Anbalagan P

A STUDY ON COMPACTION PUNCH DESIGN
ON TABLET PROPERTIES
PARTHIBAN ANBALAGAN
B. Sc. (Pharm.) (Hons.)
A THESIS SUBMITTED FOR THE DEGREE OF

DOCTOR OF PHILOSOPHY
DEPARTMENT OF PHARMACY
NATIONAL UNIVERSITY OF SINGAPORE
2017
Supervisors:
Dr. Celine Valeria Liew, Main Supervisor
Associate Professor Paul Heng Wan Sia, Co-Supervisor
Examiners:
Dr. Ng Wai Kiong, Institute of Chemical & Engineering Sciences
Professor Kunikazu Moribe, Chiba University
Dr. Effendi Widjaja, Johnson & Johnson
DECLARATION
I hereby declare that the thesis is my original work and it has been written by me in
its entirety. I have duly acknowledged all the sources of information which have
been used in the thesis.
The thesis has also not been submitted for any degree in any university previously.
________________________
Parthiban Anbalagan
21 August 2017
ii
ACKNOWLEDGEMENTS
I would like to express my heartfelt gratitude to my supervisors, Dr. Celine Valeria
Liew and Associate Professor Paul Heng Wan Sia, for their guidance, support and
encouragement throughout the candidature. I am also thankful for the teachings and
guidance from Associate Professor Chan Lai Wah and Associate Professor T. R. R.
Kurup.
I would like to thank the Department of Pharmacy, Faculty of Science and the
National University of Singapore for providing me with the research scholarship and
administrative support in these four years. My special appreciation goes to Ms.
Teresa Ang and Ms. Wong Mei Yin for their technical assistance. I would also like
to thank Ms. Vivian Chin Hui Min for her valuable contribution.
I would also like to acknowledge all members in GEA-NUS PPRL, whom I have
had the pleasure of working alongside during the course of my candidature. This
dissertation could not have been completed without your great support.
Finally, I wish to thank my family and friends for their constant support and
encouragement throughout these years.
With my sincere appreciation,
Parthiban Anbalagan
2017
iii
TABLE OF CONTENTS
DECLARATION ................................................................................................... ii
ACKNOWLEDGEMENTS ................................................................................. iii
TABLE OF CONTENTS ......................................................................................iv
SUMMARY ............................................................................................................ix
LIST OF TABLES .................................................................................................xi
LIST OF FIGURES .............................................................................................xiv
LIST OF SYMBOLS AND ABBREVIATIONS ...............................................xix
1. INTRODUCTION............................................................................................... 2
1.1 Pharmaceutical tablet manufacture ................................................................. 2
1.1.1 Tablet compaction process ....................................................................... 2
1.1.2 Particle deformation mechanism .............................................................. 3
1.1.3 Characterization of particle deformation ................................................. 5
1.1.3.1 Heckel analysis ................................................................................. 5
1.1.3.2 Stress relaxation analysis .................................................................. 6
1.1.3.3 Deformation parameters derived from compaction profiles ............. 8
1.1.3.3.1 Compaction force-time profile ................................................... 8
1.1.3.3.2 Compact deformation during high-speed tableting .................. 10
1.1.3.3.3 Compaction force-punch separation profile ............................. 11
1.1.3.4 Compaction parameters................................................................... 13
1.1.4 Instrumentation in tablet compaction ..................................................... 14
1.1.4.1 Rotary tablet press........................................................................... 14
1.1.4.1.1 Turret speed ............................................................................. 14
1.1.4.1.2 Compaction rolls ...................................................................... 15
1.1.4.1.3 Precompaction.......................................................................... 15
1.1.4.1.4 Press technology ...................................................................... 15
1.1.4.2 Compaction simulator ..................................................................... 16
1.1.4.3 Single station tablet press ................................................................ 17
1.2 Tablet physical quality .................................................................................. 17
1.2.1 Tablet tensile strength ............................................................................ 17
1.2.2 Tablet failure tendency .......................................................................... 19
1.2.3 Tablet material adherence tendency ....................................................... 20
iv
1.2.4 Factors influencing tablet physical quality ............................................ 21
1.2.4.1 Deformation characteristics of tableting materials ......................... 21
1.2.4.2 Lubrication...................................................................................... 22
1.2.4.3 Compaction force............................................................................ 23
1.2.4.4 Tableting speed ............................................................................... 24
1.2.4.5 Tablet press and tooling technology ............................................... 25
1.3 Tablet compaction tooling ............................................................................ 26
1.3.1 General concepts .................................................................................... 26
1.3.2 Punch head configuration ...................................................................... 28
1.3.3 Punch face configuration ....................................................................... 30
1.4 Computational simulation of powder compaction ........................................ 34
1.4.1 Finite element modeling (FEM) ............................................................ 34
1.4.2 Discrete element Modeling (DEM) ....................................................... 36
1.5 Research gaps ............................................................................................... 38
1.5.1 Dwell effect on tablet quality ................................................................ 39
1.5.2 Punch head configurational modification .............................................. 40
1.5.3 Punch face configurational modification ............................................... 41
2. HYPOTHESES AND OBJECTIVES ............................................................. 44
3. EXPERIMENTAL............................................................................................ 47
3.1 Part 1: Investigation of the influence of deformation mechanism of tableting
materials during dwell, and the consequent effect on tablet mechanical
properties ............................................................................................................ 47
3.1.1 Rationale of study .................................................................................. 47
3.1.2 Choice of tableting materials ................................................................. 47
3.1.3 Characterization of tableting materials .................................................. 48
3.1.3.1 Heckel analysis ............................................................................... 48
3.1.3.2 Particle size analysis ....................................................................... 49
3.1.3.3 Hausner ratio................................................................................... 50
3.1.3.4 Angle of repose ............................................................................... 50
3.1.4 Preparation of tableting blends .............................................................. 51
3.1.5 Tablet production ................................................................................... 51
3.1.6 Stress relaxation test .............................................................................. 51
3.1.7 Tablet characterization........................................................................... 52
3.1.7.1 Viscoelastic strain recovery ............................................................ 52
3.1.7.2 Tablet tensile strength ..................................................................... 53
v
3.1.8 Statistical analysis .................................................................................. 53
3.2 Part 2. Investigation of the effect of punch head modification on compaction
parameters and tablet physical quality ................................................................ 54
3.2.4 Characterization of tableting material .................................................... 55
3.2.4.2 Hausner ratio ................................................................................... 55
3.2.4.3 Angle of repose ............................................................................... 55
3.2.4.4 Heckel analysis ............................................................................... 56
3.2.4.5 Stress relaxation test........................................................................ 56
3.2.5.1 Tableting punch head configuration ............................................... 56
3.2.5.2 Rotary tableting ............................................................................... 58
3.2.6 Compaction parameters.......................................................................... 60
3.2.7 Evaluation of compact deformation ....................................................... 61
3.2.8 Tablet characterization ........................................................................... 62
3.2.8.1 Tablet elastic recovery .................................................................... 62
3.2.8.2 Tablet porosity ................................................................................ 62
3.2.8.4 Capping index ................................................................................. 63
3.2.9 Computational simulation of tablet compaction .................................... 63
3.2.10 Statistical analysis ................................................................................ 66
3.3 Part 3. Investigation of the effect of punch face modification on compaction
parameters and tablet physical quality ................................................................ 67
3.3.3.1 Tableting punch face configurations ............................................... 67
3.3.3.2 Rotary tablet production.................................................................. 69
3.3.4 Compaction energy ................................................................................ 71
3.3.5 Tablet characterization ........................................................................... 72
3.3.5.1 Tablet elastic recovery .................................................................... 72
3.3.5.2 Tablet porosity ................................................................................ 72
vi
3.3.5.4 Capping index ................................................................................. 72
3.3.6 Computational simulation of tablet compaction .................................... 72
3.3.6.1 Compact stress analysis .................................................................. 73
3.4 Part 4. Investigation of the effect of punch face modification on powder
adhesion tendency to the punch surface during powder compaction ................. 75
3.4.4 Characterization of tableting blends ...................................................... 76
3.4.4.2 Angle of repose measurement......................................................... 76
3.4.5.1 Punch face configuration ................................................................ 76
3.4.5.2 Tablet production using manual press ............................................ 76
3.4.5.3 Tablet production using compaction simulator .............................. 78
3.4.6 Assessment of powder adherence .......................................................... 79
3.4.6.1 Quantification of powder adherence ............................................... 79
3.4.6.2 Force parameters related to powder adherence............................... 80
4. RESULTS AND DISCUSSION ....................................................................... 83
materials during dwell and the consequent effect on tablet mechanical properties
............................................................................................................................ 83
4.1.1 Physical characteristics of tableting materials ....................................... 83
4.1.2 Compression behavior of tableting materials ........................................ 83
4.1.3 Compact deformation during dwell ....................................................... 85
4.1.4 Effect of formulation variables on tablet properties .............................. 92
4.1.5 Effect of dwell on tablet properties...................................................... 100
4.1.6 Summary .............................................................................................. 102
parameters and tablet physical quality .............................................................. 104
4.2.1 Physical characteristics of tableting materials ..................................... 104
4.2.2 Compression behavior of tableting materials ...................................... 104
4.2.3 Influence of punch head configuration on compaction profile ............ 106
vii
4.2.4. Influence of compaction roll configuration on compaction profile .... 110
4.2.5 Influence of punch head configuration on tablet properties................. 114
4.2.6 Finite element analysis ......................................................................... 122
4.2.7 Influence of compaction roll configuration on tablet properties .......... 126
4.2.8 Effect of tableting speed ...................................................................... 132
4.2.9 Summary .............................................................................................. 134
parameters and tablet physical quality .............................................................. 136
4.3.1 Influence of punch face edge geometry on compaction parameters .... 136
4.3.2 Influence of punch face edge geometry on tablet properties ............... 139
4.3.3 Discrete Element analysis .................................................................... 144
4.3.3.1 Comparison between simulations and experiments ...................... 144
4.3.3.2 Effect of punch face edge geometry on tablet stress distribution . 147
4.3.4 Effect of tableting speed ...................................................................... 151
4.3.5 Effect of precompaction ....................................................................... 153
4.3.6 Effect of roll displacement ................................................................... 156
4.3.7 Summary .............................................................................................. 159
adhesion tendency to the punch surface during powder compaction ................ 162
4.4.1 Physical characteristics of tableting material ....................................... 162
4.4.2 Powder adhesion model ....................................................................... 162
4.4.3 Influence of punch face configuration ................................................. 163
4.4.4 Influence of compaction force on powder adhesion ............................ 167
4.4.5 Influence of magnesium stearate concentration ................................... 169
4.4.6 Influence of compaction method (compaction simulator..................... 171
4.4.7 Influence of punch face configuration (compaction simulator) ........... 172
4.4.8 Influence of compaction force (compaction simulator) ....................... 174
4.4.9 Ejection force as sticking parameter .................................................... 175
4.4.10 Summary ............................................................................................ 177
5. CONCLUSION ............................................................................................... 180
6. REFERENCES................................................................................................ 185
7. LIST OF PUBLICATIONS AND PRESENTATIONS ............................... 195
viii
SUMMARY
Tablet compaction punches have undergone many changes to their design
specifications over the years in an effort to improve manufacturing efficiency, tablet
quality and durability of the tools. This study was initiated with the objective of
understanding the impact of developing new punch designs by modification of
punch head and face configurations, with regard to tablet physical quality.
In the first part of the work, the influence of different deformation mechanisms of
tableting materials during the dwell phase and the consequent effect on tablet
mechanical properties were investigated. Compact deformation during dwell was
observed to adopt distinct phases which were dependent on the degree of plasticity
and porosity in the compact system. Dwell time extension was found to be beneficial
when preparing tablets containing components of high plasticity by forming
mechanically strong tablets. However, the effect of dwell could possibly be
influenced by time-dependent secondary factors.
The specific impact of punch head configuration on the compaction process and
tablet mechanical properties was subsequently studied in a rotary press. Punches
with modified head flat and head radius specifications were used for high speed
tableting. Extension of head flat diameter increased dwell time and this led to
significant improvements in physical quality of tablets prepared. Dwell time
prolongation enabled greater plastic flow to occur, consequently reduced tablet
elasticity and formed more structural bonds. For punches without a head flat, a
change in the head radius was found to influence the compaction process marginally
and it led to insignificant differences in tablet properties.
The influence of punch face modification on the tableting process was next studied
and tableting performance under different compaction conditions were determined.
Flat-face punches with different face edge geometries (bevel or radius edge
ix
geometry) were used for tablet production in a rotary press. Modification of the
straight bevel edge to a curved edge enabled deeper punch penetration in the die
cavity during the compaction cycle and facilitated greater compact densification.
The presence of a bevel edge led to greater stress anisotropy in the tablet
microstructure and contributed to development of higher elastic energy within the
compact. Application of a precompaction cycle along with dwell time extension to
the tableting process amplified the tableting performance of radius edge punch face
configuration to a greater extent than the bevel edged punch.
In the final part of the study, the influence of punch face configurational
modification on tablet adhesion tendency was studied. The flat-face punches with
radius or bevel edge face configurations and the plain flat-face punches were used
for tablet compaction. Punches possessing face configurations with a concavity
(radius and bevel edge faces) were shown to be advantageous over the flat-face plain
face configuration as the presence of an edge cup reduced residual powder adherence
significantly. Powder adhesion data between the edged punch face configurations
showed similar degrees of material sticking, with only marginal differences,
suggesting equivalent tableting performance in terms of tablet adhesion tendency.
This research study on punch tools provided a comprehensive understanding on the
advantages and limitations of punch head and face configurational changes with
respect to tablet properties. The findings may aid tablet press and tools
manufacturers in process and tool design considerations for a successful tableting
program.
x
LIST OF TABLES
Table 1. Specifications of FFBE and FFRE punch face configurations (FFBE:

flat-face bevel edge, FFRE: flat-face radius edge) ............................... 68
Table 2. Physical characteristics of microcrystalline cellulose (MCC), dicalcium

phosphate (DCP) and paracetamol (APAP) particles ........................... 83
Table 3. Viscoelastic parameters obtained for APAP/MCC compacts of different

proportions (APAP/MCC; F1: 100/0, F2: 75/25, F3: 50/50, F4: 25/75,
F5: 0/100) ............................................................................................. 87
Table 4. Viscoelastic parameters obtained for DCP/MCC compacts of different

proportions (DCP/MCC; F6: 100/0, F7: 75/25, F8: 50/50, F9: 25/75) . 88
Table 5. Viscoelastic parameters obtained for APAP/DCP compacts of different

proportions (APAP/DCP; F10: 75/25, F11: 50/50, F12: 25/75) ........... 91
Table 6. Effect of dwell time on stress decay and mechanical properties of

APAP/MCC tablets of different proportions, compacted at 6 kN
(APAP/MCC; F1: 100/0, F2: 75/25, F3: 50/50, F4: 25/75, F5: 0/100) 93

APAP/MCC tablets of different proportions, compacted at 9 kN
(APAP/MCC; F1: 100/0, F2: 75/25, F3: 50/50, F4: 25/75, F5: 0/100) 94

DCP/MCC tablets of different proportions, compacted at 6 kN
(DCP/MCC; F6: 100/0, F7: 75/25, F8: 50/50, F9: 25/75) .................... 95

DCP/MCC tablets of different proportions, compacted at 9 kN
(DCP/MCC; F6: 100/0, F7: 75/25, F8: 50/50, F9: 25/75) .................... 96
APAP/DCP tablets of different proportions, compacted at 6 kN
(APAP/DCP; F10: 75/25, F11: 50/50, F12: 25/75) .............................. 98
APAP/DCP tablets of different proportions, compacted at 9 kN
(APAP/DCP; F10: 75/25, F11: 50/50, F12: 25/75) .............................. 99
xi
Table 12. Physical characteristics of paracetamol-starch and lactose granules .. 104
Table 13. Effect of punch head configuration on compaction parameters obtained

using the larger compaction roll (D240) ............................................. 107

using the smaller compaction roll (D150)........................................... 111
Table 15. Effect of punch head configuration on lactose tablet mechanical

properties at the experimental compaction forces for compaction roll of
diameter 240 mm (EHF: extended head flat, SHF: standard head flat,
RHF: reduced head flat, SRH: standard radius head, RRH: reduced radius
head).................................................................................................... 118
Table 16. Effect of punch head configuration on paracetamol-starch tablet

mechanical properties at the experimental compaction forces for
compaction roll of diameter 240 mm (EHF: extended head flat, SHF:
standard head flat, RHF: reduced head flat, SRH: standard radius head,
RRH: reduced radius head) ................................................................. 121
Table 17. Effect of punch head configuration on lactose tablet mechanical

properties at the experimental compaction forces for compaction roll of
diameter 150 mm (EHF: extended head flat, SHF: standard head flat,
RHF: reduced head flat, SRH: standard radius head, RRH: reduced radius
head).................................................................................................... 128

mechanical properties at the experimental compaction forces for
compaction roll of diameter 150 mm (EHF: extended head flat, SHF:
Table 19. Effect of punch head configuration on compaction parameters of lactose

tablets compacted at turret speed 35 rpm (47 m/min) (EHF: extended
head flat, SHF: standard head flat, RHF: reduced head flat) .............. 133
Table 20. Effect of punch head configuration on mechanical properties of lactose

tablets compacted at turret speed 35 rpm (47 m/min) (EHF: extended
head flat, SHF: standard head flat, RHF: reduced head flat) .............. 133
Table 21. Effect of punch face edge geometry on set compaction thickness at the
main compaction event ....................................................................... 137
xii
Table 22. Properties of lactose tablets prepared using single compaction cycle by
FFBE and FFRE punch face configurations ....................................... 140
Table 23. Properties of paracetamol-starch tablets prepared using single

compaction cycle by FFBE and FFRE punch face configurations ..... 143
FFBE and FFRE punch face configurations at turret speed 35 rpm (47
m/min) ................................................................................................ 152
Table 25. Properties of paracetamol-starch tablets prepared using double

compaction (mode1) cycle by FFBE and FFRE punch face configurations
............................................................................................................ 154
Table 26. Properties of paracetamol-starch tablets prepared using double

compaction (mode2) cycle by FFBE and FFRE punch face configurations
............................................................................................................ 158
xiii
LIST OF FIGURES
Figure 1. Stress relaxation curve showing stress decay in a compact under constant
strain........................................................................................................ 7
Figure 2. Schematic diagrams depicting the motion of punch head in relation to

compaction roll (A) and the graphical representation of compaction
force-time profile with the associated compaction parameters (B). ....... 9
Figure 3. Graphical representation of compaction force-punch separation profile

.............................................................................................................. 12
Figure 4. Model for vertical displacement of a punch head on a rotary press...... 12
Figure 5. Schematic showing the diametrical compression test. .......................... 18
Figure 6. Image showing absence of capping and tablet capping (left to right) after
hardness testing. .................................................................................... 19
Figure 7. Schematics showing the lower (left) and upper (right) punches. .......... 27
Figure 8. Schematic showing the punch head configuration with the associated
head flat and head radius....................................................................... 28
Figure 9. Schematics showing the FFP (A) FFBE (B) and FFRE (C) tablet designs
with difference in tablet edges shown. (FFP: flat-face plain, FFBE: flat-
face bevel edge, FFRE: flat-face radius edge) ...................................... 32
Figure 10. Schematics showing the dimensions (in mm) of the EHF (A), SHF (B),
RHF (C), SRH (D) and RRH (E) punch head configurations in terms of
their head flat (HF), head radius (HR) and head thickness (HT)
specifications. (EHF: extended head flat, SHF: standard head flat, RHF:
reduced head flat, SRH: standard radius head, RRH: reduced radius head)
.............................................................................................................. 57
Figure 11. Schematic showing the rotary press turret with positions of the die filling,
precompaction, main compaction and ejection stages. ......................... 59
Figure 12. Schematic diagrams depicting the (i) motion of the punch at
precompaction event and (ii) motion of the punch at main compaction
xiv
event (A). Graphical representation of the force-time curve at the main
compaction event, with precompaction deactivated (B). Graphical
representation of force-time curve at precompaction event with main
compaction deactivated (C). ................................................................. 59
Figure 13. Drucker-Prager Cap model with associated parameters.

.............................................................................................................. 64
Figure 14. Schematic showing the uniaxial compression test for material property
calibration. ............................................................................................ 64
Figure 15. Schematics showing the dimensions (in mm) of the FFBE (A) and FFRE
(B) punch face configurations. (FFBE: flat-face bevel edge, FFRE: flat-
face radius edge) ................................................................................... 68
Figure 16. Schematic diagrams depicting the motion of the punch at precompaction
(PC) event (i) with air compensator deactivated (mode1) and (ii) motion
of the punch at main compaction (MC) event (A). Graphical
representation of force-time curve at precompaction event with air
compensator deactivated (mode1) and the force time-time curve at the
main compaction event (B)................................................................... 70
event (i) with air compensator activated (mode2) and (ii) motion of the
punch at main compaction event (A). Graphical representation of force-
time curve at precompaction event with air compensator activated
(mode2) with the associated roll displacement curve and the force-time
curve at the main compaction event (B). .............................................. 70
Figure 18. Manual press used for tablet production.

.............................................................................................................. 77
Figure 19. Tooling setup (A) and the force feeder with the take-off force sensor (B)
in the compaction simulator.................................................................. 78
Figure 20. Schematics showing the stress factors associated with tablet ejection (A)
and take-off (B). (Fe: ejection force, Ft: take-off force, Fa: adhesive force,
Fr: compact-die wall force) ................................................................... 81
Figure 21. Heckel plots for microcrystalline cellulose (♦), dicalcium phosphate (■)
and paracetamol (●) particles ............................................................... 84
xv
Figure 22. Stress relaxation profiles of APAP/MCC compacts of different
proportions. (APAP/MCC; F1: 100/0, F2: 75/25, F3: 50/50, F4: 25/75,
F5: 0/100).............................................................................................. 87
Figure 23. Stress relaxation profiles of DCP/MCC compacts of different

proportions. (DCP/MCC; F6: 100/0, F7: 75/25, F8: 50/50, F9: 25/75) 88
Figure 24. Stress relaxation profiles of APAP/DCP compacts of different

proportions. (APAP/DCP; F10: 75/25, F11: 50/50, F12: 25/75) .......... 91
Figure 25. Heckel plots (A) and stress relaxation curves (B) for lactose (  ) and
paracetamol-starch (  ) granules. ...................................................... 105
Figure 26. Comparison of compaction profiles (6 and 9 kN) using the larger
compaction roll (D240) between SHF and RHF (A), SRH and RRH (B)
head configurations. (SHF: standard head flat, RHF: reduced head flat,
SRH: standard radius head, RRH: reduced radius head) .................... 108
Figure 27. Comparison of compaction profiles (6 and 9 kN) between compaction

rolls of diameters 240 (─) and 150 (---) mm using the SHF (A) and EHF
(B) head configurations. (EHF: extended head flat, SHF: standard head
flat) ...................................................................................................... 112
Figure 28. Effect of punch head configuration on lactose tablet tensile strength (A)
and area quotient (B) at the experimental compaction forces for
compaction roll of diameter 240 mm. (EHF: extended head flat, SHF:
Figure 29. Effect of punch head configuration on paracetamol-starch tablet tensile

strength (A) and area quotient (B) at the experimental compaction forces
for compaction roll of diameter 240 mm. (EHF: extended head flat, SHF:
Figure 30. Stress distribution contours of simulated paracetamol-starch tablets

compacted at dwell times representative of RRH (A), SRH (B), RHF (C),
SHF (D) and EHF (E) punch head configurations. ............................. 124
and area quotient (B) at the experimental compaction forces for
xvi
compaction roll of diameter 150 mm. (EHF: extended head flat, SHF:

strength (A) and area quotient (B) at the experimental compaction forces
for compaction roll of diameter 150 mm. (EHF: extended head flat, SHF:
Figure 33. Tensile strength of lactose tablets prepared using single compaction
cycle by FFBE (■) and FFRE (□) punch face configurations. ........... 140
Figure 34. Compaction force-punch separation curves for the FFBE (●) and FFRE
(○) punch face configurations. ............................................................ 141
Figure 35. Tensile strength of paracetamol-starch tablets prepared using single

compaction cycle by FFBE (■) and FFRE (□) punch face configurations.
............................................................................................................ 143
Figure 36. Comparison between simulation and experimental tablet thickness (A)
and RFA values (B) for FFP (●), FFRE (♦) and FFBE (▲) tablets. .. 145
Figure 37. (i) Compaction of particles with grid bin group in the vertical Y-Z cross-
section obtained for FFP (A), FFRE (B) and FFBE (C) tablets. Fig. 37
(ii-iii) Stress distribution in the vertical Y-Z cross-sections obtained for
FFP (A), FFRE (B) and FFBE (C) tablets at compaction force (ii) 8.0 and
(iii) 9.0 kN. The X and Y axes of the contour plots represent the distance
(mm) from the edge and bottom of the tablet respectively. ................ 148
Figure 38. Roll displacement-time curves for the FFBE (—) and FFRE (---) punch
face configurations with the activation of the air compensator (mode2)
during double compaction cycle. ........................................................ 157
Figure 39. FFP punch face showing first-degree (A) and second-degree (B) powder
adhesion. ............................................................................................. 163
Figure 40. Effect of punch face configuration on powder adherence on the upper (■)
and lower (□) punch faces after compacting 30 (A) and 50 (B) tablets
using an ibuprofen-lactose formulation containing 1 %, w/w magnesium
stearate. ............................................................................................... 164
xvii
Figure 41. Distribution of residual powder on flat-face bevel (A) and radius (B)
edge punch faces. ................................................................................ 167
and lower (□) punch faces after compacting 30 (A) and 50 (B) tablets
using an ibuprofen-lactose formulation containing 1.5 %, w/w
magnesium stearate. ............................................................................ 170
Figure 43. Effect of punch face configuration on powder adherence on the upper
punch face after tablet production in the compaction simulator, using an
ibuprofen-lactose formulation containing 1 %, w/w magnesium stearate.
............................................................................................................ 171
Figure 44. Take-off force profile during the time course of tablet production using
the compaction simulator for the FFP (○), FFRE (●) and FFBE (▲)
punch face geometries at compaction force 5.0 kN (A), 7.5 kN (B) and
10 kN (C) using an ibuprofen-lactose formulation containing 1 %, w/w
Figure 45. Ejection force profile during the time course of tablet production using
the compaction simulator for the FFP (○), FFRE (●) and FFBE (▲)
punch face geometries at compaction force 5.0 kN (A), 7.5 kN (B) and
10 kN (C) using an ibuprofen-lactose formulation containing 1 %, w/w
xviii
LIST OF SYMBOLS AND ABBREVIATIONS
AOR Angle of repose
APAP Paracetamol
API Active pharmaceutical ingredient
AQ Area quotient
a Volumetric plastic strain
CSmax Force at the beginning of the dwell phase
CS0 Force at the end of the dwell phase
Cp Compactibility coefficient
DCP Dicalcium phosphate
DEM Discrete element modeling
D150 Compaction roll of diameter 150 mm
D240 Compaction roll of diameter 240 mm
dt Tablet diameter
D Cohesion parameter
dx Particle size corresponding to Xth volume percentile

under the cumulative undersize distribution curve
EHF Extended head flat
xix
ER Tablet elastic recovery
EU European standards
F Formulation
F Tablet fracture force
FEM Finite element modeling
FFP Flat-face plain
FFBE Flat-face bevel edge
FFRE Flat-face radius edge
Ht Tablet thickness
HF Head flat
HT Head thickness
HR Head radius
MCC Microcrystalline cellulose
Np Number of tablets that capped off the press
Nh Number of tablets that capped during hardness testing
Nt Total number of tablets
Pb Cap surface parameter
Ptapped Tapped density
xx
Pbulk Bulk density
R Material parameter
RFA Rate of force application
RHF Reduced head flat
RRH Reduced radius head
rpm Revolutions per minute
RQ Relaxation quotient
SHF Standard head flat
SRH Standard radius head
STmain Set main compaction thickness
STpre Set precompaction thickness
Tcon Consolidation time
Tdwell Dwell time
Tdec Decompression time
Ttotal Total compaction time
VER Tablet viscoelastic recovery
VS Viscoelastic slope
Wnet Net work of compaction
xxi
Wexp Work of expansion
σz Axial stresses
σr Radial stresses
f
σd Radial tensile strength
f
σc Axial tensile strength
β Friction angle
ɛ Compact porosity
xxii
CHAPTER 1
INTRODUCTION
1
1. INTRODUCTION
1.1 Pharmaceutical tablet manufacture
The tablet dosage form is the most common form of solid medicament that have
been used in medicine for many years. Tablet manufacturing process has evolved
over time, with the first recorded hand-operated tablet press patented by William
Brockedon in 1843 [1]. Tablet production rate greatly increased with the invention
of the automated tablet press by John Wyeth and Bros in the late 1870s [2]. Since
then, tablet manufacturing process has evolved with advances in press and tooling
technology.
1.1.1 Tablet compaction process
A tablet is formed by reducing the volume of powder particles confined in a die into
a consolidated solid body with a pair of punches. Powder compaction process
comprises a series of sequential phases - compression, consolidation and
decompression - in a particulate solid-gas system as a result of an applied force [3,
4]. Compression involves a reduction in bulk volume caused by particle
rearrangement at relatively low pressures. Compact porosity reduces when particles
occupy the voids between them thus forming a closely packed structure. At a certain
point, the interparticulate friction and particulate packing characteristics, e.g. shape
and surface roughness, will restrict any further particle rearrangement. Application
of pressure beyond this stage results in deformation of the particles, new particle
surfaces more interparticulate contact points are created [5]. Compression processes
next lead to the consolidation phase during which bonds are formed at the contact
points. The consolidation phase represents the stage when an increase in mechanical
strength of a compact is obtained when the compaction pressure is further increased.
It is generally accepted that interparticulate forces, such as van der Waals forces and
2
hydrogen bonding, are the major bonding mechanisms [6]. Other types of bonding
mechanisms include mechanical interlocking and formation of solid bridges between
particles. Decompression occurs when the punches start to move away from each
other after reaching minimum distance apart. As the displacement between the
punches increases, the compaction pressure reduces quickly. During this phase,
some of the elastic strain present under pressure will recover with the compact still
within the confines of the die. This event consequently results in partial disruption
of the interparticulate bonds formed during the consolidation phase and brings about
a reduction in the relative density of the compact. The tablet compaction process is
therefore greatly influenced by the deformation mechanisms of the materials to be
compacted.
1.1.2 Particle deformation mechanism
Application of an external force to the powder bed results in the force being
transmitted through interparticulate points of contact leading to local deformation of
particles. Three main types of particle deformation mechanisms - elastic, plastic and
brittle fracture deformation - have been identified by researchers. Plastic
deformation and fragmentation are irreversible processes that result in permanent
change in the particle shape/size upon removal of the compaction load, whereas
elastic deformation is a reversible process with deformed particles recovering their
original shape after removal of compaction load. As a consequence of these
processes, new particle surfaces are created and interparticulate contact points
increase, leading to the formation of interparticulate bonds. Plastic deformation and
fragmentation of particles are the two processes that control the evolution of the
tablet microstructure in terms of interparticulate bonding. Examples of materials that
consolidate predominantly by plastic deformation include microcrystalline cellulose
3
and sodium chloride, and materials that undergo brittle fracture include dicalcium
phosphate and some classes of crystalline lactose [7-11]. Nonetheless, all materials
experience some degree of elastic, plastic and fragmenting characteristics. However,
usually one type of volume reduction mechanism will predominate for
pharmaceutical materials, dependent on factors such as compaction force and rate.
Some materials undergo deformation processes that are time dependent and occur at
various rates during the compaction sequence, considering that the compact mass in
the die is never in a stress/strain equilibrium during the actual compaction cycle. The
extent of compact formation for such materials is determined by its time dependent
viscoelastic behavior. Increasing the speed of the compaction process may have a
marked effect on powder compactibility, causing mechanical failure of the resultant
tablets for a formulation that produces acceptable tablets at a slower machine speed.
The rate at which the compaction load is applied and removed, and the duration
during which the compact strain is sustained have been shown to be critical factors
when viscoelastic materials are involved. During tableting, compacts usually
undergo a period of static strain known as dwell phase. Particles under compressive
forces during the dwell phase would undergo further plastic deformation and reduce
the interstitial spaces within the compact by better packing, causing a reduction in
porosity and an increase in interparticulate bonding [12]. These realignment
processes allow the internal structure of the tablet to concurrently undergo stress
relief, as characterized by a reduction in the applied force due to time dependent
plastic deformation of the materials under compression. The extent of compact
relaxation has been reported to be largely influenced by compact porosity and
amount of stored energy in the compact [13, 14].
4
1.1.3 Characterization of particle deformation
The study of powder compaction mechanism is an important part in the development
of tablets. Characterization of compaction mechanism is greatly facilitated by the
availability of equipment such as instrumented tablet presses and universal testing
machines that are fitted with force sensors and also, preferably, displacement
measurement probes.
1.1.3.1 Heckel analysis
Heckel analysis is a popular method used in pharmaceutical research to determine
the basic volume reduction mechanism during compaction at high compaction
pressures. It assumes that powder compaction adopts first-order kinetics with the
interparticulate spaces as the reactants and powder densification as the product [15-
17]. The Heckel equation (Eq. (1)) is expressed as:
1
ln = 𝑘P + A (1)
ɛ
where k and A are constants, and porosity (ɛ) of the compacts can be calculated from
the relative density of the compact at applied pressure (P) and the true density of the
material. The reciprocal of k gives the yield pressure, a material dependent constant.
The yield pressure is inversely related to the material resistance to plastic flow; high
values indicate onset of plastic deformation at higher compaction pressures. The
main utility of Heckel plots arises from their ability to identify the predominant
deformation behavior of the materials, particularly to distinguish fragmenting and
plastically deforming materials. Heckel plots allow the classification of powder
compaction behavior based on the progression of the Heckel curves [18, 19]. A
linear relationship with increasing applied pressure indicates deformation
mechanism dominated by plastic deformation [20]. An initial curved region
5
followed by a straight line indicates particles fragmentation at the early stages of the
compaction process with brittle fracture preceding plastic flow [11]. The use of
Heckel analysis for deformation characterization is debatable, attributed to the high
variability in the results from different studies largely due to employment of different
experimental conditions [21]. Variables that affect Heckel plots include compaction
kinematics (rate and duration of compaction), degree of lubrication and
configuration of tools used for compact formation. Even the method of compact
density measurement (in-die or out-of-die) has been reported to significantly impact
the Heckel plots, particularly when elastic materials are compacted. These variables
should be taken into consideration during Heckel analysis. Nonetheless, the widely
used Heckel analysis remains a valuable tool in the field of pharmaceutical
compaction studies.
1.1.3.2 Stress relaxation analysis
While the Heckel analysis evaluates compact deformation at high applied pressures,
it is not suitable for studying compact deformation during dwell phase. Stress
relaxation during the dwell phase has been used by several researchers to evaluate
the viscoelastic behavior of tableting material [22]. In a stress relaxation test,
compacts are formed at a specified force level, and the position of the punches at
maximal stated force is paused for a stipulated period of time. The reduction in the
applied force with time, which occurs due to plastic deformation, is monitored [23].
A typical stress relaxation curve is shown in Fig. 1. The amount of plastic flow
during dwell can be quantified by determining the total amount of compaction force
decay or by evaluating the viscoelastic slope from the linear portions of the stress
relaxation curves [24-26].
6
The stress relaxation test has been adopted by several authors to analyze and
differentiate particle deformation under constant strain. Augsburger et al.
demonstrated the extent of plastic flow during the dwell phase in some commonly
used directly compressible fillers and related the viscoelastic behavior to a Maxwell
model [24]. While compaction force was shown to decay with time for all fillers, a
greater total amount of compaction force decay and a more accelerated rate of decay
was found for microcrystalline cellulose and compressible starch than for lactose
and dicalcium phosphate. This was attributed to the predominantly plastically
deforming nature of the former excipients. Similar results were reported in a study
by Rees and co-workers [27]. It was reported that brittle fracturing materials showed
minimal stress relaxation during the period of constant strain, and a large total stress
relaxation was observed for materials such as pregelatinized starch and cellulose
which are relatively more ductile in nature. The contribution of lubrication to the
viscoelastic behavior of tableting materials on compact deformation during dwell
phase was further investigated by Ebba et al [28]. It was shown that the type of
lubricant used influenced the extent of stress relaxation during dwell phase, with
granules lubricated with magnesium stearate exhibiting the highest stress relaxation
followed by talc and glyceryl palmitostearate.

