University of Szeged

Faculty of Pharmacy

Department of Pharmaceutical Technology

Head: Prof. Dr. Habil. Piroska Szabó-Révész DSc.

and

Gedeon Richter Plc.

Doctoral dissertation

COMPARISON OF THE EFFECT OF GRANULATION AND DRYING

TECHNIQUES ON THE QUALITY OF A PHARMACEUTICAL PRODUCT

WITH A HIGH ACTIVE INGREDIENT CONTENT

By

Ágota Hegedűs

Pharmacist

Supervisor:

Prof. Dr. Habil. Klára Pintye-Hódi DSc.

Szeged

2007

Publications

I. Kelen Á., Hegedűs Á., Nagy T., Máthé Z., Hódi K.: A mikrohullám alkalmazásának

előnye hőérzékeny agglomerátumok szárítása esetén, Acta Pharm. Hung. 2003, 73, 65-70

II. Kelen Á., Ress S., Nagy T., Hegedűs Á., Bódis A., Erős I., Hódi K.: Mikrohullámú

vákuumszárítás során kialakuló hőeloszlás követésének lehetősége, Acta Pharm. Hung.

2005, 75, 17-22

III. Hegedűs, Á., Kelen, Á., Pintye-Hódi, K.: The effect of different drying techniques on the

porosity parameters of granules at production scale, Eur. J. Pharm. Sci. 2005, 25/Suppl.

1, S114-115

IV. Hegedűs, Á., Pintye-Hódi, K.: Comparison of the effects of different drying techniques

on properties of granules and tablets made on a production scale. Int. J. Pharm. 2007,

330, 99-104

V. Hegedűs, Á., Pintye-Hódi, K.: Influence of the type of the high-shear granulator on the

physico-chemical properties of granules. Chem. Eng. and Processing. 2007, 46, 1012-

1019

Abstracts

I. Kelen Á., Hegedűs Á.: A mikrohullámú vákuumszárítás előnye a kritikus

nedvességtartalomnál alacsonyabb nedvességtartalom elérésében (E-25). XIV Országos

Gyógyszertechnológiai Konferencia, Hévíz, 2002

II. Hegedűs Á., Máthé Z., Kelen Á., Bódis A.: Technológiai megújítási lehetőségek

örvényáramú granulálás esetén (P-46). Congressus pharmaceuticus XII., Budapest 2003.

III. Kelen Á., Hegedűs Á., Máthé Z., Nagy T., Bódis A., Hódi K.: A mikrohullám

alkalmazása hőérzékeny agglomerátumok szárítása során (E-58). Congressus

pharmaceuticus XII., Budapest 2003.

IV. Kelen Á., Hegedűs Á., Máthé Z., Angyal N., Nagy T., Bódis A., Hódi K.: A konvenkciós

és mikrohullámú energiaközlés a gyógyszeripari vákuumszárítás során (E-6).

Gyógyszerkémiai és Gyógyszertechnológiai Szimpózium, Eger 2003.

V. Máthé Z., Kelen Á., Hegedűs Á., Nagy T., Bódis A.: Optimalizációs paraméterek

meghatározása szilárd gyógyszerformák méretnövelése során (E-9). Gyógyszerkémiai és

Gyógyszertechnológiai Szimpózium, Eger 2003.

VI. Kelen Á., Ress S., Nagy T., Hegedűs Á., Erős I., Hódi K., Bódis A.: Mikrohullámú

elektromágneses tér 3D-os térképezése a gyógyszertechnológiában (P-7). Gyógyszer az

Ezredfordulón V. Konferencia, Sopron, 2004.

VII. A. Kelen, E. Pallai-Varsányi, A. Dávid, A. Hegedus, K. Pintye-Hodi: Select the most

suitable diluent to formulate a “heat sensitive” active encase of microwave vacuum

drying. Eur. J. Pharm. Sci. Vol. 25/S1. 25-27. 2005.

CONTENTS

1. Introduction…………………...…………………………………………..………………....... 1

2. Aim…………………………………………………………………….….………………........ 1

3. Literature survey……………………………………………………………………..……… 2

3.1. Granulation……………………………………………………………..................... 2

3.1.1. Melt granulation.………………………………………………................. 2

3.1.2. Dry granulation..……………………………………………….................. 3

3.1.3. Wet granulation.………………………………………………................... 3

3.2. Drying…………….……………………………………...…………........................ 4

3.2.1. Fluid-bed drying……………………………………………….................. 5

3.2.2. Microwave-vacuum drying…………………………………….................. 6

4. Experimental part…………………………………………………………………………........ 8

4.1. Materials and Equipment………...……………………………………..................... 8

4.1.1. Materials………………………………………………………………….. 8

4.1.2. Equipment..…………………………………………………………...... 10

4.1.2.1. Diosna P400………………………………………………...... 10

4.1.2.2. Collette Ultima Pro 600………………………………………. 11

4.1.2.3. Glatt WSG 200……………………………………………...... 14

5. Test of granules and tablets………………………………………………………………..... 15

5.1. Particle size analysis……………………………………………………………… 15

5.2. Bulk and tapped densities………………………………………………………… 15

5.3. Carr compressibility index……………………………………………………...... 15

5.4. Porosity……………………………………………………………...……………. 15

5.5. Moisture content…………………………………….……………………………. 16

5.6. Scanning electron microscopy (SEM)..………………………………………...... 16

5.7. Tablet evaluation……………………..…………………………………………... 16

5.8. Assay, Blend Uniformity Analysis, purity and dissolution.………..…………..... 17

6. Result and discussion 18

6.1. Influence of the type of the high-shear granulator on the physico-chemical

properties of granules…………………………………………………………………………

18

6.1.1. Manufacturing process………………….……………………………… 18

6.1.2. Particle size analysis………………….…………………………….……. 22

6.1.3. Bulk and tapped densities and Carr compressibility index...………..…. 22

6.1.4. Reproduction………………..……………………………...………..… 25

6.1.5. Tablet evaluation………………..…………………..……...………..… 28

6.1.6. Conclusion……..………………..…………………..……...………..… 28

6.2. The various aspects of granulation and drying……..………………..…………… 29

6.3. Investigation of fluid-bed and vacuum microwave drying..………..…………….. 30

6.3.1. Manufacturing process…………………………………..…………….. 30

6.3.2. Effects of different drying techniques on properties of granules and

tablets made on a production scale……………………………………………………………

33

6.3.3. Conclusions……………………………………………………………. 38

7. Summary…………………………………………………………………………………..... 39

8. References……………………………………………………………………………………. 41

1

1. Introduction

Granulation of powder to produce a pharmaceutical solid-dosage form is an essential

unit operation. Granulation in the pharmaceutical industry poses unique challenges, as it has

the additional requirements of content uniformity and consistent physical properties, such as

particle size, moisture, bulk density, porosity, hardness and compressibility.

In the past 20 years, the pharmaceutical industry has been introduced to a number of

different methods for producing pharmaceutical granulation. These methods have offered a

number of advantages, such as process efficiency, while addressing product quality and

regulatory compliance. Potent compounds can now be granulated in so-called ―one-pot‖

systems, which offer a greater measure of safety to operators by providing a single ‖pot,‖ or

bowl, to granulate and dry product. 1, 2

2. Aim

My objective was to adapt the production-scale granulation process, previously

performed with a traditional high-shear granulator (Diosna P400) and a fluid-bed drier (Glatt

WSG-200), so that it could be carried out in a single-pot high-shear granulator (Collette

Ultima Pro 600). Because of the considerable differences between the two machines, the fact

that my experiments were not preceded by laboratory and pilot tests, and the fact that I was

working on a production scale, I decided to adapt the processes in two stages: first the

granulation step, and then the overall process, including the drying step.

The first aim of this experiment was to compare the granulation results that can be

achieved in different production-scale high-shear granulator models in the case of a product

with a high active ingredient content, and the characteristics and tablet-forming properties of

the granules produced. I studied granules prepared in the Diosna P400 and the Collette Ultima

Pro 600 industrial high-shear granulators. The macroscopic and microscopic textures of the

granules prepared in these machine (which have identical manufacturing capacity) were

examined. The aim was to create, optimize and reproduce a robust technology that furnishes

granules (and the tablets formed from them) with similar physical properties in both sets of

equipment.

The second aim of this study was to compare the properties of granules prepared in the

same manner, in a high-shear granulator (Collette Ultima Pro 600 single-pot processing

equipment) and dried by using different methods (fluid-bed and microwave-vacuum drying)

and to compare the properties of tablets pressed from such granules. Experiments on a

2

production scale were performed with the Collette Ultima Pro 600 single-pot processing

equipment and a Glatt WSG 200 fluid-bed granulator and drier.

3. Literature survey

Historical survey

The term of ―granulated‖ material is derived from Latin word ―granulatum‖, meaning

grained. The granulation material can be obtained by direct size enlargement of primary

particles, or size reduction from dry compacted material.

The development of pharmaceutical granulation was driven by the invention of the tablet

press by W. Brockedon in 1843. The demands on the granulation properties were enhanced in

the 1970’s as high speed tabletting machines with automated controls were introduced. 1

3.1. Granulation

Granulation is a size-enlargement process in the course of which small particles are

formed into larger, physically strong agglomerates in which the original particles can still be

identified. The agglomeration of solid particles renders them more suitable for further

processing, such as tablet formation. Granulations are used primarily for the preparation of

materials for tabletting, but to a lesser extent are used as filler material in the encapsulating

process. 3

The main reasons for granulation can be to:

-Increase the uniformity of the active ingredients distribution

-Improve flowability properties

-Densify the materials

-Ensure optimal particle size distribution and reduce dust.

Granulation methods can be divided into three types: wet methods which utilize a liquid

in the process, dry methods in which no liquid is utilized and melt methods which involve the

use of a material which melts at relatively low temperature. All of these methods are utilized

in the pharmaceutical industry, however, wet granulation technology is more common.

3.1.1. Melt granulation

In the melt granulating process, the substance with a low melting point can be added in

the molten form over the substrate or in a solid form, which is then heated above its melting

3

point by hot air or by a heating jacket. In both cases, after melting, the substance acts like a

liquid binder, thus melt granulation does not require the use of solvents (organic or aqueous).

In melt granulation the drying step is not necessary, thus the process consumes less in terms

of time and energy than wet granulation. 4-11

3.1.2. Dry granulation

In the dry granulating process, dry powder particles may be brought together

mechanically by compression into slugs or more frequently today by roller compaction.

Slugging consists of dry-blending excipiens with an active drug substance and

compressing the powder or powder blend into a large tablet or slug on a compression

machine. After the slugs are formed, they need to be sized for final blending and tabletting

requirements. 1, 3, 12

Briquetting and compaction are densifying techniques for dry powders. Powdered

material is fed into a counterrotating pair of rolls. During the powder feed, the powder

materials rub against the roll pair surfaces and then are drawn into the nip angle area. A

compact is formed that is immediately sized for the intended needs. 13-21

3.1.3. Wet granulation

Wet granulation is a technique whereby a liquid is used to transform small solid

particles into clusters of larger ones, through a process of agglomeration. In pharmaceutical

tablet manufacturing, the agglomeration of solid particles renders them more suitable for

tablet formation.

There are two known closed-system wet granulation procedures: high-shear granulation

and fluid-bed granulation. 22, 23 These techniques differ in the modes of agitation of the

solid particles, and for this reason there are also differences in granule growth. 24-28

In the course of fluid-bed granulation, the powder bed is kept in motion by specially

treated (filtered, temperature and humidity-controlled) air, which is introduced through a

sieve plate in the base of the granulator. The binder solution is sprayed onto the fluidized

powder bed. The granules are created during the wetting of the powder bed, through the

adhesion of solid particles as the drops of liquid reach the powder bed. The agglomeration of

the powder takes place during the wetting process and, once the process of spraying the

adhesive onto the powder bed has been completed, the granules are dried through the use of

warm air. 29-34

4

In high-shear granulation, an impeller is used to agitate the solid particles within a

closed space. The binder solution is added or sprayed in from above. The mixing,

densification and agglomeration of the wet material are performed by the impeller through the

exertion of shearing and compacting forces. The process is ended before the granules begin to

grow uncontrollably, which would result in the phenomenon known as ―ball growth‖. 35, 36

High-shear granulators have long been used in the pharmaceutical industry, both for

mixing and for granulating. Originally, high-shear granulators did not have a drying

capability, which means that the wet granules produced in these machines had to be dried by

using another machine, such as a fluid-bed drier. Later, these granulators were further

developed into what are termed ―single-pot‖ systems 37, 38 , which are capable of

performing all of the processes of mixing, granulation, drying and blending. The possible

drying methods are vacuum, vacuum-microwave and gas-vacuum methods 39 , all of which

can be combined with side-wall heating. 1, 40, 41 ,

3.2. Drying

Drying involves the removal of liquid from solid material that contains moisture,

through a process of evaporation resulting from the application of heat. Thermal energy can

be applied to the granules by convection, conduction or vacuum drying. 2, 29

Convection is achieved by means of a flowing gaseous medium, in which the gaseous

particles transmit heat while in movement. Fluid-bed drying is an example of a convective

drying method. In the process of fluid-bed drying, the granules to be dried are placed in a

device fitted with a perforated screen or sieve, and air is circulated through this layer at a rate

sufficient to lift and separate the granules, which are set in motion and take on what is termed

a fluidized state. The drying occurs as a result of the consequent intensive contact between the

granules and the gaseous drying medium.

Conduction can be attained by heat exchange between adjacent particles of matter, heat

transfer through a jacketed bowl wall and vacuum drying.

In the process of vacuum drying, the material is placed in a vacuum chamber, and the heat

necessary to remove the moisture is applied directly to the solid material.

The process of pure vacuum drying requires a longer drying time, but its undisputed

advantage over other methods is that the drying takes place at a lower temperature, which

could be important when heat-sensitive materials are to be dried 29, 41 . Gas-assisted

5

vacuum drying, and more commonly microwave vacuum drying, allow quicker drying in a

single-pot processor, used consecutively or simultaneously. 42-44

In production-scale pharmaceutical manufacturing, the methods most commonly used to

produce granules are fluid-bed granulation and drying, or a combination of high-shear

granulation and fluid-bed drying. In recent years, however, single-pot technology has grown

in popularity, partly because the transfer of the moist granules from the high-shear granulator

to the fluid-bed dryer is critical. The single-pot equipment has taken the form of a

mixer/granulator retrofitted with a drying unit 40 . The drying unit is capable of pure vacuum

drying, microwave-vacuum drying, gas-assisted vacuum drying, or a combination of

microwave and gas-assisted vacuum drying.

3.2.1. Fluid-bed drying

During fluidized-bed drying moisture or solvent is removed which involves heat and

mass transfer. Heat is transferred to the product to evaporate liquid, and the mass is

transferred as a vapour in the surrounding gas, these two phenomena are interdependent.

Fluidized-bed drying is efficient because the hot air heats the free moisture on the

product’s surface. In most cases, due to its high surface area to air flow ratio, the granulated

compound allows heat to dry the liquid trapped inside the granular matrix. A liquid absorbs

the energy transmitted by the heated air. The energy absorbed by the wet granules results in

evaporation of the free moisture leaving the granule intact. 29

During fluid bed drying, the product passes three distinct temperature phases 1 . At the

beginning of the drying process, the material heats up from the ambient temperature to

approximately the wet bulb temperature of the air in the dryer (Fig. 1. Phase 1). This

temperature is maintained until the granule moisture content is reduced to the critical level

(Fig. 1. Phase 2). At this point, the material holds no free surface water, and the temperature

starts to rise (Fig. 1. Phase 3). The temperature at which moisture condenses is the dew point

temperature, this is the end of the drying process. Figure 1. shows a typical fluid-bed drying

process.

6

Fig. 1. Product temperature changes during drying in fluid bed.

3.2.1. Microwave-vacuum drying

Microwaves are waves of electromagnetic radiation, generated by magnetrons under the

combined action of electric and magnetic forces. Microwave drying is based on the absorption

of electromagnetic radiation by dielectric materials. 1 The dielectric material is placed in an

electromagnetic field, when the material becomes polarized and stores electrical energy

through polarization. The polarization produces displacement of positively and negatively

bound charges in the dielectric material. A distinction must be made between the

displacement polarization of charged particles and the orientation polarization of particles that

have permanent or induced dipole moments. 45-48 The level of polarization depends on the

state and composition of the material and the frequency of the applied electric field. For

pharmaceutical-industry drying, microwaves with a frequency of 2450 MHz (wavelength 12.2

cm) are used. The microwaves are not forms of heat, but rather forms of energy that are

manifested as heat through their interaction with materials. The permittivity ( ) of materials

sensitive to microwaves is complex and comprises two parts, the first corresponding to the

real part ( ’) or relative dielectric constant, and the second representing the imaginary part

( ‖) or loss factor. The dielectric loss factor of a material ( ‖) is a measure of how much heat

is generated inside a material per unit of time when an electric field is applied when subjected

to microwave heating.

The rate of temperature increase as the material absorbs microwave energy is as following:

T / t Pv / Cp

Pv jE2f ‖ or Pv jE

2f ’tan

7

Where: T temperature rise (K); t time (s); Pv power per unit volume (W m-3

);

density (kg m-3

); Cp heat capacity (J kg-1

K-1

); E electrical field strength (V m-1

); f

frequency (Hz); ‖ dielectric loss factor; tan loss tangent or dissipation factor; j

constant; ’ relative dielectric constant or relative permittivity.

Different materials behave differently in the presence of microwaves. The magnitude of the

dielectric characteristics of a system depends on number of factors, including moisture

content, composition, density and temperature. The total loss factor of a wetted material is

derived from the skeleton solid and the bound and the free water. 49 Most of the materials

commonly used in the pharmaceutical industry have a relatively low loss factor and absorb

microwave power only at high field strengths. By comparison, granulation liquids (water or

organic solvents) have high loss factors relative to the dry materials used. 39, 42, 50-56

Figure 2. shows a typical microwave assisted vacuum drying process. 40 Following

granulation (stage 1), the moist granules are dried. In the stage 2, the pressure is decreased to

a vacuum of 40-80 mbar absolute. As the pressure decreases, liquid evaporates and the

product temperature falls. To accelerate drying, microwave energy is radiated into the pot

when the vacuum pressure has been reached (stage 3). As the moisture content of the product

decreases, the temperature of the product rises (stage 4). As the temperature of the product

rises above a specific limit (because residual moisture content is low), the microwave system

is stopped (stage 5). If the residual moisture content is still too high, drying continues at a

reduced vacuum pressure without microwave assistance (stage 6).

