Extrusion/spheronization has been widely used to prepare pellets or microspheres from wet masses of pharmaceutical powders. The success of this process is dependent on the rheological properties of the wet powder mass.[1] Understanding the relationship between paste formulation, process conditions and the final product properties is of great importance toward better quality control and product design.[2] So far, the rheologi-cal and mechanical properties of wet masses used in the preparation of pellets by the extrusion/spheronization process have been studied via different approaches, such as ram extrusion. Harrison et al. (1987); Fielden et al. (1989) and Raines et al. (1990) used a ram extruder as a
capillary rheometer to characterize the flow properties of lactose mixtures, microcrystalline cellulose (MCC) and water in proportions, which were capable of forming pellets by the process of extrusion/spheronization.[3,4,5]. Chohan and Newton (1996) used the data of Raines et al. (1990) to demonstrate that further rheological evaluation of the wet masses could be achieved in terms of exten-sional viscosity and elastic properties.[6]. Luukkonen et al. (2001) used a ram extruder to evaluate the influence of MCC type and water content on rheological properties of the wet masses and derived flow curve models for each wet mass.[7] Newton et al. (2005) evaluated the rheo-logical properties of the wet masses formed from a self- emulsifying system and MCC by a ram extruder. They
ReseaRch aRtIcle
Rheological evaluation of wet masses for the preparation of pharmaceutical pellets by capillary and rotational rheometers
Shabnam Majidi1, Ghodratolah H. Motlagh1, Bahareh Bahramian1, Babak Kaffashi1, Seyed A. Nojoumi2, and Ismaeil Haririan3
1Department of Chemical Engineering, Faculty of Engineering, University of Tehran, Enghelab Sq., Tehran, Iran, 2Pasteur Institute of Iran, Tehran, Iran, and 3Department of Pharmaceutics, School of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran
abstractThe rheological properties of wet powder masses used in the preparation of pharmaceutical pellets by extrusion/spheronization were evaluated utilizing capillary and rotational rheometers. A ram extruder was used as a capillary rheometer to construct flow and viscosity curves for each wet mass under different extrusion rates and die geometry. As a result, shear thinning behavior was observed for all wet masses. Among the considered rheological models Power Law and Herschel-Bulkley models fitted well with the experimental results. For the majority of the wet masses, water separation and migration occurred during extrusion which led to uneven water content in the extrudate. The effect of extrusion condition including extrusion speed, die geometry and water content on the occurrence of water separation was investigated and the surface quality of the extrudates was compared. In addition, dynamic rheometry tests were done by a parallel plate rheometer to investigate the viscoelastic properties of the wet masses. The frequency sweep tests showed that as water content of the wet masses decreases storage (G’) and loss modulus (G’’) increase. The storage modulus values were much higher than those of the loss modulus showing dominated elastic rather than viscous behavior for the wet masses at low deformation rates.
