as a result of aging or plasticizer loss (Hutchings et al., 1994). The performance and stability of various plasticizers, and the effect on drug release for Aquacoat coated pellets kept at high temperature of 70 ºC was assessed in this study. The greatest change in release profiles was obtained for
Aquacoat coated pellets plasticized with citric acid esters (TEC and ATEC) (Figures
14d and 16d). The initial phase of the dissolution profiles of Aquacoat + DEP pellets appeared to be quite similar to that prior to storage at 70 ºC (Figure 18d). Slight deviation in the dissolution profiles was seen after 30 min and the total amounts of drug released at the end of the dissolution runs were comparatively lower. Pairwise analysis demonstrated that only batches stored for 3 days were found to be dissimilar to coated pellets prior to storage. Unlike other plasticized formula, Aquacoat + GTA coated pellets showed faster drug release when stored at 70 ºC for 3 days (Figure
20d). No further change was observed after 3 days, indicating that whatever major changes had already occurred during the initial 3 days. Exposing the pellets to high temperature, led to accelerated loss of GTA through volatization or degradation causing the plasticized Aquacoat films to become more brittle.
107 RESULTS AND DISCUSSION
Pellets with Aquacoat plasticized with either TEC or ATEC showed drastic changes in their dissolution profiles when stored at 70 ºC (Figures 14d and 16d). After
3 days of storage at 70 ºC the dissolution profiles of these pellets shifted considerably to the right with prolonged lag time. The lag time of Aqucoat + TEC coated pellets increased to a maximum at 2 weeks and decreased on extended storage. In contrast,
Aquacoat + ATEC coated pellets had shorter lag time on storage for more than 3 days
(Figure 17b), which also resulted in a left shift in release profile. Consequently, faster release of chlorpheniramine was obtained after storing Aquacoat + ATEC coated pellets at 70 ºC for 1 week or more. The Hugichi rate constants of these coated pellets were found to decrease significantly after 3 days and no further change was observed on prolonged storage.
The big decline in release rate constant and longer lag time at day 3 revealed a major structural change in Aquacoat films plasticized with ATEC and TEC. It was likely that these changes were driven by further coalescence of polymer, resulting in strengthening of the film structure and improved film integrity. Hence, there were less direct pathways for drug diffusion as the number of pores and cracks was greatly reduced. Moreover, the release rate decreased significantly as it became more difficult for the dissolution media and drug to penetrate through the hardened films.
Subsequent decrease in lag time for Aquacoat + ATEC (3 days) and Aquacoat + TEC
(1 week) indicated signs of film weakening on extended exposure to high temperature. This could be attributed to plasticizer loss, either through degradation or volatization. ATEC has comparatively lower permeance for Aquacoat films than TEC
(Table 7). This contributes to the earlier decrease in lag time (after 3 days) for
Aquacoat + ATEC coated pellets as constrast to 1 week for pellets coated with
Aquacoat + TEC (Figure 15b and 17b).
108 RESULTS AND DISCUSSION
Part 2. Influence of polymeric additives
A. Thermal and dynamic mechanical properties
Many methods have been employed to study interactions in polymeric blends.
They include thermoanalytical techniques, microscopy, light scattering, small-angle neutron scattering, inverse gas chromatography, rheological and spectroscopic techniques (Olabisi et al., 1979). Measurement of glass transition temperature is one of the more common approaches in studying the behaviour of polymer blends
(Sakellariou and Rowe, 1995; Song et al., 1996; Nyamweya and Hoag, 2000). Glass transition temperature is the temperature at which the molecular chains of a polymer obtain sufficient energy to surmount the energy barrier for bond rotation (Wetton et al., 1991). Consequently, the polymer changes from a frozen glass-like condition with very limited mobility to a mobile state and achieves thermodynamic equilibrium instantaneously. Glass transition is a fundamental property of an amorphous material that can determine its end-use properties such as mechanical properties, thermal properties and permeability. Depending on the nature of the interaction between individual components, polymeric blends may be miscible (single phase), partially miscible or immiscible (phase separation). Miscible blends will exhibit a single glass transition temperature at an intermediate value, between the glass transition temperatures of the individual components. For immiscible blends, glass transition temperatures of individual components will remain unchanged. When the solubility limit of one of the polymers in a miscible blend is exceeded, phase separation will occur and an additional glass transition temperature corresponding to the excess polymer will be observed. The final properties of the blend will be determined by its miscibility and phase behaviour.
109 RESULTS AND DISCUSSION
Determination of the miscibility of PVP, PV/VA or MCPVP with EC was carried out by preparing cast films aided by organic solutions. Pure EC, PVP, PV/VA and MCPVP films were also prepared and their thermal properties determined using
DSC. In the first heating cycle, thermograms were erratic in the range of 30 - 110°C for EC and 30 – 150 °C for PVP, PV/VA and MCPVP film samples, possibly due to presence of water or other volatile substances as well as other effects such as sintering of samples (Nishio and Manley, 1988). While the glass transition temperatures of EC and MCPVP film samples were easily detectable from the thermograms of first heating cycle, the thermograms of PVP and PV/VA samples did not show any characteristic drop in the heat flow to suggest glass transition temperature inflections.
However, glass transition temperatures of PVP and PV/VA film samples became visible on second heating cycle at 157.1°C and 100.6 °C respectively. The glass transition temperatures of EC and MCPVP films determined from the second heating cycles were 178.4 °C and 178.3 °C respectively.
DSC scans of film blends of EC with PVP were also determined and shown in
Figure 22. When PVP was mixed with EC, a single endothermic transition recurred in the temperature range of 178 – 180 °C, which was close to the glass transition temperature of EC. As the content of PVP increased, the endothermic inflection of EC became broader and shallower. However, the glass transition temperature remained relatively constant for EC-PVP blends up to a ratio of 4:6. An additional endothermic transition around 150 °C, representing the glass transition temperature of PVP, was observed for EC-PVP (4:6) film. At PVP concentration of 70 percent and above EC endothermic transition disappeared and only PVP glass transition endotherm was exhibited. The presence of a single endothermic transition, representing the composite glass transition temperature value, was used as a criterion for establishing polymer-
110 RESULTS AND DISCUSSION
DSC mW
4.00
k
2.00 j i
h g 0.00 f
-2.00 e
d c -4.00 b a
20 40 60 80 100 120 140 160 180 200 220 240 260 TemperatureTemp [C] (°C) Figure 22. DSC thermograms of (a) EC (b) EC-PVP K29 (9:1) (c) EC-PVP K29 (8:2) (d) EC-PVP K29 (7:3) (e) EC-PVP K29 (6:4) (f) EC-PVP K29 (5:5) (g) EC-PVP K29 (4:6) (h) EC-PVP K29 (3:7) (i) EC-PVP K29 (2:8) (j) EC-PVP K29 (1:9) (k) PVP K29
polymer miscibility. In contrast, the presence of two glass transition temperature
values corresponding to the glass transition temperatures of the individual
components was an indication of immiscible blends. PVP was completely miscible
with EC up to a concentration of 50 percent, as indicated by the presence of a single
endothermic transition. Phase separation was evident in the EC-PVP (4:6) blend and
complete miscibility observed again when the concentration of PVP increased to 70
percent or more.
The DSC thermograms of all the EC-MCPVP blends exhibited a single glass
transition endotherm (Figure 23). However, as the glass transition endotherms of EC
and MCPVP occurred in a similar temperature range of ~ 178 - 180 °C, it was hard to
distinguish if the single endothermic transition was due to the glass transition
endotherm of EC, MCPVP or fusion of both polymers. Hence, it could not be
111 RESULTS AND DISCUSSION
DSC mW 4.00
k 2.00 j
i 0.00 h g
f -2.00
e d -4.00 c b
-6.00 a 20 40 60 80 100 120 140 160 180 200 220 240 260 Temp [C] Temperature (°C)
Figure 23. DSC thermograms of (a) EC (b) EC-MCPVP (9:1) (c) EC MCPVP (8:2) (d) EC-MCPVP (7:3) (e) EC-MCPVP (6:4) (f) EC-MCPVP (5:5) (g) EC-MCPVP (4:6) (h) EC-MCPVP (3:7) (i) EC-MCPVP (2:8) (j) EC-MCPVP (1:9) (k) MCPVP
DSC mW
2.00 k
j
0.00 i
h -2.00 g
f e -4.00 d c b
-6.00 a
20 40 60 80 100 120 140 160 180 200 220 240 260 TemperatureTemp [C] (°C)
Figure 24. DSC thermograms of (a) EC (b) EC-PV/VA (9:1) (c) EC-PV/VA (8:2) (d) EC-PV/VA (7:3) (e) EC-PV/VA (6:4) (f) EC-PV/VA (5:5) (g) EC-PV/VA (4:6) (h) EC-PV/VA (3:7) (i) EC-PV/VA (2:8) (j) EC-PV/VA (1:9) (k) PV/VA.
112 RESULTS AND DISCUSSION concluded from the DSC results if MCPVP was miscible with EC. A broad endotherm
(30 – 120 °C) was observed in the DSC thermogram from MCPVP. This was attributed to the presence of water absorbed from the environment as MCPVP is relatively hygroscopic. Addition of EC at concentration as low as 10 %w/w seemed to be able to effectively reduce the amount of water absorbed by MCPVP.
The DSC thermograms of EC-PV/VA blends were similar to those of EC-PVP blends (Figure 24). A single endothermic transition occurring in the range of 170 -180
°C was observed for EC-PV/VA blends with PV/VA content up to 90 percent. The endothermic peak gradually diminished as the PV/VA concentration increased and disappeared at PVP/VA content of 80 percent and above. EC-PV/VA blends in the ratio of 1:9 and 2:8 also presented a single endothermic transition at 98 – 100 °C region, close to the glass transition temperature of PV/VA. On the otherhand, two endothermic transitions in the regions of 175 - 178 °C and 98 – 99 °C were obtained for EC-PV/VA blends with the ratios of 4:6 and 3:7. This indicated that PV/VA might be partially miscible with EC for concentration of up to 50 percent. Compared to
PVP, PV/VA showed lesser interaction with EC as the polymer blend was immiscible at a lower ratio of 3:7.
Based on the above results, Surelease films consisting of 10 - 30 %w/w PVP,
PV/VA and MCPVP respectively were prepared for study. Further attempts were made to determine any interactions between the polymeric additives and EC.
However, the DSC thermogram of the plain Surelease film did not show any glass transition temperature value, probably due to the presence of other components in
Surelease formulation. In order to study the thermal and mechanical profiles of those composite EC films, dynamic mechanical analysis (DMA) was employed. DMA has been reported to be more sensitive to macroscopic, as well as molecular relaxation
113 RESULTS AND DISCUSSION processes compared to other thermal analysis techniques which depend solely on temperature probe (Wetton et al., 1991). It is widely used to determine glass transition temperature and compatibility between polymers (Lafferty et al., 2002; Park et al.,
2001; Honary and Orafai, 2002). A dynamic thermal analysis profile generally shows
3 forms of transition relaxation. The α transition relaxation, which occurs at the highest temperature is characterized by a large decrease in E’ and a maximum in both the E” and tan δ curves near the inflection point of the E’ curve. The α transition occurs in the amorphous regions of the polymer with the initiation of cooperative micro-Brownian motion of the molecular chains (Kararli et al., 1990). Hence, α transition relaxation corresponds to the glass transition temperature of the polymer. β transition, which occurs at the second highest temperature, is thought to arise due to motion of side groups or smaller unit backbone chains. The γ transition is another transition which occurs below the β transition temperature. In this region, the main chain segments are frozen in and side group motion is made possible by defects in packing in the glassy or crystalline state. It is related to end group rotation, crystalline defects, backbone-chain motions of short segments or groups, and phase separation of impurities or diluents (Kararli et al., 1990; Murayama, 1978; Nielson, 1974; Boyer,
1974).
Figure 25 shows typical DMA thermograms of plain and composite EC films prepared from Surelease. The Surelease films without polymeric additives (EC films) produced only a single peak in the tan δ curve at 77.5°C (Figure 25a). This peak was taken to be the α transition or glass transition temperature of the film. This glass transition temperature was much lower than that of pure EC film (without plasticizer) probably due to the plasticizing effect of medium chain triglycerides in Surelease.
Composite Surelease films with 10 - 30%w/w additives, except for PVP K90, also
114 RESULTS AND DISCUSSION produced a single peak in the tan δ curve, however at higher temperature (Table 14).
Composite EC-PVP K90 film showed another smaller peak at 33 – 54 °C. Control using pure PVP or PV/VA films were not possible as these films were too brittle for
DMA test. Hence it was not certain whether the smaller peak observed was due to the
α transition of PVP polymer or β transition of EC films. As the DSC profile of pure
PVP film sample showed that its glass transition temperature is 157°C, the small peak could not be attributed to α transition of PVP. Presence of any additional peak due to incompatibility of PVP with Surelease should appear between the glass transition temperatures of Surelease (77°C) and PVP (157 °C). Since all the composite
Surelease films showed a single glass transition temperature peak in the DMA profile, it might be concluded that PVP, PV/VA and MCPVP were miscible with EC formed from Surelease at concentration of 30 %w/w or below.
Generally, addition of PVP and MCPVP caused a 5 to 22°C increase in glass transition of Surelease films. The increase in glass transition temperature of composite
Surelease films seemed to be dependent on the proportion of additives present except for PVP K29. Unlike the other film samples, unique sharp peaks were observed for loss modulus of EC-PVP K17 (9:1) and (8:2) films. The peaks occurred at much lower temperatures of 70.4 and 78.7 °C in comparison with those of tan δ. On the other hand, EC-PVP K17 (7:3) did not show such peaks in its loss modulus profile
(Figure 25b). Both changes in loss modulus and tan δ could indicate occurrence of transition of polymer blend from glass to rubbery state. It is not certain if the lower peak temperature peaks in loss modulus of EC-PVP K17 (9:1) and (8:2) films indicate a possible lower glass transition temperature than that suggested by the tan δ profiles.
115 RESULTS AND DISCUSSION
(a)
0.8
0.7
0.6
0.5 δ 0.4 Tan 0.3
EC 0.2 EC-PVP K17 (9:1) EC-PVP K17 (8:2) 0.1 EC-PVP K17 (7:3)
0 20 40 60 80 100 120 Temperature (°C) (b)
60
50
40
70.4°C 30 78.7°C
EC-PVP K17 (7:3) Loss modulus Loss 20 EC-PVP K17 (8:2) EC-PVP K17 (9:1) 10 EC
0 20 40 60 80 100 120 Temperature (°C)
Figure 25. DMA thermograms: (a) tan δ (b) loss modulus profiles of Surelease (EC) and EC-PVP K17 films.
116 RESULTS AND DISCUSSION
Table 14. Glass transition temperature (Tg), film transparency and mechanical properties of Surelease (EC) and composite Surelease films. (mean ± S.D, n = 6, unless otherwise indicated)
a Films Tg Estimated Film transparency Tensile % Elastic Work of Tensile T g at 600 nm strength elongation modulus failure strength/ (% transmittance)b (N/mm2) at break (N/mm2) (J/m2) elastic modulus x 10-2 EC 77.5 - 69.3 ± 2.6 3.17 ± 0.1 8.34 ± 0.8 123 ± 10 11000 ± 1100 2.58
EC-PV/VA (9:1) 78.2 79.1 75.2 ± 0.8 3.59 ± 0.1* 6.30 ± 0.6* 179 ± 7* 8950 ± 1100 2.09*
EC-PV/VA (8:2) 82.9 80.7 57.8 ± 1.0 3.86 ± 0.2* 7.32 ± 0.5 208 ± 7* 11700 ± 1400 1.86*
EC-PV/VA (7:3) 92.5 82.5 45.6 ± 1.7 2.93 ± 0.1 9.53 ± 0.9 155 ± 9* 11800 ± 1100 1.81*
EC-PVP K17 (9:1) 84.4/70.4# 81.1 51.7 ± 1.8 2.56 ± 0.1* 18.70 ± 5.5 75 ± 4* 20300 ± 6800 3.42 EC-PVP K17 (8:2) 91.5/78.7# 85.2 58.2 ± 0.5 1.77 ± 0.1* 15.50 ± 1.5* 85 ± 9* 12000 ± 1500 2.11 EC-PVP K17 (7:3) 96.3 89.9 54.4 ± 1.5 3.10 ± 0.5 4.54 ± 0.9* 171 ± 27 5380 ± 1900* 1.81 EC-PVP K29 (9:1) 99.7 81.1 55.2 ± 2.8 3.83 ± 0.1* 6.70 ± 1.4 198 ± 7* 10300 ± 2700 2.00*
EC-PVP K29 (8:2) 95.7 85.2 40.9 ± 2.0 3.29 ± 0.1 13.80 ± 1.6* 155 ± 8* 20300 ± 2600* 1.85*
117 EC-PVP K29 (7:3) 98.7 89.9 33.9 ± 1.5 3.87 ± 0.1* 9.44 ± 0.4 199 ± 1* 13700 ± 760* 1.94*
RESULTS AND DISCUSSION
Table 14. Glass transition temperature (Tg), film transparency and mechanical properties of Surelease (EC) and composite Surelease films. (mean ± S.D, n = 6, unless otherwise indicated) – continued a Films Tg Estimated Film transparency Tensile % Elastic Work of failure Tensile 2 (°C) Tg (°C) at 600 nm strength elongation modulus (J/m ) strength/ (% transmittance)b (N/mm2) at break (N/mm2) elastic modulus x 10-2 EC-PVP K60 (9:1) 102.0 81.1 17.8 ± 0.5 2.22 ± 0.2* 5.04 ± 1.1* 121 ± 17 4100 ± 950* 1.86 EC-PVP K60 (8:2) 99.1 85.2 9.55 ± 0.1 2.37 ± 0.1* 8.27 ± 0.6 123 ± 5 7920 ± 570* 1.92 EC-PVP K60 (7:3) 102.2 89.9 8.17 ± 0.7 2.28 ± 0.2* 10.40 ± 2.2 119 ± 16 9910 ± 2700 1.93 EC-PVP K90 (9:1) 96.6 81.1 28.8 ± 4.6 3.22 ± 0.2 9.70 ± 1.6 155 ± 12* 13200 ± 2700 1.93*
EC-PVP K90 (8:2) 97.6 85.2 8.01 ± 1.9 3.95 ± 0.1* 10.60 ± 3.0 212 ± 10* 18300 ± 5500 2.11*
EC-PVP K90 (7:3) 99.4 89.9 2.14 ± 1.5 4.03 ± 0.1* 4.93 ± 0.9* 222 ± 12* 7510 ± 780* 1.88*
EC-MCPVP (9:1) 82.7 82.1 7.33 ± 1.1 2.29 ± 0.2* 5.73 ± 2.3 111 ± 6 4530 ± 3000 2.07*
EC-MCPVP (8:2) 94.6 87.4 7.50 ± 2.1 3.21 ± 0.2 5.31 ± 1.3* 178 ± 6* 6730 ± 2200 1.80*
EC-MCPVP (7:3) 98.6 93.3 3.91 ± 0.1 3.27 ± 0.6 6.03 ± 3.9 188 ± 11* 8690 ± 6900 1.73*
a n = 1, value obtained from maxima of tan δ unless otherwise stated, bn = 3, *significant at p < 0.05 compared to EC film, # values obtained from loss modulus 118
RESULTS AND DISCUSSION
The glass transition temperature values of polymer blends may be predicted by the Gordon Taylor equation if they are dependent on the polymer composition (Nair et al., 2001). This equation is based on the additivity of free volumes of the individual components characteristic of ideal mixing (Gordon and Taylor, 1952). It can be expressed as:
Tg12 = (w1Tg1 + Kw2Tg2)/ (w1 + Kw2) (47)
where Tg12 is the glass transition temperature of the polymer blend. w1, w2, Tg1 and
Tg2 are the weight fractions and glass transition temperatures of the polymers. The constant, K, is a measure of interaction between the components and can be estimated using the following equation (Simha and Boyer, 1962):
K = ρ1Tg1/ ρ2 Tg2 (48)
where ρ1 and ρ2 refer to the true densities of the components.
The glass transition temperature values of the composite Surelease films were calculated using the Gordon Taylor equation (Table 13). The difference between the actual and estimated glass transition temperature values suggested deviation from ideal behaviour, which was often attributed to differences in strength of intermolecular interactions between individual components and those of the blend.
The glass transition temperature will be higher than expected if the two polymers bind more strongly to each other than to themselves. This is because stronger binding lowers chain mobility. In contrast, if the polymers bind less strongly to each other than to themselves, the glass transition temperatures of the blends are usually lower than expected. All the experimentally determined glass transition temperature values
119 RESULTS AND DISCUSSION of composite Surelease films were higher than the calculated values, suggesting than
Surelease interacted strongly with the additives.
Among the additives, PV/VA generally resulted in the least increase in glass transition temperature of Surelease films, followed by MCPVP and PVP. In order for the polymers to form compatible blends, favorable intermolecular interactions must occur between the different polymer chains (Hale and Blair, 1997). These interactions should result in a negative free energy of mixing which can be expressed by equation
8. As the entropy of mixing in polymeric blends is small, the enthalpy of mixing is the primary factor determining the miscibilities of components. A negative enthalpy of mixing can be produced by specific interactions between the constituent polymers, such as hydrogen bonding. Taylor and Zografi (1998) had reported that interaction by hydrogen bonding could occur in a mixture of amorphous oligosaccharides and PVP via the sugar hydroxyl groups and the proton-accepting carbonyl moieties in the pyrrolidone rings of PVP polymer chains. Since EC is a partially substituted polysaccharide with unsubstituted hydroxyl groups on the cellulose chains, it is likely that EC interacts with PVP in a similar manner (Nishio et al., 1990).
