Scholarly article on topic 'Galenic approaches in troubleshooting of glibenclamide tablet adhesion in compression machine punches'

Galenic approaches in troubleshooting of glibenclamide tablet adhesion in compression machine punches Academic research paper on "Materials engineering"

CC BY-NC-ND
0
0
Share paper
Academic journal
Saudi Pharmaceutical Journal
OECD Field of science
Keywords
{Glibenclamide / Adhesion / "Magnesium stearate" / Tableting}

Abstract of research paper on Materials engineering, author of scientific article — Janine Boniatti, Ana Lúcia Pereira Cerqueira, Alexandre Carnevale de Souza, Cristiane Rodrigues Drago Hoffmeister, Maira Assis da Costa, et al.

Abstract This study aimed to examine the adhesion of glibenclamide 5mg tablets to the tools of compression machines. This problem is not commonly reported in the literature, since it is considered as tacit knowledge. The starting point was the implementation of three technical alternatives: changing the parameters of compression, evaluating the humidity of the powder blend and the manufacturer of the lubricant magnesium stearate. The adhesion was directly related to the characteristics of magnesium stearate from different manufacturers, and the feasibility of evaluating powder flow characteristics by different techniques that are not routinely followed in various pharmaceutical companies. In vitro dissolution tests showed that the magnesium stearate manufacturer can influence on the dissolution profile of glibenclamide tablets. This study presented various aspects of tablet adhesion to compression machine punches. Troubleshooting approaches can be, most of times, conducted based on previous experience, or an experimental research needs to be implemented in order to have confident results.

Academic research paper on topic "Galenic approaches in troubleshooting of glibenclamide tablet adhesion in compression machine punches"

Saudi Pharmaceutical Journal (2013) xxx, xxx-xxx

King Saud University Saudi Pharmaceutical Journal

www.ksu.edu.sa www.sciencedirect.com

ORIGINAL ARTICLE

Galenic approaches in troubleshooting of glibenclamide tablet adhesion in compression machine punches

Janine Boniatti a, Ana LUcia Pereira Cerqueira a, Alexandre Carnevale de Souza a,

Cristiane Rodrigues Drago Hoffmeister a, Maira Assis da Costa a,

Livia Deris Prado a'*, Leandro Tasso b, Helvecio Vinícius Antunes Rocha a

a Farmanguinhos (FIOCRUZ), Av. Comandante Guaranys 447, 22775-903 Rio de Janeiro, RJ, Brazil b University of Caxias do Sul, R. Francisco GetUlio Vargas 1130, 95070-560 Caxias do Sul, RS, Brazil

Received 6 August 2013; accepted 31 August 2013

KEYWORDS

Glibenclamide; Adhesion;

Magnesium stearate; Tableting

Abstract This study aimed to examine the adhesion of glibenclamide 5 mg tablets to the tools of compression machines. This problem is not commonly reported in the literature, since it is considered as tacit knowledge. The starting point was the implementation of three technical alternatives: changing the parameters of compression, evaluating the humidity of the powder blend and the manufacturer of the lubricant magnesium stearate. The adhesion was directly related to the characteristics of magnesium stearate from different manufacturers, and the feasibility of evaluating powder flow characteristics by different techniques that are not routinely followed in various pharmaceutical companies. In vitro dissolution tests showed that the magnesium stearate manufacturer can influence on the dissolution profile of glibenclamide tablets. This study presented various aspects of tablet adhesion to compression machine punches. Troubleshooting approaches can be, most of times, conducted based on previous experience, or an experimental research needs to be implemented in order to have confident results.

© 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University.

* Corresponding author. Tel.: +55 213 348 5319; fax: +55 213 348 5050.

E-mail addresses: liviaderis@yahoo.com.br, liviaderis@gmail.com (L.D. Prado).

Peer review under responsibility of King Saud University.

1. Introduction

Solid dosage forms, in particular tablets, dominate the global pharmaceutical landscape. Different dosage forms have changed over time, mainly by the use of excipients, which have distinct functions in formulations (Sastry et al., 2000). Many difficulties in the manufacture of tablets, however, are still quite common in factories around the world. Among these difficulties is the adhesion of tablets to the tools of compression machines. Adhesion may have different causes

^jjfl I

Elsevier I Production and hosting by Elsevier

1319-0164 © 2013 Production and hosting by Elsevier B.V. on behalf of King Saud University. http://dx.doi.org/10.1016/jjsps.2013.08.002

and consequences of varying degrees, and can lead to failure in entire production batches. There are few research studies on this subject in the literature. This problem is typically resolved by trial and error, although has been reported that the use of more advanced techniques, such as determining the force exerted by the punch and dies, can predict this phenomenon (Fuhrer, 1996; Picker-Freyer, 2008). Evaluation of force exerted is of great value, since a commonly accepted hypothesis for adhesion is the interaction between the powder particles and the metal surface of punches.