Compaction force
Time
Figure 1. Stress relaxation curve showing stress decay in a compact under constant
strain.
7
1.1.3.3 Deformation parameters derived from compaction profiles
While the Heckel and stress relaxation curves allow the overall deformation and
viscoelastic character of various tableting materials to be quantified, it may not
reflect the behavior of the compacts during high speed tableting, in which the extent
of time-dependent deformation processes are often different. Compaction conditions
during Heckel and stress relaxation tests are often not representative of the rapid
compaction kinematics encountered during high speed tableting. Hence the
extrapolation of data on particulate deformation from reported studies obtained
under various compaction conditions may not be reliable. Deformation parameters
derived from compaction profiles can be alternatively used to characterize compact
deformation during high speed tableting. Two main types of compaction profiles
(force-time and force-punch separation) have been used extensively in high speed
powder compaction studies.
1.1.3.3.1 Compaction force-time profile
Compaction force-time profiles can possibly be utilized to study dwell phase
deformation during high speed tableting. Fig. 2 shows the movement of the tableting
punch head and the associated compaction force-time profile with the associated
compaction parameters - consolidation (Tcon), dwell (Tdwell) and decompression (Tdec)
times. Due to the importance of the dwell period in relation to viscoelastic
deformation within the compact, it is necessary that the concept of dwell phase is
understood in greater detail. Dwell time is defined in terms of tableting punch
displacement or compaction force profile. Tableting punches experience vertical and
linear motion upon contact with a compaction roll in a rotary press, with maximum
penetration achieved in the die at peak displacement as shown in Fig. 2A, the second
position of the punch head.
8
A
Compaction roll
Punch head
B
7000
6000
Compaction force (N)
A1 A2
5000
4000
3000
2000
1000
Time (ms)
0
0 Tcon
20 40 Tdwell 60 Tdec 80 100
Figure 2. Schematic diagrams depicting the motion of punch head in relation to

compaction roll (A) and the graphical representation of compaction force-time
profile with the associated compaction parameters (B).
Dwell time is often defined as the duration taken for the compaction roll to transit
across the head flat, as it is related to constant vertical punch displacement. This
definition might not be applicable to tableting instrumentation setups in which the
rolls are absent, e.g. compaction simulators. In such scenarios, the period under
which the separation distance between the upper and lower punches remains
constant can be identified as dwell time.
Some authors have defined dwell time as the interval during which maximum
compaction pressure is maintained by the punches during the compaction cycle [29].
However, the geometry of the majority of high speed rotary machines yield rounded
9
pressure peaks, where the true time interval at constant peak pressure may not be
identifiable despite the absence of vertical movement of the punches during the
dwell period as shown in Fig. 2B. On such machines, it is more practical to define
dwell time in terms of compaction force-time profile width at a given fixed fraction
of the peak height. Time interval in a compaction force-time profile during which
compaction force is more or equal to 90% of its peak value, is a commonly
recognized definition of dwell time among practitioners [30].
1.1.3.3.2 Compact deformation during high-speed tableting
Dwell phase compact deformation during high speed tableting can be evaluated from
specific parameters derived from compaction force-time profiles. Dwivedi and
coworkers proposed peak offset time as a deformation parameter to quantify the
extent of compact deformation during dwell phase [31]. The peak offset time is the
difference between the time of peak compaction force and the middle of the dwell
time. Peak offset time was proven as a valid measure of plastic flow during dwell;
punch stress decreased during the peak offset time as the strain remained essentially
constant. Increasing peak compaction force, tableting speed and the presence of
fragmenting materials affected the peak offset time. In a similar attempt to quantify
dwell phase deformation, Schmidt and Vogel proposed an alternative method based
on the degree of asymmetry of the compaction force-time profiles (Fig. 2B) [32].
Dwell period was identified from the compaction profile and divided into halves.
The areas under the compaction force-time curve were calculated for each half, with
both areas decreased by the rectangles under the minimum compaction force in the
dwell time. The ratio of the spatial areas (A2/A1) provided an informative parameter
that quantified deformation behavior and was demonstrated to be sensitive to
compaction load and velocities [33]. The quotient enabled differentiation of the
10
deformation characteristics of different tablet excipients; plastically deforming
materials occupying lower levels of the quotient values and materials that undergo
predominantly brittle and elastic deformations exhibiting higher quotient values [34,
35].
Both methods of plastic flow quantification have been adopted in compaction
studies. Leitriz et al. studied the impact of moisture content on deformation
mechanism of starch during the dwell phase and showed that with increasing
moisture content, the plastic deformation of excipient is amplified so that an
increasing part of the plastic deformation occurs during the consolidation phase, and
a decreasing extent of plastic flow takes place during dwell phase [36]. The
deformation quotient was also demonstrated to be more robust and sensitive to
deformational changes in the compact than peak offset time which exhibited larger
error bars.
1.1.3.3.3 Compaction force-punch separation profile
Compaction force-punch separation profiles allow the calculation of the net energy
input into the compact system which is a function of the material deformation
property [37, 38]. Fig. 3 schematically illustrates a typical compaction force-punch
separation plot. Study of the work of compaction provides information on
elastic/plastic deformation, extent of bond formation and bond disruption during
high speed tableting. Punch separation is the distance between the two punch faces
during the compaction cycle, which can be approximated based on the spatial
relation between tableting punch head, compaction roll and turret configuration as
shown in Fig. 4. Radii of the compaction roll and punch head curvature are
represented by r1 and r2 respectively, wt is the angular velocity of the turret with a
radius of r3, and x2 is the radius of the punch head flat.
11
4500
4000 C
Compaction force
3500
3000
2500
2000
1500
1000
500
0
B D Punch separation A
0.0035 0.0045 0.0055 0.0065
Figure 3. Graphical representation of compaction force-punch separation profile.
r1 Compaction x2
roll
r3 Turret
r2 Punch
wt
head
x2
Figure 4. Model for vertical displacement of a punch head on a rotary press.
12
Alternatively, punch distance can also be obtained from sensitive displacement
probes fitted into the tableting machines that are able to provide accurate punch
displacement data. The amount of energy utilized to compress a powder mass into a
consolidated tablet (net work of compaction; Wnet) and the energy recovered during
the decompression phase (work of expansion; Wexp) can be obtained by calculating
specific areas under the curve in Fig. 3. A is the punch separation at the first
measurable force and B is the minimum punch separation at peak compaction force
C. Punch separation after the decompression phase is represented by D. Area BCD
is equivalent to Wexp while Wnet is determined by the difference between area ABC
and area BCD. Wnet is an indicator of the ability of bonds formed during compaction
to survive the decompression process [39, 40]. An elastic material will exhibit a high
Wexp value and a low Wnet as the input energy for deformation will be largely
recovered during decompression. On the other hand, plastically deforming materials
usually possess high Wnet values as the energy input will be dissipated to a greater
extent for bonding.
1.1.3.4 Compaction parameters
The changes in certain compaction parameters are determined by the extent of
powder bed densification and this is dependent on the predominant deformation
mechanism of the tableting material. Ejection force (force required to push the
compact out of die) and residual die wall stress (pressure exerted onto the die wall
surface by the compact after formation) have been used by researchers to study
compact deformation [41]. A Compact that experiences a high degree of elastic
recovery usually exhibits lower ejection force and residual die wall stress values due
to the axial dimensional recovery of the compact during the decompression phase.
However, the above mentioned parameters may also be influenced by other factors
13
apart from compact deformation such as degree of adhesion at the compact-die wall
interphase. Punch penetration depth, which determines the tablet thickness at
maximum compaction in the die, is another parameter which can potentially be
utilized to study particle resistance to deformation; elastically deforming materials
generally allow reduced punch penetration.
1.1.4 Instrumentation in tablet compaction
The compaction behavior of tableting materials could be significantly altered by the
type of tablet making instrument used [42]. Research pertaining to tablet compaction
usually revolves around three types of compaction systems – rotary press,
compaction simulator and single station press.
1.1.4.1 Rotary tablet press
Commercial tablets are usually produced using rotary tablet presses for such presses
are capable of very high volume and low unit cost production. The basic compaction
operational cycle of a rotary tablet press is essentially the same from machine to
machine. It involves biaxial compaction with both the upper and lower punches
moving in unison towards one another to a specified target punch separation, upon
contact with the compaction rolls. The important operational features of the rotary
press related to tableting performance will be discussed in the subsequent sections.
1.1.4.1.1 Turret speed
Turret speed (tableting speed) is often expressed as revolutions-per-min (rpm). An
increase in turret speed narrows the compaction profile with concomitant reductions
in consolidation (increased strain rate), dwell and decompression times. Turret speed
and dwell time follow an inverse relationship, slowing the turret speed increases the
dwell time as it takes more time for the roll to transit across the head flat [43]. Tablet
14
physical quality as a function of tableting speed has been widely investigated, and
will be discussed under a later section.
1.1.4.1.2 Compaction rolls
The diameter of the compaction roll affects the compaction cycle significantly and
may impact the tablet manufacturing performance as a result. The compaction roll
diameter controls the consolidation and decompression rate, contact time with the
punch head, and the dwell time [44]. A reduction in the roll diameter reduces the
overall contact time and increases the strain rate during the consolidation phase
significantly. The changes in the roll dimensions could particularly be of importance
for materials that undergo time-dependent deformational changes.
1.1.4.1.3 Precompaction
Modern rotary tablet presses will invariably be equipped with dual compaction rolls
known as precompaction roll, positioned just after the feed frame, followed by the
main compaction roll. A compaction force that is a fraction of the main compaction
force is usually applied to the powder bed at the precompaction event. The primary
aim of this activity is to facilitate removal of air in the powder bed and increase the
effective contact time between powder particles under an applied force. During this
extended contact time, some interparticulate bonds may form and partial
neutralization of elasticity may occur [45]. Hence, application of precompaction may
aid in reducing tablet elastic recovery during decompression and strengthen the
tablets, which could reduce incidences of capping and lamination.
1.1.4.1.4 Press technology
Tablet press manufacturers have developed effective tableting systems incorporated
with innovative technologies that enable the modification of compaction parameters,
independent of turret speed and tooling configuration. Dwell duration is often the
15
target parameter for several existing press technologies. A press manufacturer,
Courtoy, manufactures tablet presses with a technology involving the use of air
compensators mounted over the compaction rolls (floating roll technology) [46].
When the compaction force applied exceeds a preset force value, the air compensator
would withdraw the roll upwards and cause displacement of the roll, thereby
maintaining a relatively constant applied force on the compact and consequently,
extending the dwell time without compromising the tableting speed. Another
innovative press technology is the use of a compaction dwell bar (Korsch) which
maintains the pressure on the compact between the precompaction and main
compaction rolls and thereby, extends the dwell time in the process [47].
1.1.4.2 Compaction simulator
A compaction simulator is a tableting machine usually fitted with a typical tablet
tooling setup - upper punch, lower punch and a die - and able to mimic vertical punch
movement during a full compaction cycle of the unit operation [48]. Simulators are
usually instrumented with sensitive sensors that allow monitoring of the
displacement and force profiles associated with the compaction event. Compaction
simulators can be programmed to mimic cycles of tablet press models or other
production conditions with relatively small amount of material. Data collected under
specified compaction parameters using the compaction simulator are relatable to
data from a high speed rotary press, due to the largely similar rapid compaction
kinematics involved [49, 50]. Compaction profiles obtained from the compaction
simulator may exhibit some subtle differences and this is partially attributed to the
lack of a compaction roll in the simulator.
16
1.1.4.3 Single station tablet press
A single station tablet press may be operated electrically or manually. The main
operational difference with single station tablet presses, differing from rotary
presses, is that the lower punch remains stationary during the compaction cycle
(single-ended compaction). The compaction kinematics of the single station presses
are often not comparable to the rapid compaction cycles observed in the above
mentioned tableting instruments. Thus, data collected from the single station tablet
presses may not be reflective of powder compaction during high speed rotary press
tableting, particularly when time-dependent particle deformation is of concern.
Nonetheless, compaction using a single station tablet press is of value when specific
tableting properties related to punch face configurational design, e.g. powder
adhesion due to embossing, or tablet properties independent of strain sensitivity are
studied.
1.2 Tablet physical quality
Tablet physical quality is often associated with the stresses necessary to disrupt the
intact tablet produced. Tablets must possess acceptable mechanical strength and
remain intact during handling and transport, between production and administration.
Optimization of tablet physical quality is an integral part of tablet formulation
development, process up-scaling to bulk manufacture. Tablet physical quality can
be characterized by evaluation of tablet tensile strength, tendency towards tablet
failure and tablet material adherence events.
1.2.1 Tablet tensile strength
Tablet tensile strength for round shaped tablets can be calculated using Eq. (2) [51].
2𝐹
Tensile strength = 𝜋 × 𝑑 (2)
𝑡 × 𝐻t
17
where F is the force needed to fracture a round shaped tablet of thickness Ht along
its diameter dt. The fracture force, often known as tablet hardness value is obtained
from a diametrical compression test (hardness test) in which the tablet is placed
against a platen and force is applied along an axis of the tablet by a movable platen
(Fig. 5). Force applied continuously increases until the tablet fails and the force at
failure is recorded as the fracture force. Tablet fracture force is influenced by the
physical dimensions of the tablet. Therefore, differences in the tablet shape must be
considered during the interpretation of tensile strength values. Tensile strength of
tablets derived from tablets of other shapes can be calculated by modification of the
existing formula. However comparison of values between different classes of tablet
shapes remains debatable.
Alternative methods of tensile strength measurement have been used in solid dosage
development. Flexure test, in which a tablet is bent, has been proposed as an
alternative to the diametrical compression test [52]. Another procedure of tensile
strength measurement involves pulling the tablet along the tablet’s main axes until
failure occurs, used primarily as a means to detect weaknesses in the compact in the
axial direction [53]. Nonetheless, the diametrical compression test remains the most
commonly used method for tablet tensile strength measurement.
Figure 5. Schematic showing the diametrical compression test.
18
1.2.2 Tablet failure tendency
Tablet failure often occurs in the form of capping. Capping refers to the partial or
complete removal of the top and/or bottom crowns of a tablet from the main body
(Fig. 6). This disruption of the tablet system can occur during decompression in the
die or upon the application of an external stress such as during diametrical
compression test. It is generally accepted that extensive tablet elastic recovery during
decompression is one of the primary factors responsible for the occurrence of defects
[54-56]. When the stresses produced by elastic recovery during decompression are
sufficient enough to disrupt the bonds that are formed during compaction, capping
occurs. Tablet capping has also been attributed to a number of other root causes: air
entrapment, mechanism of volume reduction, tableting speed, non-uniform stress
and density distribution as well as internal shear stresses due to die wall pressure.
Figure 6. Image showing absence of capping and tablet capping (left to right) after
hardness testing.
Capping tendency and tablet tensile strength share a positive relation to a certain
extent; formulation with higher capping propensity usually possess low tablet tensile
strength. Several descriptors of tablet failure propensity have been proposed by
researchers. The brittle fracture index (Eq. (3)) proposed by Hiestand and coworkers,
reflects the ability of the tablet to resist fracture during tableting [57, 58].
19
𝑇
𝐵𝐹𝐼 = 0.5 ( ) (3)
𝑇0
where T and T0 are the tensile strength of a compact without and with a center hole
respectively. Askeli et al. proposed calculation of a dimensionless index known as
capping index, an alternative method in which capping occurrence during the
production and testing stages are considered, using Eq. (4).
5𝑁p + 𝑁h
Capping Index = 𝑁t
(4)
where Np is the number of tablets that capped off the press, Nh is the number of
tablets that capped during hardness testing, and Nt is the total number of tablets tested
[59]. The equation differentiated the severity of capping incidences that occur during
rotary tablet production and hardness testing. Presence of capping during the
manufacturing process was considered more severe than capping observed during
the hardness testing.
1.2.3 Tablet material adherence tendency
Tablet material adhesion to punch face and die surfaces is a commonly encountered
problem in tablet manufacture, often referred to as ‘sticking’. Significant sticking
during tablet manufacturing leads to compromised tablet physical quality in the form
of poorer weight uniformity and marred appearance. Mild punch sticking results in
the formation of a material film on the punch face after a certain number of
compaction runs and can be visualized by inspecting the tooling. However, mild
sticking may not cause observable physical defects on the tablet surface. Severe
punch sticking leads to more visually apparent surface defects and will require
termination of the tableting process for cleaning, resulting in a drop of
manufacturing efficiency. The adhesive interactions responsible for the sticking of
particles to the surfaces are predominantly molecular in nature, mainly by van der
20
Waals forces [60-63]. However, other interfacial phenomena such as contact melting
and metallic interactions could also contribute to punch sticking potential [64]. The
sticking tendency of a particular formulation is often not easily detectable during the
early stages of development as adhesion problems usually become apparent at the
later stages of development when prolonged compaction runs are needed for larger
batch production.
1.2.4 Factors influencing tablet physical quality
Tablet physical quality is affected by interparticulate bonds formed within the tablet
matrix during the consolidation/dwell phase and lost during the bond disruptive
events during the decompression/ejection phase. Factors that influence these
processes consequently impact tablet physical quality as well. Some of these factors
related to the presented study are discussed in the subsequent sections.
1.2.4.1 Deformation characteristics of tableting materials
The predominant deformational characteristics of the drug and excipients impact the
evolution of voids and extent of bond formation in the tablet microstructure. Plastic
deformation and fragmentation of particles are the two processes that control the
degree of interparticulate bonding within the tablet microstructure. Fragmentation
affects the number of interparticulate bonds while plastic deformation relates
primarily to the effective area of contact between particles [65, 66]. Nonetheless,
both deformation mechanisms have been reported to enhance tablet tensile strength
[67].
Tablet failure events (e.g. capping) are usually related to low degrees of plasticity in
the compact system. Formulations with inadequate proportion of ductile materials
are more susceptible to bond disruption events during decompression by elastic
recovery. Plastically deforming materials allow greater absorption of the total energy
21
of compaction and less energy to be stored elastically in the compact system. This
reduces tablet elastic expansion during decompression and maintains tablet
mechanical integrity. A higher degree of plastic deformation within the tablet matrix
could also reduce powder adherence to tool surfaces by the formation of stronger
cohesive forces holding particles onto the compact body and therefore, a much larger
adhesive bond breakage energy would be required to enable particle detachment
from the compact onto the punch surface.
1.2.4.2 Lubrication
During the tableting process, friction is generated at the interface between the
compact and tool surface. Lubricants are added to tablet formulations to reduce die-
wall friction. Solid lubricants are usually of small particle size, act by boundary
mechanism in which the lubricant particles adhere onto lager particle surfaces in the
formulation and to the compaction tooling. Magnesium stearate is the most
commonly used boundary lubricant for tableting in the pharmaceutical industry. The
concentration of magnesium stearate in tablet formulations and duration of blending
with the lubricant have been shown to affect tablet mechanical strength; high
magnesium stearate concentrations and prolonged blending may be detrimental to
tablet strength [68]. This latter effect has been attributed to the formation of a
lubricant film around particles which may interfere with particle-particle bonding,
resulting in a less cohesive and mechanically weaker tablet [69]. Material that
predominantly densify via plastic deformation are more sensitive than fragmenting
materials to the deleterious effects of magnesium stearate. Fragmentation allows the
creation of fresh surfaces and reduces lubricant coated surface area that interferes
with bond formation, thus the lubricant film effect by magnesium stearate is reduced
[70]. Lubricants with anti-adherent properties may also be used to reduce powder
22
sticking by the formation of a film which reduces contact between powder
components and tool surfaces [71, 72].
1.2.4.3 Compaction force
Compaction force is a major factor in the powder densification process. An increase
in the force applied on the powder bed causes a greater degree of deformation on the
particles, increases area of contact between particles and consequently strengthens
the compact in the process. A positive linear relationship between compaction force
and tablet tensile strength can be observed for most tablet formulations until a
plateau is reached, indicative of optimum densification of the tableting material. The
linear portion of a tensile strength-force curve can be described by Eq. (5).
𝑇𝑆 = 𝐶𝑝 . 𝑃 + 𝑏 (5)
where P is the peak compaction pressure, TS is the tablet tensile strength and Cp is
the compactibility coefficient (calculated from the slope of the linear region). Cp
quantifies the tabletability of a specific compact system, indicating the effectiveness
of the applied force in enhancing interparticulate bonding within the tablet
microstructure. Plastically deforming tableting materials generally exhibit high
tabletability [73]. In contrast, fragmenting and elastically deforming materials
provide greater resistance to densification at higher compaction pressures, leading
to reduced tabletability. While increases in compaction force allow greater degree
of deformation and bonding, it is often accompanied by increases in stored elastic
energy. Hence, tendency towards tablet failure increases at higher compaction
forces. Powder tendency to adhere to punch face is similarly believed to be sensitive
to compaction force level; sticking problems could possibly be mitigated with the
use of higher compaction loads during tableting [74, 75].
23
1.2.4.4 Tableting speed
Tableting speed determines the rate of punch displacement/force application on the
compact. Adjustments to the tableting speed often alters the compaction cycle profile
and parameters (consolidation, dwell and decompression times), as described earlier
in Section 1.1.4.1.2. This affects how particles deform and therefore, determines the
integrity of the compact formed. Consolidation and dwell times are related to rate of
particle rearrangement and stress relaxation, respectively. The decompression phase
determines the rate of tablet recovery that occurs in the die.
The influence of tableting speed on tablet physical properties is dependent on the
deformation behavior of the tableting material [76-86]. Tableting materials with
viscoelastic and plastic character are more sensitive to tableting speed; increases in
tableting speed generally cause a deterioration in tablet mechanical properties.
Tablet failure events are more likely to occur under accelerated tableting conditions.
This is attributed to the shift in the elastic-plastic equilibrium in the compact brought
about by changes to the compaction cycle. At higher tableting speeds, there is
reduced time for stress relaxation and plastic flow to take place. This increases the
amount of elastic energy released, in the form of tablet expansion. As a consequence,
bonding quality is negatively affected. Studies have also shown that increases in
tableting speed may cause the yield pressure of the tableting material to increase
correspondingly; resistance to densification increases at higher tableting speed [66].
Brittle materials are believed to possess reduced sensitivity to tableting speed, as
fragmentation process occurs primarily during the consolidation phase at lower force
levels. However, contradictory results have been reported in literature.
Microcrystalline cellulose and lactose tablet properties have been demonstrated to
be independent of tableting speed, while dicalcium phosphate, which was expected
24
to possess low strain rate sensitivity, showed increased tabletability at higher
tableting speeds. This was attributed to more extensive brittle fracturing of dicalcium
phosphate particles at higher speeds.
1.2.4.5 Tablet press and tooling technology
The compaction behavior of pharmaceutical materials could be significantly altered
by the tableting technology and compaction tooling used [42]. Tablets compacted
using manual presses usually exhibit superior physical quality than tablets from
rotary presses due to the differences in compaction kinematics. Compaction cycles
in the manual press are much longer and this facilitates greater degree of time-
dependent deformation (plastic flow) to take place, leading to better tablet physical
quality. Rotary presses are often fitted with innovative tableting technologies to
improve tablet quality or productivity. Multiple compaction stations, extension of
dwell time and advanced vacuum system for removal of air within the powder bed
are available press technologies that may enhance interparticulate bonding in the
tablet structure and mitigate tableting problems. Some of these technologies have
been discussed in Section 1.1.4.1.5.
Similarly, tablet compaction tooling can affect the physical quality of the tablets
produced. Modifications to tooling surface properties and metal composition has
been reported to alter the liability of tablet sticking. Roberts investigated the
influence of punch face surface quality and the effect of chrome plated punches on
material sticking and claimed that punch surface quality is an important factor as
surface roughness positively correlated with powder adhesion [87]. However, a
smoother punch face and chrome plating on the punch faces may not necessarily
correspond to a reduction in material sticking onto punches [87]. Modifications to
punch configuration may also impact compact deformation process during the
25
compaction cycle. The influence of geometrical changes to the tableting punch
configuration on the compaction process and tablet physical quality will be
discussed in later sections.
1.3 Tablet compaction tooling
Tablet compaction tooling and formulation optimization equally contribute in
successfully ameliorating a tableting program. Tooling must be designed and
engineered to produce tablets of acceptable quality and be able to withstand stresses
associated with the compaction process for a prolonged service life. Understanding
the basic physics of powder compaction in relation to tooling modifications
enhances the ability to produce quality tablets more efficiently and provides greater
knowledge to troubleshoot and identify potential pitfalls of a tableting program.
1.3.1 General concepts
A single set of tablet tooling typically consists of an upper punch, a lower punch
(Fig. 7), and a die. Internationally recognized standards for tablet tooling includes
the TSM (Tablet Specification Manual) and EU (European) standards [88, 89]. Both
standards identify the physical tool configuration for B- and D-type compaction
tools, the majority of the tool configurations used today. Standardization of tooling
configuration allows interchangeability between tablet presses. B-type tooling
possesses smaller configurational dimensions and allows the insertion of a greater
number of punches in a turret for a higher tablet production rate. The larger body of
the D-type tooling is favorable for producing large tablets. The TSM and EU
standard manuals provide useful technical information such as critical dimensions,
associated tolerances for standard tooling and valuable references for trouble
shooting and tool inspection.
26
Head flat
Head
radius Head
Neck
Barrel
Punch face with concavity
Figure 7. Schematics showing the lower (left) and upper (right) punches.
27
Innovations in tablet press and tooling technology over the past decades have
improved efficiency by allowing increased tableting speed, improving product yield
and quality along with improvements in the sanitary and safety aspects. Tableting
tools are engineered by tool makers to withstand higher stresses, last longer and
produce tablets of better quality. In some cases, standard tooling configurations may
not achieve the expected tableting performance, particularly when tableting
materials have problematic compaction characteristics. In such cases, minor
modifications to the tooling configurations may be necessary for optimal tableting
performance. Modifications to punch head and punch face configurations are some
design tweaks that could potentially improve tableting performance.
1.3.2 Punch head configuration
Punch head configuration is clearly one of the most reworked areas on the tablet
tooling. The head flat and head radius are design features of the punch head that can
potentially be modified by tooling manufacturers to prolong tooling life and improve
product quality (Fig. 8). Punch head flat refers to the flat portion of the punch head
that makes contact with the compaction roll and determines the dwell time of the
compaction cycle. The flat top on the punch head provides a space for constant strain
onto the compact in the die while punches are sliding over the compaction rolls.
Head flat
Head radius
Figure 8. Schematic showing the punch head configuration with the associated head
flat and head radius.
28
Modifications to the punch head flat specification affect the compaction parameters,
consequently impacting the tablet mechanical properties. The diameter of the punch
head flat, turret speed and turret pitch circle diameter determine the dwell time as
shown in Eq. (6).
60 × 𝐷ℎ𝑓
𝑇𝑑𝑤𝑒𝑙𝑙 = 1000 (𝜋 × 𝑉 × 𝐷𝑡𝑢𝑟𝑟𝑒𝑡
) (6)
𝑡𝑢𝑟𝑟𝑒𝑡
where Dhf and Dturret are the diameters of the head flat (if oval, the longest dimension)
and turret respectively, Vturret is the velocity of the turret in revolutions-per-min
(rpm). Dwell time has been pointed out as one of the key compaction parameters for
tablet quality [90]. Depending on the tableting material densification characteristics,
increases in dwell time could potentially result in mechanically stronger tablets due
to a favorable shift in the elastic-plastic equilibrium in the compact and this leads to
better energy utilization towards bond formation. In view of this effect, some tooling
manufacturers provide punches with a larger head flat area. Enlarging the head flat
transverse length leads to a longer flat surface for the rolls to transit across the punch
head, consequently increasing the dwell time and providing a longer ‘relaxation’
time period for the compact. However, extension of dwell time via head flat
modification is limited by punch body configuration. The initial contact of the
compaction roll to the punch head should be within the diameter of the neck (Fig. 7)
so as to provide structural support [89]. Otherwise, punch head fracturing could
occur with prolonged use. Thus, the extent to which the head flat can be expanded
is rather limited. The type of tableting tools (B- or D-type configuration) affects the
dwell time as well. Due to the ability of the larger barrel to accommodate a larger
punch head and head flat profile, the use of D-type tooling usually leads to longer
dwell time. Despite the known effect of dwell on compact deformation, information
29
on the influence of changing punch head profile on compaction process and tablet
physical quality is rather limited in literature.
Head radius refers to the radius of the curved surface on the top of the head which
blends the middle section of the head to the head flat, often referred to as the ‘domed’
head configuration (Fig. 8). The head radius makes initial contact with the
compaction roll and allows for a smoother transition of the punches into the
compaction cycle. The presence of a head radius reduces the impact stress when the
punch makes initial contact with the compaction roll. The impact of the roll on the
punch head at high speeds and compaction forces may cause work-hardening issues
which are detrimental to both the compaction rolls and punches. Punch head radius
theoretically affects the compaction cycle by influencing the rate of compaction
force application on the compact during the consolidation phase. A reduced head
radius leads to the formation of a curved punch head with a steeper slope which
could possibly cause a reduction in the consolidation duration and increase the strain
rate during the consolidation phase.
1.3.3 Punch face configuration
Punch face configuration determines the tablet face contour for shaped tablets.
Configurational modifications to the face design can lead to profound differences in
the tooling life, powder compaction process and the resultant tablet quality.
Optimization of punch face configuration is essential for a successful tableting
program.
During compaction, the punch face surfaces experience lateral stress that causes
expansion of the sides of the punch cup outward toward the die wall momentarily.
Excessive stresses developed during compaction beyond the maximum allowable
force level for a particular face configuration may lead to permanent distortion and
30
premature failure of the punch. The maximum allowable compaction force for a
particular punch face configuration is determined by the punch face configuration,
area and stress concentration factors. Some punch face configurations possess areas
that experience high levels of stress due to transitions in their geometric profile,
resulting in a lower maximum compaction force. Punch face configurations could
be modified to strengthen the punch. Reduction of the cup depth and increasing the
cup radius are two possible modifications. A deep concave cup configuration
facilitates development of high stress points on the cup edges while configurations
with a reduced depth aid in the distribution of the stresses on the punch face surface
and could potentially lengthen tooling life. Similarly, modification of punch face
edge could also potentially improve tableting performance. The flat-face bevel edge
(FFBE) tablet shape is a popular choice among tablet manufacturers due to the
availability of a top flat surface for embossing purpose (Fig. 9B). In comparison to
the flat-face plain (FFP) shape (Fig. 9A), the FFBE shaped tablets exhibit reduced
surface attrition. A simple modification of the bevel edge to a curved surface with a
radius (Fig. 9C) strengthens the punch and allows a higher compaction force to be
applied without causing punch tip damage and eliminates edge chipping on the tablet
surface.
In relation to powder compaction, modifications to punch face configuration affect
powder flow and packing in the die under stress conditions. During the tableting
process, differential movement of powder within the die occurs; central regions
experience greater movement than peripheral regions due to frictional forces along
die wall-compact interphase. Punch face configurations that lack concavity cause
uniform but restricted movement of particles along the flat surface of the punch face
and limit the powder movement along the die walls.
31
A
Figure 9. Schematics showing the FFP (A), FFBE (B) and FFRE (C) tablet designs
with difference in tablet edges shown. (FFP: flat-face plain, FFBE: flat-face bevel
edge, FFRE: flat-face radius edge)
The introduction of concavity to the punch face promotes in-die particle movement
to a greater degree due to the inherent greater volume to fill towards the center of
the compact [91]. David et al. demonstrated that the extent of powder movement
was dependent on the relative curvature of the punch face [92]. Increasing curvature
led to a corresponding increase in the lateral particle movement within the powder
body. Similar findings were revealed in a study by Brunel et al., in which a unique
X-ray imaging method was used to study powder movement [91]. FFP punch face
was shown to promote greater lateral powder movement, primarily at the top corners
of the tablet away from the die wall with corresponding minimal radial movement.
Radial powder movement was greater in convex curved tablets at regions adjacent
to die wall and along the edges of the punch surface.
32
Differential powder movement due to punch face curvature leads to variations in the
density distribution within the tablet and formation of non-homogenous tablet
internal character [93]. Density distribution pattern of uniaxially compacted flat-
faced tablets generally adhere to the classical Train explanation; high density regions
at corners and central regions of the tablet with concomitant stress relief within the
parts of the tablet adjacent to the moving punch face [94]. Introduction of a curved
concavity to the punch face increases variation in the density distribution producing
tablets with much more heterogeneous internal structure, leading to the creation of
density boundaries between regions of high and low densities which are susceptible
to fracture failure [95, 96]. This could consequently impact tablet physical quality
and their tendency to cap.
Modifications in the punch face configurations have also been reported to cause
changes to the net resolution of the resultant compaction force within the compact
and lead to heterogeneous stress distribution within the tablet. The use of a punch
face with a curvature causes the net resolution of the resultant compaction force to
be nearer to the moving punch face and towards the central regions of the tablet [97].
It is generally believed that the formation of non-uniform stress distributions due to
different punch face configurations is correlated to the tablet mechanical properties.
Kadiri and Michrafy studied the effect of punch face configuration on compaction
force transmission through the powder compact and demonstrated that increasing
punch depth (reducing curvature radius) created a distinct stress distribution pattern
with the potential to initiate crack formation and propagation [98].
Apart from being an influential factor on tablet mechanical properties, punch face
configuration also contributes to powder adherence during tablet compaction,
related to stress concentration factors on the punch face. Optimization of punch face
33
configuration via modifications to cup design and engraving style is often required
for formulations prone to punch sticking. Waimer et al. studied the influence of
tablet engravings with respect to powder adhesion using an instrumented upper
punch and claimed that concentration of shear stresses at the lateral faces of the
engraving during compaction and stress distribution within the tablet are important
contributors to strong powder adhesion [99]. Similar findings were reported in
another study in which embossment on the punch face increased powder adhesion
considerably, attributed to the probable concentration of shear stresses at the lateral
faces of the embossed logo [100]. Modification in the punch curvature from flat-face
configurations to concave configurations was also shown to reduce powder adhesion
potential, possibly due to variation in the stress distribution on the punch face.
1.4 Computational simulation of powder compaction
Computational modeling of powder compaction has generated growing interest as
direct measurement of particle movement and behavior during the compaction
process in the die is challenging. In view of this limitation, computational modeling
has been utilized as an effective tool for predicting compaction dynamics.
Computational modeling of powder compaction has typically been carried out by
two approaches: finite element modeling (FEM) and discrete element modeling
(DEM) methods [101].
1.4.1 Finite element modeling (FEM)
FEM treats the powder as a continuous material with elastic-plastic material property
throughout the continuum solid. Although a pharmaceutical powder mass is clearly
discontinuous at the particle level, this becomes irrelevant at the larger scale of
aggregation such as when it is compacted into a relatively dense compact in a die
during tableting. Therefore, the compaction behavior of pharmaceutical powders can
34
possibly be studied using principles of continuum mechanics [102]. The Drucker-
Prager Cap model is widely accepted to be a good constitutive model for modeling
powder compaction as it considers the densification and hardening character of
powders and captures the shearing phenomenon in powders during the different
stages of compaction [103-107]. Materials and compaction condition specific
parameters are required to simulate different compaction conditions. Sinha and
coworkers demonstrated the possibility of obtaining the necessary model input
parameters via simple compaction procedures [108]. The authors compared the
simulation of powder compaction using constant property and relative density
dependent property based Drucker-Prager Cap models and demonstrated that while
the relative density dependent property model proved to be more accurate, the
constant property model could be considered reasonably accurate. The use of a
relative density dependent property model may actually be disadvantageous due to
the requirement of high computational power and time-consuming calibration steps.
The Drucker-Prager Cap model has been used in FEM to study density and stress
distribution of a consolidated powder for identification of potential tablet failure
modes in compacted tablets. Wu et al. demonstrated that FEM is able to capture the
essential aspects of the behavior of pharmaceutical powders during compaction and
can predict possible mechanism of crack propagation during the decompression
phase from the examination of the evolution of stress distribution during unloading.
Localized intensive shear bands developed from the tablet edge to the central regions
of the tablets, indicative of failure patterns within the tablet structure [109]. The
influence of punch face configuration on density/stress distribution and the relation
with tablet mechanical properties has been investigated by a number of researchers
with the employment of FEM. Han et al. demonstrated that introduction of concavity
35
to the punch face caused large shear stresses at the top corner of the compacts, due
to radial elastic recovery of the material at the top central region upon removal of
the upper punch force [110]. The simulated results provided a possible explanation
for the more frequent occurrence of capping in concave tablets. In a similar study,
Kadiri et al. demonstrated that dense regions with strong shear stresses were formed
at the upper edges and the effect of punch face design on stress distribution became
more significant with changes in punch depth [98]. The influence on tablet thickness
and punch curvature on density distribution was further investigated by Diarra and
co-workers [95]. Flat-face plain and convex tablets compacted at equivalent force
levels exhibited different density distribution at the central region of the tablets and
increases in curvature radius promoted a smoother density gradient at the tablet
edges. From a tooling perspective, FEM enables accurate identification of stress
concentration factors on the punch face and this allows optimization of tooling
design by manufacturers [111].
1.4.2 Discrete element Modeling (DEM)
DEM models consider the powder, one particle at a time. The rules of the model
determine the extent of particle interaction. The particles move in the predetermined
simulation area, and collisions or contacts between them are analyzed by the
simulation [112-114]. Since every particle is handled separately, the number of
particles largely determines the computing time and outcome of the simulation. As
such, during development of the model, the properties of the particles, such as
density and yield strength, become inputs for the model.
Compaction induces very complex states of stress in the packing structure and
involves microscopic interactions between particles. FEM models are based on
continuum mechanics that operate largely at a bulk scale and focus on
36
phenomenological description of a compaction process. Bulk response of a compact
obtained in FEM may not be related to the microscopic information. In comparison,
DEM simulations allow detailed microscopic information of compacts to be
obtained. DEM allows the processes involved during compression and consolidation
- particle rearrangement and particle deformations - to be mimicked more precisely
with accurate particle properties. Sheng et al. successfully simulated uniaxial
compaction of powder particles using a DEM model and also showed that the DEM
results correlated well with experimental results [115]. Frenning et al. utilized a
combined finite/discrete element model to simulate compaction of particles and
showed that realistic systems comprising 1000 or more particles may be successfully
analyzed within a reasonable computation time [116]. In another study, Siiria et al.
used an improvised DEM model to calculate bond strength distribution in cylindrical
tablets and showed that the estimated tablet strength increased as a function of the
compaction force in the same way as the measured breaking strength did in the
tablets [117].
While DEM has been successfully applied to simulate particle packing, it has its own
set of limitations that requires further improvement. Theoretically, DEM allows
evaluation of an infinite number of particles. However, this may not be feasible in
practice. DEM calculates the interaction of each particle. A large number of particles
will require long unrealistic computation duration for a complete analysis of the
interactions. Thus, normally, the number of particles are kept to a maximum that
allows realistic computing times. Adjustments to particle and physical parameters
are usually needed to ensure that simulated particle movement could be fitted
properly to experimental bulk powder behavior [118].
37
1.5 Research gaps
Tablet compaction process has been extensively studied, numerous publications and
theories are readily available. There have been considerable literature reports on the
effect of formulation factors on tableting performance. In comparison, the influence
of process factors on the compaction process has been less studied. Among the
limited studies that investigated tableting process factors, research on tablet
compaction tooling is scarce in published literature.
Tablet compaction tooling exerts significant influence on the densification behavior
of particle and consequently impacts the tableting outcomes. However, research on
the specific impact of tooling configurational modifications on the compaction
process and resultant tablet properties has been limited. Existing research data on
powder compaction could not be extrapolated to predict the influence of tooling
modifications on compaction parameters and tablet properties, especially during
high speed tableting as compaction studies reported in literature were often
conducted under tableting conditions that were not representative of commercial
tablet production rates as encountered during actual tablet manufacture. In addition,
discrepancies in the results of tablet properties have been reported in the literature,
mainly due to the large number of varied methods used for compaction studies and
formulation dependent outcomes. These research gaps lead to a need to better
understand the influence of tooling configurational modifications on the tableting
process and quality of tablets produced. Fundamentally, the study of compaction
tooling design is necessary to elicit the compaction physics involved as this could
aid in making judicious choices of processing considerations and tool designs for a
successful tableting program.
38
1.5.1 Dwell effect on tablet quality
As previously described under Section 1.1.2, compact deformation during the dwell
phase differs from other stages of the compaction cycle. Compacts experience
greater plastic flow under constant strain, which could potentially impact tablet
strength. There are limited publications correlating the duration of dwell phase to
the resultant tablet physical quality. In a study by Vezin et al, it was reported that
dwell time extension could be less important than lag time (separation time between
precompaction and main compaction) prolongation in affecting tablet mechanical
strength, attributed to enhanced deaeration of the powder bed during the longer
separation time between the two distinct compaction events [29]. However,
contrasting results were reported in another study where dwell time was found to
exert a more pronounced influence than lag time in affecting tablet physical quality
[119].
The influence of dwell time on tablet physical quality in relation to different
deformation mechanisms of tableting materials has also been explored. Tye and co-
workers claimed that the impact of dwell phase on tabletability was dependent on
the predominant deformation mechanism of the tableting material [120]. Highly
ductile materials were unaffected by dwell time changes while materials with
viscoelastic nature were substantially affected. However, in another study when
pectin (brittle material) was compacted, significant reductions in peak force were
observed and tablets of superior physical quality were produced with prolongation
of dwell time [121]. The results indicated some degree of viscoelastic character in
the pectin compact despite its brittle nature.
The results from the various studies often did not concur and could be attributed to
differences in the amount of plastic and viscoelastic components in the tablet
39
formulations used. Dwell duration used in existing studies were often not
representative of high speed tablet compaction and may not provide an accurate
understanding of dwell influence on compact deformation process.
1.5.2 Punch head configurational modification
Modifications to the punch head configuration affects the compaction profile and
parameters with potential consequences on mechanical properties of tablets
produced. Alterations to the head flat and radius specifications could possibly
change the dwell time and consolidation strain rate during consolidation phase.
Sarkar et al. highlighted the complex relationship between dwell time and strain rate
during the consolidation phase and the impact on the resulting tablet properties
[122]. It was shown that the influence of strain rate on tablet strength and capping
tendency could predominate over dwell time, particularly when the compaction
force employed is above the minimum force required for particle deformation.
Higher strain rates reduced the time for particles to rearrange during the
consolidation phase, and had possibly contributed to more air entrapment. These
effects consequently impacted interparticulate bond formation negatively and led to
extensive tablet capping. However, extended dwell time at a low precompaction
force followed by application of a main compaction force improved the tablet
physical quality and alleviated the capping situation. It was explained that the
prolongation of dwell time during the precompaction stage allowed more time for
particle movement and air escape, resulting in improved conditions for compact
formation at the main compaction stage and led to greater densification and bonding
quality within the tablet microstructure.
While a considerable number of publications related to tableting is available, the
impact of geometrical changes to the head flat and radius on the tableting process
40
has not been well investigated [123]. Existing studies have largely focused on the
effect of compaction parameters on tablet properties. In most studies, the compaction
parameters were modified by adjustments to the compaction kinematics, e.g.
tableting speed, which resulted in uncontrolled concomitant changes to other
parameters of the compaction cycle and could not be extrapolated in predicting the
effect of punch head and radius modifications on the parameters and the resultant
tablet properties.
1.5.3 Punch face configurational modification
Punch face configuration is an important feature of the tableting punch that affects
tableting performance significantly, as described in Section 1.3.3. Modifications to
punch face configuration impart differences to powder movement during
densification, leading to variations in the density distribution within the tablet.
Formation of a non-homogenous tablet internal character may cause the tablet matrix
to be susceptible to fracture failure which could affect tablet physical quality. In
comparison to punch head configuration, the influence of punch face configurational
modification has been more widely studied. Stress/density conditions in the tablet
matrix due to different punch face specifications have been correlated to tablet
physical quality, particularly in terms of tablet failure and powder adherence
tendency.
Majority of the studies related to punch face configuration involved concave shaped
punches, with different degrees of curvature. Research on the influence of punch
face modifications with regard to flat-face tablets of different edge geometries has
been rather limited. Considerable differences in cup depth and other geometrical
specifications between flat-face and concave punch face configurations restricts
extrapolation of results from existing publications in predicting the effect of flat-face
41
punch edge geometrical modifications on tablet compaction.
42
CHAPTER 2
HYPOTHESES AND
OBJECTIVES
43
2. HYPOTHESES AND OBJECTIVES
This study was motivated by the need to provide a more comprehensive
understanding on the influence of tooling configurational modifications with respect
to tablet properties as this knowledge could potentially aid in making process and
formulation considerations for a successful tableting program. The findings could
also facilitate the design of better compaction tools as there would be a clearer
understanding of the impact of minor tool modifications to the quality of tablets
produced.
The lack of studies specific to tooling configuration was identified as a research gap
in this area of compaction research. Process and formulation conditions adopted in
many compaction studies were often not representative of actual manufacturing
conditions and many studies produced contradicting evidence on tablet properties.
These factors restricted the extrapolation of such data for predicting the effect of
tableting punch head and face modifications on the tableting process and physical
quality of the tablets to be produced.
Therefore, the main hypothesis of this work is that tableting punch head and face
configurational modifications of flat-face punches can alter compaction parameters
and the resultant tablet physical quality.
Investigations with the following objectives was conducted to test the research
hypothesis:
A. To investigate the influence of formulation variables on compact deformation
during dwell phase;
B. To investigate the influence of punch head configuration on compaction process
and tablet physical quality;
44
C. To investigate the influence of punch face configuration on compaction process
and tablet physical quality; and
D. To investigate the influence of punch face configuration on tablet material
adhesion during tablet compaction;
The research work done was organized into four parts:
Part 1. Investigation of the influence of deformation mechanism of tableting
materials during dwell, and the consequent effect on tablet mechanical properties
Part 2. Investigation of the effect of punch head modification on compaction
parameters and tablet physical quality
Part 3. Investigation of the effect of punch face modification on compaction
Part 4. Investigation of the effect of punch face modification on powder adhesion
tendency to the punch surface during powder compaction
45
CHAPTER 3
EXPERIMENTAL
46
3. EXPERIMENTAL
materials during dwell, and the consequent effect on tablet mechanical
properties
3.1.1 Rationale of study
Tableting outcomes are closely related to the predominant deformation mechanism
of the tableting material. Compact systems with different degrees of plastic, elastic
and fracture deformation may react to constant strain during dwell phase differently,
consequently impacting resultant tablet properties. Therefore, it was of interest to
investigate the influence of dwell on plastic flow in different compact systems. Three
types of tableting materials with contrasting densification behavior were used in this
study. Influence of dwell time extension on compact deformation and mechanical
properties of tablets compacted from different formulation systems were examined.
3.1.2 Choice of tableting materials
Microcrystalline cellulose (MCC; Avicel PH101, FMC Biopharma, PA, USA),
dicalcium phosphate (DCP; Emcompress, JRS Pharma, Rosenberg, Germany) and
paracetamol (APAP; Granules India Limited, Hyderabad, India) were chosen as
model materials in this study. Microcrystalline cellulose and dicalcium phosphate
have been studied extensively in literature, representative of materials that deform
plastically and brittle fracture respectively [11, 124]. Paracetamol is generally
considered to predominantly undergo elastic deformation, often used in studies as a
model drug in formulations prone to tablet failure [59, 122].
47
3.1.3 Characterization of tableting materials
Heckel analysis was utilized to study densification behavior of the tableting
materials. Heckel plot was constructed using a universal testing machine (Autograph
AG-10KNE, Shimadzu, Kyoto, Japan). The universal testing machine allows
compaction of a powder bed at high compaction pressures, under highly controlled
settings. It is fitted with a 100 kN load cell which controls the rate of load
application. The load cell applies a compaction load on the upper punch, which
transmits the force into the powder bed confined within the customized die
(positioned in the die plate). A 10 mm diameter flat-face punch, with a
corresponding 10 mm die was used for compact formation. Heckel plot was
constructed for each of the three types of tableting material, using an out-of-die
method. An accurately weighed amount (325 mg) of powder was manually fed into
the die. The compaction speed was fixed at 10 mm/min, programmed using the in-
built software Trapezium (Shimadzu, Kyoto, Japan). Tablets were formed at
compaction pressures ranging from 25 to 150 MPa. The compaction process was
programmed to allow immediate withdrawal of the load upon reaching the stipulated
maximal pressure, without dwell time. Post compaction tablet thickness was
measured using a digital micrometer (Series 293, Mitutoyo, Kawasaki, Japan).
Compressed densities of the compact at each specified compaction pressure were
calculated based on the measured compact dimensions. The true densities of the
tableting materials were determined using a helium pycnometer (Pentapycnometer,
Quantachrome Instruments, FL, USA). The intra-tablet porosity is related to the
compressed and true densities of the deformed particles. Porosity (ɛ) values of the
compacts were calculated using Eq. (7). The Heckel equation (Eq. (1)) was
employed to characterize densification behavior of the tableting materials.
48
𝐶𝑜𝑚𝑝𝑟𝑒𝑠𝑠𝑒𝑑 𝑑𝑒𝑛𝑠𝑖𝑡𝑦
ɛ=1− 𝑇𝑟𝑢𝑒 𝑑𝑒𝑛𝑠𝑖𝑡𝑦
(7)
The Heckel profiles of the materials were analyzed based on the evolution of the
Heckel curve with increasing compression pressure. A typical Heckel plot usually
consists of an initial curved region at the lower compression pressure regions
followed by a linear segment at higher pressures. The initial curved region which is
indicative of the fragmenting tendency of the particles were compared. Materials
with a steep initial curve at the lower compressive pressures may be suggestive of a
brittle nature. Slopes of the linear portion and onset pressure at which the linear slope
begins were also analyzed to provide a relative comparison of the deformation nature
of the materials at high pressures, in terms of elastic/plastic deformation. A high
slope of the linear section of the Heckel plot is representative of plastically
deforming material with a low yield pressure. Interception of the vertical axis from
the extrapolation of the linear segment is related to the apparent density of the
powder and particle packing before force application.
3.1.3.2 Particle size analysis
Particle size analysis was conducted on the tableting materials using a laser
diffraction particle sizer (Small Volume Module LS 230, Beckman Coulter, IN,
USA). The dx value corresponded to the Xth percentile under the cumulative
undersize distribution curve. The particle size expressed in terms of the median
particle size (d50) and the size distribution was calculated in terms of span using Eq.
(8).
𝑆𝑝𝑎𝑛 = (𝑑90 − 𝑑10 )⁄𝑑50 (8)
49
3.1.3.3 Hausner ratio
Bulk and tapped densities of the tableting materials were determined for the
calculation of the Hausner ratio. The material was passed through a sieve of aperture
size 1.7 mm into a 100 mL graduated cylinder. The top of the powder was gently
levelled with a spatula, powder adhering dusted off and the mass of powder in the
cylinder accurately weighed. The powder in the graduated cylinder was then
mechanically tapped to a constant volume using a tap density tester (Sotax TD2,
Sotax, Aesch, Switzerland). The final tapped volume was noted and used to calculate
tapped density. The Hausner ratio was calculated using Eq. (9).
𝑃𝑡𝑎𝑝𝑝𝑒𝑑
𝐻𝑎𝑢𝑠𝑛𝑒𝑟 𝑟𝑎𝑡𝑖𝑜 = (9)
𝑃𝑏𝑢𝑙𝑘
where Ptapped and Pbulk represent the tapped and bulk densities respectively. The
Hausner ratio indicates the flowability of a particular powder blend; larger values
indicate poorer flow characteristics [125].
3.1.3.4 Angle of repose
The angle of repose (AOR) was determined using a powder flowability tester
(Flowability tester BEP 2, Copley, Nottingham, UK). The powder was passed
through a sieve of aperture size 1.0 mm into a funnel equipped with a 10 mm nozzle
mounted 75 mm above a 100 mm diameter circular test platform. The powder was
allowed to fall gradually onto the platform to form a conical heap. The height of the
formed heap was measured using a digital height gauge (Absolute, Mitutoyo,
Kawasaki, Japan). Angle of repose was calculated as the angle between the slopes
of the conical heap relative to the horizontal base. Smaller values for the angle of
repose indicate powders with better flow characteristics [126].
50
3.1.4 Preparation of tableting blends
Since the objective of this study was to investigate the influence of formulation
variables on compact deformation during dwell, binary blends (APAP/MCC,
DCP/MCC and APAP/DCP) were prepared by mixing the respective tableting
material in predetermined proportions (percent proportions; 0/100, 25/75, 50/50,
75/25 and 100/0) in a polyethylene bag for 10 min. The final formulation blends
were prepared by mixing the binary blends with magnesium stearate 1 % w/w
(M125; Productos Metalest, Zaragoza, Spain) in the polyethylene bag for 1 min,
prior to tableting.
3.1.5 Tablet production
Tablets were produced from each formulation blend at compaction forces of 6 and 9
kN (compaction pressures, 76.4 and 114.6 MPa) using the universal testing machine
(Autograph AG-10KNE, Shimadzu, Kyoto, Japan). An accurately weighed amount
(325 mg) of powder was manually fed into the die. Compaction speed was kept
constant at 10 mm/min. Tablets were compacted at different dwell durations ranging
0-2 s for each tableting formulation.
3.1.6 Stress relaxation test
Viscoelastic behavior of the compacts during dwell period was evaluated from stress
relaxation tests. Stress relaxation test was conducted on each formulation blend
using the universal testing machine with a 10 mm diameter flat-face punch and
compaction speed of 10 mm/min. The compaction load cell was programmed to
pause at the position where it achieved the maximal stated force for a predetermined
duration, upper punch displacement was maintained during that period. The
reduction in force of the upper punch in relation with time was monitored during the
dwell period. The force measured from the load cell during this period is
51
representative of the resistance of the compressed material to densification. Plastic
flow due to viscoelastic deformation during constant strain (dwell) causes force
decay in the compact, reflected in the stress relaxation profile [25, 26].
The degree of viscoelastic compact deformation during dwell phase was analyzed
using different viscoelastic parameters calculated from the stress relaxation profile
for each formulation. The ratio of the maximum force (CSmax) at the beginning of
the dwell phase and the force measured by the upper punch after the defined time of
dwell (CS0) provides the viscoelastic parameter, relaxation quotient (RQ).
𝐶𝑆𝑚𝑎𝑥
RQ = (10)
𝐶𝑆0
RQ provides an indication of overall plastic flow during dwell time [28]. However,
it should be noted that RQ assesses the compaction process during dwell and does
not describe the compaction process preceding dwell phase. The slopes of the linear
sections of the stress relaxation profile provides valuable information related to the
rate of plastic flow during dwell phase. Viscoelastic slopes (VS) were calculated by
subjecting the stress relaxation profile of each formulation blend to linear regression
analysis with the identification of the linear sections (R2 > 0.99) of the profiles. This
method has been reported to be informative of the compact deformation process
during dwell [23].
3.1.7 Tablet characterization
3.1.7.1 Viscoelastic strain recovery
Viscoelastic strain recovery refers to the post compaction time-dependent tablet
expansion that occurs during storage period without being restrained by die wall
[127]. Tablet viscoelastic recovery and immediate elastic recovery are related [128].
It is generally believed that majority of the stored energy due to elastic deformation
52
of particles is released via immediate elastic recovery during decompression.
However, tablets produced from materials of viscoelastic character may also
undergo retarded elastic energy release during storage leading to dimensional
changes in the tablet during storage. The percent viscoelastic strain recovery was
calculated by the equation (Eq. (11)) proposed by Maganti and Celik [129].
𝐻72hr − 𝐻dcp
VER (%) = 𝐻dcp
× 100 % (11)
where tablet thickness immediately upon ejection (Hdcp) and tablet thickness 72
hours post compaction (H72hr) were measured by an out-of-die method using a digital
micrometer (Series 293; Mitutoyo, Kawasaki, Japan).
3.1.7.2 Tablet tensile strength
Mechanical quality of compacted tablets was determined from the calculation of
tablet tensile strength (Eq. (2)) which is indicative of interparticulate bonding
quality. Breaking force of tablets was obtained from diametrical compression testing
using a hardness tester (TBF 1000, Copley, Nottingham, UK). Tablet diameter and
thickness 72 hours post compaction were measured using a digital micrometer
(Series 293, Mitutoyo, Kawasaki, Japan).
3.1.8 Statistical analysis
Statistical analysis of the data was performed using PASW Statistics 18 Software
(SPSS Inc., Chicago, IL, USA). ANOVA was used for analyzing multiple groups of
data or statistical differences. Results with p-value of less than 0.05 were statistically
significant.
53
Tableting punch head is a key component of the punch tool design, capable of
impacting tableting outcomes especially during high speed tableting. Modifications
to the punch head configuration could potentially affect dwell time along with other
associated compaction parameters. Thus, it was of interest to investigate the
influence of punch head flat and head radius configurational changes on the
compaction process and the mechanical properties of the tablets produced.
Paracetamol and lactose based formulations were used in this study. Both
formulations have been reported to be susceptible to tablet failure in the form of
capping, particularly at high tableting speeds. The elastic character of the
formulations could provide information pertaining to changes in plasticity/elasticity
balance in the tablets.
Lactose granules (Tablettose 80, Meggle Pharma, Wasserburg, Germany) were used
as provided, without further treatment. Granules containing paracetamol (Rhodadap
Dense Powder, Rhodia Wuxi Pharmaceutical, Jiangsu, China) and potato starch
(Roquette, Lestram, France) (77.9:22.1) were produced using a high shear processor
(UltimaPro 10L; GEA Collette, Wommelgam, Belgium), based on a formulation
utilized by Sarkar et al in a study on capping tendency [122]. A 15.2 %, w/w starch
paste, which was used as a binder in the later stages, was prepared by adding potato
starch to water. Dry mixing of paracetamol powder and potato starch (19.9 %, w/w)
in the high shear processor was carried out for 2 min at an impeller speed of 450 rpm
54
prior to starch paste addition and the mixture wet massed for an additional 5 min
with the impeller and chopper speeds set at 450 rpm and 2800 rpm, respectively. The
moistened granules produced were passed through a cone mill with square-hole
screen of aperture size 6350 µm (Comil 197S, Idex-Quadro Engineering, Waterloo,
Canada) and impeller speed of 1240 rpm. Granules were dried in a fluidized bed
drier (Strea-1, GEA-Aeromatic, Bubendorf, Switzerland) with inlet air temperature
of 45°C and drying was terminated when product temperature reached 40°C. The
dried granules were re-granulated through the cone mill with round-hole screen of
aperture size 1143 µm and impeller speed 1350 rpm. Final tableting blends were
formed by mixing lactose and paracetamol-starch granules separately with 1 %, w/w
magnesium stearate (M125, Productos Metalest, Zaragoza, Spain) in a polyethylene
bag for 1 min, prior to tableting.
3.2.4 Characterization of tableting material
The particle size of paracetamol-starch and lactose granules was measured using the
laser diffraction particle sizer as described in Section 3.1.3.2.
3.2.4.2 Hausner ratio
The Hausner ratio of paracetamol-starch and lactose granules was obtained from the
measurement of bulk and tapped densities of the powder blends as described in
Section 3.1.3.3.
3.2.4.3 Angle of repose
The flow characteristics of the paracetamol-starch and lactose powders were
evaluated by measuring the angle of repose as described in Section 3.1.3.4.
55
The compression behavior of the paracetamol-starch and lactose granules was
studied using the Heckel plots as described in Section 3.1.3.1.
3.2.4.5 Stress relaxation test
The general viscoelastic behavior of the paracetamol-starch and lactose granules
were determined from the stress relaxation test as described in Section 3.1.6.
3.2.5.1 Tableting punch head configuration
Five different types of 10 mm flat-face bevel edge (FFBE) punches that could be
classified into two main punch head configurations – with and without a head flat –
were used for this study [130]. The specifications of the punch head configurations
are given in Fig. 10. The punches differed mainly in terms of their head flat and head
radius specifications. Three punches with head flats - extended head flat (EHF),
standard head flat (SHF) and reduced (half) head flat (RHF) - were used. The
remaining two punch head configurations completely lacked the head flat with one
having a standard radius head (SRH) and the other having a reduced radius head
(RRH). To accommodate the changes to the head flat and head radius specifications,
the punch head thickness was modified accordingly to keep the outside diameter of
the head (25.57 mm) and overall punch length (133.60 mm) consistent with tooling
standards. All other punch specifications adhered to EU B-type tooling standards.
56
A B C
HF HR HF HR HF HR
HT HT HT
HF 13.34 HF 10.41 HF 5.21