Glass transition temperature may also be a useful parameter for determining the efficiency of plasticizers. Dechesne et al., (1984) reported that addition of plasticizers to acrylic resin copolymers lowered glass transition temperature. The magnitude of change was found to be dependent on the quantity and type of plasticizer used. In this study, the increase in glass transition temperature values by the addition of PVP (except PVP K17), MCPVP and PV/VA showed that these additives had no typical plasticizing effect on Surelease films. In contrast, when present in small amounts (10 – 20 percent), PVP K17 resulted in slight depression of glass transition temperature values of the composite Surelease films. This indicated
120 RESULTS AND DISCUSSION that PVP K17 might be capable of acting as a plasticizer when added in small amount to Surelease.
B. Film morphology
Figure 26 shows the light microscope images of Surelease and composite
Surelease films. The control Surelease films appeared smooth and relatively transparent with transmittance of 69.3 percent. A 25 percent drop in transmittance of
Surelease film was observed with the addition of PVP K17 (Figure 27). The reduction in transmittance was independent of the concentration of PVP K17 (10 to 30 %w/w) in the film.
(a) (b)
(c) (d)
121 RESULTS AND DISCUSSION
(e) (f)
Figure 26. Light microscope images of Surelease (EC) and composite Surelease films (40x magnification): (a) EC, (b) EC-PVP K17 (9:1), (c) EC-PVP K29 (9:1), (d) EC- PVP K90 (9:1), (e) EC-MCPVP (9:1) and (f) EC-PV/VA (7:3).
80 ) 70
60
50
40
30
20
10 Light transmittance at 600 nm (% 0 30% 20% EC 10% EC-PVP K17
Concentration EC-PVP K29 EC-PVP K60 EC-MCPVP EC-PV/VA of additives: EC-PVP K90 10% 20% 30%
Figure 27. Effect of type and concentration of polymeric additives on the film transparency of Surelease films.
122 RESULTS AND DISCUSSION
Addition of PVP K29 resulted in 20 to 51 percent reduction in film transparency. Unlike, PVP K17, greater reduction in transmittance was obtained with higher concentrations of PVP K29. Films containing PVP K60 and K90 were much less transparent. Addition of 10 %w/w PVP K90 reduced the film transparency by 58 percent and almost opaque films, with a transmittance of only 2.14 percent, were obtained with 30 %w/w K90. Similarly, MCPVP drastically reduced the transparency of Surelease films. As the above additives vary in molecular weights, the results suggested that molecular size of PVP might play an important role affecting the clarity of the films.
In composite polymer films, the major constituent generally forms the continuous phase while the minor component if insoluble, is dispersed as discrete individual entities, often spherical. For example, blends of EC and hydroxypropylmethyl cellulose exhibit single layer morphology with the minor polymer dispersed in the major one (Sakellariou et al., 1993). Incompatibility between two polymers could be manifested as the separation of film layers. In one study, composite EC-cellulose acetate phthalate (1:1) films were found to separate into two distinct layers, with EC rich layer on top and cellulose acetate phthalate rich layer below (Sakellariou et al., 1991). Light microscope image of Surelease film showed a continuous, homogeneous layer with predominantly circular inclusions (Figure 26a).
With the addition of low molecular weight PVPs (K17 and K29), small circular randomly dispersed crevices were found throughout the film layer (Figures 26b and c). Increasing the concentration of PVP K17 or K29 resulted in greater number of crevices, as well as slight increase in size of the crevices. Hence, it was thought that the crevices could be attributed to the presence of PVP K17 or K29 randomly dispersed throughout the Surelease matrix. Similar observation was seen for films
123 RESULTS AND DISCUSSION
(a) (b)
(c) (d)
(e)
Figure 28. Models of interaction between PVP, MCPVP or PV/VA with Surelease (EC): (a) continuous phase of EC film without polymeric additives (b) minor component existing as random spheres/crevices in the continuous phase of EC (c) minor component existing as random spheres/crevices of larger size in the continuous phase of EC (d) minor component forming a separate continuous phase (e) overcoat of one continuous phase over another.
with a low amount (10 %) of higher molecular weight PVPs (K60 and K90) (Figure
26d). However, with greater amount of PVP K60 (30 %w/w) or PVP K90 (20 %w/w),
124 RESULTS AND DISCUSSION a separate irregular network-like layer was prominently exhibited above a less obvious layer. At even higher concentration of PVP K60 or K90, the composite
Surelease films appeared as a dense layer with several small holes. This suggested that at concentration of 20 %w/w or more, higher molecular weight PVPs aggregated together to form a separate phase during film drying. A similar irregular network-like layer was seen for composite EC-MCPVP films with MCPVP concentration as low as
10%. At higher concentrations of MCPVP, the composite EC films became denser and similar findings were obtained for EC-PVP K90 films.
Results based on visual appearance and transparency of the film suggested that high concentration of large molecular weight polymers, such as PVP and MCPVP, were not fully miscible with Surelease. At a high concentration, the large molecular weight polymer had a tendency to separate out during the drying process, forming the upper layer of the film. In contrast, the polymer in low concentrations formed a homogeneous layer with the polymer interspersed in the EC matrix. Higher molecular weight PVP at low concentrations produced a greater observable effect than lower molecular weight PVP at high concentrations. EC-PV/VA (9:1) films were smooth.
Similarly, as the amount of PV/VA increased, a separate irregular network-like layer was observed, particularly at 30 %w/w. This suggested that PV/VA at low concentration was miscible with EC, forming homogeneous and smooth films.
However, beyond a threshold concentration, PV/VA aggregated to form a separate layer.
Based on the above observation, models for interaction between Surelease and
PVP, MCPVP or PV/VA were postulated (Figure 28). The interaction between
Surelease and PVP polymeric additives was found to be dependent on the molecular weight, concentration and chemical nature of the additives. When added to Surelease,
125 RESULTS AND DISCUSSION low molecular weight PVP was randomly distributed in the Surelease matrix as shown in Figure 28b. As the concentration of polymeric additives increased, low molecular weight PVP remained as a disperse phase in the form of a large number of vescicles in the continuous Surelease phase (Figure 28c). The vescicles increased in size by the fusion of adjacent vesicicles. On the otherhand, interaction between Surelease and higher molecular weights PVP was more pronounced. At a low concentration, higher molecular weight PVP existed as a disperse phase in Surelease matrix (Figure 28c).
However, as the concentration increased, the higher molecular weight additive tended to aggregate to form a continuous separate phase (Figure 28d). Formation of a continuous separate layer became more prominent with increasing concentration as indicated by huge patches or coat above the main film layer (Figure 28e). Increased molecular weight of additives accelerated the phase separation. When the molecular weight is high, a continuous separate phase may be formed even at low concentration of 10 %w/w, as in the case of MCPVP.
C. Mechanical properties - tensile test
The stability of a film coat and its ability to withstand physical stress is of major concern to the formulator. Film hardness, toughness and extendability are desired mechanical characteristics of a polymer (Mauger, 1999; Aulton et al., 1984).
Many studies have shown that the occurrence of adhesion and cohesion related film coating defects are mainly due to the building up of internal stresses within the film coat (Rowe, 1981, 1992). These internal stresses have been shown to be due to
i. shrinkage of film after its solidification point as solvent continues to
evaporate from the system (Croll et al., 1979);
126 RESULTS AND DISCUSSION
ii. difference between thermal expansion coefficients of film coat and
substrate, and difference between glass transition temperature of film coat
and actual temperature of film at any time (Sato et al., 1980); and
iii. any volumetric changes in tablet core during coating and subsequent
storage (Rowe, 1983; Okutgen et al., 1991).
When the total internal stress exceeds the tensile strength of the film, defects such as cracking, edge splitting and peeling can occur. Hence, polymeric films with high tensile strength may be able to withstand the internal stresses better and are less likely to result in film defects. Commercial EC pseudolatexes were found to produce rigid and brittle films, thus impairing the stress resistance of EC coating (Bodmeier and
Paeratakul, 1994). This property is undesirable in applications where coating of high flexibility and toughness is required, for example in tablets compressed from coated beads or cores with deep logos and break lines. Therefore, it is important to investigate if the addition of polymeric additives has any effect on the mechanical property of EC films.
Table 14 shows the mechanical properties of Surelease and composite
Surelease films. EC films formed from aqueous dispersions were found to be weaker and more brittle than those prepared using organic solvents, probably due to different mechanisms of film formation (Tarvainen et al., 2003). Except for PVP K17 and K60, the addition of PVP increased the tensile strength of composite Surelease films slightly. Tensile strength of composite Surelease films increased slightly with increasing content of PVP K90. In contrast, the increase in tensile strength of EC-PVP
K29 films did not appear to be dependent on the concentration of PVP K29.
Composite EC films containing PVP K17 and K60 showed significantly lower tensile strength compared to Surelesae films. In fact, the tensile strength of composite
127 RESULTS AND DISCUSSION
Surelease film containing 10 %w/w PVP K17 was almost 50 percent lower than that of Surelease film. Addition of PVP K17 also affected that percentage elongation at break of Surelease film significantly. Presence of PVP K17 up to 20 %w/w made
Surelease film more flexible with a higher percentage elongation at break. However, with 30 %w/w PVP K17, the Surelease film become much less flexible as the percentage elongation at break value dropped to 4.54. Similar trend was also observed for the hardness of composite EC-PVP K17 films, with significant drop in elastic modulus when 10 to 20 %w/w PVP K17 was added to Surelease films. These observations suggested that PVP K17 at low concentration (≤ 20 %w/w) has a plasticizing effect on Surelease film.
The addition of PVP of higher molecular weight than PVP K17 did not affect the percentage elongation at break of Surelease films, except for films containing 20
%w/w PVP K29 and 30 %w/w PVP K90. EC-PVP K90 (7:3) film was comparatively more brittle than EC film, as shown by its significantly lower percentage elongation value. Unlike EC-PVP K17 films, all the rest of EC-PVP films showed significantly higher elastic modulus values than the EC film. Addition of 10 – 20 %w/w PVP K90 caused the elastic modulus of EC films to increase by 26 - 72 percent. However, further increase in PVP K90 content to 30 %w/w did not result in further significant rise in elastic modulus.
The increase in elastic modulus and tensile strength indicated that the addition of PVP increased the hardness of Surelease film. EC film is rigid by nature, due to the bulky glucose subunits of the polymer, as well as the alkyl substituents serving as sites for potential inter- and intramolecular interactions (Sakellariou and Rowe, 1995;
Bodmeier and Paeratakul, 1994). PVP is also a large molecule consisting of polar imide group, four non-polar methylene groups and a methane group. Hence, these
128 RESULTS AND DISCUSSION molecules can interact through dipole-dipole attraction and form hard and clear films
(Barabas, 1989). Moreover, the ether and hydroxyl groups of EC could also interact with the imide and carbonyl groups of PVP via hydrogen bonding. These interactions may also enhance the strength of the composite films.
Interestingly, when MCPVP was added to Surelease, the tensile strength decreased or remained unchanged while the percentage elongation at break dropped to
5.3. Addition of 10 %w/w MCPVP did not result in significant change in elastic modulus. However, the elastic modulus increased significantly when the amount of
MCPVP was increased to 20 %w/w or higher. The increase in elastic modulus of EC-
MCPVP was comparatively smaller than that of EC-PVP K90 films. Although
MCPVP has a higher molecular weight than PVP K90, it has less sites available for inter-molecular interaction due to internal cross-linking. This probably explained the above observations. The EC-MCPVP films, at all concentrations of MCPVP, were also less flexible and more brittle than Surelease films. This suggested that some
MCPVP molecules had interposed among the EC rich region. Being large in size, the interposed MCPVP would weaken the intermolecular bonding between the EC chains, causing the film to be more brittle. However, larger amounts of MCPVP enabled the formation of MCPVP rich region above the EC layer, thereby forming a dense film.
The intermolecular bonding between MCPVP molecules was probably greater than those between EC and MCPVP, resulting in slightly harder and tougher films.
Addition of PV/VA increased the tensile strength and elastic modulus of the film. Beyond 20 %w/w of PV/VA, the films showed lower tensile strength, elastic modulus and work of failure. PV/VA which has similar molecular weight as PVP
K29, is less hygroscopic than PVP, due to its vinyl acetate co-polymer. In addition to the polar imide and carbonyl groups of its N-vinyl-2-pyrrolidone component, PV/VA
129 RESULTS AND DISCUSSION also has a carbonyl group from its vinyl acetate co-polymer, which can participate in intermolecular hydrogen bonding. Therefore, PV/VA would likely interact to a greater extent with EC, forming stronger films than PVP K29. However, this was not reflected in the tensile strength and elastic modulus values of EC-PV/VA films, which were comparable to those of EC-PVP K29 films.
The incidence of cracking or edge splitting was employed as a measure of mechanical properties of film coatings applied to tablets (Row, 1981; Okhamafe and
York, 1987). The occurrence of film coating defects has been associated with the magnitude of internal stress in the film. Rowe (1983) suggested that the overall internal stress, P, in a film coat can be represented as follows:
P = EM/3(1-v)[(φs - φr)/(1- φr) + ∆αcubic ∆T] (49)
where EM is the Young’s modulus of the film, v its Poisson’s ratio, φs the volume fraction of solvent at the solidification point of the coating formulation, φr the volume fraction of solvent remaining in the dry film at ambient conditions, ∆αcubic the difference between the thermal cubical expansion coefficients of the coating and the tablet core and ∆T the difference between the glass transition temperature of the coating and the ambient temperature. Cracking will occur if P is greater than or equal to the tensile strength, τ, of the film (Sato, 1980)
P ≥ τ (50)
Combining and rearranging both equations, cracking will occur if:
τ /EM ≤ 1/3(1-v)[(φs - φr)/(1- φr) + ∆αcubic ∆T] (51)
130 RESULTS AND DISCUSSION
Hence, τ/EM may be used to predict the incidence of edge splitting of a film coat
(Rowe, 1981). The larger its value, the higher the stress crack resistance of a film. In this study, the Surelease films without additives showed the highest value of 0.0258.
Addition of PVP, MCPVP and PV/VA caused a significant drop in the ratio to varying extent. Hence, there was a limit to the concentration of additives that could be added to Surelease in order not to compromise the stress crack resistance of the resultant film. The lowest τ/EM value was obtained for EC-MCPVP (7:3) films, suggesting that MCPVP might cause greater impairment to flexibility of Surelease films particularly at concentrations greater than 10 %w/w.
D. Mechanical properties - puncture test
Film coatings may be applied to particles that will be compressed into tablets.
Hence, they should have the required resistance to damage caused by compression.
Radebaugh et al., (1988) has reported a new device which can determine the process of puncture on polymeric films directly. It measures the resistance of a film sample, which is mounted on a fixed holder, to deformation with a puncturing probe attached to a force transducer. Due to its design, this device also enables the study of wet film samples, which is difficult to evaluate using the tensile tester. The mechanical properties of wet films under puncture stress are useful for predicting how polymeric films would behave during dissolution studies or under in vivo conditions (Remunan-
Lopez et al., 1997).
Two parameters, puncture strength and percentage elongation, were derived from the load - displacement profiles (Table 15). When the Surelease films were soaked in water for 24 h at 37 °C, the puncture strength only decreased slightly from
0.378 to 0.332, while a big increase in the percentage elongation was observed from
131 RESULTS AND DISCUSSION
Table 15. Mechanical properties of dry and wet Surelease (EC) and composite EC films. (mean ± S.D, n = 4)
Films Puncture strength (N/mm2) % elongation Dry Wet Dry Wet
EC 0.378 ± 0.00 0.332 ± 0.03 0.192 ± 0.06 0.821 ± 0.28
EC-PV/VA (9:1) 0.182 ± 0.01 0.192 ± 0.03 0.079 ± 0.02 0.481 ± 0.20
EC-PV/VA (8:2) 0.458 ± 0.02 0.286 ± 0.03 0.056 ± 0.00 2.170 ± 0.58*
EC-PV/VA (7:3) 0.839 ± 0.09 0.432 ± 0.04 0.258 ± 0.07 1.380 ± 0.45*
EC-PVP K17 (9:1) 0.265 ± 0.00* 0.135 ± 0.04 0.159 ± 0.03 0.136 ± 0.01*
EC-PVP K17 (8:2) 0.317 ± 0.01 0.184 ± 0.03 0.297 ± 0.06 0.262 ± 0.02*
EC-PVP K17 (7:3) 0.376 ± 0.02* 0.169 ± 0.02 0.198 ± 0.03 0.143 ± 0.04*
EC-PVP K29 (9:1) 0.531 ± 0.02* 0.322 ± 0.03 0.184 ± 0.01* 1.470 ± 0.08
EC-PVP K29 (8:2) 0.410 ± 0.02* 0.152 ± 0.02 0.122 ± 0.02 0.189 ± 0.02*
EC-PVP K29 (7:3) 0.983 ± 0.20 0.231 ± 0.06 0.280 ± 0.15 0.570 ± 0.61
EC-PVP K60 (9:1) 0.396 ± 0.05 0.145 ± 0.02 0.162 ± 0.05* 0.272 ± 0.15
EC-PVP K60 (8:2) 0.558 ± 0.06 0.130 ± 0.01 0.272 ± 0.03* 0.195 ± 0.02*
EC-PVP K60 (7:3) 0.607 ± 0.06 0.109 ± 0.00 0.280 ± 0.04* 0.180 ± 0.05*
EC-PVP K90 (9:1) 0.298 ± 0.04* 0.164 ± 0.01 0.159 ± 0.01* 0.230 ± 0.02
EC-PVP K90 (8:2) 0.434 ± 0.02* 0.156 ± 0.04 0.136 ± 0.02* 0.203 ± 0.09*
EC-PVP K90 (7:3) 0.666 ± 0.11 0.102 ± 0.02 0.254 ± 0.03* 0.170 ± 0.06*
EC-MCPVP (9:1) 0.267 ± 0.01* 0.201 ± 0.02 0.144 ± 0.02* 0.553 ± 0.21
EC-MCPVP (8:2) 0.470 ± 0.07* 0.093 ± 0.01 0.172 ± 0.01* 0.245 ± 0.05*
EC-MCPVP (7:3) 0.824 ± 0.04 0.047 ± 0.01 0.374 ± 0.09* 0.809 ± 0.12*
*significant at p < 0.05 compared to EC film
132 RESULTS AND DISCUSSION
0.192 to 0.821.
This showed that water has a plasticizing effect on Surelease film, resulting in films with greater flexibility upon wetting. Addition of PVP, except for PVP K17 and
10 %w/w PVP K90, increased the puncture strength of dry Surelease films. The increase in puncture strength appeared to be related to the amount of PVP added, with greater increase observed with a higher concentration of PVP. The influence of concentration of PVP on the percentage elongation of dry Surelease films was less obvious as the values fluctuated.
Nevertheless, addition of 30 %w/w PVP generally resulted in the highest percentage elongation for dry Surelease films. On the contrary, opposite trend was observed for wet EC-PVP films. Except for PVP K17, the highest puncture strength and percentage elongation were obtained for wet composite EC-PVP containing the lowest concentration of PVP and increasing amount of PVP generally decreased the puncture strength and percentage elongation.
The results from puncture test of dry Surelease films were similar to those obtained from tensile test. Both tests showed that the composite films became harder with the addition of PVP. However, Surelease films containing PVP K17 showed a slight drop in puncture strength and increase in percentage elongation. These observations suggested that PVP K17 might have some plasticizing effect on
Surelease films as discussed earlier.
The mechanical properties of wet EC-PVP films provided additional insight into the structural arrangement between EC and PVP. When soaked in water for 24 h, the water-soluble PVP that was interspersed within the EC matrix dissolved forming crevices filled with water. This largely weakened the EC matrix and resulted in decrease in puncture strength and percentage elongation. PVP K90 exerted a greater
133 RESULTS AND DISCUSSION weakening effect on EC matrix than PVP K29, as shown by the lower puncture strength and percentage elongation of corresponding wet EC-PVP K90 films.
The above phenomenon was also observed for both dry and wet composite
EC-MCPVP films. However, increasing PV/VA concentration resulted in increasing puncture strength and percentage elongation for both dry and wet films. Addition of
10 %w/w PV/VA reduced the puncture strength and percentage elongation of the dry and wet films by 40 to 60 percent. With greater amounts of PV/VA, the puncture strength of composite EC-PV/VA films increased to as high as 0.839 N/mm2 when dry and 0.432 N/mm2 when wet. The percentage elongation of wet EC-PV/VA films also increased to almost two-fold suggesting that the wet EC-PV/VA films were more
‘intact’ or compact, allowing the water molecules to be retained in the matrix where they exerted a plasticizing effect.
It was also noted that with 30 %w/w of additives, puncture strength of dry composite EC films decrease in the following order: EC-PVP K29 >EC-PV/VA >
EC-MCPVP > EC-PVP K90 > EC-PVP K60 > EC-PVP K17. In contrast, the puncture strength of wet composite EC films generally decreased in the following order: EC-PV/VA > EC-PVP K29 > EC-PVP K17 > EC-PVP K60 > EC-PVP K90 >
EC-MCPVP. As discussed earlier, MCPVP in high concentration probably formed a separate layer above the EC matrix. When the composite film was soaked in water,
MCPVP which was soluble in water, would dissolve, leaving behind a thin and weak
EC matrix. All composite EC films except for those containing 30 %w/w PV/VA, had lower puncture strengths than EC films when soaked in water for 24 h.