Several factors may be related to the origin of adhesion, including: manufacturing process (adjustment of the compres-sive force and of the contact area between tools and formulation powder), formulation (use of lubricants and other excipients), conditions of blends (humidity) and equipment (integrity and cleanliness of the punches) (Fuhrer, 1996; Picker-Freyer, 2008). In a manufacturing process, adhesion can be seen as a serious or critical deviation, depending on the amount of mass lost (Alderborn, 2005). In an attempt to explain this deviation, three possibilities were evaluated: (a) change in the compression process, (b) drying the blend to reduce its humidity and (c) the influence of lubricant on the compression process.

Magnesium stearate is the most widely used lubricant in the pharmaceutical industry in manufacturing of tablets and capsules, due to its capacity to reduce friction. However, it should be noted that the lubricant amount can significantly affect the product performance and quality, causing problems such as: decrease in content uniformity, decrease in tablet hardness, increase in tablet disintegration time, decrease in dissolution rate and decrease in bioavailability. Thus, it is recommended to use the minimum quantity of lubricant required (Andres et al., 2001; Wada and Matsubara, 1994). A large number of studies have reported that variations in physical and chemical properties of magnesium stearate have great influence on its lubricating action. Currently, it is known that variables such as molecular structure, crystallinity, water content, thermal stability and granular properties are able to influence the functional properties of the material in question (Bracconi et al., 2005; Wang et al., 2010). Particle size and specific surface area of magnesium stearate may be key factors influencing its lubrication efficiency (Wang et al., 2010).

It is important to note that materials supplied by different manufacturers are unlikely to be of exactly the same physical properties, but lot to lot variability of materials obtained from the same manufacturer is less likely to present a problem (Wang et al., 2010). Another parameter that can influence compression is the moisture of the powder blend and also the relative humidity of the manufacturing environment. It is necessary to identify the moisture range in which the particu-late material shows good performance to carry out the compression process. If the relative humidity can increase the rate and extent of water absorption by the formulation, it is very likely that adhesion will be evident in the final product. Furthermore, the humidity has a direct influence on the lubricant added to the formulation (Alderborn, 2005).

With regard to magnesium stearate, it is known that its functionality is based on the fact that water and air penetrate into spaces between crystalline particles, increasing the movement of these particles. This mechanism reduces the forces necessary to break the crystalline structure of magnesium stearate, which facilitates their spread on the surface to be lubricated.

However, the crystalline form found in commercial batches of magnesium stearate depends on the process for preparing the excipient and on the humidity to which the material was exposed after the manufacturing process (Rajala and Laine, 1995; Wang et al., 2010).

The adhesion of tablets to the tools of compression machines was observed by Farmanguinhos, a pharmaceutical laboratory linked to the Brazilian Ministry of Health, during the manufacture of glibenclamide 5 mg tablets. Therefore, the technical group developed a research in attempt to solve the case and to continue to manufacture the product without problems.

2. Materials and methods

2.1. Materials

Glibenclamide (Cadila Pharmaceuticals LTDA), mannitol powder (Launcher International), sodium lauryl sulfate (Cog-nis Brasil LTDA), microcrystalline cellullose 102 (Blanver Farmoquimica LTDA), silicon dioxide colloidal (Cabot Corporation) e sodium croscarmellose (Blanver Farmoquimica LTDA). Magnesium stearate was supplied by different manufacturers whose names are not mentioned here.

2.2. Methods

2.2.1. Preparation offormulations

Table 1 shows all the components used for manufacturing gli-benclamide 5 mg tablets. Glibenclamide and magnesium stearate concentrations were 4.17% and 1%, respectively. The concentrations of other components were in accordance with the recommendations of official compendium of the area.

The manufacturing process was carried out by direct compression, i.e., the active ingredient and the excipients were mixed and then, directly compressed into tablets. The powder blend was added to a "V" blender (Lawes®) and compression occurred in a 35 punch rotary tablet machine (Manesty® BB4). All tests were performed under the same environmental conditions in the manufacturing rooms, with temperature about 20 0C and relative humidity about 33%. Thus, these parameters were disregarded.

2.2.2. Tests of changing the compression parameters

All tests were conducted with a low rate of compression and with an increased pressure, i.e., higher compression force resulting in tablets with higher tensile strength. The same blend of the industrial batch with the problem of adhesion was used. The tablet hardness range was fixed from 14.0 to 19.0 Kp.

Table 1 Components of Glibenclamide 5 mg formulation.