HR 16.00 HR 16.00 HR 18.00
HT 2.89 HT 3.18 HT 3.15
D E
HR HR
HT HT
HR 33.00 HR 27.43
HT 2.51 HT 3.08
Figure 10. Schematics showing the dimensions (in mm) of the EHF (A), SHF (B),
RHF (C), SRH (D) and RRH (E) punch head configurations in terms of their head
flat (HF), head radius (HR) and head thickness (HT) specifications. (EHF: extended
head flat, SHF: standard head flat, RHF: reduced head flat, SRH: standard radius
head, RRH: reduced radius head)
57
3.2.5.2 Rotary tableting
The rotary tablet press (R190FT; GEA-Courtoy, Halle, Belgium) used in this study
consists of a turret fitted with 14 tableting stations. A schematic diagram
representing the interior of the rotary tablet press is shown in Fig. 11. The rotary
press is capable of two compaction events – precompaction and main compaction.
The precompaction event consists of a compaction roll of diameter 150 mm with an
air compensator mounted over the precompaction roll. The main compaction event
is fitted with a larger compaction roll of diameter 240 mm, without an air
compensator system connected to the roll. The compaction roll control the extent of
upper/lower punch penetration into the die. Maximum upper punch displacement at
the main compaction event could be varied between 1.8-4.0 mm, while upper punch
displacement at the precompaction event was fixed at 3.25 mm. The lower punch
displacement varies with the amount of tableting material in the die and the desired
compaction force. Main compaction and precompaction forces were determined by
the set compact thickness (distance between the upper and lower punch tips) at the
respective events. A reduced set compact thickness value for a particular punch face
configuration would lead to application of a higher compaction load on the powder
bed.
Adjustment of the compaction parameters in the rotary press enabled deactivation of
the precompaction event, allowing both compaction rolls to mimic main compaction
event. This enabled the compaction of each tablet by single compaction cycles using
either the compaction roll of diameter 240 mm (D240) or the smaller compaction
roll of diameter 150 mm (D150), with the upper punch displacement fixed at 3.25
mm at both compaction events (Fig. 12).
58
Main compaction
stage
Ejection
stage
Precompaction
stage
Die filling
stage
Figure 11. Schematic showing the rotary press turret with positions of the die filling,
precompaction, main compaction and ejection stages.
(i) (ii)
A
Precompaction Main compaction
roll roll
B
Compaction force
C Time
Compaction force
Time
Figure 12. Schematic diagrams depicting the (i) motion of the punch at
precompaction event and (ii) motion of the punch at main compaction event (A).
Graphical representation of the force-time curve at the main compaction event, with
precompaction deactivated (B). Graphical representation of force-time curve at
precompaction event with main compaction deactivated (C).
59
When the effect of one compaction event was studied, the other compaction event
was deactivated by setting a high set compact thickness at precompaction or main
compaction roll so that the die fill passed the respective event without being
subjected to any compaction force. During the compaction of tablets using the larger
compact roll, a precompaction thickness value larger than the fill depth and a
negligible precompaction force limit was set to prevent any punch penetration into
the die at the precompaction event, led to a compaction force-time curve of only the
main compaction event as shown in Fig. 12B. Similarly, when compaction using
the smaller compaction roll was desired, a main compaction thickness value larger
than the fill depth was set, ensuring that punch penetration ended above the die or
die fill at the main compaction event, giving a force-time curve of only the
precompaction event (Fig. 12C). Tablets were produced at compaction forces
ranging between 6 and 9 kN (pressure range, 76.4-114.6 MPa), Tablet weight was
kept constant at 325 mg and tableting speed range 25-35 rpm (linear speed, 34-47
m/min) was used for compaction.
3.2.6 Compaction parameters
Compaction force-time profiles (Fig. 2B) were captured by the in-built software
CDAAS (Courtoy Data Acquisition and Analysis System). Contact time (Ttotal; time
for compaction and decompression excluding ejection time), dwell time (Tdwell; time
during which compaction force is above 90 % of its peak value), consolidation time
(Tcon; time from start of compression to dwell phase), decompression time (Tdec; time
from the end of the dwell phase to end of the compaction cycle) and ejection force
were recorded from the profiles. The rate of force application (RFA) represents the
strain rate during consolidation phase (Fig. 2), was calculated using MATLAB
(R2010a; The MathWorks, MA, USA) [122]. Gaussian distribution equation (Eq.
60
(12)) was used to fit the data using the least-square error algorithm (r2 close to 1). At
the inflexion point, the second-derivative of the equation is equal to 0. By solving
d2y/dx2 = 0, a tangent (Eq. (13)) was drawn at the inflexion point (x0, y0). The slope
of the tangent, m, calculated from Eq. (14) represents the RFA during the
consolidation phase.
𝑥−𝑏 2
𝑦 = 𝑎𝑒 −( 𝑐
)
(12)
2
𝑥0 −𝑏 −(𝑥0−𝑏)
𝑦 = −2𝑎 𝑐2
𝑒 𝑐 (𝑥 − 𝑥0) + 𝑦0 (13)
𝑥0 −𝑏 −(𝑥0−𝑏)2
𝑚 = −2𝑎 𝑒 𝑐 (14)
𝑐2
where a, b and c are constants.
3.2.7 Evaluation of compact deformation
The extent of compact deformation during the dwell phase was quantified by
calculating the area quotient (AQ) for the compaction profiles using Eq. (15). AQ
indicates the extent of compaction force decay during the dwell phase, caused by
stress relaxation of the compact under constant strain [34, 35].
𝐴2
AQ = 𝐴1
(15)
The dwell phase was identified from the compaction profile and divided into halves
as shown in Fig. 2B. The areas under the force time curve were calculated for each
half (A1 and A2) using the graphing software (v9.1, OriginLab, MA, USA). Both
areas were decreased by the rectangles under the minimum compaction force in the
dwell time. The ratio of the spatial areas (A2/A1) provides the AQ (Eq. (15)) which
is a sensitive and informative parameter that quantifies deformation behavior. Lower
AQ values are indicative of greater plastic flow during dwell phase.
61
3.2.8.1 Tablet elastic recovery
Tablet elastic recovery (ER) was calculated using Eq. (16) [128]. Tablet height after
decompression process (Hdcp) was measured by an out-of-die method using an
electronic micrometer (Series 293, Mitutoyo, Kawasaki, Japan) while the tablet
height at target compaction force (Hcp) was the set compaction thickness by adjusting
the position of the respective compaction rolls to achieve the required force level
with respect to the tablet weight.
𝐻dcp − 𝐻cp
ER (%) = 𝐻cp
× 100 % (16)
3.2.8.2 Tablet porosity
Porosity of the tablets was calculated using Eq. (7). Densities of compacts at each
specified compaction force were calculated using the tablet dimensions upon tablet
ejection. The true density of the granules was determined using a helium pycnometer
(Pentapycnometer, Quantachrome Instruments, FL, USA).
Breaking force of tablets was measured using a hardness tester (TBF 1000; Copley,
Nottingham, UK). Tensile strength measurement of paracetamol-starch and lactose
tablets were calculated using Eq. (2). Tablet diameter and thickness 24 hours post
compaction were measured using a digital micrometer (Series 293; Mitutoyo,
Kawasaki, Japan). Tabletability (Eq. (5)) was quantified using the compactibility
coefficient (Cp) which is a dimensionless compaction parameter calculated from the
slope of the tensile strength-compaction force plots [15]. Tabletability describes the
effectiveness of the applied force in increasing tablet tensile strength and provides
an insight into the compaction efficiency of the tooling.
62
3.2.8.4 Capping index
Capping index was calculated using Eq. (4) that was proposed by Akseli et al, as
described in Section 1.2.2 [59]. Due to difficulties in batching product output from
a rapid tableting process run, batching of samples were based on a fixed production
duration (15 mins). The quantity of tablets produced and the number of tablets that
capped off the press during tablet production and hardness testing (TBF 1000,
Copley, Nottingham, UK) were recorded as inputs for Eq. (4).
3.2.9 Computational simulation of tablet compaction
Tablet compaction was simulated using Finite Element Modeling (FEM). The
Drucker-Prager Cap model was implemented in the FEM software (Abaqus, Ansys,
PA, USA). The use of Drucker-Prager Cap model for numerical studies on powder
compaction has been discussed under Section 1.4.1. The yield function was defined
by three surfaces represented in Fig. 13: shear failure surface Fs, elliptical surface
(cap surface) Fc and the transition surface Ft [103]. The evolution of the cap surface
was described with the hardening function Pb which is the position of the cap on
hydrostatic pressure axis for each density state. For a uniaxial cylindrical die
compaction test, the hydrostatic pressure stress (p) and Mises equivalent stress (q)
were expressed as:
1
𝑝 = − 3 ( 𝜎𝑧 + 2𝜎𝑟 ) (17)
𝑞 = |𝜎𝑧 − 𝜎𝑟 | (18)
where σz and σr are the axial and radial stresses, respectively. Six parameters were
required to define the yield surface of the Drucker-Prager Cap model (β, d, pa, R, pb
and a). In order to identify these parameters, diametrical and uniaxial compression
(Fig. 14) tests were carried out [98].
63
Mises equivalent stress q
Transition surface, Ft
Shear failure, Fs α(d + patan β)
d + pa tan β
Cap, Fc
β
d R(d + pa tan β)
pa pb
Equivalent hydrostatic pressure stress p
Figure 13. Drucker-Prager Cap model with associated parameters.
Figure 14. Schematic showing the uniaxial compression test for material property
calibration.
64
The friction angle (β) and cohesion (d) are needed to define the Drucker-Prager Cap
shear failure surface; the cap eccentricity parameter (R) and evolution (pa) are
required to define the cap surface, and pb as a function of the volumetric plastic strain
required to define the cap hardening/softening law; a was required to define the
transition surface
Material parameter identification for the Drucker-Prager Cap model:
Cohesion and internal friction angle were expressed as:
𝑓 𝑓
𝜎𝑐 𝜎𝑑 (√13−2)
𝑑= 𝑓 𝑓 (19)
𝜎𝑐 + 2𝜎𝑑
𝑓
3(𝜎𝑐 +𝑑)
𝛽 = tan−1 [ 𝑓 ] (20)
𝜎𝑐
where σfd is the radial tensile strength of tablets measured from the diametrical
compression test, and σfc is the axial tensile strength of compacts measured from the
uniaxial compression test. Radial and axial tensile strengths of the tablets were
determined using Eq. (21) and Eq. (22) respectively.
𝑓 2𝐹
𝜎𝑑 = 𝜋𝐷𝑡
(21)
𝑓 4 𝐹𝑦
𝜎𝑐 = 𝜋𝐷 2
(22)
where F is the breaking force obtained from the diametrical compression test and Fy
is the axial crushing force obtained from the uniaxial compression test, using the
universal testing machine. D is the tablet diameter and t is the tablet thickness. a is
a small number (typically 0.01 – 0.05), used to define a smooth transition surface
between the shear failure surface and the cap, was set to 0.02 [103]. Material
parameter R, that controls the shape of the cap, is given as:
65
2 (1+𝛼−𝛼/ cos 𝛽)2
𝑅= √ 3𝑞𝐴
(𝑝𝐴 − 𝑝𝑎 ) (23)
Consequently, the evolution parameter pa and pb were obtained from Eq. (24) and
Eq. (25) respectively.
[ 3𝑞𝐴 +4𝑑 tan 𝛽(1+𝛼−𝛼/ cos 𝛽)2 ]

