Intrinsic Dissolution Performance Testing
`of the USP Dissolution Apparatus 2
`(Rotating Paddle) Using Modified
`Salicylic Acid Calibrator Tablets: Proof of Principle
`John Mauger1, John Ballard2, Robert Brockson2,
`Sinjan De3, Vivian Gray4, and Dennis Robinson3
`
`email correspondence to: jmauger@deans.pharm.utah.edu
`
`Introduction
`
`K nowledge of operating variables for a dissolution
`
`apparatus is important to the pharmaceutical scientist
`interested in product development,quality assurance,
`and research applications. The dissolution performance of
`USP Apparatus 1 and 2 is dependent on both convection
`and diffusion as coupled dissolution mechanisms. The sensi-
`tivity and variability of the data may be compromised when
`either the convection or diffusion mechanism dominates.
`When convection dominates,there may be a loss of sensi-
`tivity. For example,Hamlin et al.,(1) demonstrated a loss of
`sensitivity in distinguishing real differences in dissolution
`rates due to increasing agitation intensity. When diffusion is
`the predominant mechanism,the dissolution results are
`affected by density gradients and may vary considerably (2,
`3). USP Apparatus 1 and 2 depend on a balance between
`these two mechanisms in order to produce sensitive and
`reproducible data.
`Experimental dissolution methods that employ configu-
`rations linked with convective-diffusion models,such as
`the rotating disk or the stationary disk,are useful to under-
`stand the relationship between the diffusion and convec-
`tive mechanisms. The dissolution rates determined from
`either of these two experimental methods are often
`termed “intrinsic dissolution rates”. Intrinsic dissolution
`rates have been defined as “the rate of dissolution of a pure
`pharmaceutical active ingredient when conditions such as
`surface area,temperature,agitation or stirring speed,pH,
`and ionic strength of the dissolution medium are kept con-
`stant”(4).
`This research was motivated by previous studies (2,3)
`that suggested the possibility of using modified salicylic
`acid calibrator tablets supplemented by visualization stud-
`ies (5) as a method to evaluate the intrinsic dissolution per-
`formance of USP Apparatus 2. Intrinsic dissolution perfor-
`mance in this context is the intrinsic dissolution rate due to
`the inherent convection and diffusion mechanisms that
`govern dissolution for a given apparatus. The premise for
`the current study is that these dissolution rates provide a
`framework for evaluating the dissolution performance of
`USP Apparatus 2.
`
`Therefore,the goals of this study were to:
`
`1) Demonstrate the usefulness of a simple visual analy-
`sis technique.
`2) Characterize the variability associated with the disso-
`lution data for USP Apparatus 2.
`3) Test the premise that USP Apparatus 2 operates with-
`in a convective-diffusion model that explains the
`relationship between dissolution rate and stirring
`speed as well as other hydrodynamic and fluid
`mechanical factors.
`4) Demonstrate that the convective-diffusion model
`reflects an accurate picture of the physicochemical
`aspects of dissolution within the micro environmen-
`tal region at the tablet surface.
`
`While prednisone and salicylic acid calibrators have been
`used since 1978 in connection with apparatus suitability test-
`ing,these tests are aimed primarily at demonstrating the
`influence of perturbations such as vibration (6,7).This study
`will focus on the use of modified USP Salicylic Acid Calibrator
`Tablets as a method to characterize the intrinsic dissolution
`performance characteristics of USP Apparatus 2 within the
`framework of a convective-diffusion model that explains the
`influence of stirring speed,the underlying fluid mechanical
`factors,and the diffusional and physicochemical parameters.
`
`Intrinsic Dissolution Performance Testing
`An intrinsic dissolution test should not only be based on
`a convective-diffusion model,but also on previous knowl-
`edge of the actual fluid flow regime relevant to the appara-
`tus. Previous research reported by Bocagnegra et al. (8)
`demonstrated the use of laser Doppler anemometry in
`characterizing the three-dimensional fluid flow compo-
`nents in USP Apparatus 2. Of singular interest is the finding
`that close to the bottom of the vessel of this apparatus,the
`tangential component of flow dominates and approxi-
`mates solid body rotation. This finding implies that the
`intrinsic dissolution properties of this apparatus may be
`predictable within the framework of a fluid mechanics-
`based dissolution model that incorporates and couples the
`convective and diffusion mechanisms.