Fig. 2. Typical microwave-assisted vacuum drying process.

8

4. Experimental part

4.1. Materials and Equipment

4.1.1. Materials

The given tablets contained 50% w/w metronidazole. The binding solution was an

aqueous solution of Povidone K-30 (4.5% w/w). The other excipients were corn starch (30%

w/w) as diluent; colloidal anhydrous silica (4% w/w) and glycerine (1.5% w/w) as moisture

regulator; and microcrystalline cellulose (7.9% w/w), talc (1.6% w/w) and magnesium

stearate (0.5% w/w) to improve tablet formation. I used the same composition and batch size

(150 kg).

Metronidazole (Ph. Eur.) is a white to yellowish-white crystalline powder. It is practically

insoluble in water and slightly soluble in alcohol. It is a drug frequently used in the treatment

of various anaerobic infections. The drug is useful profilactic in obstetrical and

gynaecological intervention, colorectal surgery and appendectomy. It is also used in the

treatment of susceptible protozoal infections such as amoebiasis, balantidiasis, trichomoniasis

and giardiasis. In anaerobic bacteria and sensitive protozoans the nitro group of metronidazole

metabolises ferredoxin, and the metabolite thus produced causes death of the cells by reacting

with various intracellular macromolecules. A single oral dose is generally 250 mg and the

tablets have a high active ingredient content. 57-63

Table 1. shows the physical properties, Figure 3. shows the constitutional formula of and

Figure 4. shows SEM pictures of Metronidazole.

Table 1. Metronidazole physical properties

Particle size analysis (μm)

D10 25

D50 125

D90 300

Carr compressibility index (%) 21,70

Bulk density (g/100 ml) 71,50

Tapped density (g/100 ml) 91,30

9

Fig. 3. Constitutional formula of Metronidazole

Fig. 4. SEM picture of Metronidazole

Corn starch (Ph. Eur.) consists of amylase and amylopectin, two polysaccharides based on

-glucose. Starch occurs as an odourless, tasteless, fine white powder, comprising very small

spherical granules whose size and shape are characteristic. Starch is used as an excipient

primarily in oral solid-dosage formulations, where it is utilized as a binder, diluent and

disintegrant. It is not water-soluble, but swells. 64-66

Microcrystalline cellulose (Vivapur, Avicel 101) (Ph. Eur.) is a purified, partially

depolymerised cellulose that occurs as a white, odourless, tasteless, crystalline powder

composed of porous particles. It is widely used in pharmaceuticals, primarily as binder/diluent

in oral tablet and capsule formulation, where it is applied in both wet granulation and direct

compression processes. In addition to its use as a binder/diluent, it is also has some lubricant

and disintegrant properties that make it is useful in tabletting. 65, 66

10

Povidone (Polyvinylpyrrolidone, PVP K-30) (Ph. Eur.) is a polymerization product of N-

vinylpyrrolidon. It is a white or yellowish-white powder with a slight amine or ammonia

odour. The K-value indicates the average molecular weight. It is readily soluble in water and

freely soluble in alcohol and many other organic solvents. It has adhesive and binding power

is particularly important in tabletting (wet granulation, dry granulation, direct compression). It

is generally used in the form of a solution, but it can be added to the blends in dry form. This

property is also useful in film coatings. 65- 67

Colloidal anhydrous silica (Aerosil A-200) (Ph. Eur.) is a submicroscopic fumed with a

particle size of about 15 nm. It is a light, loose, bluish-white coloured, odourless, tasteless,

nongritty amorphous powder. It is widely used in pharmaceuticals, its small particle size and

large specific surface area give it desirable flow characteristics which are exploited to

improve the flow properties of dry powders in number of processes, e. g. tabletting. It is also

used as a tablet disintegrant and as adsorbent dispersing agent for liquids in powders. 65, 66

Glycerine (Ph. Eur.) is a clear, colourless, odourless, viscous, hygroscopic liquid, it has a

sweet taste, approximately 0.6 times as sweet as sucrose. It is used in a wide variety of

pharmaceutical formulations including oral preparations, primarily for its humectant and

emollient properties. 65, 66

Magnesium stearate (Ph. Eur.) is a fine, white, precipitated or milled, impalpable powder

of low density, having a faint, characteristic odour and taste. The powder is greasy to the

touch and readily adheres to skin. It is widely used in pharmaceuticals, primarily as a

lubricant in capsule and tablet manufacturing at concentrations between 0.25-5.0 %. 65, 66

4.1.2. Equipment

I performed the granulation in a Diosna P400 (Fig. 5.) conventional high-shear mixer

(granulator) (without drying facility) or a Collette Ultima Pro 600 single-pot equipment (Fig.

6.). Technical data on the two types of equipment can be seen in Table 2.

The drying was carried out in a Glatt WSG 200 fluid-bed granulator and drier (Fig. 7.) or in

Collette Ultima Pro 600 single-pot equipment.

4.1.2.1. Diosna P400

It has a single-wall design. Neither the side walls nor the lid of the device can be

temperature-controlled. The machine is cone-shaped, with the impeller positioned vertically,

and the chopper horizontally. The impeller protrudes into the device from below. The

11

impeller blades and the special shape of the machine ensure effective mixing. Both the

impeller and the chopper have two speed settings, with no fine adjustment. The impeller and

the chopper are fitted with a time switch, which is the only means of setting an end-point.

(There is no measurement of torque or power consumption.) It is not possible to regulate the

application of the binder solution. The quality of the granules depends largely on the skill and

experience of the personnel carrying out the production. (Fig. 5.)

4.1.2.2. Collette Ultima Pro 600

This is a closed, single-pot system, which means that the entire manufacturing process

can be performed in the one device. The bowl has a jacket wall to allow the circulation of hot

or cold water, in order to regulate the temperature of the product. Both the impeller and the

chopper are positioned vertically, and protrude into the machine from above. The speeds of

the impeller and the chopper are adjustable within a given range. The liquid binder addition is

regulated, and the machine is suitable for the spraying of binder solution with high or low

viscosity. A number of parameters can be used to set up the end-point of granulation: the

processing time, the torque, the product temperature, etc., or a combination of these. The

granules can be dried by vacuum and microwave energy, which can be combined with side-

wall heating. The drying cycle of this machine is therefore more energy-efficient than other

drying processes. There are three possible drying methods: vacuum, vacuum-trans flow and

vacuum-microwave. The machine is suitable for computer-controlled, automated

manufacturing. (Fig. 6.) 68, 69

Table 2. Technical data of the Diosna P400 and the Collette Ultima Pro 600

Diosna P400 Collette Ultima Pro 600

Bowl capacity (l) 385 400

Impeller speed (rpm) 64 or 129 from 14 to 135

Chopper speed (rpm) 1450 or 2930 from 600 to 2700

Tip speed (m/s) 3,65 (64 rpm)

7,36 (129 rpm)

3,52 (64 rpm)

7,09 (129 rpm)

12

Fig. 5. Photographs of the Diosna P400 high-shear mixer granulator

13

Fig. 6. Photographs of the Collette Ultima Pro 600 single-pot equipment

14

4.1.2.3. Glatt WSG 200

This is also a single-pot system which is suitable for granulation and drying in the one

device. There is an inlet air handling unit fit for air filtering, air heating, and air cooling. The

air must be introduced at the bottom of the product container through the perforated air

distributor plate (screen type) which is important to fluidize and mix material in the container.

The spraying head with three or six nozzles can be set in three different positions over the

distribution plate. Within the expansion chamber granules are formed. There are bag filters

within the machine which retain the particles. The filter bag is made of polyester-lined

material which is of a certain mesh size. Safety air filters are built in the outlet air product.

Main processes such as air flow and spraying rate are controlled. The machine is equipped

with a data acquisition system. (Fig. 7.)

Fig. 7. Photograph of the Glatt WSG 200 fluid-bed granulator and drier

15

5. Test of granules and tablets

5.1. Particle size analysis

The particle size distribution of an approximately 25 g sample of the final granules was

determined, using a Hosokawa Alpine 200 LS air jet sieve with an array of five sieves. 70,

71

5.2. Bulk and tapped densities

100 ml of granules was poured into a 250 ml graduated tared measuring cylinder, and

the granules were then weighed and their bulk density, t, was determined in g/100 ml. 72

The density of 100 ml of granules of known weight was measured with a Stampfvolumeter

2003 (J. Engelsmann Apparatebau, Ludwigshafen, Germany). After 200–300 taps (when a

constant value had been achieved), the volume of the tapped column of granules was read off,

and the density, T, was determined in g/100 ml. 73-78

5.3. Carr compressibility index

The flow properties of the granules can be determined through compaction, and the

extent of the compaction can be defined through the relationship between the bulk and tapped

densities, which can be expressed with the Carr compressibility index 77-81 , using the

following equation:

100ρ

ρρ(%) index ilitycompressibCarr

T

tT

where T = tapped density

t = bulk density

5.4. Porosity

The properties of granules and tablets are influenced by the porosity of the granules.

Porosity can be defined through the relationship between the particle ( part) and tapped ( T)

densities, using the following equation 74-76, 82, 83 :

100ρ

ρ

part

T )1(

The particle density ( part) was determined with a Stereopycnometer SPY-5 (Quantachrome

Corp.). The pycnometric particle density was determined by measuring the volume occupied

16

by a known mass of powder, which is equivalent to the volume of helium gas displaced. 84,

85 The particle density was calculated via the following equation:

partv

w

where w = weight of sample, and

v = volume of sample.

5.5. Moisture content

The loss on drying of 2 g of granules (homogenized with the external phase) to mass

constancy at 70 °C was determined, with a Mettler Toledo HR 73 halogen moisture analyser.

The loss on drying of the final granules must be within the range of 2.5-4.5%, this range being

suitable for the tabletting of this product.

5.6. Scanning electron microscopy (SEM)

The morphological properties of the granules were examined with a JEOL JSM-5600LV

scanning electron microscope fitted with an energy dispersive X-ray spectrometer. A Polaron

sputter coating apparatus was applied to induce electric conductivity on the surface of the

sample. The air pressure was 1.3-13 mPa. 86, 87

5.7. Tablet evaluation

The granules were pressed into 500 mg tablets by using a Courtoy R190 Ft tablet press

with 36 punches. The rotational speed of the press was 65 rpm. The average and individual

masses, the thickness, the hardness (Pharma Test WHT-2ME) 88, 89 , the friability

(Pharmatest PT-TD) and the disintegration (Pharma Test PTZ-E) were measured five times in

the course of the tablet-formation process. The relative standard deviation (RSD) of the mass

of the individual tablets was determined by measuring 20 tablets. 90

17

5.8. Assay, Blend Uniformity Analysis, purity and dissolution

The assay determination is carried out by spectrophotometric method (UV-VIS) in

distilled water. The absorbancies of the test and reference solutions are measured

spectrophotometrically in 1-cm cells at the wavelength of 320 1 nm, against the blank

solution (distilled water).

Specification: Metronidazole: 250.0 mg / tablet ( 5%)

Blend Uniformity Analysis (BUA) is carried out by assay method.

The test for purity (related substance) determination is carried out by TLC method.

Specification of related substance:

-Total impurities: < 1.0%

-2-methyl-5-nitroimidazole: < 0.5%

- Other individual impurity: < 0.2% each

The amount of active substance dissolved is determined by a spectrophotometric

method. Dissolution conditions:

-Dissolution medium and volume: 0.01 M hydrochloric acid solution, 900 ml

-Apparatus: basket type

-Rotation: 100 rpm

-Dissolution time: 45 minutes

-Temperature: 37°C

Specification: Not less than 80% (Q) of metronidazole should dissolve within 45 minutes.

18

6. Result and discussion

6.1. Influence of the type of the high-shear granulator on the physico-chemical

properties of granules

6.1.1. Manufacturing process

Figure 8. shows the flowcharts of the manufacturing processes in the Diosna P400 and

the Collette Ultima Pro 600.

In the Diosna P400, the binder solution is poured manually onto the powder during the

second step, and its addition is therefore not regulated. In the Collette Ultima Pro 600, in the

course of liquid binder addition and wet massing, I varied five parameters that could affect the

characteristics of the granules. 91 I then compared the granules thus produced, and the

granulating processes used. 92 Table 3 details 11 different combinations of the following

five parameters 93-98 :

(1) Impeller speed, (2) Chopper speed, (3) Water content of the binder solution, (4)

Liquid binder flow rate, (5) Wet massing time

At the beginning of my experiments, I adapted the machine settings (the impeller and

chopper speeds, and the water content of the binder solution) to correspond to those of the

Diosna 400 (Table 3, setting C/0). With setting C/0, I was unable to produce granules in the

Collette Ultima Pro 600. I observed that the product manufacturing processes are not always

transferable (with identical technological parameters) between granulators with the same

production capacity, but with different geometric characteristics.

Through my experiments, I determined the torque values representing the granulation

end-points at various impeller speeds. 25, 99-103 These values are shown in Figure 9. In my

experience, at the torque value associated with an impeller speed of 65 rpm (530 Nm),

aggregation did not occur, and no granules were formed. The explanation for this is that the

motions of the materials differed because of the geometrical differences (with respect to both

the shape and the positioning of the impeller and the chopper) between the two machines,

with the result that the different systems required different impeller speeds to achieve the

same degree of granulation formation. 104-106 In both cases, the granules were dried as

follows:

I discharged the wet granules from the high-shear granulator equipment and loaded them

into the Glatt WSG 200 fluid-bed dryer and performed the drying at 60°C , temperature at the

19

end of the drying: 34°C. After first drying, the granules were sized in a 1.5 mm sieve, and

then followed the drying step at 70 °C to the value of the loss on drying. Dried granules were

homogenized for 2 and 5 minutes with the tabletting excipients (microcrystalline cellulose,

talc and magnesium stearate) in a container blender.

I determined the size distribution of the granules, their tapped and bulk densities, and loss on

drying, and took SEM photographs. The weight variations, thicknesses, hardnesses and

disintegration time of the tablets were examined. Analytical investigations (BUA, assay,

purity, dissolution) were also carried out.

20

Diosna P 400 Collette Ultima Pro 600

Dry mixing

Impeller speed: 64 rpm

Processing time: 6 min

Corn starch

Colloidal anhydrous silica

Metronidazole

Dry mixing

Impeller speed: 65 rpm

Processing time: 6 min

Liquid binder addition: manually

Impeller speed: 64 rpm

Chopper speed: 1,460 rpm

Povidone K-30

Glycerin

Purified water

Liquid binder addition

Impeller speed: 95 rpm

Chopper speed: 600 rpm

Liquid binder flow rate: 3; 7; 12 kg/min

Processing time: 5.5 – 13 min

Wet massing

Impeller speed: 64 rpm

Chopper speed: 2,930 rpm

Processing time: 6 - 8 min

Wet massing

Impeller speed: 65 – 135 rpm

Chopper speed: 1,500 – 2,700 rpm

Processing time 2 – 6 min

Drying: Glatt WSG 200 fluid-bed granulator and dryer

Inlet air temperature: 60°C

Temperature at the end of the drying: 34°C

Sieving: Quadro comil U 20

Rotation speed: 500 rpm, Sieve size: 1.5 mm

Drying: Glatt WSG 200 fluid-bed granulator and dryer

Inlet air temperature: 70°C

Loss on drying: 3 – 4%

Blending I. : Zanchetta Canguro container blender

Process time: 2 min (Talc, Microcrystalline cellulose)

Blending II. : Zanchetta Canguro container blender

Process time: 5 min (Magnesium-stearate)

Tabletting: Courtoy R 190 Ft tablet press

Rotation speed: 65 rpm

Fig. 8. Flow sheet of the granulation in two types of high-shear granulator and

in a fluid-bed granulator and dryer. Setting ranges of the parameters in

Diosna P400 and Collette Ultima Pro 600.

21

Fig. 9. Relationship between the torque value and impeller speed

at the granulation end-points by Collette Ultima Pro 600

Table 3. Setting parameters of the Collette Ultima Pro 600

Run Impeller speed

(rpm)

Chopper speed

(rpm)

Water content

of

the binder

solution

(kg)

Liquid binder

flow rate

(kg/min)

Wet massing

time

(min)

C/0 65 2700 26 7 12

C/1 95 1500 27 7 2

C/2 95 2700 24 7 5

C/3 80 1500 26 7 4

C/4 80 2700 26 7 5

C/5 135 1500 26 7 2

C/6 135 2700 26 7 2

C/7 95 1500 26 7 4

C/8 95 2700 26 3 2

C/9 95 2700 26 12 4

C/10 95 2700 26 7 4

0

100

200

300

400

500

600

700

800

900

0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150

impeller speed (rpm)

torq

ue (

Nm

)

22

6.1.2. Particle size analysis

The rate of granule growth is influenced by the speed of the impeller, the wet massing

time and the amount of binder. 107 I carried out my experiments on a production scale.

In the Collette Ultima 600 machine, the end-point of granulation was the torque

necessary for the given impeller speed. In order to avoid possible variations between the

different batches of active ingredient, which, for a product containing 50% active ingredient,

could result in a substantial divergence. 108 For preparations that granulated well, even with

relatively short wet massing periods, only small differences were detected, even at a variety of

impeller speeds. This means that these two parameters are not of great importance as factors

influencing particle size distribution in the composition studied.

The most important factor influencing the particle size distribution of the granules

proved to be the amount of liquid binder, as may be seen in Table 4. In experiment C/1, in

which the greatest quantity of liquid was used, the highest proportions were those of the

largest particles, i.e. >1000 m (6.4%), and of particles >355 m (58%). In experiment C/2,

where the lowest quantity of binder liquid was used, the proportion of particles <180 m was

(53.8%).