𝑝𝑎 = − 4 [ (1+𝛼−𝛼/ cos 𝛽) tan 𝛽 ]2
+
2 + 24𝑑𝑞 (1+𝛼−𝛼/ cos 𝛽)2 tan 𝛽+8(3𝑝 𝑞 +2𝑞 2 (1+𝛼−𝛼/ cos 𝛽) tan 𝛽 ]2
√9𝑞𝐴 𝐴 𝐴 𝐴 𝐴 )[
4 [ (1+𝛼−𝛼/ cos 𝛽) tan 𝛽 ]2

(24)
𝑝𝑏 = 𝑝𝑎 (1 + 𝑅 tan 𝛽) + 𝑅𝑑 (25)
significant.
66
Tableting performance of punches may however differ with different punch face
configurations. The edges of cylindrical tablets are prone to damage as the edges,
produced by the right-angle meeting two large surfaces (cylindrical and flat
surfaces), tend to contain more loosely pack particles. Thus, friability of these tablets
is higher. The solution adopted by tool manufacturers to overcome this deficit is to
provide bevel edge flat punches. Modification of the angled bevel edge geometry to
a curved radial surface leads to the formation of the flat-face radius edge punch face
configuration, a relatively new design proposed as an alternative to bevel edge
punches. Investigation of the influence of such edge geometry modifications of the
flat-face punches on compaction process and tablet properties could provide useful
information for tablet and tooling manufacturers.
Lactose and paracetamol-starch granules described in Part 2 of the study, were used
in this part of the study as well. Tableting blends were prepared following the
procedure described under Section 3.2.3.
3.3.3.1 Tableting punch face configurations
Two types of flat-face punch face configurations with different edge geometries,
with standard punch head, were used for this part of the study (Fig. 15 and Table 1)
[131]. The flat-face bevel edge (FFBE) face configuration consists of an angled
straight edge between the flat portion and the land of the punch tip, with the bevel at
an angle of 30° (Fig. 15A).
67
A
0.4
8.22
10.0 30°
0.4
7.09
10.0
R2.50
Figure 15. Schematics showing the dimensions (in mm) of the FFBE (A) and FFRE
(B) punch face configurations. (FFBE: flat-face bevel edge, FFRE: flat-face radius
edge)
Table 1. Specifications of FFBE and FFRE punch face configurations (FFBE: flat-
face bevel edge, FFRE: flat-face radius edge)
Punch Cup Cup Cup Working

face depth volume area length
3 2
configuration (mm) (mm ) (mm ) (mm)
FFBE 0.4 26.03 81.47 133.2

FFRE 0.4 25.10 80.75 133.2
68
The flat-face radius edge (FFRE) face configuration (Fig. 15B) consists of a
curvature with a radius (R) of 2.50 mm between the flat portion and the land. In
order to ensure that the radius does not exceed the 30° bevel used for the FFBE
configuration, the flat area for FFRE configuration was reduced accordingly, causing
the cup area to be reduced marginally. The punch face specifications are stated in
Table 1. All other punch specifications adhered to EU B-type tooling standards.
3.3.3.2 Rotary tablet production
Tablets were compacted using single or double compaction cycles. In single
compaction cycles, only the main compaction event (compaction roll of diameter
240 mm) was used, with the precompaction event deactivated by setting a high
precompaction set compact thickness and low precompaction force limit, which
disabled the air compensator mounted over the precompaction roll and led to a large
compaction stroke distance between the top and bottom punches so that the die fill
passed the event without being subjected to any compaction force. This method was
described earlier under Section 3.2.5.2. The main compaction force was determined
by the set compact thickness (ST main) at main compression roll. A smaller STmain
value for a particular punch face configuration led to greater compaction force on
the compact. In the double compaction cycles, both main and precompaction events
were used. Two different tableting modes (with and without the air compensator
operating) were utilized at the precompaction event (Fig. 16-17). When the air
compensator was disabled (mode1), the precompaction event acted like the main
compaction event (without floating compaction roll) as shown in Fig. 16, with the
precompaction force determined by the set compaction thickness (STpre) at
precompaction event. The precompaction force was set to 3 kN.
69
A (i) (ii)
PC roll MC roll
B
Compaction force
Main compaction
curve
Precompaction
curve
Time
(PC) event (i) with air compensator deactivated (mode1) and (ii) motion of the punch
at main compaction (MC) event (A). Graphical representation of force-time curve at
precompaction event with air compensator deactivated (mode1) and the force time-
time curve at the main compaction event (B).
A (i) (ii)
PC roll
MC roll
B
Roll displacement
Compaction force
Main compaction
Roll displacement curve
curve
Time
event (i) with air compensator activated (mode2) and (ii) motion of the punch at main
compaction event (A). Graphical representation of force-time curve at
precompaction event with air compensator activated (mode2) with the associated roll
displacement curve and the force-time curve at the main compaction event (B).
70
In the other precompaction mode (mode2), the air compensator was activated by
preseting a force limit (3 kN). When the precompaction force exceeded the force
limit, the air compensator floated the roll, causing it to displace upwards (Fig. 17),
and thereby maintaining a relatively constant applied force during the dwell phase,
with the dwell time extended accordingly.
3.3.4 Compaction energy
The amount of work done by the punches during compaction was calculated from
the compaction force-punch separation profiles for each flat-face punch
configuration [84]. Punch separation was approximated by calculating the vertical
punch displacement (z) at time t using Eq. (26).
1
z = [(𝑟1 + 𝑟2 )2 − (𝑟3 sin 𝑤𝑡 − 𝑥2 ) 2 ]2 (26)
where r1 and r2 are the radii of the compaction roll and punch head curvature
respectively, w is the angular velocity of the turret with a radius of r3, and x2 is the
radius of the punch head flat, described in Section 1.1.3.3.3 [28]. Net work of
compaction (Wnet) and work of expansion (Wexp) were obtained from the calculation
of specific areas under the force-punch separation curve using the graphing software
(v9.1, OriginLab, MA, USA) [80]. The ratio of work of expansion and compaction
provides the compaction quotient (Eq. (27)) which quantifies the change in elastic
recovery relative to the change in the degree of interparticulate bonding.

𝑊𝑒𝑥𝑝
𝐶𝑜𝑚𝑝𝑎𝑐𝑡𝑖𝑜𝑛 𝑞𝑢𝑜𝑡𝑖𝑒𝑛𝑡 = 𝑊𝑛𝑒𝑡
(27)
71
3.3.5.1 Tablet elastic recovery
Elastic recovery of the tablets produced from the different punch face geometries
were calculated as described in Section 3.2.8.1.
3.3.5.2 Tablet porosity
Porosities of tablets produced from the different punch face geometries were
calculated as described in Section 3.2.8.2.
Tensile strength of tablets produced from the different punch face geometries were
calculated as described in Section 3.2.8.3.
3.3.5.4 Capping index
Capping tendency of tablets produced from the different punch face geometries were
evaluated as described in Section 3.2.8.4.
3.3.6 Computational simulation of tablet compaction
Discrete Element Modelling (DEM) was used as a tool for examining the effect of
punch face configuration on particle compaction within the tablet in this part of the
study. DEM is capable of incorporating particle physical displacement into
simulation, likely to occur with punch concavity modifications. FEM, which was
utilized in Part 2 of the study for studying the influence of dwell time on compact
stress state, treats the simulated powder as a continuum mass, less suitable for
particulate study. The software EDEM 2.4 (DEM Solutions Ltd., Edinburgh, UK)
was used for the simulations. Particles were simulated as spheres of diameter 0.5
mm, where the contact forces were calculated using the Hysteretic Spring contact
model in EDEM for its ability to simulate elastic and plastic deformation behavior
72
[132]. The material properties used for the simulated paracetamol-starch
particles/tableting tools are as follows: density = 1347/7667 kg/m3, shear modulus =
1.0/79.3 GPa, yield pressure = 166/1830 MPa and Poisson’s ratio = 0.3/0.3. For
simplicity, the wall properties of the tooling and particles’ surface were set the same
in the simulations. Coefficients of restitution and static friction were set at 0.1 and
0.3, respectively. In order to acquire reasonable computation times, the resistance
of the particles to compaction was reduced to 0.001 times the actual resistance
(computational time = 30 ms). This reduced the interaction forces and friction forces
between the particles. However, the relative densities and particle deformation
behavior remained the same as if the original forces had been used. This approach
has been employed in a number of previous studies and found to accurately capture
the dynamics of compaction process [117].
Particle deformation behavior is significantly affected by particle size. To ensure
that the deformation behavior of simulated particles was stable and reliable, the ratio
of compaction load to total particle surface area was kept constant. Compaction
pressures for simulation purpose were calculated using experimental and simulated
particle sizes. Simulated tablet thickness at target compaction pressure and rate of
force application (RFA) were recorded as validation parameters.
3.3.6.1 Compact stress analysis
Stress distribution within the simulated tablet matrix at the target compaction force
was approximated by the simulations. Reduction of the acquisition time to construct
the stress data was enabled by using a grid bin group consisting of a suitable number
of bins at the tablet (Y-Z) cross-section. The average particle stress for each bin was
calculated. Stress distribution data was exported to a graphical software (v9.1,
OriginLab, Northampton, MA, USA) to construct contour plots representing relative
stress distribution. A calibration bar which relates the scale to the apparent stress is
73
added to allow simple understanding and comparison of the different stress
conditions in the tablets.
significant.
74
adhesion tendency to the punch surface during powder compaction
Punch face modification could possibly alter stress concentration factors at the
punch face-compact boundary and impact powder adhesion tendency during powder
compaction. Investigation of the influence of the different face configurations of flat-
face punches on the extent of powder sticking during tablet compaction could
provide valuable information for tablet and tooling manufacturers.
Ibuprofen (Hubei Granules-Biocause Pharmaceutical, Hubei, China) was used as a
model active ingredient in this study. Ibuprofen is a popular choice in powder
adhesion studies, due to its poor compaction properties and high tendency for
adhesion to compaction tools [62]. The high adhesion propensity of ibuprofen
particles has been attributed to high particle surface energy and relatively lower
melting point in comparison to other APIs such as paracetamol. Ibuprofen particle
adherence to punch surfaces is reportedly sensitive to compaction speed and force
[63]. Compressible lactose granules (Tablettose 80, Meggle Pharma, Wasserburg,
Germany) was used as the filler based on its adequate flow properties, superior
compactibility and minimal contribution to punch adhesion.
A direct compaction tableting blend, consisting of ibuprofen (70 %, w/w), lactose
(28.5 to 29 %, w/w) and magnesium stearate (1-1.5 %, w/w) (M125; Productos
Metalest, Zaragoza, Spain), was prepared. Ibuprofen and lactose mixture were
premixed manually in a polyethylene bag for 10 min. The required amount of
75
magnesium stearate was added and the formulation blended for a further 1 min, prior
to tableting.
3.4.4 Characterization of tableting blends
The particle size and size distribution of ibuprofen and lactose particles were
evaluated as described in Section 3.1.3.2.
3.4.4.2 Angle of repose measurement
The angle of repose of ibuprofen-lactose powder blend was measured as described
in Section 3.1.3.4.
3.4.5.1 Punch face configuration
Three types of 10 mm flat-face punches with different face configurations were used
for tableting in this study. Two of the experimented punch face configurations, flat-
face radius edge (FFRE) and flat-face bevel edge (FFBE) were used for tablet
compaction in Section 3.3.3.1. The FFRE configuration consists of 7.09 mm flat
circle and a curvature with a radius (R) of 2.50 mm between the flat circle and the
land. The flat-face bevel edge punch face consists of 8.22 mm flat circle and bevel
straight edge inclined at an angle of 30º between the flat circle and the land (Fig.
15). The cup volume and area for edged geometries have been reported in Table 1.
The remaining punch face configuration, flat-face plain (FFP) lacks a concavity and
cup volume with punch face area 78.5 mm2.
3.4.5.2 Tablet production using manual press
Two compaction methods were employed for tablet production. The first
compaction method involved a semi-automated manual press (NP-RD10A, Natoli,
76
St Charles, MO, USA). The manual press allows uniaxial compaction of tablets, with
a stationary lower punch. A photograph of the semi-automated manual press is
shown in Fig. 18. The hand press consists of a hydraulic system which is used to
control the vertical upper punch movement to achieve the desired compaction force,
monitored via the force sensor connected to the lower punch. Once the desired input
compaction force is reached, the upper punch withdraws automatically. Tableting
material (325 mg) was weighed out and transferred manually into the die for each
tablet preparation. Batches of tablets amounting to 30 or 50 tablets were produced
at compaction forces ranging between 5 and 10 kN. Triplicates were obtained for
each tableting batch at the various compaction forces and magnesium stearate
concentration levels. The use of the manual press for tablet compaction provides
information on the effect of uniaxial compaction and time dependent compaction
factors on powder adhesion.
Hydraulic cylinder
Ejection
lever
Upper punch
assembly Manual
pump
Force and fill

depth meter
Figure 18. Manual press used for tablet production.
77
3.4.5.3 Tablet production using compaction simulator
The second compaction method utilized an instrumented compaction simulator
(STYL’ONE EVOLUTION; MedelPharm, Beynost, France) for biaxial powder
compaction. A photograph of the tooling setup and the force feeder is shown in Fig.
19.
Upper
punch
assembly
Lower punch
and die
assembly
Force
feeder
Take-off
force sensor
Figure 19. Tooling setup (A) and the force feeder with the take-off force sensor
(B) in the compaction simulator.
78
The compaction simulator consists of a single station with upper and lower punches
capable of vertical displacement. Force and displacement sensors fitted at both the
upper and lower punch positions allowed capturing of force data and punch
movement during the time course of each compaction cycle. The compaction
simulator was programmed to compact the powder using single compaction cycle.
A feeder system was utilized to transfer the tableting material into the die during
tablet production. Batches of 50 tablets were produced at compaction forces ranging
between 5 and 10 kN. The use of the compaction simulator allowed the compaction
cycle to be shortened to milliseconds range, representative of rotary tablet
compaction.
3.4.6 Assessment of powder adherence
3.4.6.1 Quantification of powder adherence
Powder adhesion during compaction using the manual press was quantified via
spectroscopy analysis of the amount of ibuprofen adhered to punch face [87]. After
each batch of tablet production, the upper and lower punches were carefully
dismantled from the press. Powder adhered to the peripheral surfaces of the punch
tip was carefully removed using a standardized cleaning procedure, with minimal
disturbance to powder adherence state on the punch face. The punch tip was
immersed in 5 mL of 96 %, v/v ethanol to allow dissolution of ibuprofen adhered to
the punch face. The receiving solution was then analyzed by ultraviolet-visible
spectroscopy at 264 nm using a diode array spectrophotometer (U-5100, Hitachi
High-Tech Science Corporation, Tokyo, Japan). Absorption was converted to
concentration using a separately constructed standard curve. Overall, the linear range
was obtained at 0.2-0.8 absorbance, corresponding to 145-600 µg/mL concentration
of ibuprofen. The amount of ibuprofen adhered onto each punch face was recorded
79
as amount of ibuprofen adhered per unit area (µg/mm2) to remove confounding
impact due to differences in surface area. The amount of adhered residual powder
on the upper punch face during tablet production using the compaction simulator
was analyzed using spectroscopy procedure as described above. Powder adherence
on the lower punches were evaluated in terms specific force parameters measured,
discussed in the subsequent sections.
3.4.6.2 Force parameters related to powder adherence
Powder adhesion during tablet production using the compaction simulator was based
on measurement of ejection force and take-off force at the lower punch for each
tablet [133]. Ejection force is a measure of the force required to overcome the
residual die wall pressure and the frictional conditions at the interphase between the
tablet surface and the die wall during the compact travel to the die surface (Fig. 20A).
Ejection force may provide valuable information related to material-die wall
sticking.
Take-off force is the force required to detach the formed tablet off the lower punch
face after the tablet has been ejected from the die, a parameter that evaluates shear
stress during desorption of tablet from the surface of the lower punch (Fig. 20B)
[75]. Take-off force is generally considered as the most direct method to quantify
powder adhesion on the lower punch. Take-off force was measured by the take-off
piezoelectric sensor, positioned on the scraper bar located on the feeder as shown in
Fig. 19B, while the ejection force was determined by the lower punch force sensor.
Ejection and take-off force values for each tablet, over the time course of tablet
production, were individually captured by the in-built software ANALIS
(MedelPharm, Beynost, France).
80
A
Fr Tablet Fr
Fe
Lower Die
punch
Ft
B Fa
Figure 20. Schematics showing the stress factors associated with tablet ejection (A)
and take-off (B). (Fe: ejection force, Ft: take-off force, Fa: adhesive force, Fr:
compact-die wall force)
significant.
81
CHAPTER 4
RESULTS AND
DISCUSSION
82
4. RESULTS AND DISCUSSION
materials during dwell and the consequent effect on tablet mechanical
properties
4.1.1 Physical characteristics of tableting materials
Particle size and flow properties of the tableting materials (MCC, DCP and APAP)
are presented in Table 2. The particle size of MCC and APAP were significantly
smaller than DCP particles, as shown by the d50 values. DCP powder exhibited
superior flowability with low Hausner ratio and angle of repose (AOR) values.
Table 2. Physical characteristics of microcrystalline cellulose (MCC), dicalcium

phosphate (DCP) and paracetamol (APAP) particles
Tableting d50 Span Hausner AOR

material (µm) ratio (°)
MCC 56.3 (4.1) 1.31 (0.07) 1.55 (0.02) 43.0 (0.7)

DCP 164.6 (3.8) 1.13 (0.08) 1.19 (0.02) 30.6 (1.4)
APAP 48.2 (4.1) 1.16 (0.02) 1.65 (0.05) 54.9 (0.4)
Parenthesized values represent the respective standard deviation.
4.1.2 Compression behavior of tableting materials
Heckel analysis was utilized to identify the predominant deformation behavior of
the tableting materials (Fig. 21). Heckel plots provide information about the
compression behavior of substances used and allow some empirical classification of
them into basic densification mechanisms. The intercept of the vertical axis, via
extrapolation of the linear region is related to the apparent density of the powder.
While it is difficult to accurately distinguish between elastic and plastic deformation
83
components from the slopes of the linear portion of the Heckel plots, the steepness
of the slopes provides a relative comparison of the deformation nature of the
materials at high pressures. Linear section with a high slope value represents
predominant plasticity during densification. Brittle fracturing and elastically
deforming materials generally exhibit low linear slopes.
2.5
2
Ln (1/e)
1.5
0.5
0
0 50 100 150
Compression pressure (MPa)
Figure 21. Heckel plots for microcrystalline cellulose (♦), dicalcium phosphate (■)
and paracetamol (●) particles
Heckel curves for APAP, DCP and MCC are present in Fig. 21. Densification
behavior for the tableting materials largely conformed to the expected trends.
Compression of DCP showed an initial curve region, indicative of a fast volume
reduction process dominated by particle rearrangement and fragmentation. The
subsequent long quasi-linear portion with a less steep slope indicated prevailing
plastic flow, in combination with elastic deformation. After initial brittle fracturing
of the DCP particles at low pressures, the stress necessary to cause further fracture
of the smaller particles increased. As a result, the mechanism of volume reduction
84
evolved from fragmentation to elastic deformation, reflected in the reduced rate of
porosity change in the linear region. Extrapolation of linear portion suggested the
highest degree of packing in the die for the DCP powder. Compression behavior of
MCC particles was distinctively different as shown by the Heckel plot with linear
curve and steep slope. The much greater slope of the Heckel plot for MCC suggested
a higher degree of plasticity and easier densification at the higher applied pressures.
Plastic densification behavior of the MCC particles has been attributed to the
numerous slip planes and dislocations present in the structure, with hydrogen
bonding between deformed particles [134]. Heckel curve of APAP particles showed
an initial region of reducing compact porosity with increasing compression pressure,
subsequent linear region began at a relatively higher compression pressure (delayed
onset) with the lowest slope. This demonstrated a predominance of elastic
deformation in the APAP particles. The high resistance of the APAP particles to
deformation has been attributed to energetically unstable movement of molecular
slip planes under external mechanical stress, resulting in reduced compliance to
shear stress and rate of densification [135-137].
4.1.3 Compact deformation during dwell
Deformation of material under compressive force during the dwell phase differed
from other stages of the compaction cycle, Heckel analysis may not be suitable for
dwell phase deformation characterization. Therefore, stress relaxation curves and
parameters related to plastic flow were used in the evaluation of viscoelastic
deformation of the compacts during dwell. Measured force observed from the stress
relaxation profile is a direct consequence of the resistance exerted by the material
under compaction, dependent on the physical properties of the substances which may
undergo time-dependent volumetric changes. The process of compaction force
85
decay within the compact during dwell could be divided into two distinct phases.
During the consolidation stage, rearrangement of particles causes the formation of a
compact comprising certain number of pore sites. The first phase begins immediately
after the compact has reached maximum intended compaction force. During this
phase, particles tend to undergo further deformational changes and move into void
interparticulate spaces in the compact. Part of the stored elastic energy in the
compact could be released at this stage, characterized by a steep drop in force. The
second phase is a retarded deformational response of the particles, in which the
particles deformed further in a plastically manner but at a lower rate, shown by the
reduced slope within the stress profile. The onset of the second phase is indicative
of a compact microstructural state with minimal available free sites. Resistance to
movement of deformed particles increases due to high internal friction caused by
close and constrained contact between particles, consequently reducing the
possibility of plastic flow at this stage [25]. Viscoelastic slopes for the first (VS1)
and second (VS2) phases were used in conjunction with the relaxation quotient (RQ)
values for analysis of the stress relaxation profiles.
The changes to the stress relaxation process with different levels of
elasticity/plasticity were analyzed in the APAP/MCC (Fig. 22 and Table 3) and
DCP/MCC (Fig. 23 and Table 4) compact systems. Increasing the amount of MCC
in the APAP/MCC compact system increased the overall extent of force reduction
and the rate of relaxation (Fig. 22). As expected, the viscoelastic slope of the second
phase was lower than the first phase as evidenced by the VS values in Table 3. The
enhancement in structural relaxation was largely attributed to the basic deformation
mechanism of the two types of tableting materials.
86
9
8.8 F1
8.6
Force (kN)
8.4
F2
8.2
F3
8
7.8 F4
F5
7.6
0 0.5 1 1.5 2
Time (s)
Figure 22. Stress relaxation profiles of APAP/MCC compacts of different

proportions. (APAP/MCC; F1: 100/0, F2: 75/25, F3: 50/50, F4: 25/75, F5: 0/100)
Table 3. Viscoelastic parameters obtained for APAP/MCC compacts of different

proportions (APAP/MCC; F1: 100/0, F2: 75/25, F3: 50/50, F4: 25/75, F5: 0/100)
Formulation Compaction RQ VS1 VS2

force (kN) (kN/ms) (kN/ms)
F1 6.0 1.03 0.28 0.04

F2 6.0 1.10 0.93 0.08
F3 6.0 1.14 1.20 0.12
F4 6.0 1.18 1.55 0.13
F5 6.0 1.21 1.73 0.16
F1 9.0 1.02 0.32 0.03

F2 9.0 1.08 1.09 0.11
F3 9.0 1.12 1.52 0.17
F4 9.0 1.15 1.80 0.20
F5 9.0 1.16 1.95 0.20
87
9
8.8
8.6
Force (kN)
F6
8.4
8.2 F7
8
F8
7.8 F9
7.6
0 0.5 1 1.5 2
Time (s)
Figure 23. Stress relaxation profiles of DCP/MCC compacts of different

proportions. (DCP/MCC; F6: 100/0, F7: 75/25, F8: 50/50, F9: 25/75)
Table 4. Viscoelastic parameters obtained for DCP/MCC compacts of different

proportions (DCP/MCC; F6: 100/0, F7: 75/25, F8: 50/50, F9: 25/75)

F6 6.0 1.06 0.56 0.06

F7 6.0 1.12 1.10 0.08
F8 6.0 1.16 1.40 0.07
F9 6.0 1.20 1.59 0.10
F6 9.0 1.05 0.76 0.08

F7 9.0 1.11 1.37 0.10
F8 9.0 1.13 1.69 0.16
F9 9.0 1.16 1.92 0.17
88
APAP particles, majority of which were elastically deformed at the end of the
compaction cycle, reacted to the punch stop by an instantaneous elastic response by
trying to recover their original shape volume and had impeded particle movement
during dwell. Presence of MCC particles in the system brought about a more pliable
compact and allowed irreversible deformation at the beginning of the relaxation
process, thus storing much lower elasticity in the compact and hence reduced the
tendency of volume expansion during dwell. This favorably allowed further
deformational changes to the particles and facilitated movement of deformed
particles into interstitial spaces, resulting in increased plasticity.
While the introduction of plastically deforming component into the compact system
increased the overall structural relaxation in the APAP/MCC compacts, the degree
of stress relief was observed to be disproportional to MCC content, as clearly shown
by the stress relaxation profiles (Fig. 22). Initial addition of MCC into the compact
system (APAP/MCC, F2) led to a force decay of about 10 %. However, the extent
of structural relaxation was not consistent with subsequent increases in MCC
concentration; marginal differences in stress decay between APAP/MCC compacts
containing 75-100 % MCC (APAP/MCC, F4 and F5). The results could be explained
by porosity dependent secondary process that occurred in the compact
microstructure. Plastically deforming substances lead to remarkably lower porosities
in the compact than other materials with different deformation mechanisms, at
equivalent compaction loads. As a consequence of reduced internal porosity and
strain hardening characteristic of highly plastic MCC particles, resistance to
particulate displacement into interstitial spaces increased due to higher cohesiveness
and friction between particles [25]. Plastic flow was negatively affected as a result,
as reflected in the APAP/MCC stress profiles.
89
Replacement of elastically deforming APAP with brittle fracturing DCP particles,
exhibited a similar trend to the APAP/MCC systems, albeit with some differences.
In comparison to the APAP/MCC compacts (Tablet 3), presence of DCP particles
allowed a greater total stress decay (higher RQ values), at equivalent concentrations
of MCC particles in the system (Table 4). This was attributed to the reduced
deformational resistance of DCP particles under compaction load which imparted
lesser amount of elastic energy into the compact system than APAP particles. The
greater structural relaxation could have also been aided by the deformation behavior
of DCP which allowed particles to be in closer proximity with formation of a more
densely compacted coherent microstructure. This probably facilitated uniform
energy distribution throughout the compact, thus minimized localized high stress
concentrations and reduced the amount of elasticity in the compact. The enhanced
structural relaxation of DCP/MCC compacts was again observed in the viscoelastic
slopes of the first phase. VS1 values of DCP/MCC compacts (Table 4) were higher
than that of corresponding APAP/MCC compacts (Table 3). Interestingly, the
viscoelastic slopes for the subsequent phase (VS2), was noticeably lower when
compared to equivalent APAP/MCC systems, indicating reduced availability of
interstitial spaces possibly due to rapid movement of fractured particles during the
earlier phase coupled with increased interparticulate friction at the subsequent phase
in the DCP/MCC compact system.
The introduction of DCP to the compact system containing elastically deforming
APAP enhanced the structural relaxation of the compact; elastic energy decreased
via plastic flow (Fig. 24 and Table 5).
90
9
8.8
F10
8.6 F11
F12
Force (kN)
8.4
8.2
7.8
7.6
0 0.5 1 1.5 2
Time (s)
Figure 24. Stress relaxation profiles of APAP/DCP compacts of different

proportions. (APAP/DCP; F10: 75/25, F11: 50/50, F12: 25/75)
Table 5. Viscoelastic parameters obtained for APAP/DCP compacts of different

proportions (APAP/DCP; F10: 75/25, F11: 50/50, F12: 25/75)

F10 6.0 1.04 0.37 0.05

F11 6.0 1.05 0.43 0.05
F12 6.0 1.06 0.54 0.04
F10 9.0 1.03 0.45 0.06

F11 9.0 1.04 0.64 0.07
F12 9.0 1.05 0.64 0.08
91
The total amount of stress relaxation increased marginally as the concentration of
DCP increased (Fig. 24). As expected, the extent of stress relief was much reduced
in comparison to the APAP/MCC and DCP/MCC compact systems. This was
attributed to the substantial elasticity in the system by the deformation of APAP and
fractured DCP particles during the consolidation stage. As the DCP particle size
decreased due to fragmentation, yield pressure and stress necessary to cause
deformation of the particles increased and contributed to high elasticity. In addition,
the internal area of contact due to reduction in particle size within the tablet
increased. This contributed to the cohesiveness and friction within the tablet
microstructure, restricting particulate movement during dwell. These combined
effects limited structural relaxation to a large extent in the APAP/DCP compact
system, in comparison to compact systems containing MCC particles.
4.1.4 Effect of formulation variables on tablet properties
The changes to the mechanical properties of tablets containing MCC (APAP/MCC,
DCP/MCC) with respect to formulation changes are depicted in Tables 6-9. Tablet
strength is a function of the bonding created by material deformation and bond
disruptive events that occur upon punch withdrawal. Tablet tensile strength
increased significantly with corresponding increases in MCC concentration in the
system, as observed in APAP/MCC (Tables 6-7) and DCP/MCC (Tables 8-9)
compacts. The relatively lower yield pressure of MCC particles allowed high degree
of plastic deformation which greatly increased the number and area of contact points
between particles within the tablet microstructure, a prerequisite for interparticulate
bonding. The capability of MCC to form strong hydrogen bonds also contributed
greatly towards the improved compact tensile strength [124].
92
APAP/MCC tablets of different proportions, compacted at 6 kN (APAP/MCC; F1:
100/0, F2: 75/25, F3: 50/50, F4: 25/75, F5: 0/100)
Formulation Compaction Dwell RQ Tensile VER

force time strength
(kN) (ms) (MPa) (%)
F1 6.0 0 1.00 0.08 (0.01) 0.24 (0.06)

100 1.00 0.06 (0.03) 0.11 (0.10)
500 1.02 0.08 (0.01) 0.20 (0.06)
1000 1.02 0.08 (0.01) 0.29 (0.06)
2000 1.03 0.08 (0.09) 0.26 (0.05)
F2 6.0 0 1.00 0.26 (0.01) 0.35 (0.11)

100 1.01 0.28 (0.03) 0.37 (0.10)
500 1.06 0.33 (0.02) 0.39 (0.03)
1000 1.08 0.31 (0.01) 0.36(0.10)
2000 1.10 0.36 (0.01) 0.33 (0.10)
F3 6.0 0 1.00 0.89 (0.06) 0.44 (0.07)

100 1.02 1.16 (0.09) 0.38 (0.03)
500 1.08 1.17 (0.05) 0.36 (0.20)
1000 1.11 1.26 (0.05) 0.42 (0.20)
2000 1.14 1.38 (0.03) 0.47 (0.08)
F4 6.0 0 1.00 2.00 (0.06) 0.42 (0.06)

100 1.02 2.16 (0.02) 0.48 (0.11)
500 1.10 2.13 (0.03) 0.59 (0.05)
1000 1.14 2.14 (0.1) 0.59 (0.09)
2000 1.18 2.55 (0.04) 0.50 (0.20)
F5 6.0 0 1.00 3.38 (0.10) 0.64 (0.16)

100 1.02 3.79 (0.06) 0.77 (0.07)
500 1.12 3.80 (0.07) 0.63 (0.07)
1000 1.16 3.68 (0.10) 0.58 (0.12)
2000 1.21 4.27 (0.20) 0.65 (0.30)
93
APAP/MCC tablets of different proportions, compacted at 9 kN (APAP/MCC; F1:
100/0, F2: 75/25, F3: 50/50, F4: 25/75, F5: 0/100)

force time strength
(kN) (ms) (MPa) (%)
F1 9.0 0 1.00 0.14 (0.02) 0.11 (0.07)

100 1.00 0.13 (0.01) 0.12 (0.01)
500 1.01 0.13 (0.01) 0.16 (0.02)
1000 1.02 0.13 (0.02) 0.11 (0.03)
2000 1.02 0.13 (0.09) 0.26 (0.01)
F2 9.0 0 1.00 0.52 (0.03) 0.18 (0.12)

100 1.01 0.51 (0.01) 0.27 (0.10)
500 1.05 0.53 (0.01) 0.30 (0.13)
1000 1.06 0.52 (0.03) 0.18 (0.06)
2000 1.08 0.53 (0.02) 0.26 (0.07)
F3 9.0 0 1.00 1.30 (0.09) 0.38 (0.13)

100 1.01 1.54 (0.10) 0.28 (0.20)
500 1.07 1.57 (0.01) 0.40 (0.16)
1000 1.09 1.56 (0.01) 0.40 (0.13)
2000 1.12 1.72 (0.11) 0.41 (0.13)
F4 9.0 0 1.00 3.09 (0.06) 0.38 (0.12)

100 1.02 3.28 (0.05) 0.33 (0.13)
500 1.08 3.22 (0.11) 0.38 (0.10)
1000 1.12 3.23 (0.05) 0.33 (0.04)
2000 1.15 3.60 (0.11) 0.38 (0.13)
F5 9.0 0 1.00 5.50 (0.08) 0.52 (0.11)

100 1.02 5.44 (0.09) 0.60 (0.05)
500 1.09 5.60 (0.06) 0.57 (0.12)
1000 1.13 5.58 (0.05) 0.60 (0.09)
2000 1.16 5.61 (0.06) 0.50 (0.03)
94
DCP/MCC tablets of different proportions, compacted at 6 kN (DCP/MCC; F6:
100/0, F7: 75/25, F8: 50/50, F9: 25/75)

force time strength
(kN) (ms) (MPa) (%)
F6 6 0 1.00 0.11 (0.01) 0.26 (0.04)

100 1.01 0.11 (0.01) 0.32 (0.06)
500 1.04 0.13 (0.01) 0.26 (0.11)
1000 1.05 0.10 (0.00) 0.28 (0.15)
2000 1.06 0.12 (0.01) 0.30 (0.09)
F7 6 0 1.00 0.41 (0.02) 0.43 (0.05)

100 1.01 0.48 (0.03) 0.37 (0.12)
500 1.07 0.51 (0.02) 0.38 (0.10)
1000 1.09 0.50 (0.02) 0.43 (0.23)
2000 1.12 0.52 (0.04) 0.42 (0.11)
F8 6 0 1.00 1.15 (0.11) 0.48 (0.22)

100 1.02 1.38 (0.03) 0.61 (0.20)
500 1.09 1.36 (0.05) 0.62 (0.18)
1000 1.13 1.37 (0.04) 0.70 (0.14)
2000 1.16 1.53 (0.04) 0.47 (0.20)
F9 6 0 1.00 2.29 (0.08) 0.53 (0.22)

100 1.02 2.50 (0.05) 0.66 (0.11)
500 1.10 2.54 (0.04) 0.61 (0.13)
1000 1.14 2.48 (0.10) 0.57 (0.13)
2000 1.20 2.76 (0.07) 0.68 (0.04)
95
DCP/MCC tablets of different proportions, compacted at 9 kN (DCP/MCC; F6:
100/0, F7: 75/25, F8: 50/50, F9: 25/75)

force time strength
(kN) (ms) (MPa) (%)
F6 9 0 1.00 0.20 (0.01) 0.25 (0.05)

100 1.01 0.19 (0.01) 0.24 (0.13)
500 1.03 0.19 (0.01) 0.25 (0.06)
1000 1.05 0.21 (0.01) 0.23 (0.03)
2000 1.05 0.21 (0.00) 0.22 (0.06)
F7 9 0 1.00 0.76 (0.05) 0.38 (0.08)

100 1.01 0.80 (0.06) 0.27 (0.13)
500 1.06 0.87 (0.08) 0.36 (0.16)
1000 1.08 0.86 (0.02) 0.39 (0.08)
2000 1.11 0.87 (0.04) 0.37 (0.11)
F8 9 0 1.00 2.04 (0.04) 0.40 (0.07)

100 1.01 2.15 (0.02) 0.47 (0.03)
500 1.07 2.12 (0.05) 0.67 (0.07)
1000 1.10 2.12 (0.03) 0.64 (0.13)
2000 1.13 2.41 (0.10) 0.44 (0.07)
F9 9 0 1.00 3.40 (0.09) 0.44 (0.14)