`
`1 Corresponding author,University of Utah,College of Pharmacy,30 South 2000 East,Rm.
`201,Salt Lake City,UT 84112-5820,jmauger@deans.pharm.utah.edu
`2 DuPont Pharmaceuticals,Greenville,DE
`
`3 College of Pharmacy,986000 Nebraska Medical Center,Omaha,NE,68198
`4 V.A.Gray Consulting,Inc.,9 Yorkridge Trail,Hockessin,DE 19707
`
`6 Dissolution Technologies | AUGUST 2003
`
`ENDO - Ex. 2023
`Amneal v. Endo
`IPR2014-00360
`
`

`

`The fluid mechanical component of dissolution is repre-
`sented by the dimensionless Reynolds number,which is
`the ratio of inertial to frictional forces (9) where:
`1) Re = U r/ν
`The representative fluid velocity is given by U (cm/sec),ν
`is the kinematic viscosity of the fluid at a given tempera-
`ture (cm2/sec) and r is the dimension of a physical body
`over which the fluid is flowing,say the radius of a tablet
`(cm). This relationship shows that the Reynolds number is
`directly proportional to the fluid velocity and inversely pro-
`portional to the kinematic viscosity.
`The convective-diffusion component of dissolution is
`represented by the dimensionless Peclet number (9):
`
`2) Pe = U r/D
`
`where U and r are previously defined and D is the diffu-
`sion coefficient (cm2/sec) of the drug.
`For a rotating disk in an unstirred fluid,or a stationary
`disk in a rotating fluid,the Reynolds and Peclet numbers
`are modified to include the radius of the disk (r,in cm) and
`the rotational rate of the disk (ω,in radians/sec):
`3) Re = r2ω/ν and Pe = r2ω/D
`For the rotating disk,the dimensionless Reynolds and
`Peclet numbers are related to the dissolution rate per unit
`surface area,J,by:
`
`4) J = 0.62 (D/r) (Re)1/2 (Pe/Re)1/3 Cs
`
`where Cs (g/cm3) is the saturation solubility at a given
`temperature. The relationship given by equation 4 is pre-
`dictive and contains information concerning the convec-
`tive and diffusional components of the intrinsic dissolution
`properties. Furthermore,it is based on a fluid mechanics
`model for a rotating disk in an unstirred fluid (9,pp. 60-71).
`Given the findings of Bocanegra et al. (6) the fluid flow at
`the bottom of the USP Apparatus 2 is approximated by a
`model for a rotating fluid over a stationary disk. This system
`has been studied in connection with engineering and
`pharmaceutical applications (10,11,and 12). Colton and
`Smith (10) gave the following relationship for the dissolu-
`tion rate per unit area when averaged over the surface of
`the disk,,where:
`5) 〈J〉 = 0.77 (D/r) (Re)1/2 γ1/2 (Pe/Re)1/3 Cs.
`The stationary disk is similar to the rotating disk model
`since it is predictive and includes both the convective and
`diffusional components contributing to the dissolution
`rate. However,there are important differences related to
`
`the direction of fluid flow on the dissolution surface. With
`the rotating disk configuration,fluid flows outward from
`the center of the disk and the disk is uniformly accessible
`across the dissolving surface. With the stationary disk con-
`figuration,the dissolving surface is not uniformly accessi-
`ble to dissolution as the fluid sweeps from the leading
`outer edge to the center of the tablet. According to Colton
`and Smith (10) approximately 80% of the mass transport
`occurs in the outer 20% of the tablet area.
`With the stationary disk,a plume forms at the center of the
`tablet as the fluid and dissolved material exits from the tablet
`surface upward toward the center of the paddle blade.The
`three-dimensional fluid flow profiles associated with the sta-
`tionary disk have been described by Schlicting (13).