134 RESULTS AND DISCUSSION
E. Water vapour permeability
Ethylcellulose is barely soluble in water and its film has been shown to have low moisture permeability (Rekhi et al., 1995). This property of EC has made it useful in formulation of dosage forms consisting of moisture sensitive ingredients. In this connection, drug particles, pellets and tablets have been coated with EC to prevent moisture absorption. However, water vapour permeability of film may vary, depending on the film preparation method, drying temperature, type and amount of additives included (Sun et al., 1999) as well as the storage conditions employed (Guo et al., 1993). Conflicting findings were reported in some cases (Macleod, 1997) and hence it is of practical importance to determine if addition of PVP and its derivatives, which are hydrophilic, will affect the water vapour permeability of EC films.
EC film prepared from Surelease has a low water vapour permeability coefficient value of 1.75 g/h/cm per mmHg. The film permeability to water vapour remained low in the presence of small amount (10 %w/w) of low molecular weight
PVPs (K17 and K29). Addition of higher molecular weight PVPs (K60, K90 or
MCPVP) to EC, however caused a 30 - 70 percent rise in permeability to water vapour (Figure 29). The amount of PVP affected film permeability to different extents.
Increasing amount of low molecular weight PVPs generally increased film permeability proportionally while the effects of high molecular weight PVPs were less dependent on concentration. Except for EC-PVP K17, the water vapour permeabilities of composite Surelease films with higher concentrations of additives
(20 and 30 percent) were comparable, regardless of the molecular weight of PVP. The above observations indicated that the effect of molecular weight of PVP was more
135 RESULTS AND DISCUSSION
5
4.5
4
3.5
3
2.5
2 (g/h/cm per mmHg) per (g/h/cm
-3 1.5
1 × 10
0.5 Water Vapour Permeability Constant Permeability Vapour Constant Water 0 30% EC 20% EC-PVP EC-PVP K17 EC-PVP 10% K29 EC-PVP K60 EC- K90 EC- MCPVP Concentration 10% 20% 30% PV/VA of additives:
Figure 29. Water vapour permeability of Surelease (EC) and composite EC films (n = 3)
significant at low concentration. The effect was notably lessened for PVPs with molecular weight greater than 8, 000 when added at higher amount.
Several explanations have been proposed for change in water vapour permeability observed with inclusion of an additive (Macleod et al., 1997; Molina et al., 2003; Kester and Fennema, 1986, Greener and Fennema, 1989 and Martin-Polo et al., 1992). These include water sorption of film via formation of hydrogen bonding between water and hydrophilic region of the polymer, and alteration of film structure resulting in increased porosity (Macleod et al., 1997; Molina et al., 2003). The inclusion of PVP, which is hygroscopic, could facilitate the water sorption by EC film via formation of hydrogen bonding between water and polar domains on PVP as well as EC matrix. Increasing molecular weight and amount of PVP provided more polar sites for interaction with water molecules, leading to increased film permeability to water vapour. However, MCPVP, a cross-linked PVP K90 homopolymer, showed
136 RESULTS AND DISCUSSION comparable permeability constant as PVP K90, despite possessing higher molecular weight. Notably, the cross-linking of PVP homopolymer in MCPVP had reduced the number of sites available for interaction with water.
Similarly, addition of PV/VA of comparable molecular weight as PVP K29, decreased the permeability of film to water vapour (Figure 29). The vinyl acetate monomer which is known to reduce the hygroscopicity of PVP, is the most likely reason for this observation. Though EC-PV/VA films had relatively higher permeability than EC-PVP K17 films, the rate of increase in permeability was slightly lower for EC-PV/VA. In the practical application of EC films as a protective moisture barrier, the type or molecular weight as well as amount of PVP added should be carefully selected and controlled. In general, low molecular weight PVP, such as K17 and K29, or PV/VA at low concentration are recommended as they did not compromise on water vapour permeability of the EC films.
F. Drug permeability
Diffusion processes are central to drug delivery. They are particularly important in the release of drugs from controlled release devices, dissolution of tablets and particles and transport of drugs across biological barriers within the body. Hence, characterization of diffusion properties of a drug delivery system will provide useful information for formulation of controlled drug delivery systems. Two-chamber diffusion cells are most commonly used to determine diffusion coefficients in the study of drug transport across diffusion barrier.
Two of the most important factors that affect diffusion in water - swellable polymeric matrices are the physicochemical properties of the drugs and the polymer
137 RESULTS AND DISCUSSION membranes (Sung et al., 1995). Many theories have been proposed for solute diffusion through such membranes (Yasuda et al., 1971; Zentner et al., 1979). The
‘free volume’ theory, which has a wider acceptance, postulates that the solute diffuses within the hydrated polymer matrix through the fluctuating pores or channels inside the hydrated matrix. These free volumes are not occupied by the polymer backbone but constitue an integral part of the bulk volume of the polymer matrix. Free volume theory has proven to be applicable in describing the penetration of hydrophilic solutes in hydrogels, but less so for hydrophobic solutes which exhibit significant polymer- solute interaction. Free volume theory presupposes that solute diffusion in hydrated polymer membranes occurs in the largely aqueous free volume surrounding the fluctuating polymer chains. Some researchers have however, reported another possible diffusion pathway for hydrophobic solutes, which is through the relatively non-hydrated polymer segments (Zentner et al., 1979). Diffusion through this pathway involves the dissolution and diffusion of hydrophobic solute within the polymer domain. In situation where both mechanisms are possible, the non-hydrated polymer segment would be expected to become an important diffusion pathway if a greater concentration of solute dissolves in this region.
1. Effect of polymer additives on drug permeability
Generally, a membrane can be classified as a continuous, shunt or dispersed type (Flynn et al., 1974). The continuous type consists of a membrane that is uninterrupted between its 2 surfaces, as well as in the plane perpendicular to the flux vector. It may provide an uninterrupted diffusional path for a diffusing species or constitute an impervious supporting structure, depending on its composition. The shunt type consists of a membrane that is laterally discontinuous due to the presence
138 RESULTS AND DISCUSSION of pores or channels. It can be impermeable to the permeant or provide parallel diffusional pathways, depending on its composition. The dispersed type consists of one polymer phase embedded in a continuous or shunt phase. It is discontinuous along the flux vector and does not provide an uninterrupted pathway through the membrane or through any of its subphases. The mechanism of drug release from membrane- controlled drug delivery systems depends on the composition and type of membrane
(Ozturk et al., 1990; Narisawa et al., 1994). Presence of plasticizers or/and water- soluble additives can modify the pathway of drug diffusion by providing alternative routes of diffusion through the shunt rather than the continuous phase.
Ethylcellulose films cast from Surelease, with an average thickness of 200 µm is practically impermeable to all the model drugs used in this study, except for sulphanilamide (Tables 16 and 17). The presence of plasticizer, medium chain triglycerides, did not increase the permeability of Surelease film. This implied that any primary shunts (pores) present were simply too small for effective diffusion of drugs. A typical permeation profile of composite Surelease films is shown in Figure
30.
Generally, the drug permeability profile consisted of two stages – an initial slow release usually up to 120 min followed by a linear trend after 180 min. The permeability coefficient was calculated from the diffusion profile at 300 min onwards
(Table 16). The similar permeability constants and correlation coefficient calculated from both the steady state and quasi-steady model suggested that sink condition and constant drug concentration gradient were maintained throughout the study. The type and concentration of additive affected the permeability of Surelease films to chlorpheniramine to different extents. Low molecular weight additives PVP, such as
PVP K17, did not cause significant increase in permeability even when presented in
139 RESULTS AND DISCUSSION high concentration of 30 %w/w. In contrast, high molecular weight MCPVP caused a significant increase in permeability at concentration as low as 10 %w/w. The varying
Table 16. Permeability and correlation coefficients of Surelease (EC) and composite EC films to chlorpheniramine (n = 3).
2 2 Films Drug permeability Drug permeability r1 r2 -10 -10 coefficient1 × 10 coefficient 2 × 10 (cm2/s) (cm2/s) EC - - - -
EC-PV/VA (9:1) - - - -
EC-PV/VA (8:2) - - - -
EC-PV/VA (7:3) 396 ± 27 387 ± 26 0.980 0.979
EC-PVP K29 (9:1) - - - -
EC-PVP K29 (8:2) - - - -
EC-PVP K29 (7:3) 860 ± 58 848 ± 60 0.983 0.980
EC-PVP K90 (9:1) 38.7 ± 5.9 37.8 ± 5.9 0.988 0.987
EC-PVP K90 (8:2) 849 ± 93 964 ± 32 0.930 0.967
EC-PVP K90 (7:3) 2680 ± 250 2700 ± 240 0.991 0.991
EC-MCPVP (9:1) 186 ± 16 181 ± 16 0.995 0.994
EC-MCPVP (8:2) 1380 ± 150 944 ± 150 0.994 0.993
EC-MCPVP (7:3) 3660 ± 220 3720 ± 240 0.995 0.994
EC-PVP K60 (7:3) 1560 ± 90 1368 ± 141 0.980 0.979
1 calculated from equation 42 2 calculated from equation 44
140 RESULTS AND DISCUSSION
Table 17. Drug permeability coefficients of Surelease (EC) and composite EC films (n=3)
Drugs Drug permeability coefficient × 10-10 (cm2/s)
EC EC-PV/VA EC-PVP EC-PVP EC-PVP K29 K60 K90
Propranolol - 0.796 ± 0.015 66.4 ± 2.9 86.2 ± 6.4 1030 ± 150
Sulphanilamide 19.0 ± 0.55 221 ± 23 202 ± 22 1240 ± 480 4270 ± 280
Sulphaguanidine - 35.0 ± 3.6 51.4 ± 4.2 474 ± 140 4210 ± 220
Theophylline - 29.6 ± 2.5 33.1 ± 2.2 433 ± 110 570 ± 48
Metformin - 68.5 ± 1.5 74.4 ± 6.6 3570 ± 670 5350 ± 110
Verapamil - 220 ± 8.1 237 ± 14 303 ± 32 1040 ± 6.2
Chlorpheniramine - 390 ± 27 794 ± 10 1560 ± 90 2430 ± 200
0.8 EC-PV/VA (7:3) EC-PVP K29 (7:3) EC-PVP K90 (7:3) EC-MCPVP (7:3) EC-PVP K60 (7:3) 0.7
0.6
0.5
0.4
0.3
0.2
0.1
Cumulative amount of chlorpheniramine (g) 0 0 100 200 300 400 500 600 700
Time (min)
Figure 30. Permeation profiles of composite Surelease (EC) films to chlorpheniramine.
141 RESULTS AND DISCUSSION
chlorpheniramine permeability of composite EC films containing PVP K29, K60 and
K90 implied that the molecular weight of PVP played an important role in diffusion of drug through the composite EC films (Table 16). At 30 %w/w level, a linear relationship between the permeability of composite EC films containing PVP or
MCPVP and molecular weight of the additives was obtained (Figure 31). Statistical
4500 /s)
2 4000
3500
3000
2500
2000
1500
1000
500 Permeability coefficient (cm coefficient Permeability 0 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 Molecular weight of PVP
Figure 31. Relationship between permeability coefficient and molecular weight of PVP.
analysis showed that this relationship is significant at p < 0.01, with a Pearson correlation coefficient of 0.988.
At 30 %w/w of additives, the lowest permeability coefficient was obtained for
PV/VA, followed by PVP K29, PVP K60, PVP K90 and MCPVP. Interestingly, the drug permeability of EC-PV/VA film is less than half the EC-PVP K29 film, even though their molecular weights are similar. In their study, Donbrow and Friedman
142 RESULTS AND DISCUSSION
(1975) proposed that hydrophilic polymers could increase the permeability of hydrophobic films by the following mechanisms:
• by altering the polymer configuration, proportion of crystalline to amorphous
regions and hydrophilicity of the film;
• by increasing film porosity;
• by forming capillaries or a hydrated network giving direct connection between
the two sides of the film; and
• by interactions with penetrants to form complexes that have increased
membrane solubility or diffusion coefficient.
Based on the findings of other researchers, swelling of PVP and formation of water- filled pores in insoluble EC matrix would probably enhance drug release (Donbrow and Friedman 1975; Rowe et al., 1986). It was not certain if all PVP present in EC film would completely dissolve in aqueous medium, thereby creating channels or pores for more rapid drug diffusion. Alternatively, some of PVP might be partly retained in the film matrix, resulting in greater water uptake by the film. Swelling test was carried out to investigate the extent and effect of swelling of different composite
EC films on drug permeability. The extractable amounts (EXC) from composite EC films over time were calculated to give an indication of any change in porosity as the films came into contact with the dissolution medium. The ability of film matrix in absorbing water was demonstrated by its average swelling index.
A sharp rise in EXC values was observed for EC and composite EC films within 5 min (Figure 32a). The EXC values leveled off at around 9.0 percent after 5 min for EC film. The extractable amount for EC film formed from Surelease was swelling profile of EC-PV/VA films closely resembled that of EC film with a slightly higher rate of swelling in the first hour. This implies that addition of PV/VA to EC
143 RESULTS AND DISCUSSION film did not result in significant change in the amount water it could absorb. Addition of PV/VA however resulted in increased permeability of EC film to all the model drugs studied (Table 17). This increase in drug permeability was attributed to the more porous film arising from dissolution of PV/VA. Light microscope images of wetted composite EC-PV/VA film showed irregular small pore-like structures (Figure
33a). Increase in porosity, which is equivalent to decrease in effective thickness was found to be the mechanism behind the increased in the permeability of EC films containing polyethylene glycol (Donbrow et al., 1975).
The EXC value for EC-PVP K29 films increased to a maximum of 15.6 percent after 20 min while those of EC-PVP K60 and EC-PVP K90 rose more gradually to a maximum of 24.6 percent and 26.2 percent respectively at 60 min. On the other hand, EC-PVP films, in particular those containing higher molecular weight
PVP had much greater swelling than EC or EC-PV/VA films (Figure 32b). The high
EXC and swelling index values indicated that the permeability property of composite
EC-PVP film were affected by both the dissolution of PVP, creating channels or pores in the EC matrix where the drug could diffuse through, as well as, swelling of film matrix which was observed. Increase in porosity, which is equivalent to a decrease in show large patches enhanced by the presence of PVP remaining in the film. When viewed under light microscope, irregular pore - like structures were also observed for
EC-PVP K29 films, as in the case for EC-PV/VA films (Figure 33a and b). However, unlike EC-PV/VA films, EC-PVP K60 and EC-PVPK90 films showed large patches of irregular raised whitish layer (Figures 33c and d). The undefined large whitish patches seemed to be suggestive of swollen film matrix.
In a study of permeation of drug through untreated poly (methylvinylether)- maleic anhydride (PVM-MA) film, Fites et al. (1970) reported rapid increase in
144 RESULTS AND DISCUSSION
(a)
35 EC EC-PVP K29 EC-PVP K90 EC-PVP K60 EC-PV/VA
30
25
20
15
10
Extractable amount (%) amount Extractable 5
0 0 50 100 150 200 250 300 Time (min)
(b)
EC-PVP K29 EC-PVP K60 EC-PVP K 90 EC-PV/VA EC 180
160
140
120
100 (%) s
I 80
60
40
20
0 0 50 100 150 200 250 300 Time (min)
Figure 32. (a) Amount of extracted component and (b) swelling index of EC and composite EC films containing 30 %w/w of additives.
145 RESULTS AND DISCUSSION
(a) (b)
(c) (d)
Figure 33. Light microscope images of wetted composite EC films: (a) EC-PV/VA (7:3), (b) EC-PVP K29 (7:3), (c) EC-PVP K60 (7:3), (d) EC-PVP K90 (7:3)
permeability after 2 h. This rise in permeability was attributed to hydration of acid anhydride groups resulting in swelling of the film. In this study, the swelling index at
300 min for EC films containing 30 %w/w of EC-PVP K29, PVP K60 and PVP K90 films were 63.8, 90.3 and 117.8 percent respectively. This suggests that increased molecular weight of PVP resulted in greater uptake of water. This could be explained by the larger hydrophilic sites available for hydrogen bonding with water molecules.
While EC-PVP films showed proportional increase in the swelling index with increasing molecular weight of PVP added, the EXC value of EC-PVP films failed to show the same trend. It is expected that the solubility of PVP would be similar regardless of molecular weight and the difference between EXC values of EC-PVP
146 RESULTS AND DISCUSSION
K60 or K90 and EC-PVP K29 might be due to the greater effect which the molecular weight of PVP K60 or K 90 had exerted on the proportion of EC films. The pore size created by dissolution of PVP present in the film probably varied depending on the molecular weight. It might be expected that the channels created by dissolution of
PVP in the EC film matrix would be larger for higher molecular weight PVP. This possibility, coupled with the greater water uptake by PVP of higher molecular weight, might explain the greater drug permeability coefficients obtained for composite EC film consisting of high molecular weight PVP.
2. Effect of drug properties on drug permeability
The molecular weight, aqueous solubility and basicity of the drugs are given in Table 4. Chlorpheniramine maelate, propranolol HCl, sulphanilamide and sulphaguanidine constitute a series of basic drugs with comparable molecular weights but different solubilities in water whereas metformin, propranolol HCl, chlorpheniramine maelate and verapamil HCl have comparable solubilities in water (≤
0.05) but different molecular weights. Using these drugs, the influence of solubility and molecular weight of drugs on the drug permeability of the films was investigated.
With the exception of EC-PVP K29 and EC-PVP K90, all the composite EC films showed decreasing permeability to the following drugs: chlorpheniramine > sulphanilamide > sulphaguanidine > propranolol (Table 17). The swelling test results suggested that addition of PVP and derivatives might enhance the permeability of EC films through formation of water-filled channels and swelling of matrix.
Consequently, greater release was obtained for water-soluble drugs as they could dissolve or diffuse through the aqueous swollen layer better than drugs of lower solubility (Jenquin et al., 1990). It was noted that the relative solubility of
147 RESULTS AND DISCUSSION chlorpheniramine, sulphanilamide and sulphaguanidine and the corresponding drug permeability of films followed the same trend. However, despite being more soluble than sulphanilamide and sulphaguanidine, propanalol had a lower permeability constant. In their studies, Donbrow and Friedman (1975) found that EC-PEG film was more permeable to caffeine than benzoic acid or salicylic acid. The lower permeability was attributed to greater affinity of these organic acids with the EC-PEG matrix. In this study, the lower permeability constant for propranolol was probably due to interaction between propranolol and EC film matrix, thereby impeding drug diffusion.
Both sulphanilamide and sulphaguanidine showed higher permeability constants than chlorpheniramine and propranolol for EC-PVP K90 films. No significant difference between film permeability to propranolol and sulphaguanidine was observed for EC-PVP K29, but both EC-PVP K60 and K90 films were at least four to five times more permeable to sulphaguanidine than propranolol. Among the four model drugs studied, chlorpheniramine is known to have the highest aqueous solubility, followed by propranolol, sulphanilamide and sulphaguanidine.
On the other hand, sulphanilamide and sulphaguanidine are of the same chemical series (i.e. sulphonyl drugs) and whatever drug-polymer interaction present would be similar. The difference in permeability of composite EC films to these 2 drugs would most probably be affected by their solubility. It should also be noted that sulphaguanidine has a higher molecular weight (232.3) than sulphanilamide (172.2).
The higher molecular weight of sulphaguanidine might have also contributed to its lower permeability through composite EC films. This effect was more obvious for
EC-PV/VA film where the main mechanism of drug permeation is diffusion through pores. It was be easier for sulphanilamide which is smaller in molecular size to diffuse
148 RESULTS AND DISCUSSION through the pores than sulphaguanidine. Similarly, permeability of EC-PV/VA film to propranolol was markedly lower compared to EC-PVP films. This implied that the diameters of the pores or capillaries present in EC-PV/VA films were relatively small and impeded the passage of propranolol.
The effect of molecular weight of drug on permeability of composite EC films was also studied using metformin, propranolol, chlorpheniramine and verapamil. With the exception of propranolol, the drug permeability constants decreased for all composite films as the molecular weight of the drug increased. A linear relationship between molecular weights of metformin, chlorpheniramine and verapamil and their respective permeability constants was observed for EC-PVP K60 and EC-PVP K90 films. In contrast, the permeability of EC-PVP K29 and EC-PV/VA films to metformin was much lower than those for chlorpheniramine and verapamil. The lower than expected permeability could be attributed to interaction between EC and metformin. PVP K60 and K90 appeared to be able to reduce this interaction. Hence, in the presence of high molecular weight PVP, drug permeability of composite EC films could be related to the molecular weight of drugs. Drugs of higher molecular weights and therefore larger hydrodynamic size will likely diffuse more slowly through the swollen composite EC film matrix. In this respect, metformin, with the smallest molecular weight took the least amount of time to diffuse through the swollen composite EC films.
G. Correlation between physicomechanical properties and permeability of composite EC films
Several studies have reported a correlation between the changes in physicomechanical properties of film and their permeability (Aulton et al., 1981;
149 RESULTS AND DISCUSSION
Benita et al., 1986; Van Bommet et al., 1989; Jenquin et al., 1992). However, the findings were not always consistent. For instance, Jenquin et al. (1992) found that incorporation of PVP K15 and PVP K90 into acrylic resin films only resulted in a small change in the water vapour permeability, tensile strength and glass transition temperature of the matrix. In contrast, Eudragit RS PM films containing chlorpheniramine showed a reduction in drug release with addition of PVP. This decrease in drug release corresponded to increased glass transition temperature and tensile strength but not change in water vapour permeability. On the other hand, films composing of chlorpheniramine and Eudragit RL PM experienced slightly decreased water vapour permeability, a small upward shift in glass transition temperature and significantly increased tensile strength with the addition of PVP K15 or PVP K90.