100 1.01 3.61 (0.10) 0.43 (0.14)
500 1.08 3.63 (0.04) 0.37 (0.03)
1000 1.12 3.60 (0.06) 0.51 (0.32)
2000 1.16 4.05 (0.09) 0.60 (0.06)
96
Enhanced densification of powder bed led to a less restrained transmission of the
applied compaction energy and allowed greater dissipation of the built-up elastic
energy in the compact. This reduced the degree of elastic expansion of compact
during decompression which contributed to preserving the bonds formed during the
compaction cycle. Comparing the compacts formed by the two MCC containing
systems, the presence of a fragmenting material in the compact helped to form
mechanically stronger tablets at equivalent MCC concentrations, largely attributed
to the deformation mechanism of DCP particles. Fragmentation of DCP particles
with subsequent percolation of broken particles combined with occupation of voids
by plastically deformed MCC particles led to the optimal state of force utilization
and enhanced interparticulate bonding quality. Deformation of DCP particles
possibly facilitated greater dissipation of energy within the compact and reduced the
level of residual elasticity in the compact, as explained earlier.
Influence of DCP on tablet strength in the presence of APAP particles was rather
marginal, as shown by the tensile strength values in Tables 10-11. Initial introduction
of DCP (APAP/DCP, F10) to the system containing paracetamol only (F1) caused
noticeable increases in tablet strength (p < 0.05), attributed to creation of new
bonding surface and reduced elasticity conferred by the deformation of DCP
particles, as explained earlier. Interestingly, subsequent increases in the proportion
of DCP particles in the compact system (APAP/DCP, F11-F12) resulted in
insignificant changes to tablet tensile strength. This could be explained by the
differences in the bonding strength and utilization of compaction energy. Despite the
increase in effective contact area for bonding, the strength of the bonds formed was
relatively weak. The amount of elasticity from the deformation of APAP particles
was substantial, causing extensive rupture of the particle-particle bonds.
97
APAP/DCP tablets of different proportions, compacted at 6 kN (APAP/DCP; F10:
75/25, F11: 50/50, F12: 25/75)

force time strength
(kN) (ms) (MPa) (%)
F10 6 0 1.00 0.10 (0.01) 0.30 (0.04))

100 1.01 0.09 (0.00) 0.21 (0.08)
500 1.02 0.10 (0.01) 0.29 (0.10)
1000 1.03 0.10 (0.01) 0.26 (0.10)
2000 1.04 0.10 (0.00) 0.26 (0.05)
F11 6 0 1.00 0.11 (0.00) 0.19 (0.09)

100 1.00 0.11 (0.00) 0.23 (0.07)
500 1.03 0.11 (0.01) 0.33 (0.09)
1000 1.04 0.11 (0.01) 0.23 (0.08)
2000 1.05 0.11 (0.01) 0.20 (0.09)
F12 6 0 1.00 0.11 (0.00) 0.27 (0.05)

100 1.01 0.10 (0.01) 0.19 (0.02)
500 1.03 0.10 (0.01) 0.29 (0.08)
1000 1.04 0.10 (0.00) 0.24 (0.12)
2000 1.06 0.10 (0.01) 0.32 (0.15)
98
APAP/DCP tablets of different proportions, compacted at 9 kN (APAP/DCP; F10:
75/25, F11: 50/50, F12: 25/75)

force time strength
(kN) (ms) (MPa) (%)
F10 9 0 1.00 0.15 (0.01) 0.14 (0.09)

100 1.00 0.16 (0.01) 0.29 (0.07)
500 1.02 0.16 (0.01) 0.17 (0.04)
1000 1.03 0.16 (0.01) 0.19 (0.07)
2000 1.03 0.15 (0.01) 0.18 (0.11)
F11 9 0 1.00 0.18 (0.01) 0.26 (0.11)

100 1.01 0.17 (0.00) 0.22 (0.09)
500 1.03 0.17 (0.01) 0.24 (0.09)
1000 1.04 0.17 (0.01) 0.28 (0.04)
2000 1.04 0.17 (0.01) 0.25 (0.11)
F12 9 0 1.00 0.17 (0.00) 0.29 (0.04)

100 1.01 0.18 (0.01) 0.25 (0.09)
500 1.03 0.17 (0.01) 0.30 (0.10)
1000 1.04 0.17 (0.00) 0.27 (0.03)
2000 1.05 0.17 (0.01) 0.27 (0.08)
99
The effect of formulation variation on post compaction tablet strain recovery was
studied by comparing the percent viscoelastic recovery of tablets. The total tablet
elastic expansion may be defined as the sum of elastic recovery that occurs axially
immediately upon removal of the upper punch while still being constrained by the
die wall, and post compaction viscoelastic recovery that occurs during storage.
Tablets with MCC component showed distinct time-dependent viscoelastic recovery
(Tables 6-9). The percent viscoelastic recovery showed a clear trend in relation to
the proportion of MCC in the compact system; viscoelastic recovery generally
increased at higher concentrations of MCC in the system (p < 0.05). Plastic
deformation of MCC under applied constant pressure neutralized most of the stored
elastic energy and maintained high degree of bonding, preventing immediate release
of elastic energy upon punch withdrawal and facilitated viscoelastic strain recovery.
The use of a higher compaction force (9 kN) caused noticeable reductions in post-
compaction tablet expansion in both the APAP/MCC (Tables 6-7, F2-F5) and
DCP/MCC (Tables 8-9, F7-F9) compacts (p < 0.05). As the compaction load
increased, the degree of particulate deformation increased, with concomitant
increases in accumulated elasticity in the compact. However, the increase in stored
elastic energy was probably released as immediate tablet expansion (elastic
recovery) and resulted in lower energy store for viscoelastic tablet expansion [128].
In contrast to the post compaction recovery behavior of compacts containing MCC,
the physical specifications of DCP/APAP compacts remained relatively stable
throughout the storage period with consistent VER values observed (Table 10-11).
4.1.5 Effect of dwell on tablet properties
The effect of dwell time extension was observable in compact systems containing
MCC as one of the components; mechanical quality of tablets generally improved
100
with longer exposure to constant strain (dwell phase). The effect was more
noticeable in APAP/MCC (Tables 6-7) and DCP/MCC (Tables 8-9) compact
systems with 25-75 % MCC content (p < 0.05). Clearly, the increase in dwell time
caused significant improvements to the tablet mechanical quality by facilitating the
neutralization of stored elastic energy by plastic deformation during constant strain.
In consequence, reduced elastic energy contained in the compacts before the start of
punch withdrawal during the decompression phase ensured that strain recovery was
also reduced with less bond disruptive events and strengthened the tablets. These
results provided further evidence that the reaction of compacted solid to stress relief
was related to the densification mechanism of the solid.
Interestingly, it was noticed that increases in plastic flow along the time course of
constant strain did not lead to improvements in tablet strength in a proportional
manner, particularly at higher concentrations of MCC in the system (Tables 6-9, F2-
4, F7-9). Enhanced plastic flow (increasing RQ values) during dwell time 100-1000
ms very often did not cause significant changes to tablet tensile strength. In contrast,
appreciable increases in tablet strength was observed after the end of the second
phase (dwell time 2000 ms), emphasizing the time dependency of bond
consolidation in compacts showing viscoelastic character. Continuous particulate
deformation and movement during the initial phase of the relaxation process could
not provide sufficient time for building up of new interparticulate bonds. The
particulate system was in a state of flux, and without permanence, interparticulate
bonds could not be successfully formed. During the second phase of stress
relaxation, the rate of compact deformation slowed down with much reduced
movement of particles and this scenario provided conditions for greater
opportunities for fusion along with bond reinforcement. Tablet viscoelastic recovery
101
was observed to be largely independent of changes in dwell time, despite observable
changes in extent of plastic flow as indicated by the RQ values. Compact
composition and the applied compaction force were concluded to be more influential
factors in impacting viscoelastic recovery. Dwell time extension possibly impacted
immediate tablet recovery to a greater extent than time-dependent viscoelastic
recovery.
4.1.6 Summary
In this study, the influence of different deformation mechanisms of tableting
materials on plastic flow during the dwell phase and the consequent effect on tablet
mechanical properties were investigated. Plastic flow during dwell was observed in
the formulation systems investigated, at different magnitudes and rates of stress
decay. The process of stress relaxation within the compact during the time course of
constant strain showed two distinct phases. The first phase was related to particulate
deformation and movement into interstitial voids while the second phase was related
to reduced rate of plastic flow contributed by resistance to particulate movement due
to high internal friction between particles.
Presence of highly plastic material (MCC) in the compact system increased the
overall extent of force reduction and the rate of relaxation significantly. However,
the degree of stress relief during the different phases was observed to adopt a non-
linear relation with MCC content, attributed to porosity-dependent particulate
deformation. Extensive plastic deformation of MCC particles led to low internal
porosity and increased resistance to particulate movement which negatively affected
the rate of structural relaxation. The effect of dwell time extension was observable
in compact systems containing MCC; mechanical strength of tablets generally
increased with longer exposure to constant strain (dwell phase). Enhanced plastic
102
flow during the first phase of stress decay very often did not translate into tablet
strength enhancement. In contrast, appreciable increase in tablet strength was
observed after the end of second phase, emphasizing the time dependency of bond
consolidation in compacts showing viscoelastic character. The results demonstrated
the influence of formulation variation on dwell sensitivity and the limitations of
dwell extensions in different compact systems. The findings could be useful for
formulation optimization for the subsequent experiments involving dwell
modification.
103
4.2.1 Physical characteristics of tableting materials
Particle size and flow properties of the paracetamol-starch and lactose granules are
presented in Table 12. The differences in the median particle size for both types of
granules were marginal with paracetamol-starch granules exhibiting a wider size
distribution as shown by the larger span. Flow properties of both powders,
represented by the Hausner ratio and angle of repose could be considered
comparable; both powders possessed acceptable flowability suitable for high speed
tableting.
Table 12. Physical characteristics of paracetamol-starch and lactose granules
Tableting d50 Span Hausner AOR

material (µm) ratio (°)
Paracetamol -
starch 174.2 (3.80) 2.86 (0.02) 1.27 (0.01) 34.9 (1.6)
Lactose 168.3 (13.4) 1.76 (0.16) 1.28 (0.01) 33.5 (0.6)
4.2.2 Compression behavior of tableting materials
Two types of tableting materials (paracetamol-starch and lactose) that differed in
terms of deformation mechanism were used in this study. The Heckel slope and the
extent of stress relaxation were used as non-specific test parameters to provide some
general description on the deformation mechanism of the tableting materials during
the compaction cycle and the period of constant strain as shown in Fig. 25.
104
A
2.05
1.9
1.75
ln (1/e)
1.6
1.45
1.3
1.15
1
0 20 40 60 80 100 120 140 160
Compression pressure (MPa)
B
9
8.9
Force (kN)
8.8
8.7
8.6
8.5
8.4
0 100 200 300 400 500
Time (ms)
Figure 25. Heckel plots (A) and stress relaxation curves (B) for lactose (  ) and
paracetamol-starch (  ) granules.
105
The Heckel plots for both materials showed an initial curve region, indicating
particle rearrangement and different degrees of brittle fracture, followed by long
quasi-linear portions pointing to prevailing plastic flow, in combination with elastic
deformation. While it is difficult to accurately distinguish between elastic and plastic
deformation components from the slopes of the linear portion of the Heckel plots,
the steepness of the slopes provides a relative comparison of the deformation nature
of the materials at high pressures. The steeper slope of the Heckel plot for lactose
(Fig. 25A) suggested a greater degree of plasticity and easier densification at the
applied pressures. In comparison to the tableting materials used in Part 1,
paracetamol-starch and lactose granules exhibited an overall deformation
mechanism intermediate of APAP and MCC particles (Fig. 21). The stress relaxation
curves (Fig. 25B) characterized the time-dependent viscoelastic behaviors of the
tableting materials. The rate of stress decay of the tested materials resembled MCC
containing compact systems used in Part 1. It was noted that, in comparison to
paracetamol-starch compacts, the force recorded for the lactose compacts decreased
more quickly under constant strain at the compaction forces used, particularly for
the initial 100 ms of the stress profile. This suggested that lactose compacts
presented a more pronounced viscoelastic character leading to greater structural
relaxation than paracetamol-starch compacts, attributed to the development of
reduced elasticity in the lactose compacts.
4.2.3 Influence of punch head configuration on compaction profile
The compaction parameters and profiles obtained for the different configurations of
punch head at various compaction forces are presented in Table 13 and Fig. 26. The
changes in the compaction cycle in relation to the punch head configurations were
found to be similar for the compaction of both lactose and paracetamol-starch
106
tablets, thus only the parameters and compaction profiles obtained from the
compaction of lactose tablets are presented. Results on the compaction parameters
were found to differ greatly between punch heads with and without a physical head
flat. Therefore, it was necessary for the results to be discussed in accordance to these
two classifications of punch head configurations.

using the larger compaction roll (D240)
Punch Compaction RFA Tcon Tdwell Tdec Ttotal

head force
configuration (kN) (daN/ms) (ms) (ms) (ms) (ms)
EHF 6.0 26.9 (0.5) 31.8 (0.2) 27.2 (0.2) 19.0 (0.1) 78.0 (0.0)
SHF 26.5 (0.6) 33.5 (0.5) 22.8 (0.1) 17.0 (0.8) 74.0 (0.4)
RHF 26.1 (0.3) 33.7 (0.4) 16.9 (0.2) 14.3 (1.8) 64.7 (0.9)
SRH 26.0 (0.4) 36.8 (0.7) 12.3 (0.2) 12.0 (0.5) 61.0 (1.0)
RRH 26.5 (0.4) 33.8 (0.2) 11.6 (0.2) 11.0 (0.0) 56.0 (0.4)
EHF 7.5 31.5 (0.2) 37.0 (0.1) 26.8 (0.2) 20.0 (1.0) 85.0 (1.3)
SHF 31.6 (0.4) 39.0 (0.1) 23.3 (0.3) 17.5 (0.5) 80.0 (0.0)
RHF 30.8 (0.5) 38.7 (0.5) 17.2 (0.1) 15.7 (0.4) 69.9 (0.8)
SRH 30.8 (0.2) 39.8 (1.3) 12.5 (0.4) 13.0 (0.4) 65.3 (0.9)
RRH 31.0 (0.4) 37.5 (0.5) 12.0 (0.3) 12.3 (0.2) 62.0 (0.0)
EHF 9.0 37.2 (0.5) 37.0 (0.1) 27.7 (0.4) 18.7 (0.4) 84.0 (0.0)
SHF 37.7 (0.5) 39.0 (1.0) 22.8 (0.1) 18.0 (0.0) 80.7 (1.0)
RHF 37.0 (0.6) 38.0 (0.1) 17.2 (0.1) 14.7 (0.4) 70.0 (0.0)
SRH 37.1 (0.2) 40.0 (0.1) 12.9 (0.1) 14.0 (0.0) 66.0 (0.8)
RRH 37.8 (0.4) 40.3 (0.3) 12.6 (0.2) 14.0 (0.0) 66.0 (0.6)
107
A
10000
SHF
8000
RHF
6000
4000
2000
0
0 10 20 30 40 50 60 70 80 90
Time (ms)
B
10000
SRH
8000
RRH
6000
4000
2000
0
0 10 20 30 40 50 60 70 80 90
Time (ms)
Figure 26. Comparison of compaction profiles (6 and 9 kN) using the larger
compaction roll (D240) between SHF and RHF (A), SRH and RRH (B) head
configurations. (SHF: standard head flat, RHF: reduced head flat, SRH: standard
radius head, RRH: reduced radius head)
108
The RFA values for punches with head flat (EHF, SHF and RHF) were rather
similar, indicating marginal impact of the head radius curvature on the consolidation
phase (Table 13). A head curvature with a steeper slope (smaller radius) was
predicted to accelerate the consolidation phase and cause the dwell limit to be
reached earlier. However, despite the difference in the head radii between the SHF
(HR 16 mm) and RHF (HR 18 mm), the consolidation times for the two head
configurations did not vary significantly. EHF punches displayed shorter
consolidation time than SHF head configuration (p < 0.05) although both head
configurations shared common head radius, indicative of a quicker densification of
compact during the consolidation phase for the EHF punch head configuration.
Comparison of the compaction profiles of SHF and RHF head configurations (Fig.
26A) showed that the force-time curves at the early consolidation phase for the two
punches were almost identical, regardless of the compaction force used with peak
compaction force reached at similar time points. In contrast, alterations to the head
flat dimensions had a more pronounced effect on the compaction profiles. The dwell
time for the punches with a head flat increased proportionally with the head flat
diameter, with the EHF (HF 13.34 mm) configuration showing the longest dwell
time, followed by SHF (HF 10.41 mm) and RHF (HF 5.21 mm). This was expected,
as the time required for the compaction roll to transit across the flat portion of the
punch head has been reported to be independent of the required input force.
In contrast to the consolidation phase, the decompression phase was affected by the
punch head configuration. Decompression times were significantly different (p <
0.05), with the RHF head configuration having the most rapid release rate of
compaction energy during decompression and the EHF head configuration allowed
a more sustained release of compaction energy with a prolonged decompression time
109
(Table 13). However, the changes to the decompression phase was likely a result of
the dimensional modifications to the head flat, rather than changes to the head
curvature.
The punch head curvature was noticed to have a more pronounced effect on the
compaction parameters in the absence of a physical head flat. The results obtained
for the punches without head flat (SRH and RRH) showed slightly differing trends
(Table 13). While the RFA values were similar between them, there were noticeable
differences in the consolidation, dwell and decompression times. Due to the smaller
head radius, RRH (HR 27.43 mm) head configuration showed significantly shorter
consolidation, dwell and decompression durations than the SRH (HR 33.00 mm), at
lower compaction force levels of 6-7.5 kN. In consequence, total contact time
between the punch head and compaction roll for the RRH head configuration, was
lower by about 3-5 ms, largely contributed by the changes in the consolidation time.
However, the differences in the parameters were generally negligible at the higher
compaction force of 9 kN, with both head configurations sharing almost identical
compaction profiles as shown in Fig. 26B.
4.2.4. Influence of compaction roll configuration on compaction profile
The size of the compaction roll affects tablet compaction significantly, roll diameter
controls the rate and duration of force application as the punches diametrically
converge into the die. In view of this, the changes to compaction parameters and
profile due to reduction in compaction roll diameter were also recorded (Table 14
and Fig. 27). The use of a smaller roll caused the compaction profiles to undergo
significant changes, particularly at higher compaction forces. In comparison to the
compaction profiles from the larger roll, the smaller compaction roll accelerated
force application (increased RFA), reduced the consolidation and total contact times
110
for all the punch head configurations (p < 0.05). This considerably narrowed the
compaction cycle (Fig. 27A). The punch vertical displacement, which determines
the extent of punch penetration, was not affected by the change in roll dimensions.
Interestingly, amongst the head configurations with head flat, the compaction
profiles of EHF head configuration were least affected by the reduction in roll size
(Fig. 27B). This suggested that increasing the head flat diameter could possibly
counter the impact of roll size change. Punch head configurations with an extended
head flat could be useful to tablet manufacturers with the intention to preserve the
compaction profile when tablet presses fitted with a smaller compaction rolls are
utilized.

using the smaller compaction roll (D150)

head force
EHF 6.0 29.9 (0.4) 31.0 (0.4) 30.3 (0.4) 17.3 (0.9) 78.7 (0.5)
SHF 30.3 (0.2) 32.0 (0.0) 25.6 (0.1) 15.3 (0.4) 72.0 (0.8)
RHF 28.7 (0.8) 33.3 (0.8) 17.7 (0.2) 14.7 (0.4) 65.7 (0.9)
SRH 28.8 (0.2) 33.7 (1.3) 11.3 (0.2) 13.0 (0.4) 58.0 (1.6)
RRH 29.0 (0.2) 32.7 (0.8) 10.6 (0.1) 11.8 (0.2) 55.3 (0.9)
EHF 7.5 35.3 (0.2) 31.8 (0.7) 29.8 (0.7) 18.5 (0.5) 81.0 (1.0)
SHF 35.3 (0.2) 32.7 (0.9) 26.2 (0.1) 16.7 (1.3) 75.3 (0.9)
RHF 35.1 (0.6) 32.7 (0.3) 17.7 (0.5) 13.7 (1.2) 64.7 (0.4)
SRH 35.0 (0.4) 33.5 (0.5) 12.0 (0.0) 13.5 (0.5) 59.0 (1.0)
RRH 35.5 (0.3) 35.5 (1.0) 11.3 (0.2) 12.3 (0.2) 59.0 (1.0)
EHF 9.0 43.8 (0.8) 30.7 (0.4) 30.7 (0.5) 17.7 (1.2) 82.0 (1.1)
SHF 44.2 (0.5) 31.9 (0.5) 26.5 (0.1) 16.7 (0.4) 75.0 (0.3)
RHF 43.6 (0.4) 32.3 (0.5) 17.7 (0.4) 16.0 (0.8) 65.3 (0.9)
SRH 42.0 (0.3) 34.5 (1.0) 11.9 (0.1) 13.8 (0.7) 60.0 (0.0)
RRH 42.8 (0.2) 35.0 (1.1) 11.7 (0.1) 14.0 (0.4) 60.7 (0.9)
111
A
10000
Compression force (N)
8000
6000
4000
2000
0
0 10 20 30 40 50 60 70 80 90
Time (ms)
B
10000
8000
6000
4000
2000
0
0 10 20 30 40 50 60 70 80 90
Time (ms)
Figure 27. Comparison of compaction profiles (6 and 9 kN) between compaction

rolls of diameters 240 (─) and 150 (---) mm using the SHF (A) and EHF (B) head
configurations. (EHF: extended head flat, SHF: standard head flat)
112
Nonetheless, the consolidation phase was clearly independent of the changes in the
punch head radius even with a smaller roll as evidenced by the largely indifferent
RFA values and consolidation times of the SHF and RHF head configurations (Table
14). The results also indicated that the effects on dwell and decompression times
differed between the classes of punch head configurations. The values for dwell time
were found to increase significantly for the head configurations with a flattened head
geometry at all force levels studied; Tdwell increased by approximately 3 ms and 0.5
ms for the EHF/SHF and RHF head configurations respectively (p < 0.05) (Tables
13-14). Despite the reduction in roll size, the mounting surface of the compaction
rolls of different sizes onto the head flat was maintained thus theoretically caused
minimal changes in the time for the rolls to transit across the flatten surface of the
punch head. However, the reduced roll size caused an accelerated rise in the
compaction force during consolidation phase, causing the force limit for dwell phase
to be reached earlier and this resulted in increased dwell time (Table 14). In
comparison, SRH and RRH punch head configurations experienced a small
reduction in dwell times of about 1 ms, indicating the effects of not possessing a
physical flat top on the punch head; lack of head flat caused the roll to traverse at a
faster rate across the head surface. The EHF and SHF head configurations
experienced a reduction in the decompression time, similar to the effect on
consolidation phase, while the other head configurations were generally unaffected
when a smaller compaction roll was used.
The findings suggested that the head flat and compaction roll specifications are
influential factors that could cause observable changes to the compaction process.
The head radius of the punch head was predicted to impact the loading and unloading
rate of the consolidation and decompression phase respectively; larger head radius
113
should result in lower loading and unloading rates. The results, however, suggested
that the effect of the head radius on the compaction process was generally marginal,
if any. This is possibly due to the greater impact of the roll curvature in translating
the horizontal punch movement into axial compaction force than the much limited
punch head radius. Nonetheless, the punch radius curvature could play an important
role in increasing the tooling life of the punches. The curved surface of the punch
head makes initial contact with the compaction roll during rotary movement. A
larger head radius (reduced curvature) could possibly aid in reducing the initial stress
exerted on the punch head upon contact and potentially reduce the tendency to head
fracturing.
4.2.5 Influence of punch head configuration on tablet properties
The changes to the mechanical properties of lactose tablets with respect to the
different punch head configurations are depicted in Fig. 28. The results showed that
there was good agreement between the differences observed in the tablet mechanical
properties and the changes in the compaction profiles. Tablet tensile strength showed
a positive relationship with dwell time. The tablets compacted using the EHF head
configuration possessed the highest tensile strength, followed by the SHF and RHF
head configurations, while the tablets prepared using the SRH and RRH punches
displayed much lower tensile strengths (Fig. 28A) at equivalent compaction forces
(p < 0.05). It was observed that tabletability decreased in accordance with the change
of head configuration from EHF (Cp = 0.097) to RRH (Cp = 0.080), suggesting
reduced conversion of compaction energy into plastic energy. The amount of plastic
flow due to stress relaxation during the period of constant strain was evaluated by
the calculation of the ratio of spatial areas under the force-time profiles.
114
A
0.6
Tensile strength (MPa)

0.5
0.4
0.3
0.2
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.1
6 7.5 9
Compaction force (kN)
B
1.4
1.2
1
AQ
0.8
0.6
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.4
6 7.5 9
and area quotient (B) at the experimental compaction forces for compaction roll of
diameter 240 mm (D240). (EHF: extended head flat, SHF: standard head flat, RHF:
115
Apparently, the changes in dwell time had contributed to differences in the stress
relaxation process for the compacts that were prepared by the different punch head
configurations, as evidenced by the changes in AQ values (Fig. 28B). The ratio of
partial areas after and before the middle of dwell time indicates the extent of
irreversible deformation of the material under load at constant punch displacement.
During the dwell phase, the volume is nearly constant, resulting in a decrease in
force at the punch face in contact with the tablet surface, as a result of stress
relaxation. The AQ values are interpreted inversely as the relaxation quotient (RQ)
in Part 1, lower values indicate greater conversion of elasticity to plastic flow.
Materials with greater viscoelastic character would deform more readily during the
dwell phase causing reduction in AQ values. Stress relaxation during dwell
decreased at higher compaction forces, a trend similar to data on relaxation quotient
in Part 1. The extent of stress relaxation is largely influenced by the amount of
elasticity in the compact at the beginning of the dwell phase; greater elastic
deformation with increasing compaction load at the end of the consolidation phase
led to reduced plastic yielding during dwell phase.
By comparing the AQ values for the punch head configurations with a head flat (Fig.
28B), it could be seen that the EHF head configuration allowed the greatest amount
of plastic flow (lowest AQ value) during the dwell phase followed by SHF and RHF
head configurations. Clearly, the increase in dwell time by even a small extent
facilitated the conversion of stored elastic energy to plastic deformation during
constant strain and led to significant improvements to the tablet mechanical quality.
In consequence, reduced elastic energy contained in the compacts before the start of
punch withdrawal during the decompression phase ensured that axial strain recovery
was also reduced; decreasing tablet elastic expansion with increasing dwell time
116
(Table 15). The reduction in decompression time from EHF to RHF head
configurations (Tables 13-14) could also have contributed to tablet elastic recovery
by allowing accelerated release of stored elastic energy, facilitating porosity
expansion. Porosity expansion during decompression phase causes disruption of
interparticulate bonds formed during the consolidation and dwell phases, with
formation of microcracks which, if propagated, could translate to lower compact
mechanical strength during the diametrical compression test [14]. The plasticizing
action of increased periods of stress relaxation coupled with slower release of stored
energy reduced the porosity of the resultant tablets upon ejection, consequently
increased interparticle contact area. Thus, the lower porosity of tablets produced
using EHF head configuration (Table 15) suggested more interparticle bonds
preserved as compared to tablets produced from SHF and RHF punches, and
translated to better tablet mechanical strengths. In contrast, higher resistance to stress
relaxation was noted for the SRH and RRH head configurations, as evidenced by the
substantial increase in the AQ values (Fig. 28B). Despite the slightly longer dwell
time and lower strain rate by the SRH head configuration at lower force levels (Table
13), differences in the physical quality of the tablets and extent of plastic flow during
dwell for the SRH and RRH punches were, however, insignificant. The sharp
increase in AQ values could be attributed to the lack of a physical head flat that
resulted in the compacts experiencing extremely shortened periods of constant
strain, if any, during the compaction cycle. Thus, the compacts experienced a state
of continual deformation during compaction, possibly with constant increases in
elasticity without the opportunity for stress relaxation. These unfavorable conditions
possibly promoted the build-up of an abundance of elastically deformed sites within
the compacts.
117
Table 15. Effect of punch head configuration on lactose tablet mechanical properties
at the experimental compaction forces for compaction roll of diameter 240 mm
(D240) (EHF: extended head flat, SHF: standard head flat, RHF: reduced head flat,
SRH: standard radius head, RRH: reduced radius head)
Punch Compaction ER Porosity Ejection

head force force
configuration (kN) (%) (%) (N)
EHF 6.0 5.9 (0.2) 18.96 (0.11) 242 (10)

SHF 7.7 (0.3) 19.17 (0.01) 233 (13)
RHF 8.4 (0.2) 19.40 (0.05) 182 (11)
SRH 9.2 (0.1) 20.00 (0.09) 181 (12)
RRH 9.3 (0.1) 19.93 (0.07) 178 (10)
EHF 7.5 6.4 (0.5) 17.76 (0.08) 333 (10)

SHF 7.9 (0.3) 17.95 (0.11) 329 (12)
RHF 8.7 (0.2) 18.33 (0.08) 317 (12)
SRH 9.6 (0.2) 18.77 (0.02) 288 (7)
RRH 9.4 (0.2) 18.70 (0.16) 285 (9)
EHF 9.0 8.1 (0.4) 16.00 (0.10) 403 (27)

SHF 9.6 (0.3) 16.19 (0.17) 418 (14)
RHF 10.2 (0.1) 16.51 (0.03) 344 (16)
SRH 11.5 (0.3) 16.60 (0.09) 348 (18)
RRH 11.9 (0.4) 16.80 (0.11) 340 (14)
Rapid elastic recovery induced widespread microcracks with the consequence of
reduced tablet tensile strengths, as shown by the increase in the elastic recovery and
porosity values. The elastic response of the powder during compaction in accordance
with changing head configurations from EHF to RRH punches was also reflected in
the ejection force recorded (Table 15). Compacts prepared from punches with head
flat generally exhibited higher ejection force values than the punches with radial
head configurations (p < 0.05). Ejection force is a function of residual die wall
pressure and friction at the interphase between the compact and die wall surface.
Increased strain recovery for tablets produced with radial head punches caused
compacts to have greater axial recovery and by the Poisson effect, contracted
118
radially to a greater extent, thus leading to lower residual die wall pressure and
correspondingly, a lower ejection force [138-140].
The influence of the punch head configuration on the mechanical properties of
tablets compacted using the more elastically deforming paracetamol-starch granules
is presented in Fig. 29 and Table 16. Paracetamol-starch tablets generally
experienced reduced tensile strength (Fig. 29A) and greater elastic recovery (Table
16) compared to corresponding lactose tablets at equivalent compaction forces. This
was attributed to the excessive elastic deformation of paracetamol-starch granules
as supported by the Heckel plots and reduced tendency towards plastic flow during
dwell as illustrated by the stress relaxation curves (Fig. 25). Nonetheless, the benefit
of dwell time extension on paracetamol-starch tablet mechanical properties was
evident (Fig. 29A), a similar trend as exhibited by the lactose tablets (Fig. 28A),
albeit with some limitations. At compaction forces up to 7.5 kN, the tensile strength
predictably decreased from EHF to RRH, with negligible differences observed
between the tablets produced from SRH and RRH head configurations. However, it
was noticed that the tablets produced from EHF and SHF punches were equivalent
in terms of physical strength while the differences in tensile strength of tablets from
the other punches were insignificant at 9 kN. Tablet failure in the form of capping
was observed in the paracetamol-starch tablets. Analysis of the results on capping
index showed a similar outcome with capping tendency generally decreasing with
increasing dwell time (Table 16). The improved resistance to tablet failure with
increasing dwell time was observed again which reinforced the earlier discussion
that the reduction of stored elastic energy, indicated by the percent elastic recovery,
had been advantageous. Enhanced plastic yield led to decreased tablet porosity and
increased inter-particle contacts.
119
A
0.6
0.5
0.4
0.3
0.2
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.1
6 7.5 9
B
1.4
1.2
1
AQ
0.8
0.6
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
SHF
EHF
SHF
EHF
SHF
EHF
0.4
6 7.5 9

strength (A) and area quotient (B) at the experimental compaction forces for
compaction roll of diameter 240 mm (D240). (EHF: extended head flat, SHF:
standard head flat, RHF: reduced head flat, SRH: standard radius head, RRH:
reduced radius head)
120
mechanical properties at the experimental compaction forces for compaction roll of
diameter 240 mm (D240) (EHF: extended head flat, SHF: standard head flat, RHF:
Punch Compaction ER Capping Porosity Ejection

head force index force
EHF 6.0 11.3 (0.4) 0.8 17.66 (0.04) 94 (5)

SHF 12.6 (0.2) 1.1 18.18 (0.15) 97 (3)
RHF 12.9 (0.3) 1.3 18.65 (0.14) 85 (4)
SRH 13.4 (0.4) 1.8 19.00 (0.13) 87 (4)
RRH 13.8 (0.3) 1.7 19.16 (0.23) 89 (4)
EHF 7.5 13.0 (0.2) 1.1 16.30 (0.10) 109 (2)

SHF 13.9 (0.3) 1.5 16.70 (0.13) 108 (3)
RHF 14.7 (0.4) 1.8 17.96 (0.07) 107 (5)
SRH 15.4 (0.5) 2.0 17.33 (0.13) 106 (3)
RRH 16.0 (0.1) 2.2 17.33 (0.06) 103 (7)
EHF 9.0 14.4 (0.6) 2.0 15.84 (0.18) 118 (10)