`Another differing characteristic between the rotating
`and stationary disks is the γ term,which is a dimensionless
`ratio of the representative fluid velocity at some axial dis-
`tance below the paddle to the maximum paddle tip veloci-
`ty. Below the paddle and just above the surface of the dis-
`solution vessel,the fluid velocity is expected to be less than
`the paddle tip velocity and greater than zero,resulting in a
`ratio bounded between zero and one. In keeping with this
`expectation,the data from the study by Bocanegra et al. (8)
`indicates that for the USP Apparatus 2,this ratio is approxi-
`mately 0.4 in the vicinity of the dissolving surface below
`the stirring source.
`This experimentally determined value is dependent
`upon the radial distance from the center of the vessel and
`approaches zero under two conditions:1) at the center of
`the vessel,in keeping with the solid body rotation model
`where V = ωr and 2) at the surface of the vessel due to fric-
`tional forces.
`Equations 4 and 5 were tested in the present study as a
`basis for an intrinsic dissolution performance test in USP
`Apparatus 2.The test was undertaken using modified USP
`Salicylic Acid Calibrator Tablets that were coated on the rim
`and one side with a water impervious shellac,with the
`uncoated surface facing the paddle. These tablets present a
`constant surface area for dissolution during the experiment.
`
`Experimental
`Materials— Tablets for the visualization studies were
`prepared by blending a known weight of salicylic acid,
`Sigma Chemicals Lot #70K002451,with 3% by weight of
`phenolphthalein,Sigma Chemicals Lot #30K1353. Previous
`studies showed that this concentration of phenolphthalein
`was sufficient to give a clear visualization of the dissolution
`process at the surface of the tablet without flooding the
`bulk solution with color. Three hundred milligrams of this
`
`Dissolution Technologies | AUGUST 2003
`
`7
`
`

`

`Intrinsic Dissolution Performance Testing … continued
`
`blend was compressed into a tablet using 0.5 metric tons
`of pressure for 30 seconds with a Carver press. These
`tablets were used for the visualization studies after inspec-
`tion for chipping and surface imperfections. The diameter
`and thickness of these tablets approximated USP calibrator
`tablets.
`USP Salicylic Acid Calibrator Tablets,Lot N,300 mg,were
`coated on one side and around the rim with water impervi-
`ous shellac that contained as the primary ingredient
`toslyamide epoxy resin. These tablets were used in the
`intrinsic dissolution study.
`Visualization Studies – Visualization studies were per-
`formed using commercially available dissolution equip-
`ment,(VanKel),USP Apparatus 1 and 2. The visualization
`studies were performed under conditions of no agitation
`and at 25,50,and 100 rpm in 900 ml of sodium hydroxide
`solution of known concentration at 37ºC. Salicylic acid
`tablets containing phenolphthalein were placed in the ves-
`sels containing sodium hydroxide solution and the dissolu-
`tion process was visualized by observing the color at the
`surface of the tablet as dissolution of the salicylic acid and
`phenolthalein blend occurred in the alkaline solution. Pho-
`tographs were then taken between 1 and 10 minutes after
`placing the tablet in the dissolution medium.
`Dissolution Studies - Dissolution studies were performed
`on commercially available dissolution equipment,
`(VanKel),USP Apparatus 2. Test conditions were consistent
`with USP <711> Apparatus System Suitability require-
`ments. The intrinsic dissolution studies were performed
`using the USP Calibrator Tablet procedure,with 900 mL of
`0.05 M phosphate buffer (pH 7.4) as the medium. The sam-
`pling times were 5,10,15,20,25,and 30 minutes at 50 and
`100 rpm. All sampling was performed using the fiber optic
`automated system of CTechnology instruments. The solu-
`tions were read at 296 nm using the Cary 50 spectropho-
`tometer. Care was taken to locate the tablet,active face up
`and coated surface down,at the center of the vessel. Inde-
`pendent studies were undertaken to confirm that salicylic
`acid did not leak through the epoxy resin and that the resin
`did not interfere with the assay results.
`
`Results and Discussion
`
`1. Visualization Studies
`Visualization studies were undertaken as the first step in
`affirming the fluid flow patterns demonstrated previously
`using laser Doppler anemometry (8). The photograph from
`the first study is shown in Figure 1 to demonstrate the prin-
`ciple.
`This experiment was undertaken with the active side of
`the salicylic acid tablet down in 0.05 N sodium hydroxide
`solution in USP Apparatus 1 using unstirred conditions.