However, these changes in physicomechanical and physicochemical properties had decreased release rates and correlated with the changes in water vapour or drug permeabilities.
This study has shown that addition of PVP increased the glass transition temperature, tensile properties, puncture strength, as well as water vapour and drug permeabilities of composite Surelease films. The magnitude of change in these properties was not consistent in all cases. For instance, the percent increases in glass transition temperature and tensile strength were relatively small (~ 27 percent), while greater increments (76 – 80 percent) were observed for the elastic modulus and puncture strength. Presence of PVP further accentuated the permeability of film to water vapor (80 - 177 percent) and increased drug peremeability by 1000 times or more in some cases. Moreover, while a linear relationship was found between the molecular weight of PVPs and permeability to chlorpheniramine, the thermal and
150 RESULTS AND DISCUSSION mechanical properties of composite Surelease films did not show similar response to increasing molecular weight of PVPs.
Generally, it was thought that increase in glass transition temperature and tensile strength of polymer films corresponded to hardening of the films and therefore reduction in drug release rate (Benita et al., 1986; Van Bommet et al., 1989; Jenquin et al., 1992). Though addition of PVP to Surelease resulted in increase in glass transition temperature, tensile strength, elastic modulus and puncture strength of
Surelease films, these changes did not correlate with a reduction in permeabilities as suggested by other studies. This anomaly could be due to properties of additives present. PVP is a highly water-soluble and hygroscopic polymer which may increase the drug permeability through formation of pores for rapid diffusion. Similar trends were observed for changes in physicomechanical and properties and release rates of composite ethylcellulose films, regardless of the properties of drugs used. This suggested that the permeability of composite Surelease films was predominantly affected by the properties of the adjuvants (Dittgen et al., 1997).
H. Release kinetics of pellets coated with EC and polymeric additives
In this study, the influences of coating level, drug properties as well as film curing on the release kinetic of Surelease coated pellets were investigated. The effects of molecular weight and chemical nature of PVP and its derivatives on in vitro release properties of Surelease coated pellets were also then determined.
1. Effect of coating levels on drug release from Surelease - coated pellets
As seen from Figure 34, drug release decreased as the level of coating increased. In particular, the release profiles appeared to be more linear at 10 - 12 percents coating
151 RESULTS AND DISCUSSION levels. The dissolution data for 5 – 70 percents drug release were also fitted into various mathematical models (Table 18). The highest r2 values were obtained for the
Higuchi or Baker-Lonsdale models. The AIC values however distinguished Baker-
Lonsdale model as the best fit model for all coating levels. The relationship between the Baker-Lonsdale release rate constant of Surelease coated pellets and the level of
Surelease was illustrated in Figure 35a. As the coating level increased from 4 to 8
%w/w, the release rate constant decreased linearly from ~0.008 to ~0.001. While it not clearly shown in the graph, another significant rate of change in the release rate constant was seen for pellets coated with 8 and 10 %w/w EC. Thereafter, the rate of decrease became less marked despite further increase in coating thickness. Thus a minimum of 10 %w/w coating level is necessary to achieve stable and uniform controlled release of chlorpheniramine. Unlike, the change observed for release rate constant, the T50% and MDT50% profiles showed greater increase at higher coating levels of 8 – 12 %w/w (Figures 35b and c).
152 RESULTS AND DISCUSSION
4%, cured 6%, cured 8%, cured 10%, cured 12%, cured 4%, uncured 6%, uncured 8%, uncured 10%, uncured 12%, uncured 100
80
60
40
20
Chlorpheniramine released (%) released Chlorpheniramine 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
Figure 34. Dissolution profiles of pellets coated with different levels of Surelease.
Table 18. Fit factors of drug release models for chlorpheniramine pellets coated with different level of EC.
Release models Coating levels 4 % 6 % 8 % 10 % 12 % r2 Zero order 0.951 0.935 0.898 0.971 0.977 First order 0.775 0.774 0.712 0.838 0.864 Hixson-Crowell 0.851 0.841 0.788 0.897 0.915 Higuchi 0.984 0.970 0.979 0.994 0.994 Baker-Lonsdale 0.987 0.981 0.984 0.984 0.981 AIC Zero order -41.5 -39.5 -35.3 -45.3 -42.8 First order 5.6 6.4 1.9 -1.0 2.1 Hixson-Crowell -35.6 -34.6 -34.8 -39.4 -36.6 Higuchi -54.3 -50.3 -52.0 -62.7 -60.5 Baker-Lonsdale -81.9 -91.8 -107.1 -86.5 -90.3
153 RESULTS AND DISCUSSION
(a) 0.014
0.012
0.010
0.008
0.006
0.004 Release rate constant
0.002
0.000 4 6 8 10 12 Coating level (%w/w)
(b) 1600
1400
1200
1000
(min) 800 50% T 600
400
200
0 4681012 Coating level (%w/w)
154 RESULTS AND DISCUSSION
(c) 450
400
350
300
(min) 250 50% 50% 200
MDT 150
100
50
0 4681012 Coating level (% w/w)
Figure 35. Effect of coating level of Surelease on (a) Baker-Lonsdale release rate constant (b) T50% (c) MDT50% of chlorpheniramine, theophylline and paracetamol pellets. CPM – cured; CPM – uncured; THE - cured; THE - uncured; PA – cured; PA – uncured.
4%, cured 6%, cured 8%, cured 10%, cured 12%, cured 4%, uncured 6%, uncured 8%, uncured 10%, uncured 12%, uncured 100
80
60
40
20 Theophylline released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
Figure 36. Dissolution profiles of theophylline pellets coated with different levels of EC cured at 60 °C for 24 h or at 22 °C for 48 h.
155 RESULTS AND DISCUSSION
100 4%, cured 6%, cured 8%, cured 10%, cured 12%, cured 4%, uncured 6%, uncured 8%, uncured 10%, uncured 12%, uncured
80
60
40
20 Paracetamol released (%)
0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
Figure 37. Dissolution profiles of paracetamol pellets coated with different levels of Surelease cured at 60 °C for 24 h or at 22 °C for 48 h.
The effect of Surelease coating levels on two other drugs with low solubility
(theophylline and paracetamol) was also studied. As in the case for chlorpherinamine pellets, the dissolution profiles became increasingly linear as the coating thickness increased (Figures 36 – 37). Comparison of r2 values showed that theophylline pellets coated with 4 %w/w EC was best described by Baker Lonsdale model while those coated with 6 - 12 %w/w followed zero-order release (Table 19). Similarly, the r2 values indicated that paracetamol pellets coated with 4 - 6 %w/w EC was best described by Baker – Lonsdale or Higuchi square root time models while those coated with 8 - 12 %w/w followed zero-order release model (Table 20). On the other hand, based on the AIC values, release of theophllyine and paracetamols from EC coated pellets showed best fitting with Baker-Lonsdale model.
156 RESULTS AND DISCUSSION
Four possible mechanisms have been proposed to explain drug release from coated pellets (Zhuang et al., 1991):
a. transport of drug through cracks and other imperfections within the
matrix or uncoated system;
b. transport of drug through a network of capillaries (or pores) in the
coating applicable if the water-soluble components of the film leach
out;
c. transport of drug through a hydrated swollen film – applicable if the
water-soluble components are retained within the coating; and/or
d. transport of drug through the non-porous coating, which is determined
by the permeability of the drug in the film.
During film coating, the pellets were continuously layered with coating materials. Consequently, the film formed was composed of overlapping segments.
Any holes from the overlapping layers were gradually covered as more layers of film
Table 19. Fit factors of drug release models for theophylline pellets coated with different levels of Surelease.
Release models Coating levels 4 % 6 % 8 % 10 % 12 % r2 Zero order 0.888 0.995 0.999 0.997 0.995 First order 0.660 0.880 0.947 0.953 0.959 Hixson-Crowell 0.748 0.937 0.977 0.982 0.981 Higuchi 0.974 0.988 0.983 0.973 0.989 Baker-Lonsdale 0.995 0.944 0.946 0.929 0.949 AIC Zero order -25.0 -79.4 -90.9 -78.6 -83.5 First order 2.4 -0.5 -16.8 -16.3 -22.4 Hixson-Crowell -22.9 -50.9 -58.4 -59.8 -66.4 Higuchi -36.1 -68.1 -60.5 -58.1 -74.6 Baker-Lonsdale -73.7 -96.3 -94.6 -98.6 -114.0
157 RESULTS AND DISCUSSION
Table 20. Fit factors of drug release models for paracetamol pellets coated with different levels of EC.
Release models Coating levels 4 % 6 % 8 % 10 % 12 % r2 Zero order 0.894 0.988 0.997 0.989 0.989 First order 0.712 0.848 0.897 0.858 0.855 Hixson-Crowell 0.786 0.915 0.949 0.925 0.924 Higuchi 0.960 0.998 0.986 0.988 0.986 Baker-Lonsdale 0.980 0.966 0.938 0.954 0.953 AIC Zero order -21.2 -47.1 -62.2 -55.8 -59.6 First order 0.4 1.8 0.5 0.4 1.8 Hixson-Crowell -23.0 -33.5 -38.8 -41.1 -43.8 Higuchi -28.6 -64.0 -47.6 -54.4 -57.4 Baker-Lonsdale -59.6 -69.7 -66.5 -74.2 -80.8
were deposited. At low coating level of 4 %w/w, the drug was probably released through the pores or channels in the coating. As the coating thickness increased, the membrane became more complete and had less pores. Thus, the rate of release was drastically reduced. At higher coating level, the drug had to take the pathway involving diffusion through the barrier, which could be related to drug permeability in the coating and coating thickness. Hence, release of theophylline and paracetamol from pellets with high level of coating follow zero-order model (Tables 19 - 20).
However, release of chlorpheniramine pellets from the coated pellets followed Baker-
Lonsdale model despite changes in coating level. The change in release mechanism was also reflected in the drastic drop in release rate constants for theophylline and paracetamol pellets coated with 6 %w/w EC compared to 4 %w/w EC. Release constants of theophylline and paracetamol pellets coated with 6 - 12 %w/w EC showed relatively smaller change.
158 RESULTS AND DISCUSSION
The MDT50% of theophylline pellets increased as the coating levels increased from 4 - 8 %w/w and plateau at higher coating levels (Figure 35c). On the other hand, the MDT50% values of paracetamol pellets continued to increase as the coating level increased from 8 – 10 %w/w before it plateaued at 10 - 12 %w/w. The markedly lower MDT50% values for paracetamol pellets indicated that a thicker coat was required to control the release of paracetamol compared to theophylline. Unlike
MDT50%, the T50% values for both theophylline and paracetamol pellets showed almost linear increase with coating level (Figure 35b).
Analysis of the release rate constants, T50% and MDT50% demonstrated that drugs with greater solubility have faster release rate. Chlorpheniramine had the fastest release, followed by paracetamol and theophylline. When release rates were plotted against the solubilities of chlorpheniramine (50 mg/ml), paracetamol (1.43 mg/ml) and theophylline (0.833 mg/ml), notable effect was seen at low coating levels of 4 - 6
%w/w (Figure 38a). At higher coating levels, the differences in release rate constant became less significant. Plots of T50% and MDT50% against solubility showed sharp decrease as the solubility increased from 0.833 to 1.43 mg/ml at all coating levels except for 4 %w/w (Figures 38b and c). This observation could be explained by considering the mechanism of release at various coating levels. As discussed earlier, drug release occurred primarily through the pores or cracks that were present
159 RESULTS AND DISCUSSION
(a) 0.014
0.012
0.01
0.008
0.006
0.004 Release rate constant
0.002
0 0 0.05 0.1 0.15 0.2 0.25 Drug solubility (g/ml)
(b) 450
400
350
300
(min) 250 50% 200
MDT 150
100
50
0 0 0.05 0.1 0.15 0.2 0.25 Drug solubility (g/ml)
160 RESULTS AND DISCUSSION
(c) 1600
1400
1200
1000
(min) 800 50%
T 600
400
200
0 0 0.05 0.1 0.15 0.2 0.25 Drug solubility (g/ml)
Figure 38. Relationship between (a) Baker-Lonsadale release rate constant (b) MDT50% (c) T50% and drug solubility. 4% 6% 8% 10% 12%w/w coating level
in the incomplete membrane at low coating levels. Since drugs with higher solubility
would have greater driving power through the medium filled pores than drugs with
lower solubility, effect of drug solubility on release rate is heightened at low coating
levels. At higher coating levels, the film coat was better formed and release rate was
limited by the partitioning of drugs through the EC coat rather than their aqueous
solubilities. In addition, it was noted that the increase in T50% and MDT50% was more
significant at lower drug solubility range of 0.833 to 1.43 mg/ml.
2. Effect of curing on EC - coated pellets
The dissolution profiles of EC coated chlorpheniramine pellets that were cured
at 60 °C for 24 h were compared to those that were stored at ambient temperature of
22 °C, 48 h (Figure 34). Curing of chlorpheniramine pellets at 60 °C resulted in
161 RESULTS AND DISCUSSION slower release at all levels. However, the difference was not significantly displayed by the release rate constants, T50% and MDT50% values (Figures 35b and c). Independent t-test analysis showed that only chlorpheniramine pellets coated with 8 %w/w EC showed significant difference in release rate constant and MDT50% value when cured compared to uncured. In contrast, the effect of curing was more significant on theophylline and paracetamol pellets, especially those coated with low levels of EC.
The effect appeared to be most marked on the MDT50% of theophylline pellets coated with 4 - 8 %w/w EC. Paracetamol pellets coated with 4 - 6 %w/w EC have significant difference in their MDT50% values. However, it was noted that the effect of curing on
MDT50% diminished as the coating level increased to 10 %w/w. The f1 and f2 values computed from the dissolution profiles of chlorpheniramine, paracetamol and theophylline pellets further supported the above observations (Table 21). More than
10 % differences were obtained for the dissolution profiles of chlorpheniramine pellets with coating levels of 8 - 12 %w/w, cured at 60 °C for 24 h compared to those uncured. In contrast, only the dissolution profiles of paracetamol and theophylline pellets coated with low level of EC, 4 - 6 and 4 – 8 %w/w respectively were found to be dissimilar.
Table 21. Difference factors, f1, and similarity factors, f2, for chlorpheniramine (CPM), paracetamol (PA) and theophylline (THE) pellets coated with different levels of EC.
Coating levels Average f1 Average f2 (%w/w) CPM PA THE CPM PA THE 4 15.3 *39.1 *58.5 50.4 *34.5 *21.2 6 8.6 *27.7 *83.9 64.5 *40.6 *31.9 8 *29.1 6.5 *58.0 *39.4 75.7 50.5 10 *38.3 3.7 8.0 *45.0 88.2 90.6 12 *18.5 4.0 10.9 62.9 88.9 87.5 *significant at p < 0.01
162 RESULTS AND DISCUSSION
It has been reported that during curing, the pseudolatex particles would coalescence and fuse further to form a more complete film. The effect of curing would be more marked at low coating level since the polymer coats were incomplete.
This explained the reduced drug release, especially for pellets coated with low level of coating. At higher coating levels, the polymer coat would be more complete with less pores or flaws. Further coalescence during curing would not have much influence on drug release since the polymer coat was already well - formed. The effect of curing was found to vary with the type of drugs used. This could be due to the solubility of the drugs in the dissolution medium. Since chlorpheniramine has very high solubility in aqueous medium, curing pellets coated with low level of EC might not contribute significant effect on retarding its release. This was because at low coating level, drug release mainly occurred through the pores or channels which were more abundant in the coating. Curing of pellets coated with low coating level of EC aided in the fusion of some of pores or cracks present, thereby retarding release of less water - soluble drugs, such as paracetamol and theophylline, significantly. As the coating thickness increased, the effect of curing became less retarding for the less water - soluble drugs.
This suggested that at high coating levels, the film coat was thick enough such that the drug release was controlled by diffusion of the drug through the membrane coat.
Hence, curing which promoted further coalscenece and strengthened the film but not its thickness did not exert significant influence in such cases. On the other hand, curing of chlorpheniramine pellets coated with higher levels of EC showed significant reduction in drug release (Table 21). This indicated that thicker film coat was required to retard diffusion of chlorpheniramine. If the film coat was not sufficiently thick, curing alone would be insufficient in retarding the release of chlorpheniramine.
163 RESULTS AND DISCUSSION
3. Influence of PVP on EC - coated pellets
PVP, a water-soluble additive, was also added to Surelease dispersion at
concentrations ranging from 10 - 30 %w/w of total solid content. The resultant
dispersions were used to coat chlorpheniramine and theophylline pellets. The
influence of additives on the dissolution profiles of chlorpheniramine and
theophylline pellets are shown in Figures 39 - 40. Addition of PVPs at a relatively low
concentration of 10 %w/w resulted in great increase in rate of release and total
amount release. Figure 39 shows that the maximum release was achieved in about 120
min for all composite Sureleasecoated chlorpheniramine pellets coated with 10 %w/w
The dissolution data was also fitted into zero-order, first-order, Hixson-Crowell,
PVPs, except for PVP K17. Chlorpheniramine pellets coated with EC-PVP K17 (9:1)
achieved maximum release after 300 min. However, as the concentration of PVP
(a) EC EC-PVP K17 (9:1) EC-PVP K29 (9:1) EC-PVP K60 (9:1) EC-PVP K90 (9:1) EC-MCPVP (9:1) EC-PV/VA (9:1) 100
80
60
40
20 Chlorpheniramine released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
164 RESULTS AND DISCUSSION
(b) EC EC-PVP K17 (8:2) EC-PVP K29 (8:2) EC-PVP K60 (8:2) EC-PVP K90 (8:2) EC-MCPVP (8:2) EC-PV/VA (8:2) 100
80
60
40
20
Chlorpheniramine released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
(c)
EC EC-PVP K17 (7:3) EC-PVP K29 (7:3) EC-PVP K60 (7:3) EC-PVP K90 (7:3) EC-MCPVP (7:3) EC-PV/VA (7:3) 100
80
60
40
20
Chlorpheniramine released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
Figure 39. Dissolution profiles of chlorpheniramine pellets with Surelease (EC) and composite EC coats containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of PVP K17, K29, K60, K90, MCPVP and PV/VA cured at 60 °C, 24 h.
165 RESULTS AND DISCUSSION
(a) EC EC-PVP K17 (9:1) EC-PV/VA (9:1) EC-PVP K29 (9:1) EC-PVP K60 (9:1) EC-PVP K90 (9:1) EC-MCPVP (9:1) 100
80
60
40
20 Theophylline released (%)
0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
(b) EC EC-PVP K17 (8:2) EC-PVP K29 (8:2) EC-PVP K60 (8:2) EC-PV/VA (8:2) EC-PVP K90 (8:2) EC-MCPVP (8:2) 120
100
80
60
40
20 Theophylline released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
166 RESULTS AND DISCUSSION
(c) 120 EC-PVP K17 (7:3) EC-PVP K60 (7:3) EC-PVP K90 (7:3) EC-MCPVP (7:3) EC-PV/VA (7:3) EC-PVP K29 (7:3) EC 100
80
60
40
20 Theophylline released (%)
0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
Figure 40. Dissolution profiles of theophylline pellets with Surelease (EC) and composite EC coats containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of PVP K17, K29, K60, K90, MCPVP and PV/VA cured at 60 °C, 24 h.
added to Surelease coat increased, the time taken to achieve maximum release was
reduced to about 120 min for all coating formulations (Figures 39a - c). This
phenoneom was in great contrast to that observed for pellets coated with plasticized
EC (no additives) which produced an average maximum of 64 percent release even
after 12 h.
The dissolution data was also fitted into zero-order, first-order, Hixson-
Crowell, Higuchi square root time, Hopfenberg and Baker-Lonsdale equations based
on linear regression. The r2 and AIC values of the various models were showed in
Tables 22 - 23. Earlier it was found that chlorpheniramine pellets coated with
Surelease could be best described by Baker-Lonsdale model or Higuchi model
depending on the coating levels. With the presence of PVPs, majority of the coating
167 RESULTS AND DISCUSSION formula lead to a change in release model to Hopfenberg model. Similar change in mechanism of release was also indicated for theophylline pellets as inferred from the r2 and AIC values showed in Tables 24 -25. While the release of theophylline from pellets coated with 6 - 12 %w/w Surelease was characterized by zero-order release, addition of PVPs generally resulted in a shift in release model to Hopfenberg model.
Since Hopfenberg model is used to describe erosion-controlled release, these changes in model indicated a change ofmechanism from coated matrix diffusion controlled to erosion or swelling controlled release. As discussed above, release of drug from
Table 22. r2 values of chlorpheniramine pellets with composite Surelease (EC) coat cured at 60 °C, 24 h.
% of polymeric Zero First Hixson- Higuchi Hopfen- Baker- additives order order Crowell berg Lonsdale PVP K17 10 0.989 0.834 0.909 0.994 0.997 0.950 20 0.983 0.885 0.930 0.995 0.997 0.973 30 0.985 0.861 0.921 0.995 0.995 0.963 PVP K29 10 0.984 0.863 0.919 0.996 0.998 0.971 20 0.957 0.827 0.883 0.980 0.972 0.973 30 0.973 0.884 0.924 0.989 0.987 0.970 PVP K60 10 0.997 0.946 0.977 0.989 0.989 0.927 20 0.984 0.883 0.930 0.995 0.997 0.972 30 0.976 0.815 0.888 0.995 0.996 0.970 PVP K90 10 0.997 0.900 0.953 0.983 0.994 0.927 20 0.975 0.821 0.891 0.994 0.995 0.974 30 0.915 0.763 0.825 0.970 0.958 0.985 MCPVP 10 0.981 0.853 0.912 0.992 0.996 0.959 20 0.966 0.831 0.893 0.985 0.987 0.967 30 0.982 0.910 0.922 0.988 0.995 0.964 PV/VA 10 0.990 0.885 0.935 0.998 0.886 0.980 20 0.973 0.884 0.924 0.989 0.987 0.970 30 0.957 0.827 0.883 0.980 0.972 0.973
168 RESULTS AND DISCUSSION
Table 23. AIC values of chlorpheniramine pellets coated with composite Surelease (EC) coat cured at 60 °C, 24 h.
