International Journal of Pharmaceutics 244 (2002) 45–57
`
`www.elsevier.com/locate/ijpharm
`
`Crystal habit and tableting behavior
`
`Norbert Rasenack, Bernd W. Mu¨ller *
`Department of Pharmaceutics and Biopharmaceutics, Christian Albrecht Uni6ersity Kiel, Gutenbergstrasse 76,
`D-24118 Kiel, Germany
`
`Received 14 December 2001; received in revised form 7 May 2002; accepted 29 May 2002
`
`Abstract
`
`The tableting behavior of drugs can be affected by changes in the crystal habit. Different crystal habits of the
`common analgesic drugs ibuprofen and acetaminophen were prepared. Their tableting behavior was characterized. In
`the case of ibuprofen, a plate-shaped crystal was compared with the common needle-shaped form. In the case of
`acetaminophen, plate-shaped and prismatic crystals of two different particle sizes were prepared. The aim was to find
`a crystal form that is suitable for direct compression with only a low amount of excipients. This requires a substance
`that forms stable compacts at low punch forces, having a good flowability and only a low tendency to stick to the
`punches. In order to compare the tableting behavior of different substances, a comparative factor (T-factor) was
`calculated, based on typical parts of the punch force/displacement-profile and properties of the resulting compact.
`This method works with low amounts of substance and allows a rapid reproducible determination of the tableting
`behavior. The method was evaluated by characterizing different typical excipients normally used for the production
`of tablets. © 2002 Elsevier Science B.V. All rights reserved.
`
`Keywords: Tableting behavior; Preformulation studies; Crystal habit; Heckel-plot; Ibuprofen; Acetaminophen
`
`1. Introduction
`
`Tablets are the most common dosage form.
`Production should be as economical as possible.
`Especially in the case of drugs that have to be
`administered in high doses, the excipient amount
`should be kept down. The production should only
`comprise a few working steps. For example, a
`granulation step is time- and energy-intensive and
`
`* Corresponding author. Tel.: +49-431-880-1333; fax: +
`49-431-880-1352
`E-mail address: bwmueller@pharmazie.uni-kiel.de (B.W.
`Mu¨ller).
`
`exposes the formulation to water or solvent and
`heat. That is why a directly compressible powder
`is preferred which has free flowing properties, is
`able to form stable compacts at low punch forces
`and does not stick to the punches.
`By changing the crystal structure, the compres-
`sion and tableting-behavior can be affected. For
`example, acetaminophen can exist in two poly-
`morphic forms: The common crystal form is the
`thermodynamically stable form I (monoclinic)
`which leads to unstable tablets with high capping
`tendency due to a stiff construction of
`the
`molecules inside the crystal. Form II (orthorhom-
`bic) shows better compression behavior (Martino
`
`0378-5173/02/$ - see front matter © 2002 Elsevier Science B.V. All rights reserved.
`PII: S0378-5173(02)00296-X
`
`Merck Exhibit 2251, Page 1
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`46
`
`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`et al., 1996). Its physical structure contains sliding
`planes. The disadvantage of the orthorhombic
`form is the possible transition to form I.
`But not only changes of the crystal lattice can
`influence the physicochemical properties of a sub-
`stance. Isomorphic crystals can also show differ-
`ent properties due to changes in the crystal habit.
`Habit is the description of the outer appearance
`of a crystal. If only the external shape of a
`growing crystal is affected without changing the
`internal structure, a different habit results. In the
`case of tableting behavior, a free flowing powder
`can be filled homogeneously into the die. Depend-
`ing on the alignment of the crystals in the die, the
`contact area between the particles can vary.
`Therefore, an exclusive variation only of the ex-
`ternal crystal structure (the crystal habit) can
`optimize substance properties. The effect of crys-
`tal habit on tablet properties was demonstrated
`by Shell (1963). He described crystal habits by
`measurement of preferred particle orientation that
`is related to the compression characteristics of the
`powder. Optimization of tableting behavior of
`excipients was carried out by Staniforth et al.
`(1981). They examined alternative crystallization
`conditions in order to design a directly compress-
`ible mannitol and obtained a highly porous sur-
`faced mannitol by using a special crystallization
`medium. Garekani et al. (1999) compares the
`compressibility (using the Heckel-plot) of differ-
`ent crystal forms (polyhedral and thin plate-like
`crystals) of acetaminophen. The plate-like crystals
`prepared in this study had a particle size of :200
`mm, polyhedral crystals had a mean particle size
`of 100 mm. Differences in compressibility were
`found as the polyhedral crystals showed the
`higher slope in the Heckel-plot. Jbilou et al.