Address for Correspondence: Ismaeil Haririan, Department of Pharmaceutics, School of Pharmacy, Tehran University of Medical Sciences, P. O. Box: 14155- 645, Enghelab Sq., Tehran, Iran. Tel/Fax: 0098-21-66461178.E-mail: [email protected]
(Received 10 September 2011; revised 24 October 2011; accepted 03 November 2011)
assessed the shear and tension components of the flow as well as the elastic behavior of the wet masses.[8]
A major problem with the application of a capillary rheometer to evaluate the rheology of wet masses is that, under high stresses, water can separate and inde-pendently move away from the solid mass particularly in vertical capillary rheometers where gravity facilitates the water to move down the barrel. This water redistribution can result in variations in the local water content within the barrel compact paste and the extrudate, leading to inaccurate measurements or uneven product quality. Many workers have noted this phenomenon occurring and tried to evaluate the extent of water movement in dif-ferent ways. Tomer et al. (1999); Rough et al. (2004) exam-ined the extent of water movement during extrusion using a ram extruder by evaluating the water content of the extrudates.[9,10] Rough et al. (2000) investigated the effects of factors influencing the redistribution of water, such as initial water content, extrusion rate and die geometry.[11] Mascia et al. (2006) investigated the onset of liquid phase migration utilizing two different test geometries (i.e. squeeze flow and ram extrusion) to assess the impact of extensional deformation on this phenomenon.[12] From these studies, it turns out that a combination of knowl-edge regarding water mobility and rheological measure-ments is required to completely understand the behavior of the pharmaceutical formulations involved in extru-sion/spheronization process. Another approach to study materials is the application of rotational rheometers. In rotational rheometry, however, the applied strain rates are very much lower than those associated with the pro-cess but this approach may be able to provide valuable insight into rheological characteristics of the wet powder masses. MacRitchie et al. (2002) used a parallel plate controlled stress rheometer to evaluate the rheological properties of lactose mixtures and MCC and obtained the apparent viscosity and viscoelastic properties of the wet masses in dynamic mode.[13]
In the current study, the rheological properties of sugar-based wet masses for the production of pharma-ceutical pellets were studied using two distinctly differ-ent methods. These enabled their rheological properties to be measured over a wide range of deformation rates, from 10−2 to 104 s−1. The wet mass compositions were mix-tures of sucrose, microcrystalline cellulose (MCC), starch and lactose with various water contents. The suitability of the selected wet mass composition for pellet production was examined in an extrusion/spheronization process. At first a ram extruder as a capillary rheometer was used to characterize rheological properties for each wet mass under different extrusion rates and die geometry. The effect of extrusion condition, including extrusion rate, die geometry and water content on the occurrence of water migration during the extrusion of wet masses and the surface quality of the extrudates were accordingly inves-tigated. Subsequently, a parallel plate rheometer was used to evaluate viscoelastic properties of the employed wet masses.
Materials and methods
MaterialsSucrose (Varamin Co, Iran), Microcrystalline Cellulose, MCC, (Avicel PH-102 FMC Corp. Co. Cork, Ireland), Corn starch (Ebnemasuye Co, Iran) and Lactose (Meggle, UK) were used as received. Distilled water was added as a liquid binder at 32–40 wt% based on the dry mix. The proportions of the materials are given in Table 1.
Paste preparationThe powder ingredients were mixed in a planetary mixer for 5 min at about 30 rpm. The required amount of water content was gradually added and the paste was mixed for a further 10 min with occasional scraping of the walls and the blade to ensure uniform distribution of the water in the paste. Five wet masses were produced by this proce-dure and had water contents of 32, 34, 36, 38 and 40 wt.% (on a dry basis). The pastes were stored in sealed poly-amide bags for 24 h at room temperature to allow water to equilibrate throughout the mixture.
Capillary rheometryTheoryIn capillary rheometry, the shear stress is determined from the geometry of die and pressure applied by a piston on the material inside the barrel and the apparent shear rate is determined from the volumetric flow rate. The steady state extrusion pressure values were calculated by divid-ing the steady state force over the cross-sectional area of the barrel. As clearly known from standard capillary rhe-ometry, the measured steady state extrusion pressures do not represent the actual capillary pressure drop. Indeed, due to the die entrance contraction flow, the energy dis-sipation must be subtracted. Typically, this correction is done by means of the Bagley plots (Bagley 1957) where the measured pressures are plotted as a function of the die length to diameter (L/D) ratio. Extrapolation of the pressure data to zero L/D ratio yields the entrance pres-sure drop. The values of the wall shear stress (τ
w) and the
apparent shear rate (γ̇app
) are obtained by equations 1, 2:
τW = D P
L
∆4
(1)
γπ
.
app
QR
= 43
(2)
Where ∆P, L, D and Q are pressure drop, die length, die diameter and volume flow rate, respectively.
