- Email: [email protected]
Dry Granulation
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Critical Evaluation of Root Causes of the Reduced Compactability after Roll Compaction/Dry Granulation JOHANNA MOSIG, PETER KLEINEBUDDE Institute of Pharmaceutics and Biopharmaceutics, Heinrich-Heine-University, Duesseldorf 40225, Germany Received 30 May 2014; revised 19 November 2014; accepted 1 December 2014 Published online 5 January 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.24321 ABSTRACT: The influence of lubrication and particle size on the reduced compactability after dry granulation was investigated. Powder cellulose, lactose, magnesium carbonate, and two types of microcrystalline cellulose were roll compacted, granulated, and sieved into particle fractions. Particle fractions were compressed into tablets using internal and external lubrication. Internal lubrication resulted in an overlubrication of the granule material compared with the powder material. This resulted in extraordinary high reduction of compactability after dry granulation for lubricant-sensitive materials. The granule size can cause differences in strength, whereby the degree of this effect was material dependent. The loss in strength with increasing compaction force was comparable for different particles sizes of one material, suggesting a change in material properties independently of the size. Granule hardening could be one reason as for higher compaction forces the integrity of the granule structure survived the compression step. The results demonstrated that granule lubrication mainly influence the degree of the reduced compactability after dry granulation and must be considered for the evaluation of mechanism for this phenomenon. Hardening of the material as well as size enlargement will cause the loss in strength after recompression, but the C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci influence of both depends strongly on the material. 104:1108–1118, 2015 Keywords: compression; granulation; tableting; compaction; powder technology; roll compaction; lubrication; work-hardening; reduced compactability; particle size
INTRODUCTION Tableting is a central process during drug formulation. As many materials are not suitable for direct compression (DC), most times the materials are granulated prior to tableting. Roll compaction or slugging and subsequently dry granulation are common processes for the granulation without liquids. Because of the absence of water and solvents, granulation of moisture- and heat-sensitive materials is possible.1 The major drawback of dry granulation is the loss of strength after a re-compression step. Malkowska and Khan2 showed that tablets made from dry granules exhibited a lower strength as those from DC and explained it by a work hardening of the material. They defined this as the resistance to permanent deformation, which is increasing with the amount of deformation. This is caused by increasing the level of difficulty to introduce new dislocations in the crystal structure. They observed the phenomenon for the plastically deforming microcrystalline cellulose (MCC) and starch as well as for the brittle behaving dicalcium phosphate. Reduced compactability is often related to this phenomenon in the literature. However, there is no direct evidence for a work hardening of the material as the direct verification is difficult. In more recent papers, the mean yield pressure calculated by the Heckel equation was used as surrogate parameter for the work hardening.3,4 Sun and Himmelspach5 hypothesized that granule size enlargement is the primary mechanism for the reduction of strength after re-compression. In this study, two granule sieve fractions of different MCC types were tableted with same Correspondence to: Peter Kleinebudde (Telephone: +49-211-81-14220; Fax: +49-211-81-14251; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 104, 1108–1118 (2015) C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association
1108
amounts of lubricant as for the powder material. It was stated that the reduction in strength with multiple roll compaction was just an effect of the increase in granule size but independent of the total number of compaction steps. Herting and Kleinebudde3 stated that a combination of hardening and particle-size enlargement caused the loss in compactability. Here, external lubrication was performed to ensure a comparable lubrication between the powder material and different granule fractions of MCC. The impact of the particle size on the compaction of dry granules of brittle materials was investigated by Wu and Sun.6 It was suggested that the compactability of brittle granules was insensitive to size enlargement effects as extensive fracture minimized differences in the initial granule size. In the study, the roll compaction force was low and no comparison with uncompacted material was performed. Therefore, investigations on the reduced compactability could not be made here. Patel et al.7 related the loss of compactability of dry granulated MCC to the nominal single fracture strength of granules. Higher slugging pressure led to harder granules, which caused a higher reduction in compactability after a second compression step. The impact of the particle size was also studied and it was suggested that granule hardening affected the tensile strength more than the granule size enlargement. The consequences of feedstock lubrication on the mechanical strength of tablets were investigated by He et al.4 A modest increase in dynamic hardness and mean yield pressure was found for unlubricated MCC after roll compaction, which was related to a work hardening of the material. The lubrication prior to the roll compaction step led to an overlubrication especially in the milling step, which overshadowed the effect of the work hardening. It is stated in literature that magnesium stearate adhered on the surface area and formed a film around the particles during mixing.8 This resulted in interferences in particle bonding
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Table 1.
x10 x50 x90
1109
10%, 50%, and 90% Quantiles of the Particle Size Distribution of the Starting Materials (:m) (Mean ± SD; n = 3) MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
Lactose
29.36 ± 0.18 103.37 ± 0.27 219.57 ± 0.06
24.67 ± 0.32 102.37 ± 0.98 207.95 ± 0.58
5.76 ± 0.25 31.18 ± 0.36 69.83 ± 0.30
26.55 ± 0.08 66.59 ± 0.54 134.53 ± 2.56
3.77 ± 0.03 26.93 ± 0.15 91.82 ± 0.28
during the compression leading to lower tensile strength. The sensitivity for the reduction in strength with increasing lubricant amounts depends on the compression behavior of the material. De Boer et al.9 showed that materials undergoing a complete plastic deformation were most influenced. Brittle materials were less sensitive.10 Moreover, the lubricant sensitivity could be related to the particle size of the excipient by Almaya and Aburub.11 According to this, smaller particles will be less affected by magnesium stearate. Therefore, the lubrication of the fine powder material and coarser granule particles will result in different lubricant preconditions, which can offer a large impact on the results of the comparison between granule and powdered material. Aim of this study was the critical evaluation of stated reasons for the reduced compactability after dry granulation with respect to the lubrication method. Internal and external lubrication were performed to investigate the influence of the lubrication on the phenomenon of the loss of reworkability for different plastically deforming as well as for brittle behaving materials. The impact of the particle size was examined in detail taking the impact of lubrication into account.
MATERIALS AND METHODS Materials Granulation and compression experiments were performed for five different excipients. Powder cellulose (Arbocel P290; JRS ¨ Pharma GmbH & Company, Holzmuhle, Germany); two qualities of MCC, normal (Vivapur 102; JRS Pharma GmbH & Company) and high density (Vivapur 302; JRS Pharma GmbH & Company); "-lactose monohydrate (Granulac 200; Meggle Excipients and Technology, Wasserburg, Germany) and mag¨ nesium carbonate (Magnesia 18; Magnesia GmbH, Luneburg, Germany) were used as received. Magnesium stearate (Parteck LUB; Merck Millipore, Darmstadt, Germany) was served as lubricant. Particle sizes of the starting materials are listed in Table 1. All materials were stored at least for one week at 21◦ C and 45% relative humidity (rH) for equilibration. Methods Roll Compaction/Dry Granulation
Particle Size Distribution The particle size of the starting materials was determined by laser diffraction (Helos H1402+; Sympatec, ClausthalZellerfeld, Germany) using the dry dispersing unit (Rodos; Sympatec) and a dispersing pressure of 1 bar. Starting materials were measured three times and the particle size distributions were evaluated by the instrument software. The particle size distributions of granules from each specific compaction force were determined in triplicate by digital image analysis (Camsizer XT; Retsch GmbH). The X-Jet modul was used and a dispersing pressure of 0.3 bar applied to avoid agglomeration of the particles as well as a destruction. Quantiles of the particle size distribution within the fractions were calculated using the instrument software. External Surface Area External surface area of the granule fractions between 315 and 630 :m for each material was determined by air permeatry measurements in a Friedrich manometer12 (self-construction of Evonik Industries, Darmstadt, Germany) equipped with a sample holder according to Gupte.13 For external surface area determination, approximately 100 g of material was filled in the graduated powder container. The material was tapped 1250 times within the powder container (volumetric analyzer; J. Engelsmann AG Apparatebau, Ludwigshafen, Germany) to keep the granule bed porosities comparable. Flow time was determined in triplicate for each sample preparation and each material was measured three times. The detected flow times were corrected by the blank value of the instrument (3.92 s, n = 100). External surface area was calculated according to Eqs(1). and (2) determined instrument con an experimentally using = 2.27 , the corrected flow time and listed valstant − dv h ues for the air viscosity and water density at the measuring temperature.14 SV2 =
Roll compaction/dry granulation was performed with an instrumented roll compactor (Minipactor 250/25; Gerteis Maschinen + Prozessengineering AG, Jona, Switzerland). A gap width of 2 mm and a roll speed of 3 rpm were used. Ribbons were compacted with five different specific compaction forces (2, 4, 8, 10, and 12 kN/cm) and directly dry granulated through a 1 mm sieve. Therefore, a star granulator (Gerteis Maschinen + Prozessengenineering AG), rotating 120◦ clockwise and 180◦ counterclockwise, was used and the rotor speed was set to 40 rpm clockwise and 60 rpm counterclockwise. Approximately 2 kg of granules was produced and fractionated in different DOI 10.1002/jps.24321
size classes. Portions of nearly 100 g were sieved (Retsch Vibrio AS 200 control; Retsch GmbH, Haan, Germany) for 5 min with an amplitude of 1.5 mm using meshes of 800, 630, 315, and 125 :m.
1 DmgA g3 t dv (cm−2 ) k L − h (1 − g)2 0
(1)
Sv 2 cm /g D
(2)
Sm =
SV , volume-specific surface area; k, particle shape factor; Dm , density of manometer fluid; g, gravitational constant; A, crosssectional area of the powder bed; L, length of the powder bed; V, volume of manometer fluid in one arm from starting to end position; h, height difference of fluid level in the manometer arms; g, porosity of the powder bed; t, permeation time; 0, air viscosity; Sm , mass-specific surface area; D, powder density.
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1110
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Table 2.
