Twin Screw Reactor

Subscriber access provided by UNIV OF WESTERN ONTARIO
Full Paper
On the use of a twin screw extruder for continuous solid
feeding and dissolution for continuous flow processes
Arabella Marie McLaughlin, John Robertson, and Xiongwei Ni
Org. Process Res. Dev., Just Accepted Manuscript • DOI: 10.1021/acs.oprd.8b00187 • Publication Date (Web): 14 Sep 2018
Downloaded from http://pubs.acs.org on September 14, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.
Page 1 of 31 Organic Process Research & Development
1
2
3
4
5
6
7
8
On the use of a twin screw extruder for
9
10
11
12
continuous solid feeding and dissolution for
13
14
15
16
continuous flow processes
17
18
19
20
Arabella M McLaughlina‡, John Robertsonb‡ and Xiong-Wei Nia*‡
21
22
23 a
24 EPSRC Centre for Continuous Manufacturing and Crystallisation (CMAC), Centre for
25
26 Oscillatory Baffled Reactor Applications (COBRA), School of Engineering and Physical
27
28 Science, Heriot-Watt University, Edinburgh, EH14 4AS, UK
29
30
31 b
32 EPSRC Future Continuous Manufacturing and Advanced Crystallisation Research Hub,
33
34 University of Strathclyde, Technology and Innovation Centre, 99 George Street, Glasgow, G1
35
36 1RD, U.K.
37
38
39 *
40 the corresponding author, tel: 00441314513781; fax: + 441314513129; email: x.ni@hw.ac.uk
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60 ACS Paragon Plus Environment
1
Organic Process Research & Development Page 2 of 31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 Table of Contents graphic
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
2
1
2
3 ABSTRACT The progress from batch to continuous manufacture of pharmaceuticals has
4
5
6 highlighted the challenging area of dosing solid material directly, efficiently and accurately into
7
8 continuous flow systems for work up in flow chemistry processes. Twin screw extruders (TSE)
9
10 have the advantage of decoupling the dry (loss in weight feeder end) with the wet (liquid input)
11
12
13
to prevent solids from sticking around the feeder thereby enabling continuous solid feeding into a
14
15 flow process. In this study, the feasibility of a 16mm TSE as a solid feeder is investigated and
16
17 efficient dissolution of an example API is demonstrated. Paracetamol and an 80:20 mixture of
18
19
water and IPA are the solute and solvent respectively. The concentrations of paracetamol during
20
21
22 dissolution experiments are monitored using an in-line UV-ATR probe connected to a
23
24 spectrometer, and dissolution kinetics are extracted. Full dissolution of powder particles is
25
26 obtained within the residence time of the TSE, however full dissolution of granular particles is
27
28
29 achievable by lowering feed rates or having higher barrel temperatures. We have, for the first
30
31 time, proposed a methodology of estimating the power density for TSE, this enables a fair
32
33 comparison of dissolution rates between this continuous system and a batch stirred tank.
34
35
36
37
38
39 KEYWORDS Dissolution, Solid dosing, Kinetics, Continuous Pharmaceutical Manufacture,
40
41
42 Twin Screw Extruder and UV Spectrometry.
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
3
1
2
3 INTRODUCTION
4
5
6
7 Continuous manufacturing in pharmaceutical industry has gained significant attractions
8
9 recently as it offers potential flexibility, quality and economic advantages over batch operation 1-
10
11 3
. Substantial research in reaction, crystallisation and filtration have been reported in the past
12
13
14
decade 4-8, however, continuous work up, e.g. solid dissolution and dosing, remains a challenging
15
16 area yet to be addressed. The introduction of raw materials and intermediates for continuous
17
18 pharmaceutical manufacturing processes are currently based on batch feed systems, e.g. using
19
20
stirred tank vessels for dissolution of solid particles 9, 10. Depending on the physical properties of
21
22
23 the solids, such as non-wetting (hydrophobic), clumping or floating, dissolution of solids in batch
24
25 vessels is often a labour intensive and time consuming operation, with common problems
26
27 involving mass transfer limitation for solids dissolution, non-uniformity of slurry composition on
28
29
30 discharge, and nozzles plugged by solids. For suspension of sinking solids, adequate mixing for
31
32 off-bottom suspension of particles is essential, this however leads to overmixing overall,
33
34 potentially causing foaming in solid-liquid suspensions, reducing dissolution characteristics 11.