`The more dense dissolved material is easily visualized as it
`
`8 Dissolution Technologies | AUGUST 2003
`
`Figure 1. Basket apparatus, no stirring
`
`Figure 2. Paddle apparatus, no stirring
`
`falls to the bottom of the vessel,with movement created
`by the density gradients.
`The next photograph in Figure 2 shows the results from a
`visualization experiment where the active side of the
`tablet is facing up in 0.05 N sodium hydroxide solution in
`USP Apparatus 2 under unstirred conditions. This photo-
`graph shows that the dissolving material collects under
`and around the sides of the tablet due to the difference in
`density between dissolved material and the bulk medium.
`
`See Intrinsic Dissolution Performance Testing… continued on page 10
`
`

`

`Intrinsic Dissolution Performance Testing … continued
`
`Figure 3. Paddle apparatus at 25 rpm
`
`Figure 4. Paddle Apparatus at 50 rpm
`
`This visualization experiment underscores the impor-
`tance of density gradients. While it is expected that the
`more dense dissolved material will be swept into the bulk
`solution by fluid flow when the paddle speed is either 50
`or 100 RPM,this premise is open to question. In fact,other
`studies (2,3) designed to test this question have demon-
`strated that density gradients under the tablet affect the
`coefficient of variation at both 50 and 100 RPM using
`Apparatus 2. These findings imply that the outcomes from
`the USP <711> Apparatus Suitability test using salicylic
`acid calibrator tablets may be affected by density gradi-
`ents’influence on the variability of the data.
`The next photograph in Figure 3 was taken with the
`active side up in 0.025 N sodium hydroxide using USP
`Apparatus 2 at 25 RPM.
`Since stirring by the paddle creates fluid flow approxi-
`mating solid body rotation at the bottom of the vessel,the
`rotating fluid turns and sweeps inward across the tablet
`surface near the bottom of the vessel. As the rotating fluid
`encounters the outer edge of the tablet surface it slows
`due to friction and then travels radially inward toward the
`center of the tablet. Then the fluid carrying the dissolved
`material spirals upward toward the center of the paddle.
`This photograph clearly shows the plume exiting upward
`toward the center of the paddle. There is visual evidence of
`imperfect mixing of the more dense dissolved solution
`with the bulk solution at this stirring speed.
`For comparative purposes,the same experiment was run
`at 50 and 100 RPM. The resulting photographs from these
`experiments are shown in Figures 4 and 5 (Figure 4 is at 50
`RPM and Figure 5 is at 100 RPM).
`At 50 RPM the plume is well formed,and mixing with the
`bulk solution seems to be uniform. Although the shear
`pattern at the tablet surface is not evident in this photo-
`
`10 Dissolution Technologies | AUGUST 2003
`
`Figure 5. Paddle apparatus at 100 rpm
`
`graph,the color at the tablet surface is indicative of the
`underlying physicochemical process related to dissolu-
`tion. The physicochemical factors that play a role in the
`microenvironment at the tablet surface include the solu-
`bility of the dissolving material,the pKa of the dissolving
`material,the buffer capacity of the dissolution medium,
`and the pKa of the buffer relative to the pKa of the dissolv-
`ing material.
`While this idealized visualization experiment shows the
`dissolution from one surface facing the paddle,it is impor-
`tant to note that the dissolution process will be different
`for the rim and the bottom surfaces. This point was
`emphasized by Freebern (2) as well as Healy et al. (14) who
`demonstrated the effect of different surfaces on the disso-
`lution rate of benzoic acid compacts.
`At a higher stirring speed of 100 RPM the upward lift of
`the dissolved material into the bulk solution is evident and
`
`

`

`They also reflect the underlying physicochemical process
`at the tablet surface that controls dissolution.
`
`2. Characterizing the Variability Associated With the
`Intrinsic Dissolution Rates
`The data from the dissolution experiments using USP
`Apparatus 2 and coated USP Salicylic Acid Tablets,with the
`active surface facing up toward the paddle,are shown in the
`following box-whisker plots (15) as Figures 6 and 7 where
`percent dissolved is plotted versus time under stirring con-
`ditions of 50 and 100 RPM.The median for six replicates at
`each time point is represented by the horizontal line.