SHF 16.4 (0.4) 2.1 16.18 (0.22) 116 (3)
RHF 17.3 (0.5) 2.7 16.38 (0.26) 114 (11)
SRH 17.5 (0.4) 2.6 16.82 (0.16) 101 (2)
RRH 17.9 (0.7) 2.7 16.83 (0.08) 104 (2)
At higher compaction forces, the effectiveness of dwell extension on capping
tendency was reduced. While tablets produced using the EHF punches showed the
least capping tendency, the differences between the capping indices of EHF and SHF
narrowed as the compaction force increased while the differences between SHF and
RHF head configurations widened. At compaction force 9 kN, the punches could be
grouped into two distinct groups by the mechanical quality of the tablets produced -
EHF/SHF and RHF/SRH/RRH. These results suggested that dwell extension may
not be as beneficial, particularly at higher compaction forces and for materials that
undergo high degree of elastic deformation. The results obtained were similar to the
outcome observed in Part 1, in which increases in dwell time did not contribute to
121
the physical strength of the tablets, attributed to insufficient time for bond
reinforcement after plastic yielding during dwell.
The AQ values for paracetamol-starch tablets (Fig. 29B) were higher than that of
lactose tablets (Fig. 28B) at equivalent compaction forces, indicating greater
resistance to plastic flow attributed to the presence of high elasticity at the end of the
consolidation phase. The AQ values as expectedly adhered to a positive relation with
compaction force, AQ increased at higher force levels. In a trend parallel to that of
lactose tablets, plastic flow during dwell was greatest in the tablets compacted using
EHF punches; AQ values increased from EHF to RHF with a significant rise in
values for the SRH and RRH punch head configurations at all compaction forces. In
relation to the tablet mechanical property data, the enhanced stress relaxation with
increased dwell time was not entirely reflected by the tensile strength and capping
index data. Tablets compacted using EHF head configuration experienced greater
stress relaxation and lower elastic expansion than tablets from the SHF head
configuration at all forces.
The enhanced plastic yielding, however, did not lead to an improved mechanical
quality of the tablets at the highest compaction force of 9 kN. A similar observation
was also made when tablets from RHF and SRH/RRH head configurations were
compared. These results could again be related to the data in Part 1, in which
consistent increases in overall stress decay experienced by the compact with
increasing dwell time was not translated into enhancement in the tablet strength.
4.2.6 Finite element analysis
FEM is an effective tool to simulate compaction mechanics of powder. In this study,
FEM was utilized to investigate the transmission of the applied load by the bevel
shaped punches and the reaction of the compacted solid to dwell time changes. The
122
mechanical behavior of the powder was modeled using the Drucker-Prager Cap
model, using material properties (β = 72.41, R = 0.622, a = 0.02, d = 0.5, Pb = 69.6)
calibrated from experiments with the instrumented die described in Section 3.2.9,
using the paracetamol-starch granules. A constant material property model was used,
as it consumed a smaller amount of computational time [108]. Symmetric
compaction was simulated, upper and lower punches moved at the same time and in
opposite directions, with compaction force fixed at 9 kN.
The stress distribution within the powder compact was produced by contour
mapping of the stress experienced by every element in the simulated compact. The
relative stress distribution at the end of different periods of dwell representative of
head flat change, before the decompression process, is shown in Fig. 30. In all cases,
the load of compaction imposed by the punches was not completely transmitted
within the compact and led to development of stress gradients throughout the height
of the simulated tablet. This heterogeneity of the stress distribution was principally
due to the friction between the tools and powder, and the influence of punch face
surface on the resolution of load through the compact. The effective stress at the top
half of the compact was observed to be generally lower than the bottom half of the
compressed powder bed indicating that the powder near the top generally relaxed
more than that at the bottom half of the compact, consistent with the results reported
in existing studies [98, 103].
Examination of the stress distribution revealed development of maximum stresses in
the regions adjacent to the angled surface of the punch face. The degree of stress was
almost constant along the regions in contact with the flat portion of the punch face.
123
A B
C D
S, mises (relative)
E
9.7
8.9
8.1
7.3
6.5
5.7
5.0
4.2
3.4
2.5
1.8
1.0
Figure 30. Stress distribution contours of simulated paracetamol-starch tablets

compacted at dwell times representative of RRH (A), SRH (B), RHF (C), SHF (D)
and EHF (E) punch head configurations.
Stress intensity rose when approaching the angled surface of the punch face followed
by the development of an intensive shear band that ran from the top edge towards
the mid-center of the compact. Formation of shear bands has been suggested to
facilitate crack propagation during the decompression phase during which a part of
the tablet is free to dilate axially and other parts affected by the die wall effect,
contributing to capping [109]. This clearly showed the great influence of the punch
form on stress distribution in convex tablets.
124
Stress distribution within the compact, which corresponded to the development of
elastic energy, was observed to react significantly to changes in dwell duration.
Strain deformation during the consolidation phase led to the accumulation of stored
energy. The introduction of shortened dwell phases, representative of RRH and SRH
punch head configurations, allowed certain regions of the compacts to undergo
minimal stress relief (Fig. 30A-B). Extensive stress gradients continued to exist
between the central regions and the regions adjacent to the punch face. Further
extension of the dwell phase led to significant change in the stress distribution and
promoted reduction in the intensity of the stress gradient and high stress regions,
particularly in the central regions of the compact. This indicated different rates of
stress relief in different regions of the compact. Plastic flow under constant strain in
the compact was observed to initiate from the central regions of the compact, with
the shear band from the peripheral regions still preserved. The total compact stress
at punch face level was distributed more equally at extended dwell times, thus
reduced local shear intensity experienced by the particles in the compact and reduced
stress gradients between the different regions in the compacts. At the end of the
dwell phase allowed by the EHF head geometry, it could be seen that the regions of
high accumulated stress were localized at the regions opposite the beveled surface
of the punch face. Clearly, the extension of time for viscoelastic deformation reduced
the overall amount of stored energy in the compact.
While the numerical data proved valuable in studying the stress state in the
compacts, the presented simulation model was unable to differentiate between the
different deformation mechanisms e.g. plastic and brittle fracturing, of the tableting
materials. More localized bands may occur in some types of materials which could
not be predicted by the simulation. The powder mass in the simulation was treated
125
as a continuum mass with uniform bulk density throughout the compact, which is
often not the case after die filling, particularly during high speed tableting. Powder
compaction studies have shown that material properties of powders can vary
substantially during the compaction process itself due to changes in the relative
densities within the powder bed. Numerical predictions of force transmission could
also be enhanced by considering the inter-particle cohesive strength, which is not
straight forward and further investigations will be required to the implication of such
particle properties used in the simulation system [109]. Nonetheless, the presented
FEM model was sufficient for the investigation of dwell phase compact deformation.
4.2.7 Influence of compaction roll configuration on tablet properties
The changes to the mechanical properties of lactose tablets with respect to the
smaller compaction roll size are depicted in Fig. 31 and Table 17. In comparison to
corresponding tablets compacted using the bigger roll (Fig. 28 and Table 15) at
equivalent compaction forces, the use of the smaller compaction roll caused the
tensile strengths (Fig. 31A) and elastic recovery (Table 17) of the tablets to decrease
and increase significantly (p < 0.05), respectively, indicative of increased elastic
content within the compact. Increased strain recovery resulted in greater tablet
porosity expansion (bond disruptive event) and consequently led to mechanically
weaker resultant tablets. The effect of the change in compaction roll size was also
noticed in the ejection forces; reduction in roll size led to reduced ejection force,
explained by the increased post compaction axial recovery during decompression
phase. Nonetheless, the overall benefit of dwell time extension even with reduced
roll dimension was again evident.
126
A 0.6

0.5
0.4
0.3
0.2
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.1
6 7.5 9
B 1.4
1.2
AQ
0.8
0.6
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.4
6 7.5 9
and area quotient (B) at the experimental compaction forces for compaction roll of
127
Table 17. Effect of punch head configuration on lactose tablet mechanical properties
at the experimental compaction forces for compaction roll of diameter 150 mm
(D150) (EHF: extended head flat, SHF: standard head flat, RHF: reduced head flat,
SRH: standard radius head, RRH: reduced radius head)
Punch Compaction ER Porosity Ejection

head force force
EHF 6.0 6.1 (0.1) 20.47 (0.08) 175 (12)

SHF 8.1 (0.2) 20.58 (0.09) 183 (7)
RHF 8.6 (0.1) 20.89 (0.11) 147 (5)
SRH 8.9 (0.1) 21.13 (0.12) 130 (10)
RRH 9.1 (0.3) 21.32 (0.11) 141 (17)
EHF 7.5 7.2 (0.1) 18.18 (0.07) 230 (11)

SHF 9.4 (0.2) 18.39 (0.08) 241 (7)
RHF 9.8 (0.2) 19.10 (0.16) 222 (9)
SRH 10.3 (0.1) 18.94 (0.09) 206 (5)
RRH 10.9 (0.3) 19.06 (0.06) 209 (10)
EHF 9.0 9.8 (0.2) 16.92 (0.09) 316 (6)

SHF 11.9 (0.3) 17.12 (0.03) 327 (12)
RHF 12.1 (0.4) 17.21 (0.07) 280 (7)
SRH 13.1 (0.3) 17.70 (0.16) 236 (15)
RRH 13.3 (0.4) 17.84 (0.10) 242 (15)
Tablet mechanical quality improved from RRH to EHF head configuration at all
experimented force levels with EHF punches still possessing the highest tabletability
properties. While the shift to a smaller roll size led to weaker tablets, the impact on
compact deformation during dwell phase was observed to be more positive (Fig.
31B). The AQ values for the punches with flat head geometries decreased noticeably
due to increased dwell time (p < 0.05). This suggested that the particles underwent
enhanced stress relaxation within the confines of the die cavity when a smaller roll
was used for compaction purpose. EHF and SHF punches experienced increased
plastic yielding as evidenced by the greater reduction in the AQ values than RHF
punch head configuration due to the greater increase in dwell time for EHF and SHF
128
punches than RHF punches (Table. 17). The contrasting outcome on tablet
mechanical properties despite greater plasticity development during dwell phase is
attributed to the changes in the densification during consolidation phase. The use of
a smaller roll for tablet compaction caused the strain rate to increase dramatically
during the consolidation phase, increased elastic content in the materials under stress
and contributed to increased air entrapment.
By comparing the data obtained from lactose and paracetamol-starch tablets, it could
be seen that the use of a smaller compaction roll altered the mechanical properties
of the paracetamol-starch tablets similarly but with some differences. The
mechanical quality of the paracetamol-starch tablets, as expected, deteriorated with
the use of a smaller roll and this could be assigned to the rapid elastic expansion
upon punch withdrawal (Fig. 32A). Interestingly, the positive impact of dwell time
extension on tablet mechanical strength also declined with a smaller roll.
While the tablets compressed using EHF punches continued to exhibit the highest
mechanical quality, tensile strength values of the tablets produced from SHF and
RHF punches were statistically indifferent but of higher values than those compacted
using SRH/RRH punches at compaction forces of 6-7.5 kN. At compaction force 9
kN, tensile strength of tablets produced using the EHF and SHF punches were
equivalent, despite observable differences in the extent of elastic recovery between
the two punches (Table 18), and only marginal differences existed between the
tablets from RHF and SRH/RRH punches, a similar trend observed in paracetamol-
starch tablets compacted using the bigger roll.
129
A 0.4
Tesnile strength (MPa)
0.3
0.2 RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.1
6 7.5 9
B 1.4
1.2
1
AQ
0.8
0.6
RRH
RRH
RRH
SRH
SRH
SRH
RHF
RHF
RHF
EHF
SHF
EHF
SHF
EHF
SHF
0.4
6 7.5 9

strength (A) and area quotient (B) at the experimental compaction forces for
compaction roll of diameter 150 mm (D150). (EHF: extended head flat, SHF:
standard head flat, RHF: reduced head flat, SRH: standard radius head, RRH:
reduced radius head)
130
mechanical properties at the experimental compaction forces for compaction roll of
Punch Compaction ER Capping Porosity Ejection

head force index force
EHF 6.0 11.4 (0.3) 1.2 18.30 (0.24) 66 (6)

SHF 13.5 (0.2) 1.3 18.94 (0.13) 64 (10)
RHF 13.6 (0.3) 1.8 18.99 (0.11) 76 (13)
SRH 14.3 (0.4) 2.2 19.67 (0.18) 64 (8)
RRH 14.1 (0.3) 2.2 19.41 (0.08) 67 (11)
EHF 7.5 13.5 (0.6) 1.5 17.20 (0.26) 108 (3)

SHF 15.6 (0.2) 1.8 17.36 (0.25) 86 (12)
RHF 16.5 (0.3) 2.1 17.98 (0.20) 82 (10)
SRH 17.8 (0.8) 2.5 18.24 (0.26) 68 (8)
RRH 17.1 (0.5) 2.7 18.25 (0.17) 69 (9)
EHF 9.0 16.5 (0.2) 2.2 16.28 (0.27) 111 (9)

SHF 18.1 (0.3) 2.3 16.48 (0.08) 97 (13)
RHF 20.0 (0.3) 2.8 16.73 (0.15) 89 (10)
SRH 19.5 (0.1) 2.9 17.00 (0.05) 87 (4)
RRH 19.8 (0.1) 3.0 17.08 (0.10) 90 (2)
Capping indices increased for all punch configurations at equivalent compaction
forces with the smaller roll and this can be attributed to increased elasticity (Table.
18). However, the inter-punch trend for capping tendency differed from the data
obtained using the bigger compaction roll. The differences in capping indices for the
tablets from EHF and SHF were generally smaller when compared to the capping
index values obtained from the larger roll. A distinct difference in capping index
values between RHF and SRH/RRH punch head configurations existed until
compaction force 7.5 kN. At 9 kN, the differences in capping index between RHF
and SRH/RRH punches could be considered marginal. This consistent plateauing of
the tablet mechanical properties at higher force levels of the paracetamol-starch
131
tablets compacted by either of the roll sizes suggested the presence of different levels
of elasticity thresholds in the compact system under the experimented tableting
conditions.
4.2.8 Effect of tableting speed
Manipulation of the rotary tableting speed affects the time of contact between the
compaction roll and punch head, impacting the compaction cycle as a result.
Tableting speed and dwell time follow an inverse relation; slowing the turret speed
increases the dwell time as it takes more time for the roll to transit across the head
flat. In view of this, the relationship between tableting speed, punch head geometry
and the resultant tablet properties was also evaluated (Tables 19-20). Lactose tablets
were produced using the larger compaction roll with punch configurations having
head flats at a higher turret speed of 35 rpm (47 m/min). The increased tableting
speed narrowed the differences in compaction profiles, with increased RFA
accompanied by a drastic reduction in dwell time by about 30 % (Table 19). The
force-time curves during dwell time became more symmetrical, resulting in
significant increases in the AQ values indicative of lesser overall stress relaxation
during dwell time (p < 0.05) (Table 20). As a result, the tablets compressed using
the higher turret speed displayed reduced tensile strengths. Contrastingly, tablet
elastic recovery decreased at equivalent compaction forces. This could be explained
by the higher compact porosities, shown by the increased set compaction thickness
values obtained at the higher tableting speed. As the punch vertical velocity
increased, the rate of load application provided less time for powder consolidation
and air exhaustion, resulting in compacts of increased porosities [78, 81]. These
effects reduced the degree of interparticulate bonding in the tablets and therefore,
tabletability decreased at higher tableting speeds.
132
Table 19. Effect of punch head configuration on compaction parameters of lactose
tablets compacted at turret speed 35 rpm (47 m/min) (EHF: extended head flat, SHF:
standard head flat, RHF: reduced head flat)

head force
EHF 6.0 36.2 (0.2) 24.1 (0.1) 18.5 (0.1) 14.7 (0.5) 58.0 (1.6)
SHF 35.8 (0.2) 25.0 (0.4) 15.7 (0.2) 13.2 (0.6) 53.3 (0.9)
RHF 36.1 (0.3) 26.1 (0.6) 12.1 (0.1) 10.3 (0.9) 48.0 (0.6)
EHF 9.0 49.5 (0.6) 26.1 (0.2) 19.2 (0.2) 14.3 (0.2) 60.0 (0.1)
SHF 50.9 (0.3) 27.2 (1.2) 15.5 (0.1) 13.2 (0.2) 56.0 (1.6)
RHF 51.8 (0.1) 26.3 (0.6) 11.6 (0.1) 10.8 (0.6) 49.6 (0.9)
Table 20. Effect of punch head configuration on mechanical properties of lactose

tablets compacted at turret speed 35 rpm (47 m/min) (EHF: extended head flat, SHF:
standard head flat, RHF: reduced head flat)
Punch Compaction Tensile ER Porosity Ejection AQ

head force strength force
configuration (kN) (MPa) (%) (%) (N)
EHF 6.0 0.23 (0.02) 5.4 (0.3) 20.58 (0.17) 179 (11) 0.79
SHF 0.21 (0.02) 7.3 (0.3) 20.83 (0.03) 162 (7) 0.82
RHF 0.19 (0.01) 7.9 (0.2) 21.06 (0.21) 165 (12) 0.88
EHF 9.0 0.48 (0.02) 8.1 (0.2) 17.02 (0.08) 287 (6) 0.91
SHF 0.42 (0.01) 9.7 (0.2) 17.25 (0.08) 293 (13) 0.96
RHF 0.38 (0.01) 10.4 (0.2) 17.39 (0.15) 281 (8) 1.02
133
Nonetheless, the trend of enhanced plastic yielding with increasing head flat
diameter was similar for turret speeds of 25 (Fig. 28) and 35 (Table 20) rpm. Tablets
compacted using the EHF punch head configuration continued to exhibit the best
mechanical quality with highest tensile strengths and least tablet expansion,
followed by SHF and RHF head configurations (Table 20).
4.2.9 Summary
In this study, the influence of tableting punch head geometry, in relation to
compaction roll profile, on tablet mechanical properties using lactose and
paracetamol-starch granules was investigated. The use of punches with head
geometry capable of increasing the dwell time was advantageous as the increased
dwell time often improved the mechanical strength of tablets and reduced capping
tendencies. Analysis of the lactose and paracetamol-starch tablet properties revealed
tablets produced from punches with a flat head profile consistently displayed better
mechanical quality than tablets produced from punches that lacked a head flat. The
presence of a head flat on the punch head appeared to be crucially important to allow
a state of constant strain for the compacts to undergo stress relief. Tablet tensile
strength showed a positive relationship with dwell time indicating that an increase
in dwell time by even a small extent caused significant improvements to the tablet
mechanical quality. Prolongation of dwell time was advantageous as it allowed
greater plastic flow during dwell phase and consequently reduced tablet expansion
during decompression, demonstrated using experimental and computational data. In
contrast, the effect of the punch head radius modifications on the compaction
parameters was considered marginal at the experimented force levels.
The positive impact of dwell time extension was also noticeable at a higher tableting
speed. The use of a smaller compaction roll caused the strain rate to increase
134
dramatically, facilitated air entrapment in the tablets and thus caused the tensile
strength and elastic recovery of the tablets to decrease and increase, respectively.
The increase in dwell time by the smaller roll enhanced stress relaxation in the
compacts but was insufficient to counteract the rise in elasticity in the compact due
to increased strain rate during consolidation stage. The findings could be used to
provide a comprehensive understanding on the impact of punch head and
compaction roll modifications on tablet properties and aid tablet/tooling/press
manufacturers to make processing and designing considerations.
135
4.3.1 Influence of punch face edge geometry on compaction parameters
The compaction parameters obtained for the different punch face edge geometries at
main compaction event were studied. The punch face edge modification was found
to cause negligible effect on the force-time profiles, with the consolidation (35 ms),
dwell (23 ms) and decompression (20 ms) times showing almost identical values for
both configurations. The punch face edge modification was however found to have
noticeable effect on the set main compaction thickness (STmain), related to the tablet
thickness at the specified force level and in-die porosity of the compact formed
(Table 21). In comparison to the FFBE geometry, the STmain values for FFRE
geometry were consistently lower at the investigated compaction forces, indicating
deeper penetration for the FFRE punch in the die cavity during the compaction cycle
to achieve a desired compaction load. This caused the volume, and consequently the
in-die porosity, of the compact formed by the radius edge punch geometry to be
relatively lower than that of the beveled edge geometry at equivalent compaction
forces, despite the FFRE punch face configuration having a smaller cup volume
(25.10 mm3) than the FFBE configuration (26.03 mm3). This suggested that changes
observed in the packing efficiency and resistance to densification of the powder
particles were due to the alterations in punch face edge geometry.
During a compaction cycle, there are differential movements of powder within the
die, largely initiated from the punch face governed by frictional conditions between
the compact and tooling surface. Central regions of the compact show relatively
greater movement than at the peripheral regions in contact with the die walls [91].
136
Table 21. Effect of punch face edge geometry on set compaction thickness at the
main compaction event
Punch Compaction STmain

face force (mm)
configuration (kN) Lactose Paracetamol
-starch
FFBE 6.0 2.51 2.80

7.0 2.44 2.72
8.0 2.39 2.65
9.0 2.32 2.60
FFRE 6.0 2.47 2.73

7.0 2.38 2.65
8.0 2.32 2.59
9.0 2.24 2.53
The powder mass in the die cavity during consolidation is believed to be in a
fluidized state with air attempting to escape through the clearance between the punch
tips and die wall, thus powder movement and migration of air pockets within the
powder mass are largely believed to be inter-related. From the punch penetration
data (Table 21), it is reasonable to suggest that the radius edge geometry facilitated
the smooth transition of powder particles along the moving punch face and provided
lesser resistance to powder movement during the consolidation phase, allowing
increased radial movement of powder particles [92]. Axial movement of particles at
the die wall/compact interphase relative to the center of the compact was possibly
enhanced due to accelerated particle movement along the radius punch face. The
favorable powder movement by the curved edges of FFRE geometry could also have
facilitated migration of air from the central regions of the powder mass to the
peripheral regions adjacent to the die wall. This consequently allowed greater
powder densification and possibly aided in reducing variation in density distribution
137
within the compact. In contrast, the sharp angled edges and extended flat surface of
the FFBE geometry likely provided greater resistance to movement of powder
particles along the punch face surface thus affecting the particle rearrangement
during the consolidation stage of the compaction process. The radial powder
movement from the peripheral regions of the compact to the center of the compact
was relatively reduced and impeded the extent of axial powder movement along the
die walls, subsequently resisted the flow of air through the powder mass in the
confined cavity of the die. The packing efficiency of the radius edge geometry was
further demonstrated by the widening difference in ST main values between the two
geometries with increasing compaction force during production of lactose tablets
(Table 21). At higher compaction forces, the rate of compaction force application
rises dramatically, providing lesser time for powder movement and air exhaustion.
However, with the curved surface of the FFRE geometry, this negative effect was
reduced by a greater extent than by the angled edges of the FFBE geometry.
The results obtained for the paracetamol-starch tablets showed that the design
modification in the punch face edge affected the compaction parameters of the
paracetamol-starch tablets similarly, with some differences. The STmain values for the
paracetamol-starch granules were greater than the values observed for the
compaction of lactose granules (Table 21) under equivalent compaction conditions.
While the trend of lower STmain values for the FFRE geometry was maintained
(Table 21), the differences between the STmain values for the two edged punches
during paracetamol-starch tablet production were generally consistent at increasing
compaction force. Both observations could be attributed to the elastically deforming
nature of the paracetamol-starch particles [78]. At higher compaction forces, the rate
of force application increased considerably and probably exceeded the rate at which
138
the particles could react. This caused the paracetamol-starch granules to experience
much more elastic deformation at higher forces and increased resistance to
densification, altering the in-die movement of the particles.
4.3.2 Influence of punch face edge geometry on tablet properties
The changes to the mechanical properties of lactose tablets produced using single
compaction cycles with respect to the different punch face edge geometries were
studied. The FFRE tablets consistently displayed higher tensile strength than FFBE
tablets at the investigated compaction forces as shown in Fig 33. Tabletability
decreased in accordance with the change of punch face configuration from FFRE
(Cp = 0.102) to FFBE (Cp = 0.090). By comparing the extent of tablet elastic recovery
(Table 22), it could be seen that the radius edge geometry reduced elasticity in the
compact, prior to punch withdrawal during the decompression phase. This ensured
that the final tablet porosities (Table 22), measured upon ejection, followed a trend
parallel to STmain values, with the FFRE tablets displaying lower tablet porosities
than the FFBE tablets.
It should be noted that the compaction profiles for radius and bevel punch face
configurations were nearly identical, without observable changes to the symmetry
of the curve during the dwell phase. This showed that the differences in tablet
expansion was contributed significantly by elasticity development during the
consolidation phase. Apparently, the different extent of punch penetration by the two
types of punches during tableting led to changes to the tablet internal character as
previously described and contributed to differences in the tablet physical quality.
The FFRE configuration, with enhanced particle movement and air removal,
probably reduced variation in density distribution within the compact accompanied
by greater degree of densification causing the particles to be packed closer together.
139
0.8
0.6
0.4
0.2
0
6 7 8 9
Figure 33. Tensile strength of lactose tablets prepared using single compaction cycle
by FFBE (■) and FFRE (□) punch face configurations.
FFBE and FFRE punch face configurations
Punch Compaction ER Porosity Compaction Ejection

face force quotient force
FFBE 6.0 7.1 (0.2) 19.80 (0.04) 0.077 212 (1.4)

7.0 7.6 (0.1) 18.61 (0.25) 0.108 321 (1.2)
8.0 8.5 (0.3) 17.76 (0.18) 0.130 346 (1.4)
9.0 9.4 (0.1) 16.35 (0.05) 0.126 405 (3.1)
FFRE 6.0 6.8 (0.1) 19.02 (0.04) 0.070 224 (1.0)

7.0 7.0 (0.2) 18.13 (0.13) 0.082 330 (3.6)
8.0 7.7 (0.3) 17.16 (0.10) 0.092 354 (1.6)
9.0 8.7 (0.2) 15.46 (0.21) 0.102 420 (2.5)
140
In comparison, beveled edges introduced more heterogeneity in the compact internal
structure that probably facilitated development of much more stored elastic energy
in the tablet, as evidenced from the tablet expansion data. The elastic response of the
powder during compaction in accordance with changing edge geometries was also
reflected in the ejection force recorded, which is a function of residual die wall
pressure and friction conditions in the die (Table 22). Compacts prepared from FFRE
punch generally exhibited higher ejection force values than the compacts produced
from the punches with the bevel edge (p < 0.05). Increased strain recovery for FFBE
tablets caused compacts to have greater axial recovery and contracted radially to a
greater extent, thus leading to lower residual die wall pressure and correspondingly,
a lower ejection force. The differences in compaction efficiency between the edge
geometries was further demonstrated by the force-punch displacement curves.
Radius edge punches consistently exhibited greater area under the curve, indicating
a greater work of compaction and reduced energy utilization for expansion (Fig. 34).
10000
8000
6000
4000
2000
0
0 2 4 6 8
Punch separation (mm)
Figure 34. Compaction force-punch separation curves for the FFBE (●) and FFRE
(○) punch face configurations.
141
The influence of the punch face edge geometry on the mechanical properties of
tablets compacted using paracetamol-starch granules at main compaction event was
also studied. Paracetamol-starch tablets generally experienced reduced tensile
strength (Fig. 35) and greater elastic recovery accompanied by tablet failure in the
form of capping (Table 23), compared to corresponding lactose tablets at equivalent
compaction forces. This was attributed to the excessive elastic deformation of
paracetamol-starch granules as previously described under Section 4.2.2. However,
the benefit of punch face edge modification on tablet mechanical properties was
evident, a similar trend as exhibited by the lactose tablets, albeit with some
limitations.
FFBE tablets generally displayed a higher propensity to cap than FFRE tablets as
evidenced by the higher capping index values (Table 23). Capping incidences during
high speed rotary production was observed to be more frequent when beveled
punches were used. The greater capping severity of the FFBE tablets was also
observed during the hardness testing during which double capping (capping of top
and bottom of tablet crowns) occurred, particularly at higher compaction forces. The
improved resistance to tablet failure with face edge modification again reinforced
the earlier discussion that the favorable particle densification state and reduced
internal density differences within the compact formed by FFRE geometry had been
advantageous. Extensive elastic expansion in FFBE tablets promoted propagation of
fissures or cracks along the density gradient boundaries of the high pressure regions,
often situated near the moving punch faces and the central body of the compact,
during the unloading phase, amplifying the capping tendency in FFBE tablets.
142
0.5
0.4
0.3
0.2
0.1
0
6 7 8 9
Compression force (kN)
Figure 35. Tensile strength of paracetamol-starch tablets prepared using single

compaction cycle by FFBE (■) and FFRE (□) punch face configurations.
Table 23. Properties of paracetamol-starch tablets prepared using single compaction

cycle by FFBE and FFRE punch face configurations
Punch Compaction Capping ER Porosity Compaction Ejection

face force index quotient force
FFBE 6.0 1.53 13.1 (0.4) 19.03 (0.31) 0.164 87 (1.6)

7.0 1.82 13.7 (0.2) 18.08 (0.30) 0.171 88 (2.9)
8.0 2.27 15.9 (0.1) 17.12 (0.07) 0.186 96 (1.4)
9.0 2.69 17.0 (0.4) 15.85 (0.07) 0.198 105 (2.3)
FFRE 6.0 0.98 11.8 (0.4) 18.45 (0.13) 0.140 93 (2.6)

7.0 1.36 12.9 (0.4) 17.39 (0.17) 0.155 98 (2.3)
8.0 1.81 15.1 (0.2) 16.28 (0.13) 0.166 104 (3.9)
9.0 2.79 15.7 (0.3) 14.60 (0.14) 0.175 103 (1.1)
143
The positive effect of the curved edges in alleviating the tendency of the tablets to
cap tapered off at the highest compaction force of 9 kN with both tablets showing
comparable capping indices indicating the limitation of edge modification. The
outcomes concurred with the results reported in Part 2, in which dwell time extension
had limited effect on tablet properties at higher compaction forces, attributed to
development of high elasticity in the tablet structure.
4.3.3 Discrete Element analysis
DEM is an effective tool to simulate compaction mechanics of granular particles.
Setting up of detailed models can be time consuming and it can also be limited by
computational power, which restricts the number of particles and the size of the
system that can be processed. In this study, approximately 3700 spherical particles
were simulated to form a theoretical 325 mg tablet. This ensured the completion of
a simulation run within a feasible computing timeframe. A constant compaction load
to total particle surface area ratio was maintained. The compaction process was
simulated using three types of punch face configurations – FFP, FFRE and FFBE.
Computational data of powder compaction from the FFP face configuration was
used as a reference.
4.3.3.1 Comparison between simulations and experiments
Usability of DEM simulations to predict the performance of real-world biaxial
compaction process was validated by comparing experimental measurements and
computed responses derived from the simulations [141]. Tablet thickness at target
compaction force and rate of compaction force application (RFA) were chosen as
the simulation variables for the comparison (Fig. 36).
Tablet thickness. As shown in Fig. 36A, qualitatively the effects of punch edge
geometry on tablet height were similar in both experimental results and simulations.
144
A
4.8
Simulation tablet thickness (mm)
4.6
4.4
4.2
3.4 3.6 3.8 4
Experimental tablet thickness (mm)
B
800
Simulation RFA (N/ms)
700
600
500
400
150 200 250 300 350
Experimental RFA (N/ms)
Figure 36. Comparison between simulation and experimental tablet thickness (A)
and RFA values (B) for FFP (●), FFRE (♦) and FFBE (▲) tablets.
145
FFP tablets consistently exhibited the lowest simulated thickness, followed by FFRE
and FFBE tablets (p < 0.05). The simulated thickness values for FFP, FFRE and
FFBE tablets correlated positively with their corresponding experimental results
respectively. One-to-one quantitative comparisons showed that the experimental
values were lower than the simulation values. This could be attributed to the larger
particle size used in simulation and reduced particle rearrangement during the
compaction process.
Rate of Force Application. In both simulations and experiments, simulated RFA
was shown to increase with increasing compaction force for all three types of flat
punch edge geometries. It was also noted that simulated RFA values (Fig. 36B)
correlated positively with experimental RFA values (R2 = 0.99). At any one
compaction force, simulated RFA did not differ significantly between the punches,
agreeing with experimental results (p > 0.05). Quantitatively, experimental RFA
values were lower than simulation values at all investigated compaction forces.
Generally, the trends observed for the mentioned variables showed good qualitative
agreement with experimental values. Quantitative comparison between the two
methods showed some differences as can be expected due to limitations of the
simulation model (larger and spherically shaped particles, approximated contact
parameters, neglected effects of air movement or frictional effects among others).
These comparisons however indicated that the numerical calculations with the
current simulation parameters could accurately capture essential compaction
dynamics, and simulation calculations took place within a feasible computational
time. DEM proved to be a suitable tool for predicting trends with regard to powder
compaction.
146
4.3.3.2 Effect of punch face edge geometry on tablet stress distribution
DEM was employed to investigate how the compaction force would be transmitted
within a powder bed during biaxial compaction with flat-face punches of different
configurations. Computational stress analysis of the tablet cross-sectional area was
prepared for examination. A grid bin group consisting of 96 bins was formed at the
tablet (Y-Z) cross-section as shown in Fig. 37(i). The mean apparent particle stress
of FFP tablet at the lowest compaction force was used as reference value for the
calibration bar. This allowed meaningful comparison between the different types of
tablets by the use of a common stress scale.
As expected, the force of compaction imposed by the upper and lower punches was
not uniformly transmitted within the three tablet types. The differences in the stress
distribution plots were more noticeable at the higher compaction forces. Fig.
37A(ii)-(iii) show the computed relative stress distribution present in FFP tablets
prepared using compaction forces from 8-9 kN. Stress distributions were observed
to be fairly symmetrical with higher shear surfaces created at the central regions of
the tablets and at the interface between the die wall and particles. This form of
symmetrical stress distribution is consistent for a biaxial compaction. The lack of
perfect symmetry may reflect the variation in initial particle arrangement caused by
imperfect die filling and random movement of particles during compaction. Shear
friction formed at the regions adjacent to the die wall indicated higher resistance to
particle movement during the compaction process at those regions. As the
compaction force increased, force components acting on the particles were
accommodated by an increase in local stresses directed towards the central regions
of the tablets, producing loci of high stress. The findings were in general agreement
with those reported in other studies [94, 109].
147
A B C
(i)
(ii)
(iii)
Figure 37. (i) Compaction of particles with grid bin group in the vertical Y-Z cross-section obtained for FFP (A), FFRE (B) and FFBE (C) tablets.
Fig. 37 (ii-iii) Stress distribution in the vertical Y-Z cross-sections obtained for FFP (A), FFRE (B) and FFBE (C) tablets at compaction force (ii)
8.0 and (iii) 9.0 kN. The X and Y axes of the contour plots represent the distance (mm) from the edge and bottom of the tablet respectively.
148
The introduction of a cup to the punch face configuration was shown to cause
significant changes in the stress contour plots. Fig. 37(B-C) shows the computed
relative stress distribution in the FFRE and FFBE tablets with the corners of the plots
representing the regions adjacent to the radial/bevel edges of the punches. The
variation in stress distribution within the FFRE tablet (Fig. 37B(ii)-(iii)) appeared
much greater and less homogenous than in the corresponding FFP tablet (Fig.
37A(ii)-(iii)). The upper half of the tablets generally experienced more stress with
the formation of a high stress core near to the symmetry axis. This indicated that the
radius edge tool significantly increased interaction between particles and punch
surface, thus consequently changing the distribution of compaction load during
compaction. Resolution of increasing compressive force exerted by the FFRE
geometry on the particles interacting with the curved edge of the punch surface
showed that the resultant forces acted towards the center of the tablet at an increased
rate compared to FFP punches. This caused the stress zones at the tablet core to
increase more in intensity and area than in corresponding FFP tablets. This condition
occurred despite the greater tablet volume measured for the FFRE tablets at
equivalent compaction forces. The compressive stress loci occurred in a similar
manner as in the FFP tablet, primarily towards the core regions. However, when a
higher compaction force was used, regions around the circumference of the die wall
experienced reduced stress relative to the central regions due to decreased particulate
interaction possibly caused by increased radial movement of particles. This resulted
in a steep stress gradient between the tablet edges and core with large amounts of
stored energy within the tablet itself.
149
From comparisons of the FFRE contour plots (Fig. 37B(ii)-(iii)) with the FFBE plots
(Fig. 37C(ii)-(iii)), it was apparent that the change from a curved edge to an angled
straight slope caused stress contour of the central regions of the tablet to change
significantly. Horizontal component of forces acting on particles interacting with the
bevel edges allowed a more intensive transfer of shear in the direction of the internal
volume of the tablet with reduced shear surfaces formed at the interface between the
particles and the punch surface. This resulted in formation of shear bands running
from the top edges towards the central regions with subsequent dissipation of the
forces acting between the die wall and particles. The results for the FFBE specimens
showed that the highest stress regions tend to be concentrated at the symmetry axis
with the maximum stress achieved at the central regions shown to be higher than in
corresponding FFP or FFRE tablets. This occurred despite the reduced state of
densification achieved in FFBE tablets. These observations implied that bevel edges
direct force dissemination in a manner that facilitate the development of much more
stored energy in the tablet.
Stress analysis results provided further support to the experimental data on tablet
properties. Curved edge surface of FFRE punches facilitated a more favorable
distribution of the compaction load among the particles compared to the FFBE
punches, thus allowing particles to attain a lower average stress within the tablet
volume and to be compressed to a greater extent to achieve the target compaction
force. High stress regions for the FFRE tablet also seemed to be reduced in intensity
and area. In comparison, the presence of a bevel edge elevated the non-uniformity
of load distribution, leading to greater stress anisotropy in the tablet microstructure.
Compressive forces were directed towards the central regions of the tablet
facilitating localization of stress and created intensive stress gradient in the regions
150
between the constrained and unconstrained tablet body. Upon removal of the applied
force, the release of a higher amount of stored elastic energy caused extensive elastic
recovery in FFBE tablets. This facilitated the disruption of bonds between particles
and propagation of fissures or cracks along the stress gradient boundaries in the
FFBE tablets during the unloading phase. Particle deformation and interparticulate
bonding were reduced as a consequence, thus directly reducing compact mechanical
strength and amplifying the capping tendency in FFBE tablets.
4.3.4 Effect of tableting speed
The relationship between tableting speed, punch face edge geometry and the
resultant tablet properties was also evaluated (Table 24). The increased tableting
speed narrowed the force-time profiles for both edged punches similarly, with
reductions in consolidation (26 ms), dwell (16 ms) and decompression (14 ms) times.
The higher tableting speed caused the STmain values to increase for both punches,
indicating reduced punch penetration and higher in-die compact porosities at the
respective compaction forces. This could be explained by the greater resistance to
compaction by particles at the increased rate of punch displacement, explained
earlier under Section 4.2.8. Nonetheless, the deeper punch penetration by the radius
edge punch was well preserved. Interestingly, the differences in the STmain values
between the two punch face configurations increased at the higher tableting speed.
This could again be explained by the better particle flow and favorable force
transmission by the radius edges. The accelerated compaction process (increased
rate of force application) restricted particulate movement and reduced time available
for particle rearrangement and air escape through the powder mass.
151
Table 24. Properties of lactose tablets prepared using single compaction cycle by FFBE and FFRE punch face configurations at turret speed 35 rpm
(47 m/min)
Punch Compaction STmain Tensile ER Porosity Compaction Ejection