Added Amount of Magnesium Stearate in% for the Surface Proportional Lubrication of Dry Granules (315–630 :m)
Specific Compaction Force (kN/cm)
MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
Lactose
2 4 8 10 12
0.25 0.19 0.16 0.14 0.13
0.21 0.16 0.13 0.12 0.11
0.20 0.20 0.17 – 0.16
0.24 0.19 0.19 0.22 0.19
0.15 0.13 0.12 0.10 0.10
Lubrication Internal and external lubrication were performed. For internal lubrication, the materials were blended for 2 min with magnesium stearate in a Turbula mixer (T2C; Willy A. Bachofen AG, Basel, Switzerland) and tableted. External lubrication was achieved by manually lubricating punches and die by an eye shadow applicator. Internal lubrication was performed for the powdered materials and the granule fractions from 315 to 630 :m for each material. Granule materials were lubricated with an amount of magnesium stearate proportionally to the determined specific external surface area (2.5 :g/cm2 14 ) resulting in amounts of approximately 0.2% of magnesium stearate (Table 2). The powdered materials were mixed with 0.2% magnesium stearate as the calculated amounts for a surface proportionally lubrication were too high (0.8%–2.1%).14 External lubrication was performed for the granule fractions 630–800 and 125–315 :m as well as for the powdered material. Moreover, the granule fraction from 315 to 630 :m of MCC was lubricated externally, additionally to the internal lubrication, as well as the amount of fines (particles < 125 :m) of MCC.
granule and the corresponding DC AUC was calculated for each type of granule.2 X-ray Tomography For the analysis of the internal structure, tablets of 10 mm diameter and 350 mg mass of granules between 315 and 630 :m were produced with a compression pressure of 76 MPa. X-ray microtomography was performed with a Skyscan 1172/F instrument (Skyscan, Kontich, Belgium, control software v1.5.13). Tablets were scanned over 180◦ of rotation and 798 shadow images were acquired. Cross-sectional images were reconstructed using the Skyscan software package resulting in images of 2564 × 2564 pixels (pixel size 4.29 :m). Statistical Analysis The slopes of the AUC ratio curves were determined for each granule fraction and the different lubrication methods. Differences of the slopes were investigated using a Student t-test. For linear regressions resulting in not significant different slopes (p value > 0.01), also the differences of the intercepts were statistically investigated. p Values were calculated.
Tableting Granules as well as the raw materials were compressed into 8 mm flat-faced tablets with a mass of 200 mg. An instrumented rotary die press (PressIMA; IMA Kilian, Cologne, Germany) with a rotation speed of 10 rpm was used and compression pressures between 60 and 420 MPa were applied. For the compression of the particle fraction 315 to 630 :m of MCC with external lubrication, not all compression pressures were used. The batch of the 2 kN/cm granules was tableted with pressures between 120 and 420 MPa, granules from 8 kN/cm with 180– 420 MPa. Tablets were stored for 48 h after production at 21◦ C and 45% rH. Characterization of Tablets After storage, tablets were characterized with respect to their weight (CP 224S; Sartorius AG, G¨ottingen, Germany), height, diameter (Digimatic Caliper, Mitutoyo, Hoshima, Japan), and crushing force (TBH 210; Erweka GmbH Apparatebau, Heusenstamm, Germany). The tensile strength was calculated according to Fell and Newton.15 In this work, compactability is defined as the ability of a material to form compacts of a certain tensile strength under pressure. Some authors denominate this compaction pressure–tensile strength relationship as tabletability and to refer to compactability as compaction pressure–solid fraction relationship.16 The areas under the compactability curves were determined for each granule compression and the DC. To evaluate the degree of the reduced compactability, the ratio (AUC/AUCDC ) between the AUC of the
RESULTS AND DISCUSSION Internal versus External Lubrication A direct comparison between internal and external lubrication can be performed for MCC of the fraction 315–630 :m. A loss in strength between DC and granule compression was detectable for both lubrication methods (Figs. 1c and 1d), according to the former observed reduction in strength after dry granulation.2 The loss in strength increased with an increasing specific compaction force during roll compaction. Achieved tensile strength for tablets produced by internal lubrication was lower compared with the external lubrication. The usage of 0.2% magnesium stearate reduced the strength of the direct compressed tablets more than one third. External lubrication of the granules resulted in maximal tablet strengths in the range from 11.3 to 9 MPa, whereas with internal lubrication just 7.5– 4.4 MPa could be achieved. Moreover, the differences between the various specific compaction forces were less pronounced for the external lubrication. For the internal lubrication, tensile strength of granule tablets produced with 2 kN/cm were still close to the strength from direct compressed tablets, whereas for 10 and 12 kN/cm, a strong decrease in strength was observed. To evaluate the loss in strength over an increase in the specific compaction force during granule production in more detail, Figure 2a presented the different ratios between the AUCs of the granule compactabiliy curves and the corresponding, internal or external lubrication, DC curve. The decrease of the AUC ratio over an increase in the specific compaction
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
1111
Figure 1. Compactability curves normal density MCC, (a) <125 :m, (b) 125–315 :m, (c) 315–630 :m (internal lubrication), (d) 315–630 :m (external lubrication), and (e) 630–800 :m (mean ± SD; n = 10).
force was stronger for the internal lubrication compared with the external lubrication. This can be explained by the lubricant sensitivity of the material. Materials undergoing a plastic deformation were strongly influenced by the addition of lubricant and tensile strength of the tablets decreased much with an increasing amount of lubricant.9 It is stated in literature that magnesium stearate adhered on the surface and coated the particles.8 The generated lubricant film interfered with the particle bonding and weakened the tablets, which resulted in lower tensile strength.8,17 This film formation of magnesium stearate will be more pronounced for the granule particles if the powdered material and the granule particles were lubricated with same amounts of magnesium stearate. Therefore, the coarser granule material will be overlubricated in comparison with the fine starting material. Even if the lubricant amount was adjusted to the surface area of the granules, overDOI 10.1002/jps.24321
lubrication occurred. In a former study, it was shown that the approach of a surface proportional lubrication was not successful for the generation of a comparable lubrication and overlubrication of the materials was constituted.14 The overlubrication will decrease the strength of the granule tablets. Consequently, two effects will influence the tablet strength if an internal lubrication is used. On the one hand, the overlubrication of the coarser granule particles will lead to a decrease in compactability; on the other hand, a loss in strength due to the roll compaction step occurred. Therefore, the decrease in compactability over an increasing compaction force for the internal lubrication is a result of these two effects. Lubricant sensitivity and loss of reworkability are summed up here resulting in a stronger loss in strength compared with the external lubrication. In case of the external lubrication, just the loss in strength due to the dry granulation process was detected, which resulted
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1112
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 2. Ratios of the areas under the curve over the specific compaction forces, (a) normal density MCC, (b) high-density MCC, (c) powder cellulose, (d) magnesium carbonate, and (e) lactose.
in a lower decrease of the AUC ratio for the external lubrication. He et al.,4 who investigated the influence of lubrication during the roll compaction process of MCC, observed also an impact of magnesium stearate when it was added intragranular prior to the roll compaction process. In contrast to the unlubricated material, the loss in tensile strength for the lubricated material could not be explained by hardening of the material and was attributed to an overlubrication of the material. For other investigated materials, results according to MCC were found. Although internal and external lubrication was performed for different granule sizes, an influence of the lubrication method on the reduction in compactability was detectable for the lubricant-sensitive materials. Maximal achievable tensile strength is higher for the usage of external lubrication for the lubricant-sensitive high-density MCC and
powder cellulose, whereas for magnesium carbonate and lactose comparable tablet strength could be generated using internal and external lubrication. High-density MCC and powder cellulose offered a much stronger decrease for the AUC ratios using an internal lubrication (Figs. 2b and 2c). The decrease for granule fractions, which were compressed by external lubrication, was less pronounced. Internal lubrication decreased the AUC ratio about 0.65, whereas the external lubrication resulted in a decrease of about 0.15 for the high-density MCC. For powder cellulose, the effect is even more pronounced as internal lubrication led to decrease of about 0.85, whereas for external lubrication, the AUC ratio decreased just about 0.1. For both investigated brittle materials, magnesium carbonate and lactose, no effect of the lubrication method was detectable. The usage of internal and external lubrication resulted in comparable loss in
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
1113
Resulting from this, lubrication effects have to be considered for the compression of lubricant-sensitive materials. Not only the usage of intragranular lubricant affected the strength of dry granule tablets negatively, as it was described by He et al.4 Moreover, even the addition of extragranular magnesium stearate reduced tablet strength dramatically. The degree of the reduction in strength after roll compaction was overestimated by the usage of internal lubrication as the lubrication sensitivity of the materials was also detected. Hence, especially literature reporting the reduced compactability of dry granules of plastically deforming materials should be reconsidered, as internal lubrication will strengthen the observed phenomena. Particle Size
Figure 3. Compactability curves lactose, (a) 315–630 :m (internal lubrication) and (b) 630–800 :m (mean ± SD; n = 10).
strength for magnesium carbonate, as the AUC ratio is reduced for both lubrication methods about 0.2 (Fig. 2d). Jarosz and Parrot10 showed for the brittle behaving dicalcium phosphate that tablet tensile strength did not decrease with an increasing amount of lubricant. Because of the occurring fracture of the material, new binding positions were offered and strength of the tablets was unaffected by the lubrication method. Resulting from this, even the comparatively high amounts of magnesium stearate for the internal lubrication showed no influence on the tablet strength. The tablet strength of lactose tablets is also unaffected by the lubrication method (Fig. 2e). Moreover, roll compaction also seemed to offer no impact on the tablet strength (Fig. 3). The ratios for the AUCs were comparable for the different specific compaction forces and thus nearly straight lines parallel to the x-axis were observable with an increasing specific compaction force. With this, tablet strength was not influenced whether a dry granulation step was applied to the material or a high or low compaction forces were used for the dry granulation process. Riepma et al.18 observed for "-lactose monohydrate and roller-dried $-lactose that tablet strength and porosity of dry granule tablets was comparable to those of the slugs if same compression forces were used. Here, lactose was also unaffected of the applied slugging force and presented no loss in reworkability after dry granulation as it was also observed in this study. DOI 10.1002/jps.24321
As the lubrication can severely influence the degree of the reduction in strength (see section Internal versus External Lubrication), external lubrication was performed for the particle fractions 125–315 and 630–800 :m in order to exclude the lubricant effect. For MCC, two more fractions were available for the evaluation of the particle size effect (315–630 :m, <125 :m). As for the brittle behaving magnesium carbonate and lactose, no influence of the lubrication was observable (see Internal versus External Lubrication); for these two materials, also the internally lubricated fraction 315–630 :m can be used for investigations of the particle size effect. Tablet tensile strength decreased with increasing specific compaction force for the different granule fractions of both types of MCC (Figs. 1a, 1b, 1d, and 1e and 4a and 4c), powder cellulose (Figs. 5a and 5c), and magnesium carbonate (Figs. 6a-6c). Even the smallest particle fraction from 125 to 315 :m as well as the fines (<125 :m) of MCC offered a reduction in strength over the different specific compaction forces (Figs. 1a and 1b, 4a, 5a, and 6a). Just lactose presented no influence of the specific compaction force on the tablet strength, as it was already observed for the different lubrication methods (see Internal versus External Lubrication). The reduction in strength for the different granule fractions was clearly recognizable in the diagrams for the ratios of the AUCs (Figs. 2a–2d). Ratios decreased over an increasing specific compaction force for the smaller granules as well as for the coarser ones. For MCC, high-density MCC and powder cellulose differences in the ratio were detectable for the different granule sizes. The granule fractions with the smallest particles offered higher ratios of the areas under the curves over the range of the applied specific compaction forces. Compression of the granules from 125 to 315 :m led to strength ratios between 0.93 and 0.68 for the normal density MCC (Fig. 2a). Increasing the granule size decreased the ratio to 0.85 and 0.55 for the middle granule fraction and to 0.7 to 0.61 for the coarsest granules. Just for the highest applied specific compaction force of 12 kN/cm, the compactability of the coarser granules was higher as for the middle sized granules (ratio 0.55–0.61). For high-density MCC, ratios between 1.02 and 0.87 were achieved for the smaller granules, whereas bigger granules resulted in ratios between 0.91 and 0.76. This effect was also observable for powder cellulose. Here, the smaller sized granules offered ratios between 0.91 and 0.82, whereas for the coarser granules compactability decreased and ratios between 0.83 and 0.72 were achieved. The materials offered varying differences in strength (the ratios for the areas under the curve) between the different
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1114
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 4. Compactability curves high-density MCC, (a) 125–315 :m, (b) 315–630 :m (internal lubrication), and (c) 630–800 :m (mean ± SD; n = 10).