`The percent dissolved is a linear function of time for
`these data at both 50 and 100 RPM.
`The variability of the data is indicated by the box length.
`At 50 RPM,the box lengths are approximately the same
`length over time while at 100 RPM the box lengths tend to
`increase with increasing time. This finding suggests that
`the variability may be time dependent at the higher stir-
`ring speed.
`The plots also show the upper and lower whisker
`lengths,which represent the largest and smallest values in
`each data set,respectively. The whisker length appears to
`increase with time at both 50 and 100 RPM,suggesting
`that the range of data may be correlated to time. At 50
`RPM the whiskers are reasonably symmetric with respect
`to the median at each time point while at 100 RPM the
`whiskers are increasingly asymmetric with increasing time.
`At 100 RPM,the median drifts away from the center of the
`box as time increases.
`In summary,the box-whisker plot analysis has demon-
`strated that 1) at 100 RPM the values are increasingly asym-
`metric with increasing time,2) the box length appears to
`increase with time,particularly at 100 RPM,and 3) the
`whisker length appears to increase with increasing time at
`both 50 and 100 RPM. Since apparatus suitability testing is
`usually linked with one pre-selected time point for the USP
`Salicylic Acid Calibrator Tablet,this finding suggests that
`the apparent time dependence of the variability may be a
`factor in selecting the time point associated with the
`acceptance range for suitability testing.
`Given the small sample size of this study,additional
`studies are needed to determine the within-laboratory
`variability associated with the intrinsic dissolution rates.
`Within-lab variability will provide a basis on which to more
`rigorously evaluate the apparent time-dependent variabil-
`ity and will also allow a comparison of the level of differ-
`ence in variability between 50 and 100 RPM.
`
`3. Evaluation of the Intrinsic Dissolution Rates Within
`the Framework of a Convective Diffusion Model
`All dissolution data were generated from the modified
`USP Salicylic Acid Calibrator Tablet,with one face and the
`
`Dissolution Technologies | AUGUST 2003
`
`11
`
`Figure 6. Box Plot of paddle apparatus at 50 rpm
`
`Figure 7. Box plot of paddle apparatus at 100 rpm
`
`the plume becomes more diffuse. Other visualization experi-
`ments (5) have shown that at even higher agitation rates,the
`color at the dissolving surface becomes absent or unevenly
`dispersed across the tablet surface as the dissolved material
`is lifted from the surface at a faster rate. In this case the con-
`vection component of the dissolution mechanism out-
`weighs the diffusional component,and it is anticipated that
`the sensitivity of the test would be compromised.
`In summary,these visualization experiments are simple
`to perform and provide qualitative information about the
`underlying fluid flow patterns as well as evidence about
`the influence of stirring speed on density gradients and
`mixing of the dissolved material with the bulk solution.
`
`

`

`Intrinsic Dissolution Performance Testing … continued
`
`equation that relates the intrinsic dissolution rate to the
`stirring speed and the Reynolds number.
`This analysis has shown that USP Apparatus 2 operates
`within a convective-diffusion model within specified stir-
`ring speed ranges. This finding leads to the idea that intrin-
`sic dissolution testing may have potential as an apparatus
`suitability test to detect perturbations in the hydrodynam-
`ic conditions that affect the intrinsic dissolution rate.
`
`4. Reflecting an Accurate Picture of the
`Microenvironment at the Dissolving Tablet Surface
`The commonly used Nernst model and convective-diffu-
`sion models such as equations 4 and 5 both relate the dis-
`solution rate per unit surface area to the saturation solubil-
`ity of the pure drug (Cs) as follows:
`7) 〈J〉 ∝Cs.
`Other studies have shown that the solubility term is
`dependent upon the micro environmental pH,which,in
`turn,is coupled with the buffering capacity of a saturated
`solution of the drug (17). In fact,the micro environmental
`pH is dependent upon several factors including the pKa of
`the buffer,the pKa of the drug,the pH and buffer capacity of
`the buffer system,and the saturation solubility of the drug
`(17). Therefore,the bulk pH of the buffer is often an inaccu-
`rate approximation of the pH at the dissolving surface.