% of polymeric Zero First Hixson- Higuchi Hopfen- Baker- additives order order Crowell berg Lonsdale PVP K17 10 -50.0 0.5 -35.0 -59.0 -79.1 -67.3 20 -35.7 -6.2 -30.8 -43.8 -57.2 -53.8 30 -42.2 -2.6 -32.3 -49.6 -52.6 -61.5 PVP K29 10 -35.8 -3.6 -24.6 -46.1 -61.0 -53.3 20 -44.1 7.3 -36.7 -60.7 -77.1 -101.2 30 -47.7 1.2 -41.0 -51.0 -72.9 -83.6 PVP K60 10 -36.1 -8.9 -28.7 -29.0 -37.0 -36.7 20 -41.0 -6.5 -34.8 -51.4 -66.8 -61.2 30 -53.6 3.2 -41.8 -73.1 -93.9 -91.4 PVP K90 10 -92.8 0.9 -58.8 -68.1 -108.3 -99.1 20 -54.6 3.4 -43.6 -73.7 -94.4 -96.4 30 -38.5 1.3 -39.4 -51.6 -64.4 -97.5 MCPVP 10 -57.4 3.9 -44.2 -68.4 -96.5 -91.6 20 -45.8 0.7 -39.7 -55.4 -72.8 -82.3 30 -45.8 -12.1 -37.6 -51.1 -71.1 -69.6 PV/VA 10 -61.1 -7.3 -46.8 -78.0 -78.5 -90.8 20 -43.5 -12.1 -42.4 -51.6 -63.4 -69.9 30 -88.5 -5.3 -72.1 -101.9 -129.4 -170.0
169 RESULTS AND DISCUSSION
Table 24. r2 values of theophylline pellets coated with composite Surelease (EC) coat cured at 60 °C, 24 h.
% of polymeric Zero First Hixson- Higuchi Hopfen- Baker- additives order order Crowell berg Lonsdale PVP K17 10 0.996 0.894 0.946 0.989 0.996 0.936 20 0.982 0.837 0.907 0.993 0.997 0.971 30 0.963 0.843 0.899 0.976 0.982 0.963 PVP K29 10 0.996 0.860 0.931 0.993 0.999 0.953 20 0.994 0.893 0.943 0.990 0.997 0.949 30 0.984 0.902 0.947 0.944 0.983 0.908 PVP K60 PVP K60 10 0.990 0.868 0.931 0.983 0.997 0.970 20 0.990 0.910 0.952 0.967 0.990 0.921 30 0.990 0.904 0.951 0.966 0.990 0.927 PVP K90 10 0.998 0.949 0.981 0.982 0.990 0.918 20 0.987 0.880 0.938 0.965 0.986 0.926 30 0.974 0.855 0.911 0.981 0.989 0.961 MCPVP 10 0.980 0.793 0.880 0.996 0.996 0.975 20 0.990 0.895 0.945 0.975 0.992 0.930 30 0.993 0.905 0.952 0.983 0.993 0.943 PV/VA 10 0.998 0.927 0.964 0.994 0.999 0.960 20 0.979 0.857 0.914 0.989 0.993 0.967 30 0.989 0.883 0.934 0.990 0.996 0.958
170 RESULTS AND DISCUSSION
Table 25. AIC values of theophylline pellets coated with composite Surelese (EC) coat cured at 60 °C, 24 h.
% of polymeric Zero First Hixson- Higuchi Hopfen- Baker- additives order order Crowell berg Lonsdale
PVP K17 10 -46.1 -2.8 -34.4 -49.4 -67.6 -66.6 20 -50.1 2.8 -38.2 -60.4 -87.4 -80.2 30 -44.2 3.9 -38.1 -48.6 -70.9 -81.7 PVP K29 10 -40.4 -1.5 -25.1 -36.3 -55.6 -45.5 20 -81.4 -1.2 -55.9 -74.0 -115.2 -101.5 30 -72.3 2.1 -42.6 -53.0 -96.2 -101.8 PVP K60 10 -50.4 0.2 -35.4 -45.2 -75.3 -74.8 20 -58.6 -5.0 -47.0 -46.7 -76.3 -73.8 30 -58.8 -3.1 -46.8 -45.0 -76.1 -74.5 PVP K90 10 -45.4 -9.6 -31.8 -30.9 -42.3 -42.3 20 -66.8 -0.1 -52.4 -52.6 -84.9 -86.4 30 -46.3 0.9 -39.1 -50.1 -72.5 -78.2 MCPVP 10 -40.1 0.7 -29.4 -53.2 -66.3 -57.2 20 -56.4 -2.3 -43.3 -47.5 -75.7 -71.6 30 -46.6 -5.8 -36.2 -39.4 -58.8 -55.9 PV/VA 10 -81.9 -12.6 -53.8 -70.1 -108.4 -95.3 20 -45.2 -1.4 -36.8 -51.6 -70.5 -72.8 30 -42.4 -4.2 -33.5 -43.3 -64.0 -56.8
pellets with 12 %w/w coating was most likely diffusion-controlled since the film coating was sufficiently thick and uniform with minimum number of pores or flaws.
The release rate would therefore depend on the permeability of the drug in the polymer coat and solubility in dissolution fluids. As discussed in previous sections, addition of small amount of PVPs resulted in the formation of almost uniform film with PVP dispersed in the continuous EC phase. When drug-loaded pellets were encapsulated with these composite films, it was expected that the film coat would still be intact and act as a semi-permeable membrane when the water-soluble components,
171 RESULTS AND DISCUSSION
(a)
0
-0.005
-0.01
-0.015
-0.02
-0.025
-0.03 Release rate constant
-0.035
-0.04 0 500 1,000 1,500 2,000 Mw of PVP (x103)
(b) 100
90
80
70
60
(min) 50
50% 50% 40 T
30
20
10
0 0 500 1,000 1,500 2,000 Mw of PVP (x103)
172 RESULTS AND DISCUSSION
(c) 100 90
80
70
60 50% 50
MDT 40
30
20
10
0 0 500 1,000 1,500 2,000 Mw of PVP (x 103)
Figure 41. Influence of molecular weight of PVP on the (a) Hopefenberg release rate constant (b) T50% of chlorpheniramine pellets and (c) MDT50% of theophylline pellets coated with Surelease, cured at 60 °C. 24 h. 10 %w/w, 20%w/w and 30 %w/w PVP
which is mainly PVP dissolved in the aqueous medium. Hence, drug release could be greatly enhanced by diffusion through the water-filled channels.
Figure 41a shows the relationship between the Hopfenberg constant for chlorpheniramine and molecular weight of PVP added to the Surelease coat. At 10
%w/w level, a sharp increase in rate constant was obtained as the molecular weight of
PVP increased from 8, 000 to 58, 000 and a relatively smaller rise as the molecular weight of PVP increased to 400, 000. The rate constants did not show any particular trend of increase as the molecular weight of PVP became larger. Similar trend was also observed for T50% of chlorpheniramine pellets (Figure 41b). At 10 %w/w PVP, a steep drop in T50% was obtained as the molecular weight of PVP rise from 8, 000 to
58, 000. This change was followed by a leveling off. It was thought that the molecular weight of PVP might influence the release of drugs by causing formation of larger
173 RESULTS AND DISCUSSION and/or greater number of pores available for drug diffusion. It was noted that the molecular weight of PVP exerted noticeable influence in the range of 8, 000 to about
400, 000. Beyond 400, 000, the molecular weight of PVP did not result in much change in T50%. In contrast, at 10 %w/w level, increase in molecular weight of PVP did not produce any steep increase in rate constant of theophylline concentrations of
PVPs (20 - 30 %w/w), sharp increase in both zero-order and Higuchi rate constant was observed as the molecular weight increased from 8,000 to 58, 000. In addition, a gradual increase in rate constant was observed as the molecular weight of PVP increased further. Similar trend was observed for MDT50% and T50% of theophylline pellets coated with composite EC film coat (Figure 41c). The two phases of change in release rate with increasing size of PVP suggested a possible indication of change of mechanism of drug release. The initial phase being controlled mainly via diffusion of drug molecules through aqueous pores or channels formed by dissolution of PVPs which were randomly dispersed in the continuous EC membrane. Hence as the molecular weight of PVP increased from 8, 000 to 58, 000, a significant and sharp increase in release rate constant and drop in time taken for 50 % of the drug to be released was obtained. As the molecular weight of PVP increased further, another mechanism of release, such as swelling and/erosion-controlled release also come into play.
Investigation on the effects of swelling and erosion on EC-PVP coat were carried out using composite EC-PVP films containing 10 - 30 %w/w of PVPs. Figures
42 - 43 show the swelling index and extractable amount of the composite EC films over time. Generally both the swelling index and extractable amount increased gradually within the first 2 hours and levelled off thereafter. In general, PVP with higher molecular weight showed higher swelling index (Figure 42). The highest
174 RESULTS AND DISCUSSION swelling index within the first hour was determined for each additive. It was found to decrease in the following order:
for 10 % level; PVP K60 > MCPVP > PVP K90 > PVP K29 > PVP K17,
(a) 30
25
20
15
10
Extractable amount (%) 5
0 0 30 60 90 120 150 180 210 240 270 300 Time (min)
(b) 35
30
25
20
15
10
Extractable amount (%) 5
0 0 30 60 90 120 150 180 210 240 270 300 Time (min)
175 RESULTS AND DISCUSSION
(c) 35
30
25
20
15
10
Extractable amount (%) 5
0 0 30 60 90 120 150 180 210 240 270 300 Time (min)
Figure 42. Extractable amount of Surelease (EC) and composite EC films containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of additives. EC-PVP K17; EC- PVP K29; EC-PVP K60; EC-PVP K90; EC-MCPVP; EC-PV/VA; EC.
(a) 90 80 70 60 50 40 (%w/w)
s I 30
20
10 0 0 30 60 90 120 150 180 210 240 270 300 Time (min)
176 RESULTS AND DISCUSSION
(b) 180
160
140
120
100
80 (%w/w) s I 60
40
20
0 0 30 60 90 120 150 180 210 240 270 300 Time (min)
(c) 250
200
150 (%w/w)
s 100 I
50
0 0 30 60 90 120 150 180 210 240 270 300 Time (min)
Figure 43. Swelling index of Surelease (EC) and composite EC films containing (a) 10 %w/w (b) 20 %w/w and (c) 30 %w/w of additives. EC-PVP K17; EC-PVP K29; EC-PVP K60; EC-PVP K90; EC-MCPVP; EC-PV/VA; EC.
177 RESULTS AND DISCUSSION
for 20 % level; MCPVP > PVP K90 > PVP K29 > PVP K60 > PVP K17, and
for 30% level; MCPVP > PVP K60 > PVP K90 > PVP K29 > PVP K17.
With the exception of PVP K60, the rate of increase of swelling index was directly related to the molecular weight of PVPs. However, it was also noted that for
30 %w/w level, the swelling index profile of EC-PVP K90 started to decline after 20 min to a similar level as EC films (Figure 43c). This phenomenon was most obvious for composite Surelease films containing 30 %w/w PVP K90. This result could be explained by the amount of PVP that was lost due to dissolution into the aqueous media.
From Figures 42a-c, the highest extractable amount for each PVP was determined. It was found to decrease in the following order:
for 10 % level; PVP K60 > PVP K90 > PVP K29 > MCPVP > PVP K17,
for 20 % level; PVP K90 > PVP K60 > PVP K29 > MCPVP > PVP k17, and
for 30% level; PVP K90 > PVP K60 > PVP K17 ≥ MCPVP > PVP K29.
Since the extractable amounts were related to the differences between the weights of dried films before and after exposure to the dissolution media, they indicated the amounts of water-soluble components that were lost due to dissolution. Apparently,
PVP K90 showed the greatest tendency to be dissolved or eroded in the aqueous media. With the dissolution of PVP K90, the resultant composite Surelease films consisted of numerous pores or water-filled voids. Since most of the PVP K90 component in the composite Surelease films were not retained, the swelling index profile gradually declined. In contrast, MCPVP which has the highest molecular weight showed the second lowest extractable amount, suggesting that the MCPVP component was still retained in the composite Surelease film structure after exposure
178 RESULTS AND DISCUSSION to water. Hence, the wetted EC-MCPVP films were able to absorb high amount of water.
Low molecular weight PVPs, such as PVP K17 and K29, exhibited both low swelling index as well as extractable amount. It was possible that a certain portion of
PVP K17 and PVP K29 in the composite Surelease films was dissolved while the rest remained trapped in the film. Generally, PVP K29 also showed a higher swelling index and extractable amount compared to PVP K17. This indicated that a greater amount of PVP K29 could be lost due to dissolution compared to PVP K17 and/or the amount of PVP K29 retained in the polymer film had a higher water retention power.
The above phenomeon aptly accounted for the great difference in T50% value between pellets coated with EC-PVP K17 and EC-PVP K29. In addition, the greater swelling tendency of PVP K29 also suggested that the resultant swollen composite EC films could also enhanced the dissolution and diffusion of drug, especially water - soluble drugs through the polymer coat. Moreover, the difference in T50% values between pellets coated with EC-PVP K17 and EC-PVP K29 were most marked at higher concentration of 20 - 30 %w/w for theophylline. On the other hand, release of chlorpheniramine was more marked at lower concentrations of 10 - 20 %w/w PVP.
Since theophylline has a lower aqueous solubility than chlorpheniramine, this implied that swelling-controlled mechanism might have a greater influence on EC-PVP K17 and K29 coated pellets at higher concentration of PVPs. At low concentrations (10 -
20 %) of PVP, significantly greater amounts of PVP were leached out from EC-PVP
K29 while the difference in swelling index between EC-PVP K29 and EC-PVP K17 films was comparatively smaller. At low concentrations of PVP, composite EC-PVP
K29 films showed similar swelling tendencies as EC-PVP K17, and greater amounts of PVP K29 were also dissolved resulting in more and perhaps larger pores.
179 RESULTS AND DISCUSSION
Consequently, chlorpheniramine which has a higher aqueous solubility, showed greater difference in release rate, whereas the release of theophylline which has low aqueous solubility, was not greatly enhanced with the change in additives form PVP
K29 to PVP K17. However, as the concentration of PVP in the composite Surelease coat increased, more PVP could be dissolved and leached out from the coat, thereby enhancing the release of both chlorpheniramine and theophylline. Moreover at high concentration, relatively greater amount of PVP K17 was extracted and lost, resulting in much lower swelling profile compared to PVP K29.
Except for PVP K60, the PVPs showed quite comparable or small difference in the extractable amount and swelling index at 10 percent level. Hence, this explained the lack of influence of molecular weight of PVPs on the release rate of chlorpheniramine and theophylline pellets coated with 10 %w/w PVP. At higher concentrations, the higher molecular weight PVPs showed greater swelling indices and leachable amounts, indicating a shift in mechanism towards swelling-controlled and erosion-controlled release. It was possible that PVPs of medium molecular weights such as PVP K29 and K60 exhibited drug release that was controlled by swelling and diffusion mechanisms. In the earlier part of this study, it was found that the higher molecular weight PVPs, such as PVP K90 and MCPVP tended to form a separate layer when present at high concentration. Since PVP K90 also showed a high tendency to dissolve in aqueous media, it was likely that the PVPs present on top of the Surelease membrane were dissolved and eroded away when exposed to the aqueous media. The resultant film coat would consist of a thinner Surelease coat with flaws or pores created by the embedded PVP molecules. This was illustrated by the greater increase in release rate of theophylline at 20 %w/w compared to 30 %w/w.
180 RESULTS AND DISCUSSION
4. Effect of curing on drug release from EC-PVP coated pellets
In the above discussion, it was shown that curing Surelease coated pellets at
60 °C for 24 h impeded drug release. In order to investigate whether curing had similar effect on composite Surelease coat, the dissolution profiles of cured and uncured pellets coated with composite Surelease films were compared. As seen from
Figures 44a - c, the dissolution profiles of uncured pellets were very similar to those cured. The f1 and f2 values in Table 26 showed that none of the dissolution profiles of pellets cured at 60 °C were more than 10 percent different from those of uncured pellets. This finding was not unexpected since release of drugs occurred mainly through water-filled channels created by dissolution of water - soluble PVP. Curing of composite Surelease coated pellets is not likely to affect the dissolution of water - soluble PVP from the composite Surelease.
Table 26. Statistical analysis on influence of curing on chlorpheniramine release from pellets coated with composite Surelease (EC).
Film coat Average f1 Average f2 EC-PVP K29 (9:1) 4.1 78.3 EC-PVP K29 (8:2) 9.8 57.7 EC-PVP K29 (7:3) 8.9 65.6 EC-PVP K90 (9:1) 11.4 58.7 EC-PVP K90 (8:2) 7.2 69.7 EC-PVP K90 (7:3) 6.5 68.3 EC-MCPVP (9:1) 11.5 61.3 EC-MCPVP (8:2) 8.8 64.2 EC-MCPVP (7:3) 11.7 57.1 EC-PV/VA (9:1) *19.2 53.4 EC-PV/VA (8:2) 4.6 72.9 EC-PV/VA (7:3) 7.7 63.5
181 RESULTS AND DISCUSSION
(a)
EC-PVP K29 (9:1), cured EC-PVP K29 (8:2), cured EC-PVP K29 (7:3), cured EC-PVP K29 (9:1), uncured EC-PVP K29 (8:2), uncured EC-PVP K29 (7:3), uncured EC 100
80
60
40
20
Chlorpheniramine released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
(b) EC-PVP K90 (9:1), cured EC-PVP K90 (8:2), cured EC-PVP K90 (7:3), cured EC-PVP K90 (9:1),uncured EC-PVP K90 (8:2), uncured EC-PVP K90 (7:3), uncured EC 100
80
60
40
20 Chlorpheniramine released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
182 RESULTS AND DISCUSSION
(c) EC-MCPVP (9:1), cured EC-MCPVP (8:2), cured EC-MCPVP (7:3), cured EC-MCPVP (9:1), uncured EC-MCPVP (8:2), uncured EC-MCPVP (7:3), uncured EC 100
80
60
40
20
Chlorpheniramine released (%) 0 0 60 120 180 240 300 360 420 480 540 600 660 720 Time (min)
Figure 44. Dissolution profiles of chlorpheniramine pellets coated with Surelease (EC) and composite EC containing 10 - 30 %w/w (a) PVP K29, (b) PVP K90 and (c) MCPVP, cured at 60 °C, 24 h or uncured.
5. Influence of PV/VA on drug release from EC - coated pellets
The effect of PV/VA, a cross-linked PVP co-polymer, on drug release from
Surelease coated pellets was also studied (Figures 45 - 46). As mentioned above,
PV/VA has similar molecular weight as PVP K29, pellets coated with EC-PV/VA was used as comparator to determine the effect of chemical nature of additive on
Surelease coat. Generally, the r2 and AIC values suggested that chlorpheniramine pellets coated with EC-PV/VA followed Higuchi or Baker-lonsdale models while theophylline pellets release profiles seem to suggest Hopfenberg model (Tables 27 -
29). Comparison between the Hopfenberg release rate constants of EC-PV/VA and
EC-PVP K29, showed that at 10 % level, EC-PV/VA seems to be able retarded
183 RESULTS AND DISCUSSION release of chlorpheniramine better than EC-PVP K29 (Table 28). However as the concentration of PV/VA increased, the difference between rate constant of EC-
PV/VA and EC-PVP K29 diminished. In contrast, significant difference was observed for the both the zero-order and Higuchi rate constants of theophylline pellets coated with EC-PV/VA and EC-PVP K29 at all levels (Table 30).
Unlike, chlorpheniramine pellets, the differences for theophylline pellets appeared to be greater at higher concentration of additives. At 30 %w/w level, PV/VA was found to have slower rate constant compared to EC-PVP K17. These observations could be explained from the swelling results of EC-PV/VA and EC-PVP
K29 films. As seen in Figures 42 and 43, at low concentration, EC-PVP K29 showed slightly higher swelling index and extractable amount compared to EC-PV/VA.
However, as the concentration of additives increased to 20-30 %w/w, the EC-PVP
K29 showed significantly greater swelling index, though amount extracted was quite comparable to EC-PV/VA. This implied that swelling-controlled mechanism would have exerted a greater influence on drug release from EC-PVP K29 coated pellets compared to EC-PV/VA. Consequently, a marked increase in release rate was observed for theophylline pellets as composite EC-PVP K29 coat was more ‘porous’ and had higher water-content. This would help to enhance the release of less water- soluble drugs, such as theophylline as there is greater volume for dissolution. On the otherhand, chlorpheniramine is very soluble in aqueous medium, and presence of a more swollen layer did not greatly affect its release rate.
184 RESULTS AND DISCUSSION
100
80
60
40 4%, cured 6%, cured 8%, cured 10%, cured 12%, cured 4%, uncured 20 6%, uncured 8%, uncured 10%, uncured 12%, uncured Chlorpheniramine released (%) 0 0 30 60 90 120 150 180 210 240 270 300 330 360 Time (min)
Figure 45. Dissolution profiles of chlorpheniramine pellets coated with composite EC containing 30 %w/w PV/VA, cured at 60 °C for 24 h or uncured.
