45 tayangan

Diunggah oleh Riyo U. Rahadi

sedikit tentang Drug Release

- tut16
- RevModPhys.57.437
- KF Wele Welw
- Solubility Product
- Mass Transfer
- Assessment of the Leachability and Mechanical Stability of Mud From a Zinc-plating Plant and Waste Zeolite Binding With Portland Cement
- Codigo Fuente Fortran
- 10.1016@j.chemolab.2016.07.011
- Hydrocarbon
- Mass Transfer and Diffusion
- Design of Optimal Process Flow Sheet for Fractional Crystallization
- TQ Science 7 1st Grading
- williams1934.pdf
- pub-mb_2820
- Battery1 BatteryDesignStudio RS
- 201210 Smit Help Ho
- Mercer 1990
- Chemistry Reviewer
- Fix
- Itraconazole E100

Anda di halaman 1dari 18

TARAVAT GAFOURIAN,1,2 AREZOO SAFARI,1 KHOSRO ADIBKIA,3 FATEMEH PARVIZ,1 ALI NOKHODCHI1,2 Drug Applied Research Center and School of Pharmacy, Tabriz University of Medical Sciences, Daneshgah Street, Tabriz 51664, Iran

2 3 1

Medway School of Pharmacy, Universities of Kent and Greenwich, Chatham, Maritimes, Kent ME4 4TB, England School of Pharmacy, Zanjan University of Medical Sciences, Zanjan, Iran

Received 20 June 2006; revised 15 February 2007; accepted 22 February 2007 Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/jps.20990

ABSTRACT: This investigation is aimed at characterization of the mode of release from two different substitution types of HPMC and the effect of chemical structure of drugs using the QSPR (Quantitative - StructureProperty Relationship) technique. To this end, release proles of HPMC matrices of several drugs containing the same formulation and compressed at a constant pressure were studied. QSPR method was used to establish statistically signicant relationships between release parameters and the structural descriptors. Structural descriptors consisted of molecular mechanical, quantum mechanical and graph-theoretical parameters, as well as the partition coefcient and the aqueous solubility of the drugs. The results showed that the most important factors determining the release prole from both HPMC K4M and HPMC E4M matrices were the aqueous solubility of drugs (which could be substituted efciently by dipole moment) and the size of the drug molecules. Comparison of drug release from matrices prepared using the two grades of HPMC showed very distinct differences for some drugs, as evaluated by the similarity factor. The results indicated that the source of the difference could be sought in the drug properties (as exemplied by the aqueous solubility and surface area) as well as the rate of erosion (that depends mainly on the polymer type). 2007 Wiley-Liss, Inc. and the American Pharmacists Association J Pharm

Sci 96:33343351, 2007

Keywords:

INTRODUCTION

A method of obtaining a sustained-release product is to embed or disperse the solid medicinal compound in an insoluble matrix by compression of a physical mixture of the compound and a polymeric material.1 This has attracted considerable attention and has been described by several

Correspondence to: Taravat Gafourian (Telephone: 441634-883846; Fax: 44-1634-883927; E-mail: t.ghafourian@kent.ac.uk; ghafourian_t@yahoo.com)

Journal of Pharmaceutical Sciences, Vol. 96, 33343351 (2007) 2007 Wiley-Liss, Inc. and the American Pharmacists Association

researchers.25 Matrix tablets have long been used to obtain sustained drug delivery and it was Higuchi who rst presented a detailed mathematical analysis of this release.6 Hyroxypropylmethylcellulose (HPMC) is one of the most widely used polymers in the preparation of oral controlled drug delivery systems. To achieve controlled release through the use of a water-soluble polymer such as HPMC, the polymer must quickly hydrate on the outer tablet skin to form a gelatinous layer. A rapid formation of a gelatinous layer is critical to prevent wetting of the interior and disintegration of the tablet core. Once the protective gel layer is formed, it controls

3334

3335

the penetration of additional water into the tablet. As the outer gel layer fully hydrates and dissolves, a new inner layer must replace it and be cohesive and continuous enough to retard the inux of water and control drug diffusion. Hydroxypropyl methylcellulose (HPMC) products vary chemically and physically. The major chemical differences are in degree of methoxyl substitution, moles of hydroxypropoxyl substitution (MS), and degree of polymerization. Varying ratios of hydroxypropyl and methyl substitution in different products inuence properties such as organic solubility and the thermal gelation temperature of aqueous solutions and swelling behavior.7 It has been shown that mechanism of release and the release proles from matrices depend not only on the type of the polymer but also on the properties of the drug.810 For example the release mechanisms of propranolol HCl and indomethacin from HPMC matrices are signicantly different.8,9 In a study on the release of drugs from polyvinylalcohol matrices, it was observed that the release rates of potassium chloride, phenylpropanolamine hydrochloride and bovine serum albumin decrease as the molecular size of the drug increases.2 Although the phenomenon was attributed to decreased diffusivity and molecular weight was taken as the criterion of molecular size, the study was not aimed at establishing any explicit relationship. In another investigation where molecular weights of the drugs classied in groups of roughly similar solubilities were compared, molecular weight and solubility were indicated as the possible factors affecting drug release from matrices.11 Baveja et al. investigated release characteristics of six watersoluble bronchodilators from HPMC K4M matrices in order to nd correlation between release rate and molecular geometry of the drugs.12 They showed that despite almost identical aqueous solubilities different drug molecules showed different release rates from HPMC matrices, which was related to the accessible surface area of the drugs. However, it must be stressed that as the drugs used in the study consisted of only structurally related beta-blockers, the ndings cannot be extrapolated to other drugs. Quantitative StructureProperty Relationship (QSPR) is a valuable tool that employs specialized statistical techniques to relate the property under investigation to the molecular structure of the chemicals represented by physico - chemical properties or structural descriptors. The resulting QSPR models facilitate understanding of the property in terms of chemical structure, ultimately

DOI 10.1002/jps

enabling the investigators to estimate the property for other similar compounds. The aim of the present investigation was to rationalize the release characteristics of different drugs from HPMC matrices in terms of the chemical structure of the drugs and the properties of the polymer. To achieve this goal, separate QSPR models were established for drug release from HPMC K4M or HPMC E4M matrices and the resulting models were compared. The models were obtained using the release data of 15 drugs (belonging to different chemical classes) from matrices prepared using HPMC K4M and E4M. Chemical structure of the drugs were represented by a wide range of molecular descriptors such as partition coefcient (log P), pKa, atomic charges, orbital energies, dipole moment, length, surface area and electrostatic potentials on the surface, molecular connectivity indexes and shape indices, solubility and solubility parameter.