Formulation Function

Glibenclamide Mannitol powder Sodium lauryl sulfate Microcrystalline cellulose 102 Colloidal silicon dioxide Magnesium stearate Sodium croscarmellose Active principle (hypoglycemic) Diluent Surfactant Diluent/agent compression Flow enhancer/absorbent Lubricant/nonstick/sliding Disintegrant

2.2.3. Tests with decreasing blend moisture

The same blend of the previous test was used, but it was oven dried until reaching the moisture content of 2.2%, considering the initial moisture of 3.6%. The dried blend was compressed and it was observed the appearance of adhesion to it.

2.2.4. Tests with different lubricant manufacturers and assessment of blend storage time

The blend that generated the problem of adhesion during the industrial manufacture was used; therefore it contained the same batch of magnesium stearate. To evaluate the influence of magnesium stearate batches, three formulations sets (1, 2 and 3) were produced, each one composed of 5 formulations (FMS1, FMS2, FMS3, FMS4 and FMS5). The variation among them was only the time between the powder blending and compression. In group 1, formulations were blended and compressed on the same day; in group 2 compression occurred 7 days after; group 3 had a period of 14 days between blending and compression.

2.2.5. Preparation of powder blend

The quantitative formulation (drug and excipient composition) used was the same for all tests. Each batch had a total mass of 2 kg. All raw materials were sieved through a 16 mesh sieve. The powders were added in a V blender at a rotation speed of 66 rpm for 15 min. Then, magnesium stearate was added and blending continued for 5 min more. At the end of this step, samples from different batches were taken and evaluated by the following tests: loss on drying, particle size (sieving), flow measurement, angle of repose and bulk (loose) and tapped (compressed) densities.

2.2.6. Evaluation of powder blend properties

Powder blend properties were evaluated by different tests, such as particle size and size distribution, flow index, angle of repose, Hausner ratio and Carr's index and loss on drying.

2.2.6.1. Powder particle size and size distribution. Particle size determination (powder particle size and size distribution) was performed as described in the Brazilian Pharmacopoeia 5th ed. (2010). This test is a mechanical analysis, which uses a shaker to vibrate the sieves. Standardized sieves (80, 100, 115, 150, 200 and 250 mesh) were stacked on top of one another. An amount of 25 g was added to the larger mesh sieve and submitted to vibration for 60 min. Calculations were performed to determine the powder formulation homogeneity and representativeness of each particle size.

2.2.6.2. Apparent and tapped densities. For bulk and tapped volumes determination, 10.0 g of powder was added in a 100.0 mL graduated cylinder properly calibrated. The volume occupied was measured and to it was given the name of apparent (bulk) volume. Bulk density was calculated as the ratio of the mass to its apparent volume. After observing the initial volume, the cylinder was mechanically tapped, using Erweka SVM22 apparatus, and volume readings were taken. Vertical movements occurred at a fixed drop of 3 mm (±10%) and at a nominal rate of 250 taps per minute (Picker-Freyer, 2008). After 500 and 750 taps, the corresponding tapped

volume was read. Only when the difference between V500 and V750 was smaller than 2%, 500 additional taps were performed. Tapped density was determined as the ratio of the mass to its final tapped volume.

2.2.6.3. Flow properties. Evaluation of blend flow properties was based on four parameters: angle of repose, Hausner ratio, Carr's (compressibility) index and flow through an orifice, all listed and recommended by USP 32nd ed. (2009). A granulate tester Erweka GTB, which provides an automated measurement of the angle of repose, was used. In order to determine the Hausner ratio (HR) and the compressibility index (IC), it was needed to know the apparent (Vi) and tapped (Vf) volumes, defined according to methodology described in previous item. Hausner ratio and compressibility index were calculated using Eqs. 1 and 2, respectively.

HR ^ Vf (1)

Vf - Vi x 100

The flow rate of each formulation was also studied using a flow tester equipment Erweka GTB. The powder blends were poured through funnels with opening diameters of 15 and 25 mm; the first one agitation (speed 3) was used. The funnel was opened and the time needed for 100 g to flow was measured (USP, 2009).

2.2.6.4. Determination of loss on drying. The moister content was determined by measuring the loss on drying using an infrared radiation analyzer Gehaka IV2000. Samples were homogeneously distributed on aluminum plates and heated by infrared radiation to constant weight (Brazilian Pharmacopeia, 2010). The endpoint determination was performed in the automatic mode.

2.2.7. Manufacturing and quality control of the tablets

2.2.7.1. Manufacturing of glibenclamide tablets. Glibenclamide 5 mg tablets were produced by direct compression using a rotary tablet machine (Manesty®/BB4) with a 7.0 mm round flat punch.