Figure 5. Compactability curves powder cellulose, (a) 125–315 :m, (b) 315–630 :m (internal lubrication), and (c) 630–800 :m (mean ± SD; n = 10).
particle sizes. Differences between the coarsest and finest granules were more pronounced for the normal density MCC as for powder cellulose and the high-density MCC. It is remarkable that for the granule fraction 125–315 :m of high-density MCC granule tablet strength exceeded the strength of direct compressed tablets (Fig. 4a). For the specific compaction forces of 2, 4, and 8 kN/cm, maximal tablet strength was higher as for the compression of the powdered material. Nevertheless, for compression pressures ≤ 179 MPa DC led to higher tablet strength as the granule compression. This phenomenon was also observable for the next largest granules between 315 and 630 :m (Fig. 4b). Compression of granules produced with 2 kN/cm offered higher tablet strength as the DC. Because of the overlubrication and the bigger size of the particles, the phenomenon is just observable for the lowest applied specific compaction force. Compression of the coarsest granules presented
the expected results. DC led to the highest tablet strength and a decrease over an increase in the specific compaction force is observable (Fig. 4c). For compression pressures ≥ 300 MPa, the 4 kN/cm granule tablets presented a slightly higher strength because of the shape of the curve. The compactability curves for the other granule particles offered the trend to reach a plateau, here an increase in strength is detectable even for high compression forces. Inghelbrecht and Remon19 observed for high-density MCC a decrease in tensile strength with an increase of the force during roll compaction, unfortunately no DC experiment were performed. Therefore, no evaluation was possible if here also lower roll compaction forces increased the compactability compared with the raw material. Nevertheless, if the whole compactability behavior was taken into account by using the areas under the curve instead of one single tensile strength, the increase in strength is not excessively available.
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 6. Compactability curves magnesium carbonate, (a) 125– 315 :m, (b) 315–630 :m (internal lubrication), and (c) 630–800 :m (mean ± SD; n = 10).
The total increase of the compactability over the tested pressure range was just 2% and 5%. For magnesium carbonate, the different particle sizes were not distinguishable in strength creation as the curves of the AUC ratios were crossing each other (Fig. 2d). Even if internal lubrication was used during the compression, the reduction of the ratios differed not. Wu and Sun6 investigated the influence of the particle size on compaction properties of brittle materials. For lactose, anhydrous dibasic calcium phosphate and mannitol a relatively insensitivity of compactability to dry granule size was observed. Brittle particles undergo fragmentation during the compaction step and therefore larger particles were destroyed by fracture resulting in negligible impact of the original granule size. However, the slugging force was low in all cases. DOI 10.1002/jps.24321
1115
Lactose presented for both size fractions no decrease over an increase in the specific compaction force. It can therefore be concluded that the dry granulation process did not affect the strength formation of the granule tablets (see Internal versus External Lubrication). However, differences between the particle fractions were observable. If the roll compaction has no influence on the tablet strength it can be expected that for the AUC ratios lines parallel to the x-axis should result at a value of one. Although here straight lines parallel to the x-axis were observed, none of the both was at a ratio of one. For the compression of the particles between 315 and 630 :m, the DC offered the lowest compactability (Fig. 3a) and therefore ratios of approximately 1.1 were calculated. Granules between 630 and 800 :m were compressed with external lubrication and because of this, a different DC curve has to be used for the calculation of the ratios. For this curve, slightly higher strength compared with the granule compression was achieved resulting in ratios of around 0.8 (Fig. 3b). The differences for the DC could be caused by the poor flowing properties of the material. Resulting from this, tableting, especially die filling, was problematic. As lactose is brittle behaving and therefore relative lubricant insensitive the observed differences can be attributed rather to the tableting process than to lubrication effects. Particle size effects should be as small as for magnesium carbonate because of the intensive fragmentation. Therefore, variations of the intercepts of the lines should not be over interpreted as a size enlargement effect. Sun and Himmelspach5 hypothesized that the primary mechanism for the reduced compactability of roller compacted dry granules is granule size enlargement and that multiple compaction has no effect on tablet strength if particles of the same size were compressed. However, an internal lubrication with same amounts of magnesium stearate for the MCC powder and granules of different sizes was used. It can be expected, that the film formation of the magnesium stearate was more completed for the coarser particles and interferences in particle bonding will be more pronounced. As discussed in section Internal versus External Lubrication, lubrication will mainly effect the reduction of tablet strength and the differing extent of lubrication between the different sized granules will already result in differences in the final tablet regardless of other effects. Therefore, the influence of the particle size enlargement was overestimated as the impact of the incomparable lubrication was neglected. Resulting from this, the observed huge differences in strength (>60%) will be a combination of the incomparable lubrication and the size enlargement itself. In this study, the effect of lubrication was excluded resulting in smaller effects of the size enlargement on the decrease of tablet strength. The results of this study suggested that the particle size enlargement can have an influence on the tablet strength but the observed reduced compactability after dry granulation cannot be explained in total by this phenomenon. These findings were supported by the study of Herting and Kleinebudde.3 Here, also external lubrication was performed and it was pointed out, that not solely particle size enlargement caused the reduction in strength. According to them, a combination of work hardening and size enlargement were responsible. Patel et al.7 also studied the influence of size enlargement and hardening on the compacatbility of dry granules of MCC. Here, unlubricated materials were investigated and just small effects of the granule size on the compactability were found, even for middle and high slugging pressures. Results of this study showed that also
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1116
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Table 3.
x50 —Values of the Granule Fractions (Mean ± SD; n = 3)
Material MCC
High-density MCC
Powder cellulose
Magnesium carbonate
Lactose
Specific Compaction Force (kN/cm) 2 4 8 10 12 2 4 8 10 12 2 4 8 10 12 2 4 8 10 12 2 4 8 10 12
<125 :m
125–315 :m
315–630 :m
630–800 :m
66.2 ± 0.3 64.7 ± 1.2 63.7 ± 0.9 64.5 ± 0.3 63.5 ± 0.4
187.7 ± 0.1 194.6 ± 0.7 204.7 ± 0.5 206.7 ± 0.8 210.4 ± 0.4 189.9 ± 0.2 197.9 ± 0.1 207.0 ± 0.4 209.3 ± 0.3 211.7 ± 0.4 161.1 ± 1.6 170.9 ± 0.6 182.8 ± 0.2 186.4 ± 0.3 190.0 ± 1.2 193.3 ± 2.3 206.8 ± 0.9 217.6 ± 0.3 219.6 ± 0.8 221.3 ± 0.5 174.4 ± 1.6 197.3 ± 0.8 210.6 ± 0.2 214.5 ± 0.8 216.2 ± 1.4
453.4 ± 0.8 470.9 ± 1.1 478.4 ± 2.3 479.2 ± 1.2 481.1 ± 0.6 463.3 ± 0.7 472.7 ± 1.0 477.0 ± 1.0 479.0 ± 0.9 481.0 ± 1.2 486.0 ± 4.0 488.6 ± 1.1 490.7 ± 0.3 490.9 ± 1.1 489.8 ± 0.5 478.8 ± 1.7 490.1 ± 1.3 493.1 ± 1.5 490.1 ± 0.9 493.7 ± 1.3 488.2 ± 4.0 483.4 ± 2.0 484.9 ± 1.7 485.0 ± 0.5 485.4 ± 0.5
715.2 ± 1.9 718.3 ± 0.1 717.1 ± 0.5 716.4 ± 0.2 716.9 ± 0.8 720.9 ± 1.5 720.8 ± 1.1 719.3 ± 0.3 719.8 ± 0.2 719.4 ± 0.2 711.9 ± 1.2 713.3 ± 0.3 713.4 ± 1.2 714.4 ± 1.0 715.0 ± 1.2 710.8 ± 5.5 717.8 ± 2.8 719.0 ± 1.1 720.2 ± 1.8 719.4 ± 0.6 717.6 ± 5.7 717.6 ± 0.9 719.1 ± 1.5 717.8 ± 1.9 717.4 ± 2.8
for other materials than MCC rather a combination of effects caused the reduction in strength than solely the particle size enlargement. Work Hardening Because of the use of sieve fractions, particle size variations between the batches of the different specific compaction forces should be negligible. Table 3 presented the x50 values for the different granule fractions of the materials. For the particle fraction between 125 and 315 :m, a certain increase in size was detectable for all materials. Magnesium carbonate and both types of MCC also offered a small increase in size in the fraction 315–630 :m, whereas for the other two materials as well as for the other fraction no increase in the mean particle size was observable. Resulting from this, the decreases in strength over the increase in the specific compaction forces were difficult to explain by the size enlargement. Moreover, magnesium carbonate presented a loss in strength with increasing force during roll compaction even if no effect of particle size variations was observable. Hence, other effects will be responsible for the reduction in strength. Malkowska and Khan2 stated that the materials were work hardened during the first compaction step, offering a higher resistance to deformation. As direct verification of the work hardening phenomenon is difficult, Patel et al.7 argued to describe the phenomenon as a granule hardening. According to them, it will be more useful to consider the strength of granules in the case of granule compression, as granules with higher yield strength tend to pack less efficiently. In this study, corresponding results were found. X-ray microtomography measurements of dry granule tablets (Fig. 7) showed that the integrity of the granule structure was preserved af-
ter compression for granules from higher specific compaction forces. The usage of higher specific compaction forces will result in higher granule strength, resulting in a higher resistance to deformation in the second compression step. This resistance to deformation can contribute to the reduction in strength after recompression. The reduction in strength over an increasing roll compaction force is comparable for the different particle sizes if external lubrication was used for the compression (Table 4a). This supported the assumption that a mechanism, like the above discussed granule hardening, led to the reduction in strength which is independent of the particle size. For the lubricantsensitive materials MCC and powder cellulose, the differences in the decrease of the AUC ratio were significant comparing the data of the internal and external lubrication. This underlined the observation that the lubrication effect and hardening effects were summed up (see Internal versus External Lubrication). For the brittle magnesium carbonate, slopes of the decrease differed not significantly, even if an internal lubrication was used. This confirmed the assumption that the mechanism of the reduced reworkability is unaffected by lubrication effects. Resulting from this, the slope of the decrease could be seen as a scale for the susceptibility to hardening effects of the material, which occurred independently of the granule size. Beside the slopes, the intercepts for the regressions without significantly differing slopes were compared to each other (Table 4b). Statistical analysis confirmed the observation from section Particle Size. For magnesium carbonate, intercepts differed not significantly and with this no effect of the particle size enlargement was detectable. The intercepts for both compression experiments with external lubrication of powder cellulose
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
1117
Figure 7. :-CT images of MCC tablets compressed at 76 MPa, (a) direct compression, (b) 2 kN/cm granules, and (c) 8 kN/cm granules.