35
36
37
38
39
40
41 For suspension of floating solids, entrainment is often achieved using a mixer designed to
42
43 provide downward pulling drag force to offset the upward buoyancy force, leading to incomplete
44
45
solids wetting, shearing and breaking up of agglomerates 12, 13. If solids are sticky, agglomeration
46
47
48 and accumulation on the impellers, baffles and supports are the norm, leading to batch to batch
49
50 variation on product quality.
51
52
53
54
55
56
57
58
59
4
1
2
3 The use of static mixers has addressed various mixing issues 14, 15, including solids blending in a
4
5
6 fluid phase, the dispersion of additives in a suspension, and solid dispersion by breaking
7
8 agglomerates in a fluid phase and in pulp and paper processes 16. There are however still
9
10 problems in maintaining slurry homogeneity in flow, resulting blockages in valves and nozzles
11
12
13
during downstream processing when solids are present. A continuous solid dosing system
14
15 incorporating complete dissolution would overcome the above issues; this is the focus of our
16
17 work where a twin screw extruder is investigated as a novel solid dosing and dissolution system.
18
19
Twin screw extruders have extensively been studied for Hot Melt Extrusion (HME) 17-20 and wet
20
21
22 granulation 21-26. The objectives of this work are to design and adopt the twin screw extruder as a
23
24 continuous dissolution system; to carry out a systemic investigation of the effects of design and
25
26 operational parameters on solid dissolution; to gain scientific understanding on dissolution
27
28
29 kinetics and to establish a comparison on dissolution kinetics between batch and continuous
30
31 dosing systems.
32
33
34
35
36
37
38
EXPERIMENTAL SECTION
39
40
41 MATERIALS
42
43
44 Two grades (powder and granular) of paracetamol (99% purity) were supplied by Mallinckrodt
45
46
Chemical Limited (UK). The mean particle size and particle size distributions (see Figure 1)
47
48
49 were analysed by a mastersizer (HYDRO, Malvern 3000) and given in Table 1. Raw materials,
50
51 directly purchased from the manufacturer, were powders not agglomerates. The samples were
52
53 dispersed in hexane and added directly to the mastersizer.
54
55
56
57
58
59
5
1
2
3
25
4
5 Paracetamol Powder
6 20 Paracetamol Granular
7
Volume Density (%)
8
9 15
10
11
10
12
13
14 5
15
16
17 0
18 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0
19 Size classes (μm)
20
21
22
Figure 1 Particle size distribution of powder and granular paracetamol
23
24
25
26
27
28 Table 1 Particle sizes for two types of paracetamol
29
30
31 Dx (10) (µm) Dx (50) (µm) Dx (90) (µm)
32
33
34
Powder 12.6 44.9 124
35
36
37
38 Granular 263 374 516
39
40
41
42
43
44 Propan-2-ol (IPA) (>99.5% purity) was sourced from Sigma-Aldrich (Gillingham, UK).
45
46
47 Deionised water was produced using the in-house Millipore Milli-Q system.
48
49
50
51
52
53
54
55
56
57
58
59
6
1
2
3 METHODS
4
5
6
7 Flow Properties
8
9
10 FT4 Powder Rheometer (Freeman Technology Ltd., Tewkesbury, UK) was used to measure
11
12 bulk properties and dynamic flow for each of the grades of paracetamol including 1) Stability
13
14
15
and Variable Flow, 2) Permeability, 3) Aeration, 4) Compressibility and 5) Shear Cell 9kPa. All
16
17 tests were conducted in 25mm cells. Data was collected using the FT4 Powder Rheometer
18
19 software version 4.0 (Freeman Technology Ltd., Tewkesbury, UK) and analysed with FT4 Data
20
21
Analysis software version 3.01.0057 (Freeman Technology Ltd., Tewkesbury, UK).
22
23
24
25
26
27
28 Continuous Twin Screw Extruder
29
30 A 16mm diameter twin screw extruder (TSE) (Eurolab 16, Thermo Fisher Scientific, Stone,
31
32
33 UK) is shown schematically in Figure 2. The barrel has a length of 400 mm with a length to
34
35 diameter ratio of 25:1. Liquids are dispensed into the barrel using a peristaltic pump (Watson
36
37 Marlow, Falmouth, UK) and solids added via a loss in weight (LIW) gravimetric feeder. Two
38
39
40
types of feeders were used in this work including a Brabender MT-S LIW Feeder and a
41
42 Brabender FW-18 Flexwall Classic LIW Feeder (see Figure 3). The former is a rigid frame
43
44 laboratory scale feeder with twin concave screws suitable for low feed rates (< 10 g min-1) of
45
46
high bulk density materials (e.g. granular paracetamol), while the latter is a universal flexible
47
48
49 wall feeder with a single spiral screw suitable for materials with poor flowability and low bulk
50
51 density (e.g. powder paracetamol).