`(1999) examined the properties of ibuprofen. A
`directly compressible ibuprofen was developed,
`not by crystal engineering but by spherical
`agglomeration.
`Especially for drugs that have to be adminis-
`tered in high doses (which means that there is a
`high drug load in the tablet), tableting behavior of
`the pure drug plays an important role. It is known
`that most drugs can exist in different (pseudo-)
`polymorphic forms. But even isomorphic forms
`can exhibit different crystal habits. The choice of
`
`the suitable crystal form can affect the physico-
`chemical properties of the drug. It is important to
`have the drug substance in the final form fairly
`early in the development scheme (Carstensen et
`al., 1993). Accordingly,
`in early-phase-develop-
`ment the appropriate crystal has to be found by
`analyzing powder flow characteristics, dissolution
`and tableting properties, so that the biopharma-
`ceutical and manufacturing properties can be af-
`fected and consequently optimized. At this stage
`of development,
`in most cases there is only a
`small amount of drug available in high quality.
`Therefore, especially for characterizing the tablet-
`ing properties of different crystal forms, a method
`has to be used which needs only low quantities of
`materials, but which allows the development of
`reliable data. For this reason, the T-factor used in
`this study for comparing the tableting behavior
`can be used.
`For characterizing the processability of a sub-
`stance on a tableting machine, different terms
`characterizing the forming of a tablet have to be
`distinguished (Joiris et al., 1998). The compress-
`ibility describes the reduction of the volume in the
`die at applied punch force. It is characterized by
`the relation powder density versus force. Com-
`pactability describes the formation of stable com-
`pacts under
`the effect of compression.
`It
`is
`characterized by the relation stability versus den-
`sity of the resulting compact. The properties of
`the resulting compact depending on the applied
`punch force are described by the term tabletabil-
`ity. It is characterized by the relation stability
`versus force. When calculating the T-factor, the
`compressibility (represented by the factor sFmax/
`Fmax) and the tabletability (crushing strength/
`Fmax) are considered. Another important aspect in
`the tableting process is the relationship between
`elastic and plastic energy. Elastic deformation is a
`reversible phenomenon hindering the formation
`of stable tablets. Plastic deformation and brittle
`fracture are irreversible and promote tableting.
`In the literature, several methods for character-
`izing the compressibility are given (Panelli and
`Filho, 1998). The most common is the Heckel-
`plot (Heckel et al., 1961). The density of the
`powder at different applied punch forces is calcu-
`lated. A high slope in the plot (density versus
`
`Merck Exhibit 2251, Page 2
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`47
`
`pressure) shows high compressibility. However,
`the measurement by the static Heckel-plot (calcu-
`lation is based on the force-maximum) requires
`measurements at different maximal punch forces.
`Therefore, this method is not suitable for sub-
`stances that are only available in very small
`amounts. Many other equations (Table 1) have
`been proposed to support a relation between
`porosity and applied pressure.
`To compare different crystal habits of ibupro-
`fen and acetaminophen, a punch force/displace-
`ment-profile of
`the pure drug and powder
`mixtures was recorded using an instrumented sin-
`gle punch tablet machine press. By mathematical
`calculation based on characteristic data of the
`compression process and the resulting compact, a
`factor
`(T-factor)
`for comparing the tableting
`properties was obtained. In a first step,
`the
`method was evaluated by measuring different
`excipients.
`
`2. Materials
`
`Avicel® PH 102 (FMC Corp. PA), AcDiSol®
`(FMC Corp.), Elcema® G250 (Degussa, Frank-
`furt, Germany), Emcompress® (Mendell, Patter-
`son, NY), Granulatum simplex (Ph. Eur. quality),
`maize starch, mannitol, PharmDC® 93000 (all
`Cerestar, Krefeld, Germany),
`sieved lactose
`(SpheroLac®100, Meggle, Wasserburg, Germany)
`and preagglomerated lactose (Tablettose®, Meg-
`gle) were of Ph. Eur. quality. Ibuprofen and acet-
`aminophen were
`supplied
`by BASF AG
`
`(Ludwigshafen, Germany). Methanol and isopro-
`pyl alcohol were obtained from Merck KG
`(Darmstadt, Germany). Water was used in dou-
`ble-distilled quality.