Table 4. p Values of the Statistical Analysis of (a) the Differences of the Slopes of the Linear Regression of the Decrease of AUC/AUCDC ; (b) the Differences of the Intercepts of the Linear Regression of the Decrease of AUC/AUCDC for Regressions with no Significant Differences in Slopes (a)
630–800/315–630 :m 630–800/315–630 (ex) :m 630–800/125–315 :m 630–800/<125 :m 315–630/125–315 :m 315–630 (ex)/125–315 :m 315–630/<125 :m 315–630 (ex)/<125 :m 125–315/<125 :m
MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
Lactose
0.0013 0.0277 0.0011 0.0185 0.0120 0.7574 0.0026 0.0984 0.0108
0.0039 – 0.7997 – 0.0021 – – – –
0.4177 – 0.3921 – 0.1675 – – – –
0.0002 – 0.7352 – 0.0002 – – – –
0.2061 – – – – – – – –
(b)
630–800/315–630 :m 630–800/315–630 (ex) :m 630–800/125–315 :m 630–800/<125 :m 315–630/125–315 :m 315–630 (ex)/125–315 :m 315–630/<125 :m 315–630 (ex)/<125 :m 125–315/<125 :m
MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
– 0.1630 – 4 × 10−8 0.0015 0.0014 – 4 × 10−5 9 × 10−5
– – 0.0022 – – – – – –
0.1825 – 0.8538 – 0.2188 – – – –
– – 0.0005 – – – – – –
and high-density MCC presented significant differences. For the normal density MCC differences in the granule size were also significant except between the fraction the middle sized fraction and the coarsest granules. Herting and Kleinebudde3 stated that a combination of work hardening and particle size enlargement caused the loss in compactability for MCC types of different sizes. Because of the results of this study, this assumption can be confirmed for a broader range of excipients. The impact of both effects on the reduction in strength differed between the materials. The brittle behaving magnesium carbonate offered just a hardening effect, whereas for plastically deforming materials both effects contributed to the reduction in strength. Moreover, the susceptibility for the hardening effect differed between the materials. Whereas the brittle magnesium carbonate offered a comparable reduction in strength as normal density MCC over an increase in the roll compaction force, reduction in strength is less pronounced for powder cellulose and high-density MCC. Therefore, DOI 10.1002/jps.24321
high-density MCC seemed to present a superior applicability to for dry granulation processes compared to the normal density MCC.
CONCLUSIONS Lubrication mainly influences the degree of the reduction in strength after dry granulation. The usage of internal lubrication for the compression of dry granules resulted in an overestimation of the effect of the reduced compactability for lubricantsensitive materials. Here, the reduction in strength due to lubricant interferences in particle bonding and the roll compaction step were summed up. The brittle behaving materials were less affected by the lubrication and the reduction in strength occurred in the same manner using internal or external lubrication. Therefore, the common practice of internal lubrication should be carefully considered, especially for plastically deforming materials. An incomparable lubrication will
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1118
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
here dramatically distort the results for investigations of the phenomenon of the reduced compactability. Excluding the influence of the lubrication, the impact of the particle size enlargement on the reduced compcatability after dry granulation is less pronounced compared with the literature. Particle size enlargement can cause differences in tablet strength, whereby materials differed in susceptibility for this effect. For high-density MCC and powder cellulose, the effect is less pronounced compared with MCC of normal density. Tablet strength of the brittle behaving magnesium carbonate and lactose was unaffected by size enlargement effects. The loss in strength increased with increasing force during roll compaction and can be explained by a hardening of the granule particles. :-CT measurements showed that the resistance to deformation increased with increasing roll compaction force. The hardening was comparable for different granule sizes of one material but differed between the investigated materials. The brittle behaving lactose offered no loss in reworkability after dry granulation, whereas magnesium carbonate presented a reduction in strength despite the capacity to fragment. Moreover, this loss in compactability occurred in the same manner as for the normal density MCC. Remarkable was the lower susceptibility for strength reduction after the roll compaction step of the high-density MCC compared with the standard density MCC. Both effects, hardening and size enlargement, contribute to the phenomenon of the reduced compactability. The impact of both effects on the loss in strength differed between the materials presenting in total different susceptibilities for the strength reduction. Resulting from this, selection of the lubrication method as well as the material should be well considered for the production of dry granules to achieve desired tablet properties.
ACKNOWLEDGMENT The authors would like to thanks Dr. Axel Zeitler (University of Cambridge) for the performance of the :-CT analysis.
REFERENCES 1. Kleinebudde P. 2004. Roll compaction/dry granulation: Pharmaceutical applications. Eur J Pharm Biopharm 58(2):317–326. 2. Malkowska S, Khan KA. 1983. Effect of re-compression on the properties of tablets prepared by dry granulation. Drug Dev Ind Pharm 9(3):331–347.
3. Herting MG, Kleinebudde P. 2008. Studies on the reduction of tensile strength of tablets after roll compaction/dry granulation. Eur J Pharm Biopharm 70(1):372–379. 4. He X, Secreast PJ, Amidon GE. 2007. Mechanistic study of the effect of roller compaction and lubricant on tablet mechanical strength. J Pharm Sci 96(5):1342–1355. 5. Sun C, Himmelspach MW. 2006. Reduced tabletability of roller compacted granules as a result of granule size enlargement. J Pharm Sci 95(1):200–206. 6. Wu S-J, Sun C. 2007. Insensitivity of compaction properties of brittle granules to size enlargement by roller compaction. J Pharm Sci 96(5):1445–1450. 7. Patel S, Dahiya S, Calvin Sun C, Bansal AK. 2011. Understanding size enlargement and hardening of granules on tabletability of unlubricated granules prepared by dry granulation. J Pharm Sci 100(2):758– 766. 8. Strickland WA, Nelson E, Busse LW, Higuchi T. 1956. The physics of tablet compression IX. Fundamental aspects of tablet lubrication. J Am Pharm Assoc 45(1):51–55. 9. De Boer AH, Bolhuis GK, Lerk CF. 1978. Bonding characteristics by scanning electron microscopy of powders mixed with magnesium stearate. Powder Technol 20(1):75–82. 10. Jarosz PJ, Parrot EL.1984. Effect of lubricants on tensile strengths of tablets. Drug Dev Ind Pharm 10(2):259–273. 11. Almaya A, Aburub A. 2008. Effect of particle size on compaction of materials with different deformation mechanisms with and without lubricants. AAPS PharmSciTech 9(2):414–148. ¨ zur Messung der spezifischen Oberflache ¨ 12. Friedrich W. 1957. Gerat ¨ empfindlicher Guter. Chem Ingenieur Technik 29(2):104–107. ¨ 13. Gupte AR. 1976. Messung der spezifischen Oberflache grober Granulate und der mittleren Porengr¨oße von Tabletten. Acta Pharm Technol 22(3):153–168. 14. Mosig J, Kleinebudde P. 2014. Evaluation of lubrication methods: How to generate a comparable lubrication for dry granules and powder material for tableting processes. Powder Tech 266:156–166. 15. Fell JT, Newton JM. 1970. Determination of tablet strength by the diametral-compression test. J Pharm Sci 59(5):688–691. 16. Tye CK., Sun CC., Amidon GE. 2005. Evaluation of the effects of tableting speed on the relationships between compaction pressure, tablet tensile strength, and tablet solid fraction. J Pharm Sci 94:465– 472. 17. Shotton E, Lewis CJ. 1964. Some observations on the effect of lubrication on the crushing strength of tablets. J Pharm Pharmacol 16(S1):111T–120T. 18. Riepma KA, Vromans H, Zuurman K, Lerk CF. 1993. The effect of dry granulation on the consolidation and compaction of crystalline lactose. Int J Pharm 97(1–3):29–38. 19. Inghelbrecht S, Remon JP. 1998. Roller compaction and tableting of microcrystalline cellulose/drug mixtures. Int J Pharm 161(2):215– 224.