In comparison, the granules prepared in the Diosna P400 granulator displayed a

considerable variance in their particle size distribution, as shown Table 5. The proportion of

particles >1000 m was between 0.9% and 14.8%, while the fraction <180 m varied

between 24.0% and 60.5%. These differences could have been caused by the initial uneven

distribution of moisture, and the subjectivity involved in determining the end-point of

granulation.

6.1.3. Bulk and tapped densities and Carr compressibility index

The Carr compressibility index is widely used to analyse the flow properties of granules.

If it is between 5 and 10, then the granules have excellent flow properties, while values of 12-

16 indicate good, 18-21 acceptable, and 23-28 poor flowability. In this case the Carr

compressibility index indicates weakly or poorly-flowing granules, as the Carr

compressibility index was 18.04. (Table 4.) This result can be considered acceptable. The

Carr compressibility index, with a value of 7.40, showed that the granules produced in

experiment C/7 (a medium quantity of binder solution: 26 kg, a medium flow rate: 7 kg/min, a

medium impeller speed: 95 rpm, and a low chopper speed: 1,500 rpm) had excellent flow

properties. The granules prepared in experiments C/2 (Carr compressibility index 10.72) and

23

C/3 (10.71) were similarly good. The bulk and tapped densities of the granules produced in

experiments C/2 and C/7 were low. For compositions with a high active ingredient content,

relatively high bulk and tapped densities are favourable from the point of view of tablet

formation, since the volume of die filling is proportionally reduced. 82 With respect to the

flow properties of the granules, the settings used in experiment C/3 (a medium quantity of

binder solution: 26 kg, a medium flow rate: 7 kg/min, a low impeller speed: 80 rpm, and a

low chopper speed: 1,500 rpm) yielded the best results.

In comparison, there were no significant differences in the Carr compressibility index

values of the granules prepared in the Diosna P400, given in Table 5. The best Carr

compressibility index was 11.51, but all the granules had good flow properties, with values

ranging between 11.51 and 15.97. The bulk density ranged between 68.49 and 75.76, while

the tapped density varied from 78.47 to 87.50.

24

Table 4. Granules properties in the Collette Ultima Pro 600 with different setting parameters

C/1 C/2 C/3 C/4 C/5 C/6 C/7 C/8 C/9 C/10

Bulk density (g/100 ml) 69,44 68,49 75,76 70,42 70,42 72,46 71,43 69,44 73,53 68,49

Tapped density (g/100 ml) 84,72 76,71 84,85 80,28 82,39 84,06 77,14 81,94 83,82 80,82

Carr compressibility index 18,04 10,72 10,71 12,28 14,53 13,80 7,40 15,26 12,28 15,26

Loss on drying (%)

3,44 2,90 3,15 3,06 3,36 2,99 3,45 3,49 3,55 3,25

Particle size analysis (%)

0,090 mm 18,7 18,6 17,6 21,0 13,6 17,6 15,5 17,1 12,9 19,1

0,090 – 0,180 mm 7,5 35,2 16,2 17,3 19,9 19,4 19,8 18,1 15,5 14,0

0,180 – 0,355 mm 15,8 26,7 29,5 23,5 33,4 33,3 32,0 29,3 24,7 23,1

0,355 – 0,710 mm 33,7 18,3 24,9 29,9 19,1 19,7 21,1 26,8 33,1 29,5

0,710 – 1,000 mm 17,9 1,2 8,9 7,0 8,7 8,0 9,2 8,7 9,5 11,4

1,000 mm 6,4 0 3,2 1,3 5,3 2,0 2,4 0 4,3 2,9

25

Table 5. Granules properties in the Diosna P400

D/1 D/2 D/3 D/4 D/5 D/6

Bulk density (g/100 ml) 73,53 74,63 75,76 69,44 68,49 69,44

Tapped density (g/100 ml) 87,50 86,57 87,88 79,86 80,14 78,47

Carr compressibility index 15,97 13,79 13,79 13,05 14,54 11,51

Loss on drying (%)

3,06 3,10 3,11 3,09 3,04 3,50

Particle size analysis (%)

0,090 mm 11,7 16,0 15,7 19,1 20,8 25,5

0,090 – 0,180 mm 18,8 8,0 9,7 29,8 39,7 16,2

0,180 – 0,355 mm 32,6 14,9 26,2 28,2 18,7 24,6

0,355 – 0,710 mm 16,2 27,5 23,2 15,8 13,7 22,3

0,710 – 1,000 mm 11,6 18,8 14,4 6,2 5,9 8,5

1,000 mm 9,1 14,8 10,8 0,9 1,2 2,9

6.1.4. Reproduction

I performed my experiments with production-scale machinery. Reproducibility is important,

and a prerequisite for validation in the pharmaceutical industry. For this reason, setting 10 was

selected as the medium value to manufacture a further 5 batches in the Collette Ultima Pro 600 (the

results are shown in Table 6.), and the granules with those prepared in the Diosna P400 were

compared.

Table 6. Granules properties of reproduction batches in the Collette Ultima Pro 600

R/1 R/2 R/3 R/4 R/5 R/6

Bulk density (g/100 ml) 68,49 69,44 69,44 69,44 71,43 69,44

Tapped density (g/100 ml) 80,82 82,64 81,94 80,55 84,29 81,94

Carr compressibility index 15,26 15,97 15,26 13,79 15,26 15,26

Loss on drying (%)

3,00 2,96 3,45 3,42 3,35 2,70

Particle size analysis (%)

0,090 mm 19,7 16,2 20,6 20,2 18,9 19,7

0,090 – 0,180 mm 20,4 15,6 19,2 15,9 15,7 14,7

0,180 – 0,355 mm 26,7 27,0 25,4 24,7 24,8 21,9

0,355 – 0,710 mm 24,9 28,4 28,2 29,8 29,9 32,1

0,710 – 1,000 mm 6,8 10,4 5,8 8,3 9,2 11,6

1,000 mm 1,5 2,4 0,8 1,1 1,5 0

26

With respect to particle size distribution, for all the reproduction (R) batches, the fraction of

particles >1000 m was <3%, but for D/2 and D/3 (14.8% and 10.8%) it exceeded this value. In all

cases, the fraction of the particles <90 m in the R batches were <20%, but for D/4 and D/5 it was

20.8% and 25.5%. For the R batches, the majority of the granules (>50%) fell into the fraction 180-

710 m, in contrast to those produced in the Diosna P400, none of which were in this range.

The Carr compressibility index demonstrated that the granules prepared in both sets of

equipment had good flow properties, but relative standard deviation (RSD) is higher (10.80%) for

the D batches than the R batches (4.74%). The bulk and tapped densities varied within a wider

range and higher relative standard deviation for the D batches (bulk density: 68.49–75.76 g/100 ml,

RSD: 4.34%; tapped density: 79.86-87.88 g/100 ml, RSD: 5,21%) than for the R batches (bulk

density: 68.49–71.43 g/100 ml, RSD: 1.39%; tapped density: 80.55-84.29 g/100 ml, RSD: 1.65%).

Figure 10. shows SEM pictures of granules prepared in the Diosna P400. The particles making

up the granules prepared in the Diosna P400 were dense and relatively large. The spherical structure

was retained during the fluid-bed drying process that followed the granulation. The surfaces of the

granules displayed little wear.

Figure 11. presents SEM pictures of granules prepared in the Collette Ultima 600. These

granules had a looser structure, were less spherical and smaller, and cracked during drying. Any

irregular protrusions of the particles broke away from the granules.

27

Fig. 10. SEM pictures of granules prepared in the Diosna P400

Fig. 11. SEM pictures of granules prepared in the Collette Ultima Pro 600

28

The differences in the texture of the granules could be caused by the differing geometries of

the two machines as concerns the shape and positions of the impeller and chopper blades. Because

of these factors, the materials exhibit completely different types of motion during granulation, with

the Collette Ultima 600 producing an ―undulating‖ effect, and the Diosna P400 granulator

employing a ―folding‖ action. This difference can be made even more distinct by varying the

parameter settings.

The result of the homogeneity study (Blend Uniformity Analysis) demonstrated a mean value

of 99.1+0.4% for the active ingredient content with the relative standard deviation less than 1.5 %

in both cases.

6.1.5. Tablet evaluation

The tablet parameters were satisfactory in both cases. The friability was <0.33%, the thickness

was between 4.00 and 4.19 mm, the disintegration time was <2 min for all batches, and the average

hardness was between 51 and 66 N. The weight variation of the tablets was 0.60-1.00% for the

experimental batches, 1.01-1.12% for the D batches, and 0.57-0.7% for the R batches.

Assay, impurity and dissolution test of the tablets were also determined. For all batches the

metronidazole content varied between 98.8 – 101.0 %, the impurity spot was not visible and

dissolution values varied between 99.1 – 100. 4 %.

6.1.6. Conclusion

For two high-shear granulators with different constructions, I established the ranges of

parameter settings which ensure the safe transference of the technologies for a preparation with a

high content of active ingredient.

I determined the optimal setting ranges for mass production (impeller speed: 80-135 rpm;

ideal torque associated with the impeller speed: 560-800 Nm; chopper speed: 600-2,700 rpm; ideal

water content of the binder solution: 26 kg; liquid binder flow rate: 3-12 kg/min; massing time: 2-6

min), within which parameter ranges a satisfactory product could be manufactured in a manner such

that the drying stage took place within the same fluid-bed drying equipment. The experiment

demonstrated that, although the two technological devices perform granulation according to similar

principles of operation, their different geometric properties require different technical settings in

order for the end-products to have the same physical characteristics.

The textures of the granules prepared in the two types of machine differed considerably, but

the differences between the measured physical parameters were not as great. The granulation

process was highly controllable, the product was suitably robust and the results were easy to

29

reproduce in the Collette Ultima 600 granulator, which allowed elimination of the inconsistencies

resulting from the use of the Diosna P 400.

6.2. The various aspects of granulation and drying

Although the geometrical differences between high shear granulators and the divergences in

their technical specifications and construction did result in a physical discrepancy between the

textures of the granules, this structural difference had no significant effect on their tablet-formation

properties. This phenomenon is clearly attributable to the fact that these macroscopic changes are

offset by the robustness of the technology, and therefore their impact on the further processability of

the material system is not substantial. The granules have good pressability characteristics, and

therefore the structural changes did not affect pressability. In the case of both samples the physical

attributes of the pressed products were well within the required specifications. I have reached this

conclusion the structural characteristics used to analyse the agglomeration process.

This justifiably gives rise to the question of what effect the following drying stage will have

on the structure of the granules. Do the structural properties change depending on the drying

method, or, once the bonds have been formed between the particles, does the moisture leave the

system without having any impact on the final structure? To determine this, I studied granules that

had been granulated using different methods but dried in the same way, as well as granules that

were granulated and dried by different means. I examined the effect of microwave vacuum drying

and fluid-bed drying on the binding forces created during the granulation stage. The primary

objective of this series of experiments was to analyse the impact of the forces and energies released

during the drying process on the systems granulated in the previous phases.

Approaching the drying process from the standpoint of energy transfer theory, one would

expect to find considerable discrepancies, since in the case of fluid-bed drying the flow of warm air

extracts the moisture from the solid particles at atmospheric pressure, while the particles collide,

and can thus be damaged by these extremely high-velocity collisions. In the case of microwave

vacuum drying the layer of granules being dried is more or less static, and the moisture exits the

granules in a sub-atmospheric chamber, as the result of a change in the partial solvent pressure

above the material. The bowl temperature stated for the product only changes the speed of drying,

but does not influence the change of the structure of the granules. However, the microwave-

generated energy could have an effect on the granules’ physical structure. The heat generated in this

way is instrumental in extracting the moisture, since the solvents only affect the molecules

indirectly. The energy applied in this way plays a very important role, since in the case of uneven

heat dissipation, localised hotspots can arise in the material, which could in turn lead to burning

30

points. This phenomenon can be eliminated through the regulation of the microwave energy, and by

the slow, intermittent – or where necessary continuous – agitation of the system.

Examining the systems from the standpoint of mechanical actuation, structural discrepancies

can also be expected in the course of drying, if the velocity and temperature of the fluidisation

medium, as well as the volume of the batch to be dried, are not selected optimally. In the case of

microwave drying, the optimal thickness of the layer to be dried primarily depends on the duration

of the process; however, a low-intensity yet efficient agitation will result in a consistent change in

the moisture content of the system.

Thus it can be established that, besides optimisation of the manufacturing technology, the

ideal technological parameters are also affected by the robustness of the given operation, and,

owing to the fact that the drying stage involves a series of complex physical processes, these

parameters have a profound influence on the quality of the end product.

It was therefore logical to conclude that, in the course of the technology transfer, I should

study the steps in the drying process in as much detail as the steps in the granulation process. I

compared the physical characteristics of the dry particles from both the granulometric standpoint

and in terms of their tablet-formation properties. My conclusions clearly corroborate the results of

the impurity profile and physical-morphological tests of the finished tablets. Consequently, in my

second experimental project I studied the questions related to drying.

6.3. Investigation of fluid-bed and vacuum microwave drying

6.3.1. Manufacturing process

I performed the granulation in the Collette Ultima Pro 600 processing equipment. The active

ingredient, the corn starch and the colloidal anhydrous silica were homogenized (impeller speed: 65

rpm, process time: 6 min). The liquid binder was added to the powder mixture (impeller speed: 95

rpm, chopper speed: 600 rpm, liquid binder flow rate: 7 kg/min, process time: approx. 4 minutes).

After addition of the liquid binder, mixing was continued to the torque value (Wet massing:

impeller speed: 95 rpm, chopper speed: 2,700 rpm, torque value: 6.5 kW).

The granules were dried to the value of the loss on drying by using two different methods:

In one case, I discharged the wet granules from the Collette Ultima Pro 600 equipment and loaded

them into the Glatt WSG 200 fluid-bed dryer and performed the drying at 60 °C (process time: 35

minutes, maximum product temperature 35.5 °C).

In the other case, drying was carried out in Collette Ultima Pro 600 equipment by microwave-

vacuum drying. The aim was to achieve the shortest possible drying time, and I therefore used the

31

maximum forward energy (vacuum: 50 mbar, microwave forward energy: 22 kW, continuous

mixing: 20 rpm, process time: 58 minutes, maximum product temperature: 43 °C).

After drying, the granules were sized in a 1.5 mm sieve (rotation speed: 500 rpm), and then

homogenized for 2 and 5 minutes with the tabletting excipients (microcrystalline cellulose, talc and

magnesium stearate) in a container blender.

Figure 12. shows the flowcharts of the detailed manufacturing processes.

I determined the size distribution of the granules, their tapped and bulk densities, porosity and loss

on drying, and took SEM photographs. The individual and average masses, thicknesses, hardnesses

and disintegration time of the tablets were examined. Analytical investigations (BUA, assay, purity,

dissolution) were also carried out.

32

Dry mixing

Collette Ultima Pro 600 single-pot processor

Impeller speed: 65 rpm, Process time: 6 min

Corn starch

Colloidal anhydrous silica

Metronidazole

Liquid binder addition

Collette Ultima Pro 600 single-pot processor

Impeller speed: 95 rpm, Chopper speed: 600 rpm

Liquid binder flow rate: 7 kg/min

Process time: approx. 4 min

Povidone K-30

Glycerin

Purified water

Wet massing

Collette Ultima Pro 600 single-pot processor

Impeller speed: 95 rpm, Chopper speed: 2,700 rpm

Torque value: 6.5 kW

Drying

Glatt WSG 200 fluid-bed granulator and dryer

Inlet air temperature: 60°C

Dry to approx. 3.0 – 4.0 %

Drying

Collette Ultima Pro 600 single-pot processor

Continous mixing: 20 rpm, Vacuum: 50 mbar,

Microwave forward energy: 22 kW,

Dry to approx. 3.0 – 4.0 %

Sieving

Quadro comil U 20

Rotation speed: 500 rpm, Sieve size: 1.5 mm

Blending I.

Zanchetta Canguro container blender

Process time: 2 min

Talc,

Microcrystalline cellulose

Blending II.

Zanchetta Canguro container blender

Process time: 5 min

Magnesium-stearate

Tabletting

Courtoy R 190 Ft tablet press

Rotation speed: 65 rpm

Fig. 12. Flow sheet of the granulation in two types of dryers and tabletting.

33

6.3.2. Effects of different drying techniques on properties of granules and tablets made on a

production scale

Depending on the composition of the material system and the solvents used (organic or water)

and their quantities, preference is given to different drying techniques (e.g. fluid-bed or vacuum) in

the pharmaceutical industry. However, for certain material systems, the differences between the

drying technologies are not marked enough to make one or the other unambiguously preferable. The

products under study do not contain organic solvents or materials that are sensitive to heat or

oxygen, or which contain toxic or potent compounds, in which cases the single-pot technology

would be clearly preferable. 42, 52 On the other hand, I am not using a liquid binder with a high

water content, and it is therefore not necessary to ensure a low loss on drying when the granules are

dried (Table 7.), in which case fluid-bed drying would be preferable. For this reason, with these

products I had the opportunity to perform a comparative granulometric analysis of different

techniques used for drying wet granules prepared by using the same method. The advantage of

fluid-bed drying is the short drying time, in contrast with pure vacuum drying, which entails a long

processing time. Accordingly, I combined vacuum drying with microwave drying, since the

duration of processing is an important consideration in the pharmaceutical industry. In the drying

process, the primary goal was to shorten the processing time. In the experiments, the difference

between the maximum product temperatures attained with the two drying techniques (35.5 °C and

43 °C) had no impact on the product quality.

In the case of high-shear granulation granule size increase is influenced by the impeller speed, the

wet massing time and the amount of liquid. In this study, the granules were granulated by means of

the same technology, but dried with different methods. The powder fraction was relatively high for

both vacuum and fluid-bed drying, at 21% and 23%, respectively, as shown in Table 7., but a

significant difference in the powder fraction was not detected (which is consistent with earlier

reported results. 109 ). In respect to the composition under study, the mean particle size ((D50) was

larger for the granules dried by using microwaves than for the fluid-bed dried sample. This is

because the granules collide with each other and the wall of the equipment during the fluid-bed

drying process. Particles therefore constantly break off and are eroded.