52
53
54
55
56
57
58
59
7
1
2
3
4
5
6
7
8
9 PAT Liquids Solids
10
probe
11
12
13
14
15
16
17
18 Dissolution
19
20 Figure 2 Schematic of continuous twin screw extruder
21
22
23 The extruder is connected to a central control unit where temperature and screw speed can be
24
25
varied. The temperatures of different sections along the barrel are controlled by electrical heating
26
27
28 bands and monitored by thermocouples. A bespoke discharge coupling, and tubing was
29
30 connected to the exit of the twin screw to provide downward output of material. This prevented
31
32 build-up of material at the extruder exit. The UV probe was mounted on a retort stand and
33
34
35 inserted into the tubing. Absorbance data is collected continuously using a UV-ATR probe
36
37 inserted at the flow exit, interfaced with a Carl Zeiss MC600 Spectrometer and a PC for real-
38
39 time display, logging and data analysis.
40
41
42
43
The residence time of liquid within the barrel was measured to aid the determination of
44
45 dissolution kinetics. A dye was injected at liquid entry ports and minimum residence times of the
46
47 dye were recorded. When liquid enters the barrel at port 6, the residence time is merely 3
48
49
seconds, which is the same as the UV probe capture time, hence the UV probe is insufficient to
50
51
52 allow direct measurement for this case, a digital stopwatch was used instead. The variations
53
54 obtained were + 0.2 seconds for five measurements.
55
56
57
58
59
8
1
2
3
4 Single Screw
5 Flexwall
6
Feeder
7
8
9
Mini Twin
10 Screw
11 Feeder
12
13
14
15 Solid Feeder
16 Cone
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 Figure 3 Set up of continuous twin screw extruder with LIW gravimetric feeders
40
41
42
43
In dissolution work, solids are dosed at Port 1 of the barrel as shown in Figure 4; liquid flows
44
45 from the right to left and can be pumped in at any of the Ports from 2 to 6 (see Figure 4). In this
46
47 way, the decoupling of the dry solids from the wet liquid is achieved, preventing solids from
48
49
sticking around the feeder. Liquid coming at Port 2 has the longest residence time of 13 seconds
50
51
52 within the barrel whereas liquid at Port 6 the shortest residence time of 3 seconds.
53
54
55
56
57
58
59
9
1
2
3
4 Solid Input
5 Liquid Input
6
7
8
9
10
11
12 Port Port Port Port Port Port
13 6 5 4 3 2 1
14 Solution
15
Out
16
17
18 Figure 4 Schematic of barrel showing input port positions
19
20
21
In order to obtain a concentration-time profile, calibration curves of absorbance versus
22
23
24 concentration were generated from known amounts of paracetamol in the Water/IPA (80:20)
25
26 solvent system, with a maximum absorbance peak at 248nm. A complete set of sequential runs
27
28 were then undertaken at each port, e.g. Run 1 at Port 6, Run 2 at Port 5, Run 3 at Port 4, and so
29
30
31 on, due to the short residence time. The absorbance measurements were recorded using the in-
32
33 line UV-ATR probe positioned at the TSE exit. Compiling the output concentrations at each port,
34
35 a dissolution profile was finally assembled.
36
37
38
39
RESULTS AND DISCUSSION
40
41
42 Material Characterisation
43
44
45 In order to feed solids consistently and accurately in continuous manufacturing processes even
46
47
at small flow rates (< 5 g min-1), the choice of feeder is of utmost importance, as feeder’s
48
49
50 performance is strongly dependent upon flow properties of materials 27. For instance, the use of
51
52 LIW feeders improved the ability to control feed rates for powders with high cohesion and
53
54 electrostatics 28; the use of flexible frame LIW feeders with a single spiral screw was good for
55
56
57
58
59
10
1
2
3 materials with low bulk density and poor flow ability 26. For materials with higher values of bulk
4
5
6 density (0.5 g ml-1), a rigid frame LIW feeder with concave twin screws worked well.