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
Critical Evaluation of Root Causes of the Reduced Compactability after Roll Compaction/Dry Granulation JOHANNA MOSIG, PETER KLEINEBUDDE Institute of Pharmaceutics and Biopharmaceutics, Heinrich-Heine-University, Duesseldorf 40225, Germany Received 30 May 2014; revised 19 November 2014; accepted 1 December 2014 Published online 5 January 2015 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jps.24321 ABSTRACT: The influence of lubrication and particle size on the reduced compactability after dry granulation was investigated. Powder cellulose, lactose, magnesium carbonate, and two types of microcrystalline cellulose were roll compacted, granulated, and sieved into particle fractions. Particle fractions were compressed into tablets using internal and external lubrication. Internal lubrication resulted in an overlubrication of the granule material compared with the powder material. This resulted in extraordinary high reduction of compactability after dry granulation for lubricant-sensitive materials. The granule size can cause differences in strength, whereby the degree of this effect was material dependent. The loss in strength with increasing compaction force was comparable for different particles sizes of one material, suggesting a change in material properties independently of the size. Granule hardening could be one reason as for higher compaction forces the integrity of the granule structure survived the compression step. The results demonstrated that granule lubrication mainly influence the degree of the reduced compactability after dry granulation and must be considered for the evaluation of mechanism for this phenomenon. Hardening of the material as well as size enlargement will cause the loss in strength after recompression, but the C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association J Pharm Sci influence of both depends strongly on the material. 104:1108–1118, 2015 Keywords: compression; granulation; tableting; compaction; powder technology; roll compaction; lubrication; work-hardening; reduced compactability; particle size
INTRODUCTION Tableting is a central process during drug formulation. As many materials are not suitable for direct compression (DC), most times the materials are granulated prior to tableting. Roll compaction or slugging and subsequently dry granulation are common processes for the granulation without liquids. Because of the absence of water and solvents, granulation of moisture- and heat-sensitive materials is possible.1 The major drawback of dry granulation is the loss of strength after a re-compression step. Malkowska and Khan2 showed that tablets made from dry granules exhibited a lower strength as those from DC and explained it by a work hardening of the material. They defined this as the resistance to permanent deformation, which is increasing with the amount of deformation. This is caused by increasing the level of difficulty to introduce new dislocations in the crystal structure. They observed the phenomenon for the plastically deforming microcrystalline cellulose (MCC) and starch as well as for the brittle behaving dicalcium phosphate. Reduced compactability is often related to this phenomenon in the literature. However, there is no direct evidence for a work hardening of the material as the direct verification is difficult. In more recent papers, the mean yield pressure calculated by the Heckel equation was used as surrogate parameter for the work hardening.3,4 Sun and Himmelspach5 hypothesized that granule size enlargement is the primary mechanism for the reduction of strength after re-compression. In this study, two granule sieve fractions of different MCC types were tableted with same Correspondence to: Peter Kleinebudde (Telephone: +49-211-81-14220; Fax: +49-211-81-14251; E-mail: [email protected]) Journal of Pharmaceutical Sciences, Vol. 104, 1108–1118 (2015) C 2015 Wiley Periodicals, Inc. and the American Pharmacists Association
1108
amounts of lubricant as for the powder material. It was stated that the reduction in strength with multiple roll compaction was just an effect of the increase in granule size but independent of the total number of compaction steps. Herting and Kleinebudde3 stated that a combination of hardening and particle-size enlargement caused the loss in compactability. Here, external lubrication was performed to ensure a comparable lubrication between the powder material and different granule fractions of MCC. The impact of the particle size on the compaction of dry granules of brittle materials was investigated by Wu and Sun.6 It was suggested that the compactability of brittle granules was insensitive to size enlargement effects as extensive fracture minimized differences in the initial granule size. In the study, the roll compaction force was low and no comparison with uncompacted material was performed. Therefore, investigations on the reduced compactability could not be made here. Patel et al.7 related the loss of compactability of dry granulated MCC to the nominal single fracture strength of granules. Higher slugging pressure led to harder granules, which caused a higher reduction in compactability after a second compression step. The impact of the particle size was also studied and it was suggested that granule hardening affected the tensile strength more than the granule size enlargement. The consequences of feedstock lubrication on the mechanical strength of tablets were investigated by He et al.4 A modest increase in dynamic hardness and mean yield pressure was found for unlubricated MCC after roll compaction, which was related to a work hardening of the material. The lubrication prior to the roll compaction step led to an overlubrication especially in the milling step, which overshadowed the effect of the work hardening. It is stated in literature that magnesium stearate adhered on the surface area and formed a film around the particles during mixing.8 This resulted in interferences in particle bonding
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Table 1.
x10 x50 x90
1109
10%, 50%, and 90% Quantiles of the Particle Size Distribution of the Starting Materials (:m) (Mean ± SD; n = 3) MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
Lactose
29.36 ± 0.18 103.37 ± 0.27 219.57 ± 0.06
24.67 ± 0.32 102.37 ± 0.98 207.95 ± 0.58
5.76 ± 0.25 31.18 ± 0.36 69.83 ± 0.30
26.55 ± 0.08 66.59 ± 0.54 134.53 ± 2.56
3.77 ± 0.03 26.93 ± 0.15 91.82 ± 0.28
during the compression leading to lower tensile strength. The sensitivity for the reduction in strength with increasing lubricant amounts depends on the compression behavior of the material. De Boer et al.9 showed that materials undergoing a complete plastic deformation were most influenced. Brittle materials were less sensitive.10 Moreover, the lubricant sensitivity could be related to the particle size of the excipient by Almaya and Aburub.11 According to this, smaller particles will be less affected by magnesium stearate. Therefore, the lubrication of the fine powder material and coarser granule particles will result in different lubricant preconditions, which can offer a large impact on the results of the comparison between granule and powdered material. Aim of this study was the critical evaluation of stated reasons for the reduced compactability after dry granulation with respect to the lubrication method. Internal and external lubrication were performed to investigate the influence of the lubrication on the phenomenon of the loss of reworkability for different plastically deforming as well as for brittle behaving materials. The impact of the particle size was examined in detail taking the impact of lubrication into account.
MATERIALS AND METHODS Materials Granulation and compression experiments were performed for five different excipients. Powder cellulose (Arbocel P290; JRS ¨ Pharma GmbH & Company, Holzmuhle, Germany); two qualities of MCC, normal (Vivapur 102; JRS Pharma GmbH & Company) and high density (Vivapur 302; JRS Pharma GmbH & Company); "-lactose monohydrate (Granulac 200; Meggle Excipients and Technology, Wasserburg, Germany) and mag¨ nesium carbonate (Magnesia 18; Magnesia GmbH, Luneburg, Germany) were used as received. Magnesium stearate (Parteck LUB; Merck Millipore, Darmstadt, Germany) was served as lubricant. Particle sizes of the starting materials are listed in Table 1. All materials were stored at least for one week at 21◦ C and 45% relative humidity (rH) for equilibration. Methods Roll Compaction/Dry Granulation
Particle Size Distribution The particle size of the starting materials was determined by laser diffraction (Helos H1402+; Sympatec, ClausthalZellerfeld, Germany) using the dry dispersing unit (Rodos; Sympatec) and a dispersing pressure of 1 bar. Starting materials were measured three times and the particle size distributions were evaluated by the instrument software. The particle size distributions of granules from each specific compaction force were determined in triplicate by digital image analysis (Camsizer XT; Retsch GmbH). The X-Jet modul was used and a dispersing pressure of 0.3 bar applied to avoid agglomeration of the particles as well as a destruction. Quantiles of the particle size distribution within the fractions were calculated using the instrument software. External Surface Area External surface area of the granule fractions between 315 and 630 :m for each material was determined by air permeatry measurements in a Friedrich manometer12 (self-construction of Evonik Industries, Darmstadt, Germany) equipped with a sample holder according to Gupte.13 For external surface area determination, approximately 100 g of material was filled in the graduated powder container. The material was tapped 1250 times within the powder container (volumetric analyzer; J. Engelsmann AG Apparatebau, Ludwigshafen, Germany) to keep the granule bed porosities comparable. Flow time was determined in triplicate for each sample preparation and each material was measured three times. The detected flow times were corrected by the blank value of the instrument (3.92 s, n = 100). External surface area was calculated according to Eqs(1). and (2) determined instrument con an experimentally using = 2.27 , the corrected flow time and listed valstant − dv h ues for the air viscosity and water density at the measuring temperature.14 SV2 =
Roll compaction/dry granulation was performed with an instrumented roll compactor (Minipactor 250/25; Gerteis Maschinen + Prozessengineering AG, Jona, Switzerland). A gap width of 2 mm and a roll speed of 3 rpm were used. Ribbons were compacted with five different specific compaction forces (2, 4, 8, 10, and 12 kN/cm) and directly dry granulated through a 1 mm sieve. Therefore, a star granulator (Gerteis Maschinen + Prozessengenineering AG), rotating 120◦ clockwise and 180◦ counterclockwise, was used and the rotor speed was set to 40 rpm clockwise and 60 rpm counterclockwise. Approximately 2 kg of granules was produced and fractionated in different DOI 10.1002/jps.24321
size classes. Portions of nearly 100 g were sieved (Retsch Vibrio AS 200 control; Retsch GmbH, Haan, Germany) for 5 min with an amplitude of 1.5 mm using meshes of 800, 630, 315, and 125 :m.
1 DmgA g3 t dv (cm−2 ) k L − h (1 − g)2 0
(1)
Sv 2 cm /g D
(2)
Sm =
SV , volume-specific surface area; k, particle shape factor; Dm , density of manometer fluid; g, gravitational constant; A, crosssectional area of the powder bed; L, length of the powder bed; V, volume of manometer fluid in one arm from starting to end position; h, height difference of fluid level in the manometer arms; g, porosity of the powder bed; t, permeation time; 0, air viscosity; Sm , mass-specific surface area; D, powder density.
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1110
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Table 2.