Besides D50, my findings were also corroborated by the SEM images shown in Figures 11. and 13.

The granules dried in the vacuum chamber were more geometrically regular and spherical, and thus

had a different external physical structure from that of the granules dried with the fluid-bed

technology.

34

Table 7. Granule properties of batches dried

in the Collette Ultima Pro 600 and in the Glatt WSG 200

Glatt WSG 200 Collette Ultima Pro 600

Bulk density (g/100 ml) 68.49 – 71.43 79.37 – 83.30

Tapped density (g/100 ml) 80.55 – 84.29 94.53 – 104.17

Loss on drying (%) 2.70 – 3.45 3.07 – 4.08

Fine particles (%) 21 23

D50 (μm) 310 – 370 360 – 420

Porosity (%) 73.9 63.1

Fig. 13. SEM photograph of granules dried in the Collette Ultima Pro 600

The physical differences between the granules could result partly from the drying time, and partly

from the nature of the drying curves (Fig. 14). In order for a material system with the same moisture

content to be achieved by the end of the drying process, approximately 1.5 times the drying time is

necessary in the case of vacuum drying than in the case of fluid-bed drying. In other words the

expulsion of moisture is slower, gentler and more even, with the result that the primary physical

structure of the granules remains more intact. In the case of fluid-bed drying, the raggedness and

erosion of the granules arises not only from the impact, but also as a result of the sudden

temperature change, owing to the rapid expulsion of moisture. This rapid evaporation inflicts more

intensive damage on the granules.

35

10

20

30

40

50

0 10 20 30 40 50

Time (min)

Pro

du

ct

tem

pera

ture (

°C)

Fig. 14. Drying curves of the vacuum-microwave ( - - - - ) and fluid-bed ( )drying technology

It is known from the literature that the porosity of granules is affected considerably by the impeller

speed and the wet massing time. 107 However, less research has been conducted into the extent to

which the porosity of granules prepared by using the same granulation technology is influenced by

the subsequent use of different drying methods.

The data in Table 7. shows that the granules dried in the vacuum chamber had a lower level of

porosity than those dried by using the fluid-bed process, although the drying process was slower.

This is due to the mechanism by which the moisture is forced out of the capillaries in the granules

under sub-atmospheric pressure, which results in the formation of ―channels‖ in the interior of the

granules as the moisture leaves the granules. In the course of fluid-bed drying, which takes place at

atmospheric pressure, the granules dry from their surface inwards, which results in a higher level of

porosity.

The lower porosity values entail higher bulk and tapped density values, as shown in Table 7.

A reduction in porosity generally leads to a deterioration in compressibility. In the systems I

examined, this took the form of a shift in the range of the compressing force required to produce a

tablet of the same hardness. 110-114

The correlation between compressing force and hardness is shown in Fig. 15. The granules prepared

by using microwave-vacuum drying are denser, with the result that the tablets are thinner and easily

compressible, but a higher pressure force must be applied than in the case of the granules dried with

the fluid-bed technology. The correlation between hardness and thickness is shown in Fig. 16. The

thickness of the compressed tablets from the granules dried with the microwave-vacuum technology

was lower than that of the tablets compressed from granules of the same hardness, dried with the

36

fluid-bed technology. The differences in pressability can be attributed to the differences in the

granules’ structure, which are caused by the differing drying technologies. The use of the different

drying techniques had no effect on the individual mass distribution or disintegration time of the

tablets; they had a relatively low mass distribution and short disintegration time (< 1 minute) in

both cases.

For all batches the metronidazole content varied between 99.0 – 101.3 %, the impurity spot

was not visible and dissolution values varied between 97.6 – 103.0 %.

The result of the homogeneity study (Blend Uniformity Analysis) demonstrated a mean value

of 99.6 + 0.3% for the active ingredient content with a relative standard deviation (RSD) of less

than 1.3 % in all batches.

37

Higher pressing power had to be applied when pressing granules dried using vacuum-microwave

technology (+) than in the case of fluid-bed dried granules ( ).

0

10

20

30

40

50

60

70

80

10 12 14 16 18 20 22 24 26

pressing power (kN)

hard

ness

(N

)

Fig. 15. The correlation between hardness and pressing power

Tablets pressed from granules dried in the Collette Ultima Pro 600

Tablets pressed from granules dried in the Glatt WSG 200

Fig. 16. The correlation between hardness and thickness

0

10

20

30

40

50

60

70

3,85 3,9 3,95 4 4,05 4,1 4,15 4,2

thickness (mm)

hard

ness

(N

)

38

6.3.3. Conclusions

Following the wet massing process, the drying technologies applied in the pharmaceutical

industry were selected on the basis of a number of criteria, such as the properties of the active

ingredient, the type of solvent, the processing time, etc. The choice of the most suitable technology

for the given purpose requires careful consideration and testing. Two drying techniques, based on

differing principles (fluid-bed and microwave-vacuum) were selected for the purposes of the

present research, and the properties of the granules produced by using these methods were

compared.

The granules produced in the traditional high-shear granulator and dried in a vacuum chamber had a

lower level of porosity, and higher bulk and tapped densities, owing to the special characteristics of

the drying process. They retained their spherical form, in contrast to the granules dried by using the

fluid-bed technology. These characteristics of the granules also determined the properties of the

tablets pressed from them, and made it necessary to apply a greater compressing force in the case of

the granules prepared by using the microwave-vacuum drying process. At the same time, the mass

distribution and disintegration time were not affected.

Despite the measurable physical differences arising from the differing principles of the two drying

methods, both drying technologies proved highly suitable for production-scale manufacturing of the

compositions under study.

39

7. Summary

The objective of my research was to facilitate the technology transfer of a product prepared in

a production-scale traditional high-shear granulator and dried using fluid-bed technology, into a

―single-pot‖ type machine, without any change to its composition. In ―single-pot‖ or ―one-pot‖

equipment, the entire manufacturing process, the granulation process, as well as the drying stage

that follows, takes place in the one device. With respect to the granulation process, I performed my

experiments in traditional high-shear and single-pot machines with differing geometrical attributes.

The drying process did not follow the fluid-bed principle, but was carried out using vacuum and

microwave technology. This study aims to shed light on the similarities and differences between the

two procedures, through a study of the – primarily physical – attributes of the granules and tablets

produced.

I have reached the following conclusions:

The aggregation process of the granules is influenced considerably by the geometrical

properties of the equipment, as well as the location and shape of the impeller and chopper.

The quality characteristics of the particles can only be objectively analysed by performing

several physical examinations.

In the case of granules containing metronidazole, the fluid-bed (moving layer) drying method

results in a smaller average granule size (D50) than the vacuum-microwave drying method.

However, in the case of aqueous systems, the process time for vacuum-microwave drying is

considerably longer than that of fluid-bed drying. Vacuum-microwave technology is unable to

match the speed of fluid-bed drying.

In the case of vacuum-microwave drying the dry granules are more spherical and

geometrically regular, as a result of abrasion between the granules.

In the course of fluid-bed drying, the erosion caused by the particles colliding is what causes

greater reduction in granule size. In the case of the vacuum-microwave drying method, less erosion

results from a combination of the longer process time and the constant agitation necessary to ensure

even heat dissipation.

The texture of the aggregates is mainly influenced by the drying method, rather than by the

granulation process.

The porosity value of granules prepared using vacuum-microwave drying is lower that those

prepared using fluid-bed drying, owing to the means by which the moisture exits the system.

In the case of lower-porosity products, a higher pressure must be applied when pressing

tablets.

40

In the course of pressing, the differences in porosity result in changes to the hardness and

thickness of the tablets.

In the case of robust technologies, when selecting the drying method, its effect on the quality

and stability of the material systems must also be taken into consideration.

Practical usefulness:

Pharmaceutical companies often find that they have to present scientific evidence to justify the

replacement or modernization of equipment that could be up to one or two decades old.

Pharmaceutical companies endeavour to achieve more cost-effective and better-regulated processes,

in line with the standards of Good Manufacturing Practice (GMP). The advantages of single-pot

high-shear granulators include the facts that the entire process takes place in one set of equipment,

GMP requirements are met, the processing time is reduced, cleaning is simplified through the use of

integrated, programmed cleaning systems, and the granulators are equipped with the appropriate

safety systems. 39-40

The purpose of my experiments was to search for the correlations that can be identified in the

course of technology transfers, and which influence the quality of the products. In an industrial

environment technology transfers are very common. Pharmaceutical technology should not only

focus on the behaviour of material systems, but should also take into consideration the attributes of

the equipment used, and the affect they have on the end product.

In the course of the technology transfer I compared the processes that take place during granulation

and drying in the case of industrial equipment operating under identical and differing principles,

while optimising the technological parameters of a product that is currently in use. At the same

time, an opportunity was created to learn about the operating mechanisms of a relatively new

technological culture. The experience thus gained could also be of practical use in the course of

future technology transfers.

41

8. References

1 Parikh, D. Handbook of Pharmaceutical Granulation Technology; Drugs and Pharmaceutical

Sciences, 1997, Marcel Dekker, Inc.: New York

2 Lachman, L.; Libermann, H.A.; Kanig, J.L. The theory and practice of industrial pharmacy,

1986, Lea & Febiger, Philadelphia

3 Rácz, I.; Selmeczi, B. Gyógyszertechnológia, 3 kötet 2001, Medicina, Budapest

4 Kristensen, G.H.; Schaefer, T., Granulations. In: Swarbrick, J., otherwise difficult to

understand. Boylan, J.C. (Eds.). Encyclopedia of Pharmaceutical Technology, Marcel

Dekker, New York, 1993, 7. 121–160.

5 Passerini, N.; Albertini, B.; Gonzales-Rodriguez, M.L.; Cavallari, C.; Rodriguez. L.

Preparation and characterisation of ibuprofen-poloxamer 188 granules obtained by melt

granulation, Int. J. Pharm. 2002, 15, 71-78

6 Passerini, N.; Albertini, B.;. Perissutti, B.; Rodriguez. L. Evaluation of melt granulation and

ultrasonic spray congealing as techniques to enhance the dissolution of praziquantel Int. J.

Pharm. 2006, 318, 92-102

7 Walker, G.M.; Holland, C.R.; Ahmad, M.M.N.; Craig, D.Q.M. Influence of process

parameters on fluidised hot-melt granulation and tablet pressing of pharmaceutical powders,

Chem. Eng. Science 2005, 60, 3867–3877

8 Walker, G.M.; Andrews G.; Jones D. Effect of process parameters on the melt granulation

of pharmaceutical powders, Powder Technol. 2006, 165, 161-166

9 Walker, G.M.; Bell, S.E.J.; Andrews G.; Jones D. Co-melt fluidised bed granulation of

pharmaceutical powders: Improvements in drug bioavailability, Chem. Eng. Science 2007,

62, 451–462

10 Van Melkebeke, B.; Vermeulen, B.; Vervaet, C.; Remon J.P. Melt granulation using a

twin-screw extruder: A case study, Int. J. Pharm. 2006, 326, 89-93

11 Yang, D.; Kulkarni, R.; Behme, R.J.; Kotiyan, P.N. Effect of the melt granulation

technique on the dissolution characteristics of griseofulvin, Int. J. Pharm. 2007, 329, 72-80

12 Riepma, K.A.; Vromans, H.; Zuurman K.; Lerk C.F. The effect of dry granulation on the

consolidation and compaction of crystalline lactose, Int. J. Pharm. 1993, 97, 29-38

42

13 Kleinebudde, P. Roll compaction/dry granulation: pharmaceutical applications, Eur. J.

Pharm. and Biopharm. 2004, 58, 317–326

14 Freitag, F.; Kleinebudde, P. How do roll compaction/dry granulation affect the tableting

behaviour of inorganic materials? Comparison of four magnesium carbonates, Eur. J.

Pharm. Sci. 2003, 19, 281–289

15 Freitag, F.; Reincke, K.; Runge, J.; Grellmann, W.; Kleinebudde, P. How do roll

compaction/dry granulation affect the tableting behaviour of inorganic materials?:

Microhardness of ribbons and mercury porosimetry measurements of tablets, Eur. J. Pharm.

Sci. 2004, 22, 325–333

16 Rambali, B.; Baert, L.; Jans, E.; Massart, D.L. Influence of the roll compactor parameter

settings and the compression pressure on the buccal bio-adhesive tablet properties, Int. J.

Pharm. 2001, 220, 129-140

17 Inghelbrecht, S.; Remon, J.; Fernandes de Aguiar, P.; Walczak, B.; Massart, D.; Van De

Velde, F.; De Baets, P.; Vermeersch H.; De Backer P. Instrumentation of a roll compactor

and the evaluation of the parameter settings by neural networks, Int. J. Pharm. 1997, 148,

103-115

18 Weyenberg, W.; Vermeire, A.; Vandervoort, J.; Remon , J.P.; Ludwig, A. Effects of roller

compaction settings on the preparation of bioadhesive granules and ocular minitablets, Eur.

J. Pharm. and Biopharm. 2005, 59, 527–536

19 Bindhumadhavan, G.; Seville, J.P.K.; Adams, M.J.; Greenwood, R.W.; Fitzpatrick, S. Roll

compaction of a pharmaceutical excipient: Experimental validation of rolling theory for

granular solids, Chem. Eng. Science 2005, 60, 3891-3897

20 Yusof, Y.A.;. Smith, A.C.; Briscoe, B.J. Roll compaction of maize powder, Chem. Eng.

Science 2005, 60, 3919 - 3931

21 Lecompte, T.; Doremus, P.; Thomas, G.; Perier-Camby, L.; Le Thiesse, J.; Masteau J.;

Debove L. Dry granulation of organic powders—dependence of pressure 2D-distribution on

different process parameters, Chem. Eng. Science 2005, 60, 3933–3940

22 Hausman, D.S. Comparison of low shear, high shear, and fluid bed granulation during low

dose tablet process development, Drug Dev. Ind. Pharm. 2004, 30, 259-266

43

23 Gao, J.Z.H.; Jain, A.; Motheram, R.; Gray, D.B.; Hussain, M.A. Fluid bed granulation of a

poorly water soluble, low density, micronized drug: comparison with high shear granulation,

Int. J. Pharm. 2002, 237, 1–14

24 Kristensen, G.H.; Holm, P.; Schaefer, T. Mechanical properties of moist agglomerates in

relation to granulation mechanisms, Part II. Effect of particle size distribution, Powder

Technol. 1985, 44, 239-247

25 Kristensen, G.H.; Holm, P.; Schaefer, T. Granulation in high-speed mixers, Part VI.

Effects of process conditions and power consumption and granule growth, Powder Technol.

1985, 43, 225-233

26 Iveson, S.M.; Lister, J.D. Growth regime map for liquid-bound granules, AIChE Journal

1998, 44, 1510-1518

27 Iveson, S.M.; Wauters, P.A.L.; Forrest, S.; Lister, J.D.; Meesters, G.M.H.; Scarlett, .

Growth regime map for liquid-bound granules: further development and experimental

validation, Powder Technol. 2001, 117, 83-97

28 Knight, P.C. Structuring agglomerated products for improved perfomance, Powder

Technol. 2001, 119, 14-25

29 Fox, B. True Grit: Granulation & Drying of Delicate Products, Chem. Eng. 2005, 115, 35-

38

30 Cryer, S.A.; Scherer P.N. Observations and process parameter sensitiveties in fluid-bed

granulation, AIChE Journal 2003, 49, 2802-2809

31 Menon, A., Dhodi.;N.; Mandella, W.; Chakrabarti, S. Identifying fluid-bed parmeters

affecting product variability, Int. J. Pharm. 1996, 140, 207–218

32 Pont, V.; Saleh, K.; Steinmetz, D.; Hémati, M. Influnce of the physicochemical properties

on the growth of solid particles by granulation in fluidized bed, Powder Technol. 2001, 120,

97-104

33 Zank, J.; Kind, M.; Schlünder, E.-U. Partical growth and droplet deposition in fluidised

bed granulation, Powder Technol. 2001, 120, 76-81

34 Heinrich, S., Mörl, L. Fluidized bed spray granulation-A new model for the desccription

of particle wetting and of temperature and concentration distribution, Chem. Eng. and

Processing, 1999, 38, 635-663

44

35 Landin, M.; York, P.; Cliff, M.J.; Rowe, R. C. Scaleup of a pharmaceutical granulation in

planetary mixers, Pharm. Dev. Technol. 1999, 4, 145–150

36 Sirois, P.J.; Craig, G.D. Scaleup of a high-shear granulation process using a normalized

impeller work parameter, Pharm. Dev. Technol. 2000, 5, 365–374

37 Vermeire, A.; Keleb, E.I.; Kiekens, F.; Van Driessche, Hoste, S.; Remon, J.P.; Vervaet, C.

Tablets prepared by single-step granulation/tabletting:Interparticulate binding mechanism

and stability, Pharm. Dev. Technol. 2005, 10, 397–403

38 Keleb, E.I.; Vermeier, A.; Vervaet, C.; Remon, J.P. Single-step granulation/tabletting of

different grades of lactose: a comparison with high shear granulation and compression, Eur.