`The importance of this point has been emphasized by
`Rohrs (18) in a discussion of dissolution methods for poor-
`ly soluble compounds. In the case of salicylic acid,the
`micro environmental pH near the dissolving surface was
`approximated by determining the pH of a saturated solu-
`tion of salicylic acid in 0.05 M pH 7.4 buffer.
`The experimentally determined value under these con-
`ditions was 3.2,which is in reasonable agreement with the
`previously determined value of 3.4 (16). The saturation sol-
`ubility in pH 7.4 buffer is 0.073 M (16). This value,converted
`to g/cm3,was then used in equation 5 to generate the cal-
`culated value for the stationary disk model. We conclude
`that the stationary disk model reflects an accurate picture
`of the micro environment at the tablet surface where rate
`controlling interactions occur between the buffer species
`and the dissolving salicylic acid. This finding has practical
`implications when selecting buffer species,buffer concen-
`tration,and buffer pH for the dissolution medium. More-
`
`Table I – Intrinsic Dissolution Rates Determined under
`Compendial Conditions at pH 7.4 and at 50 and 100
`RPM Paddle Speeds
`
`Paddle
`Speed
`(RPM)
`
`Intrinsic
`Dissolution
`Rate (g/sec)
`
`Standard Error
`of the Intrinsic
`Dissolution Rate
`
`Intrinsic Dissolution
`Rate Per Unit Surface
`Area (g/sec/cm2)
`
`50
`
`100
`
`9.06E-06
`
`1.33E-05
`
`±4.02E-07
`
`±8.74E-07
`
`1.25E-05
`
`1.84E-05
`
`Figure 8. Aggregate data, percent dissolved
`
`rim coated with water impervious shellac. The surface area
`was constant during the experiment and tablet surfaces
`did not show excess wear patterns at the completion of the
`run. The aggregate data for each set of six tablets are
`shown in Figure 8.
`The percent dissolved data were transformed to amount
`dissolved and the data from six tablets at each time period
`were pooled and analyzed using least squares analysis,
`which provided an estimate of the slope and the standard
`error of the slope. These slopes represent the intrinsic dis-
`solution rates,which were then normalized to amount dis-
`solved per unit time per unit area. These results are given in
`Table I.
`Other relevant physical chemical data for salicylic acid
`and related fluid mechanical data needed to test equations
`4 and 5 are given in Table II.
`The data in Table II were used with in conjunction with
`equations 4 and 5 to independently calculate the intrinsic
`dissolution rates at 50 and 100 RPM from the rotating disk
`and stationary disk models. Table III compares calculated
`intrinsic dissolution rates with the experimental rates.
`The calculated values using the rotating disk model,
`equation 4,are considerably greater than the experimental
`values at both 50 and 100 RPM. In contrast there is good
`agreement between the experimental and calculated val-
`ues for the stationary disk at both 50 and 100 RPM,with γ =
`0.40. Therefore,we conclude that the experimentally
`determined intrinsic dissolution rates are in reasonable
`agreement with equation 5,the stationary disk model.
`These findings imply that the experimental intrinsic disso-
`lution rates are in agreement between 50 and 100 RPM
`when they are related to the stirring rate by:
`6) 〈J〉 ∝ (Reγ)1/2.
`Therefore,within these commonly used stirring ranges,
`intrinsic dissolution results follow from one underlying
`
`Dissolution Technologies | AUGUST 2003
`
`12
`
`

`

`Table II - Relevant Physical Chemical Data for Salicylic Acid and Salicylic
`Acid Calibrator Tablets, Lot N, with Relevant Fluid Mechanical Data
`
`Saturation Solubility of Salicylic Acid in Water
`at 37 degrees C1
`Saturation Solubility of Salicylic Acid in pH 7.4
`Phosphate Buffer at 37 degrees C1
`Final pH of a Saturated Solution of Salicylic
`Acid in 0.05M Phosphate Buffer at 37 degrees C
`
`1.87E-02M,2.58 mg/mL
`
`7.30E-02M,10.07 mg/mL
`
`3.2
`
`Diffusion Coefficient for Salicylic Acid2
`
`1.132E-05 cm 2/sec
`
`Salicylic Acid Calibrator Tablet Weight
`
`300 mg.