100
80
60 4 % 40 6 % 8 % 20 10 %
Theophylline released (%) 12 %
0 0 30 60 90 120 150 180 210 240 270 300 330 360 Time (min)
Figure 46. Dissolution profiles of theophylline pellets coated with composite Surelease (EC) coating containing 30 %w/w PV/VA, cured at 60 °C for 24 h.
185 RESULTS AND DISCUSSION
Table 27. Fit factors of drug release models for chlorpheniramine pellets coated with composite Surelease (EC) containing 30 %w/w of PV/VA cured at 60 °C, 24 h
Release models Coating levels 4 % 6 % 8 % 10 % 12 % r2 Zero order 0.933 0.887 0.802 0.783 0.957 First order 0.734 0.687 0.605 0.593 0.827 Hixson-Crowell 0.815 0.769 0.680 0.664 0.883 Higuchi 0.991 0.968 0.907 0.895 0.980 Hopfenberg 0.975 0.939 0.867 0.849 0.972 Baker-Lonsdale 0.994 0.983 0.943 0.933 0.973 AIC Zero order -43.3 -35.2 -31.5 -31.3 -88.5 First order 2.7 11.6 11.1 16.3 -5.3 Hixson-Crowell -39.5 -34.1 -35.5 -35.6 -72.1 Higuchi -62.8 -53.2 -42.5 -43.2 -101.9 Hopfenberg -73.2 -63.0 -57.4 -60.2 -129.4 Baker-Lonsdale -113.6 -103.2 -93.0 -100.6 -170.0
Table 28. Hopfenberg release rate constants of chlorpheniramine (CPM) and theophylline (THE) pellets coated with composite Surelease (EC) containing 30 %w/w of PV/VA cured at 60 °C, 24 h
Coating levels CPM THE 4 % -0.0218 -0.0436 6 % -0.0191 -0.0401 8 % -0.0144 -0.0373 10 % -0.0098 -0.0335 12 % -0.0212 -0.0046
186 RESULTS AND DISCUSSION
Table 29. Fit factors of theophylline pellets coated with composite Surelease (EC) containing 30 %w/w of PV/VA cured at 60 °C, 24 h
Release models Coating levels 4 % 6 % 8 % 10 % 12 % r2 Zero order 0.981 0.993 0.984 0.989 0.963 First order 0.845 0.890 0.854 0.900 0.843 Hixson-Crowell 0.908 0.943 0.916 0.945 0.899 Higuchi 0.995 0.985 0.988 0.984 0.976 Hopfenberg 0.994 0.994 0.994 0.994 0.982 Baker-Lonsdale 0.961 0.937 0.955 0.949 0.963 AIC Zero order -38.8 -47.0 -43.4 -51.5 -44.2 First order -3.0 -4.4 -0.1 -5.0 3.9 Hixson-Crowell -31.7 -35.0 -33.5 -41.1 -38.1 Higuchi -50.6 -43.0 -46.5 -48.6 -48.6 Hopfenberg -60.3 -61.9 -67.2 -71.4 -70.9 Baker-Lonsdale -57.3 -57.0 -64.0 -68.1 -81.7
187 CONCLUSION
V. CONCLUSION
It has been shown in this study that EC film stability is influenced by several factors, including type and amount of plasticizers remaining in the film, as well as the storage conditions. Plasticizers interact with EC and affect its film properties primarily in the following three ways. Firstly, plasticizers with low permanence, such as GTA and DEP can cause formation of brittle films as the plasticizers volatilize or degrade on extended storage. Secondly, the extent of coalescence or ageing on EC films varies with different plasticizers. In addition, the stability of EC films to storage conditions is also dependent of type of plasticizers.
Efficiency of plasticizers is one factor which contributes tremendously to the performance of EC film coat in controlling drug release. At a commonly applied concentration of 30 percent, the citrate ester, particularly TEC, had been shown to interact well with Aquacoat, while GTA and DEP were not able to plasticize
Aquacoat films as effectively. Storage humidities and temperature could potentially affect the expected performance of coated products through modulation on the ageing process. Exposing the film membrane to low humidity or high temperature accelerate loss of bound water, thus enhancing the coalescence or fusion of polymer molecules.
Storage environments with high moisture content, on the other hand could delay and reduce the coalescence of films but not prevent it. With the exception of GTA, chlorpheniramine release from pellets coated with plasticized Aquacoat films were affected to varying degree by storage conditions. Higher storage temperature tends to effect greater changes for pellets coated with Aquacoat containing citric acid class of plasticizers (TEC and ATEC) compared to other plasticizers. This study demonstrated that the physicochemical properties of the plasticizer coupled with the storage conditions have an important influence on the stability and performance of the final
188 CONCLUSION film coat. Hence it is important to select carefully the type of plasticizers for film coating, ensure the stability and optimal performance of the dosage forms.
In the second part of the study, the influence of secondary polymer, PVP and its derivatives on the properties of EC film and coated pellets was investigated. EC films formed from aqueous dispersion were found to be smooth and relatively transparent. Addition of PVP, especially high molecular weight PVP, and MCPVP resulted in drastic reduction in film transparency, suggesting that molecular weight of
PVP greatly affects the clarity of composite EC films. The interaction between EC and PVP polymeric additives was found to be dependent on the molecular weight, concentration and chemical nature of the additives. When added to EC, low molecular weight PVP would be randomly distributed through the EC matrix as a disperse phase. Increased concentration up to 30% did not alter the phase distribution. In contrast, greater interaction was exhibited between EC and higher molecular weight
PVP as represented by PVP K60, K90 and MCPVP. At low concentration, higher molecular weight PVP may exist as a disperse phase in EC matrix. However, as the concentration increased, the higher molecular weight additives tend to aggregate together forming a separate continuous phase. Formation of separate continuous layers became more prominent with increasing concentration. Increase molecular weight of additives accelerated the progression to formation of separate layers.
At low concentration, PVP increased the hardness of EC films probably due to intermolecular interaction between EC and PVP. However, addition of large amount of high molecular weight PVP or MCPVP would result in polymer phase separation.
Incorporation of PVP increased the puncture strength and percentage elongation of dry films but lowered the puncture strength and percentage elongation of wet films.
On the other hand, PV/VA increased the percentage elongation of wet films, while the
189 CONCLUSION dry films remained brittle. This suggested that wet EC-PV/VA films were more
‘intact’ or compact than EC-PVP films, allowing the water molecule to be retained in the matrix where they exerted their plasticizing effect. The concentration of PVP,
MCPVP or PV/VA used had a significant influence on the flexibility of the resultant films. Addition of PVP, MCPVP and PV/VA resulted in increase in glass transition temperature, tensile strength and elastic modulus.
Depending on its molecular weight and concentration, addition of PVP affected the water vapour permeability of EC films to varying extent. The water vapour permeability of composite EC films was significantly increased with addition of high molecular weight PVP, but not low molecular weight PVP (PVP K17, K29) at low concentration. However, with greater amount of PVP present, the effect of molecular weight of PVP on water vapour permeability became less significant.
PV/VA a N-vinyl-2-pyrrolidone and vinyl acetate co-polymer which is less hygroscopic than PVP K29, showed a lower permeability to water vapour. In order not to affect the application of EC film as a protective moisture barrier, the type or molecular weights as well as amount of PVP added should be carefully selected and controlled in order not to compromise its water vapour permeability too much. Low molecular weight PVP, such as K17 and K29, or PV/VA at low concentration were found to be the more suitable for additives that does not compromise the water vapour permeability of EC films.
Addition of PVP markedly increased the permeability of drug EC. The increase was found to be dependent on the molecular weight of PVP added. Higher molecular weight PVP generally increased the drug permeability of EC film to a greater extent. Generally, the permeability of composite EC-PV/VA films was almost similar to those of composite EC-PVP K29 films for all the model drugs studied
190 CONCLUSION expect for propranolol and sulphaguanidine. The dissimilarity in chemical nature between PV/VA and PVP K29 might exert different effect on the structural properties of composite EC films thereby resulting differences in permeabilities of the respective composite EC films to several model drugs. The swelling test result suggested that increase in drug permeability for EC-PV/VA film was likely due to diffusion of drug mainly through pores or channels due to dissolution of PV/VA present rather than through formation of hydrated network. On the other hand, permeability of composite
EC-PVP film may be affected by both the dissolution of PVP in the EC matrix creating channels or pores where the drug could diffuse through, as well as swelling of film matrix encouraged by the presence of PVP remaining in the film. The greater increase in drug permeability of EC-PVP films containing higher molecular weight
PVP may be due to dual effects of greater water uptake by composite films and formation of larger pores.
Drug permeability of composite EC films is also affected by the solubility of drugs. Permeability is enhanced for drugs of greater aqueous solubility, which can dissolve and permeate through the water-filled channels and swollen composite EC film matrix more effective than hydrophobic drugs. Exception was found for propranolol which might have interactions with EC film matrix, resulting in much slower release. A linear relationship was also found between molecular weights of model drugs studied and drug permeability of composite EC films containing high molecular weight PVP, such as PVP K60 and K90.
Dissolution study of pellets coated with EC and composite EC films showed drug release was dependent on the molecular weight and concentration of polymeric additives, as well as the physicochemical properties of the drugs, such as their solubility in the dissolution media. Modification of drug release is determined by the
191 CONCLUSION degree of influence each factors exert and the corresponding change in the release mechanisms. For instance, at low concentration of additives, PVPs with molecular ranging from 8, 000 to 58, 000 resulted in great increase in amount of chlorpheniramine released, while release of a less soluble drug, theophylline remained unchanged. Generally, as the concentration of PVPs added to EC coat increased the time taken to achieve maximum release was also greatly reduced. The reduction in release time was attributed mainly to a change in mechanism of release from diffusion-controlled to erosion or swelling controlled release. However, it was noted that while difference in release rate was observed for PVPs with low molecular weight (< 58,000), release profiles of chlorpheniramine pellets coated with composite
EC containing high molecular weight PVPs show negligible difference. In contrast, release of theophylline from pellets coated with composite EC containing high molecular weight PVP increase proportionally with the molecular weight of PVP as the amount of PVP added increased. This phenomenon adequately illustrated the solubility of drugs in influencing the release profiles of composite EC coated dosage forms. In summary, there is a great potential in the use of PVP and its derivatives as polymeric additives for modification of drug release from EC coated dosage forms.
Modification of drug release via use of polymeric additives has also proven to be advantageous compared to the convectional mean of varying the coating level. Unlike dosage forms coated with high level of polymer, composite polymers coating can enhance the release and allow most of the drug to be released within the gastrointestinal transit period. This would greatly reduce the amount of drug lost, which usually happened for dosage forms. Knowledge of the behaviour of EC and
PVPs and its derivatives could be of use in formulation of appropriate new coating for modified drug delivery.
192 REFERENCES
VI. REFERENCES
1. Aldeyeye, C. M., and Barabas, E., Polyvinylpyrrolidone (Povidone). Analytic Profiles Drug Substances Excipients 22 (1993) 555 – 685. 2. American Society for Testing and Materials (ASTM). Designation: D 882-97: Standard test methods for tensile properties of thin plastic sheeting. 3. American Society for Testing and Materials (ASTM). Designation: D 1653 - 93: Standard test methods for water vapor transmission of organic coating films. 4. American Society for Testing and Materials (ASTM). Designation: E 96-95: Standard test methods for water vapor transmission of materials. 5. Aulton M.E., Abdul Razzak M.H., Hogan J.E., Mechanical properties of hydroxypropylmethyl cellulose films derived from aqueous systems. Part 1. Influence of plasticizers Drug Development and Industrial Pharmacy 7(1981) 649-668 6. Aulton M.E., Abdul Razzak M.H., Hogan J.E., Mechanical properties of hydroxypropylmethyl cellulose films derived from aqueous systems. Part 2. Influence of solid inclusions Drug-Development-and-Industrial-Pharmacy 10 (1984) 541 - 561 7. Aulton, M.E., Assessment of the mechanical properties of film coating materials. International Journal of Pharmaceutical Technology and Product Manufacturing 3 (1982) 9 - 16. 8. Banker G.S., Film coating theory and practice. Journal of Pharmaceutical Sciences 55 (1966) 81 - 89. 9. Banker G.S., Gore A.Y., Swarbrick J., Water Vapour Transmission Properties of Free Polymer Films. Journal of Pharmacy and Pharmacology 18 (1966) 457 - 466. 10. Banker G.S., Pharmaceutical coating composition, and preparation and dosages so coated. US patent 4,330,338 (1982). 11. Barabas, E. S., Vinyl alkyl ether polymers. In: Mark, H.F. (Ed.), Encyclopedia of Polymer Science and Engineering: Vol. 17. Wiley-Interscience, New York, 1989, pp. 198. 12. BASF Fine Chemicals (BASF Corporation Chemicals Division), 1986. Kollidone® grades, MEF 129e. 13. Bechgaard, H., Nielsen, G.H., Controlled-release multiple-units and single- unit doses. Drug Development and Industrial Pharmacy 4 (1978) 53 -67. 14. Benita, S., Dor, P., Aronhime, M., Marom, G., Permeability and mechanical properties of a new polymer: cellulose hydrogen phthalate. International Journal of Pharmaceutics 33 (1986) 71 - 80. 15. Bertelsen, P., Christensen, F.N., Holm, P., Jorgensen, K., Comparison of organic solvent-based ethylcellulose coating on KCl crystals applied by top and bottom spraying in fluidized-bed equipment. International Journal of Pharmaceutics 111 (1994) 117-125. 16. Bicerano (Ed) Prediction of Polymer Properties. 2nd Edition Marcel Dekker, New York 1996. 17. Bilmeyer, F.W. (Ed.), Textbook of Polymer Science. Wiley, New York, 1984, pp. 246-247. 18. Bindschaedler C., Gurny R., Doelker E., Theoretical concepts regarding the formation of films from aqueous microdispersions and application to coating. Labratory Pharmaceutical Problems Technology 31 (1983) 389 - 394.
193 REFERENCES
19. Bindschaedler, C., Gurny, R., Doelker, E., Mechanically strong films produced from cellulose acetate latexes. Journal of Pharmacy and Pharmacology 39 (1987) 335 - 338. 20. Binschaedler C., Gurny R., Doelker E., Influence of emulsifiers on film formation from cellulose acetate latexes experimental study of phase separation phenomena due to sodium dodecyl sulfate. I. Journal of Applied Polymer Science 34 (1987) 2631 - 2647. 21. Bindschaedler C., Gurny R., Doelker E., Influence of emulsifiers on film formation from cellulose acetate latexes. Modeling approach to the fate of emulsifiers in highly plasticized films. II. Journal of Applied Polymer Science 37 (1989) 173 - 182. 22. Blanchon, S., Couarraze, G., Rieg-Falson, F., Cohen, G., Puisieux, F., Permeability of progesterone and synthetic progestin through methacrylic films. International Journal of Pharmaceutics 72 (1991) 1 - 10. 23. Blecher, L., Lorenz, D. H., Lowd, H. L., Wood, A. S., and Wyman, D. P., Polyvinylpyrrolidone. In: Davidson, R. L. (Ed.), Handbook of Water-Soluble Gums and Resins. Mc-Graw-Hill, 1980, Chap. 21. 24. Bodmeier, R., Paeratakul, O., Leaching of water-soluble plasticizers from polymeric films prepared from aqueous colloidal polymer dispersions. Drug Development and Industrial Pharmacy 18 (1992) 1865 - 1882 25. Bodmeier, R., Paeratakul, O., Dry and wet strengths of polymeric films prepared from an aqueous colloidal polymer dispersion: Eudragit RS 30D. International Journal of Pharmaceutics 96 (1993) 129 - 138. 26. Bodmeier, R., Paeratakul, O., The distribution of plasticizers between aqueous and polymer phases in aqueous colloidal polymer dispersions. International Journal of Pharmaceutics 103 (1994) 47 - 54. 27. Bodmeier, R., Paeratakul, O., Mechanical properties of dry and wet cellulosic and acrylic films prepared from aqueous colloidal polymer dispersions used in the coating of solid dosage forms. Pharmaceutical Research 11(1994) 882 - 888. 28. Bodmeier, R., Paeratakul, O., Plasticizer uptake by aqueous colloidal polymer dispersions used for the coating of solid dosage forms. International Journal of Pharmaceutics 152 (1997) 17 - 26. 29. Boyer, R. F., In: Bear, E., Radcliffe, S. V. (Ed.). Polymer Materials: Relationship between Structure and Mechanical Behaviour, American Society for Metals: Metals Park. OH. 1974, pp 227 - 368. 30. Brown G.L., Formation of films from polymer dispersions. Journal of Polymer Science 22 (1956) 423 - 434. 31. Burrell, H., In: Bandrup. J., and Immergut, E.H. (Eds.), Polymer Handbook, 2nd edition, P. IV. 337, Wiley Interscience, New York, 1975. 32. Bussemer, T., Dashevsky, A. and Bodmeier, R. A pulsatile drug delivery system based on rupturable coated hard gelatin capsules. Journal of Controlled Release 93 (2003) 331 - 339. 33. Chang R.K., Hsiao C.H., Robinson J.H., A review of aqueous coating techniques and preliminary data on release from a theophylline product. Pharmaceutical Technology 11 (1987) 56 - 68. 34. Chas. Ffizer & Co., Technical Bulletin No. 31 Citroflex Plasticizers, Brooklyn, New York, 1961
194 REFERENCES
35. Chen, B.H., Lee, D.J., Finite element analysis of slow drug release through deformed coating films: effects of morphology and average thickness of coating film. International Journal of Pharmaceutics 234 (2002) 25-42. 36. Chopra, R., Podczeck, F., Newton, J.M., Alderborn, G. The influence of pellet shape and film coating on the filling of pellets into hard shell capsules. European Journal of Pharmaceutics and Biopharmaceutics 53 (2002) 327- 333. 37. Chowhan, Z.T., Amaro, A.A., Chi, L.H., Comparative evaluations of aqueous film coated tablet formulations by high humidity aging. Drug Development and Industrial Pharmacy 8 (1982) 713 - 737. 38. Cole, G.C., Coating pans and coating columns. In: G. Cole (Ed.), Pharmaceutical Coating Technology. Taylor & Francis, London, 1995, pp. 205 - 239. 39. Costa, F.O., Sousa, J.J.S., Pais, A.A.C.C., Formosinho, S.J. Comparison of dissolution profiles of ibuprofen pellets. Journal of Controlled Release 89 (2003) 199-212. 40. Costa, P., An alternative method to evaluation of similarity factor in dissolution testing. International Journal of Pharmaceutics 220 (2001) 77 - 83. 41. Costa, P., Sousa Lobo, J.M. Modeling and comparison o dissolution profiles. European Journal of Pharmaceutical Sciences 13 (2001) 123 - 133. 42. Croll, S. G., The origin of residual internal stress in solvent-cast thermoplastic coatings. Journal of Applied Polymer Science 23 (1979) 847 - 858. 43. Dechesne, J. P., Vanderschueren, J., and Jaminet, F., Influence des plastifiants sur la temperature de transition vitreuse de filmogenes gastroresistants enterosolubles. Journal de pharmacie de Belgique 39 (1984) 341 - 347. 44. Dillon, R.E., Matheson, L.A., and Bradford, E.B., Sintering of synthetic latex particles. Journal of Colloid Science 6 (1951) 108 - 117 45. Dittgen, M., Relationship between film properties and drug release from acrylic films. Drug Development and Industrial Pharmacy 11 (1985) 269 - 279. 46. Dittgen, M., Durrani, M., and Lehmann, K., Acrylic polymers - a review of pharmaceutical applications. STP Pharma Science 7 (1997) 403 - 437. 47. Donbrow, M., and Samuelov, Y., Zero order drug delivery from double- layered porous films: release rate profiles from ethyl cellulose, hydroxypropyl cellulose and polyethylene glycol mixtures. Journal of Pharmacy and Pharmacology 32 (1980) 463 - 470. 48. Donbrow, M., and Friedman, M., Enhancement of permeability of ethylcellulose films for drug penetration. Journal of Pharmacy and Pharmacology 27 (1975) 633-646. 49. Donbrow, M., and Friedman, M., Permeability of films of ethylcellulose and PEG to caffeine. Journal of Pharmacy and Pharmacology 26 (1974) 148-150. 50. Dressman, J.B., Derbin, G.M., Ismailos, G., Jarvis, C., Ozturk, A., Palsson, B.O., and Wheatley, T.A., Circumvention of pH dependent release from ethylcellulose-coated pellets. Journal of Controlled Release 36: (1995) 251- 260. 51. Eastman Chemical Products, Myvacet - Distilled Acetylated Monoglycerides as Fodd Coatings, Kingsport, Tennessee, 1981. 52. Fan, T. Y., Wei, S. L., Yan, W. W., Chen, D. B. and Li, J., An investigation of pulsatile release tablets with ethyl cellulose and Eudragit L as film coating
195 REFERENCES
materials and cross-linked polyvinylpyrrolidone in the core tablets. Journal of Controlled Release 77 (2001) 245 - 251. 53. Fites, A.L., Banker, G.S., Smolen, V.F., Controlled drug release through polymeric films. Journal of Pharmaceutical Sciences 59 (1970) 610 - 613 54. Flynn, G.L., Yalkowsky, S.H., Rosemn, T.J., Mass transport phenomena and models: Theoretical concepts. Journal of Pharmaceutical Science 63 (1974) 479 - 510. 55. Frohoff-Hulsmann, M. A., Lippold, B. C., and Mcginity, J. W., Aqueous ethylcellulose dispersion containing plasticizers of different water solubility and hdroxypropyl methylellulose as coating material for diffusion pellets II: properties of sprayed films. European Journal of Pharmaceutical and Biopharmaceutics 48 (1999) 67 - 75. 56. Frohoff-Hulsmann, M.A., Schmitz A., Lippold, B.C., Aqueous ethylcellulose dispersions containing plasticizers of different water solubility and hydroxypropyl methyl cellulose as coating material for diffusion pellets I. Drug release rates from coated pellets. International Journal of Pharmaceutics 177 (1999) 69 - 82. 57. Ghebre-Sellassie, I., Pellets: a general overview. In: Ghebre-Sellassie, I. (Ed) Pharmaceutical Pelletization Technology, Marcel Dekker, 1989, New York, USA, pp 1 – 14. 58. Gibaldi, M., Feldman, S., Establishment of sink conditions in dissolution rate determinations - theoretical considerations and application to non disintegrating dosage forms. Journal of Pharmaceutical Science 56 (1967) 1238-1242. 59. Gibson, S.H.M., Rowe, R.C., White, E.F.T., Determination of critical pigment volume concentrations of pigmented film coating formulations using gloss measurement. International Journal of Pharmaceutics 45 (1988) 245-248. 60. Gilligan C.A., Wan Po A.L., Factors affecting drug release from a pellet system coated with an aqueous dispersion. International Journal of Pharmaceutics 73 (1991) 51 - 68. 61. Giovannini, F., Sistemi famaceutici a rilascio controllato basati su gel di alginate, Graduate Thesis, Università di Trieste, Facoltà di Ingegneria, Dipartimeno di Ingegneria Chimica (DICAMP), 1996 62. Goodhart, F.W., Harris, M.R., Murthy, K.S., Nesbitt, R.U., Evaluation of aqueous film forming dispersions for controlled release. Pharmaceutical Technology 8 (1984) 64, 66, 68, 70-71. 63. Gordon and Taylor, J. S., Ideal copolymers and the second order transition of synthetic rubbers 1. Non-crystalline co-polymers. Journal of Applied Chemistry 2 (1952) 493 - 500. 64. Grassi, M., Colombo, I., Mathematical modelling of drug permeation through a swollen membrane. Journal of Controlled Release 59 (1999) 343 - 359. 65. Greener, I.K., and Fennema, O., Evaluation of edible bilayer films for use as moisture barriers for food. Journal of Food Science 54 (1989) 1400-1406. 66. Gunder, W., Lippold, B. H., Lippold, B. C., Release of drugs from ethyl cellulose microcapsules (diffusion pellets) with pore formers and pore fusion. European Journal of Pharmaceutical Science 3 (1995) 203-214. 67. Guo, J.H., Robertson, R.E., Amidon, G.L., Influence of physical aging on mechanical properties of polymer free films: The prediction of long-term aging effects on the water permeability and dissolution rate of polymer film- coated tablets. Pharmaceutical Research 8 (1991) 1500 - 1504.