J. Pharm. and Biopharm. 2004, 58, 77–82

39 Péré, C.; Rodier, E. Microwave vacuum drying of porous media: experimental study and

qualitative considerations of internal transfers, Chem. Eng. and Processing, 2002, 41, 427-

436

40 Stahl, H. Single-pot system for drying pharmaceutical granules, Pharm. Technol. Eur.,

2000, 12, 192-201

41 Stahl, H. Comparing different granulation techniques, Pharm. Technol. Eur., 2004, 16, 23-

33

42 Fox, B.; Bohle, L.B. Vacuum and Microwave Dry Granulated Products, Chem. Eng.,

2001, 108, 135-141

43 McMinn, W.A.M.; McLoughlin, C.M.; Magee, T. R. A. Microwave-convective drying

characteristics of pharmaceutical powders, Powder Technol. 2005, 153, 23-33

44 Chen, G.; Wang W.; Mujumdar, A.S. Theoretical study of microwave heating patterns on

batch fluidized bed drying of porous material, Chem. Eng. Science 2001, 56, 6823–6835

45 Getoff, N. Generation of O2 by microwave discharge and some characteristic reactions: A

short review, Radiat Phy. Chem. 1995, 45, 609-614

46 Berlan, J. Microwave in Chemistry: Another way of heating reaction mixtures, Radiat

Phy. Chem. 1995, 45, 581-589

47 Kaatze, U. Fundamental of microwaves, Radiat Phy. Chem. 1995, 45, 539-548

48 Kaatze, U. Microwave dielectric properties of liquids, Radiat Phy. Chem. 1995, 45, 549-

566

45

49 McLoughlin, C.M.; McMinn, W.A.M.; Magee, T. R. A. Physical dielectric properties of

pharmaceutical powders, Powder Technol. 2003, 134, 40-51

50 Kelen, Á.; Hegedűs, Á.; Nagy, T.; Máthé, Z.; Hódi, K. A mikrohullám alkalmazásának

előnye hőérzékeny agglomerátumok esetén, Acta Pharmaceutica Hungarica, 2003, 72, 65–

70

51 Kelen, Á.; Ress, S.; Nagy, T.; Hegedűs, Á.; Bódis, A.; Erős, I.; Hódi, K. Mikrohullámú

vákuumszárítás során kialakuló hőeloszlás követésének lehetősége, Acta Pharmaceutica

Hungarica, 2005, 75, 17–22

52 Kelen, Á.; Pallai-Varsányi, E.; Dávid, Á.; Hegedűs, Á.; Pintye-Hódi, K. Select the most

suitable diluent to formulate a ―heat sensitive‖ active in case of microwave vacuum drying,

Eur. J. Pharm. Sci. 2005, 25, 77–79

53 Hegedűs, Á.; Kelen, Á.; Pintye-Hódi, K. The effect of different drying techniques on the

porosity parameters of granules at production scale, Eur. J. Pharm. Sci. 2005, 25/S1, 114–

115

54 Dávid, Á.; Benkóczy, Z.; Ács, Z.; Greskovits, D.; Dávid, Á.Z. The theoretical basis for

scaling-up of the method of microwave granulation, Drug Dev. Ind Pharm. 2000, 26, 943-

951

55 Duschler, G.; Carius, W.; Bauer, K.H. Single-step granulation method with microwaves:

preliminary studies and pilot scale results, Drug Dev. Ind Pharm. 1995, 21, 1599-1610

56 Mandal, T.K. Evaluation of microwave drying for pharmaceutical granulations, Drug Dev.

Ind Pharm. 1995, 21, 1683-1688

57 Sweetman, S.C. Martindale The Extra Pharmacopoeia 34. Edition, 2005, Pharmaceutical

Press, London

58 Vizi, E.Sz. Humán farmakológia, 1997, Medicina, Budapest

59 Fürst, Zs. Gyógyszertan, 1998, Medicina, Budapest

60 Knoll, J. Gyógyszertan, 1993, Medicina, Budapest

61 Szász, Gy., Takács, M., Végh, A.; Gyógyszerészi kémia, 1990, Medicina, Budapest

62 Odeku, O.A.; Patani, B.O. Evaluation of Dika Nut Mucilage (Irvingia gabonensis) as

binding agent in metronidazole tablet formulations, Pharm. Dev. Technol. 2005, 10, 439–

446

46

63 European Pharmacocopoeia, 5th

Edition, 2005, Council of Europe, Strasbourg

64 Weiner, M.; Bernstein, I.L. Adverse reactions to drug formulation agents: A handbook of

excipiens, 1989, Marcel Dekker, Inc.: New York

65 Wade, A., Weller, P.J. Handbook of Parmaceutical Excipients, 1994, American

Pharmaceutical Association, Washington

66 Fiedler, H.P. Encyclopedia of excipiens for pharmaceuticals, cosmetics and related areas,

Volume I-II. 2002, Editio Cantor Verlag, Anlendorf

67 Bührer, V. KollidonR

Polyvinylpirolidon for pharmaceutical industry, 1999, BASF

Aktiengesellschaft Feinchemie: Ludwigshafen

68 Van Vaerenbergh, G. The influence of a swinging bowl on granulate properties, Pharm.

Technol. Eur. 2001 march, 36-43

69 Faure, A.; Grimsey, I.M.; Rowe, R.C.; York, P.; Cliff, M.J. Applicabbility of a scale-up

methodology for wet granulation processes in Collette Gral high shear mixer-granulators,

Eur. J. Pharm. Sci. 1999, 8, 85–93

70 Laitinen, N.; Antikainen, O.; Yliruusi, J. Does a powder surface contain all necessary

information for particle size distribution analysis? Eur. J. Pharm. Sci. 2002, 17, 217-227

71 Obae, K.; Iijima, H.; Imada, K. Morphological effect of microcrystalline cellulose

particles on tablet tensile strength Int. J. Pharm. 1999, 182, 155-164

72 Murakami, H.; Yoneyama, T.; Nakajima, K.; Kobayashi, M. Correlation between loose

density and compactibility of granules prepared by various granulation methods, Int. J.

Pharm. 2001, 216, 159–164

73 Schüssele, A.; Bauer-Brandl, A. Note on themeasurement of flowability according to the

European Pharmacopoeia, Int. J. Pharm. 2003, 257, 301-304

74 Carstensen, J. T. Pharmaceutics of solids and solid dosage forms, 1977, J. Wiley & Sons

Co., Canada

75 Carstensen, J. T. Solid pharmaceutics mechanical properties and rate phenomena, 1980,

Academic Press Inc.: London

76 Carstensen, J. T. Pharmaceutical principles of solid dosage forms, 1993, J. Technomic

Publishing Company Inc.: Pennsylvania

47

77 Podczeck, F.; Newton, J.M. Powder filling into hard gelatine capsules on a tamp filling

machines, Drug Dev. Int. J. Pharm. 1999, 185, 237–254

78 Szabó-Révész, P.; Göcző, H.; Pintye-Hódi, K.; Kása, P.; Erős, I.; Hasznos-Nezdei, M.;

Farkas, B. Development of spherical crystal agglomerates of an aspartic acid salt for dirrect

tablet making, Powder Technol. 2001, 114, 118-124

79 Carr, R.L. Evaluating flow properties of solids. Chem. Eng. 1965, 72, 163–168.

80 Shah, S.N.; Killeen, M. Use of Carr index for determination of machinability of phase I

formulation, Pharm. Eng. 2004, march/april, 74–82

81 Podczeck, F.; Newton, J.M. Powder and capsule filling properties of lubricated granulated

cellulose powder, Drug Dev. Ind. Pharm. 2000, 50, 373–377

82 Kumar, V.; Reus-Medina, M.L.; Yang, D. Preparation Characterization, and tabletting

properties of a new cellulose-based pharmaceutical aid, Int. J. Pharm., 2002, 235, 129-140

83 Gluba, T. The energy input of bed processing during drum granulation, Chem. Eng. and

Processing, 2005, 44, 237-243

84 Chantraine, F.; Viana, M.; Cazalbou, S.; Brielles, N.; Mondain-Monval, O.; Pouget, C.;

Branlard, P.; Rubinstenn, G.; Chulia D. From compressibility to structural investigation of

sodium dodecyl sulphate - Part 2: A singular behavior under pressure, Powder Technol.

2007, 177, 41-50

85 Viana, M.; Jouannin, P.; Pontier, C.; Chulia, D. About pycnometric density measurements,

Talanta, 2002, 57, 583-593

86 Bedow, J.K. Partical characterisation in technology, Volume I. Application and

Microanalysis, 1984, CRC Press, Boca Raton, Florida

87 Bedow, J.K. Partical characterisation in technology, Volume II. Morphological Analysis,

1984, CRC Press, Boca Raton, Florida

88 Kotoe, M.O.; Fuji, M.; Takei, T.; Chikazawa, M. Effect of geometric structure and surface

wetability of glidant on tablet hardness, Int. J. Pharm. 2003, 262, 75–82

89 Davies, P.N.; Worthington, H.E.C.; Podczeck, F.; Newton, J.M. The determination of the

mechanical strength of tablets of different shapes, Eur. J. Pharm. Sci. 2007, 67, 268-276

90 Rubinstein, M.H. Pharmaceutical technology, Volume I. Tabletting technology, 1987, J.

Wiley & Sons Co., England

48

91 Knight, P.C.; Johansen, A.; Kristensen, H.G.; Schaefer, T.; Seville, J.P.K. An

investigation of the effect on the agglomeration of the changing the speed of a mechanical

mixer, Powder. Technol. 2000, 110, 204-209

92 Bouwman, A.M.; Henstra, M.J.; Westerman, D.; Chung, J.T.; Zhang, Z.; Ingram, A.;

Seville, J.PK.; Frijlink, H.W. The effect of amount of binder liquid on mechanisms and

structure of microcrystalline cellulose granules prepared by high shear granulation, Int. J.

Pharm. 2005, 290, 129–136

93 Saleh, K.; Vialatte, L.; Guigon, P. Wet granulation in a batch high shear mixer, Chem.

Eng. Science 2005, 60, 3763-3775

94 Dévay, A.; Mayer, K.; Pál Sz.; Antal, I. Investigation on drug dissolution and particle

characteristics of pellets related to manufacturing process variables of high-shear

granulation, J. Biochem. Biophys. Methods 2006, 69, 197-205

95 Ameye, D.; Keleb, E.; Vervaet, C.; Remon, J.P,; Adams, E.; Massart. D.L. Scaling-up of a

lactose wet granulation process in Mi-Pro high shear mixers, Eur. J. Pharm. Sci. 2002, 17,

247–251

96 Faure, A.; York, P.; Rowe, R.C. Process control and scale-up of pharmaceutical wet

granulation process: a review, Eur. J. Pharm. Sci. 2001, 52, 269–277

97 Van den Dries, K.; de Vegt, O.M.; Girard, V.; Vromans, H. Granule breakage phenomena

in high shear mixer; influence of process and formulation variables and consequenes on

granule homogeneity, Powder Technol. 2003, 133, 228-236

98 Ohno, I.; Hasegawa, S.; Yada, S.; Kusai, A.; Moribe, K.; Yamamoto K. Importance of

evaluating the consolidation of granules manufactured by high shear mixer, Int. J. Pharm.

2007, 338, 79-86

99 Betz, G.; Bürgin, P.J.; Leuenberer, H. Power consumption profile analysis and tensile

strength measurements during moist agglomeration, Int. J. Pharm. 2003, 252, 11-25

100 Betz, G.; Bürgin, P.J.; Leuenberer, H. Power consumption measurement and temperature

recording during granulation, Int. J. Pharm. 2004, 272, 137-149

101 Ritala, M.; Holm, P.; Schaefer, T.; Kristensen, H.G. Influence of liquid bonding strength

on power consumption during granulation in a high shear-mixer, Drug Dev. Ind Pharm.

1988, 14, 1041-1060

49

102 Paris, L.; Stamm, A. Optimal massing liquid volume determination by energy

consumption measurement: Study of the influence of some physical properties of solvents

and products used, Drug Dev. Ind Pharm. 1985, 11, 361- 386

103 Terashita, K.; Watano, S.; Miyanami, K. Determination o fend-pont by frequency of

power consumption in agitation granulation, Chem. Pharm. Bull. 1990, 38, 3120–3123

104 Holm, P.; Schaefer, T.; Kristensen, H.G. Granulation in high-speed mixers. Part. V. Power

consumption and temperature changes during granulation, Powder Technol. 1985, 43, 213-

223

105 Holm, P.; Schaefer, T.; Kristensen, H.G. Granulation in high-speed mixers. Part. VI.

Effects of process condition on power consumption and granule growth, Powder Technol.

1985, 43, 225-233

106 Holm, P.; Schaefer, T.; Larsen, C. End-Pont Detection in a Wet Granulation Process,

Pharm. Dev. Technol. 2001, 6, 181-192

107 Badawy, S.I.F.; Menning, M.M.; Gorko, M.A; Gilbert D.L. Effect of process parameters

on compressibility of granulation manufactured in a high-shear mixer, Int. J. Pharm., 2000,

198, 51-61

108 Keleb, E.I.; Vermeier, A.; Vervaet, C.; Remon, J.P. Extrusion granulation and high shear

granulation of different grades of lactose and highly dosed drugs: A comparative study,

Drug Dev. Ind. Pharm. 2004, 30, 679-691

109 Vromans, H.; Poels-Janssen, H.G.M.; Egermann, H. Effect of High.Shear Granulation on

Granulate Homogeneity, Pharm. Dev. Technol. 1999, 4, 297-303

110 Zuurman, K.; Bolhuis, G.K.; Vromans, H. The relationship between bulk density and

compactibility of lactose, Int. J. Pharm. 1994, 102, 1-9

111 Zuurman, K.; Bolhuis, G.K.; Vromans, H. Effect of binder on the relationship between

bulk density and compactibility of lactose granulations, Int. J. Pharm. 1995, 119, 65-69

112 Landin, M.; Martinez-Pacheco, R.; Gomez-Amoza, J.L.; Souto, C.; Concheiro, A.; Rowe,

R. C. The effect of county of orogin ont he properties of dicalcium phosphate dihydrate

powder, Int. J. Pharm. 1994, 103, 9-18

50

113 Landin, M.; Martinez-Pacheco, R.; Gomez-Amoza, J.L.; Souto, C.; Concheiro, A.; Rowe,

R. C. Dicalcium phosphate dihydrate for direct compression: Characterisation and

intermanufacturer variability, Int. J. Pharm. 1994, 109, 1-8

114 Johansson, B.; Alderborn, G. The effect of shape and porosity ont he compression

behavior and tablet forming ability of granular materials formed from microcrystalline

cellulose, Eur. J. Pharm. and Biopharm. 2001, 52, 347–357

Acknowledgements

I am very grateful to my supervisor

Professor Dr. Klára Pintye-Hódi DSc.

for her support. I greatly appreciate her continuous help during the preparation of my thesis.

I owe my warm gratitude to her for her criticism, encouragement and numerous discussions

during my Ph. D. work.

My warmest thanks go to

Dr. Attila Bódis

Head of the Pharmaceutical Technology Department, for his help, encouragement and invaluable

advice.

I would like to thank

Professor Dr. István Erős DSc.

Head of the Ph.D. programme Pharmaceutical Technology

and

Professor Dr. Piroska Szabó-Révész DSc.

present Head of Department of Pharmaceutical Technology,

for providing me with the possibility to complete my work.

I express my grateful thanks to

Mátyás Koncz

Head of the Solid Dosage Forms Plant

and

Gábor Toma

Deputy-head of the Solid Dosage Forms Plant

for providing possibility for me to carry out scientific research in the Plant.

I thank all members of the Gedeon Richter Plc. for their collaboration in this work.

ANNEX

Related articles

I.

II.

III.

S114 Posters / European Journal of Pharmaceutical Sciences 25S1 (2005) S1–S226

P-45

The effects of different drying techniques on the porosityparameters of granules at production scale

A. Hegedusa, A. Kelena, K. Pintye-HodibaGedeon Richter Ltd., H-1475 Budapest, P.O. Box 27.,Hungary; bDepartment of Pharmaceutical Technology,University of Szeged H-6720 Szeged, E¨otvos Str. 6., Hungary

1. Introduction

There are two known closed-system wet granulation pro-cedures: high-shear granulation and fluid-bed granulation.These techniques differ in the modes of agitation of the solidparticles, and for this reason there are also differences ingranule growth.

In the course of fluid-bed granulation, the powder bed iskept in motion by specially treated (filtered, temperature andhumidity-controlled) air, which is introduced through a sieveplate in the base of the granulator.

In high-shear granulation, an impeller is used to agitate thesolid particles within an enclosed space. The binding agentis added or sprayed in from above. The mixing, densificationand agglomeration of the wet material is performed by theimpeller through the exertion of shearing and compactingforces.

Originally, high-shear granulators did not have a dryingcapability, which means that the wet granules produced inthese machines had to be dried in a separate machine, suchas a fluid-bed drier. Later, these granulators were further de-veloped into what are termed “single-pot” systems, whichare capable of performing all the processes of mixing, gran-ulation, drying and mixing. The possible drying methods arevacuum, vacuum-microwave and gas-vacuum methods, all ofwhich can be combined with side-wall heating [1–3].

Our objective was to compare the porosities of granulesthat were all produced using a traditional high-shear granu-lation, but were dried using either a fluid-bed dryer (GlattWSG 200), or vacuum-drying technology (Collette UltimaPro 600).

2. Materials and Methods

The given tablets contained 50% w/w active ingredient.The binding solution was an aqueous solution of PVP K-30(4.5% w/w). The other excipients were corn starch (30% w/w)as diluent; colloidal anhydrous silica (4% w/w) and glycerine(1.5% w/w) as moisture regulator; and microcrystalline cel-lulose (7.9% w/w), talc (1.6% w/w) and magnesium stearate(0.5% w/w) to improve tablet formation. We used the samecomposition and batch size (150 kg) in both sets of equip-ment.

In both cases we performed the granulation in a ColletteUltima Pro 600 single-pot machine, while the drying wascarried out in either a Glatt WSG 200 fluid bed granulator

and drier or a Collette Ultima Pro 600 single-pot granulator.The two machines are shown in Figs. 1 and 2.

The properties of granules and tablets are also influencedby the porosity of the granules. Porosity can be definedthrough the relationship between the true (�true) and tapped(�tapp) densities, using the following equation[4] :(

1 − ρtapp

ρtrue

)× 100

The true density (�true) was determined using Stereopyc-nometer SPY-5 (Quantachrome Corp.). The pycnometric truedensity is determined by measuring the volume occupied bya known mass of powder which is equivalent to the volumeof helium gas displaced. The true density was calculated thefollowing equation:

ρtrue = w

v

where w = weight of samples, v = true volume of samples.

3. Results and Discussion

We know from the specialist literature that the porosity ofgranules is affected considerably by the impeller speed andthe wet massing time [5]. However, less research has beenconducted into the extent to which the porosity of granulesprepared using the same granulation technology is influencedby the subsequent use of different drying methods.