`
`Surface Area of Salicylic Acid Calibrator Tablet
`
`0.7238 cm2
`
`Kinematic Viscosity of Water at 37 degrees C3
`
`0.00699 cm2/sec
`
`50 RPM;100 RPM
`
`Reynolds Number
`
`5.235 radians per sec.;10.47 radians per sec.
`
`1.73E02 at 50 RPM;3.45E02 at 100 RPM
`
`1.07E05 at 50 RPM;2.13E05 at 100 RPM
`
`Peclet Number
`ν/D
`Dimensionless stirring speed factor,γ4
`1Reference 16
`2M. Gibaldi,S. Feldman,and N. Weiner,Chem. Pharm. Bull. 18,715-723 (1970).
`3H. Nogami,T. Nagai,and A. Suzuki,Chem. Pharm. Bull.,14,329-338,1966.
`4 Reference 8.
`
`617
`
`0.4
`
`solution data tends to increase with
`increasing time during the dissolution
`process and that the relative magnitude
`of data dispersion is consistently higher
`at 100 RPM than at 50 RPM.
`These findings provide a basis to fur-
`ther develop and refine apparatus suit-
`ability test requirements using chemi-
`cal calibrators. One can envision an
`acceptance range that is based not only
`on the statistical analysis of an experi-
`mental data set,but also on an interval
`that is based on the difference between
`the experimental value and the calcu-
`lated value from the stationary disk
`model.
`In addition,it may be of value to eval-
`uate calibrator data for relationships
`between variability and time when
`selecting the time point for suitability
`testing. Finally,it would be worthwhile
`in future studies to determine the cut-
`off limits where the level of agitation
`begins to confound the ability to detect
`hydrodynamic and fluid mechanical perturbations.
`
`Table III - Comparison of Experimental Dissolution
`Rates (g/sec/cm2) with Rates Calculated Using the
`Rotating Disk and Stationary Disk Models
`
`Stirring
`Speed
`(RPM)
`
`Experimental
`Dissolution
`Rate per Unit
`Surface Area
`
`Rotating Disk
`Dissolution Rate
`per Unit Surface
`Area (equation 4)
`
`50
`
`100
`
`1.25E-05,n=6
`
`1.65E-05
`
`1.84E-05,n=6
`
`2.33E-05
`
`Stationary Disk
`Dissolution Rate
`per Unit Surface
`Area (equation 5)
`1.29E-05,γ = 0.40
`1.83E-05,γ = 0.40
`
`over,these findings support the principle that the micro
`environmental pH rather than the bulk pH controls the dis-
`solution rate process for the salicylic acid calibrator tablet.
`
`Summary
`Visualization experiments coupled with intrinsic dissolu-
`tion rates studies have demonstrated that this methodolo-
`gy is useful in characterizing the performance of USP Appa-
`ratus 2 within the framework of a convective-diffusion
`model. The experimentally determined dissolution rates
`were found to be in good agreement with those calculated
`independently from the stationary disk model at both 50
`and 100 RPM. This model accounts for convective effects
`and provides an accurate picture of the physicochemical
`interactions occurring in the microenvironment at the
`tablet surface. The model also demonstrates a quantitative
`relationship between the expected dissolution rate and
`stirring speed. Furthermore,through box-whisker plot
`analysis,the data demonstrate that the variability of the dis-
`
`References
`1) W.E. Hamlin,E. Nelson,B.E. Ballard,and J.G. Wagner,
`“Loss of Sensitivity in Distinguishing Real Differences
`In Dissolution rates due to Increasing Agitation Inten-
`sity”,J. Pharm.Sci.,51,432-435,1962.
`2) K. Freebern,“Dissolution Kinetics of Calibrator and
`Matrix Tablets In the USP Dissolution Apparatus”,Mas-
`ters Thesis,The University of Nebraska Medical Center,
`Department of Pharmaceutical Sciences,1993.
`3) K. Freebern,D. Robinson,and J. Mauger,“In Vitro Disso-
`lution of Salicylic Acid Calibrator Tablets”,AAPS Mid-
`west Regional Meeting,Chicago,Illinois,1991.