`
`3. Methods
`
`3.1. Crystallization procedures
`
`Crystals that are available on the market are
`called habit
`I
`(Acetaminophen crystal
`resp.
`Ibuprofen 50, BASF). The other crystal forms
`were prepared by the following method: all crys-
`tallizations were carried out using the solvent
`change method. A double-walled glass vessel with
`thermostat was used. First, the drug was dissolved
`in the solvent. The concentration was below the
`saturation concentration to avoid remaining crys-
`tals that would affect the crystallization process.
`After precipitation by the addition of water, the
`crystals were collected by filtration under vacuum.
`They were dried in a desiccator under vacuum.
`
`3.1.1. Ibuprofen
`A total of 45 g ibuprofen was dissolved in 100
`ml of isopropyl alcohol at room temperature.
`Precipitation was carried out by solvent change
`method as described by Rasenack et al. (2001).
`
`3.1.2. Acetaminophen
`
`II).
`3.1.2.1. Large prismatic crystals (=habit
`Acetaminophen (30 g) was dissolved in 100 ml
`
`Table 1
`Most common equations for characterizing the compressibility
`
`Author
`
`y=mx+b
`
`Heckel
`
`ln
`
`1−D
`
`=kP+A
`
` 1
`
`
`
`1−D0
`
`Density versus pressure
`
`D=rel. density at pressure p
`
`Density versus pressure
`
`Pressure versus volume
`
`D=rel. density at pressure p; D0=rel.
`density without pressure
`p=pressure; Vr=spec. volume
`
`log
`
`ln
`
`1−D
`
`=klog P+c
`
`Density versus pressure
`
`D=rel. density at pressure p
`
`
` 1
`
` 1
`
` 1
`
`ln
`
`1−D
`ln p=−LVr+c
`
`=kP+ln
`
`n
`
`Konopicky
`
`Balshin
`
`Ge
`
`Merck Exhibit 2251, Page 3
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`48
`
`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`methanol at 40 °C. Precipitation was carried out
`by adding 300 ml of water (5 °C) continuously
`over 120 min under stirring conditions. During
`this process, the temperature was lowered contin-
`uously to 10 °C. The crystals were collected
`immediately.
`
`3.1.2.2. Small prismatic crystals (=habit III). A
`total of 30 g acetaminophen was dissolved in 100
`ml methanol at 40 °C. Precipitation was carried
`out by adding 300 ml of water (5 °C) continu-
`ously over 30 min under stirring conditions. Dur-
`ing
`this process
`temperature was
`lowered
`continuously to 20 °C. The crystals were collected
`immediately.
`
`3.1.2.3. Large plate-like crystals (=habit IV).
`Acetaminophen (30 g) was dissolved in 100 ml
`methanol at 40 °C. Precipitation was carried out
`by adding 300 ml of water (5 °C) over 30 s under
`stirring conditions. The resulting suspension was
`cooled down to 2 °C under stirring over 2 h to
`induce further crystallization and crystal growth.
`
`3.1.2.4. Small plate-like crystals (=habit V).
`Acetaminophen (30 g) was dissolved in 100 ml
`methanol at 40 °C. Precipitation was carried out
`by adding 300 ml of water (5 °C) over 30 s under
`stirring conditions. The crystals were collected
`immediately.
`
`3.2. Characterizing techniques
`
`3.2.1. Method for tableting
`The compressions were carried out using a
`Fette Exacta 11 (Wilhelm Fette Inc., Schwarzen-
`bek, Germany) single punch press, equipped with
`flat punches of 12 mm diameter. Data (punch
`force,
`displacement,
`time) were
`recorded
`(piezoelectric, resp. inductive).
`For measuring
`the mechanical properties
`(crushing strength) of the compacts a Pharma
`Test PTB300 (Frankfurt, Germany) was used.
`To obtain comparable data, constant true vol-
`umes were poured manually into the previously
`cleaned die. The amount of powder required was
`calculated from the true density. Every powder
`was tableted ten times. The R.S.D. of the calcu-
`lated T-factors was B5%.
`
`3.2.2. True density
`True density of excipients and drug substances
`were determined using the helium gaspycnometer
`AccuPyc 1330 (Micromeritics Instrument Corp.,
`Norcross).