Posters / European Journal of Pharmaceutical Sciences 25S1 (2005) S1–S226 S115

Table 1

Porosity of granules subjected to different drying techniques

Glatt WSG 200 Collette Ultima Pro 600

Porosity� (%) 73.9 63.1

The data in Table 1 show that granules dried in a vacuumchamber are more porous than those dried using a fluid-bedprocess, although the drying process is slower. This is dueto the mechanism by which the moisture is forced out of thegranule’s capillaries under sub-atmospheric pressure, whichresults in the formation of “channels” in the granule’s interioras the moisture leaves the granule. In the course of fluid-bed drying, which takes places at atmospheric pressure, thegranules dry from their surface inwards, which results in alower level of porosity.

A reduction in porosity generally leads to a deteriorationin compressibility. In the case of the systems examined byus, this took the form of a shift in the range of pressing forcerequired in order to produce a tablet of the same hardness.

Besides the porosity test, our findings are also corrobo-rated by the SEM (Scanning Electron Microscopy) imagesshown in Figs. 3 and 4.

The granules dried in a vacuum chamber are moregeometrically regular, and spherical, and thus they havea different external physical structure than that of thegranules dried using fluid-bed technology. In terms of theirtablet-forming properties, these granules fill the die moreevenly.

4. Conclusion

Granules produced in a traditional high-shear granulatorand dried in a vacuum chamber are more porous, due to thespecial characteristics of the drying process; and, in contrastto granules dried using fluid-bed technology, they retain theirspherical shape. The porosity test shows the ratio of the totalvolume of the pores in the granules to the total volume of thegranules, thereby making it suitable for a more discriminativeexamination of the differences between the granules, and en-abling us to draw conclusions regarding their tablet-formingproperties.

References

[1] Parikh, D., 1997. Handbook of Pharmaceutical Granu-lation Technology; Drugs and Pharmaceutical Sciences,Marcel Dekker Inc. New York, USA, p. 81, p. 151–204.

[2] Stahl, H., 2000. Pharm. Technol. Eur. 12, 192–201.[3] Stahl, H., 2004. Pharm. Technol. Eur. 16, 23–33.[4] Kumar, V., Reus-Medina, M. L., Yang, D., 2002. Int. J.

Pharm. 235, 129–140.[5] Badawy, S, I. F., Menning, M. M., Gorko, M. A, Gilbert

D. L., 2000. Int. J. Pharm. 198, 51–61.

P-46

Preparation and evaluation of ketoprofen loaded self-microemulsifying systems

M. Homara, M. Markocicb, M. GasperlinbaLek Pharmaceuticals d.d., Research and Development,SI-1526 Ljubljana, Verovˇskova Str. 57., Slovenia;bFacultyof Pharmacy, University of Ljubljana, SI-1000 Ljubljana,Askerceva Str. 7., Slovenia

1. Introduction

In recent years more and more ingredients are known thatare active but exhibit low bioavailability due to their low sol-ubility. Formulation of emulsions and microemulsions (ME)can greatly enhance the bioavailability of many such sub-stances. The main advantages of ME are thermodynamicstability, spontaneous formation, increased drug solubilityand permeability enhancement. Self-emulsifying (SES) andself-microemulsifying (SMES) drug delivery systems, whichform emulsions and ME on gentle agitation in aqueous media,offer additional advantages, the most important being theirgreater stability, compared to emulsions or even microemul-sions, and greater drug loading capacity [1,2]. Ketoprofenwas used as a model drug.

IV.

A

(wgafl©

K

1

wafisibapig

cft2

m

0d

International Journal of Pharmaceutics 330 (2007) 99–104

Comparison of the effects of different drying techniques on properties ofgranules and tablets made on a production scale

Agota Hegedus a, Klara Pintye-Hodi b,∗a Gedeon Richter Ltd., H-1475 Budapest 10, PO Box 27, Hungary

b Department of Pharmaceutical Technology, University of Szeged, H-6720 Szeged, Eotvos u. 6, Hungary

Received 24 April 2006; received in revised form 21 August 2006; accepted 6 September 2006Available online 10 September 2006

bstract

The aims of this study were to compare the properties of granules prepared in a high-shear granulator and dried by using different methodsfluid-bed and microwave-vacuum drying) and to compare the properties of tablets pressed from such granules. Experiments on a production scaleere performed with Collette Ultima Pro 600 single-pot processing equipment and a Glatt WSG 200 fluid-bed granulator and drier. The particles

ranulated in the traditional high-shear granulator and dried in a vacuum chamber had a higher porosity and higher bulk and tapped densities, asconsequence of the special characteristics of the drying process. They retained their spherical form, in contrast with the particles dried via theuid-bed technology. The two types of granules required different compressing forces for tabletting.2006 Elsevier B.V. All rights reserved.

; Parti

cdtotoot

aw

im

td

eywords: Single-pot processing equipment; Fluid-bed drier; Production scale

. Introduction

Granulation is a size-enlargement process in the course ofhich small particles are formed into larger, physically strong

gglomerates in which the original particles can still be identi-ed. The agglomeration of solid particles renders them moreuitable for further processing, such as tablet formation. Itmproves the flowability, ensures optimal particle size distri-ution and better homogenization of the active ingredient, andllows control of the granules, making them suitable for com-ression. In the wet-granulating process, a granulating liquids used to facilitate the agglomeration process, and the moistranules are then dried (Parikh, 1997).

Drying involves the removal of liquid from solid material thatontains moisture, through a process of evaporation resultingrom the application of heat. Thermal energy can be applied tohe granules by convection, conduction or vacuum drying (Fox,

005).

1.1. Convection is achieved by means of a flowing gaseousedium, in which the gaseous particles transmit heat while

∗ Corresponding author. Tel.: +36 62 545 576; fax: +36 62 545 571.E-mail address: klara.hodi@pharm.u-szeged.hu (K. Pintye-Hodi).

tSmp2

o

378-5173/$ – see front matter © 2006 Elsevier B.V. All rights reserved.oi:10.1016/j.ijpharm.2006.09.001

cle size distribution; Bulk and tapped densities; Porosity; SEM

hanging place. Fluid-bed drying is an example of a convectiverying method. In the process of fluid-bed drying, the granuleso be dried are placed in a device fitted with a perforated screenr sieve, and air is circulated through this layer at a rate sufficiento lift and separate the granules, which are set in motion and taken what is termed a fluidized state. The drying occurs as a resultf the consequent intensive contact between the granules andhe gaseous drying medium.

1.2. Conduction can be attained by heat exchange betweendjacent particles of matter, heat transfer through a jacked bowlall and vacuum drying.1.3. In the process of vacuum drying, the material is placed

n a vacuum chamber, and the heat necessary to remove theoisture is applied directly to the solid material.The process of pure vacuum drying requires a longer drying

ime, but its undisputed advantage over other methods is that therying takes place at a lower temperature, which could be impor-ant when heat-sensitive materials are to be dried (Fox, 2005;tahl, 2004). Gas-assisted vacuum drying, and more commonlyicrowave-vacuum drying, allow quicker drying in a single-pot

rocessor, used consecutively or simultaneously (Fox and Bohle,001; McMinn et al., 2005).

In production-scale pharmaceutical manufacturing, the meth-ds most commonly used to produce granules are fluid-bed

1 l Journal of Pharmaceutics 330 (2007) 99–104

gltfflt(iaMbnetmpcewTtacrioasoho(t2

gua

2

2

T(a(l(c

2

Uc

(e

2

eTwttaufAgabsfaac

2

u

00 A. Hegedus, K. Pintye-Hodi / Internationa

ranulation and drying, or a combination of high-shear granu-ation and fluid-bed drying. In recent years, however, single-potechnology has grown in popularity, partly because the trans-er of the moist granules from the high-shear granulator to theuid-bed drier is critical. The single-pot equipment has taken

he form of a mixer/granulator retrofitted with a drying unitStahl, 2000). The drying unit is capable of pure vacuum dry-ng, microwave-vacuum drying, gas-assisted vacuum drying, orcombination of microwave and gas-assisted vacuum drying.icrowaves are waves of electromagnetic radiation, generated

y magnetrons under the combined action of electric and mag-etic forces. Microwave drying is based on the absorption oflectromagnetic radiation by dielectric materials. The dielec-ric material is placed in an electromagnetic field, when the

aterial becomes polarized and stores electrical energy througholarization. The level of polarization depends on the state andomposition of the material and the frequency of the appliedlectric field. For pharmaceutical-industry drying, microwavesith a frequency of 2450 MHz (wavelength 12.2 cm) are used.he microwaves are not forms of heat, but rather forms of energy

hat are manifested as heat through their interaction with materi-ls. The permittivity (ε) of materials sensitive to microwaves isomplex and comprises two parts, the first corresponding to theeal part or relative dielectric constant, and the second represent-ng the imaginary part or loss factor. The dielectric loss factorf a material is a measure of how much heat is generated insidematerial per unit time when an electric field is applied, when

ubjected to microwave heating (McLoughlin et al., 2003). Mostf the materials commonly used in the pharmaceutical industryave a relatively low loss factor and absorb microwave powernly at high field strengths. By comparison, granulation liquidswater or organic solvents) have high loss factors relative tohe dry materials used (Fox and Bohle, 2001; Pere and Rodier,002).

The purpose of this study was to compare the properties ofranules produced in the same manner, through high-shear gran-lation, but dried by using two different techniques (fluid-bednd microwave-vacuum drying).

. Materials and equipment

.1. Materials

The given tablets contained 50% (w/w) active ingredient.he binding solution was an aqueous solution of PVP K-30

4.5%, w/w). The other excipients were corn starch (30%, w/w)s diluent; colloidal anhydrous silica (4%, w/w) and glycerine1.5%, w/w) as moisture regulators; and microcrystalline cel-ulose (7.9%, w/w), talc (1.6%, w/w) and magnesium stearate0.5%, w/w) to improve tablet formation. We used the sameomposition and batch size (150 kg) in both sets of equipment.

.2. Equipment

In both cases, we performed the granulation in a Colletteltima Pro 600 single-pot processor (Fig. 1). The drying was

arried out in a Glatt WSG 200 fluid-bed granulator and drier

aTtw

Fig. 1. Photograph of Collette Ultima Pro 600 single-pot equipment.

Fig. 2) and in Collette Ultima Pro 600 single-pot processingquipment.

.2.1. Collette Ultima Pro 600This is a closed, single-pot system, which means that the

ntire granulation process can be performed in the one device.he bowl has a jacket wall to allow the circulation of hot or coldater, in order to regulate the temperature of the product. Both

he impeller and the chopper are positioned vertically, and pro-rude into the machine from above. The speeds of the impellernd the chopper are adjustable within a given range. The liq-id binder addition is regulated, and the machine is suitableor the spraying of binder solution with high or low viscosity.

number of parameters can be used to set up the end-point ofranulation: the processing time, the torque, the product temper-ture, etc., or a combination of these. The granules can be driedy vacuum and microwave energy, which can be combined withide-wall heating. The drying cycle of this machine is there-ore more energy-efficient than other drying processes. Therere three possible drying methods: vacuum, vacuum-trans flownd vacuum-microwave. The machine is suitable for computer-ontrolled, automated manufacturing.

.2.2. Glatt WSG 200This is also a single-pot system which is suitable for gran-

lation and drying process in the one device. There is an inlet

ir handling unit fit for air filtering, air heating, and air cooling.he air must be introduced at the bottom of the product con-

ainer through the perforated air distributor plate (screen type)hich is important to fluidize and mix material in the container.

A. Hegedus, K. Pintye-Hodi / International Jour

Tdstpaae

3

3

past6ic9

o

Ufl3

U

atmp

tt5

p

tth

3

3

iA

3

gw

mracm

3

prt

ε

oppvc

ρ

w

Fig. 2. Photograph of Glatt WSG 200 fluid-bed granulator and drier.

he spraying head with three or six nozzles can be set in threeifferent positions over the distribution plate. Within the expan-ion chamber, granules are formed. There are bag filters withinhe machine which retain the particles. The filter bag is made ofolyester-lined material which is of a certain mesh size. Safetyir filters are built in the outlet air product. Main processes suchs air flow and spraying rate are controlled. The machine isquipped with a data acquisition system.

. Methods

.1. Preparation of granules and tablets

We performed the granulation in the Collette Ultima Pro 600rocessing equipment. The active ingredient, the corn starchnd the colloidal anhydrous silica were homogenized (impellerpeed: 65 rpm, process time: 6 min). The liquid binder was addedo the powder mixture (impeller speed: 95 rpm, chopper speed:00 rpm, liquid binder flow rate: 7 kg/min, process time: approx-mately 4 min). After addition of the liquid binder, mixing wasontinued to the torque value (wet massing—impeller speed:5 rpm, chopper speed: 2700 rpm, torque value: 6.5 kW).

The granules were dried to the prescribed value of the lossn drying by using two different methods.

In one case, we removed the wet granules from the Colletteltima Pro 600 equipment, loaded them into the Glatt WSG 200

uid-bed drier and performed the drying at 60 ◦C (process time:5 min, maximum product temperature: 35.5 ◦C).

In the other case, drying was carried out in the Colletteltima Pro 600 equipment by microwave-vacuum drying. The

3

ea

nal of Pharmaceutics 330 (2007) 99–104 101

im was to achieve the shortest possible drying time, and weherefore used the maximum forward energy (vacuum: 50 mbar,

icrowave forward energy: 22 kW, continuous mixing: 20 rpm,rocess time: 58 min, maximum product temperature: 43 ◦C).

After drying, the granules were sized in a 1.5 mm sieve (rota-ion speed: 500 rpm), and next homogenized for 2 min with theabletting excipients (microcrystalline cellulose, talc) and thenmin (magnesium stearate) in a container blender.

Fig. 3 shows the detailed flowcharts of the manufacturingrocesses.

We determined the size distribution of the granules, theirapped and bulk densities, porosity and moisture content, andook SEM photographs. The individual and average masses,eights and hardnesses of the tablets were examined.

.2. Testing of granules and tablets

.2.1. Particle size analysisWe determined the particle size distribution of an approx-

mately 25 g sample of the final granules, using a Hosokawalpine 200 LS air jet sieve with an array of five sieves.

.2.2. Bulk and tapped densitiesHundred millilitres of granules was poured into a 250 ml

raduated tared measuring cylinder, and the granules were theneighed and their bulk density, ρT, was determined in g/100 ml.The density of 100 ml of granules of known weight was

easured with a Stampfvolumeter 2003 (J. Engelsmann Appa-atebau, Ludwigshafen, Germany). After 200–300 taps (whenconstant value had been achieved), the volume of the tapped

olumn of granules was read off, and the density, ρT, was deter-ined in g/100 ml.

.2.3. PorosityThe properties of granules and tablets are influenced by the

orosity of the granules. Porosity can be defined through theelationship between the particle (ρpart) and tapped (ρT) densi-ies, using the following equation (Kumar et al., 2002):

=(

1 − ρT

ρpart

)× 100

The particle density (ρpart) was determined with a Stere-pycnometer SPY-5 (Quantachrome Corp.). The pycnometricarticle density was determined by measuring the volume occu-ied by a known mass of powder, which is equivalent to theolume of helium gas displaced. The particle density was cal-ulated via the following equation:

part = w

v

here w is the weight of sample and v is the volume of samples.

.2.4. Moisture determinationThe loss on drying of 2 g of granules (homogenized with the

xternal phase) to mass constancy at 70 ◦C was determined, withMettler Toledo HR 73 halogen moisture analyser. The loss on

102 A. Hegedus, K. Pintye-Hodi / International Journal of Pharmaceutics 330 (2007) 99–104

in tw

dr

3

b5dro

3

CsiWtsw

4

s

iisnpvcpwit is therefore not necessary to ensure a low loss on drying whenthe granules are dried (Table 1), in which case fluid-bed dryingwould be preferable. For this reason, with these products we hadthe opportunity to perform a comparative granulometric analysis

Table 1Granule properties of batches dried in the Collette Ultima Pro 600 and in theGlatt WSG 200

Glatt WSG 200 Collette Ultima Pro 600

Bulk density (g/100 ml) 68.49–71.43 79.37–83.30Tapped density (g/100 ml) 80.55–84.29 94.53–104.17

Fig. 3. Flow sheet of granulation

rying of the granules must be within the range 2.5–4.5%, thisange being suitable for the tabletting of this product.

.2.5. Scanning electron microscopy (SEM)The morphological properties of the granules prepared in

oth sets of equipment were examined with a JEOL JSM-600LV scanning electron microscope fitted with an energyispersive X-ray spectrometer. A Polaron sputter coating appa-atus was applied to induce electric conductivity on the surfacef the sample. The air pressure was 1.3–13 mPa.

.2.6. Tablet evaluationThe granules were pressed into 500 mg tablets by using a

ourtoy R190 Ft tablet press with 36 punches. The rotationalpeed of the press was 65 rpm. We measured the average andndividual masses, the thickness, the hardness (Pharma Test

HT-2ME) and the disintegration (Pharma Test PTZ-E) fiveimes in the course of the tablet-formation process. The relativetandard deviation (R.S.D.) of the mass of the individual tabletsas determined by measuring 20 tablets.

. Results and discussion

Depending on the composition of the material system and theolvents used (organic or water), and their quantities, preference

LFDP

o types of dryer and tabletting.

s given to different drying techniques (e.g. fluid-bed or vacuum)n the pharmaceutical industry. However, for certain materialystems, the differences between the drying technologies areot marked enough to make one or the other unambiguouslyreferable. The products under study do not contain organic sol-ents or materials that are sensitive to heat or oxygen, or whichontain toxic or potent compounds, in which cases the single-ot technology would be clearly preferable. On the other hand,e are not using a liquid binder with a high water content, and

oss on drying (%) 2.70–3.45 3.07–4.08ine particles (%) <21 <23

50 (�m) 310–370 360–420orosity, ε (%) 73.9 63.1

A. Hegedus, K. Pintye-Hodi / International Journal of Pharmaceutics 330 (2007) 99–104 103

obiwbotsbtp

wsmTflTntgstP

Ft

ictg

pdsctImtttoi

a

Fig. 4. SEM photographs of granules dried in Glatt WSG 200.

f different techniques used for drying wet granules preparedy using the same method. The advantage of fluid-bed dryings the short drying time, in contrast with pure vacuum drying,hich entails a long processing time. Accordingly, we com-ined vacuum drying with microwave drying, since the durationf processing is an important consideration in the pharmaceu-ical industry. In the drying process, the primary goal was tohorten the processing time. In the experiments, the differenceetween the maximum product temperatures attained with thewo drying techniques (35.5 ◦C and 43 ◦C) had no impact on theroduct quality.