7
8
9
10
11
12 Prior to the decision of feeder and screw used in this study, the flow properties of powder and
13
14
15
granular paracetamol were determined using the FT4 Powder Rheometer as listed in Table 2.
16
17
18 Table 2 Particle properties of paracetamol
19
20
21 Mean Bulk Permeability Basic Compressibility Cohesion
22
23
Particle Density (cm2) Flowability (% @ 15kPa) (kPa)
24
25
26 Size (g ml-1) Energy (mJ)
27
28 (µm)
29
30
31
Granular 374 0.728 6.6x10-5 630.2 29.1 0.49
32
33
34
35 Powder 45 0.357 4.1x10-6 150.3 51.7 1.68
36
37
38
39
40
41 It is clear that powder has lower bulk density and poorer flow properties (indicated by the
42
43
44 numbers of the Basic Flowability Energy) than granular; while granular has a lower %
45
46 compressibility suggesting that this is a non-cohesive material which is less likely to compact on
47
48 the feeder. The stability and variable flow tests indicated that two grades of the paracetamol did
49
50
51 not show any signs of de-agglomeration or segregation and were stable during flow, although the
52
53 powder was more sensitive to changes in flow rate, mainly as a result of high air content in the
54
55
56
57
58
59
11
1
2
3 cohesive material. The measurements were made on the different feeders and in combination
4
5
6 with the FT4, an agreement with previous work 26-28 is seen.
7
8
9
10
11
12 Dosing trials at a solid feed rate of 3.33 g min-1 confirmed the above selections as the other
13
14
15
combinations caused significant problems, e.g. the single spiral screw was unable to convey
16
17 granular paracetamol due to an increase in the frictional resistance to flow because of
18
19 entrainment, alarming feeder and shutting down the operation. Likewise, the powder presented
20
21
challenges in the rigid frame feeder due to the increase in torque that has compacted powder
22
23
24 within and between the screws, as well as the barrel housing. This also led to the feeder shutting
25
26 down.
27
28
29
30
31
32
33 Dissolution tests
34
35
36 The solubility for paracetamol in water/IPA (80:20) at 40˚C is 11g paracetamol in 100g of
37
38 solvent 29, i.e. at a solid: liquid ratio of 1:9. The dissolution tests were carried out at a solid:
39
40
41
liquid ratio of 1:11 to ensure complete dissolution. Paracetamol powder was continuously dosed
42
43 into the TSE using the single screw flexible wall LIW feeder at a feed rate of 3.33 g min-1 and
44
45 the solvent (Water/IPA 80:20) at a flow rate of 37 g min-1, giving a target output concentration of
46
47
9.0 g per 100 g solvent. Figure 5 shows that complete dissolution was obtained within the
48
49
50 residence time of the barrel (13 seconds) where the concentration of paracetamol at the exit was
51
52 consistently 9.0 g + 0.1 g per 100 g solvent (see Figure 5).
53
54
55
56
57
58
59
12
1
2
3 The same tests were carried out for granular paracetamol using a twin screw rigid wall LIW
4
5
6 feeder at the same feed rates, complete dissolution was not achieved within the residence time of
7
8 the barrel (see Figure 5) with some undissolved paracetamol being observed in the exit. The
9
10 concentration of paracetamol in the exit remained at 7.7 + 0.3 g per 100 g solvent for 3000
11
12
13
seconds. 160 seconds run time for granular paracetamol is plotted for the purpose of comparison
14
15 with that for powder paracetamol.
16
17
18
19
20
21
22 12.0
Solubility Limit
23
24 10.0 (11 g/100 g)
Concentration (g/100 g)
25
26 8.0
27
28 6.0
29
30 4.0
Granular
31 Powder
32
2.0
33
34
35 0.0
36 0 20 40 60 80 100 120 140 160
37 Time (s)
38
39
40
41 Figure 5 Concentration-time profile post TSE for dissolution of paracetamol (Temperature =
42
43
44
40˚C, solid feed rate = 3.33 g min-1, liquid flow rate = 37 g min-1)
45
46
47
48
49
50 Dissolution tests were carried out at a solid feed rate of 2.5 g min-1 and a liquid flow rate of 29 g
51
52
min-1, giving a target output concentration of 8.6 g per 100 g solvent. Dissolution profiles were
53
54
55 compiled from the concentrations at exit of the TSE for liquid input at each port as shown in
56
57
58
59
13
1
2
3 Figure 6. Note that each concentration measurement was repeated 3 times and the data in Figure
4
5
6 5 are the averaged value from three repeats. Error bars are not shown on the graph as they are too
7
8 small to see clearly. The variation obtained for the runs at each port are given for granular
9
10 paracetamol - standard error 0.2, 0.3, 0.2, 0.2, and 0.1.