`
`3.2.3. Differential scanning calorimetry (DSC)
`A differential
`scanning calorimeter
`(DSC7,
`Perkin Elmer, CT) was used. The equipment was
`calibrated using indium and zinc. Samples were
`heated at 10 °C/min in sealed aluminium pans
`under nitrogen atmosphere. The onsets of
`the melting points and enthalpies of
`fusion
`were calculated by the software (Pyris, Perkin
`Elmer).
`
`3.2.4. X-ray diffractometry
`Powder X-ray diffraction (PXRD) patterns
`were collected in transmission using an X-ray
`diffractometer (Stoe, PSD supply unit, Darm-
`stadt, Germany) with Cu Ka1
`radiation
`(monochromator: Germanium) generated at 30
`mA and 40 kV. Powder was packed into the
`rotating sample holder between two films (PETP).
`
`3.2.5. Flowability
`Flowability was quantified using avalanche
`analysis
`to
`quantify
`powder
`flowability
`(AeroFlow—TSI Modell 3250, TSI, Aachen,
`Germany). The powder sample was put in a cylin-
`drical drum that slowly rotated about its horizon-
`tal axis at a constant rate. When the incline angle
`of the powder’s surface became too great for its
`molecular structure to support, the powder col-
`lapsed down toward the bottom. This event is
`referred to as an ‘avalanche’. The time interval
`between avalanches and the amplitude of
`the
`avalanche is recorded. Before measurement, the
`powder was disagglomerated through a sieve
`(710). Some 60 ml of each powder were used and
`measurement was carried out over 300 s with 1
`UpM. Factors characterizing the flowability are
`the mean time between avalanches, the scatter and
`the maximum time. A high mean and high max
`show cohesivity; irregular flow characteristics re-
`sult in a high scatter.
`
`Merck Exhibit 2251, Page 4
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`49
`
`Table 2
`Readings used for the calculation
`
`Factor
`
`Plastic
`deformation
`Total energy
`Crushing
`strength Fc
`Fmax
`sFmax (up)
`
`e
`
`Unit
`
`%
`
`Nm
`N
`
`kN
`mm
`
`cm3
`
`Percentage of the plastic
`energy
`Plastic and elastic energy
`Stability of the compact
`
`Upper punch force (maximum)
`Displacement of upper punch
`in Fmax
`Volume (true) tableted
`
`3.2.6. Scanning electron microscopy (SEM) and
`microscopy
`Electron-micrographs of crystals were obtained
`using a scanning electron microscope (Philips XL
`20, Philips, Eindhoven, Netherlands). Samples
`were fixed on an aluminium stub with conductive
`double-sided adhesive tape (Leit-Tabs, Plannet
`GmbH, Wetzlar, Germany) and coated with gold
`in an argon atmosphere (50 Pa) at 50 mA for 50
`s (Sputter Coater, Bal-Tec AG, Liechtenstein).
`Photographs of the big acetaminophen crystals
`(habit II and IV) were obtained using a micro-
`scope (Zeiss, Heidenheim, Germany).
`
`3.2.7. Method for comparing the tableting
`properties
`Due to the fact that most of the common ways
`only analyze the compressibility and do not look
`at the resulting compacts (=tabletability), here
`another way for comparing the tableting proper-
`ties was evaluated. For determination of
`the
`tableting properties (especially in view of suitabil-
`ity for direct compression) of a drug, this is an
`important aspect. For these reasons, a method
`was developed which includes the quality of the
`resulting compacts (inserted in the equation as
`crushing strength). Together with other proper-
`ties, such as flowability or sticking to the punches,
`a complete evaluation of a drug/powder mixture
`can be achieved.
`The T-factor [J×cm4] is calculated from char-
`acteristic data (Table 2) using Eq. (1). A high
`value shows good tableting behavior. The equa-
`tion is suitable for comparing different powders at
`
`the same machine parameters: the values of the
`most important parameters do not seem to be
`transferable when using different adjustments. So
`for comparability, the same experimental set-ups
`(same adjustment of upper and lower punch, same
`true volume of each substance compressed) are
`required.
`
`T=plastic def×energytotal ×
`
`Fc
`Fmax
`
`×sFmax(up)×e
`(1)
`
`The relation between elastic and plastic defor-
`mation is important for the compaction behavior
`of any substance. Beside this, the absolute value
`for the plastic energy (in relation to the maximum
`of punch force) is important. The factor (plastic
`deformation [%]×total energy [Nm]) character-
`izes the effectiveness of the compression process.