Added Amount of Magnesium Stearate in% for the Surface Proportional Lubrication of Dry Granules (315–630 :m)
Specific Compaction Force (kN/cm)
MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
Lactose
2 4 8 10 12
0.25 0.19 0.16 0.14 0.13
0.21 0.16 0.13 0.12 0.11
0.20 0.20 0.17 – 0.16
0.24 0.19 0.19 0.22 0.19
0.15 0.13 0.12 0.10 0.10
Lubrication Internal and external lubrication were performed. For internal lubrication, the materials were blended for 2 min with magnesium stearate in a Turbula mixer (T2C; Willy A. Bachofen AG, Basel, Switzerland) and tableted. External lubrication was achieved by manually lubricating punches and die by an eye shadow applicator. Internal lubrication was performed for the powdered materials and the granule fractions from 315 to 630 :m for each material. Granule materials were lubricated with an amount of magnesium stearate proportionally to the determined specific external surface area (2.5 :g/cm2 14 ) resulting in amounts of approximately 0.2% of magnesium stearate (Table 2). The powdered materials were mixed with 0.2% magnesium stearate as the calculated amounts for a surface proportionally lubrication were too high (0.8%–2.1%).14 External lubrication was performed for the granule fractions 630–800 and 125–315 :m as well as for the powdered material. Moreover, the granule fraction from 315 to 630 :m of MCC was lubricated externally, additionally to the internal lubrication, as well as the amount of fines (particles < 125 :m) of MCC.
granule and the corresponding DC AUC was calculated for each type of granule.2 X-ray Tomography For the analysis of the internal structure, tablets of 10 mm diameter and 350 mg mass of granules between 315 and 630 :m were produced with a compression pressure of 76 MPa. X-ray microtomography was performed with a Skyscan 1172/F instrument (Skyscan, Kontich, Belgium, control software v1.5.13). Tablets were scanned over 180◦ of rotation and 798 shadow images were acquired. Cross-sectional images were reconstructed using the Skyscan software package resulting in images of 2564 × 2564 pixels (pixel size 4.29 :m). Statistical Analysis The slopes of the AUC ratio curves were determined for each granule fraction and the different lubrication methods. Differences of the slopes were investigated using a Student t-test. For linear regressions resulting in not significant different slopes (p value > 0.01), also the differences of the intercepts were statistically investigated. p Values were calculated.
Tableting Granules as well as the raw materials were compressed into 8 mm flat-faced tablets with a mass of 200 mg. An instrumented rotary die press (PressIMA; IMA Kilian, Cologne, Germany) with a rotation speed of 10 rpm was used and compression pressures between 60 and 420 MPa were applied. For the compression of the particle fraction 315 to 630 :m of MCC with external lubrication, not all compression pressures were used. The batch of the 2 kN/cm granules was tableted with pressures between 120 and 420 MPa, granules from 8 kN/cm with 180– 420 MPa. Tablets were stored for 48 h after production at 21◦ C and 45% rH. Characterization of Tablets After storage, tablets were characterized with respect to their weight (CP 224S; Sartorius AG, G¨ottingen, Germany), height, diameter (Digimatic Caliper, Mitutoyo, Hoshima, Japan), and crushing force (TBH 210; Erweka GmbH Apparatebau, Heusenstamm, Germany). The tensile strength was calculated according to Fell and Newton.15 In this work, compactability is defined as the ability of a material to form compacts of a certain tensile strength under pressure. Some authors denominate this compaction pressure–tensile strength relationship as tabletability and to refer to compactability as compaction pressure–solid fraction relationship.16 The areas under the compactability curves were determined for each granule compression and the DC. To evaluate the degree of the reduced compactability, the ratio (AUC/AUCDC ) between the AUC of the
RESULTS AND DISCUSSION Internal versus External Lubrication A direct comparison between internal and external lubrication can be performed for MCC of the fraction 315–630 :m. A loss in strength between DC and granule compression was detectable for both lubrication methods (Figs. 1c and 1d), according to the former observed reduction in strength after dry granulation.2 The loss in strength increased with an increasing specific compaction force during roll compaction. Achieved tensile strength for tablets produced by internal lubrication was lower compared with the external lubrication. The usage of 0.2% magnesium stearate reduced the strength of the direct compressed tablets more than one third. External lubrication of the granules resulted in maximal tablet strengths in the range from 11.3 to 9 MPa, whereas with internal lubrication just 7.5– 4.4 MPa could be achieved. Moreover, the differences between the various specific compaction forces were less pronounced for the external lubrication. For the internal lubrication, tensile strength of granule tablets produced with 2 kN/cm were still close to the strength from direct compressed tablets, whereas for 10 and 12 kN/cm, a strong decrease in strength was observed. To evaluate the loss in strength over an increase in the specific compaction force during granule production in more detail, Figure 2a presented the different ratios between the AUCs of the granule compactabiliy curves and the corresponding, internal or external lubrication, DC curve. The decrease of the AUC ratio over an increase in the specific compaction
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
1111
Figure 1. Compactability curves normal density MCC, (a) <125 :m, (b) 125–315 :m, (c) 315–630 :m (internal lubrication), (d) 315–630 :m (external lubrication), and (e) 630–800 :m (mean ± SD; n = 10).
force was stronger for the internal lubrication compared with the external lubrication. This can be explained by the lubricant sensitivity of the material. Materials undergoing a plastic deformation were strongly influenced by the addition of lubricant and tensile strength of the tablets decreased much with an increasing amount of lubricant.9 It is stated in literature that magnesium stearate adhered on the surface and coated the particles.8 The generated lubricant film interfered with the particle bonding and weakened the tablets, which resulted in lower tensile strength.8,17 This film formation of magnesium stearate will be more pronounced for the granule particles if the powdered material and the granule particles were lubricated with same amounts of magnesium stearate. Therefore, the coarser granule material will be overlubricated in comparison with the fine starting material. Even if the lubricant amount was adjusted to the surface area of the granules, overDOI 10.1002/jps.24321
lubrication occurred. In a former study, it was shown that the approach of a surface proportional lubrication was not successful for the generation of a comparable lubrication and overlubrication of the materials was constituted.14 The overlubrication will decrease the strength of the granule tablets. Consequently, two effects will influence the tablet strength if an internal lubrication is used. On the one hand, the overlubrication of the coarser granule particles will lead to a decrease in compactability; on the other hand, a loss in strength due to the roll compaction step occurred. Therefore, the decrease in compactability over an increasing compaction force for the internal lubrication is a result of these two effects. Lubricant sensitivity and loss of reworkability are summed up here resulting in a stronger loss in strength compared with the external lubrication. In case of the external lubrication, just the loss in strength due to the dry granulation process was detected, which resulted
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1112
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 2. Ratios of the areas under the curve over the specific compaction forces, (a) normal density MCC, (b) high-density MCC, (c) powder cellulose, (d) magnesium carbonate, and (e) lactose.
in a lower decrease of the AUC ratio for the external lubrication. He et al.,4 who investigated the influence of lubrication during the roll compaction process of MCC, observed also an impact of magnesium stearate when it was added intragranular prior to the roll compaction process. In contrast to the unlubricated material, the loss in tensile strength for the lubricated material could not be explained by hardening of the material and was attributed to an overlubrication of the material. For other investigated materials, results according to MCC were found. Although internal and external lubrication was performed for different granule sizes, an influence of the lubrication method on the reduction in compactability was detectable for the lubricant-sensitive materials. Maximal achievable tensile strength is higher for the usage of external lubrication for the lubricant-sensitive high-density MCC and
powder cellulose, whereas for magnesium carbonate and lactose comparable tablet strength could be generated using internal and external lubrication. High-density MCC and powder cellulose offered a much stronger decrease for the AUC ratios using an internal lubrication (Figs. 2b and 2c). The decrease for granule fractions, which were compressed by external lubrication, was less pronounced. Internal lubrication decreased the AUC ratio about 0.65, whereas the external lubrication resulted in a decrease of about 0.15 for the high-density MCC. For powder cellulose, the effect is even more pronounced as internal lubrication led to decrease of about 0.85, whereas for external lubrication, the AUC ratio decreased just about 0.1. For both investigated brittle materials, magnesium carbonate and lactose, no effect of the lubrication method was detectable. The usage of internal and external lubrication resulted in comparable loss in
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
1113
Resulting from this, lubrication effects have to be considered for the compression of lubricant-sensitive materials. Not only the usage of intragranular lubricant affected the strength of dry granule tablets negatively, as it was described by He et al.4 Moreover, even the addition of extragranular magnesium stearate reduced tablet strength dramatically. The degree of the reduction in strength after roll compaction was overestimated by the usage of internal lubrication as the lubrication sensitivity of the materials was also detected. Hence, especially literature reporting the reduced compactability of dry granules of plastically deforming materials should be reconsidered, as internal lubrication will strengthen the observed phenomena. Particle Size
Figure 3. Compactability curves lactose, (a) 315–630 :m (internal lubrication) and (b) 630–800 :m (mean ± SD; n = 10).
strength for magnesium carbonate, as the AUC ratio is reduced for both lubrication methods about 0.2 (Fig. 2d). Jarosz and Parrot10 showed for the brittle behaving dicalcium phosphate that tablet tensile strength did not decrease with an increasing amount of lubricant. Because of the occurring fracture of the material, new binding positions were offered and strength of the tablets was unaffected by the lubrication method. Resulting from this, even the comparatively high amounts of magnesium stearate for the internal lubrication showed no influence on the tablet strength. The tablet strength of lactose tablets is also unaffected by the lubrication method (Fig. 2e). Moreover, roll compaction also seemed to offer no impact on the tablet strength (Fig. 3). The ratios for the AUCs were comparable for the different specific compaction forces and thus nearly straight lines parallel to the x-axis were observable with an increasing specific compaction force. With this, tablet strength was not influenced whether a dry granulation step was applied to the material or a high or low compaction forces were used for the dry granulation process. Riepma et al.18 observed for "-lactose monohydrate and roller-dried $-lactose that tablet strength and porosity of dry granule tablets was comparable to those of the slugs if same compression forces were used. Here, lactose was also unaffected of the applied slugging force and presented no loss in reworkability after dry granulation as it was also observed in this study. DOI 10.1002/jps.24321
As the lubrication can severely influence the degree of the reduction in strength (see section Internal versus External Lubrication), external lubrication was performed for the particle fractions 125–315 and 630–800 :m in order to exclude the lubricant effect. For MCC, two more fractions were available for the evaluation of the particle size effect (315–630 :m, <125 :m). As for the brittle behaving magnesium carbonate and lactose, no influence of the lubrication was observable (see Internal versus External Lubrication); for these two materials, also the internally lubricated fraction 315–630 :m can be used for investigations of the particle size effect. Tablet tensile strength decreased with increasing specific compaction force for the different granule fractions of both types of MCC (Figs. 1a, 1b, 1d, and 1e and 4a and 4c), powder cellulose (Figs. 5a and 5c), and magnesium carbonate (Figs. 6a-6c). Even the smallest particle fraction from 125 to 315 :m as well as the fines (<125 :m) of MCC offered a reduction in strength over the different specific compaction forces (Figs. 1a and 1b, 4a, 5a, and 6a). Just lactose presented no influence of the specific compaction force on the tablet strength, as it was already observed for the different lubrication methods (see Internal versus External Lubrication). The reduction in strength for the different granule fractions was clearly recognizable in the diagrams for the ratios of the AUCs (Figs. 2a–2d). Ratios decreased over an increasing specific compaction force for the smaller granules as well as for the coarser ones. For MCC, high-density MCC and powder cellulose differences in the ratio were detectable for the different granule sizes. The granule fractions with the smallest particles offered higher ratios of the areas under the curves over the range of the applied specific compaction forces. Compression of the granules from 125 to 315 :m led to strength ratios between 0.93 and 0.68 for the normal density MCC (Fig. 2a). Increasing the granule size decreased the ratio to 0.85 and 0.55 for the middle granule fraction and to 0.7 to 0.61 for the coarsest granules. Just for the highest applied specific compaction force of 12 kN/cm, the compactability of the coarser granules was higher as for the middle sized granules (ratio 0.55–0.61). For high-density MCC, ratios between 1.02 and 0.87 were achieved for the smaller granules, whereas bigger granules resulted in ratios between 0.91 and 0.76. This effect was also observable for powder cellulose. Here, the smaller sized granules offered ratios between 0.91 and 0.82, whereas for the coarser granules compactability decreased and ratios between 0.83 and 0.72 were achieved. The materials offered varying differences in strength (the ratios for the areas under the curve) between the different
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1114
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 4. Compactability curves high-density MCC, (a) 125–315 :m, (b) 315–630 :m (internal lubrication), and (c) 630–800 :m (mean ± SD; n = 10).