Materials Acetaminophen (Acros Organic, UK), diclofenac sodium (Sobhan Co., Iran), uoxetine HCl (ParsDaru, Iran), naproxen (Pars-Daru, Iran), piroxicam (Sigma, USA), propranolol HCl (Acros Organic, UK), sulfamethoxazole (Logman, Iran), diltiazem HCl (Acros Organic, UK), ibuprofen (Acros Organic, UK), atenolol (Daru-Pakhsh, Iran), diphenhydramine HCl (Acros Organic, UK), imipramine HCl (Logman, Iran), theophylline monohydrate (Acros Organic, UK), triuoperazineHCl (Sobhan Co., Iran) and trimethoprim (Logman, Iran) were obtained. Two HPMC (Hypromellose) grades, HPMC (Methocel) K4M Premium CR and HPMC (Methocel) E4M Premium CR were gifts from Colorcon, UK. Preparation of Tablets The matrices were prepared by mixing 20 g of the drug (with particle size in the range of 45125 mm) with 20 g of HPMC K4M or HPMC E4M for 10 min using a small double cone mixer. Magnesium stearate was then added to the mixture and mixed for a further 1 min. The nal mixtures were compressed on an 8-mm punch and die using single punch machine (Erweka, Germany) at a constant pressure of 10 kN. The weight of each tablet was 202 mg which included 100 mg drug (49.5%), 100 mg HPMC K4M (49.5%) and 1% magnesium stearate.

3336

NOKHODCHI ET AL.

Dissolution Studies The united state pharmacopoeia (USP) basket method (Erweka, DT 6R, Heusenstamm, Germany) was used for all the in vitro dissolution studies. In this method, phosphate buffer (pH 6.8) without enzyme, was used as dissolution medium. The buffer was prepared according to United States Pharmacopoeia by adding 50 mL of 0.2 M monobasic potassium phosphate and 22.4 mL of 0.2 M sodium hydroxide into a 200 mL volumetric ask and adding water to volume.13 Matrices were placed in 900 mL of the dissolution medium and maintained at 37 0.1 8C for 8 h at pH 6.8. The amount of drug was 100 mg in all formulations. The rate of stirring was 100 2 rpm. At appropriate intervals (15, 30, 60, 90, 120, 180, 240, 300,

360, 420, and 480 min), 5 mL of samples was taken and ltered through a 0.45 mm Millipore lter. The dissolution media was then replaced by 5 ml of fresh dissolution uid to maintain a constant volume. The samples were then analyzed by ultraviolet/visible spectrophotometer at the maximum absorption wavelengths (see Fig. 1). The mean of three determinations was used to calculate the drug release rate from each of the formulations.

Dissolution Parameters The parameters obtained from dissolution studies were the time required for 50% of drug release (T50%), percentage of drug released after 8 h (Q8 h),

Figure 1. Chemical structures and the lmax wavelengths of the drugs used in this study.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 12, DECEMBER 2007 DOI 10.1002/jps

3337

release rates obtained using various kinetic models (k), and dissolution efciency (DE8). Naproxen, piroxicam, triuoperazine, and trimethoprim showed less than 50% release during the 8 h of dissolution studies. The slope and the intercept of the zero-order release were calculated for these formulations (correlation coefcient above 0.99) and the results were used for the estimation of the time required for the release of 50% drug. The dissolution efciency (DEt) of a pharmaceutical dosage form is dened as the area under the dissolution curve up to the time, t, expressed as the percentage of the area of the rectangle (Eq. 1).14

t R

photometer (Shimadzu, Japan). The mean of three determinations for each drug was reported.

Structural Parameters A total of 75 structural descriptors for each drug were obtained from various software packages. Table 1 gives a summary of the descriptors calculated for the drugs and the software used. The COSMIC force eld was used for energy minimization prior to molecular mechanical parameter calculations. For calculation of molecular orbital parameters, the three-dimensional structures of the drugs were minimized using the MNDO Hamiltonian in MOPAC version 7.0 (QCPE, Department of Chemistry, Indiana University, 800 East Kirkwood Ave., Bloomington, IN 47405-7102). The structural descriptors obtained from MOPAC consisted of Atomic charges, orbital energies, dipole moment, length, and principle moments of inertia. SMILES strings were entered into the MOLCONN-Z software. MOLCONN-Z was used to calculate graph theoretical descriptors. ACD/Log D Suite release 7.0 was used to obtain log P, log D at pH 6.8, and pKa. The calculated values were only used when the experimental values were not available in the database. The fractions of drugs that are ionized to anions (A) or cations (B), and the unionized fraction of drugs (fu) at pH 6.8 were calculated using Eqs. (6)(8). Note that only the rst acidic pKa and the rst basic pKa were considered in the calculations. fiB fiA 1 1 anti log6:8 pKa 1 1 anti logpKa 6:8 (6)

yd t 100% (1)

DEt

y100 t

where y is the percent of drug dissolved at time t. Eq. (1) was used to calculate the dissolution efciency of the matrices where t was 8 h.

Kinetic Models In order to obtain the rate of release, the release data from the matrices were tted to the following mathematical models: zero-order kinetic (Eq. 2), rst-order kinetic (Eq. 3), square-root of time equation (Higuchi equation, Eq. 4), and Peppas equation (Eq. 5). Q kt ln 100 Q ln Q0 kt Q kt1=2 Q ktn (2) (3) (4) (5)

In Eqs. (2)(5), Q is the percent of drug released at time t and k is the coefcient of the equations. In Eq. (3)Q0 equals 100. In Eq. (5), k is the constant incorporating structural and geometric characteristics of the release device and n is the release exponent indicative of the mechanism of release. Solubility Measurements Solubility measurements were carried out in phosphate buffer pH 6.8 at 37 18C. Saturated solutions were prepared by adding excess amounts of drugs to water and shaking on a shaker bath (Velp, Italy) for 48 h under constant vibration. After this period, the solutions were ltered, diluted and analyzed by UV spectroDOI 10.1002/jps

(7) (8)

fu 1 fiB 1 fiA

Solubility parameter and energy of vaporization was calculated using a group contribution method.15

Development of QSPRs The QSPR endpoints were dissolution parameters discussed earlier under dissolution parameters, as well as the percentage released at different times. Stepwise regression analysis was used to determine statistically signicant relationships between structural parameters discussed in latter

3338

NOKHODCHI ET AL.