`The factor (sFmax (up) [mm]/Fmax [kN]) character-
`izes the compressibility (analogous to the Heckel-
`plot): a high displacement of the upper punch at a
`low maximal force shows good compressibility
`(corresponding to a high slope in Heckel-plot).
`The factor (Fc[N]/Fmax [kN]) represents the prop-
`erties of the resulting compact (characterizing the
`tabletability). The adjustment of the tableting ma-
`chine and the compressed powder-volume (true
`volume) must be the same in one run—small
`deviations in weight are corrected by the factor e.
`If the die is overfilled, the maximum punch force
`increases and thus the plastic deformation de-
`creases. This would effect a wrongly lowered T-
`factor. Therefore, the volume e is inserted in the
`equation for correction of small deviations in
`compressed volume.
`
`4. Results and discussion
`
`4.1. Comparison of different excipients
`
`The excipients were chosen to represent a vari-
`ety of deformation characteristics—microcrys-
`(Avicel®) and pregelatinized
`talline
`cellulose
`starch (PharmDC® 93000) exhibit high plastic de-
`formation. Even at low punch forces, they are
`able to form stable tablets. Dicalciumphosphate
`(Emcompress®), sieved lactose (SpheroLac® 100)
`
`Merck Exhibit 2251, Page 5
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`50
`
`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`and preagglomerated lactose (Tablettose®) are
`brittle materials. Tablettose®
`shows
`better
`tabletability than sieved lactose. The method em-
`ployed for calculating the T-factor was able to
`show this difference (T-factor 0.35 for sieved and
`for
`preagglomerated
`T=0.50
`lactose, P=
`0.0012). Granulatum simplex
`(lactose/potato
`
`starch/PVP) is an example of a granulate. Fig.
`1(b) shows the calculated T-factors for these ex-
`cipients and some drugs. Characteristic force/dis-
`placement-profiles
`of
`excipients
`and
`acet-
`aminophen are presented in Fig. 1(a). Same true
`volumes of each substance were tableted with
`same machine adjustments.
`
`Fig. 1. (a) Force/displacement-profiles of excipients and the drug acetaminophen for comparison. Same true volumes of each
`substance were tableted with same machine adjustments. (b) Tableting properties of excipients and drugs.
`
`Merck Exhibit 2251, Page 6
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`51
`
`Fig. 2. SEM photographs of ibuprofen. (a) Habit I; (b) habit II.
`
`Table 3
`DSC-data
`
`Onset of Tm
`(°C) (9S.D.)
`DHf (J/g)
`(9S.D.)
`
`(a) DSC-data ibuprofen
`
`(b) DSC-data acetaminophen
`
`Needle-shaped
`habit
`
`Plate-shaped
`habit
`
`Habit I
`
`Habit II
`
`Habit III
`
`Habit IV
`
`Habit V
`
`75.290.1
`
`75.190.1
`
`169.590.1
`
`169.390.1
`
`169.490.1
`
`169.790.1
`
`169.290.1
`
`125.893.3
`
`126.192.9
`
`177.794.3
`
`173.795.6
`
`175.793.9
`
`176.195.4
`
`172.793.7
`
`Results: mean of three measurements.
`
`Merck Exhibit 2251, Page 7
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`52
`
`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`Fig. 3. SEM and microscope pictures of acetaminophen. (a) Habit I; (b) habit II; (c) habit III; (d) habit IV; (e) habit V.
`
`Merck Exhibit 2251, Page 8
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`53
`
`4.2. Influencing of tableting properties by crystal
`habit: ibuprofen and acetaminophen
`
`The analgesic drugs ibuprofen and acetamino-
`phen show bad manufacturing behavior: ibupro-
`fen is a cohesive and adhesive powder, the flow
`properties are poor and it has a high tendency to
`stick to the punches. The compacts are mechani-
`
`cally unstable and consequently, for the produc-
`tion of tablets, a high amount of bonding agents
`and a granulation step are necessary. Tabletability
`of acetaminophen is much worse than that of
`ibuprofen, as confirmed by the T-factor (Fig. 1).
`Therefore, both drugs mostly have to be granu-
`lated before tableting.
`In this study, two habits of ibuprofen and five
`
`Fig. 4. Powder flow characteristics of ibuprofen. (a) Habit I; (b) habit II.