`4) T. Viegas,R. Curatella,L. VanWinkle,and G. Brinker,
`“Measurement of Intrinsic Drug Dissolution Rates
`Using Two Types of Apparatus”,Pharmaceutical Tech-
`nology,June,2001.
`5) J. W. Mauger,“Physicochemical and Fluid Mechanical
`Factors Related to Dissolution Testing”,Dissolution
`Technologies,3(1),7-11,1996.
`6) Thakker,K.D.;Naik,N.C.;Gray,V.A. and Sun,S.,Fine-Tun-
`ing of Dissolution Apparatus for Apparatus Suitability
`Test Using the USP Dissolution Calibrators,Pharm.
`Forum,6(4),177-185,1980.
`7) PhRMA Subcommittee on Dissolution Calibration:
`Brune,S.,Bucko J.,Emr,S.,Gray,V.,Hippeli,K.,Kentrup,
`
`See Intrinsic Dissolution Performance Testing… continued on page 15
`
`Dissolution Technologies | AUGUST 2003 13
`
`

`

`Intrinsic Dissolution Performance Testing … continued
`
`A.,Whiteman,D.,Loranger,M.,Oates,M.,‘Dissolution
`Calibrator:Recommendations for Reduced Chemical
`Testing and Enhanced Mechanical Calibration”,Pharm.
`Forum,26(4),1149-1166,2000.
`8) L.M. Bocanegra,G. Morris,J. Jurewich,and J.W. Mauger,
`“Fluid and Particle Laser Doppler Velocity Measure-
`ments and Mass Transfer Predictions for the USP Pad-
`dle Method Dissolution Apparatus”,Drug Develop-
`ment and Industrial Pharmacy,16 (9),1441-1464,1990.
`9) V. Levich,Physicochemical Hydrodynamics,Prentice-
`Hall,Inc.,Englewood Cliffs,NJ,1962,p. 52.
`10) C. Colton and K. Smith,“Mass Transfer to a Rotating
`Fluid. Transport from a Base of an Agitated Cylindrical
`Tank”,AIChE Journal,18 (5),958-967,1972.
`11) N. Khoury,J.W. Mauger,and S. Howard,“Dissolution
`Rate Studies from a Stationary Disk/Rotating Fluid Sys-
`tem”,Pharmaceutical Research,5,495-500,1988.
`12) T.Viegas,R. Curatella,L. VanWinkle,and G. Brinker,“Intrin-
`sic Drug Dissolution Testing Using the Stationary Disk
`System”,Dissolution Technologies,8(3),19-23,2001.
`13) H. Schlicting,Boundary Layer Theory,6th.ed.,McGraw Hill
`Book Company,New York,1968,pp. 93-97,pp. 213-218.
`14) A.M. Healy,L.G. McCarthy,K.M. Gallagher,and O.I. Corri-
`
`gan,“Sensitivity of Dissolution Rate to Location in the
`Paddle Dissolution Apparatus”,Journal of Pharmacy
`and Pharmacology,54,441-444,2002.
`15) J. W. Tukey,Exploratory Data Analysis,Addison-Wesley
`Publishing Co.,Reading Mass.,1977,pp. 39-46.
`16) Z. Ramtoola and O.I. Corrigan,“Dissolution Characteris-
`tics of Benzoic Acid and Salicylic Acid Mixtures in Reac-
`tive Media”,Drug Development and Industrial Pharma-
`cy,13,1703-1720,1987.
`17) D. French and J.W. Mauger,“Evaluation of the Physico-
`chemical Properties and Dissolution Characteristics of
`Mesalamine:Relevance to Controlled Intestinal Drug
`Delivery”,Pharmaceutical Research,10,1285-1290,1993.
`18) B. Rohrs,“Dissolution Method Development for Poorly
`Soluble Compounds”,Dissolution Technologies,8(3),
`August 2001.
`
`Acknowledgements
`Assistance and support from the USP is gratefully
`acknowledged. Specific recognition is given to Will Brown,
`Walter Hauck,and Margareth Marques for their valuable
`suggestions and support.
`
`Dissolution Technologies | AUGUST 2003 15
`
`