Method

Nemesis

Parameter

Solvent accessible surface area (SA), dipole moment calculated by the Charge-2 method (m), the highest and the lowest electrostatic potentials on the solvent accessible surface (ESP and ESP, respectively) The energies of the highest occupied and the lowest unoccupied molecular orbitals (EHOMO and ELUMO, respectively), dipole moment (m), the highest and the lowest atomic charges in the molecule (Q and Q), principle moments of inertia (IM), length of the molecule (L), molecular weight (MW) Simple and valence corrected molecular connectivity indexes including zero- through 4 4 v 4 v fourth-order path (0xp4xp and 0 xv p xp ), fourth order path-cluster ( xpc and xpc ),

4 v third- and fourth-order cluster (3xc, 4xc, 3 xv c and xc ), and fth- and sixth-order chain (5xch and 6xch), the highest atomic electrotopological index (S(I)), molecular shape indexes (0k 3k and 0ka3ka), and delta connectivity indexes (X1X3) Partition coefcient (log P), distribution coefcient at pH 6.8 (log D6.8), rst acidic and rst basic pKa values, the fraction unionized at pH 6.8 (fu), polarizability (a), molecular volume (V), molar refractivity (MR) and parachor (PA) Aqueous solubility (mg/mL) measured at 378C, in phosphate buffer pH 6.8

ACD/log D

The total number of oxygen and nitrogen atoms (NNO), number of hydrogen atoms connected to oxygen or nitrogen (NH), number of double bonds (N ), number of rotatable bonds (Nrotat), total number of bonds (Nbond), number of aromatic carbon atoms (Carom), number of aliphatic carbon atoms (Calip), logarithms of molecular surface area and weight (log SA and log MW), 1/3 and 1/2 power of molar volume (V1/3 and V1/2), energy of vaporization and solubility parameter (Ev and d)

section and the dissolution parameters. Distribution of the dissolution parameters and the logarithmically transferred dissolution parameters were studied and those with the least deviation from normality were used in stepwise regression analyses. The statistical analyses were performed using the MINITAB (release 13.1) statistical software. In order to avoid the risk of chance correlations, loss of interpretability and predictability, the number of parameters in the models was kept as low as possible. Accordingly, only two parameters were allowed in stepwise regression analysis. The following statistical details of the models were noted: n, the number of observations; R2, the squared correlation coefcient; s, the standard deviation; F, the Fisher statistic; and the p value. The gures in brackets with the regression coefcients were standard errors of coefcients.

where, Y is discriminant score, that is, the dependent variable, x1xn represent the specic descriptors, and b are corresponding to weights associated with the respective descriptors. Stepwise LDA was performed using the software TSAR (Accelrys Software Inc.).

Various factors could be accounted for the drug release mechanism from hydrophilic matrices.16 These include the geometry of the matrix,17 particle size of polymer and matrix swelling ratio (which depends on the HPMC type and controls water and drug diffusion coefcients),18,19 polymer and drug concentrations,20,21 chain length and degree of substitution of the HPMC,19 as well as the drug characteristics.22,23 An additional factor that might affect the drug release prole is the constituents of the dissolution media. For example, it has been suggested that phosphate mitigates the hydration of HPMC compacts, an interaction that is further amplied if the compact

DOI 10.1002/jps

Linear Discriminant Analysis (LDA) The basic theory of LDA is to classify the dependent by dividing an n-dimensional descriptor space into two regions that are separated by a hyperplane dened by a linear discriminant

3339

incorporates diclofenac sodium.24 However, it was shown that release of chlorpheniramine maleate and theophylline from HPMC matrices was not signicantly different in phosphate buffer compared to water medium.25 In this study, the matrices consist of the same weight percentages of the drug and HPMC E4M or HPMC K4M, compressed under the similar pressure, with the same die geometry. Furthermore, the same dissolution medium has been used throughout the investigation. Therefore, the release proles from matrices prepared using the same polymer type are expected to be related to the intrinsic

properties of the drug and the interaction between drug and the polymer. Moreover, physicochemical characteristics of drug alone should be able to determine which of the release mechanisms, polymer swelling, dissolution/erosion at the matrix periphery and drug diffusion is the governing process in drug release. Figures 2 and 3 show the dissolution characteristics of HPMC matrices of these drugs. The gures show different release proles from HPMC matrices containing the same polymer type for different drugs. Following the QSPR approach in this study, the relationships were sought between

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 12, DECEMBER 2007

3340

NOKHODCHI ET AL.

measured release characteristics and the structural parameters of the drugs. The drugs belonged to various chemical classes covering a broad range of water solubility values, as well as acidic and basic properties (Table 2).

QSPR Models for Release Parameters (Rate Constants and Model Independent Parameters) Various parameters have been employed in order to evaluate and compare drug dissolution proles from pharmaceutical formulations. Among the

most widely used parameters are release rate constants from different kinetic models,26,27 and model independent parameters such as difference ( f1) and similarity ( f2) factors,28,29 dissolution efciency,14 mean dissolution time,30 and percent released after a given time period.31 In this study, several informative parameters including kinetic and model independent release parameters were calculated from dissolution data. These included release rates, dissolution efciency, the amount of drug released after 8 h, and the time required for the release of 50% of drug. Table 3 shows the release rate constants calculated using different

DOI 10.1002/jps

3341

Table 2. Water Solubility (mg/mL), pKa, Molecular Volume (V) and Molecular Weight (MW) Values of the Drugs Used in the Study

Drug

Acetaminophen Diclofenac sodium Fluoxetine HCl Naproxen Piroxicam Propranolol HCl Sulfamethoxazole Diltiazem HCl Ibuprofen Atenolol Diphenhydramine HCL Imipramine HCl Theophylline monohydrate Triuoperazine HCl Trimethoprim

a b

Acidic pKa

9.86 4.01b 4.84a 5.07b 13.84a 5.81a 4.41a 13.88a 8.54b

a

Basic pKa

1.72 2.26a 9.62b 2.33b 9.47b 1.39a 8.06b 9.16a 9.02b 9.4b 1.7a 8.08b 7.12b

a

22.1 17.76 50.0 0.193 0.172 254 0.51 1071 0.23 24.75 1010 750 8.6 610 0.4

Va

120.9 206.8 266.7 192.2 211.9 237.1 173.1 327.6 200.3 236.6 249.2 269.2 112 328.7 231.8

MW

151.2 296.1 309.3 230.3 331.3 259.3 253.3 414.5 206.3 266.3 255.4 280.4 182.2 407.5 290.3

kinetic models. Comparing the correlation coefcients (r2 values) of different kinetic models in Table 3, it becomes evident that no single kinetic model can be selected as the best model for all the drugs. For example, the release of propranolol hydrochloride best follows a Higuchian, Peppas or rst order pattern whereas uoxetine hydrochloride follows a zero-order release pattern best. As the kinetic constants from different models could not be compared with each other, the release rates from each of the kinetic models were separately correlated against structural descriptors of the drugs using stepwise regression analysis and the QSPR models were developed. The results of these analyses have been presented in Table 4 for HPMC K4M and in equation 14 for HPMC E4M. In stepwise regression analyses, water solubility was the rst parameter that entered the models. In step 2 of the analyses, molecular volume (V) entered the models for release from HPMC K4M. Therefore, it can be concluded from stepwise regression analysis that the most important molecular property increasing the release rate from both HPMC E4M and HPMC K4M matrices is water solubility (SW) of the drugs. Furthermore, the larger molecular volume (V) of the drugs results in the smaller release rates from HPMC K4M matrices. It must be noted that in case of HPMC E4M matrices, the only signicant QSPR for release rates was obtained for the release rate constant of Peppas kinetic

DOI 10.1002/jps

model (Eq. 14). log k 0:35 0:10 log SW 0:04L n 11 s 0:101R2 0:748 F 11:9 p 0:004 14

In Eq. (14), length of the drug molecules (L) is the second descriptor to be selected by stepwise regression. Model independent dissolution parameters of DE8, T50%, and Q8 h for different drugs are listed in Table 5. The relationships between these parameters and physicochemical/structural descriptors (QSPR models) obtained from stepwise regression analysis have been presented in Table 6. The QSPR models in Table 6 conrm the conclusion that drug release from HPMC K4M and E4M matrices is primarily controlled by water solubility. Moreover, in QSPRs for dissolution of HPMC K4M matrices (Eqs. 1517), the second parameter is either V or 1xp, with the latter being the rst order molecular connectivity index, mostly dening the size of the molecule.32 However, in QSPRs for dissolution of HPMC E4M matrices (Eqs. 1820), the second most signicant parameter describing the dissolution is the length of the molecule. The effect of drug solubility on release rate is well documented.9,10,33,34 The theoretical basis can be explained using Ficks rst law (Eq. 21).