11
12
13
14
We see that both grades gradually dissolve along the barrel of the TSE with faster and fuller for
15
16 powder than for granular grade, which is consistent with what has been shown in Figure 5 as
17
18 well as previous work 30 in a stirred tank vessel.
19
20
21
22
23
24
25 10.0
26 9.0
27
8.0
28
29 7.0
30 6.0
Port 3 Port 2
31 5.0 Port 4
32 Port 5
4.0 Port 6
33
3.0
34
35 2.0 Granular
36 1.0 Powder
37 0.0
38 0 2 4 6 8 10 12 14 16 18
39 Time (s)
40
41
42
43
44 Figure 6 Dissolution profile for paracetamol dissolution (Temperature = 40˚C, solid feed rate
45
46 = 2.5 g min-1, liquid flow rate = 29 g min-1)
47
48
49
50
51
52
53 Dissolution Kinetics
54
55
56
57
58
59
14
1
2
3 Fitting the first order kinetics, Figure 7 plots of ln (C2/C1) vs time where C2 and C1 are the
4
5
6 concentrations of paracetamol (g 100g-1) at the starting and dissolution times. The straight line fit
7
8 confirms the first order kinetics and the slope of which gives the rate constant of dissolution =
9
10 0.0162 s-1 and 0.0211 s-1 for granular and powder respectively.
11
12
13
14
15
16
17
18
19
20 0.35
21 Granular y = 0.0211x - 0.0344
22 0.3 R² = 0.8103
Powder
23
24 0.25
25
LN (C2/C1)
26 0.2
27 0.15
28
29 0.1
30 y = 0.0162x - 0.026
31 0.05 R² = 0.9679
32
0
33
0 2 4 6 8 10 12 14 16 18
34
35 Time (s)
36
37
38
39 Figure 7 Dissolution kinetic plots of ln (C2/C1) vs time for granular and powder paracetamol
40
41
42
43
44
45
46 Achieving full dissolution
47
48
49 From Figure 6, we see that the granular paracetamol does not fully dissolve within the
50
51 residence time of the barrel. The target concentration for full dissolution at the set feed rates (2.5
52
53
54 g min-1 solid and 29 g min-1 liquid) is 8.6 g (paracetamol)/100 g (solvent). This was achieved for
55
56 powder grade but not for granular where 6.9 g (paracetamol)/100 g (solvent) was dissolved.
57
58
59
15
1
2
3 Several operational parameters in the TSE can however be manipulated to afford full dissolution,
4
5
6 including liquid flow rate, solid feed rate, screw speed, screw configuration and barrel
7
8 temperature. These parameters are investigated in turn.
9
10
11
12
13
14
15
Effect of Liquid Flow Rate
16
17
18 At a fixed solid feed rate of 2.5 g min-1 and a fixed solution temperature of 40˚C, Figure 8
19
20 shows the dissolution concentrations for various liquid flow rates from 15 to 35 g min-1.
21
22
Increasing liquid flow rates to achieve sink conditions results in a higher driving force for
23
24
25 dissolution but reduces the residence time for mixing within the barrel as the degree of screw fill
26
27 is higher. Increasing the liquid feed rate decreases the concentration of the output solution and
28
29 does not achieved full dissolution of the solute. Solvent flow affects the dissolution process by
30
31
32 physical abrasion of the solid, thereby reducing the diffusion layer thickness around each particle
33
31
34 . The decrease of liquid flow rate increases the residence time within the barrel from 11 to 15
35
36 seconds allowing more time for dissolution. Table 3 shows the correlations between the
37
38
39
solid/liquid feed rates and concentrations.
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
16
1
2
3
12.0
4
5 15 g/min
6 10.0
20 g/min
7
25 g/min
8 8.0
9 30 g/min
10 6.0
35 g/min
11
12
4.0
13
14
15 2.0
16
17 0.0
18 0 50 100 150 200 250
19 Time (s)
20
21
22
23
Figure 8 Concentration-time profile of granular paracetamol in water/IPA (80:20)
24
25
26 (Temp=40°C, solid feed rate = 2.5 g min-1).