Table 1. Formulation of the wet mass used in this work.Ingredients Mass fraction (wt.%)Sucrose 50Avicel PH 102 35Corn starch 10Lactose 5Distillated water 32, 34, 36, 38, 40 mL/100 g dry mass
The apparent shear rate corresponds to Newtonian behavior (constant viscosity). A correction is neces-sary for non-Newtonian materials. The Weissenberg-Rabinowitsch shear rate correction accounts for the fact that the true shear rate is often larger than the apparent shear rate for pseudoplastic materials. The true shear rates γ̇
true can be calculated using the following equation:
γ γ. .
.true app
n
n= +
3 1
4 (3)
Where n is the slope of shear stress versus apparent shear rate in logarithmic plot. The die shear stresses are plotted against the true shear rates to construct a flow curve which provides a representation of the rheological properties of pastes.[14] For convenience, the rheology of pastes is often described as a single-phase continuum using a single constitutive equation such as the Power Law equation:
τ γw pnk= ⋅ (4)
The value of kp is the Power Law viscosity constant and n
is the Power Law index. Wet masses possess an apparent yield stress, and their tendency to display Power Law rate dependence, often makes the Herschel-Bulkley constitu-tive model a suitable choice.
τ τ γw y
nk= + ⋅
HB (5)
Where τy is the yield stress, k
HB and n are the Herschel-
Bulkley viscosity constant and the Herschel-Bulkley index, respectively.[15]
ExperimentalThe extrusion apparatus consisted of a computer-controlled ram extruder, incorporating a cylindrical stainless steel barrel with an inside diameter of 20 mm and various cylindrical square entry dies (1 mm in diameter and 3, 6, 12 mm in length). The ram extruder was mounted in the frame of a 20 kN universal test-ing machine (Gotech TCS 2000). In all experiments, approximately 30 g of the paste was filled into the bar-rel and manually packed by inserting a stainless steel ram until a reference initial height of 10 cm from the die entry was reached. This was considered as starting point for all the experiments. The pastes were extruded at four different speeds (20, 50, 100, 150 mm/min) by the ram attached to the cross-head of the universal machine at room temperature. The applied force was monitored using a load transducer situated within the cross-head while the displacement of the ram was recorded. The steady state extrusion force for each die length to diameter ratio of (L/D = 3, 6, 12) and each ram speed (20, 50, 100, 150 mm/min) was determined. The results were the mean of three experiments.
Rotational rheometryOscillatory measurement is a nondestructive tech-nique which is ideal for investigating structure/
structural changes in materials. The oscillation tech-nique requires applying a sinusoidal oscillating stress wave to a material and measuring the resulting strain wave. Measurements are made over a range of frequen-cies to generate a family of curves. Data of most interest derived from these oscillation strain control measure-ments are Complex viscosity (η*), Storage and Loss modulus (G´& G˝), as a function of frequency (ω). The relative magnitude of G˝ to G´ describes the viscoelas-tic response of the material over the range of angular frequency.
In this work, dynamic rheometery tests were carried out using a parallel plate rheometer at room temperature. The paste sample was placed between the upper moving plate and the lower fixed base plate (diameter 25 mm). The distance between the two plates was adjusted to 2 mm for all measurements. Each experiment has been done three times.
Preparation of samplesApproximately 3 g of the paste was compressed to form a disk with 25 mm in diameter and 2 mm in thickness. The sample was transferred to the lower plate of the rheometer then the upper plate was lowered onto the sample meanwhile the temperature of the plates was maintained at 25°C. To ensure the consistency of the tests, the water content was measured for some of the disks over the time scale of the test and did not show significant reduction.
Strain sweep testInitially, a strain sweep test was applied to each sample in order to determine the linear viscoelastic region (LVR) and to characterize the strain dependent viscoelastic behavior of the samples. It was performed at a frequency of 1.0 s−1 in the range of deformation from 0.1% to 10% for each sample.
Frequency sweep testFrequency sweep test was performed over the range of angular frequency (0.1–100 rad/s) under a constant stain of 0.1% which was in the linear viscoelastic region obtained from the strain sweep test. The elastic modulus (G´), loss modulus (G˝) and complex viscosity (η*) were measured as a function of the angular frequency (ω).