2.2.7.2. Quality control of glibenclamide tablets. 2.2.7.2.1. Visual characteristics' control. Some visual characteristics evaluated during the production process were shape, color and presence of any foreign material. In addition to visual assessment, optical microscopy was used for better visualization of tablets' surface characteristics. A magnification of 6.25x was used in an optical stereo microscope Olympus SCX9.

2.2.7.2.2. Average weight. For each formulation, samples were taken at the beginning, middle and end of the compression run. Twenty tablets were weighed individually (for each step). The average weight of 120.0 mg was established, with a variation of 7.5% (Brazilian Pharmacopeia, 2010).

2.2.7.2.3. Thickness, diameter and hardness. The thickness, diameter and hardness of tablets were simultaneously determined using a specific equipment Erweka ZT71. For each formulation, 10 tablets were sampled at the beginning, middle and end of the compression run (Brazilian Pharmacopeia, 2010), in a total of 30 tablets for each step. In this study, tablets'

hardness range was held from 4.0 to 12.0 kgf.

2.2.7.2.4. Friability test. Twenty tablets were weighed and placed in the friability tester Erweka that revolved at 25 rpm for 4 min. After this time, the tablets were weighed and the percentage loss was determined. No cracked, cleaved or broken tablets can be present at the end of the test. A maximum weight loss from the samples of not more than 1.5% was considered acceptable (Brazilian Pharmacopeia, 2010).

2.2.7.2.5. Disintegration test. Disintegration test was conducted as described in the Brazilian Pharmacopoeia 5th ed. (2010), using an automatic disintegrator Erweka ZT71. The time limit established was 15 min and the averages of disintegration time were used.

2.2.7.2.6. In vitro dissolution studies. Dissolution tests were conducted, according to the previously described procedure (Qureshi and McGilveray, 1999), at 37 ± 0.5 0C in 900 mL of phosphate buffer solution pH 7.40 ± 0.05 at 75 rpm. A USP dissolution apparatus II (paddle method) (Distek 6100) was used for studies. Aliquots of 10 mL were withdrawn at 10, 20, 30, 60, 90 and 120 min and immediately assayed using a UV Shimadzu UV-1800 spectrophotometer at a wavelength of 227 nm. Cumulative percentages of the dissolved drug from the tablets were calculated and plotted versus time. Dissolution studies for each formulation were performed in six replicates. The dissolution profiles were compared through a model independent method, using the difference factor (f1) and the similarity factor (f2), Eqs. 3 and 4, respectively.

f1 = 100

ËlRt - T

f2 = 50 log.

1) t R

Figure 1 Microphotographs of Glibenclamide 5 mg tablets. (A) Intact tablet, with no signs of adhesion and (B) Glibenclamide 5 mg tablet with significant grip on the top.

Table 2 Physical test results of glibenclamide 5 mg tablets (n = 30) for formulations varying from the manufacturer of magnesium

stearate.

Test Formulation

FMS1 FMS2 FMS3 FMS4 FMS5

Average weight (mg)

1 120.00 ± 2.60 120.36 ± 2.67 121.16 ± 3.63 119.60 ± 2.99 120.93 ± 3.39

2 122.03 ± 4.03 120.36 ± 1.66 120.53 ± 1.67 122.16 ± 1.26 120.40 ± 1.68

3 121.21 ± 1.45 120.21 ± 1.07 120.31 ± 1.42 120.43 ± 1.12 120.30 ± 1.27

Hardness (kgf)

1 8.45 ± 0.24 7.78 ± 0.63 8.26 ± 0.46 8.25 ± 1.27 8.04 ± 1.09

2 8.42 ± 0.55 7.73 ± 1.03 7.73 ± 1.03 8.45 ± 0.55 7.24 ± 0.57

3 7.84 ± 0.35 8.95 ± 0.32 7.81 ± 0.27 7.92 ± 0.41 7.95 ± 0.34

Friability (%)

1 0.07 ± 0.05 0.05 ± 0.05 0.04 ± 0.00 0.09 ± 0.04 0

2 0 0 0 0 0.12 ± 0.04

3 0.03 ± 0.02 0.02 ± 0.04 0.03 ± 0.02 0.01 ± 0.02 0.03 ± 0.02

Disintegration (seg)

1 63.11 ± 18.61 27.50 ± 7.93 43.33 ± 15.77 43.33 ± 16.69 35.44 ± 20.56

2 35.44 ± 15.48 19.66 ± 9.98 45.66 ± 25.09 33.55 ± 15.33 15.50 ± 10.11

3 35.22 ± 13.07 20.88 ± 17.91 23.22 ± 17.71 25.66 ± 14.78 26.77 ± 16.27

Particle .size (im)

1, 2 e 3 15.83 64.22 87.62 85.30 82.73

3. Results and discussion

The phenomenon of adhesion is quite common in the manufacture of tablets. Blending, granulation, compression and coating steps increase the interaction forces between the active ingredients and excipients and this can be problematic during the compression process because it can increase the adhesion between tablets and tools of compression machines. In time consuming manufacturing processes, in which the contact between the metal surfaces of the dies and punches with the powder is exacerbated, adhesion can become more evident. There were addressed three possibilities for solving the problem of glibenclamide 5 mg tablet adhesion: increase hardness, dry the blend and use magnesium stearate supplied by different manufacturers.