`
`Table 4
`Flowability
`
`(a) Flowability of ibuprofen
`
`(b) Flowability of acetaminophen
`
`Needle-shaped habit
`
`Plate-shaped habit
`
`Habit I
`
`Habit III
`
`Habit V
`
`Mean (s)
`Scatter (s)
`Max (s)
`
`4.9
`2.3
`12.0
`
`3.1
`1.0
`5.0
`
`3.9
`1.1
`6.2
`
`3.7
`0.8
`5.0
`
`3.2
`0.8
`5.0
`
`Fig. 5. Tableting properties of ibuprofen.
`
`Merck Exhibit 2251, Page 9
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`54
`
`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`Fig. 6. Force/displacement-profiles of different crystal forms of acetaminophen. Same true volumes of each substance were tableted
`with same machine adjustments. (b) Tableting properties of acetaminophen.
`
`different crystal forms of acetaminophen were
`compared (Figs. 2 and 3). These crystals were
`determined as isomorphs by DSC (Table 3) and
`powder X-ray diffraction.
`Properties that affect the handling and manu-
`facturing processes of these crystals vary dramati-
`cally. It was observed that different crystal forms
`have different flow characteristics (different adhe-
`sion/cohesion behavior). Because of the especially
`
`bad flowability of ibuprofen-commodity (habit I)
`differences in flow-characteristics are particularly
`remarkable. In Table 4, the parameters of ibupro-
`fen are presented and in Fig. 4, the profile of
`measured avalanches is shown. It is clear that the
`prepared crystals show a regular flowability. The
`flow characteristic of the common crystal form is
`irregular which represents cohesive behavior. Dif-
`ferences in flowability of acetaminophen are not
`
`Merck Exhibit 2251, Page 10
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`55
`
`so dramatic because of the better flowability of
`the common crystals of this drug (Table 4).
`Differences in tableting properties of the pure
`drug between the different habits are shown in
`Fig. 5 and Fig. 6(b). In the case of ibuprofen, the
`free flowing habit II (thin plates) shows the best
`compaction behavior. The properties of both
`compacts are compared in Table 5. The difference
`in mechanical strength of the compact of pure
`
`drug at comparable punch forces in particular is
`remarkable. Because of the better flowability, the
`die is filled very homogeneously and the contact
`between the crystals is intensive. So,
`input of
`energy into the powder is much more effective.
`The total energy is higher and the elastic deforma-
`tion lower. In the case of acetaminophen, small
`plates (habit V) or prismatic crystals (habit III)
`are preferable
`(force/displacement-profiles are
`
`Fig. 7. SEM photographs of the ibuprofen tablet surface. (a) Habit I; (b) habit II.
`
`Fig. 8. Ibuprofen-tablets (drug content 91%). (b) Acetaminophen tablets.
`
`Merck Exhibit 2251, Page 11
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`56
`
`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`Table 5
`Comparison of ibuprofen compacts: characteristic data
`
`Needle-shaped
`habit
`
`Plate-shaped habit
`
`Flowability
`
`Cohesive
`
`Sticking to the
`punches
`Elastic recovery
`(%)
`Fmax (kN)
`Total energy (J)
`Crushing
`strength (N)
`
`Adhesive
`
`24
`
`16
`4.2
`32
`
`Free flowing, not
`cohesive
`Not adhesive
`
`19
`
`15
`4.9
`55
`
`Same true volumes of pure ibuprofen were tableted ten times
`with same machine adjustments.
`
`shown in Fig. 6(a) and calculated T-factors in
`Fig. 6(b)).
`To test the suitability for direct compression, a
`powder with 91% drug content with 4% Avicel®
`PH102, 4% AcDiSol®, 0.5% Aerosil® and 0.5%
`magnesium stearate was tableted. In the case of
`habit I, the flowability was lower which resulted
`in a higher S.D. (1.1%) of the tablet mass (S.D.=
`0.4% in the case of ibuprofen habit II). The
`T-factors of the powder mixtures are shown in
`Fig. 8(a). Because of the good tableting behavior
`of the excipients, the T-factors are higher than
`those of the pure drug. Differences between the
`two crystal forms are still present in the powder
`mixture and are detected by the used method.