27
28
29
30
31
32 Table 3 Correlations between concentrations and liquid flow rate at fixed solid feed rates
33
34
35 Solid Liquid Target Actual Maximum %
36 flow flow Concentration Concentration Dissolution Dissolved
37 rate rate at full (g solute/100 g Achieved
38 (g/min) (g/min) dissolution (g solvent)
39 solute/100 g
40
41
solvent)
42
43
2.5 15 10.7 10.6 Yes 99
44 (saturated)
45
46
47
2.5 20 10.7 9.6 No 90
48 (saturated)
49
50 2.5 25 10.0 8.4 No 84
51
52 2.5 30 8.3 7.7 No 93
53
54 2.5 35 7.1 6.5 No 92
55
56
57
58
59
17
1
2
3
4
5
6
7 Effect of Solid Feed Rates
8
9
10 At a fixed liquid flow rate of 30 g min-1 (at which full dissolution was not achievable) and at a
11
12 fixed solution temperature of 40 ˚C, Figure 9 shows the effect of varying solid feed rates on the
13
14
15
dissolution concentration. Increasing the solid feed rate increases the concentration of the output
16
17 solution but does not achieve full dissolution of the solute. Decreasing the solid feed rate from
18
19 3.33 g min-1 to 1.05 g min-1 decreases the saturation level of the solution and the degree of screw
20
21
fill. The TSE is a starve fed system hence when the throughput is decreased at constant rpm more
22
23
24 mixing occurs as the materials being processed have a longer residence time in the mixing
25
26 elements. Full dissolution was achieved at 1.67 g min-1. Table 4 shows the correlations between
27
28 the solid/liquid feed rates and concentrations.
29
30
31
32 Drug development studies are often carried out with limited amount of materials, the lowest feed
33
34 rate of solids is thus of significant interest for this type of work. Our tests show that the lowest
35
36 feed rate of solids for the TSE producing consistent concentrations measurements at the exit was
37
38
39
1.67 g min-1. Observations in this study indicate more variability in the exit concentration with
40
41 decreasing feed rate.
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
18
1
2
3
12.0
4
5 1.05 g/min 1.67 g/min 2.5 g/min 3.33 g/min
6 10.0
7
8 8.0
9
10 6.0
11
12
4.0
13
14
15 2.0
16
17 0.0
18 0 50 100 150 200
19 Time (s)
20
21
22
23
Figure 9 Concentration-time profile of granular paracetamol in water/IPA (80:20)
24
25
26 (Temp=40°C, liquid flow rate = 30 g min-1)
27
28
29
30
31
32 Table 4 Correlations between concentrations and solid feed rate at fixed liquid flow rates
33
34
35
36 Solid Liquid Target Actual Maximum %
37
38 feed flow Concentration at Concentration Dissolution Dissolved
39
40 rate rate full dissolution (g (g solute/100 g Achieved
41
42
43 (g/min) (g/min) solute/100 g solvent) solvent)
44
45
46 3.33 30 10.7 (saturated) 8.7 No 81
47
48
49 2.5 30 8.3 7.6 No 92
50
51
52
53 1.67 30 5.6 5.7 Yes 100
54
55
56
57
58
59
19
1
2
3 1.05 30 3.5 3.5 Yes 100
4
5
6
7
8
9
10 Effect of Screw Speed
11
12
13 The screw in the TSE conveys the solids forward and the speed of which can affect dissolution
14
15
16 rate. The investigations were carried out for the screw speed from 100 to 500rpm (50 rpm
17
18 resulted in accumulation of solids at the input port) at a fixed liquid flow rate of 30 g min-1, a
19
20 fixed solid feed rate of 2.5 g min-1 and a solution temperature of 40˚C. The target concentration
21
22
23
for full dissolution at the set feed rates is 8.3 g (paracetamol)/100 g (solvent). While shear
24
25 mixing and power consumption intensify with the increase of the screw speed, allowing solute
26
27 molecules to encounter fresh solvent molecules faster, dissolution rate does not change
28
29
significantly and complete dissolution of granular paracetamol was not achieved by increasing
30
31
32 the screw speed alone. This is due to the fact that the residence time of the solute within the
33
34 barrel is reduced as the screw speed is increased (see Table 5).