3342

NOKHODCHI ET AL.

Table 3. Release Rate Constants of HPMC Matrices Obtained From Different Kinetic Models; The Corresponding Squared Correlation Coefcients are Reported in the Parentheses

0.471 (0.992) 0.181 (0.967) 0.181 (0.964) 0.05 (0.924) 0.073 (0.994) 0.276 (0.997) 0.115 (0.998) 0.36 (0.988) 0.179 (0.797) 0.123 (0.967) 0.402 (0.993) 0.248 (0.966) 0.343 (0.974) 0.075 (0.994) 0.046 (0.997) 0.967 0.978 0.966 0.946 0.376 (0.996) 0.213 (0.936) 0.193 (0.966) 0.089 (0.971) 0.13 (0.989) 0.36 (0.992) 0.14 (0.997) 0.366 (0.999) 0.093 (0.951) 0.205 (0.936) 0.475 (0.994) 0.267 (0.989) 0.322 (0.989) 0.098 (0.977) 0.057 (0.990) 0.504 (0.993) 0.311 (0.966) 0.32 (0.965) 0.134 (0.926) 0.179 (0.99) 0.349 (0.998) 0.247 (0.986) 0.406 (0.996) 0.303 (0.809) 0.247 (0.966) 0.385 (0.994) 0.347 (0.971) 0.417 (0.989) 0.188 (0.973) 0.128 (0.965) 0.109 0.115 0.073 0.031 0.030 0.104 (0.807) (0.984) (0.804) (0.959) (0.992) (0.955) 0.104 (0.991) 0.102 (0.992) 0.047 (0.975) 0.067 (0.944) 0.111 (0.993) 0.0999 (0.953)

0.294 (0.987) 0.188 (0.856) 0.063 (0.890) 0.017 (0.970) 0.016 (0.991) 0.141 (0.985) 0.135 (0.833) 0.105 (0.956) 0.028 (0.991) 0.055 (0.981) 0.139 (0.917) 0.111 (0.997) 0946

0.368 0.242 0.277 0.096 0.098 0.297 (0.978) (0.999) (0.949) (0.991) (0.97) (0.996) 0.244 (0.979) 0.208 (0.989) 0.140 (0.992) 0.285 (0.980) 0.225 (0.976) 0.256 (0.995) 0.978

Zero Order

Higuchi

0.392 (0.997) 0.394 (0.994) 0.264 (0.918) 0.108 (0.987) 0.100 (0.965) 0.361 (0.995) 0.350 (0.966) 0.346 (0.976) 0.161 (0.996) 0.232 (0.988) 0.374 (0.982) 0.348 (0.997) 0.970

Acetaminophen Diclofenac sodium Fluoxetine HCl Naproxen Piroxicam Propranolol HCl Sulfamethoxazole Diltiazem HCl Ibuprofen Atenolol Diphenhydramine Imipramine HCl Theophylline Triuoperazine HCl Trimethoprim

0.224 (0.998) 0.102 (0.942) 0.09 (0.999) 0.04 (0.937) 0.052 (0.984) 0.123 (0.963) 0.073 (0.991) 0.16 (0.987) 0.106 (0.896) 0.071 (0.942) 0.149 (0.948) 0.126 (0.996) 0.164 (0.975) 0.056 (0.990) 0.038 (0.998)

Mean R2 values

0.970

DOI 10.1002/jps

3343

Table 4. QSPR Models for Release Rate Constants from HPMC K4M Matrices: a, b and c are Coefcients in the Equation, log k a b. log SW cV, where k is the Release Rate Constant, Sw is Water Solubility, and V is the Molecular Volume, n is the Number of Drugs, R2 is the Correlation Coefcient, s is Standard Deviation, and F is the Fisher Statistic (Ratio of Variance Between Groups to the Overall Variance in Analysis of Variance)

Eq.

10 11 12 13

Kinetic Model

Zero order First order Peppas Higuchi 0.584 0.210 0.296 0.232

a

(0.147) (0.202) (0.150) (0.115) 0.160 0.231 0.207 0.123

b

(0.031) (0.042) (0.031) (0.024)

c

0.00286 0.00380 0.00301 0.00212 (0.0007) (0.0010) (0.0007) (0.0006)

n

15 15 15 15

R2

0.704 0.719 0.784 0.693

s

0.137 0.189 0.140 0.108

F

14.3 15.3 21.8 13.5

p

0.001 0.000 0.000 0.001

Accordingly, when the drug is poorly watersoluble, dissolved and non-dissolved drug are both present within the matrix, from which only the dissolved drug can diffuse into the dissolution media. In other words, the concentration gradient, being smaller for such drugs, results in a reduced diffusion according to Ficks rst law.35 J D dC dx (21)

In Eq. (21), J is the ux (the amount of material owing through a unit cross-section of a barrier in unit time), D is the diffusion coefcient, dC/dx is the concentration gradient. With concentration gradient being dependent on water solubility, the second parameter in Ficks law (D) is expected to be controlled by structural characteristics of the drug. Diffusion coefcient has been theoretically

related to the radius of the diffusing particle in StokesEinstein equation for diffusion of a spherical particle within a continuous uid.35 This has been used to justify the inverse proportionality between diffusion constant and cross sectional area of solute,36 or between diffusion constant and cube root of volume, V1/3 for diffusions through skin.37 Although diffusion in polymers does not obey the Stokes Einstein equation, it is still reasonable to expect that the larger the diffusing molecule the more difcult its movement within the HPMC gel medium will be. This could explain the negative effect of molecular size (volume or length terms) observed in the QSPR models (Eqs. 1020) on the release rate or efciency. Although all of the equations (Eqs. 1020) are statistically signicant ( p < 0.05), the statistical quality of the equations can be compared based on

Table 5. Dissolution Efciency (DE8), Time Required for 50% of Drug Release (T50%), and the amount Released after 8 h (Q8 h) of HPMC Matrices

Acetaminophene Diclofenac sodium Fluoxetine HCl Naproxen Piroxicam Propranolol HCl Sulfamethoxazole Diltiazem HCl Ibuprofen Atenolol Diphenhydramine HCL Imipramine HCl Theophylline monohydrate Triuoperazine Trimethoprim

DOI 10.1002/jps

HPMC E4M Q8

h

DE8

75.98 57.31 44.12 21.23 28.52 66.38 35.71 70.10 42.29 51.38 78.61 61.60 70.97 26.01 16.60

T50%

1.6 3 4.8 10.6 9.1 1.9 5.8 1.8 5 3.6 1.1 3.2 2.1 11 12.8

DE8

61.91 58.69 54.71 19.25 17.01 62.83 53.36 50.47 28.26 48.53 54.63 58.31

T50%

1.0 3 1.9 15.8 14.1 2.5 3.5 4.35 8.5 3.5 3 2.8

Q8

99.42 99.23 79.79 39.04 47.51 96.27 61.16 96.92 92.65 82.10 97.57 98.38 98.18 44.77 31.42

100.19 98.64 71.55 28.44 30.28 92.95 96.13 86.07 44 68.69 94.72 87.08

3344

NOKHODCHI ET AL.