`During the tableting process in the case of habit I,
`there was substance sticking to the punches and
`the surface of the tablets was damaged. In the
`case of the plate-like habit, the surface of the
`tablets was even and there was no substance
`sticking to the punches. Pictures of the ibuprofen
`tablets are shown in Fig. 7. Holes can be detected
`in tablets prepared from the common crystal form
`(Fig. 7a).
`To compare the tableting behavior of powder
`mixtures with excipients and acetaminophen, the
`drug was mixed with excipients (80% drug con-
`tent: 15.5% Avicel® PH102, 4% AcDiSol® and
`0.5% magnesium stearate; 60% drug content: 34%
`Avicel® PH102, 5.5% AcDiSol® and 0.5% magne-
`sium stearate) and analyzed. The T-factor shows
`
`higher values now because of the good com-
`paction behavior of the excipients. However, in
`Fig. 8(b) it can be seen that the differences be-
`tween the crystals
`still exist
`in the powder
`mixture.
`
`5. Conclusions
`
`Differences in crystal modification affect the
`properties of a drug even if, as in isomorphic
`drugs, there are only differences in the crystal
`habit. Tableting behavior, flowability and the ten-
`dency to stick to the punches can be affected by
`the choice of an optimal crystallization method—
`influencing the crystal habit of the drug. Even for
`drugs with bad compaction behavior, a habit can
`be developed that shows better tableting proper-
`ties. In the case of ibuprofen, a directly compress-
`ible form was found (possible drug content 90%).
`Also, in the case of acetaminophen, differences in
`tableting properties were observed.
`The importance of the choice of the optimal
`crystal form was demonstrated. This has to be
`carried out in the first steps of drug formulation
`development. Determination of the tabletability
`using the employed calculation allows a rapid,
`economic and reproducible comparison of pure
`drugs or powder-mixtures. Thus, the development
`of a suitable crystal form, which ideally is directly
`compressible, can be achieved.
`
`Acknowledgements
`
`We would like to thank BASF AG for financial
`support and for the supply of ibuprofen and
`acetaminophen.
`
`References
`
`Carstensen, J.T., Ertell, C., Geoffroy, J.-M., 1993. Physico-
`chemical properties of particulate matter. Drug Dev. Ind.
`Pharm. 19, 195–219.
`Garekani, H.A., Ford, J.L., Rubinstein, M.H., Rajabi-Siah-
`boomi, A.R., 1999. Formation and compression character-
`istics of prismatic polyhedral and thin plate-like crystals of
`paracetamol. Int. J. Pharm. 187, 77–89.
`
`Merck Exhibit 2251, Page 12
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`

`

`N. Rasenack, B.W. Mu¨ller /International Journal of Pharmaceutics 244 (2002) 45–57
`
`57
`
`Heckel, R.W., 1961. Density-pressure relationships in powder
`compaction. Trans. Metall. Soc. AIME 221, 671–675.
`Jbilou, M., Ettabia, A., Guyot-Hermann, A.-M., Guyot, J.C.,
`1999. Ibuprofen agglomerates preparation by phase separa-
`tion. Drug Dev. Ind. Pharm. 25, 297–305.
`Joiris, E., Martino, P., Berneron, C., Guyot-Herman, A.-M.,
`Guyot, J.-C., 1998. Compression behavior of orthorhombic
`paracetamol. Pharm. Res. 15, 1122–1130.
`Martino, P., Guyot-Hermann, A.-M., Conflant, P., Drache, M.,
`Guyot, J.-C., 1996. A new pure paracetamol for direct com-
`pression: the orthorhombic form. Int. J. Pharm. 128, 1–8.
`
`Panelli, R., Filho, F.A., 1998. Compaction equation and its use
`to describe powder consolidation behavior. Powd. Metal. 41,
`131–133.
`Shell, J., 1963. X-ray and crystallographic applications in
`pharmaceutical systems. III: Crystal habit quantitation. J.
`Pharm. Sci. 52, 100–101.
`Staniforth, J.N., Rees, J.E., Kayes, J.B., Priest, R.C., Cotterill,
`N.J., 1981. The design of a direct compression tablet
`excipient. Drug Dev. Ind. Pharm. 7, 179–190.
`Rasenack, N., Mu¨ller, B.W., Einig, H., 2001. German Patent
`Application No. 0050/53009.
`
`Merck Exhibit 2251, Page 13
`Mylan Pharmaceuticals Inc. v. Merck Sharp & Dohme Corp.
`IPR2020-00040
`
`