Moisture adsorption by solid forms determines the need to control humidity during production and storage of medicines. Virtually moisture interacts with a pharmaceutical solid during all production steps (Miller and York, 1985). The two fundamental forces that can affect flow properties of powders are friction and cohesion (Miller and York, 1985). Cohesion is the mutual attraction and resistance to separation of contacting powder of similar material. With regard to pharmaceutical formulations, interactions occur not only between individual particles of active ingredients, but also between them and the excipients present in formulations.

Moisture can influence the strength of interaction between solid particles in at least three ways: (a) can be adsorbed in the surface and influence the surface energy, (b) can change the surface conductivity and thus the particle electrostatic charge and (c) can precipitate in regions adjacent to capillaries where there is a real contact (Hiestand, 1996).

Both alternatives, increased hardness and drying the final blend, were able to solve the adhesion problem. Because drying the blend was effective, it was evident that the humidity had an impact on the observed adhesion and this moisture must be coming from the manufacturing process. However, it should be considered that the method used was direct compression and that the environment humidity was adequate. Thus, this variable is dependent only on the amount of moisture in the formulation components. The increase in hardness due to lower machine speed and increase in compression strength also showed an improvement in the appearance of adhesion of the tablets. This alternative, however, aggravates the process productivity.

The increase in tablet hardness was primarily evaluated by reducing the manufacturing speed and increasing the compression force applied to the tablets. The test results were effective, eliminating the problem of adhesion. The values obtained with the new specifications were hardness of 17.09 kp, thickness of 2.50 mm, diameter of 6.88 mm, friability of 0% and disintegration in 67 s. Despite the good results obtained, it should be mentioned that the use of high hardness value fixed for compression, is not a good alternative for the pharmaceutical industry, because production can be impaired due to the slower process.

The drying of the powder blend before compression was carried out in a drying oven until reaching moisture of 2.2%. After the drying process, it was not observed the adhesion of tablet and furthermore lower hardness values were obtained. The results showed hardness of 7.89 kp, thickness of 2.60 mm, diameter of 6.89 mm, friability of 0.12% and disintegration in 21 s. It is noteworthy that for both tests the results were expressed according to official recommendations.

Table 3 Results of the assessment of the flow of mixtures prepared.

Test Formulation

FMS1 FMS2 FMS3 FMS4 FMS5

Angle of repose (°)

1 45.13 ± 0.65 44.63 ± 0.16 45.06 ± 1.06 44.33 ± 0.61 44.60 ± 0.32

2 44.66 ± 0.04 45.80 ± 0.96 45.16 ± 0.59 44.83 ± 1.00 44.20 ± 0.99

3 44.96 ± 0.05 44.80 ± 0.37 45.76 ± 0.09 45.30 ± 0.65 46.06 ± 0.30

Apparent density (g/mL)

1 0.45 0.50 0.41 0.47 0.41

2 0.43 0.41 0.50 0.45 0.43

3 0.41 0.47 0.41 0.50 0.45

Bulk density (g/mL)

1 0.55 0.58 0.50 0.55 0.50

2 0.52 0.50 0.50 0.52 0.50

3 0.50 0.55 0.50 0.50 0.52

Hausner ratio

1 1.22 1.17 1.20 1.16 1.20

2 1.21 1.20 1.05 1.10 1.21

3 1.20 1.16 1.20 1.20 1.21

Carr's index

1 18.18 15.00 16.66 14.28 16.66

2 17.39 16.66 5.00 9.09 17.39

3 16.67 14.27 16.66 16.66 16.66

Table 4 Results of flow through orifice of mixture loss on drying (in seconds/100 g).

Group Formulationsa

FMS1 FMS2 FMS3 FMS4 FMS5

15 mm 25 mm 15 mm 25 mm 15 mm 25 mm 15 mm 25 mm 15 mm 25 mm

1 12.7 3.0 13.3 3.0 13.9 4.3 13.4 3.3 12.9 3.7

2 12.6 2.9 13.3 2.7 13.4 4.1 13.4 3.5 13.5 3.6

3 12.8 2.8 12.8 2.8 12.0 3.6 12.4 3.4 13.1 3.4

a Tests performed with opening of 15 mm were with agitation speed in 3 and were performed with 25 mm without shaking.