HPMC Type

the value of R2. R2 shows the percentage of the variation in Y (dependent variable) that can be explained by the independent variables (aqueous solubility, volume or length). Other variations in Y can be due to the factors other than those investigated here, or the experimental error. In the QSPR models (10)(20), R2 values vary between 0.548 and 0.792 ( p values below 0.028). Generally, the QSPR models for Peppas release constants (Eqs. 12 and 14) with R2 values of 0.784 ( p 0.000) and 0.748 ( p 0.004) are statistically superior to the models for other release rate constants. A particularly good quality model is Eq. (15) for the estimation of DE8 with R2 value of 0.792 ( p 0.000).

R2

Structure - Based QSPR Models for Release Parameters The QSPR models (10)(20), are able to predict different release endpoints using water solubility of drugs and the molecular volume/length. However it is most desirable to be able to estimate release properties using an entirely structure-based model, without the need to obtain drug solubility. Such QSPR models were obtained from stepwise regression analysis between drug release parameters and structural descriptors (Tables 7 and 8). Table 7 shows that release rates are correlated with dipole moment (m) and a size term that is cube root of molar volume (V1/3), molecular weight (MW), or length of molecules (L). The QSPRs in Table 8 also contain dipole moment and a size parameter (mostly MW). It is possible that in these equations dipole moment is merely describing the drug solubility. In order to test this hypothesis, the structural descriptors that are related to solubility were explored using stepwise regression analysis, from which Eq. (32) resulted. Eq. (32) shows that dipole moment is the main structural parameter describing the water solubility of this set of drugs. The second term in this equation is fraction of drug molecules that are in anionic form at pH 6.8. The negative sign of A shows that acidic drugs (which ionize to anions) are less water soluble than the basic drugs in the set (see Fig. 4). log SW 0:3740:31 0:1270:021m 1:57 0:32fiA n 15 r2 0:879 s 0:547 F 43:4 P 0:000 (32)

Table 6. QSPR Models for Model Independent Dissolution Parameters of HPMC Matrices

Eq.

15 16 17 18 19 20

DE8 701 (88) 123 (18)log SW 2.01 (0.43)V log T50% 0.003 (0.205) 0.239 (0.043) log SW 0.00391 (0.0010)V Q8 h 127 (19.5) 14.4(3.4) log SW 7.07 (2.17) 1xp DE8 108 (24.6) 8.74 (3.30)log SW 5.98 (2.12)L T50% 5.13 (4.78) 2.67(0.64) log SW 1.20 (0.41)L Q8 h 126 (31.8) 12.3 (4.27)log SW 5.83 (2.74)L

QSPR model

15 15 15 12 12 12

DOI 10.1002/jps

DOI 10.1002/jps

Table 7. QSPR Models for Drug Release Rates using Only the Structural Parameters

Eq.

15 15 15 15 11 0.556 0.556 0.681 0.581 0.705 0.168 0.237 0.178 0.126 0.109 7.5 7.5 7.8 8.3 9.6 0.008 0.008 0.004 0.005 0.008

Kinetic Model n R2 S F p

QSPR Model

HPMC Type

K4M K4M K4M K4M E4M

22 23 24 25 26

k 0.436 (0.52) 0.292 (0.092) V1/3 0.0276 (0.0076)m k 1.11 (0.73) 0.384 (0.130) V1/3 0.0399 (0.011)m k 2.00 (0.78) 0.804 (0.23) V1/3 0.0540 (0.012) m 0.132 (0.053) L k 0.539 (0.386) 0.221 (0.069) V1/3 0.0221 (0.0057)m k 0.566 (0.156) 0.0231 (0.0053)m 0.00145 (0.0006) MW

Table 8. QSPR Models for Model Independent Drug Release Parameters using Only the Structural Parameters

Eq.

QSPR Model

n

15 15 15 12 12

R2

0.538 0.577 0.601 0.572 0.628

S

122.4 0.237 17.4 15.4 3.33

F

7.0 8.2 9.0 6.0 7.6

p

0.010 0.006 0.004 0.022 0.012

HPMC Type

K4M K4M K4M E4M E4M

27 28 29 30 31

DE8 631 (129) 17.9 (5.19)m 1.53 (0.52) MW log T50% 0.263 (0.198) 0.0261 (0.009)m 0.398 (0.120) 3xc Q8 h 115 (18.3) 2.89 (0.73) m 0.247 (0.073) MW DE8 96.5 (20.1) 2.22 (0.74)m 7.84 (2.52) 1xp T50% 0.11 (3.9) 0.613 (0.160)m 0.0461 (0.017) MW

3345

3346

NOKHODCHI ET AL.

QSPR models for the percentage of drug released The second approach in this investigation was the incorporation of all the release data in the form of percentage released (Q) at various release times in the QSPR analysis. Table 9 is a list of the QSPR models obtained from stepwise regression of Q against time and the structural descriptors of drugs. Different forms of Q time relationships, adopted from Peppas, Higuchi, zero order, and rst order kinetic models, were examined. Comparing the statistics of the resulting QSPR models (Table 9), it can be seen that drug release from both HPMC K4M and HPMC E4M matrices best follows the Peppas kinetic model. This is of course expected due to the exibility of Peppas model that can t into the release proles of both diffusion, and erosion controlled release mechanisms. Figures 5 and 6 show the plot between the observed and calculated percentage released by Eqs. (33) and (37), respectively. The QSPRs show that apart from the time, drug aqueous solubility is the most important parameter in determining the percentage released from matrices prepared from both grades of HPMC. However, it is noted that coefcients of log SW are higher in QSPRs for release from HPMC K4M matrices in comparison with those in QSPRs for release from HPMC E4M matrices. In other words, solubility of drugs has less effect on the release from HPMC E4M. Moreover, coefcients of time are also higher for HPMC K4M, indicating a generally higher drug release rates from matrices prepared with this HPMC grade. In order to make a better comparison of the coefcients of time and log SW in QSAR models for release from HPMC E4M and HPMC K4M, the same variables and the same drug set were used for the development of the