35
36
37 Table 5 Effect of screw speed on residence time
38
39
40
41 Screw Speed (rpm) Mean Concentration of Paracetamol in Residence
42
43 Solution (g 100 g-1) Time (s)
44
45
46 100 7.9 13.2
47
48
49
50 200 7.1 8.8
51
52
53 300 7.5 6.9
54
55
56
57
58
59
20
1
2
3 400 7.8 6.7
4
5
6
7 500 8.1 6.0
8
9
10
11
12
13 Effect of screw configuration
14
15
16
17 The twin screws in the TSE are made up of individual elements of either concave conveying or
18
19 bi-lobe mixing (see Figure 10A), delivering different shear energy to the materials. The effect of
20
21 the screw configuration on dissolution was investigated by using one, two and three mixing
22
23
elements at the discharge (left) end of the screws (Figure 10B), i.e. at Ports 5 and 6 (Figure 4).
24
25
26 The dissolution profile is shown in Figure 11. Adding mixing elements to the screw
27
28 configuration increases dissolution, more for Ports 6-5 than the earlier ones, due to the increase
29
30 in shear mixing. The impact on residence time of solute within the barrel was < 2 seconds.
31
32
33 Complete dissolution of the granular paracetamol was not attained for any mixing elements,
34
35 however more consistent concentration of solution is obtained with more mixing elements.
36
37
38
39
40 Conveyor
41 element
42 0° 90° mixing
43 mixing element
44 element
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
21
1
2
3 (A)
4
5
6
7
8
9
10
11
12
13
14
15 Mixing Conveying
16
17
18
19 (B)
20
21
22
23
Figure 10 (A) - Conveying and Mixing Elements, (B) twin screws within the barrel
24
25
26
27
28
29
30
31
32
33 8.0
34 Port 4 Port 3 Port 2
7.0 Port 5
35 Port 6
36 6.0
37
38 5.0
39 4.0
40
41 3.0 No Mix
42 1 Mix
43 2.0
44 2 Mix
1.0
45 3 Mix
46 0.0
47 0 2 4 6 8 10 12 14
48 Time (s)
49
50
51
52
53 Figure 11 Effect of adding mixing elements on dissolution profiles (Temperature = 40˚C, solid
54
55 feed rate = 2.5 g min-1, liquid flow rate = 30 g min-1, screw speed = 100 rpm)
56
57
58
59
22
1
2
3
4
5
6
7 Effect of Barrel Temperature
8
9
10 The barrel is effectively a heating jacket and temperatures in each of the sections (Port 2 to
11
12 Port 6) can individually or collectively be controlled. Figure 12 shows the effect of barrel
13
14
15
temperature on dissolution from room temperature to 75°C. It is expected that an increase in
16
17 barrel temperature enhances dissolution and full dissolution of the granular paracetamol was
18
19 achieved when the barrel temperature reached 60°C.
20
21
22
23 30
24 Solubility Curve
25 25
26
27 20
28
29 15
30
31 10
32 R² = 0.9976
33 5
34
35 0
36 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0
37 Solution Temperature (˚C)
38
39
40
41 Figure 12 Effect of barrel temperature on dissolution of paracetamol (solid feed rate = 2.5 g
42
43
44
min-1, liquid flow rate = 30 g min-1, screw speed = 100 rpm), Solubility curve overlaid
45
46
47
48
49
50 In summary, the flow rates of either liquid or solid together with the barrel temperature can lead
51
52 to complete dissolution of granular paracetamol.
53
54
55
56
57
58
59
23
1
2
3 Batch v Continuous Dissolution
4
5
6
7 Dissolution of solid materials in a batch vessel is the current norm for feeding in the
8
9 pharmaceutical industry, we carried out the same dissolution experiments in a stirred tank (temp
10
11 = 40 ˚C, agitation rate = 750 rpm, mass of solute = 83 g, mass of solvent = 1000 g) and compare
12
13
14
the dissolution profiles in Figure 13. It is clear that dissolution is much faster in the TSE than
15
16 that in the stirred tank vessel, e.g. 13 seconds to dissolve 83g of paracetamol vs. 126 seconds to
17
18 dissolve the same. The increase in the dissolution rate in the TSE is due to much more aggressive
19
20
local shear mixing and higher thermal energy generated by the rotation of the screws 32. This also
21
22
23 delivers good uniformity of the solute in the solution.