face force strength quotient force
configuration (kN) (mm) (MPa) (%) (%) (N)
FFBE 7.0 2.47 0.27 (0.03) 7.1 (0.1) 19.90 (0.06) 0.125 200 (4.6)
9.0 2.34 0.45 (0.01) 9.0 (0.3) 17.63 (0.17) 0.148 320 (0.8)
FFRE 7.0 2.40 0.32 (0.02) 6.2 (0.2) 19.43 (0.03) 0.083 209 (4.6)
9.0 2.25 0.50 (0.02) 8.3 (0.2) 16.72 (0.41) 0.110 339 (5.0)
152
The ability of the curved surface of the FFRE punch face to allow particle movement
with reduced slide resistance led to better tableting performance under the less
unfavourable tablet formation conditions. In addition, the radius edge’s ability to
reduce the generation of localized stress regions probably aided particle movements
during the consolidation phase.
Tablets compressed at the higher tableting speed displayed poorer tablet physical
quality. However, tablet expansion was reduced in contrast to tablets compacted at
the lower turret speed 25 rpm (Table 22), indicating that the amount of stored elastic
strain was possibly reduced when tablets were prepared with increased compaction
velocity. Thus, the reduction in tablet mechanical strength was attributed to the
higher compact porosities and the resultant reduced interparticulate bonding in the
tablets, rather than due to disruption of formed bonds during decompression. When
compared to the results obtained at turret speed 25 rpm (Table 22), the compaction
efficiency expectedly decreased for both punch face configurations at the higher
tableting speed, as indicated by the values for the compaction quotient (Table 24).
Interestingly, between the two punch face edge geometries, the compaction quotient
for radius edge punch was least affected by the increase in tableting speed. This
suggested that the modification of the punch face curvature could possibly counter
the adverse impact of high tableting speed and preserve compaction efficiency of
compaction tooling to some extent.
4.3.5 Effect of precompaction
The effect of precompaction to the mechanical properties of paracetamol-starch
tablets with respect to the different punch face edge geometries are depicted in Table
25. A precompaction cycle (Tcon: 30 ms, Tdwell: 20 ms, Tdec: 20 ms) with a peak force
of 3 kN was introduced prior to the application of the main compaction forces.
153
Table 25. Properties of paracetamol-starch tablets prepared using double compaction (mode1) cycle by FFBE and FFRE punch face configurations
Punch Compaction STpre STmain Tensile Capping ER Porosity Compaction Ejection

face force strength index quotient force
configuration (kN) (mm) (mm) (MPa) (%) (%) (N)
FFBE 6.0 3.21 2.76 0.42 (0.01) 0.78 11.1 (0.5) 17.31 (0.17) 0.118 124 (1.4)
7.0 3.21 2.69 0.48 (0.02) 0.92 11.8 (0.3) 16.41 (0.08) 0.144 135 (2.1)
8.0 3.21 2.63 0.52 (0.02) 1.20 14.2 (0.7) 15.89 (0.11) 0.162 148 (1.8)
9.0 3.21 2.58 0.56 (0.03) 1.59 14.6 (0.5) 15.37 (0.04) 0.173 147 (2.0)
FFRE 6.0 3.11 2.70 0.47 (0.01) 0.29 9.6 (0.2) 17.01 (0.09) 0.090 117 (2.4)
7.0 3.11 2.62 0.53 (0.02) 0.27 10.7 (0.2) 15.99 (0.21) 0.095 141 (0.5)
8.0 3.11 2.55 0.60 (0.01) 0.53 11.4 (0.3) 15.27 (0.09) 0.113 143 (1.4)
9.0 3.11 2.50 0.64 (0.02) 0.93 12.7 (0.3) 14.54 (0.19) 0.123 141 (6.0)
154
Main compaction cycle was unaffected by the precompaction cycle; consolidation,
dwell and decompression times remained largely indifferent. STmain values however,
decreased further for both punch face configurations (Table 25), indicating a positive
effect on powder densification. Introduction of precompaction led to greater
reduction in porosities of the compacts in the die aided by removal of significant
amount of air in the powder bed at precompaction event which reduced resistance to
powder movement during the main compaction stage. Nonetheless, the punch
penetration to achieve the desired load was still different between the two edge
geometries; FFRE punches continued to display deeper punch penetration at both
precompaction (STpre 3.11 mm) and main compaction events. Interestingly, the
differences between the bevel and radius punches’ ST main values widened at
increasing compaction force, a trend that was not observed in paracetamol-starch
tablets compacted using single compaction cycles. This indicated that the positive
effect of the radius edge on powder packing was amplified when double compaction
cycle was utilized.
The positive influence of precompaction was also reflected by the improvement in
tablet physical quality for both tablet designs. Tablet tensile strength significantly
increased with corresponding reduction in elastic recovery, porosity and capping
tendency (Table 25), probably associated with the neutralization of elastic energy
during the precompaction phase which resulted in lesser tablet recovery and
interparticulate bond disruption. Despite the pronounced effect of precompaction on
tablet properties, the differences in tablet quality between the FFRE and FFBE
tablets were still evident; FFRE tablets displayed better tablet physical quality than
the FFBE tablets. This again was attributed to the deeper punch penetration enabled
by the FFRE geometry, at both pre and main compaction.
155
The FFRE punch face’s ability to disseminate elastic energy in tablets and
consequently reduce capping tendency was well demonstrated by the increased
disparity in the capping index values between the two punch face configurations
(compared to capping index values from single compaction), even at the highest
compaction force of 9 kN as shown in Table 25. The total compact stress at punch
face level was distributed more equally due to the growing interparticle contacts
within the compact formed by the radius punch face, thus reduced the local shear
stress experienced by the particles in the compact and minimized the tendency to
cap. Overall, the findings demonstrated the agonistic effect of precompaction on the
tableting performance of radius edged punches and could provide an alternative in
resolving tablet failure issues.
4.3.6 Effect of roll displacement
The effect of precompaction on the compaction parameters and mechanical
properties of paracetamol-starch tablets with respect to the different edge geometries
was further investigated with activation of the floating roll technology. The air
compensator was activated at the precompaction stage by fixing the precompaction
force (3 kN) and set precompaction thickness (STpre 2.82 mm). This caused the
displacement of the roll and extended the dwell phase (Tdwell: 28 ms) at
precompaction event. Due to the ability of the FFBE face configuration to achieve
the desired compaction load with reduced punch travel, FFBE punches experienced
greater displacement as compared to the FFRE punches, as shown in Fig. 38. The
dwell times for both punch configurations (evaluated from the displacement-time
curve) remained relatively similar despite the different degrees of displacement.
156
0.4
Roll displacement (mm)

0.3
0.2
0.1
0
0 20 40 60 80
Time (ms)
Figure 38. Roll displacement-time curves for the FFBE (—) and FFRE (---) punch
face configurations with the activation of the air compensator (mode2) during double
compaction cycle.
The extended dwell time facilitated further removal of air from the powder bed at
precompaction and consequently reduced in-die compact porosities at the
precompaction stage (evaluated from the addition of STpre and roll displacement).
This subsequently allowed easier particle consolidation and lower in-die porosities
at the main compaction phase, reflected by the reduced STmain values (Table 26). The
FFRE punch continued to show greater punch travel with lower ST main values then
the FFBE punch. Tablet physical properties were significantly improved with the
extension of dwell phase at precompaction with the FFRE tablets particularly
experiencing a substantial increase in tablet strength (Table 26). The results were
again attributed to the efficient particle packing by the radius edge geometry and
aided by greater interparticulate bond reinforcement and plastic flow within the
compact during the extended contact time [45, 119].
157
Table 26. Properties of paracetamol-starch tablets prepared using double compaction (mode2) cycle by FFBE and FFRE punch face configurations
Punch Compaction STpre STmain Tensile ER Porosity Compaction Capping Ejection

face force strength quotient index force
configuration (kN) (mm) (mm) (MPa) (%) (%) (N)
FFBE 7.0 2.82 2.69 0.55 (0.07) 11.2 (0.3) 16.02 (0.12) 0.132 0.20 137 (2.4)
9.0 2.82 2.57 0.81 (0.05) 13.1 (0.1) 13.84 (0.06) 0.157 0 146 (3.0)
FFRE 7.0 2.82 2.60 0.76 (0.02) 9.7 (0.2) 14.47 (0.10) 0.086 0 133 (3.0)
9.0 2.82 2.47 0.99 (0.03) 11.8 (0.2) 12.87 (0.06) 0.106 0 144 (2.1)
158
However, it was found that the tablets produced using both FFBE and FFRE punches
could be considered equivalent in terms of capping tendency, despite obvious
differences in the states of elastic strain condition in both types of tablets as indicated
by the significantly different elastic recovery values (Table 26). The results
demonstrated the considerable influence of dwell time extension in effectively
counteracting the elastic strain built-up in the compact. Nonetheless, the value of the
punch face edge curvature at improving the tableting performance is clear. FFRE
tablets compacted at equal compaction forces to FFBE tablets generally displayed
better physical properties, attributed to deeper punch penetration and enhanced
compaction efficiency of the FFRE punches as previously mentioned. This
advantage of the FFRE configuration can also be extrapolated when tableting at
equal thickness is involved, as used by most tablet manufacturers. Tablets
compacted by the FFRE punch configuration would be able to attain equal thickness,
as tablets formed using the FFBE punch face, at relatively lower compaction forces.
This allows the punch tip to experience lesser stress during tablet production and
could possibly also lengthen the tooling lifespan.
4.3.7 Summary
The use of flat-face punches with a radius edge face configuration was advantageous
in tablet production as it allowed greater powder densification which often improved
the mechanical strength of tablets and reduced capping tendencies. Analysis of the
lactose and paracetamol-starch tablet mechanical properties revealed that tablets
produced using the radius edge punches consistently displayed better mechanical
quality than tablets from the bevel edge punches. The presence of the radius
curvature on the punch face edge was able to allow deeper punch penetration in the
die cavity during the compaction cycle, causing the compact structure to experience
159
enhanced powder consolidation which is necessary for interparticulate bond
formation and efficient dissemination of compaction load through the compact. This
latter effect allowed better utilization of the compaction force energy and reduced
local elastic strain energy developments, thus consequently reduced tablet expansion
during decompression. Stress analysis results provided further support to the
experimental data on tablet properties. Curved edge surface of FFRE punches
facilitated a more favorable distribution of the compaction load through the
particulate system as compared to the FFBE punches, thus allowing particles to
attain a lower average stress within the tablet volume and to be compressed to a
greater extent to achieve the target compaction force. High stress regions for the
FFRE tablet also seemed to be reduced in intensity and area. In comparison, the
presence of a bevel edge elevated the non-uniformity of load distribution, leading to
formation of localized shear stresses in certain regions and contributed to
development of elasticity within the compact.
The positive impact of the radius face configuration on compaction was also
noticeable at a higher turret speed. The application of a precompaction force and
dwell phase extension (via roll displacement) at the precompaction event amplified
the tableting performance of radius face punches to a greater extent when compared
to bevel face punches, due to enhanced stress relaxation of particles under
compaction by the radius face at precompaction stage.
This study has provided a better understanding on the implications of edge
modifications on flat-face punches on the tableting process. Tooling manufacturers
have shown that the radius edge face configuration is an upgrade over the bevel edge
face configuration in terms of tooling strength. The findings from this study have
proven that edge modification could also be beneficial in overcoming compression
160
challenges such as capping and could provide an alternative to the commonly used
bevel shaped punch face configuration.
161
adhesion tendency to the punch surface during powder compaction
4.4.1 Physical characteristics of tableting material
The particle size properties of lactose granules are presented in Section 4.2.1. The
ibuprofen particles had a median particle size 130 ± 6.2 µm, with span 2.32 ± 0.05.
In comparison to the paracetamol-starch and lactose granules used in Part 2 & 3 of
the study, the flowabilty of ibuprofen-lactose blend was poorer (Hausner ratio 1.35
± 0.03, AOR 40.8 ± 1.3°).
4.4.2 Powder adhesion model
Tablets were produced from three different punch face configurations using the
manual press. Based on visual observation of residual powder distribution on the
punch faces at the end of the tableting cycle (30/50 tablets), it could be concluded
that powder adhesion in the current investigation adopted the adhesion model
proposed in literature [61]. The extent of powder adhesion found could be attributed
to various factors impacting the interrelationships between API-tooling surface
adhesive forces, API-excipient adhesive forces and API-API cohesive forces. When
the forces between the powder particles (API-API and API-excipient) exceed that of
the adhesive forces formed between the particles and the punch surface, sticking will
not occur as the stronger cohesion within the compact prevents powder transfer to
the tool surface. Conversely, when the adhesion of powder to punch surface is
stronger than the cohesive strength of the powder components in the compact,
powder sticking will occur. Powder adhesion to punches could be classified into two
basic categories - first and second-degree adhesion. First degree powder adhesion
was characterized by the formation of a relatively uniform monolayer on the punch
face as shown in Fig. 39A.
162
A B
Figure 39. FFP punch face showing first-degree (A) and second-degree (B) powder
adhesion.
Non-uniform powder adhesion with uneven powder masses on the punch face was
associated with second-degree adhesion (Fig. 39B). First-degree powder adhesion is
considered as a minor form of punch sticking, often referred to as “filming”. Filming
occured due to equivalent API-excipient adhesive and API-API cohesive forces,
with corresponding higher API-punch surface adhesive interactions. Formation of
film on the punch face led to dull grainy appearance of the tablet. Filming accelerated
thickening of the API layer on the tooling surface with continuation of the tableting
process, eventually leading to second-degree powder adhesion scenario. This
occurred when the cohesive forces between the API-API particles were greater than
that of the adhesive forces between the API and other tablet components
(excipients), causing the API particles from the tablet body to detach and bond with
the adhered API on the punch surface.
4.4.3 Influence of punch face configuration
The sticking quantification results were corrected for surface area of the punch face
geometries. Examination of the powder adhesion results indicated that the FFP
punches exhibited the highest propensity towards punch sticking at equivalent
compaction forces (Fig. 40).
163
A 70
Powder adherence (µg/mm2)
60
50
40
30
20
FFBE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
10
FFP
FFP
FFP
FFP
FFP
FFP
0
5 7.5 10
B 70
60
50
40
30
20
FFRE
FFBE
FFBE
FFBE
FFRE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
10
FFP
FFP
FFP
FFP
FFP
FFP
0
5 7.5 10
and lower (□) punch faces after compacting 30 (A) and 50 (B) tablets using an
164
Modifications in the punch face configuration have clearly produced differences in
the cohesion-adhesion equilibrium in the tablet system. As described earlier under
Section 4.3.1, particle movement and compressive force resolution are greatly
affected by the punch face configuration. The flattened surface of the FFP punch
face restricted movement of particles at the powder bed boundaries and within the
compact matrix. Due to the absence of undulations on the FFP surface, shear stress
variations on the tablet undergoing compaction was largely avoided and resulted in
a relatively more homogenous stress distribution across the top and lower flat punch
surfaces. As a consequence, powder particles at the compact-tool boundary
experienced lower amount of disruptive side sliding shear during the consolidation
phase and facilitated formation of a reduced stress gradient along the punch face-
compact interphase. This facilitated greater opportunity for contact fusion of the
compact’s surface particles to the flat punch face. Introduction of concavity to the
punch tip (FFBE and FFRE) facilitated axial and radial movement of particles,
particularly from the peripheral to the central regions of the compact. As a result,
particles in contact with punch surface during the consolidation phase were afforded
much lesser constant contact time due to greater material sliding shear forces at the
punch boundaries. At maximum compaction, during which particulate movements
are minimized, the radius/bevel edges imparted variations in the shear force
distribution in the compact. The edged punch face configurations exerted pressure
gradients across the compacting material surface and created high density areas in
the compact particularly in regions adjacent to the edges. This led to reduced liability
of the particles adhering to tool surfaces, in particular, at the punch edges. These
combined effects weakened the adhesive forces of powder to the FFRE/FFBE punch
faces and led to reduced powder adhesion.
165
The findings illustrated the importance of punch face configuration optimization for
mitigation of different types of tableting problems. FFP configuration allowed a
homogenous stress and density distribution in the tablet, favorable for densification
and elasticity reduction in the compact. However, the enlarged flat area of the FFP
geometry accentuated particle adherence (Fig. 40). Angled/curved edges of FFBE
and FFRE face configurations, which were shown to facilitate unfavorable
localization of stresses and elasticity buildup within the tablet, were beneficial in
reducing material adhesion tendency (Fig. 40). The results were similar to those
reported by Roberts et al., whose study demonstrated the advantage of concave
tooling over flat tooling in mitigating powder sticking issues [100].
The advantage of the radius edge punch in terms of compact densification was again
evident in this part of the study. FFRE tablets exhibited significantly reduced post
compaction thickness than the FFBE tablets at equivalent compaction forces (results
not shown); tablets produced by the radius edge punches were denser than those by
the bevel edge punches. This suggested greater utilization of the compaction energy
for densification purpose which probably increased cohesion within the tablet
matrix. Nonetheless, these positive effects on compact densification led to minimal
differences in terms of powder adhesion. Comparison of powder adherence data
between the edged configurations (FFBE and FFRE) showed that overall
differences, if any, were rather small in magnitude and inconclusive (Fig. 40).
Distribution of powder on the FFBE and FFRE punch faces was however found to
be different during visual inspection. Powder adhesion onto the radius edge punches
was found to extend beyond the central flat region as compared to the bevel edge
punches in which powder sticking was limited to the central flat zone and did not
extend to the angled edges (Fig. 41). These observations may be attributed to the
166
differences in stress concentration factors on the edge geometries of the two flat-
face punches. Straight angled edges of the FFBE configuration experienced much
higher stress when compared to stress concentration along the central flat zone. The
high stress regions limited powder adhesion to the flat surfaces by imparting sliding
shear force along the bevel slope, causing particles to pack better, which facilitated
the formation of stronger bonds along the edges. On the other hand, curved edges on
the FFRE face configuration dissipated stress more homogenously during
compaction and minimized differences in stress concentration factors between the
flat and edged surfaces. Hence, less densely packed surface material was presented
at the vicinity of the radius edge as compared to that of the bevel edge. This resulted
in powder adherence beyond the central flat face, along the radial curvature at lower
compaction forces.
A B
Figure 41. Distribution of residual powder on flat-face bevel (A) and radius (B) edge
punch faces.
4.4.4 Influence of compaction force on powder adhesion
Powder adherence was observed to decrease with increasing compaction force,
regardless of punch face configuration (Fig. 40). This was attributed to the enhanced
state of particulate deformation at the higher force levels. Ibuprofen particles
167
reportedly undergo predominantly elastic deformation and lactose granules
experience a mixed mechanism of brittle fracture and plastic deformation.
Nonetheless, the greater degree of physical deformation under higher compaction
loads increased the available interfacial bonding surface and led to formation of
stronger interparticulate bonds. Stronger cohesion within the tablet matrix increased
the minimum force of adhesion required to cause particle detachment off the tablet
surface. However, the effect of compaction force is debatable as contrasting results
has been reported in studies [61, 87]. This could be attributed to different pressure-
sensitive characteristics of the compaction materials used in experiments.
Quantification of powder adherence on the punch faces indicated a greater sticking
tendency on the lower punches, for all punch face configurations and force levels
investigated. This phenomenon was attributed to the force application method
(uniaxial compaction) during tableting by the manual press. Compaction load was
exerted on the powder bed confined in the die by upper punch displacement while
the lower punch remained stationary. However, force transmission across the
compact between the two punches was probably incomplete largely due to partial
utilization of compaction energy to overcome frictional resistance. As a
consequence, the force experienced at the top half of the tablet was greater than at
the bottom regions of the compact. Interparticulate bonds at the peripheral regions
adjacent to the upper punch surface were enforced, leading to less powder adhesion
on the upper punch face. Conversely, the relatively longer contact time between the
tableting material and the lower punch face could also have strengthened adhesive
forces at the punch face boundary interface.
168
4.4.5 Influence of magnesium stearate concentration
Magnesium stearate is a widely used boundary lubricant with anti-adherent
properties and could possibly be used to negate material adhesion issue during
tableting. It acts by readily adsorbing onto surfaces of metal or particles to form
protective lubricant film layers. These layers reduce contact between tablet feed
particles to compaction tools, and could possibly weaken the bond strength of
particles to tool surfaces when formed thus providing anti-adherent function.
However, the use of a higher concentration of magnesium stearate in the
experimented formulation unexpectedly increased powder adherence levels on both
the upper and lower punches for all punch face configurations (Fig. 42) (p < 0.05).
This effect has been reported in literature, attributed to the formation of a eutectic
system between ibuprofen and magnesium stearate which possibly reduced the
thermal stability of ibuprofen [71].
During the compaction process, friction conditions within the die created localized
high temperature areas that exceeded the melting point of ibuprofen. The increased
temperature caused some degree of pressurized melting and concomitant changes on
the tablet surface, while under compressive stress. Rapid recrystallization occurred
on the surface of the compact with the removal of the compaction load and possibly
accelerated adhesion of ibuprofen to the punch surface. In addition, the formation of
thicker lubricant film surrounding the particles at higher magnesium stearate
concentrations probably weakened the cohesive forces within the tablet matrix by
hindering interparticulate bond formation between the powder components, shifting
the cohesion-adhesion equilibrium towards metal surface adhesion.
169
A 70
60
50
40
30
20
FFBE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
10
FFP
FFP
FFP
FFP
FFP
FFP
0
5 7.5 10
B 70
60
50
40
30
20
FFBE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
FFBE
FFBE
FFRE
FFRE
10
FFP
FFP
FFP
FFP
FFP
FFP
0
5 7.5 10
and lower (□) punch faces after compacting 30 (A) and 50 (B) tablets using an
ibuprofen-lactose formulation containing 1.5 %, w/w magnesium stearate.
170
4.4.6 Influence of compaction method (compaction simulator)
The influence of punch face configuration on powder adhesion was similarly
investigated under accelerated compaction conditions using the compaction
simulator. The compaction simulator enables tablets to be produced under conditions
representative of rotary production. Tablets were compacted using single
compaction cycle (total compaction time 100 ms, dwell time 20 ms). Fig. 43 shows
the amount of powder adhered on the upper punches after a continuous tablet
production cycle (50 tablets) using a compaction simulator, quantified by UV
spectroscopic assay measurements. Powder residue on the lower punches was not
evaluated due to possible inaccuracies caused by the horizontal shear caused by the
tablet removal scrapping action. In comparison to upper punch sticking data from
the single station press (Fig. 40), some quantitative differences in powder adherence
were noted although the trend remained and the overall amounts generally decreased
when tablets were compacted using the compaction simulator (Fig. 43).
20
16
12
4
FFBE
FFRE
FFBE
FFRE
FFBE
FFRE
FFP
FFP
FFP
0
5 7.5 10
Figure 43. Effect of punch face configuration on powder adherence on the upper
punch face after tablet production in the compaction simulator, using an ibuprofen-
lactose formulation containing 1 %, w/w magnesium stearate.
171
This was attributed to differences in the compaction kinematics between the
simulator and the manual press. A much shortened compaction cycle in the simulator
caused more rapid punch movements and accelerated the movement of particles and
air pockets from the central regions of the compact towards the annular rings of the
die, creating a highly fluidized state throughout the powder bed. Material shear along
the punch face-compact boundary increased considerably as a consequence and
weakened the adhesive bonding between the particles at the compact surface and
punch during compaction cycle. In contrast, the reduced punch vertical velocity
during compaction in the manual press afforded much more time for particle
rearrangement and migration of air pockets within the powder bed, resulting in a
relatively reduced material shear at the interface. Increased contact fusion time
between particles and punch surface during the use of the manual press had probably
contributed to the disparity in the degree of powder adherence observed between the
two compaction equipment.
4.4.7 Influence of punch face configuration (compaction simulator)
The effect of punch face configuration on powder adhesion tendency was analyzed
using powder quantification data of the upper punches (Fig. 43) and take-off force
values measured from the lower punches (Fig. 44). FFP punches consistently
exhibited higher powder adherence propensity than the edged face configurations
used. Under ideal tableting conditions with formulations not prone to sticking issues,
a much lower take-off force would be required to dislodge the tablet from the lower
punch of the FFP face type than the edged punch faces. However, the amount of
residual powder and take-off forces for the FFP tablets were much higher than the
other two corresponding tablet types, at equivalent compaction forces. Therefore,
results supported the conclusion that the apparent adhesion behavior was clearly a
172
A 60
50
Take-off force (N)