models. Therefore, SA (surface area) rather than V (volume) and the same set of drugs as those in Eq. (37) for release from HPMC E4M was used for the correlation of the amount released from HPMC K4M. The resulting equation below for release from HPMC K4M still has a higher log t and log SW coefcients than Eq. (37), but a smaller (absolute) coefcient of SA. log Q 1:57 0:053 0:622 0:021 log t 0:148 0:008 log SW 0:00147 0:0002SA (41) n 132 R2 0:0907 s 0:111 F 418:6 p 0:000 It should be emphasized that the nding pertains to average release proles of all the drugs and some drugs may show a higher release rate from the HPMC E4M matrices. For example, according to Table 3, diclofenac sodium, uoxetine HCl, naproxen and ibuprofen have higher Peppas release rates from HPMC E4M matrices than from HPMC K4M matrices. It is generally thought that HPMC K4M is a more efcient retardant than HPMC E4M.38 However, in a comparative study on the release of propranolol HCl from HPMC matrices, it has been shown that the order of release rate from different grades of HPMC depends dramatically on the drug: Polymer ratio. At low polymer: Drug ratio HPMC K4M has a lower drug release rate. However, at higher polymer: drug ratio of 1:1 HPMC E4M becomes more retardant than HPMC K4M.39 A stepwise linear discriminant analysis was performed in order to nd structural descriptors that are able to classify drugs correctly into two groups of 0 for

Figure 4. Regression plot of log SW against the log SW calculated by Eq. 32.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 12, DECEMBER 2007 DOI 10.1002/jps

3347

Table 9. QSPR Models for the Amount of Drug Released at Different Time Intervals (Q) from Matrices Prepared using two Grades of HPMC; t is time, SW is Solubility, V is Molar Volume, and SA is Solvent Accessible Surface Area

HPMC Type

which release from HPMC E4M matrices is faster than release from HPMC K4M and 1 for which release is faster from HPMC K4M matrices. This resulted in the following discriminant axis with 83% correct classication: Y 0:236SA 0:972 log fu 0:706 confidenceestimate 0:833 (42)

p F

log Q 1.68 (0.045) 0.645 (0.023) log t 0.177 (0.009) log SW 0.00271 (0.0002)V Q 26.3 (4.43) 30.2 (1.23)t1/2 13.8 (0.81) log SW 0.222 (0.02)V Q 47.2 (4.24) 8.78 (0.37) t 13.8 (0.84) log SW 0.222 (0.02)V ln(100 Q) 3.26 (0.220) 0.303 (0.019) t 0.457 (0.044) log SW 0.0089 (0.001)V log Q 1.82 (0.057) 0.552 (0.022) log t 0.132 (0.009) log SW 0.00239 (0.00022) SA Q 38.8 (6.1) 27.5 (1.38)t1/2 9.83 (0.87) log SW 0.200 (0.022) SA Q 57.9 (6.03) 7.99 (0.42)t 9.84 (0.90) log SW 0.200 (0.022) SA ln(100 Q) 1.56 (0.100) 0.094 (0.007) t 0.095 (0.0148) log SW 0.0020 (0.0003) SA

The analysis shows that drugs with larger surface area and percentage unionized at pH 7.4 are more likely to have faster release rates from HPMC K4M matrices than from HPMC E4M matrices. The two grades of hydroxypropylmethylcellulose, HPMC K4M and E4M share a similar molecular weight but differ in degrees of substitution, with the methoxyl percent being 22 for K4M and 29 for E4M. Hydroxypropyl percent for K4M and E4M grades is very similar (8.1 and 8.5, respectively). In order to further investigate the effect of polymer type on the release kinetics, the similarity factor ( f2) was calculated using Eq. (43) and matrices of the same drug prepared with HPMC E4M or HPMC K4M were compared (ca. Table 10). f2 50 log 8" 9 #0:5 < = m X E4Mj K 4Mj 2 100 1 1=m : ; j1 (43) Average f2 for all the drugs was 38.04 showing differences between matrices prepared from HPMCs E4M and K4M. Table 10 shows that not all the drugs are liberated with a different prole from HPMC K4M and E4M matrices. The most different release prole is observed for acetaminophen with f2 value of 20. On the other hand, the release proles of diclofenac sodium from the two matrices are very similar with f2 value of 76. Differences between release proles of the same drug from matrices prepared using different grades of HPMC could be determined by diffusion (if soluble) through the gel and by the rate of tablet erosion. Also listed in Table 10 is Peppas coefcient n calculated for matrices of HPMC E4M and K4M. The table shows that average n is higher for matrices prepared from HPMC K4M, indicating higher contribution of erosion in the release process from these matrices in comparison with the matrices prepared from HPMC E4M.

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 12, DECEMBER 2007

DOI 10.1002/jps

33 34 35 36 37 38 39 40

Peppas Higuchi Zero order First order Peppas Higuchi Zero order First order

R2

3348

NOKHODCHI ET AL.

Figure 5. Plot of observed vs. calculated log Q (logarithm of percentage released) from HPMC K4M matrices according to Eq. (33).

From the above discussion it can be deduced that the overall lower release kinetics observed for HPMC E4M in comparison with HPMC K4M should be attributed to the lower erosion of the corresponding matrices. The lower erosion of these matrices implies a higher dependence of the release process on diffusion, a process requiring small drug molecular size, which is conrmed by the larger coefcient of SA in Eq. (37) in comparison with that in Eq. (41). Figure 7 shows that the higher the n of the HPMC E4M matrices, the higher the similarities of the matrices prepared from the two HPMC grades. The deviations from

this relationship (error) observed in Figure 7 is likely to be due to the varying contribution of diffusion and the effect of drug solubility that is not accounted for in the graph.

CONCLUSION

The process of drug release from HPMC matrices involves the two routes of erosion of the matrix, and diffusion of the dissolved drug through the HPMC gel, each to a varying extent depending on the properties of the drug as well as the polymer

Figure 6. Plot of observed vs. calculated log Q (logarithm of percentage released) from HPMC E4M matrices according to Eq. (37).

JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 12, DECEMBER 2007 DOI 10.1002/jps

3349

Table 10. Similarity Factor between Matrices Prepared Using HPMC E4M and K4M and Peppas n Coefcients for the Matrices

Drug

Acetaminophen Diclofenac sodium Fluoxetine HCl Naproxen Piroxicam Propranolol HCl Sulfamethoxazole Diltiazem HCl Ibuprofen Atenolol Diphenhydramine HCL Imipramine HCl Theophylline monohydrate Triuoperazine HCl Trimethoprim Average

f2

20.36 76.03 43.57 69.73 48.61 67.38 39.85 48.95 32.10 29.25 58.68 47.04

n (E4M) n (K4M)

0.560 0.683 0.532 0.536 0.469 0.595 0.560 0.643 0.549 0.424 0.630 0.659 0.644 0.676 0.579 0.657 0.604 0.496 0.703 0.505 0.908 0.777 0.406 0.546 0.623 0.714 0.783 0.647

of HPMC K4M and HPMC E4M matrices revealed that the matrices perform differently in most cases. Comparison of the QSPRs for release from two types of HPMC matrices led to the conclusion that HPMC E4M matrices have a lower tendency to erode resulting in overall inferior drug release kinetics for some drugs from this grade of HPMC in comparison with HPMC K4M. Furthermore, physicochemical properties of the drugs (surface area and percentage unionized at pH 6.8) determine whether the release is faster from HPMC K4M or from HPMC E4M.