Other characterizationsThe water content of wet masses, extrudates or test sam-ples for dynamic rheometry was measured by putting the wet samples in a vacuum oven at 60°C for 3 h. The surface quality of the extrudates was qualitatively examined by means optical microscopy.
Results
Capillary rheometry or ram extrusionThe extrusion of wet masses gave pressure-ram displace-ment profiles with three stages: initial compaction, steady
Phar
mac
eutic
al D
evel
opm
ent a
nd T
echn
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
r. I
smae
il H
arir
ian
on 0
1/06
/12
For
pers
onal
use
onl
y.
4 S. Majidi et al.
Pharmaceutical Development and Technology
state and force flow. During the compaction stage, the wet mass was compacted to a density close to its saturation level until the material in the barrel and the die began to flow. This was followed by the steady state stage in which the extrusion pressure almost remained constant as the displacement increased and then by forced flow, resulting in an increased extrusion pressure with displacement. For a majority of the wet masses studied here, water separa-tion and migration occurred during extrusion leading to variations of up to 10% of the mean pressure value in the
steady state region as shown in Figure 1. The mean values featured an average standard error of 2% from the mean values calculated from three repeated experiments.
Effect of initial water contentFigure 1 shows the effect of initial water content upon extrusion pressure profile for a ram speed of 100 mm/min through a die with L/D = 12. As the water content increased, the length of the compaction stage decreased. The reason might be due to the higher rate of water release at higher water content so that sufficient water among par-ticles required for the onset of flow builds up at the lower compaction. It is like squeezing a sponge with higher water content. By raising water content, the initial value of the extrusion pressure decreased and the duration of the steady state flow increased. This can be explained by the reduction in bulk yield stress for softer masses, which in turn results in less of a stress gradient within the deforming wet mass, thus decreasing the potential for water migration. Similar measurements were taken for the extrusion speeds of 20, 50 and 150 mm/min through dies with L/D = 3 and 6 and the same trend was observed, that is, shorter compaction region, lower steady pressure and longer steady flow region with increasing water con-tent. The effect of water content on surface quality of the extrudates is shown in Figure 2. It seems that by decreas-ing water content, the extrudate surface becomes better and less surface irregularities appear.
Figure 1. Extrusion pressure-ram displacement profiles for wet masses of initial water content 32 wt.%, 34 wt.%, 36 wt.% and 38 wt.% at ram speed of 100 mm/min; L/D = 12 (n = 3). (See colour version of this figure online at www.informahealthcare. com/phd)
Figure 2. The effect of water content on surface quality of the extrudates by optical microscopy, (A) 32 wt.%, (B) 34 wt.%, (C) 36 wt.% and (D) 38 wt.% all runs at a ram speed of 100 mm/min; L/D = 12. (See colour version of this figure online at www.informahealthcare.com/phd)
Effect of extrusion speedFigure 3 shows the effect of extrusion speed on the extrusion pressure. By increasing extrusion speed, the compaction length decreased and the onset of extrusion shifted to the lower value. As increasing speed gives a shorter time for particles to rearrange, therefore com-paction is less efficient and extrudates come out earlier. The initial value of the extrusion pressure increased with increasing extrusion speed and both the extrusion
pressure fluctuation and consequently the potential for water migration was minimized. This can be explained by the fact that, at higher speeds, less time was available for the water to find a path of least resistance through the voids between the particles and to move down the barrel before it is extruded. At lower speed (20 mm/min), the wet mass was seen to drip out of the die before extrusion. The first flow of extrudate was wetter than the main flow causing an uneven water content and surface defects for the extrudate as shown in Figure 4 (A). However, higher speeds might lead to surface fracture of the extrudates due to larger shear stress applied on the wet mass at the die wall.
The effect of extrusion speed on surface quality of the extrudates is shown in Figure 4. For the wet mass studied, little dependence of speed on the surface fracture was observed. It seems that the speed of 100 mm/min ben-efits from a better surface quality as compared to others and at higher speeds, for example, 150 mm/min, the sur-face starts to roughen.