Table 5 Results of flow through an orifice (in seconds/100 g).

Group Formulations

FMS1 FMS2 FMS3 FMS4 FMS5

1 5.7 4.7 5.2 4.4 6.1

2 4.4 3.6 3.7 4.3 4.4

3 4.7 3.6 3.7 3.7 3.7

Figure 2 Dissolution profiles of glibenclamide tablets with different magnesium stearate batches.

As a third possibility, some batches were produced with magnesium stearate supplied by different manufacturers. In the specific case of the formulation under study, magnesium stearate is the lubricant and its properties can have a direct impact on the final properties of the tablets produced. Excipients have been widely discussed in the literature, especially those with lubricant action. No report, however, was published (in the research carried out) showing in detail the characteristics necessary for a better lubricant action. It is known that magnesium stearate has different degrees of hydration (Ertel and Carstensen, 1988; Swaminathan and Kildsig, 2001), crystal structure (Ertel and Carstensen, 1988), fatty acid composition (Sharpe et al., 1997), and surface area (Swaminathan et al., 2006), among other important features to its lubricating action.

It is of great importance for the pharmaceutical industry to establish very rigorous specifications to avoid gaps in quality of medicines manufactured. It is known that small variations in the characteristics of the active ingredients and excipients

can have direct effect in the manufacturing process (Wada and Matsubara, 1994), stability (Airaksinen, 2005) and bioavailability (Andres et al., 2001; Wada and Matsubara, 1994) of medicines.

Pharmaceutical raw materials specification must conform to materials performance in a real situation of production. Thus, the basic formulation described in Table 1 was used varying only the quality of magnesium stearate with the aim of evaluating different samples and determine the most adequate one to be used in the industrial manufacturing of this product. Five different lubricant manufacturers were evaluated and the excipients showed to be physical and chemically different (results not presented here).

The main objective was to evaluate the compression performance of each formulation and, consequently, each lubricant. Other work is under development to establish the structure-performance correlation.

The main objective of these tests was to evaluate the adhesion to the tablet tools, and also check the different

physical properties presented by tablets produced with different manufacturers of magnesium stearate. The results of the physical tests performed are shown in Table 2. Tablets were also analyzed by microscopy for better visualization of adhesion points in tablets. The microphotographs are shown in Fig. 1. The adhesion area is indicated with a red arrow.

It is a common practice in the pharmaceutical industry to prepare blends for batches of a campaign to manufacture a particular product. These blends were not compressed immediately, but only a few days after. During this time, some changes may occur, including an increase in the amount of water adsorbed on the surface of the powder blend. This adsorption can lead to adhesion of the tablet to tools. To evaluate the impact of storage time on the adhesion results, blends prepared with different manufacturers of magnesium stearate were stored for 7 and 14 days, after this period they were compressed. These periods were selected in order to simulate the expected time for compression in routine production.

According to above observations, all values found in process control tests (average mass, disintegration, friability and hardness) were within the established parameters. The formulation that showed a significant adhesion was FMS2 - group 2, however it should be noted that the FMS1 - group 1 also showed signs of this problem, although less intense. The values of average mass did not present wide variations, even with formulation FMS2, as shown in Fig. 1. Granulometric analysis showed difference in terms of particle sizes of the blends FMS1 and FMS2.

An evaluation of major importance for industrial application is the powder flow. This property has a direct impact on various parameters of the process and in the actual final product (Rios, 2006). Flowability is not exclusively an inherent property of the material but results from a combination of physical properties and equipment used in processing (Prescott and Barnum, 2000). Among the tests used to evaluate flowabil-

ity of pharmaceutical powders are Carr index (or compressibility), Hausner ratio, flow through an orifice and the angle of repose. Of all of them, the flow through an orifice is a more realistic prediction of the material behavior, since it is the only direct analysis of the flowability. The other tests are indirect parameters, which take into account calculations based on other measures.

Magnesium stearate, in addition to its lubricating function, has also a sliding action, which facilitates the flow from hoppers into the die cavity. Thus, this magnesium stearate action should be reflected in powder flow depending on the sliding capacity of this excipient. In the present study, this evaluation was also performed (Table 3).

Table 3 shows the angle of repose obtained for different formulation. The angle of repose can be defined as the constant three-dimensional angle measured relatively to the horizontal base, assumed by a cone-like pile of material formed when the powder is passed through a funnel. An angle of repose lower than 40° indicates good flowability, conversely an angle superior to 40° is an indication of cohesiveness (Sar-raguca et al., 2010), although USP describes other ranges of classification. This test did not allow distinguishing between flow properties of the formulations. The results were very close and there was no correlation between them and adhesion. Hausner ratio and Carr index were more suitable to differentiate between formulations because the values obtained were able to successfully distinguish the flow properties. Furthermore, Hausner ratio and Carr index showed a closer correlation with tablet adhesion.