REFERENCES

1. Christenson GL, Dale LB. 1962. Sustained release tablets. U.S. Pat. 3065143. 2. Korsmeyer RW, Gurny R, Doelker E, Buri P, Peppas NA. 1983. Mechanism of solute release from porous hydrophilic polymers. Int J Pharm 15: 2535. 3. Lapidus H, Lordi NG. 1966. Some factors affecting the release of a water-soluble drug from compressed hydrophilic matrices. J Pharm Sci 55: 840843. 4. Huber HE, Christenson GL. 1968. Utilization of hydrophilic gums for the control of drug substance release from tablet formulations II. Inuence of tablet hardness and density on dissolution behaviour. J Pharm Sci 57:164166. 5. Buri P, Doelker E. 1980. Formulation des comprimes a liberation prolongee. II. Matrices hydrophilies. Pharm Act Helv 55:189197.

38.04

0.570

grade. QSPR method employed in this investigation was able to outline the major drug properties responsible for the control of release from HPMC matrices. These included drug solubility and the molecular size, with enhancing and reducing effects on the drug release, respectively. The effect of these properties can be explained on the basis of the effect of solubility and molecular size on the diffusion of a drug within a uid medium. The similarity factor calculated in order for comparison

Figure 7. Plot of similarity factor ( f2) between matrices prepared either with HPMC K4M or HPMC E4M vs. Peppas n coefcient for HPMC E4M matrices.

DOI 10.1002/jps JOURNAL OF PHARMACEUTICAL SCIENCES, VOL. 96, NO. 12, DECEMBER 2007

3350

NOKHODCHI ET AL.

6. Higuchi T. 1963. Mechanism of sustained-action medication: Theoretical analysis of rate of release of solid drugs dispersed in solid matrices. J Pharm Sci 52:1451149. 7. Rajabi-Siahboomi AR, Bowtell RW, Manseld P, Henderson A, Davies MC, Melia CD. 1994. Structure and behavior in hydrophilic matrix sustainedrelease dosage forms. 2. NMR-imaging studies of dimensional changes in the gel layer and core of HPMC tablets undergoing hydration. J Control Release 31:121128. 8. Ford JL, Rubinstein MH, Hogan JE. 1985. Propranolol hydrochloride and aminophylline release from matrix tablets containing hydroxypropylmethylcellulose. Int J Pharm 24:339350. 9. Ford JL, Rubinstein MH, Hogan JE. 1985. Dissolution of a poorly water soluble drug, indomethacin, from hydroxypropylmethylcellulose controlled release tablets. J Pharm Pharmacol 37:33P. 10. Tahara K, Yamamoto K, Nishihata T. 1996. Application of model-independent and model analysis for the investigation of effect of drug solubility on its release rate from hydroxypropyl methylcellulose sustained release tablets. Int J Pharm 133: 1727. 11. Rao KVR, Devi KP, Buri P. 1990. Inuence of molecular-size and water solubility of the solute on its release from swelling and erosion controlled polymeric matrices. J Control Release 12:133141. 12. Baveja SK, Ranga Rao KV, Singh A, Gombar VK. 1988. Release characteristics of some bronchodilators from compressed hydrophilic polymeric matrices and their correlation with molecular geometry. Int J Pharm 41:5562. 13. United States Pharmacopoeia, 2003. vol. 26, p. 25242525. 14. Khan KA. 1975. Concept of dissolution efciency. J Pharm Pharmacol 271:4849. 15. Fedors RF. 1974. A method for estimating both the solubility parameters and polar volumes of liquids. Poly Eng Sci 14:147154. 16. Siepmann J, Peppas NA. 2001. Modeling of drug release from delivery systems based on hydroxypropyl methylcellulose (HPMC). Adv Drug Deliv Rev 48:139157. 17. Ford JL, Rubinstein MH, McCaul F, Hogan JE, Edgar PJ. 1987. Importance of drug type, tablet shape and added diluents on drug release kinetics from hydroxypropylmethylcellulose matrix tablets. Int J Pharm 40:223234. 18. Velasco MV, Ford JL, Rowe P, Rajabi-Siahboomi AR. 1999. Inuence of drug:hydroxypropylmethylcellulose ratio, drug and polymer particle size and compression force on the release of diclofenac sodium from HPMC tablets. J Control Release 57:7585. 19. Campos-Aldrete ME, Villafuerte-Robles L. 1997. Inuence of the viscosity grade and the particle size of HPMC on metronidazole release from matrix tablets. Eur J Pharm Biopharm 43:173178.

20. Mitchell K, Ford JL, Armstrong DJ, Elliott PNC, Rostron C, Hogan JE. 1993. The inuence of concentration on the release of drugs from gels and matrices containing Methocel. Int J Pharm 100: 155163. 21. Fu XC, Wang GP, Liang WQ, Chow MSS. 2004. Prediction of drug release from HPMC matrices: effect of physicochemical properties of drug and polymer concentration. J Control Release 95: 209216. 22. Ford JL, Mitchell K, Rowe P, Armstrong DJ, Elliott PNC, Rostron C, Hogan JE. 1991. Mathematical modelling of drug release from hydroxypropylmethylcellulose matrices: Effect of temperature. Int J Pharm 71:95104. 23. Mitchell K, Ford JL, Armstrong DJ, Elliott PNC, Hogan JE, Rostron C. 1993. The inuence of drugs on the properties of gels and swelling characteristics of matrices containing methylcellulose or hydroxypropylmethylcellulose. Int J Pharm 100: 165173. 24. Rajabi-Siahboomi AR, Adler J, Davies MC, Melia CD. 1994. Particle swelling and the mechanism of failure of HPMC matrices. Proc 3rd Assoc: 21. 25. Levina M, Rajabi-Siahboomi AR. 2004. The inuence of excipients on drug release from hydroxypropyl methylcellulose matrices. J Pharm Sci 93: 27462754. 26. Vueba ML, Batista de Carvalho LA, Veiga F, Sousa JJ, Pina ME. 2004. Inuence of cellulose ether polymers on ketoprofen release from hydrophilic matrix tablets. Eur J Pharm Biopharm 58:5159. 27. Kim H, Fasihi R. 1997. Application of a binary polymer system in drug release rate modulation. 1. Characterization of release mechanism. J Pharm Sci 86:316322. 28. Moore JW, Flanner HH. 1996. Mathematical comparison of dissolution proles. Pharm Technol 20: 6474. 29. Koester LS, Ortega GG, Mayorga P, Bassani VL. 2004. Mathematical evaluation of in vitro release proles of hydroxypropylmethylcellulose matrix tablets containing carbamazepine associated to bcyclodextrin. Eur J Pharm Biopharm 58:177179. 30. Tanigawara Y, Yamaoka K, Nakagawa T, Uno T. 1982. New method for the valuation of in vitro dissolution time and disintegration time. Chem Pharm Bull 30:10881090. 31. Huang YB, Tsai YH, Lee SH, Chang JS, Wu PC. 2005. Optimization of pH-independent release of nicardipine hydrochloride extended-release matrix tablets using response surface methodology. Int J Pharm 289:8795. 32. Hall LH, Kier LB, Murray WJ. 1975. Molecular Connectivity. 2. Relationship to water solubility and boiling-point. J Pharm Sci 64:19741977. 33. Tahara K, Yamamoto K, Nishihata T. 1995. Overall mechanism behind matrix sustained release (SR)