40
30
20
10
0
0 10 20 30 40 50
Tablet sequence
B 60
50
Take-off force (N)
40
30
20
10
0
0 10 20 30 40 50
Tablet sequence
C
60
50
Take-off force (N)
40
30
20
10
0
0 10 20 30 40 50
Tablet sequence
Figure 44. Take-off force profile during the time course of tablet production using
the compaction simulator for the FFP (○), FFRE (●) and FFBE (▲) punch face
geometries at compaction force 5.0 kN (A), 7.5 kN (B) and 10 kN (C) using an
173
consequence of the higher sticking potential of the FFP face configuration. Based on
the powder quantification data (Fig. 43) and take-off force profiles (Fig. 44), the
differences in the powder adhesion tendency between FFBE and FFRE face
configurations could be considered marginal and inconclusive in attempts to rank
the sticking behaviors between the two punch edge geometries. Nonetheless, the
visually observed adherence pattern of powder on the FFBE and FFRE punches
resembled the results obtained using the manual press (Fig. 41). Powder adhesion
on the beveled punch surface was largely restricted to the central flat region of the
face configuration with minimal sticking on the angled surface, while residual
powder often extended beyond the flat area and occupied portions of the curved
surface of the radius edge geometry. This may again be due to the phenomenon of
homogenous stress distribution of compaction load along the radius edge of the
FFRE configuration, as explained earlier.
4.4.8 Influence of compaction force (compaction simulator)
Based on Fig. 43, it could be seen that the powder adherence on the upper punch
surface generally decreased with increasing compaction force, matching the trend
observed for the manual press. Analysis of the take-off force profiles (Fig. 44) during
the time course of tablet production indicated an influential effect of compaction
force on powder adherence process. At the lowest compaction force (5 kN), huge
variations in the take-off values were recorded in the profiles obtained for the various
punch face configurations. The variability seen on the profiles narrowed at higher
compaction force levels, suggesting a more consistent material transfer process at
the tablet-punch surface interphase. At the lower compaction force level (5 kN),
cohesion within the compact matrix was relatively weak due to reduced particle
deformation, thus the energy required for detachment of material from the tablet
174
surface was relatively reduced. This allowed varying amounts of material to
accumulate on the punch face. As the material mass on the punch face attained a
critical size, it broke off and become integrated into the subsequent tablet,
manifesting in a higher measured take-off force. At higher compaction forces, the
interparticulate bonds and cohesive forces within the compact were strengthened.
This increased the energy required to detach particles off the tablet surface and
facilitated in maintaining a relatively consistent bidirectional transfer of material
between the tablet and punch face.
4.4.9 Ejection force as sticking parameter
Tablet ejection force is a function of the pressure required to push the tablet out of
the die (influenced by residual die wall pressure) and the adhesion between the
circumferential side of the tablet and the die wall (influenced by material property
and contact surface area). Ejection force could therefore reflect the extent of powder
sticking to tool surface. The general trend of the ejection force values for the FFP
tablets was significantly higher than that of the FFRE and FFBE tablets over the
entire force range (5-10 kN) (p < 0.05), attributed to increased contact surface area
of FFP compact with the die wall and the resultant higher residual die wall pressure
(Fig. 45). Differences in ejection force between FFRE and FFBE tablets could be
considered marginal.
During the time course of tablet production, it was observed that the force required
to eject tablets varied greatly for the tablets produced using the FFP punches at the
lowest compaction force 5 kN, as demonstrated by the larger scatter of the ejection
forces in the profile. Ejection forces for the FFP punch started at lower magnitudes
and increased to levels higher than those observed for the tablets compacted using
the edged punches and varied drastically during the course of tableting.
175
A
400
350
Ejection force (N)
300
250
200
150
100
0 10 20 30 40 50
Tablet sequence
B 400
350
Ejection force (N)
300
250
200
150
100
0 10 20 30 40 50
Tablet sequence
C 400
350
Ejection force (N)
300
250
200
150
100
0 10 20 30 40 50
Tablet sequence
Figure 45. Ejection force profile during the time course of tablet production using
the compaction simulator for the FFP (○), FFRE (●) and FFBE (▲) punch face
geometries at compaction force 5.0 kN (A), 7.5 kN (B) and 10 kN (C) using an
176
The ejection force profiles showed less variability and reduced in overall force
magnitude at higher compaction force levels. The reduction in ejection force values
at increasing compaction forces validates the occurrence of more powder adhesion
to the die wall at lower compaction forces. In comparison to the FFP compacts, the
ejection forces for the radius and straight beveled punches remained relatively
constant over time at the compaction force levels studied
4.4.10 Summary
In this study, the influence of punch face configuration on powder adhesion tendency
onto the punch face under different compaction conditions was investigated. The use
of punch face configuration with a concavity (radius and bevel edge geometry) was
significantly advantageous as the presence of a cup reduced residual powder
adherence significantly. The outcome was similar to those reported by researchers
who have studied the relationship between punch face configuration and powder
adhesion [99-100]. The flat-face plain configuration reduced movement of particles
along the punch surface-compact boundaries. Absence of undulations on the flat-
face plain configuration ensured a relatively more homogenous stress distribution
across the punch face surfaces and this reduced the level of shear stress variations
across the compact surface. As a consequence, powder particles at the compact-tool
boundary experienced lower amount of disruptive sliding material shear during the
compaction process and this clearly facilitated particle fusion to the punch surface.
Powder adhesion data between the edged punch face configurations showed that the
overall differences were generally marginal and inconclusive, with similar degrees
of material sticking were observed. Process factors, apart from tooling configuration,
were observed to be more influential determinants of powder adhesion. Powder
adherence was observed to decrease with increasing compaction force, attributed to
177
the enhanced state of particulate consolidation at the higher force levels.
Acceleration of the compaction process was found to reduce punch sticking
tendency. Punch-material sticking increased significantly with an increase in
magnesium stearate concentration in the formulation, attributed to interaction
between magnesium stearate and ibuprofen particles which possibly reduced thermal
stability of ibuprofen.
The results from this experiment clearly showed the limitations of edge
modifications on flat-face punches. While the introduction of a curvature to the
punch face edge could strengthen the punch and improve tablet mechanical
properties, it may not provide a solution to tablet adhesion during tableting.
Manipulation of process factors such as tableting speed and compaction force could
be more effective in mitigating tablet adhesion incidences. The findings could be
useful to tablet manufacturers in making tooling considerations to overcome sticking
issues during production.
178
CHAPTER 5
CONCLUSION
179
5. CONCLUSION
In this thesis, the influence of tablet compaction tooling configuration on compaction
process and the resultant effect on tablet physical quality was investigated. Tableting
punch head and punch face configurations are often reworked by tooling
manufacturers for optimal tableting performance. Changes to punch head
configuration impacts different phases of the compaction process, particularly the
dwell phase.
In the first part of the study, the effect of formulation variables on dwell phase
compact deformation was better understood. Tableting formulations possessing
different densification characteristics were employed and reaction of compaction
systems to changes in dwell phase was studied. Plastic flow during the period of
constant compact strain (dwell phase) was observed in the experimented compact
systems, albeit at different extents. Compact systems containing plastically
deforming components reacted positively to dwell time increases; more significant
plastic flow was observed with dwell extension. Rate of stress decay was negatively
influenced by reductions in compact internal porosity caused by extensive
deformation and strain hardening of the plastic components. While dwell time
extension generally improved physical quality of tablets produced from more plastic
compact systems, the changes in tablet strength were however not consistent with
dwell changes and this could be indicative of time dependent bond reinforcement
process in compacts, showing also the viscoelastic character during the dwell phase.
In the subsequent experimental studies, effect of punch head configurational (head
flat and radius) modifications on tableting performance during rotary production was
investigated. Tablet formulations with viscoelastic deformation characteristics were
utilized for high speed tableting. As the dwell period was adjusted via head
180
configurational modifications rather than changes to the compaction kinematics, the
findings provided fresh insights into the effect of dwell during high speed tableting.
The head flat was found to be a much more influential feature than the head radius
in impacting the compaction cycle and tableting outcomes. The presence of a
physical head flat on the punch head was found to be advantageous, as tablets
produced from punches with a flat head configuration consistently displayed better
mechanical properties than tablets made from punches that lacked the head flat. The
flat top on the punch head allowed a state of constant strain for the compacts to
undergo stress relief (greater plastic flow). Tableting performance was however
greatly influenced by inherent deformation nature of the tableting materials and
process factors such as loading rate and compaction force, conditions that facilitated
the build-up of elasticity in the tablet matrix. Benefits of dwell time extension was
limited in more elastic compact systems, particularly at high compaction forces. Use
of a small compaction roll for tablet production decreased the compaction cycle time
that could potentially increase the contribution of dwell duration by proportion and
consequently enhanced stress relaxation in the compact but was insufficient to
counteract the rise in elasticity in the compact due to the higher loading rate during
the consolidation phase.
In the subsequent parts of this study, effect of edge modifications of flat-face
punches on the tableting process were studied. The quality of tablets produced from
different flat-face punch configurations have not been reported in literature, the data
from the next two parts of this study provided new knowledge on the tableting
performance of flat-face punch configurations under various compaction conditions.
Flat-face punch configurations with different edge geometries were used for tablet
production under different compaction conditions. Modification of the angled edge
181
of the bevel face (FFBE configuration) from straight to a curved edge with a radius
(FFRE configuration), enabled deeper punch penetration in the die cavity during the
compaction cycle, bringing about greater compact densification. As a result, tablets
produced from the FFRE face configuration consistently displayed better physical
quality; higher tensile strength and lower capping tendency. Improved die fill
packing by the radius edge geometry increased interparticulate bond formation and
helped to dissipate destructive elasticity within the compact, consequently reduced
tablet expansion during the decompression phase. The positive impact of face edge
modifications of flat-face punches was also more noticeable at a higher turret speed.
The application of the precompaction force along with dwell time extension
amplified the tableting performance of FFRE face configuration to a greater extent
when compared to the FFBE face configuration, attributed to the enhanced packing
efficiency at both precompaction and main compaction stages.
In the final part of the study, the impact of face configurational modification on
tablet material sticking tendency was studied using flat-face plain, bevel and radius
edged face configurations. The use of punch face configuration with a concavity
(radius and bevel edge geometry) was shown to be advantageous as the presence of
a cup reduced residual powder adherence significantly. Flattened surface of the flat-
face plain (FFP) configuration restricted movement of particles at the powder bed
boundaries and within the compact matrix to a greater extent than the other
experimented punch face configurations. Modification of the angled edge of the
FFBE geometry to a curved radius edge did not improve powder sticking to punch
surfaces, indicating equivalent tableting performance in terms of tablet material
adhesion tendency.
182
Thus, through the studies conducted in this study, design modifications to tableting
punch head and face proved to be influential factors in determining tableting
outcomes. The findings provided a better understanding on the advantages and
limitations of punch head and face configurational changes with respect to tablet
properties and may aid in making processing considerations for a successful
tableting program. The findings could also facilitate the design of better tablet
compaction tools.
Finally, two research areas with potential scientific impact are identified for future
investigation. First, a comprehensive evaluation of punch face configuration on the
tableting process should also include other tablet shape categories e.g. capsule and
oval shaped configurations. The most challenging aspect will probably be the need
to standardize the tablet characterization procedure, which may vary with different
tablet shapes. Second, the effect of tooling specialty metal and coating on tablet
quality is a relatively unexplored area of research. It is generally understood that
certain types of specialty metal and coating, usually of the chromium variant, could
be beneficial in reducing sticking tendency. However, benefits of such tooling
modifications have yet to be proven in a scientific approach.
183
CHAPTER 6
REFERENCES
184
6. REFERENCES
[1] M. Turkoglu, A. Sakr, Tablet dosage forms, in: G.S. Banker, C.T. Rhodes,
ed(s), Modern pharmaceutics, Fourth edition, New York: Marcel Dekker,
2009, pp. 481-497.
[2] J.F. Skaftason, T. Johannesson, Tablets and tablet production - with special
reference to Icelandic conditions, Laeknabladid, 99 (2013) 197-202.
[3] S. Mohan, Compression physics of pharmaceutical powders: a review,
International Journal of Pharmaceutical Sciences and Research, 3 (2012)
1580-1592.
[4] M. Çelik, Compaction of multiparticulate oral dosage forms, in: I. Ghebre-
Sellassie, ed(s), Multiparticulate oral drug delivery, New York: Marcel
Dekker, 1994, pp. 181-215.
[5] E.N. Hiestand, J.E. Wells, C.B. Peot, J.F. Ochs, Physical processes of
tableting, Journal of Pharmaceutical Sciences, 66 (1977) 510-519.
[6] B.V. Derjaguin, I.I. Abrikosova, E.M. Lifshitz, Direct measurement of
molecular attraction between solids separated by a narrow gap, Progress in
Surface Science, 40 (1992) 77-82.
[7] M. Duberg, C. Nyström, Studies on direct compression of tablets XVII.
Porosity—pressure curves for the characterization of volume reduction
mechanisms in powder compression, Powder Technology, 46 (1986) 67-75.
[8] D. Sixsmith, The compression characteristics of microcrystalline cellulose
powders, Journal of Pharmacy Pharmacology, 34 (1982) 345-346.
[9] M.J. Adams, M.A. Mullier, J.P.K. Seville, Agglomerate strength
measurement using a uniaxial confined compression test, Powder
Technology, 78 (1994) 5-13.
[10] S.A. Altaf, S.W. Hoag, Deformation of the Stokes B2 rotary tablet press:
quantitation and influence on tablet compaction, Journal of Pharmaceutical
Sciences, 84 (1995) 337-343.
[11] P.C. Schmidt, R. Herzog, Calcium phosphates in pharmaceutical tableting,
Pharmacy World and Science, 15 (1993) 105-115.
[12] E.T. Cole, J.E. Rees, J.A. Hersey, Relations between compaction data for
some crystalline pharmaceutical materials, Pharmaceutica acta Helvetiae, 50
(1975) 28-32.
[13] K. Van der Voort Maarschalk, K. Zuurman, H. Vromans, G.K. Bolhuis, C.F.
Lerk, Stress relaxation of compacts produced from viscoelastic materials,
International Journal of Pharmaceutics, 151 (1997) 27-34.
[14] K. van der Voort Maarschalk, K. Zuurman, H. Vromans, G.K. Bolhuis, C.F.
Lerk, Porosity expansion of tablets as a result of bonding and deformation of
particulate solids, International Journal of Pharmaceutics, 140 (1996) 185-
193.
[15] J.M. Sonnergaard, Quantification of the compactibility of pharmaceutical
powders, European Journal of Pharmaceutics and Biopharmaceutics, 63
(2006) 270-277.
[16] B. Johansson, M. Wikberg, R. Ek, G. Alderborn, Compression behaviour and
compactability of microcrystalline cellulose pellets in relationship to their
pore structure and mechanical properties, International Journal of
Pharmaceutics, 117 (1995) 57-73.
[17] P.J. Denny, Compaction equations: a comparison of the Heckel and Kawakita
equations, Powder Technology, 127 (2002) 162-172.
185
[18] R.J. Roberts, R.C. Rowe, Brittle/ductile behaviour in pharmaceutical
materials used in tabletting, International Journal of Pharmaceutics, 36 (1987)
205-209.
[19] J.S.M. Garr, M.H. Rubinstein, The effect of rate of force application on the
properties of microcrystalline cellulose and dibasic calcium phosphate
mixtures, International Journal of Pharmaceutics, 73 (1991) 75-80.
[20] F. Kiekens, A. Debunne, C. Vervaet, L. Baert, F. Vanhoutte, I. Van Assche,
F. Menard, J.P. Remon, Influence of the punch diameter and curvature on the
yield pressure of MCC-compacts during Heckel analysis, European Journal
of Pharmaceutical Sciences, 22 (2004) 117-126.
[21] J.M. Sonnergaard, A critical evaluation of the Heckel equation, International
Journal of Pharmaceutics, 193 (1999) 63-71.
[22] D.W. Danielson, W.T. Morehead, E.G. Rippie, Unloading and
postcompression viscoelastic stress versus strain behavior of pharmaceutical
solids, Journal of Pharmaceutical Sciences, 72 (1983) 342-345.
[23] S. Shlanta, G. Milosovich, Compression of pharmaceutical powders I.Theory
and instrumentation, Journal of Pharmaceutical Sciences, 53 (1964) 562-564.
[24] S.T. David, L.L. Augsburger, Plastic flow during compression of directly
compressible fillers and its effect on tablet strength, Journal of Pharmaceutical
Sciences, 66 (1977) 155-159.
[25] M. Rehula, R. Adamek, V. Spacek, Stress relaxation study of fillers for
directly compressed tablets, Powder Technology, 217 (2012) 510-515.
[26] L. Casahoursat, G. Lemagnen, D. Larrouture, The use of stress relaxation
trials to characterize tablet capping, Drug Development and Industrial
Pharmacy, 14 (1988) 2179-2199.
[27] J.E. Rees, P.J. Rue, Time-dependent deformation of some direct compression
excipients, Journal of Pharmacy Pharmacology, 30 (1978) 601-607.
[28] F. Ebba, P. Piccerelle, P. Prinderre, D. Opota, J. Joachim, Stress relaxation
studies of granules as a function of different lubricants, European Journal of
Pharmaceutics and Biopharmaceutics, 52 (2001) 211-220.
[29] W.R. Vezin, H.M. Pang, K.A. Khan, S. Malkowska, The effect of
precompression in a rotary machine on tablet strength, Drug Development and
Industrial Pharmacy, 9 (1983) 1465-1474.
[30] M. Celik, Pharmaceutical powder compaction technology, Informa
Healthcare, New York, 2011.
[31] S.K. Dwivedi, R.J. Oates, A.G. Mitchell, Peak offset times as an indication of
stress relaxation during tableting on a rotary tablet press, Journal of Pharmacy
Pharmacology, 43 (1991) 673-678.
[32] P.C. Schmidt, P.J. Vogel, Force-time-curves of a modern rotary tablet
machine I. Evaluation techniques and characterization of deformation
behaviour of pharmaceutical substances, Drug Development and Industrial
Pharmacy, 20 (1994) 921-934.
[33] P.J. Vogel, P.C. Schmidt, Force-time curves of a modern rotary tablet machine
II. Influence of compression force and tableting speed on the deformation
mechanisms of pharmaceutical substances, Drug Development and Industrial
Pharmacy, 19 (1993) 1917-1930.
[34] P.C. Schmidt, M. Leitritz, Compression force/time-profiles of
microcrystalline cellulose, dicalcium phosphate dihydrate and their binary
mixtures—a critical consideration of experimental parameters, European
Journal of Pharmaceutics and Biopharmaceutics, 44 (1997) 303-313.
186
[35] P. Konkel, J.B. Mielck, Associations of parameters characterizing the time
course of the tabletting process on a reciprocating and on a rotary tabletting
machine for high-speed production, European Journal of Pharmaceutics and
Biopharmaceutics, 45 (1998) 137.
[36] M. Leitritz, M. Krumme, P.C. Schmidt, Force‐time curves of a rotary tablet
press. Interpretation of the compressibility of a modified starch containing
various amounts of moisture, Journal of Pharmacy and Pharmacology, 48
(1996) 456-462.
[37] O. Antikainen, J. Yliruusi, Determining the compression behaviour of
pharmaceutical powders from the force–distance compression profile,
[38] G. Ragnarsson, J. Sjogren, Work of friction and net work during compaction,
Journal of Pharmacy and Pharmacology, 35 (1983) 201-204.
[39] H. Nakamura, Y. Sugino, S. Watano, In-die evaluation of capping tendency
of pharmaceutical tablets using force-displacement curve and stress relaxation
parameter, Chemical & Pharmaceutical Bulletin (Tokyo), 60 (2012) 772-777.
[40] A. Nokhodchi, M.H. Rubinstein, H. Larhrib, J.C. Guyot, The effect of
moisture content on the energies involved in the compaction of ibuprofen,
[41] H. Takeuchi, S. Nagira, H. Yamamoto, Y. Kawashima, Die wall pressure
measurement for evaluation of compaction property of pharmaceutical
materials, International Journal of Pharmaceutics, 274 (2004) 131-138.
[42] G.F. Palmieri, E. Joiris, G. Bonacucina, M. Cespi, A. Mercuri, Differences
between eccentric and rotary tablet machines in the evaluation of powder
densification behaviour, International Journal of Pharmaceutics, 298 (2005)
164-175.
[43] P. Anbalagan, C.V. Liew, P.W.S. Heng, Role of dwell on compact
deformation during tableting: an overview, Journal of Pharmaceutical
Investigation, 47 (2017) 173-181.
[44] K.M. Picker, The 3-D model: Comparison of parameters obtained from and
by simulating different tableting machines, AAPS PharmSciTech, 4 (2003)
55-61.
[45] C.E. Ruegger, M. Çelik, The influence of varying precompaction and main
compaction profile parameters on the mechanical strength of compacts,
Pharmaceutical Development and Technology, 5 (2000) 495-505.
[46] Courtoy, GEA - Courtoy tabletting equipment, 2014 [Accessed 11 December
2016]. Available from: http://www.s3process.co.uk/solid-dosage/gea-
courtoy/tabletting
[47] Korsch, High speed, fully automated production, 2016 [Accessed 11
December 2016]. Available from:
http://www.korsch.de/en/applications/pharmaceutical-industry/production
[48] S.D. Bateman, M.H. Rubinstein, R.C. Rowe, R.J. Roberts, P. Drew, A.Y.K.
Ho, A comparative investigation of compression simulators, International
[49] F. Michaut, V. Busignies, C. Fouquereau, B.H. de Barochez, B. Leclerc, P.
Tchoreloff, Evaluation of a rotary tablet press simulator as a tool for the
characterization of compaction properties of pharmaceutical products, Journal
[50] M. Çelik, K. Marshall, Use of a compaction simulator system in tabletting
research, Drug Development and Industrial Pharmacy, 15 (1989) 759-800.
187
[51] J.T. Fell, J.M. Newton, Determination of tablet strength by the diametral-
compression test, Journal of Pharmaceutical Sciences, 59 (1970) 688-691.
[52] S.T. David, L.L. Augsburger, Flexure test for determination of tablet tensile
strength, Journal of Pharmaceutical Sciences, 63 (1974) 933-936.
[53] C. Nystrom, K. Malmqvist, J. Mazur, W. Alex, A.W. Holzer, Measurement
of axial and radial tensile strength of tablets and their relation to capping, Acta
pharmaceutica Suecica, 15 (1978) 226-232.
[54] A. Adam, L. Schrimpl, P.C. Schmidt, Factors influencing capping and
cracking of mefenamic acid tablets, Drug Development and Industrial
Pharmacy, 26 (2000) 489-497.
[55] I. Akseli, A. Stecula, X. He, N. Ladyzhynsky, Quantitative correlation of the
effect of process conditions on the capping tendencies of tablet formulations,
Journal of Pharmaceutical Sciences, (2014).
[56] A. Belič, I. Škrjanc, D.Z. Božič, R. Karba, F. Vrečer, Minimisation of the
capping tendency by tableting process optimisation with the application of
artificial neural networks and fuzzy models, European Journal of
Pharmaceutics and Biopharmaceutics, 73 (2009) 172-178.
[57] E.N. Hiestand, Principles, tenets and notions of tablet bonding and
measurements of strength, European Journal of Pharmaceutics and
Biopharmaceutics, 44 (1997) 229-242.
[58] E.N. Hiestand, D.P. Smith, Indices of tableting performance, Powder
Technology, 38 (1984) 145-159.
[59] I. Akseli, N. Ladyzhynsky, J. Katz, X. He, Development of predictive tools to
assess capping tendency of tablet formulations, Powder Technology, 236
(2013) 139.
[60] K.K. Lam, J.M. Newton, Influence of particle size on the adhesion behaviour
of powders, after application of an initial press-on force, Powder Technology,
73 (1992) 117-125.
[61] S. Paul, L.J. Taylor, B. Murphy, J. Krzyzaniak, N. Dawson, M.P. Mullarney,
P. Meenan, C.C. Sun, Mechanism and kinetics of punch sticking of
pharmaceuticals, Journal of Pharmaceutical Sciences, 106 (2017) 151-158.
[62] J.J. Wang, T. Li, S.D. Bateman, R. Erck, K.R. Morris, Modeling of adhesion
in tablet compression I. Atomic force microscopy and molecular simulation,
Journal of Pharmaceutical Sciences, 92 (2003) 798-814.
[63] J.J. Wang, M.A. Guillot, S.D. Bateman, K.R. Morris, Modeling of adhesion
in tablet compression II. Compaction studies using a compaction simulator
and an instrumented tablet press, Journal of Pharmaceutical Sciences, 93
(2004) 407-417.
[64] S.W. Booth, J.M. Newton, Experimental investigation of adhesion between
powders and surfaces, Journal of Pharmacy and Pharmacology, 39 (1987)
679-684.
[65] M. Eriksson, G. Alderborn, The effect of particle fragmentation and
deformation on the interparticulate bond formation process during powder
compaction, Pharmaceutical Research, 12 (1995) 1031-1039.
[66] R.J. Roberts, R.C. Rowe, The compaction of pharmaceutical and other model
materials - a pragmatic approach, Chemical Engineering Science, 42 (1987)
903-911.
[67] C. Nyström, G. Alderborn, M. Duberg, P.G. Karehill, Bonding surface area
and bonding mechanism-two Important factors for the understanding of
powder comparability, Drug Development and Industrial Pharmacy, 19
(1993) 2143-2196.
188
[68] M. Rahmouni, V. Lenaerts, D. Massuelle, E. Doelker, J.C. Leroux, Influence
of physical parameters and lubricants on the compaction properties of
granulated and non-granulated cross-linked high amylose starch, Chemical &
Pharmaceutical Bulletin (Tokyo), 50 (2002) 1155-1162.
[69] B. van Veen, G.K. Bolhuis, Y.S. Wu, K. Zuurman, H.W. Frijlink, Compaction
mechanism and tablet strength of unlubricated and lubricated (silicified)
microcrystalline cellulose, European Journal of Pharmaceutics and
Biopharmaceutics, 59 (2005) 133-138.
[70] M.J. Mollan, M. Çelik, The effects of lubrication on the compaction and post-
compaction properties of directly compressible maltodextrins, International
[71] M. Roberts, J.L. Ford, P.H. Rowe, A.M. Dyas, G.S. MacLeod, J.T. Fell, G.W.
Smith, Effect of lubricant type and concentration on the punch tip adherence
of model ibuprofen formulations, Journal of Pharmacy and Pharmacology, 56
(2004) 299-305.
[72] F. Waimer, M. Krumme, P. Danz, U. Tenter, P.C. Schmidt, A novel method
for the detection of sticking of tablets, Pharmaceutical Development and
Technology, 4 (1999) 359-367.
[73] L. Shi, C.C. Sun, Overcoming poor tabletability of pharmaceutical crystals by
surface modification, Pharmaceutical Research, 28 (2011) 3248-3255.
[74] K.K. Lam, J.M. Newton, Investigation of applied compression on the
adhesion of powders to a substrate surface, Powder Technology, 65 (1991)
167-175.
[75] K. Kakimi, T. Niwa, K. Danjo, Influence of compression pressure and
velocity on tablet sticking, Chemical & Pharmaceutical Bulletin (Tokyo), 58
(2010) 1565-1568.
[76] R.J. Roberts, R.C. Rowe, The effect of punch velocity on the compaction of a
variety of materials, Journal of Pharmacy and Pharmacology, 37 (1985) 377-
384.
[77] R.J. Roberts, R.C. Rowe, The effect of the relationship between punch
velocity and particle size on the compaction behaviour of materials with
varying deformation mechanisms, Journal of Pharmacy and Pharmacology,
38 (1986) 567-571.
[78] N. Armstrong, L.P. Palfrey, The effect of machine speed on the consolidation
of four directly compressible tablet diluents, Journal of Pharmacy and
Pharmacology, 41 (1989) 149-151.
[79] R. Ishino, H. Yoshino, Y. Hirakawa, K. Noda, Influence of tabletting speed
on compactibility and compressibility of two direct compressible powders
under high speed compression, Chemical & pharmaceutical bulletin, 38
(1990) 1987.
[80] J.S.M. Garr, M.H. Rubinstein, An investigation into the capping of
paracetamol at increasing speeds of compression, International Journal of
Pharmaceutics, 72 (1991) 117-122.
[81] P.V. Marshall, P. York, J.Q. Maclaine, An investigation of the effect of the
punch velocity on the compaction properties of ibuprofen, Powder
Technology, 74 (1993) 171-177.
[82] O.F. Akande, M.H. Rubinstein, P.H. Rowe, J.L. Ford, Effect of compression
speeds on the compaction properties of a 1:1 paracetamol–microcrystalline
cellulose mixture prepared by single compression and by combinations of pre-
compression and main-compression, International Journal of Pharmaceutics,
157 (1997) 127-136.
189
[83] C.E. Ruegger, M. Çelick, The effect of compression and decompression speed
on the mechanical strength of compacts, Pharmaceutical Development and
Technology, 5 (2000) 485-494.
[84] H.A. Garekani, J.L. Ford, M.H. Rubinstein, A.R. Rajabi-Siahboomi, Effect of
compression force, compression speed, and particle size on the compression
properties of paracetamol, Drug Development and Industrial Pharmacy, 27
(2001) 935-942.
[85] H. Larhrib, J.I. Wells, M.H. Rubinstein, Compressing polyethylene glycols:
The effect of compression pressure and speed, International Journal of
Pharmaceutics, 147 (1997) 199-205.
[86] A. Nokhodchi, J.L. Ford, P.H. Rowe, M.H. Rubinstein, The effects of
compression rate and force on the compaction properties of different viscosity
grades of hydroxypropylmethylcellulose, International Journal of
Pharmaceutics, 129 (1996) 21-31.
[87] M. Roberts, J.L. Ford, G.S. MacLeod, J.T. Fell, G.W. Smith, P.H. Rowe,
Effects of surface roughness and chrome plating of punch tips on the sticking
tendencies of model ibuprofen formulations, Journal of Pharmacy and
Pharmacology, 55 (2003) 1223-1228.
[88] D. Natoli, Tooling for pharmaceutical processing, in: L.L. Augsburger, S.W.
Hoag, ed(s), Pharmaceutical dosage forms - Tablets, Third edition, New York:
Informa Healthcare USA, 2008, pp. 1165-1212.
[89] American Pharmacists Association, Tableting Specification Manual,
American Pharmacists Association, 2006.
[90] L.L. Augsburger, S.W. Hoag, Pharmaceutical dosage forms: tablets, Informa
Healthcare USA, New York, 2008.
[91] B. Eiliazadeh, K. Pitt, B. Briscoe, Effects of punch geometry on powder
movement during pharmaceutical tabletting processes, International Journal
of Solids and Structures, 41 (2004) 5967-5977.
[92] D. Sixsmith, D. McCluskey, The effect of punch tip geometry on powder
movement during the tableting process, Journal of Pharmacy and
Pharmacology, 33 (1981) 79-81.
[93] A. Djemai, I.C. Sinka, NMR imaging of density distributions in tablets,
[94] V. Busignies, B. Leclerc, P. Porion, P. Evesque, G. Couarraze, P. Tchoreloff,
Quantitative measurements of localized density variations in cylindrical
tablets using X-ray microtomography, European Journal of Pharmaceutics
and Biopharmaceutics 64 (2006) 38-50.
[95] H. Diarra, V. Mazel, V. Busignies, P. Tchoreloff, Investigating the effect of
tablet thickness and punch curvature on density distribution using finite
elements method, International Journal of Pharmaceutics, 493 (2015) 121-
128.
[96] J.M. Newton, I. Haririan, F. Podczeck, The influence of punch curvature on
the mechanical properties of compacted powders, Powder Technology, 107
(2000) 79-83.
[97] D.C. Andersson, P. Larsson, A. Cadario, P. Lindskog, On the influence from
punch geometry on the stress distribution at powder compaction, Powder
Technology, 202 (2010) 78-88.
[98] M.S. Kadiri, A. Michrafy, The effect of punch's shape on die compaction of
pharmaceutical powders, Powder Technology, 239 (2013) 467-477.
[99] F. Waimer, M. Krumme, P. Danz, U. Tenter, P.C. Schmidt, The influence of
engravings on the sticking of tablets. Investigations with an instrumented
190
upper punch, Pharmaceutical Development and Technology, 4 (1999) 369-
375.
[100] M. Roberts, J.L. Ford, G.S. MacLeod, J.T. Fell, G.W. Smith, P.H. Rowe, A.M.
Dyas, Effect of punch tip geometry and embossment on the punch tip
adherence of a model ibuprofen formulation, Journal of Pharmacy and
Pharmacology, 56 (2004) 947-950.
[101] R.S. Ransing, D.T. Gethin, A.R. Khoei, P. Mosbah, R.W. Lewis, Powder
compaction modelling via the discrete and finite element method, Materials
& Design, 21 (2000) 263-269.
[102] D.T. Gethin, R.S. Ransing, R.W. Lewis, M. Dutko, A.J.L. Crook, Numerical
comparison of a deformable discrete element model and an equivalent
continuum analysis for the compaction of ductile porous material, Computers
and Structures, 79 (2001) 1287-1294.
[103] C.Y. Wu, O.M. Ruddy, A.C. Bentham, B.C. Hancock, S.M. Best, J.A. Elliott,
Modelling the mechanical behaviour of pharmaceutical powders during
compaction, Powder Technology, 152 (2005) 107-117.
[104] Y. Hayashi, T. Miura, T. Shimada, Y. Onuki, Y. Obata, K. Takayama,
Prediction of tablet characteristics from residual stress distribution estimated
by the finite element method, Journal of Pharmaceutical Sciences, 102 (2013)
3678-3686.
[105] J.C. Cunningham, I.C. Sinka, A. Zavaliangos, Analysis of tablet compaction.
I. Characterization of mechanical behavior of powder and powder/tooling
friction, Journal of Pharmaceutical Sciences, 93 (2004) 2022-2039.
[106] I.C. Sinka, J.C. Cunningham, A. Zavaliangos, The effect of wall friction in
the compaction of pharmaceutical tablets with curved faces: a validation study
of the Drucker–Prager Cap model, Powder Technology, 133 (2003) 33-43.
[107] I.C. Sinka, J.C. Cunningham, A. Zavaliangos, Analysis of tablet compaction.
II. Finite element analysis of density distributions in convex tablets, Journal
[108] T. Sinha, R. Bharadwaj, J.S. Curtis, B.C. Hancock, C. Wassgren, Finite
element analysis of pharmaceutical tablet compaction using a density
dependent material plasticity model, Powder Technology, 202 (2010) 46-54.
[109] C.Y. Wu, B.C. Hancock, A. Mills, A.C. Bentham, S.M. Best, J.A. Elliott,
Numerical and experimental investigation of capping mechanisms during
pharmaceutical tablet compaction, Powder Technology, 181 (2008) 121-129.
[110] L.H. Han, J.A. Elliott, A.C. Bentham, A. Mills, G.E. Amidon, B.C. Hancock,
A modified Drucker-Prager Cap model for die compaction simulation of
pharmaceutical powders, International Journal of Solids and Structures, 45
(2008) 3088-3106.
[111] H. Sato, T. Miura, H. Furuichi, FEM Analysis of tablet hardness and punch
strength using ANSYS software, Journal of Computational Science and
Technology, 5 (2011) 120-133.
[112] H.P. Zhu, Z.Y. Zhou, R.Y. Yang, A.B. Yu, Discrete particle simulation of
particulate systems: A review of major applications and findings, Chemical
Engineering Science, 63 (2008) 5728-5770.
[113] A. Mehrotra, B. Chaudhuri, A. Faqih, M.S. Tomassone, F.J. Muzzio, A
modeling approach for understanding effects of powder flow properties on
tablet weight variability, Powder Technology, 188 (2009) 295-300.
[114] M. Khanal, W. Schubert, J. Tomas, DEM simulation of diametrical
compression test on particle compounds, Granular Matter, 7 (2005) 83-90.
191
[115] Y. Sheng, C.J. Lawrence, B.J. Briscoe, C. Thornton, Numerical studies of
uniaxial powder compaction process by 3D DEM, Engineering Computations,
21 (2004) 304-317.
[116] G. Frenning, An efficient finite/discrete element procedure for simulating
compression of 3D particle assemblies, Computer Methods in Applied
Mechanics and Engineering, 197 (2008) 4266-4272.
[117] S.M. Siiriä, O. Antikainen, J. Heinämäki, J. Yliruusi, 3D Simulation of
internal tablet strength during tableting, AAPS PharmSciTech, 12 (2011) 593-
603.
[118] H.P. Zhu, Z.Y. Zhou, R.Y. Yang, A.B. Yu, Discrete particle simulation of
particulate systems: Theoretical developments, Chemical Engineering
Science, 62 (2007) 3378-3396.
[119] O.F. Akande, J.L. Ford, P.H. Rowe, M.H. Rubinstein, The effects of lag-time
and dwell-time on the compaction properties of 1:1
paracetamol/microcrystalline cellulose tablets prepared by pre-compression
and main compression, Journal of Pharmacy and Pharmacology, 50 (1998)
19.
[120] C.K. Tye, C. Sun, G.E. Amidon, Evaluation of the effects of tableting speed
on the relationships between compaction pressure, tablet tensile strength, and
tablet solid fraction, Journal of Pharmaceutical Sciences, 94 (2005) 465-472.
[121] K. Hyun Jo, V. Gopi, Original Articles : Compaction simulator study on
pectin introducing dwell Time, International Journal of Pharmaceutical
Investigation, 35 (2005) 243-247.
[122] S. Sarkar, S.M. Ooi, C.V. Liew, P.W. Heng, Influence of rate of force
application during compression on tablet capping, Journal of Pharmaceutical
Sciences, 104 (2015) 1319-1327.
[123] S. Aoki, J. Uchiyama, M. Ito, Development of new punch shape to replicate
scale‐up issues in laboratory tablet press II: A new design of punch head to
emulate consolidation and dwell times in commercial tablet press, Journal of
Pharmaceutical Sciences, 103 (2014) 1921-1927.
[124] G. Thoorens, F. Krier, B. Leclercq, B. Carlin, B. Evrard, Microcrystalline
cellulose, a direct compression binder in a quality by design environment - A
review, International Journal of Pharmaceutics, 473 (2014) 64-72.
[125] H.Y. Saw, C.E. Davies, A.H.J. Paterson, J.R. Jones, Correlation between
powder flow properties measured by shear testing and Hausner ratio, Procedia
Engineering, 102 (2015) 218-225.
[126] L.E. Dahlinder, M. Johansson, J. Sjögren, Comparison of methods for
evaluation of flow properties of powders and granulates, Drug Development
and Industrial Pharmacy, 8 (1982) 455-461.
[127] C. Cahyadi, L.W. Chan, P.W.S. Heng, The reality of in-line tablet coating,
Pharmaceutical Development and Technology, 18 (2013) 2-16.
[128] R.V. Haware, I. Tho, A. Bauer-Brandl, Evaluation of a rapid approximation
method for the elastic recovery of tablets, Powder Technology, 202 (2010)
71-77.
[129] L. Maganti, M. Çelik, Compaction studies on pellets I. Uncoated pellets,
[130] P. Anbalagan, S. Sarkar, C.V. Liew, P.W.S. Heng, Influence of the punch
head design on the physical quality of tablets produced in a rotary press,
Journal of Pharmaceutical Sciences, 106 (2016) 356-365.
192
[131] P. Anbalagan, P.W.S. Heng, C.V. Liew, Tablet compression tooling – Impact
of punch face edge modification, International Journal of Pharmaceutics, 524
(2017) 373-381.
[132] M. Ucgul, J.M. Fielke, C. Saunders, 3D DEM tillage simulation: Validation
of a hysteretic spring (plastic) contact model for a sweep tool operating in a
cohesionless soil, Soil and Tillage Research, 144 (2014) 220-227.
[133] S. Paul, L.J. Taylor, B. Murphy, J.F. Krzyzaniak, N. Dawson, M.P.
Mullarney, P. Meenan, C.C. Sun, Powder properties and compaction
parameters that influence punch sticking propensity of pharmaceuticals,
[134] D. Khossravi, W.T. Morehead, Consolidation mechanisms of pharmaceutical
solids: A multi-compression cycle approach, Pharmaceutical Research, 14
(1997) 1039-1045.
[135] S.R. Perumalla, L. Shi, C.C. Sun, Ionized form of acetaminophen with
improved compaction properties, CrystEngComm, 14 (2012) 2389-2390.
[136] S. Karki, T. Friscic, L. Fabian, P.R. Laity, G.M. Day, W. Jones, Improving
mechanical properties of crystalline solids by cocrystal formation: New
compressible forms of paracetamol, Advanced Materials, 21 (2009) 3905-
3909.
[137] C.C. Sun, Y.H. Kiang, On the identification of slip planes in organic crystals
based on attachment energy calculation, Journal of Pharmaceutical Sciences,
97 (2008) 3456-3461.
[138] E. Shotton, B.A. Obiorah, Effect of physical properties on compression
characteristics, Journal of Pharmaceutical Sciences, 64 (1975) 1213-1216.
[139] S. Abdel-Hamid, M. Koziolek, G. Betz, Study of radial die-wall pressure
during high speed tableting: effect of formulation variables, Drug
Development and Industrial Pharmacy, 38 (2012) 623-634.
[140] H. Takeuchi, S. Nagira, H. Yamamoto, Y. Kawashima, Die wall pressure
measurement for evaluation of compaction property of pharmaceutical
materials, International Journal of Pharmaceutics, 274 (2004) 131-138.
[141] A.P. Grima, P.W. Wypych, Development and validation of calibration
methods for discrete element modelling, Granular Matter, 13 (2011) 127-132.
193
CHAPTER 7
LIST OF PUBLICATIONS
AND PRESENTATIONS
194
7. LIST OF PUBLICATIONS AND PRESENTATIONS
International Journals
1. Anbalagan P, Sarkar S, Liew CV, Heng PWS 2016. Influence of the punch
head design on the physical quality of tablets produced in a rotary press.
Journal of Pharmaceutical Sciences 106:356-365.
2. Anbalagan P, Liew CV, Heng PWS 2017. Role of dwell on tablet

compaction. Journal of Pharmaceutical Investigation 47(3):173-181.
3. Anbalagan P, Heng PWS, Liew CV 2017. Tablet compression tooling -

Impact of punch face edge modification. International Journal of
Pharmaceutics 524:373-381.
4. Desai PM, Anbalagan P, Heng PWS, Liew CV 2017. Evaluation of tablet

punch configuration on mitigating capping by a quality by design approach.
Drug Delivery and Translational Research (in press).
Oral presentations
1. Anbalagan P, Liew CV, Heng PWS 2016. Influence of tableting punch head
design on tablet compression in a rotary press: An experimental and
computational approach. Asian Graduate Congress on Pharmaceutical
Technology. Singapore.
2. Anbalagan P, Liew CV, Heng PWS 2016. Tablet compression tooling design
modifications to optimize tablet properties. Globalization Pharmaceutics
Education Network (GPEN) conference. Kansas, USA.
Poster presentations
1. Anbalagan P, Liew CV, Heng PWS 2014. Tablet compression tooling design
modifications for optimal tablet manufacture. International Society for
Pharmaceutical Engineering (ISPE) conference. Singapore.
2. Anbalagan P, Liew CV, Heng PWS 2015. Influence of punch geometry on

tablet capping during manufacturing: An experimental and computational
approach. 7th Asian Association of Schools of Pharmacy (AASP) conference.
Taipei, Taiwan.
3. Anbalagan P, Liew CV, Heng PWS 2016. Influence of tableting punch head
design on tablet compression in a rotary press: An experimental and
computational approach. American Association of Pharmaceutical Scientists
Annual Meeting and Exposition. Denver, USA.
195

Anbalagan P

Caricato da

Informazioni sul documento

Descrizione originale:

Titolo originale

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Anbalagan P

Caricato da

Copyright:

Formati disponibili

A STUDY ON COMPACTION PUNCH DESIGN

A THESIS SUBMITTED FOR THE DEGREE OF

been used in the thesis.

I would like to express my heartfelt gratitude to my supervisors, Dr. Celine Valeria

administrative support in these four years. My special appreciation goes to Ms.

encouragement throughout these years.

With my sincere appreciation,

Tablet compaction punches have undergone many changes to their design

specifications over the years in an effort to improve manufacturing efficiency, tablet

understanding the impact of developing new punch designs by modification of

mechanical properties were investigated. Compact deformation during dwell was

when preparing tablets containing components of high plasticity by forming

mechanically strong tablets. However, the effect of dwell could possibly be

influenced by time-dependent secondary factors.

tablet mechanical properties was subsequently studied in a rotary press. Punches

significant improvements in physical quality of tablets prepared. Dwell time

prolongation enabled greater plastic flow to occur, consequently reduced tablet

and it led to insignificant differences in tablet properties.

and tableting performance under different compaction conditions were determined.

microstructure and contributed to development of higher elastic energy within the

compact. Application of a precompaction cycle along with dwell time extension to

configuration to a greater extent than the bevel edged punch.

for tablet compaction. Punches possessing face configurations with a concavity

showed similar degrees of material sticking, with only marginal differences,

suggesting equivalent tableting performance in terms of tablet adhesion tendency.

This research study on punch tools provided a comprehensive understanding on the

manufacturers in process and tool design considerations for a successful tableting

Table 1. Specifications of FFBE and FFRE punch face configurations (FFBE:

Table 2. Physical characteristics of microcrystalline cellulose (MCC), dicalcium

Table 3. Viscoelastic parameters obtained for APAP/MCC compacts of different

Table 4. Viscoelastic parameters obtained for DCP/MCC compacts of different

Table 5. Viscoelastic parameters obtained for APAP/DCP compacts of different

Table 6. Effect of dwell time on stress decay and mechanical properties of

Table 7. Effect of dwell time on stress decay and mechanical properties of

Table 8. Effect of dwell time on stress decay and mechanical properties of

Table 9. Effect of dwell time on stress decay and mechanical properties of

Table 13. Effect of punch head configuration on compaction parameters obtained

Table 14. Effect of punch head configuration on compaction parameters obtained

Table 15. Effect of punch head configuration on lactose tablet mechanical

Table 16. Effect of punch head configuration on paracetamol-starch tablet

Table 17. Effect of punch head configuration on lactose tablet mechanical

Table 18. Effect of punch head configuration on paracetamol-starch tablet

Table 19. Effect of punch head configuration on compaction parameters of lactose

Table 20. Effect of punch head configuration on mechanical properties of lactose

Table 23. Properties of paracetamol-starch tablets prepared using single

Table 25. Properties of paracetamol-starch tablets prepared using double

Table 26. Properties of paracetamol-starch tablets prepared using double

Figure 2. Schematic diagrams depicting the motion of punch head in relation to

Figure 3. Graphical representation of compaction force-punch separation profile

Figure 4. Model for vertical displacement of a punch head on a rotary press...... 12

Figure 5. Schematic showing the diametrical compression test. .......................... 18

Figure 13. Drucker-Prager Cap model with associated parameters.

Figure 18. Manual press used for tablet production.

Figure 23. Stress relaxation profiles of DCP/MCC compacts of different

Figure 24. Stress relaxation profiles of APAP/DCP compacts of different

Figure 27. Comparison of compaction profiles (6 and 9 kN) between compaction

Figure 29. Effect of punch head configuration on paracetamol-starch tablet tensile

Figure 30. Stress distribution contours of simulated paracetamol-starch tablets

Figure 32. Effect of punch head configuration on paracetamol-starch tablet tensile

Figure 35. Tensile strength of paracetamol-starch tablets prepared using single

AOR Angle of repose

API Active pharmaceutical ingredient