Granule size increase is influenced by the impeller speed, theet massing time and the amount of liquid in the case of high-

hear granulation. In this study, the granules were granulated byeans of the same technology, but dried with different methods.he powder fraction was relatively high for both vacuum anduid-bed drying, at <21% and <23%, respectively, as shown inable 1, but a significant difference in the powder fraction wasot detected (Vromans et al., 1999). As conserns the composi-ion under study, the mean particle size (D50) was larger for theranules dried by using microwaves than for the fluid-bed dried

ample. This is because the granules collide with each other andhe wall of the equipment during the fluid-bed drying process.articles therefore constantly break off and are eroded.

Fig. 5. SEM photographs of granules dried in Collette Ultima Pro 600.

icpt

Ft

ig. 6. Drying curves of the vacuum-microwave (- - -) and fluid-bed (—) dryingechnology.

Besides D50, our findings were also corroborated by the SEMmages shown in Figs. 4 and 5. The granules dried in the vacuumhamber were more geometrically regular and spherical, andhus had a different external physical structure from that of theranules dried with the fluid-bed technology.

The physical differences between the granules could resultartly from the drying time, and partly from the nature of therying curves (Fig. 6). In order for a material system with theame moisture content to develop by the end of the drying pro-ess, approximately 1.5 times the drying time is necessary inhe case of vacuum drying than in the case of fluid-bed drying.n other words, the expulsion of moisture is slower, gentler andore even, with the result that the primary physical structure of

he granules remains more intact. In the case of fluid-bed drying,he raggedness and erosion of the granules arise not only fromhe impact, but also as a result of the sudden temperature change,wing to the rapid expulsion of moisture. This rapid evaporationnflicts more intensive damage on the granules.

It is known from the literature that the porosity of granules isffected considerably by the impeller speed and the wet mass-ng time (Badawy et al., 2000). However, less research has been

onducted into the extent to which the porosity of granules pre-ared by using the same granulation technology is influenced byhe subsequent use of different drying methods.

ig. 7. Correlation between hardness and pressing power. Vacuum-microwaveechnology (×), fluid-bed dried granules (�).

104 A. Hegedus, K. Pintye-Hodi / International Jour

Fgd

ctTowtoth

s

coa

svlbbitlsfbit

TPa

R

TD

ma(

5

aodocosec

upssfldmcdi

dtf

R

B

F

F

K

M

ig. 8. Correlation between hardness and thickness. (�) Tablets pressed fromranules dried in Collette Ultima Pro 600. (�) Tablets pressed from granulesried in Glatt WSG 200.

The data in Table 1 show that the granules dried in the vacuumhamber had a lower level of porosity than those dried by usinghe fluid-bed process, although the drying process was slower.his is due to the mechanism by which the moisture is forced outf the capillaries in the granules under sub-atmospheric pressure,hich results in the formation of “channels” in the interior of

he granules as the moisture leaves the granules. In the coursef fluid-bed drying, which takes place at atmospheric pressure,he granules dry from their surface inwards, which results in aigher level of porosity.

The lower porosity values entail higher bulk and tapped den-ity values, as shown in Table 1.

A reduction in porosity generally leads to a deterioration inompressibility. In the systems we examined, this took the formf a shift in the range of compressing force required to producetablet of the same hardness.

The correlation between compressing force and hardness ishown in Fig. 7. The granules prepared by using microwave-acuum drying are denser, with the result that the tablets areower and easily compressible, but a higher pressure force muste applied than in the case of the granules dried with the fluid-ed technology. The correlation between hardness and heights shown in Fig. 8. The height of the compressed tablets fromhe granules dried with the microwave-vacuum technology wasower than that of the tablets compressed from granules of theame hardness, dried with the fluid-bed technology. The dif-

erences in compressibility can be attributed to the differencesetween the structures of the granules, caused by the differ-ng drying technologies. As can be seen in Table 2, the use ofhe different drying techniques had no effect on the individual

able 2roperties of tablets pressed from granules dried in the Collette Ultima Pro 600nd the Glatt WSG 200

Glatt WSG200

Collette UltimaPro 600

elative standard deviation (R.S.D.) ofindividual mass from average mass (%)

<1.00 <1.07

hickness (mm) 4.01–4.17 3.91–4.09isintegration (min) <1 <1

M

P

P

S

S

V

nal of Pharmaceutics 330 (2007) 99–104

ass distribution or disintegration time of the tablets; they hadrelatively low mass distribution and short disintegration time

<1 min) in both cases.

. Conclusions

Following the wet massing process, the drying technologiespplied in the pharmaceutical industry were selected on the basisf a number of criteria, such as the properties of the active ingre-ient, the type of solvent, the processing time, etc. The choicef the most suitable technology for the given purpose requiresareful consideration and testing. Two drying techniques, basedn differing principles (fluid-bed and microwave-vacuum) wereelected for the purposes of the present research, and the prop-rties of the granules produced by using these methods wereompared.

The granules produced in the traditional high-shear gran-lator and dried in a vacuum chamber had a lower level oforosity, and higher bulk and tapped densities, owing to thepecial characteristics of the drying process. They retained theirpherical form, in contrast with the granules dried by using theuid-bed technology. These characteristics of the granules alsoetermined the properties of the tablets pressed from them, andade it necessary to apply a greater compressing force in the

ase of the granules prepared by using the microwave-vacuumrying process. At the same time, the mass distribution and dis-ntegration time were not affected.

Despite the measurable physical differences arising from theiffering principles of the two drying methods, both dryingechnologies proved highly suitable for production-scale manu-acturing of the compositions under study.

eferences

adawy, S.I.F., Menning, M.M., Gorko, M.A., Gilbert, D.L., 2000. Effect ofprocess parameters on compressibility of granulation manufactured in ahigh-shear mixer. Int. J. Pharm. 198, 51–61.

ox, B., Bohle, L.B., 2001. Vacuum and microwave dry granulated products.Chem. Eng. 108, 135–141.

ox, B., 2005. True grit: granulation and drying of delicate products. Chem.Eng. 115, 35–38.

umar, V., Reus-Medina, M.L., Yang, D., 2002. Preparation characterization,and tabletting properties of a new cellulose-based pharmaceutical aid. Int. J.Pharm. 235, 129–140.

cLoughlin, C.M., McMinn, W.A.M., Magee, T.R.A., 2003. Physical dielectricproperties of pharmaceutical powders. Powder Technol. 134, 40–51.

cMinn, W.A.M., McLoughlin, C.M., Magee, T.R.A., 2005. Microwave-convective drying characteristics of pharmaceutical powders. Powder Tech-nol. 153, 23–33.

arikh, D., 1997. Handbook of Pharmaceutical Granulation Technology; Drugsand Pharmaceutical Sciences, vol. 81. Marcel Dekker Inc., New York, pp.151–204.

ere, C., Rodier, E., 2002. Microwave-vacuum drying of porous media: experi-mental study and qualitative considerations of internal transfers. Chem. Eng.Process. 41, 427–436.

tahl, H., 2000. Single-pot system for drying pharmaceutical granules. Pharm.Technol. Eur. 12, 192–201.

tahl, H., 2004. Comparing different granulation techniques. Pharm. Technol.Eur. 16, 23–33.

romans, H., Poels-Janssen, H.G.M., Egermann, H., 1999. Effect of high-shear granulation on granulate homogeneity. Pharm. Develop. Technol. 4,297–303.

V.

A

gibmaeoa©

K

1

ttmmeeg

dta

k

0d

Chemical Engineering and Processing 46 (2007) 1012–1019

Influence of the type of the high-shear granulatoron the physico-chemical properties of granules

Agota Hegedus a,∗, Klara Pintye-Hodi b

a Gedeon Richter Ltd., H-1475 Budapest 10, P.O.B. 27, Hungaryb Department of Pharmaceutical Technology, University of Szeged, H-6720 Szeged,

Eotvos str. 6, Hungary

Received 23 May 2006; received in revised form 19 April 2007; accepted 17 May 2007Available online 20 June 2007

bstract

The purpose of this experiment was to compare the results of the granulation that can be achieved in different production-scale high-shearranulator models, and the characteristics and tablet-forming properties of the granules produced, in the case of a product with a high activengredient content. We studied granules prepared in Diosna P400 and Collette Ultima Pro 600 industrial high-shear granulators. The main differencesetween the two machines relate to the geometry of the bowl and the positioning of the chopper and impeller. The granules produced in theseachines (which have identical manufacturing capacity) have different macroscopic and microscopic structures. The aim was to create, optimize

nd reproduce a robust technology that furnishes granules (and the tablets formed from them) with similar physical properties in both sets of

quipment. With the currently studied product, it was possible to establish the optimal ranges of the impeller and chopper speeds, the water contentf the binder solution, the liquid binder flow rate, the wet massing time and the torque, thereby rendering the manufacturing process controllablend reproducible.

2007 Elsevier B.V. All rights reserved.

ulk a

hpoisacpt

soe

eywords: High-shear granulator; Production scale; Particle size distribution; B

. Introduction

Wet granulation is a technique whereby a liquid is usedo transform small solid particles into clusters of larger ones,hrough a process of agglomeration. In pharmaceutical tablet

anufacturing, the agglomeration of solid particles renders themore suitable for tablet formation. It improves the flowability,

nsures optimal particle size distribution and a better homog-nization of the active ingredient, and allows control of theranules, making them suitable for compression.

There are two known closed-system wet granulation proce-ures: high-shear granulation and fluid-bed granulation. Theseechniques differ in the modes of agitation of the solid particles,

nd for this reason there are also differences in granule growth.

In the course of fluid-bed granulation, the powder bed isept in motion by specially treated (filtered, temperature and

∗ Corresponding author. Tel.: +36 62 545 576; fax: +36 62 545 571.E-mail address: hodi k@freemail.hu (K. Pintye-Hodi).

tew

chm

255-2701/$ – see front matter © 2007 Elsevier B.V. All rights reserved.oi:10.1016/j.cep.2007.05.017

nd tapped densities; SEM

umidity-controlled) air, which is introduced through a sievelate in the base of the granulator. The binder solution is sprayednto the fluidized powder bed. The granules are created dur-ng the wetting of the powder bed, through the adhesion ofolid particles as the drops of liquid reach the powder bed. Thegglomeration of the powder takes place during the wetting pro-ess and, once the process of spraying the adhesive onto theowder bed has been completed, the granules are dried throughhe use of warm air [1].

In high-shear granulation, an impeller is used to agitate theolid particles within a closed space. The binder solution is addedr sprayed in from above. The mixing, densification and agglom-ration of the wet material are performed by the impeller throughhe exertion of shearing and compacting forces. The process isnded before the granules begin to grow uncontrollably, whichould result in the phenomenon known as “ball growth”.

High-shear granulators have long been used in the pharma-

eutical industry, both for mixing and for granulating. Originally,igh-shear granulators did not have drying capability, whicheans that the wet granules produced in these machines had

gineer

tLtaTas

somaMhtttg[

tgsu

dmftassrr

2

2

T(a(lulose (7.9%, w/w), talc (1.6%, w/w) and magnesium stearate

A. Hegedus, K. Pintye-Hodi / Chemical En

o be dried by using another machine, such as a fluid-bed drier.ater, these granulators were further developed into what are

ermed “single-pot” systems, which are capable of performingll of the processes of mixing, granulation, drying and blending.he possible drying methods are vacuum, vacuum-microwavend gas-vacuum methods [2], all of which can be combined withide-wall heating [3–5].

Pharmaceutical companies often find that they have to presentcientific evidence to justify the replacement or modernizationf equipment that could be up to one or two decades old. Phar-aceutical companies endeavour to achieve more cost-effective

nd better-regulated processes, in line with the standards of Goodanufacturing Practice (GMP). The advantages of single-pot

igh-shear granulators include the facts that the entire processakes place in one set of equipment, GMP requirements are met,he processing time is reduced, cleaning is simplified throughhe use of integrated, programmed cleaning systems, and theranulators are equipped with the appropriate safety systems4,5].

Our objective was to adapt the production-scale granula-

ion process, previously performed with a traditional high-shearranulator (Diosna P400) and a fluid-bed drier (Glatt WSG-200),o that it could be carried out in a single-pot high-shear gran-lator (Collette Ultima Pro 600). In view of the considerable

Fig. 1. Photographs of the Diosna P400 mixer granulator.

(c

F

ing and Processing 46 (2007) 1012–1019 1013

ifferences between the two machines, the fact that our experi-ents were not preceded by laboratory and pilot tests, and the

act that we were working on a production scale, we decidedo adapt the processes in two stages: first the granulation step,nd then the overall process, including the drying step. We alsoet out to compare the granulation process of a specific compo-ition in two sets of equipment, which, although identical withespect to capacity and operating principles, differ in many otherespects.

. Materials and equipment

.1. Materials

The given tablets contained 50% (w/w) active ingredient.he binder solution was an aqueous solution of PVP K-30

4.5%, w/w). The other excipients were cornstarch (30%, w/w)s diluent; colloidal anhydrous silica (4%, w/w) and glycerine1.5%, w/w) as moisture regulators; and microcrystalline cel-

0.5%, w/w) to improve tablet formation. We used the sameomposition and batch size (150 kg) in both sets of equipment.

ig. 2. Photographs of the Collette Ultima Pro 600 single-pot equipment.

1014 A. Hegedus, K. Pintye-Hodi / Chemical Engineer

Table 1Technical data of the Diosna P400 and the Collette Ultima Pro 600

Diosna P400 Collette Ultima Pro 600

Bowl capacity (l) 385 400Impeller speed (rpm) 64 or 129 From 14 to 135Chopper speed (rpm) 1450 or 2930 From 600 to 2700

T

2

cCc[fl

f

2

lccf

mciisrtp

2

eTcuaiTavepu

Fa

ip speed (m/s) 3.65 (64 rpm) 3.52 (64 rpm)7.36 (129 rpm) 7.09 (129 rpm)

.2. Equipment

We performed the granulation in a Diosna P400 (Fig. 1)onventional mixer-granulator (without drying facility) and aollette Ultima Pro 600 single-pot equipment (Fig. 2). Techni-al data on the two types of equipment can be seen in Table 13,6]. In both cases, drying was carried out in a Glatt WSG 200uid-bed granulator and drier.

The main differences between the two granulators are asollows.

.2.1. Diosna P400It has a single-walled design. Neither the side-walls nor the

id of the device can be temperature-controlled. The machine isone-shaped, with the impeller positioned vertically, and thehopper horizontally. The impeller protrudes into the devicerom below. The impeller blades and the special shape of the

codu

ig. 3. Flow sheet of the granulation in two types of high-shear granulator and in a flnd Collette Ultima Pro 600.

ing and Processing 46 (2007) 1012–1019

achine ensure effective mixing. Both the impeller and thehopper have two speed settings, with no fine adjustment. Thempeller and the chopper are fitted with a time switch, whichs the only means of setting an end-point. (There is no mea-urement of torque or power consumption.) It is not possible toegulate the application of the binder solution. The quality ofhe granules depends largely on the skill and experience of theersonnel carrying out the production.

.2.2. Collette Ultima Pro 600This is a closed, single-pot system, which means that the

ntire granulation process can be performed in the one device.he bowl has a jacket wall to allow the circulation of hot orold water, in order to regulate the temperature of the prod-ct. Both the impeller and the chopper are positioned vertically,nd protrude into the machine from above. The speeds of thempeller and the chopper are adjustable within a given range.he liquid binder addition is regulated, and the machine is suit-ble for the spraying of the binder solution with high or lowiscosity. A number of parameters can be used to set up thend-point of granulation: the processing time, the torque, theroduct temperature, etc., or a combination of these. The gran-les can be dried by vacuum and microwave energy, which

an be combined with side-wall heating. The drying cyclef this machine is therefore more energy-efficient than otherrying processes. There are three possible drying methods: vac-um, vacuum-trans flow and vacuum-microwave. The machine

uid-bed granulator and dryer. Setting ranges of the parameters in Diosna P400

gineering and Processing 46 (2007) 1012–1019 1015

i[

3

3

i

ps6wtac

(((((

so4dt(wctieouoett

Fl

s[

tdoavftb

3

3

iA

3

m

TS

R

1

A. Hegedus, K. Pintye-Hodi / Chemical En

s suitable for computer-controlled, automated manufacturing2,6,7].

. Methods

.1. Manufacturing process

Fig. 3 shows the flowcharts of the manufacturing processesn the Diosna P400 and the Collette Ultima Pro 600.

In the Diosna P400, the binder solution is poured onto theowder during the second step, and the addition of the binderolution is therefore not regulated. In the Collette Ultima Pro00, in the course of liquid binder addition and wet massing,e varied five parameters that could affect the characteristics of

he granules [8]. We then compared the granules thus produced,nd the granulating processes used. Table 2 details 11 differentombinations of the following five parameters [9,10]:

1) impeller speed,2) chopper speed,3) water content of the binder solution,4) liquid binder flow rate,5) wet massing time.

At the beginning of our experiments, we adapted the machineettings (the impeller and chopper speeds, and the water contentf the binder solution) to correspond to those of the Diosna00 (Table 2, setting 0). With setting 0, we were unable to pro-uce granules in the Collette Ultima Pro 600. We observed thathe product manufacturing processes are not always transferablewith identical technological parameters) between granulatorsith the same production capacity, but different geometric

haracteristics. Through our experiments, we determined theorque values representing the granulation end-points at variousmpeller speeds [11]. These values are shown in Fig. 4. In ourxperiment, at the torque value associated with an impeller speedf 65 rpm (530 Nm), aggregation did not occur, and no gran-les were formed. The explanation for this is that the motions

f the materials differed because of the geometrical differ-nces (with respect to both the shape and the positioning ofhe impeller and the chopper) between the two machines, withhe result that the different systems required different impeller

m

wr

able 2etting parameters of the Collette Ultima Pro 600

un Impeller speed (rpm) Chopper speed (rpm) Water content of the binde

0 65 2700 261 95 1500 272 95 2700 243 80 1500 264 80 2700 265 135 1500 266 135 2700 267 95 1500 268 95 2700 269 95 2700 260 95 2700 26

ig. 4. Relationship between the torque value and impeller speed at the granu-ation end-points by Collette Ultima Pro 600.

peeds to achieve the same degree of granulation formation12,13].