Flow through an orifice results are reported in Table 4. First, it is important to mention that this specific flow test demands a methodology development work. There are different hole diameters, which allow a more realistic evaluation depending on the flow properties of the evaluated powders. The smaller the hole diameter, the more difficult will be to flow

Table 6 Results of the pairwise comparison of formulation of glibenclamide tablets, employing the difference (f1) and similarity (f2)

criteria.

Formulation /-criteria FMS1 FMS2 FMS3 FMS4 FMS5

FMS1 /1

- 13.96 5.99 5.66 9.29

42.05 51.32 61.20 48.44

FMS2 /1

13.96 - 20.51 8.75 16.26

42.05 34.69 53.99 35.44

FMS3 /1

5.99 20.51 - 9.76 7.86

51.32 34.69 45.61 51.21

FMS4 /1

5.66 8.75 9.76 - 7.25

61.20 53.99 45.61 44.66

FMS5 /1

9.29 16.26 7.86 7.25 -

48.44 35.44 51.21 44.66

the powder. During the methodology development, the smaller diameter holes are used, because they can be more discriminatory between different samples. The flow, however, must be uniform, showing no significant variation in the analysis.

In this work, hole diameter of 15 mm was selected to begin the test, but the powder did not flow and portions were retained in the funnel. Thus, a possibility was to use the equipment with a shaker, which facilitates the flow by revolutions at a constant rate. The results with agitation were very reproducible. In order to evaluate a configuration that would analyze the flow without the need for agitation, which would be a more "real" assessment of the powder intrinsic flowability, it was used a 25 mm opening. The results were quite reproducible and discriminatory. Thus, this configuration should be adopted for future analysis of this material. Moreover, the test also allowed discrimination between the samples that did and did not presented good results during compression. Those that adhered to the punches (FMS1 and FMS2) presented a lower flow rate than those that did not adhered. Therefore, this test, for its simplicity, proved to be an interesting evaluation for the magnesium stearate performance, complementing its functionality.

Table 5 shows the values of loss on drying of the blends. A considerable change in moisture with time was observed, but the results were not uniform. Some values decreased, while others increased. Within group 2, FMS2 presented the highest adhesion, and there was no correlation between the values of loss on drying with the adhesion of the blends.

In vitro dissolution test is a valuable tool to identify the influence of key formulation and manufacturing factors, like excipients, binder and mixing effects. Therefore, dissolution testing plays an important role, providing better control of production process and assuring consistent quality of products (Maggio et al., 2008). The results of in vitro dissolution studies showed the influence of magnesium stearate characteristics in dissolution profile (Fig. 2).

The corresponding f1 and f2 values were calculated using data acquired at all points. All possible comparisons were carried out and the results are given in Table 6.

When comparing FMS1 against FMS3 and FMS4, one can see that the difference (5.99 and 5.66, respectively) and the similarity factors (51.32 and 61.20, respectively) indicated that their dissolution profiles were similar. FMS2 and FMS4 exhibited acceptable difference and similarity factors (8.75 and 53.99, respectively). Finally, comparing FMS3 to FMS5, both difference and similarity factors (7.86 and 51.21, respectively) were acceptable, indicating that their dissolution profiles should be considered similar (Table 6). The other comparisons showed unacceptable difference and/or similarity values, indicating that these formulations were not similar. Therefore, in some cases the use of magnesium stearate from different manufacturers affected the dissolution of glibenclamide tablets.

4. Conclusion

This study presented various aspects of tablet adhesion to compression machine punches. The case was glibenclamide 5 mg tablets in an industrial situation. Three possibilities were proposed, and the implementation and refinement of the specifications of the lubricant magnesium stearate was the one more feasible. The other two (increased hardness and drying

the blend) are prone to reduce productivity, and lead to changes in the process already registered in the Brazilian Health Surveillance Agency (ANVISA) of Brazil. The results showed that the adhesion was directly related to the characteristics of magnesium stearate from different manufacturers, and the feasibility of evaluating powder flow characteristics by different techniques that are not routinely followed in various pharmaceutical companies. The results also contributed to the development of excipient functionality studies. In vitro dissolution tests showed that the magnesium stearate manufacturer can influence on the dissolution profile of glibenclamide tablets. Troubleshooting approaches can be, most of times, conducted based on previous experience, or an experimental research needs to be implemented in order to have confident results.