DOI 10.1002/jps

3351

tablets prepared with hydroxypropyl methylcellulose 2910. J Control Release 35:5966. 34. Akbari J, Adrangui M, Farid J, Siahi-Shadbad MR, Nokhodchi A. 2000. Factors affecting the release rate of a poorly soluble drug (carbamazepine) from HPMC matrices. STP Pharm Sci 10:473 478. 35. Martin A, Bustamante P, Chun AHC. 1993. Physical Pharmacy. Baltimore, USA: Lippincott Williams & Wilkins. p. 400401. 36. Mitragotri S. 2002. A theoretical analysis of permeation of small hydrophobic solutes across the stratum corneum based on scaled particle theory. J Pharm Sci 91:744752.

37. Buchwald P, Bodor N. 2001. A simple, predictive, structure based skin permeability model. J Pharm Pharmacol 53:10871098. 38. Rajabi-Siahboomi AR, Bowtell RW, Manseld P, Davies MC, Melia CD. 1996. Structure and behavior in hydrophilic matrix sustained release dosage forms. 4. Studies of water mobility and diffusion coefcients in the gel layer of HPMC tablets using NMR imaging. Pharm Res 13:376380. 39. Mitchell K, Ford JL, Armstrong DJ, Elliott PNC, Hogan JE, Rostron C. 1993. The inuence of substitution type on the performance of methylcellulose and hydroxypropylmethylcellulose in gels and matrices. Int J Pharm 100:143 154.

DOI 10.1002/jps

- tut16Diunggah olehAndre Oliveira
- RevModPhys.57.437Diunggah olehbhagaban70442393
- KF Wele WelwDiunggah olehdewi
- Solubility ProductDiunggah olehuniquestar
- Mass TransferDiunggah olehRijith Balakrishnan
- Assessment of the Leachability and Mechanical Stability of Mud From a Zinc-plating Plant and Waste Zeolite Binding With Portland CementDiunggah olehInternational Journal of Research in Engineering and Technology
- Codigo Fuente FortranDiunggah olehGabriel Menchu
- 10.1016@j.chemolab.2016.07.011Diunggah olehJessica
- HydrocarbonDiunggah olehBezawit Temesgen Belete
- Mass Transfer and DiffusionDiunggah olehAdrianio Lozhada
- Design of Optimal Process Flow Sheet for Fractional CrystallizationDiunggah olehVinay Bhaya
- TQ Science 7 1st GradingDiunggah olehRocaya Sarip
- williams1934.pdfDiunggah olehKhalya Fathonah Luwu
- pub-mb_2820Diunggah olehOlie
- Battery1 BatteryDesignStudio RSDiunggah olehdoowh
- 201210 Smit Help HoDiunggah olehamir8100
- Mercer 1990Diunggah olehAnonymous dkML2wzU
- Chemistry ReviewerDiunggah olehDavid Tulagan
- FixDiunggah olehVicky Andrean
- Itraconazole E100Diunggah olehabhijit_gothoskar6039
- Q1- How Many Grams Are There InDiunggah olehMandeep Singh
- fulltext(2)Diunggah olehAries Arcega
- mt test 1 qpDiunggah olehHaribabu Krishnan
- Chapter 13 Review chemDiunggah olehMindy Le
- Appendix 2Diunggah olehdjf113
- IJAIEM-2014-08-19-32Diunggah olehAnonymous vQrJlEN
- 5_Diffusion in SolidsDiunggah olehSam Magomnang
- Solids Liquids and GasesDiunggah olehJane Seymour
- Solubility of FibresDiunggah olehNasir Sarwar
- School Tutorials 15Diunggah olehVenkat Ram

- 1409908607Diunggah olehparag7676
- Atomic Model TimelineDiunggah olehLayneGelleiMeneses-MillaminaBuenavente
- Tutorial thermodynamic Chapter 2 : Properties of Pure SubstanceDiunggah olehNaim Zulkifli
- 56-PODJDiunggah olehRavish Malhotra
- Optical fiber ADSS - 12 96 cores english versionDiunggah olehtstefan88
- Analysis of Autogenous and Drying Shrinkage of ConcreteDiunggah olehAndrei Bora
- Chapter 8 - Cost of ManufacturingDiunggah olehRaja Maizatulakmal
- Mantenimiento de calderas industriales BOSCH.pdfDiunggah olehfrankz89
- physicclass guide EASADiunggah olehAngelo Martinez
- 25895327 Dna Repair MechanismsDiunggah olehayesha
- Fenol UOPDiunggah olehQuynh Nguyen Van
- Tesla Bladeless Microturbine - AerodynamicsDiunggah olehPaulFerry
- Insulating Koint CatalougeDiunggah olehDenstar Ricardo Silalahi
- 35. Petrochem Tech.pdfDiunggah olehThirunavukarasu
- GUIDE - Elect Roles Nickel PlatingDiunggah olehratnasdatar
- Quantum Dots.pdfDiunggah olehsoumendra ghorai
- Sucker Rod String Design by LufkinDiunggah olehachmad mahmudy
- Ocean Circulation DeepseaDiunggah olehDeenar Tunas Rancak
- Seafloor Hydrothermal SystemsDiunggah olehraj_h1980
- CV_sunil_wajeDiunggah olehYogesh Kathrecha
- User Guide EkomilkDiunggah olehEdwin Chila Choque
- Chima et alDiunggah olehVictor Nwaugo
- 1 Sand CastingDiunggah olehMayank Grover
- PneumaticsDiunggah oleher9824925568
- Word ListDiunggah olehMohd Safuan Zakaria
- Pneumatic Conveying Demo GuideDiunggah olehmontoyazumaeta
- Understanding Rheology of ThermosetsDiunggah olehchristopher_kephart1
- C-OV1.1_09.09Diunggah olehmohamednibin
- Guidelines_Using_Thermal_Mass_Concrete_BuildingsDiunggah olehnatalia_campelo
- Anti-mycobacterium Activity and Bioassay Guided Fractionation of the Leave Extract of Ximenia Americana L.Diunggah olehrobert0rojer