Figure 5. Compactability curves powder cellulose, (a) 125–315 :m, (b) 315–630 :m (internal lubrication), and (c) 630–800 :m (mean ± SD; n = 10).
particle sizes. Differences between the coarsest and finest granules were more pronounced for the normal density MCC as for powder cellulose and the high-density MCC. It is remarkable that for the granule fraction 125–315 :m of high-density MCC granule tablet strength exceeded the strength of direct compressed tablets (Fig. 4a). For the specific compaction forces of 2, 4, and 8 kN/cm, maximal tablet strength was higher as for the compression of the powdered material. Nevertheless, for compression pressures ≤ 179 MPa DC led to higher tablet strength as the granule compression. This phenomenon was also observable for the next largest granules between 315 and 630 :m (Fig. 4b). Compression of granules produced with 2 kN/cm offered higher tablet strength as the DC. Because of the overlubrication and the bigger size of the particles, the phenomenon is just observable for the lowest applied specific compaction force. Compression of the coarsest granules presented
the expected results. DC led to the highest tablet strength and a decrease over an increase in the specific compaction force is observable (Fig. 4c). For compression pressures ≥ 300 MPa, the 4 kN/cm granule tablets presented a slightly higher strength because of the shape of the curve. The compactability curves for the other granule particles offered the trend to reach a plateau, here an increase in strength is detectable even for high compression forces. Inghelbrecht and Remon19 observed for high-density MCC a decrease in tensile strength with an increase of the force during roll compaction, unfortunately no DC experiment were performed. Therefore, no evaluation was possible if here also lower roll compaction forces increased the compactability compared with the raw material. Nevertheless, if the whole compactability behavior was taken into account by using the areas under the curve instead of one single tensile strength, the increase in strength is not excessively available.
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Figure 6. Compactability curves magnesium carbonate, (a) 125– 315 :m, (b) 315–630 :m (internal lubrication), and (c) 630–800 :m (mean ± SD; n = 10).
The total increase of the compactability over the tested pressure range was just 2% and 5%. For magnesium carbonate, the different particle sizes were not distinguishable in strength creation as the curves of the AUC ratios were crossing each other (Fig. 2d). Even if internal lubrication was used during the compression, the reduction of the ratios differed not. Wu and Sun6 investigated the influence of the particle size on compaction properties of brittle materials. For lactose, anhydrous dibasic calcium phosphate and mannitol a relatively insensitivity of compactability to dry granule size was observed. Brittle particles undergo fragmentation during the compaction step and therefore larger particles were destroyed by fracture resulting in negligible impact of the original granule size. However, the slugging force was low in all cases. DOI 10.1002/jps.24321
1115
Lactose presented for both size fractions no decrease over an increase in the specific compaction force. It can therefore be concluded that the dry granulation process did not affect the strength formation of the granule tablets (see Internal versus External Lubrication). However, differences between the particle fractions were observable. If the roll compaction has no influence on the tablet strength it can be expected that for the AUC ratios lines parallel to the x-axis should result at a value of one. Although here straight lines parallel to the x-axis were observed, none of the both was at a ratio of one. For the compression of the particles between 315 and 630 :m, the DC offered the lowest compactability (Fig. 3a) and therefore ratios of approximately 1.1 were calculated. Granules between 630 and 800 :m were compressed with external lubrication and because of this, a different DC curve has to be used for the calculation of the ratios. For this curve, slightly higher strength compared with the granule compression was achieved resulting in ratios of around 0.8 (Fig. 3b). The differences for the DC could be caused by the poor flowing properties of the material. Resulting from this, tableting, especially die filling, was problematic. As lactose is brittle behaving and therefore relative lubricant insensitive the observed differences can be attributed rather to the tableting process than to lubrication effects. Particle size effects should be as small as for magnesium carbonate because of the intensive fragmentation. Therefore, variations of the intercepts of the lines should not be over interpreted as a size enlargement effect. Sun and Himmelspach5 hypothesized that the primary mechanism for the reduced compactability of roller compacted dry granules is granule size enlargement and that multiple compaction has no effect on tablet strength if particles of the same size were compressed. However, an internal lubrication with same amounts of magnesium stearate for the MCC powder and granules of different sizes was used. It can be expected, that the film formation of the magnesium stearate was more completed for the coarser particles and interferences in particle bonding will be more pronounced. As discussed in section Internal versus External Lubrication, lubrication will mainly effect the reduction of tablet strength and the differing extent of lubrication between the different sized granules will already result in differences in the final tablet regardless of other effects. Therefore, the influence of the particle size enlargement was overestimated as the impact of the incomparable lubrication was neglected. Resulting from this, the observed huge differences in strength (>60%) will be a combination of the incomparable lubrication and the size enlargement itself. In this study, the effect of lubrication was excluded resulting in smaller effects of the size enlargement on the decrease of tablet strength. The results of this study suggested that the particle size enlargement can have an influence on the tablet strength but the observed reduced compactability after dry granulation cannot be explained in total by this phenomenon. These findings were supported by the study of Herting and Kleinebudde.3 Here, also external lubrication was performed and it was pointed out, that not solely particle size enlargement caused the reduction in strength. According to them, a combination of work hardening and size enlargement were responsible. Patel et al.7 also studied the influence of size enlargement and hardening on the compacatbility of dry granules of MCC. Here, unlubricated materials were investigated and just small effects of the granule size on the compactability were found, even for middle and high slugging pressures. Results of this study showed that also
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1116
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
Table 3.
x50 —Values of the Granule Fractions (Mean ± SD; n = 3)
Material MCC
High-density MCC
Powder cellulose
Magnesium carbonate
Lactose
Specific Compaction Force (kN/cm) 2 4 8 10 12 2 4 8 10 12 2 4 8 10 12 2 4 8 10 12 2 4 8 10 12
<125 :m
125–315 :m
315–630 :m
630–800 :m
66.2 ± 0.3 64.7 ± 1.2 63.7 ± 0.9 64.5 ± 0.3 63.5 ± 0.4
187.7 ± 0.1 194.6 ± 0.7 204.7 ± 0.5 206.7 ± 0.8 210.4 ± 0.4 189.9 ± 0.2 197.9 ± 0.1 207.0 ± 0.4 209.3 ± 0.3 211.7 ± 0.4 161.1 ± 1.6 170.9 ± 0.6 182.8 ± 0.2 186.4 ± 0.3 190.0 ± 1.2 193.3 ± 2.3 206.8 ± 0.9 217.6 ± 0.3 219.6 ± 0.8 221.3 ± 0.5 174.4 ± 1.6 197.3 ± 0.8 210.6 ± 0.2 214.5 ± 0.8 216.2 ± 1.4
453.4 ± 0.8 470.9 ± 1.1 478.4 ± 2.3 479.2 ± 1.2 481.1 ± 0.6 463.3 ± 0.7 472.7 ± 1.0 477.0 ± 1.0 479.0 ± 0.9 481.0 ± 1.2 486.0 ± 4.0 488.6 ± 1.1 490.7 ± 0.3 490.9 ± 1.1 489.8 ± 0.5 478.8 ± 1.7 490.1 ± 1.3 493.1 ± 1.5 490.1 ± 0.9 493.7 ± 1.3 488.2 ± 4.0 483.4 ± 2.0 484.9 ± 1.7 485.0 ± 0.5 485.4 ± 0.5
715.2 ± 1.9 718.3 ± 0.1 717.1 ± 0.5 716.4 ± 0.2 716.9 ± 0.8 720.9 ± 1.5 720.8 ± 1.1 719.3 ± 0.3 719.8 ± 0.2 719.4 ± 0.2 711.9 ± 1.2 713.3 ± 0.3 713.4 ± 1.2 714.4 ± 1.0 715.0 ± 1.2 710.8 ± 5.5 717.8 ± 2.8 719.0 ± 1.1 720.2 ± 1.8 719.4 ± 0.6 717.6 ± 5.7 717.6 ± 0.9 719.1 ± 1.5 717.8 ± 1.9 717.4 ± 2.8
for other materials than MCC rather a combination of effects caused the reduction in strength than solely the particle size enlargement. Work Hardening Because of the use of sieve fractions, particle size variations between the batches of the different specific compaction forces should be negligible. Table 3 presented the x50 values for the different granule fractions of the materials. For the particle fraction between 125 and 315 :m, a certain increase in size was detectable for all materials. Magnesium carbonate and both types of MCC also offered a small increase in size in the fraction 315–630 :m, whereas for the other two materials as well as for the other fraction no increase in the mean particle size was observable. Resulting from this, the decreases in strength over the increase in the specific compaction forces were difficult to explain by the size enlargement. Moreover, magnesium carbonate presented a loss in strength with increasing force during roll compaction even if no effect of particle size variations was observable. Hence, other effects will be responsible for the reduction in strength. Malkowska and Khan2 stated that the materials were work hardened during the first compaction step, offering a higher resistance to deformation. As direct verification of the work hardening phenomenon is difficult, Patel et al.7 argued to describe the phenomenon as a granule hardening. According to them, it will be more useful to consider the strength of granules in the case of granule compression, as granules with higher yield strength tend to pack less efficiently. In this study, corresponding results were found. X-ray microtomography measurements of dry granule tablets (Fig. 7) showed that the integrity of the granule structure was preserved af-
ter compression for granules from higher specific compaction forces. The usage of higher specific compaction forces will result in higher granule strength, resulting in a higher resistance to deformation in the second compression step. This resistance to deformation can contribute to the reduction in strength after recompression. The reduction in strength over an increasing roll compaction force is comparable for the different particle sizes if external lubrication was used for the compression (Table 4a). This supported the assumption that a mechanism, like the above discussed granule hardening, led to the reduction in strength which is independent of the particle size. For the lubricantsensitive materials MCC and powder cellulose, the differences in the decrease of the AUC ratio were significant comparing the data of the internal and external lubrication. This underlined the observation that the lubrication effect and hardening effects were summed up (see Internal versus External Lubrication). For the brittle magnesium carbonate, slopes of the decrease differed not significantly, even if an internal lubrication was used. This confirmed the assumption that the mechanism of the reduced reworkability is unaffected by lubrication effects. Resulting from this, the slope of the decrease could be seen as a scale for the susceptibility to hardening effects of the material, which occurred independently of the granule size. Beside the slopes, the intercepts for the regressions without significantly differing slopes were compared to each other (Table 4b). Statistical analysis confirmed the observation from section Particle Size. For magnesium carbonate, intercepts differed not significantly and with this no effect of the particle size enlargement was detectable. The intercepts for both compression experiments with external lubrication of powder cellulose
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
1117
Figure 7. :-CT images of MCC tablets compressed at 76 MPa, (a) direct compression, (b) 2 kN/cm granules, and (c) 8 kN/cm granules.