Acknowledgements

Farmanguinhos-FIOCRUZ for granted the stage of Janine Boniatti. The entire staff of the Laboratory of Pharmaceutical Technology and the technical support of Altivo Pitaluga and Rafael Seiceira from Laboratory of Solid State Studies of this institution. The sincere thanks to Professors Valeria Weiss Angeli and Marco Aurelio Dorneles from University of Caxias do Sul for the support and partnership.

References

Airaksinen, S., 2005. Role of Excipients in Moisture Sorption and Physical Stability of Solid Pharmaceutical Formulations. Academic Dissertation, University of Helsinki. Alderborn, G., 2005. Tablets and compaction. In: Aulton, M.E. (Ed.), Pharmaceutics - The Science of Dosage Form Design, 2nd ed. Churchill Livingstone, London, pp. 197-210. Andres, C., Bracconi, P., Pourcelot, Y., 2001. On the difficulty of assessing the specific surface area of magnesium stearate. Int. J. Pharm. 218 (1-2), 153-163. Bracconi, P., Andres, C., N'Diaye, A., Pourcelot, Y., 2005. Thermal analyses of commercial magnesium stearate pseudopolymorphs. Thermochim. Acta 429 (1), 43-51. Brazilian Pharmacopeia, 2010. Brazilian Pharmacopeia, fifth ed.

Atheneu, Sao Paulo. Ertel, K.D., Carstensen, J.T., 1988. An examination of the physical properties of pure magnesium stearate. Int. J. Pharm. 42, 171-180. Fuhrer, C., 1996. Interparticulate bonding characteristics of pharmaceutical compacts. In: Alderborn, G., Nystrom, C. (Eds.), . In: Pharmaceutical Powder Compaction Technology. Drugs Pharm. Sciences, vol. 71. Marcel Dekker, pp. 1-16. Hiestand, E.N., 1996. Powder: particle interactions. J. Pharm. Sci. 55, 1325-1343.

Maggio, R.M., Castellano, P.M., Kaufman, T.S., 2008. A new principal component analysis-based approach for testing "similarity" of drug dissolution profiles. Eur. J. Pharm. Sci. 34 (1), 66-77. Miller, T.A., York, P., 1985. Physical and chemical characteristics of some high purity magnesium stearate and palmitate powders. Int. J. Pharm. 23, 55-67. Picker-Freyer, K.M., 2008. Tablet production systems. In: Gad, S.C. (Ed.), Pharmaceutical Manufacturing Handbook - Production and Processes. John Wiley & Sons, New Jersey, pp. 053-1058. Prescott, J.K., Barnum, R.A., 2000. On powder flowability. Pharm.

Technol. 24 (10), 60-85. Qureshi, S.A., McGilveray, I.J., 1999. Typical variability in drug dissolution testing: study with USP and FDA calibrator tablets and

a marketed drug (glibenclamide) product. Eur. J. Pharm. Sci. 7, 249-258.

Rajala, R., Laine, E., 1995. The effect of moisture on the structure of magnesium stearate. Thermochim. Acta 248, 177-188.

Rios, M., 2006. Developments in powder flow testing. Pharm. Technol. 30, 38-49.

Sarraguca, M.C., Cruz, A.V., Soares, S.O., Amaral, H.R., Costa, P.C., Lopes, J.A., 2010. Determination of flow properties of pharmaceutical powders by near infrared spectroscopy. J. Pharm. Biom. Analysis 52 (4), 484-492.

Sastry, V.S., Nyshdham, J.R., Fix, J.A., 2000. Recent technological advances in oral drug delivery - a review. Sci. Technol. Today 3, 138-145.

Sharpe, S.A., Celik, M., Newman, A.W., Brittain, H.G., 1997. Physical characterization of the polymorphic variations of magnesium stearate and magnesium palmitate hydrate species. Struct. Chem. 8, 73-84.

Swaminathan, V., Kildsig, D.O., 2001. An examination of the moisture sorption characteristics of commercial magnesium stearate. AAPS PharmSciTech 2 (4), 73-79.

Swaminathan, V., Cobb, J., Saracovan, I., 2006. Measurement of the surface energy of lubricated pharmaceutical powders by inverse gas chromatography. Int. J. Pharm. 312, 158-165.

United States Pharmacopeia (USP) National Formulary, 2009. Ed. 32 - NF. U.S. Pharmacopeial Convention, Rockville.

Wada, Y., Matsubara, T., 1994. Pseudopolymorphism and lubricating properties of magnesium stearate. Powder Technol. 78 (2), 109114.

Wang, J., Wen, H., Desai, D., 2010. Lubrication in tablet formulations. Eur. J. Pharm. Biopharm. 75 (1), 1-15.