Effect of die lengthA number of extrusion runs were performed through dies of 1 mm diameter having different lengths of 3, 6, 12 mm. Figure 5 shows the effect of die length on the steady state pressure for the wet mass (initial water content 32 wt.%) at ram speed of 150 mm/min. As the die length increased,
Figure 3. Extrusion pressure-ram displacement profiles for wet mass (initial water content 32 wt.%) through a 12 × 1 mm die at ram speeds of 20 mm/min, 50 mm/min, 100 mm/min and 150 mm/min (n = 3). (See colour version of this figure online at www.informahealthcare.com/phd)
Figure 4. The effect of extrusion rate on the surface of extrudates for wet mass (initial water content 32 wt.%) through a 12 × 1 mm die at ram speeds (A) 20 mm/min, (B) 50 mm/min, (C) 100 mm/min and (D) 150 mm/min. (See colour version of this figure online at www.informahealthcare.com/phd)
Phar
mac
eutic
al D
evel
opm
ent a
nd T
echn
olog
y D
ownl
oade
d fr
om in
form
ahea
lthca
re.c
om b
y D
r. I
smae
il H
arir
ian
on 0
1/06
/12
For
pers
onal
use
onl
y.
6 S. Majidi et al.
Pharmaceutical Development and Technology
the steady state extrusion pressure showed an increase in value, with no considerable effect on fluctuations. This indicates that the die length has no significant effect on the mechanism of water migration.
Image analysis by optical microscopy showed that increasing the die length has a beneficial effect on surface quality of the extrudate. The depth of fracture decreases with increasing die L/D ratio, as shown in Figure 6.
Capillary flow analysisFigure 7 shows the Bagley plots for all the specimens at each extrusion speed. It has demonstrated that all the wet masses provide a linear regression line for the steady state extrusion pressure as a function of the L/D ratio at each extrusion speed. The steady state pressure was raised by increasing both the die L/D ratio and the
Figure 5. Extrusion pressure-ram displacement profiles for wet mass (initial water content 32 wt.%) through dies of 1mm of various lengths 3, 6 and 12 mm; ram speed = 150 mm/min (n = 3). (See colour version of this figure online at www.informahealthcare.com/phd)
Figure 6. The effect of die length on surface quality of the extrudates for the wet mass (initial water content 32 wt.%) at various die lengths (A) 3mm, (B) 6 mm and (C) 12 mm; ram speed = 150 mm/min. (See colour version of this figure online at www.informahealthcare.com/phd)
Figure 7. Bagley plots of the wet masses, (A) 32% (B) 34% (C) 36% (D) 38% water content. (See colour version of this figure online at www.informahealthcare.com/phd)
extrusion speed. These results coincide with previous findings by other workers.
Figure 8 shows the flow curves for the wet masses with varying water content. Generally, the shear stress decreased with increasing water content and the varia-tion of shear stress with shear rate decreased when the water content was increasing.
On the log–log scale, a linear correlation between shear stress and apparent shear rate was adequately observed. As the Weissenberg-Rabinowitsch correction (eq. 3) was applied, the true shear rate was calculated and then viscosity was measured by dividing shear stress over true shear rate. The resulting curves are given in Figures 9 and 10. As anticipated, the derived flow curves show shear thinning behavior, that is, decrease in viscos-ity with increase in shear rate for all the wet masses.
Flow curve modelsThe shear stress versus shear rate data was adjusted to the Power Law and Herschel-Bulkley models which are usu-ally considered for the rheological properties of pastes. Results for the rheological models are shown in Table 2. As shown, the Power Law and Herschel-Bulkley models were well-fitted with high determination coefficients (R2 = 0.99). The Power Law viscosity constant (k) decreases with increasing water content and the power index (n) is almost constant at all water contents. Up to 40% decrease in k is observed by increasing water content from 32 wt%
to 38 wt.%. For the Herschel-Bulkley model, by increas-ing water content, the yield stress (τ
y) decreases up to
20% and the Herschel-Bulkley index (n) increases. This indicates that as water content in the wet mass increases, its flow behavior approaches Newtonian flow. The values of k deceases with increasing water content as well, but are smaller than those for the Power Law model.