Table 4. p Values of the Statistical Analysis of (a) the Differences of the Slopes of the Linear Regression of the Decrease of AUC/AUCDC ; (b) the Differences of the Intercepts of the Linear Regression of the Decrease of AUC/AUCDC for Regressions with no Significant Differences in Slopes (a)
630–800/315–630 :m 630–800/315–630 (ex) :m 630–800/125–315 :m 630–800/<125 :m 315–630/125–315 :m 315–630 (ex)/125–315 :m 315–630/<125 :m 315–630 (ex)/<125 :m 125–315/<125 :m
MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
Lactose
0.0013 0.0277 0.0011 0.0185 0.0120 0.7574 0.0026 0.0984 0.0108
0.0039 – 0.7997 – 0.0021 – – – –
0.4177 – 0.3921 – 0.1675 – – – –
0.0002 – 0.7352 – 0.0002 – – – –
0.2061 – – – – – – – –
(b)
630–800/315–630 :m 630–800/315–630 (ex) :m 630–800/125–315 :m 630–800/<125 :m 315–630/125–315 :m 315–630 (ex)/125–315 :m 315–630/<125 :m 315–630 (ex)/<125 :m 125–315/<125 :m
MCC
MCC (High Density)
Magnesium Carbonate
Powder Cellulose
– 0.1630 – 4 × 10−8 0.0015 0.0014 – 4 × 10−5 9 × 10−5
– – 0.0022 – – – – – –
0.1825 – 0.8538 – 0.2188 – – – –
– – 0.0005 – – – – – –
and high-density MCC presented significant differences. For the normal density MCC differences in the granule size were also significant except between the fraction the middle sized fraction and the coarsest granules. Herting and Kleinebudde3 stated that a combination of work hardening and particle size enlargement caused the loss in compactability for MCC types of different sizes. Because of the results of this study, this assumption can be confirmed for a broader range of excipients. The impact of both effects on the reduction in strength differed between the materials. The brittle behaving magnesium carbonate offered just a hardening effect, whereas for plastically deforming materials both effects contributed to the reduction in strength. Moreover, the susceptibility for the hardening effect differed between the materials. Whereas the brittle magnesium carbonate offered a comparable reduction in strength as normal density MCC over an increase in the roll compaction force, reduction in strength is less pronounced for powder cellulose and high-density MCC. Therefore, DOI 10.1002/jps.24321
high-density MCC seemed to present a superior applicability to for dry granulation processes compared to the normal density MCC.
CONCLUSIONS Lubrication mainly influences the degree of the reduction in strength after dry granulation. The usage of internal lubrication for the compression of dry granules resulted in an overestimation of the effect of the reduced compactability for lubricantsensitive materials. Here, the reduction in strength due to lubricant interferences in particle bonding and the roll compaction step were summed up. The brittle behaving materials were less affected by the lubrication and the reduction in strength occurred in the same manner using internal or external lubrication. Therefore, the common practice of internal lubrication should be carefully considered, especially for plastically deforming materials. An incomparable lubrication will
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
1118
RESEARCH ARTICLE – Pharmaceutics, Drug Delivery and Pharmaceutical Technology
here dramatically distort the results for investigations of the phenomenon of the reduced compactability. Excluding the influence of the lubrication, the impact of the particle size enlargement on the reduced compcatability after dry granulation is less pronounced compared with the literature. Particle size enlargement can cause differences in tablet strength, whereby materials differed in susceptibility for this effect. For high-density MCC and powder cellulose, the effect is less pronounced compared with MCC of normal density. Tablet strength of the brittle behaving magnesium carbonate and lactose was unaffected by size enlargement effects. The loss in strength increased with increasing force during roll compaction and can be explained by a hardening of the granule particles. :-CT measurements showed that the resistance to deformation increased with increasing roll compaction force. The hardening was comparable for different granule sizes of one material but differed between the investigated materials. The brittle behaving lactose offered no loss in reworkability after dry granulation, whereas magnesium carbonate presented a reduction in strength despite the capacity to fragment. Moreover, this loss in compactability occurred in the same manner as for the normal density MCC. Remarkable was the lower susceptibility for strength reduction after the roll compaction step of the high-density MCC compared with the standard density MCC. Both effects, hardening and size enlargement, contribute to the phenomenon of the reduced compactability. The impact of both effects on the loss in strength differed between the materials presenting in total different susceptibilities for the strength reduction. Resulting from this, selection of the lubrication method as well as the material should be well considered for the production of dry granules to achieve desired tablet properties.
ACKNOWLEDGMENT The authors would like to thanks Dr. Axel Zeitler (University of Cambridge) for the performance of the :-CT analysis.
REFERENCES 1. Kleinebudde P. 2004. Roll compaction/dry granulation: Pharmaceutical applications. Eur J Pharm Biopharm 58(2):317–326. 2. Malkowska S, Khan KA. 1983. Effect of re-compression on the properties of tablets prepared by dry granulation. Drug Dev Ind Pharm 9(3):331–347.
3. Herting MG, Kleinebudde P. 2008. Studies on the reduction of tensile strength of tablets after roll compaction/dry granulation. Eur J Pharm Biopharm 70(1):372–379. 4. He X, Secreast PJ, Amidon GE. 2007. Mechanistic study of the effect of roller compaction and lubricant on tablet mechanical strength. J Pharm Sci 96(5):1342–1355. 5. Sun C, Himmelspach MW. 2006. Reduced tabletability of roller compacted granules as a result of granule size enlargement. J Pharm Sci 95(1):200–206. 6. Wu S-J, Sun C. 2007. Insensitivity of compaction properties of brittle granules to size enlargement by roller compaction. J Pharm Sci 96(5):1445–1450. 7. Patel S, Dahiya S, Calvin Sun C, Bansal AK. 2011. Understanding size enlargement and hardening of granules on tabletability of unlubricated granules prepared by dry granulation. J Pharm Sci 100(2):758– 766. 8. Strickland WA, Nelson E, Busse LW, Higuchi T. 1956. The physics of tablet compression IX. Fundamental aspects of tablet lubrication. J Am Pharm Assoc 45(1):51–55. 9. De Boer AH, Bolhuis GK, Lerk CF. 1978. Bonding characteristics by scanning electron microscopy of powders mixed with magnesium stearate. Powder Technol 20(1):75–82. 10. Jarosz PJ, Parrot EL.1984. Effect of lubricants on tensile strengths of tablets. Drug Dev Ind Pharm 10(2):259–273. 11. Almaya A, Aburub A. 2008. Effect of particle size on compaction of materials with different deformation mechanisms with and without lubricants. AAPS PharmSciTech 9(2):414–148. ¨ zur Messung der spezifischen Oberflache ¨ 12. Friedrich W. 1957. Gerat ¨ empfindlicher Guter. Chem Ingenieur Technik 29(2):104–107. ¨ 13. Gupte AR. 1976. Messung der spezifischen Oberflache grober Granulate und der mittleren Porengr¨oße von Tabletten. Acta Pharm Technol 22(3):153–168. 14. Mosig J, Kleinebudde P. 2014. Evaluation of lubrication methods: How to generate a comparable lubrication for dry granules and powder material for tableting processes. Powder Tech 266:156–166. 15. Fell JT, Newton JM. 1970. Determination of tablet strength by the diametral-compression test. J Pharm Sci 59(5):688–691. 16. Tye CK., Sun CC., Amidon GE. 2005. Evaluation of the effects of tableting speed on the relationships between compaction pressure, tablet tensile strength, and tablet solid fraction. J Pharm Sci 94:465– 472. 17. Shotton E, Lewis CJ. 1964. Some observations on the effect of lubrication on the crushing strength of tablets. J Pharm Pharmacol 16(S1):111T–120T. 18. Riepma KA, Vromans H, Zuurman K, Lerk CF. 1993. The effect of dry granulation on the consolidation and compaction of crystalline lactose. Int J Pharm 97(1–3):29–38. 19. Inghelbrecht S, Remon JP. 1998. Roller compaction and tableting of microcrystalline cellulose/drug mixtures. Int J Pharm 161(2):215– 224.
Mosig and Kleinebudde, JOURNAL OF PHARMACEUTICAL SCIENCES 104:1108–1118, 2015
DOI 10.1002/jps.24321