In both cases, the granules were dried as follows.We discharged the wet granules from the high-shear granula-

or equipment and loaded them into the Glatt WSG 200 fluid-bedryer and performed the drying at 60 ◦C, temperature at the endf the drying: 34 ◦C. After first drying, the granules were sized in1.5 mm sieve, and then followed the drying step at 60 ◦C to thealue of the loss on drying. Dried granules were homogenizedor 2 and 5 min with the tabletting excipients (microcrys-alline cellulose, talc and magnesium stearate) in a containerlender.

.2. Testing of granule characteristics

.2.1. Particle size analysisWe determined the particle size distribution of an approx-

mately 25 g sample of the final granules, using a Hosokawalpine 200 LS air jet sieve with an array of five sieves.

.2.2. Bulk and tapped densitiesWe poured 100 ml of granules into a 250 ml graduated tared

easuring cylinder, and then weighed the granules and deter-

ined their bulk density, ρt, in g/100 ml.We measured the density of 100 ml of granules of known

eight with a Stampfvolumeter 2003 (J. Engelsmann Appa-atebau, Ludwigshafen, Germany). After 200–300 taps (when a

r solution (kg) Liquid binder flow rate (kg/min) Wet massing time (min)

7 127 27 57 47 57 27 27 43 2

12 47 4

1 gineering and Processing 46 (2007) 1012–1019

cti

3

ctwu

C

w

3

gadr

3

pspws

4

4

iW

lo

Table 3Active ingredient physical properties

Particle size analysis (�m)D10 25D50 125D90 300

Carr compressibility index (%) 21.70BT

oirlond

baqo>b(

gdc<cto

4i

TG

BTCL

P

016 A. Hegedus, K. Pintye-Hodi / Chemical En

onstant value had been achieved), we read off the volume of theapped column of granules, and the density, ρT, was determinedn g/100 ml [14].

.2.3. Carr compressibility indexThe flow properties of the granules can be determined through

ompaction, and the extent of the compaction can be definedhrough the relationship between the bulk and tapped densities,hich can be expressed with the Carr compressibility index [15],sing the following equation:

arr compressibility (%) = ρT − ρt

ρT× 100

here ρT is the tapped density and ρt is the bulk density.

.2.4. Moisture determinationWe determined the loss on drying to mass constancy of 2 g of

ranules (homogenized with the external phase) at 70 ◦C, usingMettler Toledo HR 73 halogen moisture analyser. The loss onrying of the granules must be within the range 2.5–4.5%. Thisange is suitable for the tabletting of this product.

.2.5. Scanning electron microscopyWe examined the morphological properties of the granules

repared in both sets of equipment, using a JEOL JSM-5600LVcanning electron microscope (SEM) fitted with an energy dis-ersive X-ray spectrometer. A Polaron sputter coating apparatusas applied to induce electric conductivity on the surface of the

ample. The air pressure was 1.3–13 mPa.

. Result and discussion

Table 3 shows the active ingredient’s physical properties.

.1. Particle size analysis

The rate of granule growth is influenced by the speed of thempeller, the wet massing time and the amount of binder [16].

e carried out our experiments on a production scale.In the Collette Ultima 600 machine, the end-point of granu-

ation was the torque necessary for the given impeller speed (inrder to avoid possible variations between the different batches

flbea

able 4ranules properties in the Collette Ultima Pro 600 of the different setting parameters

1 2 3 4

ulk density (g/100 ml) 69.44 68.49 75.76 70.42apped density (g/100 ml) 84.72 76.71 84.85 80.28arr compressibility index 18.04 10.72 10.71 12.28oss on drying (%) 3.44 2.90 3.15 3.06

article size analysis (%)<0.090 mm 18.7 18.6 17.6 21.00.090–0.180 mm 7.5 35.2 16.2 17.30.180–0.355 mm 15.8 26.7 29.5 23.50.355–0.710 mm 33.7 18.3 24.9 29.90.710–1.000 mm 17.9 1.2 8.9 7.0>1.000 mm 6.4 0 3.2 1.3

ulk density (g/100 ml) 71.50apped density (g/100 ml) 91.30

f active ingredient, which, for a product containing 50% activengredient, could result in a substantial divergence). Even withelatively short wet massing periods, for preparations that granu-ated well, only small differences were detected, even at a varietyf impeller speeds. This means that these two parameters areot of great importance as factors influencing the particle sizeistribution in the composition studied.

The most important factor influencing the particle size distri-ution of the granules proved to be the amount of liquid binder,s may be seen in Table 4. In experiment 1, in which the greatestuantity of liquid was used, the highest proportions were thosef the largest particles, i.e. >1000 �m (6.4%), and of particles355 �m (58%). In experiment 2, where the lowest quantity ofinder liquid was used, the proportion of particles <180 �m was53.8%).

In comparison, the granules prepared in the Diosna P400ranulator displayed a considerable variance in their particle sizeistribution, as is to be seen in Table 4. The proportion of parti-les >1000 �m was between 0.9% and 14.8%, while the fraction180 �m varied between 24.0% and 60.5%. These differencesould have been caused by the initial uneven distribution of mois-ure, and the subjectivity involved in determining the end-pointf granulation.

.2. Bulk and tapped densities and Carr compressibilityndex

The Carr compressibility index is widely used to analyse the

ow properties of granules. If the Carr compressibility index isetween 5 and 10, then the granules have excellent flow prop-rties, whiles values of 12–16 indicate good, 18–21 acceptable,nd 23–28 poor flowability. In this case the Carr compressibility

5 6 7 8 9 10

70.42 72.46 71.43 69.44 73.53 68.4982.39 84.06 77.14 81.94 83.82 80.8214.53 13.80 7.40 15.26 12.28 15.263.36 2.99 3.45 3.49 3.55 3.25

13.6 17.6 15.5 17.1 12.9 19.119.9 19.4 19.8 18.1 15.5 14.033.4 33.3 32.0 29.3 24.7 23.119.1 19.7 21.1 26.8 33.1 29.5

8.7 8.0 9.2 8.7 9.5 11.45.3 2.0 2.4 0 4.3 2.9

gineering and Processing 46 (2007) 1012–1019 1017

iceom7spatFhvpttly

Ctiwrf

4

mv1biP

d<tia(

TG

BTCL

P

Table 6Granules properties of reproduction batches in the Collette Ultima Pro 600

R/1 R/2 R/3 R/4 R/5 R/6

Bulk density (g/100 ml) 68.49 69.44 69.44 69.44 71.43 69.44Tapped density (g/100 ml) 80.82 82.64 81.94 80.55 84.29 81.94Carr compressibility index 15.26 15.97 15.26 13.79 15.26 15.26Loss on drying (%) 3.00 2.96 3.45 3.42 3.35 2.70

Particle size analysis (%)<0.090 mm 19.7 16.2 20.6 20.2 18.9 19.70.090–0.180 mm 20.4 15.6 19.2 15.9 15.7 14.70.180–0.355 mm 26.7 27.0 25.4 24.7 24.8 21.90.355–0.710 mm 24.9 28.4 28.2 29.8 29.9 32.1

tr

up(10.80%) for the D batches than the R batches (4.74%). Thebulk and tapped densities varied within a wider range andhigher relative standard deviation for the D batches (bulkdensity: 68.49–75.76 g/100 ml, R.S.D.: 4.34%; tapped density:

A. Hegedus, K. Pintye-Hodi / Chemical En

ndex indicate weakly or poorly flowing granules, as the Carrompressibility index was 18.04. This result can be consid-red acceptable. The Carr compressibility index, with a valuef 7.40, showed that the granules produced in experiment 7 (aedium quantity of binder solution: 26 kg, a medium flow rate:kg/min, a medium impeller speed: 95 rpm, and a low chopper

peed: 1500 rpm) had excellent flow properties. The granulesrepared in experiments 2 (Carr compressibility index 10.72)nd 3 (10.71) were similarly good. The bulk and tapped densi-ies of the granules produced in experiments 2 and 7 were low.or compositions with a high active ingredient content, relativelyigh bulk and tapped densities are favourable from the point ofiew of tablet formation, since the volume of die filling is pro-ortionally reduced [17]. With respect to the flow properties ofhe granules, the settings used in experiment 3 (a medium quan-ity of binder solution: 26 kg, a medium flow rate: 7 kg/min, aow impeller speed: 80 rpm, and a low chopper speed: 1500 rpm)ielded the best results.

In comparison, there were no significant differences in thearr compressibility index values of the granules prepared in

he Diosna P400, given in Table 5. The best Carr compressibilityndex was 11.51, but all the granules had good flow properties,ith values ranging between 11.51 and 15.97. The bulk density

anged between 68.49 and 75.76, while the tapped density variedrom 78.47 to 87.50.

.3. Reproduction

We performed our experiments with production-scaleachinery. Reproducibility is important, and a prerequisite for

alidation in the pharmaceutical industry. For this reason, setting0 was selected as medium value to manufacture a further fiveatches in the Collette Ultima Pro 600 (the results are shownn Table 6), and the granules with those prepared in the Diosna400 were compared.

With respect to the particle size distribution, for all the repro-uction (R) batches, the fraction of particles >1000 �m was3%, but for D/2 and D/3 (14.8% and 10.8%) it exceeded

his value. In all cases, the fraction of the particles <90 �mn the R batches < 20%, but for D/4 and D/5 it was 20.8%nd 25.5%. For the R batches, the majority of the granules>50%) fell into the fraction 180–710 �m, in contrast with

able 5ranules properties in the Diosna P400

D/1 D/2 D/3 D/4 D/5 D/6

ulk density (g/100 ml) 73.53 74.63 75.76 69.44 68.49 69.44apped density (g/100 ml) 87.50 86.57 87.88 79.86 80.14 78.47arr compressibility index 15.97 13.79 13.79 13.05 14.54 11.51oss on drying (%) 3.06 3.10 3.11 3.09 3.04 3.50

article size analysis (%)<0.090 mm 11.7 16.0 15.7 19.1 20.8 25.50.090–0.180 mm 18.8 8.0 9.7 29.8 39.7 16.20.180–0.355 mm 32.6 14.9 26.2 28.2 18.7 24.60.355–0.710 mm 16.2 27.5 23.2 15.8 13.7 22.30.710–1.000 mm 11.6 18.8 14.4 6.2 5.9 8.5>1.000 mm 9.1 14.8 10.8 0.9 1.2 2.9

0.710–1.000 mm 6.8 10.4 5.8 8.3 9.2 11.6>1.000 mm 1.5 2.4 0.8 1.1 1.5 0

hose produced in the Diosna P400, none of which were in thisange.

The Carr compressibility index demonstrated that the gran-les prepared in both sets of equipment had good flowroperties, but relative standard deviation (R.S.D.) is higher

Fig. 5. SEM pictures of granules prepared in the Diosna P400.

1018 A. Hegedus, K. Pintye-Hodi / Chemical Engineer

F

7(d

PDsfl

lsp

ca

bsoodem

ddi

4

to1taPctwttb0b

5

wta

(iiflpistamt

mmpa6r

R

tal study and qualitative considerations of internal transfers, Chem. Eng.

ig. 6. SEM pictures of granules prepared in the Collette Ultima Pro 600.

9.86–87.88 g/100 ml, R.S.D.: 5.21%) than for the R batchesbulk density: 68.49–71.43 g/100 ml, R.S.D.: 1.39%; tappedensity: 80.55–84.29 g/100 ml, R.S.D.: 1.65%).

Fig. 5 shows SEM pictures of granules prepared in the Diosna400. The particles making up the granules prepared in theiosna P400 were dense and relatively large. The spherical

tructure was retained during the fluid-bed drying process thatollowed the granulation. The surfaces of the granules displayedittle wear.

Fig. 6 presents SEM pictures of granules prepared in the Col-ette Ultima 600. These granules had a looser structure, were lesspherical and smaller, and cracked during drying. Any irregularrotrusions of the particles broke away from the granules.

In both cases, the starch particles (which have a dense texture)ould be easily differentiated from the active ingredient crystalsnd excipients.

The differences in the texture of the granules could be causedy the differing geometries of the two machines as concerns thehape and positions of the impeller and chopper blades. Becausef these factors, the materials exhibit completely different typesf motion during granulation, with the Collette Ultima 600 pro-

ucing an “undulating” effect, and the Diosna P400 granulatormploying a “folding” action. This difference can be made evenore distinct by varying the settable parameters.

ing and Processing 46 (2007) 1012–1019

The result homogeneity study (Blend Uniformity Analysis)emonstrated a mean value of 99.1 ± 0.4% for the active ingre-ient content with the relative standard deviation less than 1.5%n both case.

.4. Tablet evaluation

We pressed the granules into 500 mg tablets by using a Cour-oy R190 Ft tablet press with 36 punches. The rotational speedf the press was 65 rpm, and the main pressure applied was6 ± 1 kN. We measured the average and individual masses,he thickness and the hardness (Pharmatest WHT-2ME), the fri-bility (Pharmatest PT-TD) and the disintegration (PharmatestTZ-E) five times in the course of the tablet-formation pro-ess. We determined the R.S.D. of the weight of the individualablets by measurements of 20 tablets. The tablet parametersere satisfactory in both batches. The friability was <0.33%,

he thickness was between 4.00 and 4.19 mm, the disintegrationime was <2 min for all batches, and the average hardness wasetween 51 and 66 N. The weight variation of the tablets was.60–1.00% for the experimental batches, 1.01–1.12% for the Datches, and 0.57–0.7% for the R batches.

. Conclusion

For two high-shear granulators with different constructions,e established the ranges of parameter settings which ensure

he safe transference of the technologies for a preparation withhigh content of active ingredient.

We determined the optimal setting ranges for mass productionimpeller speed: 80–135 rpm; ideal torque associated with thempeller speed: 560–800 Nm; chopper speed: 600–2700 rpm;deal water content of the binder solution: 26 kg; liquid binderow rate: 3–12 kg/min; massing time: 2–6 min), within whicharameter ranges a satisfactory product could be manufacturedn a manner such that the drying stage took place within theame fluid-bed drying equipment. The experiment demonstratedhat, although the two technological devices perform granulationccording to similar principles of operation, their different geo-etric properties require different technical settings in order for

he end-products to have the same physical characteristics.The textures of the granules prepared in the two types of

achine differed considerably, but the differences between theeasured physical parameters were not as great. The granulation

rocess was highly controllable, the product was suitably robustnd the results were easy to reproduce in the Collette Ultima00 granulator, which allowed elimination of the inconsistenciesesulting from the use of the Diosna P 400.

eferences

[1] B. Fox, True grit: granulation & drying of delicate products, Chem. Eng.115 (2005) 35–38.

[2] C. Pere, E. Rodier, Microwave vacuum drying of porous media: experimen-

Process. 41 (2002) 427–436.[3] D. Parikh, Handbook of Pharmaceutical Granulation Technology, 81, Drugs

and Pharmaceutical Sciences, Marcel Dekker Inc., New York, 1997, pp.151–204.

gineer

[

[

[

[

[

[

[

A. Hegedus, K. Pintye-Hodi / Chemical En

[4] H. Stahl, Single-pot system for drying pharmaceutical granules, Pharm.Technol. Eur. 12 (5) (2000) 192–201.

[5] H. Stahl, Comparing different granulation techniques, Pharm. Technol. Eur.16 (11) (2004) 23–33.

[6] A. Faure, I.M. Grimsey, R.C. Rowe, P. York, M.J. Cliff, Applicability of ascale-up methodology for wet granulation processes in Collette Gral highshear mixer-granulators, Eur. J. Pharm. Sci. 8 (1999) 85–93.

[7] A. Faure, P. York, R.C. Rowe, Process control and scale-up of pharma-ceutical wet granulation process: a review, Eur. J. Pharm. Sci. 52 (2001)269–277.

[8] P.C. Knight, A. Johansen, H.G. Kristensen, T. Schaefer, J.P.K. Seville, Aninvestigation of the effect on the agglomeration of the changing the speedof a mechanical mixer, Powder Technol. 110 (2000) 204–209.

[9] K. Saleh, L. Vialatte, P. Guigon, Wet granulation in a batch high shear

mixer, Chem. Eng. Sci. 60 (2005) 3763–3775.

10] A. Devay, K. Mayer, Sz. Pal, I. Antal, Investigation on drug dissolutionand particle characteristics of pellets related to manufacturing process vari-ables of high-shear granulation, J. Biochem. Biophys. Methods 69 (2006)197–205.

[

ing and Processing 46 (2007) 1012–1019 1019

11] G. Betz, P.J. Burgin, H. Leuenberer, Power consumption measurementand temperature recording during granulation, Int. J. Pharm. 272 (2004)137–149.

12] P. Holm, T. Schaefer, H.G. Kristensen, Granulation in high-speed mixers.Part. VI. Effects of process condition on power consumption and granulegrowth, Powder Technol. 43 (1985) 225–233.

13] P. Holm, T. Schaefer, C. Larsen, End-point detection in a wet granulationprocess, Pharm. Dev. Technol. 6 (2) (2001) 181–192.

14] A. Schussele, A. Bauer-Brandl, Note on the measurement of flowabil-ity according to the European Pharmacopoeia, Int. J. Pharm. 257 (2003)301–304.

15] R.L. Carr, Evaluating flow properties of solids, Chem. Eng. 72 (1965)163–168.

16] S.I.F. Badawy, M.M. Menning, M.A. Gorko, D.L. Gilbert, Effect of process

parameters on compressibility of granulation manufactured in a high-shearmixer, Int. J. Pharm. 198 (2000) 51–61.

17] V. Kumar, M.L. Reus-Medina, D. Yang, Preparation characterization, andtabletting properties of a new cellulose-based pharmaceutical aid, Int. J.Pharm. 235 (2002) 129–140.