Rotational rheometryThe changes in G´ and G˝ of the wet mass (at 36% water content) as a function of strain at frequency of 1.0 s−1 are shown in Figure 11. The strain of 0.1% is in the linear vis-coelastic region. Subsequently, a frequency sweep test was performed at a strain of 0.1%.
Figure 12 shows the changes in G´ and G˝ of the wet mass (at 36% water content) as a function of the angular frequency. Generally, G´ and G˝ values of all wet masses increased with increasing frequency. The G˝ values were much lower than the G´ values at all ω values (0.1–100 rad/s) with a small frequency dependency that confirms the elastic nature of the wet masses.
The complex viscosity as a function of angular fre-quency for different water contents is shown in Figure 13.
Decreasing complex viscosity by increasing frequency demonstrated that the wet masses are shear thinning and as water content increased the complex viscos-ity decreased confirming similar results from capillary rheometry.
Comparison between dynamic and capillary measurementsComparison between dynamic and capillary measure-ments for the wet masses of initial water content 36 wt.%
and 38 wt.% is shown in Figure 14. Extrapolation of capil-lary data, in the range of the frequency sweep test, shows that complex viscosity is larger than steady shear viscos-ity measured from capillary rheometry and the slopes of curves are different (n = −0.9 for rotational rheometer and n = −0.6 for capillary rheometer). It obviously indi-cates that the wet masses are elastic in nature and their structure changes by application of stress. High shear rates lead to breaking of the structure and a decrease in viscosity.
Discussion
A common difficulty in characterizing paste materials in ram extrusion is its variability in experimental data. It is common to observe pressure variations of 10% within a single experiment. This difficulty is usually due to the lack of homogeneities in the pastes, such as air pockets, large agglomerates, or phase migration. In this study for the majority of the wet masses, water migration occurred. This resulted in the first extrudates emerging from the die being wetter than the initial water content. With time, the water content in the extrudates declines until the last extrudates are normally drier than the original water content. This happens in most pharma-ceutical wet masses because water finds the path of least resistance, moving down the barrel faster than sol-ids. To understand how to control this issue, the effect of extrusion condition including extrusion speed, die geometry and water content on the occurrence of water separation, the water content of the extrudates were investigated over extrusion time directly from drying the extrudates. The results showed that higher extrusion speed and water content cause less water separation while no significant effect was observed for the die length. On the other hand, decreasing water content at higher speed has a beneficial effect on surface quality of the extrudates. The flow curves for the wet masses with varying water content showed shear-thinning behavior, that is, decrease in viscosity with increase in shear rate, as reported elsewhere other pharmaceutical
Figure 12. Storage and loss modulus as a function of the angular frequency of wet mass at water content of 36 wt.% (n = 3, log–log scale).
Figure 13. Complex viscosity as a function of angular frequency for different water content (log–log scale).
Figure 14. Comparison between dynamic and capillary measurements for wet masses of initial water content 36 wt.% and 38 wt.% (See colour version of this figure online at www.informahealthcare.com/phd)
Figure 11. G′ and G″ versus shear strain of the wet mass at water content of 36% (n = 3).
wet masses. The shear stress versus true shear rate data was well fitted to Power law and Herschel-Bulkley mod-els. The parameters of these models were calculated. The results showed that the power law viscosity con-stant decreases with increasing water content and the power index is almost constant at all water content. For Herschel-Bulkley model, by increasing water content, the values of yield of stress decrease and the Herschel-Bulkley index increases. The result of dynamic measure-ments showed that as water content of the wet masses decreases storage (G´) and loss modulus (G˝ ) increase. The storage modulus values were much higher than those of the loss modulus showing dominated elastic rather than viscous behavior for the wet masses at low deformation rates and decreasing complex viscosity by raising the frequency showed that the wet masses were shear thinning in behavior.