196 REFERENCES
68. Guo, J.H., Robertson, R.E., Amidon, G.L., An investigation into the mechanical and transport properties of aqueous latex films: a new hypothesis for the film-forming mechanism of aqueous dispersion system. Pharmaceutical Research 10 (1993) 405 - 410. 69. Gutierrez-Rocca, J.C., McGinity, J.W., Influence of water soluble and insoluble plasticizers on the physical and mechanical properties of acrylic resin copolymers. International Journal of Pharmaceutics 103 (1994) 293- 301. 70. Hale, A., and Blair, H. E., Polymer blends and block copolymers. In: Turi, E. A., (Ed.), Thermal Characterization of Polymeric Materials, (2nd Ed), Academic Press, San Diego, 1997, pp. 745 - 886. 71. Harris, F. W., Introduction to polymer chemistry. In State of the Art III: Polymer Chemistry, American Chemical Society, 1981, pp. 837 - 843. 72. Heng, P.W.S., Wan, L.S.C., Tan, Y.T.F., Relationship between aggregation of HPMC coated spheroids and tackiness/viscosity/additives of the coating formulations. International Journal of Pharmaceutics 138 (1996) 57 - 66. 73. Hennig, D., and Kala, H., Application of polyethylene glycol (PEG) in order to control the permeability of poly(meth)acrylate coatings Pharmazie 42 (1987) 26 - 28. 74. Higuchi, T., Mechanism of sustained action medication. Journal of Pharmaceutical Science 52 (1963) 1145 - 1149. 75. Hildebrand, J., and Scott, R., The solubility of non-electrolytes, 3rd ed., Reinhold, New York, 1949. 76. Hixson, A.W., Crowell, J.H., Dependence of reaction velocity upon surface and agitation. Industrial and Engineering Chemistry 23 (1931) 923 - 931. 77. Hjärtstam J., Hjertberg T., Swelling of pellets coated with a composite film containing ethylcellulose and hydroxypropyl methylcellulose. International Journal of Pharmaceutics 161 (1998) 23 - 28. 78. Hoechst Aktiengesellschaft, Properties and Uses of Polyethlene Glycols, Frankfurt, West Germany, 1977. 79. Hogan, J.E., Film-coating materials and their properties. In: Cole, G. (Ed.), Pharmaceutical Coating Technology. Taylor & Francis, London, 1995, pp. 6- 52. 80. Honary, S., Orafai, H., The effect of different plasticizer molecular weights and concentrations on mechanical and thermomechanical properties of free films. Drug Development and Industrial Pharmacy 28 (2002) 711-715. 81. Hood, D. K., Senak, L., Kopolow, S. L., Tallon, M. A., Kwak, Y. T., Patel, D. and Mckittrick, J., Structural insights into a novel molecular-scale composite of soluble poly(vinyl pyrrolidone) supporting uniformly dispersed nanoscale poly(vinyl pyrrolidone) particles. Journal of Applied Polymer Science 89 (2003) 734 - 741. 82. Hopfenberg, H.B., Controlled release from erodible slabs, cyclinders, an spheres. In: Paul, D.R. and Harris, F.W. (Eds.) Controlled release polymeric formulations. ACS symposium series 33, Washington, 1976 83. Hutchings, D., Clarson, S., Sakr, A., Studies of the mechanical properties of free films prepared using an ethylcellulose pseudolatex coating system. International Journal of Pharmaceutics 104 (1994) 203 - 213. 84. Hutchings, D., Sakr, A., Influence of pH and plasticizers on drug release from ethyl cellulose pseudolatex coated pellets. Journal of Pharmaceutical Science 83 (1994) 1386 – 1390.
197 REFERENCES
85. Iyer, U., Hong, W. M. N., Das, N., Ghebre-Sellassie, I., Comparative evaluation of three organic solvent and dispersion-based ethyl cellulose coating formulations. Pharmaceutical Technology 14 (1990) 68- 86. 86. Iyer, R.M., Augsburger, Parikh, D.M., Evaluation of drug layering and coating: effect of process mode and binder level. Drug Development and Industrial Pharmacy 19 (1993) 981 - 998. 87. Jenquin, M.R., Sarabia, R.E., Liebowitz, S.M., McGinity, J.W., Relationship of film properties to drug release from monolithic films containing adjuvants, Journal of Pharmaceutical Sciences 81 (1992) 983 – 989. 88. Johansson, B., Alderborn, G., Degree of pellet deformation during compaction and its relationship to the tensile strength of tablets formed of microcrystalline cellulose pellets. International Journal of Pharmaceutics 132 (1996) 207 - 220. 89. Jones, D.M., Percel, P.J., Coating of multiparticulates using molten materials: Formulation and process considerations. In: Ghebre-Sellasie, I. (Ed.), Multiparticulate Oral Drug Delivery. Dekker, New York, 1994, pp. 113 - 142. 90. Kallstrand, G., and Ekman, B., Membrane coated tablets: system for the controlled release of drugs. Journal of Pharmaceutical Science 72 (1983) 772 - 775 91. Kararli, T. T., Hurlbut, J. B., Needham, T. E., Glass-rubber transitions of cellulosic polymers by dynamic mechanical analysis. Journal of Pharmaceutical Science 79 (1990) 845 - 848. 92. Katzhendler, I., Hoffman, A., Golderger, A., Friedman, M., Modelling of drug release from erodible tablets. Journal of Pharmaceutical Sciences 86 (1997) 110 - 115. 93. Keshikawa T., Nakagami H., Film formation with coating systems of aqueous suspensions and latex dispersions of ethyl cellulose. Chemical and Pharmaceutical Bulletin 42 (1994) 656 - 662. 94. Kester, J.J., and Fennema, O., Edible films and coatings: a review. Food Technology 40 (1986) 47 – 56. 95. Kim, J.E., Kim, S.R., Lee, S.H., Lee, C.H. and Kim, D.D. The effect of pore formers on the controlled release of cefadroxil from a polyurethane matrix. International Journal of Pharmaceutics 201 (2000) 29 - 36. 96. Kumar, V., Yang, T., and Yang, Y., Interpolymer complexation. Part 1. Preparation and characterization of a polyvinyl acetate phthalate- polyvinylpyrrolidone (PVAP-PVP) complex. International Journal of Pharmaceutics 188 (1999) 221 - 232. 97. Lafferty, S. V., Newton, J. M., Podczeck, F., Dynamic mechanical thermal analysis studies of polymer films prepared from aqueous dispersion. International Journal of Pharmaceutics 235 (2002) 107 - 111. 98. Lapidus, H. and Lordi, N.G., Drug release from compressed hydrophilic matrices. Journal of Pharmaceutical Science 57 (1968) 1292 - 1301. 99. Lecomte F., Siepmann J., Walther M., MacRae R.J. and Bodmeier R., Blends of enteric and GIT-insoluble polymers used for film coating: physicochemical characterization and drug release patterns. Journal of Controlled Release 89 (2003) 57 - 471. 100. Lecomte, F., Siepmann, J., Walther, M., MacRae, R.J., Bodmeier, R. Polymer blends used for the coating of multiparticulates: Comparison of aqueous and organic coating techniques. Pharmaceutical Research 21 (2004) 882 – 890.
198 REFERENCES
101. Lehmann, K.O.R., Chemistry and application properties of polymethacrylate coating systems. In: Aqueous polymeric coatings for pharmaceutical dosage forms. McGinity, J.W. (Ed.), Marcel Dekker, New York, 1997 pp 101 - 176. 102. Lindholm, T., Juslin, M., Controlled release Tablets: Part 3: Ethylcellulose coats containing surfactant and powdered matter. Pharmazeutische Industrie 44 (1982) 937 - 941. 103. Lindholm, T., Juslin, M., Lindholm, B.A., Poikala, M., Tiilikainen,S., Varis, H., Properties of free ethylcellulose films containing surfactant and particulate matter. Pharmazeutische Industrie 49 (1987) 740-746. 104. Lippold, B. C., Sutter, B. K., and Lippold, B. C., Parameters controlling drug release from pellets coated with aqueous ethylcellulose dispersion. International Journal of Pharmaceutics 54 (1989) 15 - 25. 105. Lippold B.C., Lippold B.H., Sutter B.K., Gunder W., Properties of aqueous, plasticizer-containing ethylcellulose dispersions and prepared films in respect to the production of oral extended release formulations. Drug Development and Industrial Pharmacy 16 (1990) 1725 – 1747. 106. Lu, D.R., Abu-Izza, K., Mao, F., Nonlinear data fitting for controlled release devices: an integrated computer program. International Journal of Pharmaceutics 129 (1996) 243 - 251. 107. Macleod, G. S., Fell, J. T., and Collett, J. H., Studies on the physical properties of mixed pectin/ethylcellulose films intended for colonic drug delivery. International Journal of Pharmaceutics 157 (1997) 53 - 60. 108. Macleod, G.S., Collett, J.H., Fell, J.T., Potential use of mixed films of pectin, chitosan and HPMC for bimodal drug release. Journal of Controlled Release 58 (1999) 303 - 310. 109. Martin-Polo, M., Mauguin, C., and Voilley, A., Hydrophobic films and their efficiency against moisture transfer. 1. Influence of the film preparation technique. Journal of Agricultural Food Chemistry 40 (1992) 407 - 412. 110. Masilungan F.C., Lordi N.G., Evaluation of film coating compositions by thermomechanical analysis. Part 1. Penetration mode. International Journal of Pharmaceutics 20 (1984) 295 – 305. 111. Mathiowitz, E. (Ed.) Encyclopedia of Controlled Drug Delivery. John Wiley & Son Inc, New York, 1999. 112. Mauger, J. W., Experimental methods to evaluate diffusion coefficients and investigate transport processes of pharmaceutical interest. In: Amidon, G. L., Lee, P. I. and Topp, E. M. (Eds.), Transport Processes in Pharmaceutical systems. Marcel Dekker, New York, 1999, pp. 87 - 107. 113. Mauger, J.W., Chilko, D., Howard, S., On the analysis of the dissolution data. Drug Development and Industrial Pharmacy 12 (1986) 969 - 992. 114. McAinsh, J., and Rowe, R.C., U.S. Patent 4, 138,475 (1979) 115. Mehta, A.M., Processing and equipment considerations for aqueous coatings. In: McGinity, J.W. (Ed.), Aqueous polymeric coatings for pharmaceutical dosage forms. Dekker, New York, 2nd Ed. 1997 pp. 287 - 326. 116. Molina, E. B., LegarretaaI, I.G., and Carterb, E. J.V., Moisture barrier properties and morphology of mesquite gum–candelilla wax based edible emulsion coatings. Food Research International 36 (2003) 885 – 890. 117. Moore, J.W., Flanner, H.H., Mathematical comparison of dissolution profiles. Pharmaceutical Technology June (1996) 93 - 100.
199 REFERENCES
118. Mooter, G.V.D., Samyn, C., Kinget, R., Characterization of colon-specific azo polymers: A study of the swelling properties and the permeability of isolated polymeric films. International Journal of Pharmaceutics 111 (1994) 127-136. 119. Murayama, T., In: Dynamic Mechanical Properties of Polymeric Materials. Material Science Monographs. Elsevier, Amsterdam, 1978, Vol. 1 120. Myvacet - Distilled Acetylated Monoglycerides as Fodd Coatings, Eastman Chemical Products Inc., Kingsport, Tennessee, 1981. 121. Nair, R., Nyamweya, N., Gönen, S., Martínez-Miranda, L. J., Hoag, S. W., Influence of various drugs on the glass transition temperature of poly(vinylpyrrolidone): a thermodynamic and spectroscopic investigation. International Journal of Pharmaceutics 225 (2001) 83-96. 122. Nakano, T., Yuasa, H., Suppression of agglomeration in fluidized bed coating. IV. Effects of sodium citrate concentration on the suppression of particle agglomeration and the physical properties of HPMC film. International Journal of Pharmaceutics 215 (2000) 3 - 12. 123. Narisawa S., Yoshino H., Hirakawa Y., Noda K., Porosity- controlled ethylcellulose film coating. I. Formation of porous ethylcellulose film in the casting process and factors affecting film-density. Chemical Pharmaceutical Bulletin 41 (1993) 329 - 334. 124. Narisawa, S., Yoshino, H., Hirakawa, Y., Noda, K., Porosity- controlled ethylcellulose film coating. II. Spontaneous porous film formation in the spraying process and its solute permeability. International Journal of Pharmaceutics 104 (1994) 95 - 106. 125. Nash R.A., Pharmaceutical suspensions. In: Lieberman, H.A., Rieger, M.M., and Banker, G.S., (Eds.) Pharmaceutical Dosage Forms: Disperse Systems, Vol. 2. Marcel Dekker, New York and Basel, 1996, pp 1 - 46. 126. Nesbitt, R.U., Effect of formulation components on drug release from multiparticulates. Drug Development and Industrial Pharmacy 20 (1994) 3207 - 3236. 127. Nesbitt R.U., Mahjour M., Mills N.L., Fawzi M.B., Effect of substrate on mass release from ethylcellulose latex coated pellets. Journal of Controlled Release, 32 (1994) 71-77 128. Nielson, L. E., In: Mechanical Properties of Polymers and Composites, Marcel Dekker, New York, 1974, Vol. 1. 129. Nishio, Y., Haratani, T., Takahashi, T., Miscibility and orientation behaviour of poly(vinyl alcohol)/poly(vinyl pyrrolidone) blends. Journal of Polymer Science: Part B: Polymer Physics 28 (1990) 377 - 386. 130. Nishio, Y., Manley, R. S., Cellulose/poly(vinyl alcohol) blends prepared from solutions in N,N-dimethylacetamide-lithium chloride. Macromolecules 21 (1988) 1270 - 1277. 131. Nyamweya, N., and Hoag, S. W., Assessment of polymer-polymer interactions in blends of HPMC and film forming polymers by modulated temperature differential scanning calorimetry. Pharmaceutical Research 17 (2000) 625- 631. 132. O'Hara, T., Dunne, A., Bulter, J., Devane, J., A review of methods used to compare dissolution profile data. Pharmaceutical Science & Technology Today 1 (1998) 214 - 223. 133. Okhamafe A.O., York P., Analysis of the Permeation and Mechanical Characteristics of Some Aqueous-Based Film Coating Systems. Journal of Pharmacy and Pharmacology 35 (1983) 409 - 415.
200 REFERENCES
134. Okhamafe, A. O., and York, P., Interaction phenomena in pharmaceutical film coatings and testing methods. International Journal of Pharmaceutics 39 (1987) 1 - 21. 135. Okhamafe, A.O., York, P., Adhesion characteristics of some pigmented and unpigmented aqueous based film coatings applied to aspirin tablets. Journal of Pharmacy and Pharmacology 37 (1985) 849 - 853. 136. Okutgen, E., Hogan, J. E., and Aulton, M. E., Effects of tablet core dimensional instability on the generation of internal stresses within film coats. Part 3. Exposure to temperatures and relative humidities which mimic the film coating process. Drug Development and Industrial Pharmacy 17 (1991) 2005 - 2016. 137. Olabisi, O., Robeson, L. M., and Shaw, M. T., Polymer - polymer miscibility. New York, Academic Press, 1979. 138. Onions A., Films from water based colloidal dispersions. Manufacturing Chemist 57 (1986) 55, 57, 59. 139. Osterwald, H. P., Properties of film-formers and their use in aqueous systems. Pharmaceutical Research 2 (1985) 14 - 18. 140. Ozturk, A.G., Ozturk, S.S., Palsson, B.O., Wheatley, T.A., Dressman, J.B., Mechanism of release from pellets coated with an ethylcellulose-based film. Journal of Controlled Release 14 (1990) 203 - 213. 141. Paeratakul, O., Bodmeier, R., Drug release from beads coated with different ethyl cellulose latexes. Pharmaceutical Research 6 (1989) S72. 142. Parikh N.H., Porter S.C., Rohera B.D., Tensile properties of free films cast from aqueous ethyl cellulose dispersions. Pharmaceutical Research 10 (1993) 810 - 815. 143. Park, J. S., Park, J. W. and Ruckenstein, E., A dynamic mechanical and thermal analysis of unplasticized and plasticized poly(vinyl alcohol)/ methyl cellulose blends. Journal of Applied Polymer Science 80 (2001) 1825 - 1834. 144. Phuapradit W, Shah N.H., Railkar A., Williams L., Infeld M.H., In vitro characterization of polymeric membrane used for controlled release application. Drug Development and Industrial Pharmacy 21 (1995) 955 - 963. 145. Pickard, J.F., (1979) Ph.D Thesis, CNAA. 146. Pillay, V., Fessihi, R., Evaluation and comparison of dissolution data derived from different modified release dosage forms: an alternative method. Journal of Controlled Release 55 (1998) 45 - 55. 147. Podczeck, F., Comparison of in vitro dissolution profiles by calculating mean dissolution time (MDT) or mean residence time (MRT). International Journal of Pharmaceutics 97 (1993) 93 - 100. 148. Polli, J.E., Rekhi, S., Augsburger, L.L., Shah, V.P., Methods to compare dissolution profiles and a rationale for wide dissolution specifications for metoprolol tartrate tablets. Journal of Pharmaceutical Sciences 86 (1997) 690 - 700. 149. Porter S.C., Aqueous polymeric dispersions for film coating of pharmaceutical solid-dosage forms. In: Mahendra, K.S., and Micale, F.J. (Eds.) Surface phenomena and fine particles in water-based coatings and printing technology. Plenum Press, New York, 1991, pp. 71 - 108. 150. Porter, S. C., Controlled-release film coatings based on ethylcellulose. Drug Development and Industrial Pharmacy 15 (1989) 1495 - 1521.
201 REFERENCES
151. Porter, S., Bruno, C., In: Pharmaceutical dosage forms: Tablets, Lieberman, H.A., Lachman, L., and Schwartz, J.B. (Eds.), 2nd edition, Marcel Dekker, New York, 1990, Vol. 3, Chpt. 2. 152. Porter, S.C., Bruno, C.H., Coating of pharmaceutical solid dosage forms. In: Lieberman, H.A., Lachman, L., Schwartz, J.B. (Eds) Pharmaceutical dosage forms: Tablets 3, Marcel Dekker, New York, 1990 153. Porter, S.C., Ghebre-Sellassie, I. Key factors in the development of modified release pellets. In Ghebre-Sellassie, I., (Ed). Multiparticulate Oral drug Delivery. Marcel Dekker: New York, 1994, pp 217 – 284 154. Porter, S.C., Use of Opadry, Sureteric, and Surelease for the aqueous film coating of pharmaceutical oral dosage forms. In: Aqueous polymeric coatings for pharmaceutical dosage forms. McGinity, J.W. (Ed.), Marcel Dekker, New York, 1997 pp 327 – 272. 155. Radebaugh, G. W., Murtha, J. L., Julian, T. N., Bondi, J. N., Methods for evaluating the puncture and shear properties of pharmaceutical polymeric films. International Journal of Pharmaceutics 45 (1988) 39 – 46. 156. Ragnarsson, G., Sandberg, A., Johansson, M.O., Lindstedt, B., Sjogren, J., In vitro release characteristics of a membrane-coated pellet formulation – influence of drug solubility and particle size. International Journal of Pharmaceutics 79 (1992) 223 - 232. 157. Ragnarsson, G., Johansson, M.O., Coated drug cores in multiple unit preparations: influence of particle size. Drug Development and Industrial Pharmacy 14 (1988) 734 - 739. 158. Ranga Rao, K.V., Padmalatha Devi, K. and Buri, P., Influence of molecular size and water solubility of the solute on its release from swelling and erosion controlled polymeric matrices. Journal of Controlled Release 12 (1990) 133- 141. 159. Rao, B.S., Ramana, M.K.V., Studies on rifampicin release from ethylcellulose coated nonpareil beads. International Journal of Pharmaceutics 231 (2002) 97 - 106. 160. Rekhi G. S., Porter S.C., Jambhekar S.S., Factors affecting the release of propranolol hydrochloride from beads coated with aqueous polymeric dispersions. Drug Development and Industrial Pharmacy 21 (1995) 709 - 729 161. Rekhi, G. S., and Jambhekar, S. S., Ethylcelluose - a polymer review. Drug Development and Industrial Pharmacy 21 (1995) 61 - 77. 162. Remunan-Lopez, C., and Bodmeier, R., Mechanical water uptake and permeability properties of crosslinked chitosan glutamate and alginate films. Journal of Controlled Release 44 (1997) 215 - 225. 163. Rohera, B.D., Parikh, N.H., Influence of type and level of water-soluble additives on drug release and surface and mechanical properties of Surelease films. Pharmaceutical Development and Technology 7 (2002) 421 - 432. 164. Rowe, R. C., Correlations between the in-situ performance of tablet film coating formulations based on hydroxypropyl methyl cellulose and data obtained from the tensile testing of free films. Acta Pharmaceutical Technology 29 (1983) 205 - 207. 165. Rowe, R. C., Defects in film-coated tablets: aetiology and solutions. In: Granderton, D., and Jones, T. M. (Eds.), Advances in Pharmaceutical Sciences, Vol. 6, Academic Press, London, 1992, pp. 65-100. 166. Rowe, R.C., Prediction of compatibility/incompatibility in blends of ethylcellulose with hydroxypropyl methyl cellulose or hydroxypropyl
202 REFERENCES
cellulose using 2-dimensional solubility parameter maps. Journal of Pharmacy and Pharmacology 38 (1986) 214 vcb- 215 167. Rowe, R.C., The effect of the molecular weight of ethylcellulose on the drug release properties of mixed films of ethylcellulose and hydroxypropyl methyl cellulose. International Journal of Pharmaceutics 29 (1986) 37-41. 168. Rowe, R.C., Correlations between the in-situ performance of tablet film coating formulations based on hydroxypropyl methyl cellulose and data obtained from the tensile testing of free films. Acta Pharmaceutical Technology 29 (1983) 205 - 207. 169. Rowe, R. C., The cracking of film coatings on film-coated tablets - a theoretical approach with practical implications. Journal of Pharmacy and Pharmacology. 323 (1981) 423 - 426. 170. Rowe, R.C., Molecular weight and molecular weight distribution of hydroxypropyl methylcellulose used in the film coating of tablets Journal of Pharmacy and Pharmacology 32 (1980) 116 - 119 171. Sadeghi, F., Ford, J.L., Rubinstein, M.H., Rajabi-Siahboomi, A.R., Comparative study of drug release from pellets coated with HPMC or Surelease. Drug Development and Industrial Pharmacy 26 (2000) 651 - 660. 172. Sangalli, M.E., Zema, L., Maroni, A., Foppoli, A., Giordano, F. and Gazzaniga, A. Influence of betacyclodextrin on the release of poorly soluble drugs from inert and hydrophilic heterogeneous polymeric matrices. Biomaterials 22 (2001) 2647 – 2651. 173. Sakellariou P., Rowe R.C., White E.F.T., Polymer/polymer interaction in blends of ethylcellulose with both cellulose derivatives and polyethylene glycol 6000. International Journal of Pharmaceutics 34 (1986) 93 - 103. 174. Sakellariou, P., and Rowe, R.C., Interactions in cellulose derivative films for oral drug delivery. Progress in Polymer Science 20 (1995) 889 - 942. 175. Sakellariou, P., and Rowe, R.C., Phase-separation and morphology in ethylcellulose cellulose-acetate phthalate blends. Journal of Appiled Polymer Science 43 (1991) 845 - 855. 176. Sakellariou, P., Effect of polymer compatibility on surface enrichment in polymer blends. Polymer 34 (1993) 3408 - 3415. 177. Sánchez-Lafuente, C., Faucci, M.T., Fernández-Arévalo, M., Álvarez-Fuentes, J., Rabasco, A.M., and Mura, P., Development of sustained release matrix tablets of didanosine containing methacrylic and ethylcellulose polymers. International Journal of Pharmaceutics 234 (2002) 213 - 221. 178. Sathe, P.M., Tsong Y., Shah, V.P. In vitro dissolution profile comparison: statistics and analysis, model dependent approach. Pharmaceutical Research 13 (1996) 1799 – 1803. 179. Sato, K. The internal stress of coating films. Progress in Organic Coating 8 (1980) 143 - 160. 180. Savage, G.V., Rhodes, C.T., The sustained release coating of solid dosage forms: A historical review. Drug Development and Industrial Pharmacy 21 (1995) 93 - 118. 181. Savage, A.B., In: Encyclopedia of Polymer Science and Technology, 3 p 492, Mark, H.F., Gaylord, N.G. and Bikales, N.M. (Eds) Interscience, New York 1965 182. Scheflan, L., Jacobs, M.B., The Handbook of Solvents, Macmillan, London, 1953
203 REFERENCES
183. Schultz, P., Kleinebudde, P., A new multiparticulate delayed release system. Part I: Dissolution properties and release mechanism. Journal of Controlled Release 47 (1997) 181 - 189. 184. Shah, J.C., Analysis of permeation data: evaluation of the lag time method. International Journal of Pharmaceutics 90 (1993) 161-169. 185. Shah, V. P., Tsong, Y., Sathe, P., Liu, J. P., In Vitro Dissolution Profile Comparison - Statistics and Analysis of the Similarity Factor, f2. Pharmaceutical Research 15 (1998) 889 - 896. 186. Shah, V.P., Lesko, L.J, Fan, J., Fleischer, N., Handerson, J., Malinowski, H., Makary, M., Ouderkirk, L., Roy, S., Sathe, P., Singh, G.J.P., Tillman, L., Tsong, Y., Williams, R.L., FDA guidance for industry: dissolution testing of immediate release solid oral dosage forms. Dissolution Technology 4 (1997) 15 - 22. 187. Simha, R., Boyer, R. F., On a general relation involving the glass temperatures and coefficients of expansion of polymers. Journal of Chemical Physic 37 (1962) 1003 - 1007. 188. Sinko, C.M., Amidon, G.L., Plasticizer-induced changes in the mechanical rate of response of film coatings: an approach to quantitating plasticizer effectiveness. International Journal of Pharmaceutical 55 (1989) 241 - 256. 189. Skultety, F., and Sims, S.M., Evaluation of the loss of propylene glycol during aqueous film coating. Drug Development and Industrial Pharmacy 13(1987): 2209 - 2219. 190. Song, M., Hammiche, A., Pollock, H. M., Hourston, D. J., and Reading, M., Modulated differential scanning calorimetry: 4. Miscibility and glass transition behaviour in poly (methylmethacrylate) and poly(epichlorohydrin) blends. Polymer 37 (1996) 5661 - 5665. 191. Sousa, J.J., Sousa, A., Moura, M.J., Podczeck, F., Newton, J.M., The influence of core materials and film coating on the drug release from coated pellets. International Journal of Pharmaceutics 233 (2002) 111 - 122. 192. Steward P.A., Hearn J., Wilkinson M.C. et al., Studies on permeation through polymer latex films, I. Films containing no or only low levels of additives. Polymer International 38 (1995) 1 - 12. 193. Sun, Y. M., Huang, W. F., Chang, C. C., Spray-coated solution-cast ethylcellulose pseudolatex membranes. Journal of Membrane Science 157 (1999) 159 - 170. 194. Sung, K.C., Topp, E.M., Effect of drug hydrophilicity and membrane hydration on diffusion in hyaluronic acid ester membranes. Journal of Controlled Release 37 (1995) 95 - 104. 195. Surelease E-7-19010 - Product information Bulletin (Colorcon, US). 196. Tarvainen, M., Sutinen, R., Peltonen, S., Mikkonen, H., Maunus, J., Vähä- Heikkilä, K., Lehto, V. P. and Paronen, P. Enhanced film-forming properties for ethyl cellulose and starch acetate using n-alkenyl succinic anhydrides as novel plasticizers. European Journal of Pharmaceutical Science 19 (2003) 363 - 371. 197. Tarvainen, M., Sutinen, R., Peltonen, S., Tiihonen, P., Paronen, P., Starch acetate - a novel film-forming polymer for pharmaceutical coatings. Journal of Pharmaceutical Sciences 91 (2002) 282 - 289. 198. Taylor, L.S., and Zografi, G., Sugar-polymer hydrogen bond interactions in lyophilized amorphous mixtures. Journal of Pharmaceutical Sciences 87 (1998) 1615 - 1621.
204 REFERENCES
199. Lachmann, L. (Ed.), The theory and practice of industrial pharmacy. Lea & Febiger, Philadelphia, 1986. 200. Thioune O., Briancon S., Devissaguet J.P., Fessi H., Development of a new ethylcelluose pseudolatex for coating. Drug Development Research 50 (2000) 157 - 162. 201. Tunon, A., Grasjo, J., Alderborn, G., Effect of intragranular porosity on compression behaviour of and drug release from reservoir pellets. European Journal of Pharmaceutical Science 19 (2003) 333 - 344. 202. Umprayn K., Chitropas P., Amarekajorn S., Development of Terbutaline Sulfate Sustained-Release Coated pellets. Drug Development and Industrial Pharmacy 25 (1999) 477- 491. 203. Union Carbide Corp., Carbowax-Polyethylene Glycols, Danbury, Connecticut, 1981 204. Van Bommel, E.M.G., Fokkens, J.G., Crommelin, D.J.A., Proceedings pf 5th International Conference of Pharmaceutical Technology 2 (1989) 39 - 45. 205. Van Bommel, E.M.G., Fokkens, J.G., and Crommelin, D.J.A., Effects of additives on the physicochemical properties of sprayed ethylcellulose films. Acta Pharmaceutical Technonology. 35 (1989) 232 - 237. 206. Vanderhoff, J.W., and El-Asser, M.S., Polymer emulsification process. U.S. Patent 4,177,177 (1978). 207. Verma, R.K. and Garg, S. Development and evaluation of osmotically controlled oral drug delivery system of glipizide. European Journal of Pharmaceutics and Biopharmaceutics 57 (2004) 513-525. 208. Vrentas, J.S., Duda, J. L., Ju, S.T., and Ni, L.W., Oscillatory diffusion experiments for solvent diffusion in polymer films. Journal of Membrane Science18 (1984) 161 - 175 209. Wade and Weller, P.J., Povidone. In: Kibbe, A.H., (Ed.) Handbook of Pharmaceutical Excipients, 2nd edition, Pharmaceutical Press, London, 1994, pp. 392–399. 210. Wan, L.S.C., Lai, W.F., The influence of antitack additives on drug release from film-coated granules. International Journal of Pharmaceutical 94 (1992) 39 - 47. 211. Wesdyk, R., Joshi, Y.M., Vincentis, J.D., Newman, A.W., Jain, N.B., Factors affecting differences in film thickness of beads coated in fluidized bed units. International Journal of Pharmaceutical 93 (1993) 101 - 109. 212. Wetton, R.E., Marsh, R.D.L., Van deVelde, J.G., Theory and application of dynamic mechanical thermal analysis. Thermochimica Acta 175 (1991) 1 - 11. 213. Wheatley T.A., Steuernagel C.R., Latex emulsions for controlled drug delivery. In: Aqueous polymeric coatings for pharmaceutical dosage forms. McGinity, J.W. (Ed.), Marcel Dekker, New York, 1997 pp 1 - 54. 214. Wicks, Z.W, Jr., Jones, F.N., Pappas, S.P., (Eds). Organic coatings : science and technology Wiley-Interscience, New York , 1999, 2nd edition. 215. Wiederhold P.R. (Ed.) Water vapor measurement: methods and instrumentation. Marcel Dekker, New York, 1997. 216. Wong, T.W., Heng, P.W.S., Yeo, T.N., Chan, L.W., Influence of polyvinylpyrrolidone on aggregation propensity of coated spheroids. International Journal of Pharmaceutics 242 (2002) 357 - 360. 217. Woods, M.E., Dodge, J.S., Krieger, I.M. and Pierce, P.E., Emulsion polymerization with mixtures of anionic and nonionic surfactants. Journal of Paint Technology 40 (1968) 541 – 548.
205 REFERENCES
218. Wu C., McGinity J.W., Influence of relative humidity on the mechanical and drug release properties of theophylline pellets coated with an acrylic polymer containing methylparaben as a non-traditional plasticizer. European Journal of Pharmaceutics and Biopharmaceutics 50 (2000) 277 - 284. 219. Yamada, T., Onishi, H., and Machida, Y., Sustained release ketoprofen microparticles with ethyl cellulose and carboxymethylethyl cellulose, Journal of Controlled Release 75 (2001) 271-282. 220. Yang S.T., Sellassie I.G., The effect of product bed temperature on the microstructure of Aquacoat-based controlled-release coatings. International Journal of Pharmaceutics 60 (1990) 109 - 124. 221. Yang, S.T., Savage, G.V., Weiss, J., Ghebre-Sellassie, I., The effect of spray mode and chamber geometry of fluid-bed coating equipment and other parameters on an aqueous-based ethyl cellulose coating. International Journal of Pharmaceutics 86 (1992) 247 - 257. 222. Yasuda, H. and Lamaze, C.E., Permselectively of solutes in homogenous water-swollen polymer membranes. Journal of Macromolecular Science Physic B5 (1971) 1303 - 1305. 223. Yu, L.X., and Amidon, G.L., Analytical Solutions to Mass Transfer. In: Amidon, G.L., Lee, P.I., Topp, E.M. (Eds) Transport Processes in Pharmaceutical systems. Marcel Dekker, New York 1999. 224. Yuasa, H., Nakano, T., Kanaya, Y., Suppression of agglomeration in fluidized bed coating I. Suppression of agglomeration by adding NaCl. International Journal of Pharmaceutics 158 (1997) 195 - 201. 225. Yuksel, N., Kanik, A.E., Baykara, T., Comparison of in vitro dissolution profiles by ANOVA-based, model-dependent and independent methods. International Journal of Pharmaceutics 209 (2000) 57 - 67. 226. Zentner, G.M., Cardinal, J.R., Feijen, J., Song, S., Progestin permeation through polymer membranes IV: mechanism of steroid permeation and functional group contributions to diffusion through hydrogel films. Journal of Pharmaceutical Science 68 (1979) 970 - 975. 227. Zhuang, G.H., Schwartz, J.B., Schnaare, R.L., Bead coating. I. Change in release kinetics (and mechanism) due to coating Levels. Pharmaceutical Research 8 (1991) 331 - 335. 228. Zingone, G., Rubessa, F., Release of carbamazepine from solid dispersions with polyvinylpyrrolidone/vinyl acetate copolymer (PVP/VA). STP Pharma Science 4 (1994) 122 – 127.
206 APPENDICES
VII. APPENDICES
Symbols a0 initial radius for a sphere or cylinder or the half-thickness for a slab a1 a constant incorporating structural and geometric characteristics of the dosage form A area through which diffusion occurs or effective area for mass transfer
Acs cross-sectional area of the film
Ai initial cross-sectional area of the sample AIC Akaike information criterion AUC area under the curve
Cb, Cs concentration of drug at drug-coating interface and the bulk respectively
Cd concentration of the drug
Cfs drug solubility in the liquid surrounding the matrix
Ci , Cm interior and media drug concentrations respectively
C0 initial concentration of the drug in the polymer matrix. 3 Cr concentration of the drug in the receptor fluid (g/cm )
Ce solubility of drug in dissolution medium
Cs solubility of drug in the matrix cw water concentration in the system
Dp displacement of the probe from point of contact to point of puncture D diffusion coefficient or diffusivity De Deborah number
Deff effective diffusion coefficient
Diw diffusion coefficient of drug through the pores
Dm diffusion coefficient across the membrane
Dr diffusivity of the drug through the membrane dC/dx gradient in concentration (C) in the x direction. dL/dm slope of the linear potion of the elastic deformation dM/dt rate of diffusion E’ storage modulus E” loss modulus
EL percent elongation at break (tensile test)
207 APPENDICES
EM elastic modulus EXC amount of extracted components, f1 difference factor f2 similarity factor ft fraction of drug released at any time t. F load required for puncture hm thickness of the membrane K partition coefficient
Kf filtration coefficient ke erosion rate constant
K0 zero order release rate constant
K1 first order release rate constant
KH Higuchi release rate constant
Ks a constant incorporating the surface–volume relation for Hixson and Crowell model K′ a constant related to the surface, the shape and the density of the particle
Kβ Hixson and Crowell release constant.
Is swelling index j flux per unit area J total flux li initial gauge length
Lmax maximum load M mass change
Md, Ma the respective amounts of diffusing substance in the donor and acceptor compartment when t = 0 MDT mean dissolution time n value in order to characterise different release mechanisms or number of dissolution data points N number of particles p number of the parameters of the model P permeability constant
Peff effective permeability coefficient
208 APPENDICES
Pm permeability coefficient of the polymer membrane Q amount of drug released in time t per unit area,
Q0 initial amount of drug in the solution
Qt amount of drug dissolved in time t, r radius of a sphere ro radius of the device r2 correlation coefficient R radius of the film exposed across the open screw cap
Rj percent dissolved of reference products at each time point j
Rtot total resistance
Rm membrane’s diffusional resistance
Rwvp water vapour permeation rate
R1, R2 aqueous resistance T absolute temperature t time tL lag time tx% parameter corresponds to the time necessary to the release of a determined percentage of drug
Td dissolution time
Tj percent dissolved of test products at each time point j
Ts tensile strength W total amount of the drug per unit volume of the matrix
Wd weight of the dried polymer film
We weight of the film that was dried to constant weight wi optional weighing factor
W0 initial amount of drug in the dosage form,
Ws weight of the swollen film
Wt amount of drug in the dosage form at time t WSSR weighed sum of square of residues v Poisson’s ratio
V volume of the receptor fluid
V1, V2 molar volumes
Vd, Va volumes of the donor and receptor compartments.
209 APPENDICES
Vm total volume of the mixture ε void fraction (porosity) τ tortuosity of the membrane θ one-half the angle of coalescence (contact angle)
θ d diffusion time η viscosity of the spheres β chain immobilization factor λ relaxation time ∂ % elongation (puncture test) δ solubility parameter
δc coating thickness
δd thickness of the diffusion layer
δs cross-head speed σ reflection coefficient of the coating ω work of failure φ volume fraction ∆π osmotic pressure difference
∆αcubic difference between the thermal cubical expansion coefficients of the coating and the tablet core ∆E energy of vaporization ∆E/V cohesive energy density (CED) ∆G Gibbs free energy of mixing ∆H heat of mixing
∆lb increase in length at break point ∆p vapour pressure difference ∆S entropy pf mixing ∆W amount of water vapour permeated through the film γ surface tension
γp puncture strength
210 APPENDICES
Published communications and presentations
The following have been published or presented in advance of this thesis.
PUBLICATIONS (PAPER)
1. Heng, P.W.S., Chan, L.W., Ong, K.T., Influence of storage conditions and type of plasticizers on ethyl cellulose and acrylate films formed from aqueous dispersions. Journal of Pharmacy and Pharmaceutical Science 6 (2003) 334- 344. 2. Chan L.W., Ong K.T., Heng P.W.S., Novel film modifiers to alter the physical properties of composite ethyl cellulose films. Pharmaceutical Research 22 (2005) 476-489.
PUBLICATIONS (POSTERS)
1. Heng, P.W.S., Chan, L.W., and Ong, K.T., The influence of storage humidity on pseudolatex films. American Association of Pharmaceutical Scientists Annual Meeting and Exposition, Canada, 2002
2. Ong, K.T., Chan, L.W., Heng, P.W.S. The influence of storage conditions on polymeric films. 11th International Pharmaceutical Technology Symposium on “Intelligent Drug Delivery Systems Better and Safer Therapy” Istanbul, Turkey (9-11 September 2002)
3. Heng, P.W.S, Er, D.Z.L., Ong, K.T., Chan, L.W., A Study of Composite Ethylcellulose Films. 31st Annual Meeting and Exposition of the Controlled Release Society, Hawaii, US, 12-16 June 2004.
4. K.T. Ong, Paul W.S. Heng and L.W. Chan. Influence of polyvinylpyrrolidone and its derivatives on properties of ethylcellulose films American Association of Pharmaceutical Scientists Annual Meeting and Exposition, Baltimore, US, 2004.
5. K.T. Ong, Paul W.S. Heng and L.W. Chan. Investigation on effect of storage conditions and type of plasticizers on drug release from ethylcellulose coated pellets. American Association of Pharmaceutical Scientists Annual Meeting and Exposition, San Antonia, US, 2006.
211