conclusion
The rheological properties of sugar-based wet masses were comparatively evaluated using a ram extruder and rotational rheometer For the majority of the wet masses studied here, water migration occurred during extrusion leading to variations up to 10% of the mean pressure value and resulted in uneven water content in the extrudate and also surface fracture of the extru-dates. Image analysis indicated that decreasing water content at higher speed has a beneficial effect on surface quality of the extrudate and the depth of frac-ture decreased with increasing L/D ratios of the dies. Shear thinning behavior of all wet masses in the ram extruder was observed and Power Law and Herschel-Bulkley models fitted well the experimental data with high determination coefficients (R2 = 0.99). By increas-ing water content, the behavior of the wet mass was approached to that of Newtonian fluids. The results of dynamic rheometry measurements showed that at low deformation rates, the wet masses are elastic in nature but according to ram extruder results, they can flow at higher deformation rates. This can be explained by the fact that by increasing the shear rate, the structure of the wet masses change and the viscosity reduces so that they can flow through a die at suitable processing conditions.
Declaration of interest
The authors report no declarations of interest.
References 1. Dukic-Ott A, Thommes M, Remon JP, Kleinebudde P, Vervaet C.
Production of pellets via extrusion-spheronisation without the incorporation of microcrystalline cellulose: a critical review. Eur J Pharm Biopharm 2009;71:38–46.
2. Cheyne A, Barnes J, wilson DI. Extrusion behavior of cohesive starch pastes: І-Rheological characterization. J Food Eng 2005;66:1–12.
3. Harrison PJ, Newton JM, Rowe RC. The application of capillary rheometry to the extrusion of wet masses. Int J Pharm 1987;35:235–242.
4. Fielden KE, Newton JM, Rowe RC. The effect of lactose particle size on the extrusion properties of microcrystalline cellulose-lactose mixtures. J Pharm Pharmacol 1989;41:217–221.
5. Raines CL, Newton JM. (1990). Extrusion of microcrystalline cellulose formulations. In: Carter RE, ed. Rheology of Food and Biological Materials with General Rheology. London: Elsevier Applied Science; 240–257.
6. Chohan R K, Newton JM. Analysis of extrusion of some wet powder masses used in extrusion/spheronisation. Int J Pharm 1996;131:201–207.
7. Luukkonen P, Newton JM, Podczeck F, Yliruusi J. Use of a capillary rheometer to evaluate the rheological properties of microcrystalline cellulose and silicified microcrystalline cellulose wet masses. Int J Pharm 2001;216:147–157.
8. Newton JM, Bazzigialuppi M, Podczeck F, Booth S, Clarke A. The rheological properties of self-emulsifying systems, water and microcrystalline cellulose. Eur J Pharm Sci 2005;26:176–183.
9. Tomer G, Newton JM. Water movement evaluation during extrusion of wet powder masses by collecting extrudate fractions. Int J Pharm 1999;182:71–77.
10. Rough SL, Wilson DI. The production of homogeneous extrudates of microcrystalline cellulose pastes. Int J Pharm 2004;276:185–189.
11. Rough SL, Bridgwater J, Wilson DI. Effects of liquid phase migration on extrusion of microcrystalline cellulose pastes. Int J Pharm 2000;204:117–126.
12. Mascia S, Patel MJ, Rough SL, Martin PJ, Wilson DI. Liquid phase migration in the extrusion and squeezing of microcrystalline cellulose pastes. Eur J Pharm Sci 2006;29:22–34.
13. MacRitchie KA, Newton JM, Rowe RC. The evaluation of the rheological properties of lactose/microcrystalline cellulose and water mixtures by controlled stress rheometry and the relationship to the production of spherical pellets by extrusion/spheronization. Eur J Pharm Sci 2002;17:43–50.
14. ASTM D 3835-02. Standard Test Method for Determination of Properties of Polymeric Materials by Means of a Capillary Rheometer.
15. Chhabra R P, Richardson JF. (2008). Non-Newtonian flow and applied rheology: engineering application. 2nd ed. Oxford, UK: Butterworth-Heinemann.