24
25
26
27
28
29
120
30
31 Solubility Limit
32 100 (110g/1000g)
Concentration (g/1000g)
33
34 80
35
36 60
37
38
40
39 Stirred Tank
40
41 20 Twin Screw
42
43 0
44 0 50 100 150 200 250
45 Time (s)
46
47
48
49
Figure 13 Dissolution profile of granular paracetamol in water/IPA (80:20) at 40˚C according
50
51
52 to two methods
53
54
55
56
57
58
59
24
1
2
3 Table 6 Dissolution rate constants from a stirred tank vessel and a twin screw extruder
4
5
6
7 Dissolution Rate Dissolution rate constant per
8
9 constant, k (s-1) power density (m3 W-1 s-1)
10
11
12 Stirred Tank 0.0175 5.48 × 10-6
13
14
15
16 Twin Screw Extruder 0.0441 7.39 × 10-6
17
18
19
20
21
22
23
The dissolution rate constants are given in Table 6 for both devices, however power density
24
25 should be the basis for such a comparison. The power density of stirred tanks is well reported 33,
26
34
27 as:
28
29
30
31
32
33

34 = (1)
35
36
37
38
39
40
41 Where P/V is the power density (W m-3), ρ the fluid density (kg m-3 at 40 °C), Ns the speed of
42
43
44
the stirrer (rps), DS the diameter of the stirrer (m), VL the volume of liquid in the STC (m3) and
45
46 PO the dimensionless power number of the agitator, which was estimated as 2.3 based on data
47
48 presented by Nienow and Miles 35 for the type of impellor used in our work.
49
50
51
52
53
54
55
56
57
58
59
25
1
2
3 Previous estimations of power density in a twin screw extruder36, 37
were based on a non-
4
5
6 isothermal melt and a non-Newtonian fluid, covering various conditions including melt
7
8 temperatures, feed rates and scale of equipment. However, the definition of power density from
9
10 previous work differs from ours in that we estimate the power dissipated into the liquid or power
11
12
13
experienced by the liquid, not the power input by the motor as in previous power consumption
14
15 calculations. By treating the twin screw extruder as a stirred tank working horizontally, we could
16
17 estimate the power dissipation as follows. The diameter of the stirrer in eq. (1) becomes the
18
19
diameter of the screw, while the rotational speed remains the same. There is no similar power
20
21
22 number for the twin screw, but the helical screw impellor is the closest. In terms of the liquid
23
24 volume, only half of the liquid in the TSE is experiencing the effect of shearing imposed by the
25
26 twin screws at any given time, as the screw pitch is only 45% filled38.
27
28
29
30
31 Applying these values to eq. (1), the power density for the stirred tank and the TSE are 3196 and
32
33 5968 W m-3 respectively. It is observed that faster dissolution rate is achieved in the TSE
34
35
36
however the power density alone is an insufficient descriptor for comparison with the stirred
37
38 tank. Other factors influencing the dissolution rate include the short mass transfer distances and
39
40 the efficient shear mixing in the TSE.
41
42
43
Start up and shut down losses encountered in this work were low. It took approximately 90
44
45
46 seconds to reach steady state on starting the equipment from empty which equates to losses of
47
48 approximately 3 g solid and 50 ml solvent. On shutdown the material remaining in the barrel
49
50 equates to losses of approximately 1g of solid and 8 ml of solvent.
51
52
53
54 CONCLUSIONS
55
56
57
58
59
26
1
2
3 In this work, we have demonstrated that the twin screw extruder enables the decoupling of
4
5
6 liquid from the solid feed, thus eliminating any potential fouled solid feeding even at low solid
7
8 feed, e.g. 1.67 g min-1. It also allows the controlled and synchronised input flows, together with
9
10 intense mixing, to deliver either a dissolved solution or suspension of controlled composition
11
12
13
ready for the next unit operation in the process train. These are the novelties of this work, they
14
15 fill the gap for continuous reaction and crystallisation in the pharmaceutical industry. This study
16
17 also highlights the flexibility of the TSE to cope with different raw material feed stocks;
18
19
achieving full dissolution of powder paracetamol within the residence time of the TSE and
20
21
22 achieving full dissolution of granular paracetamol by altering key variables such as solid or
23
24 liquid feed rates and barrel temperature.
25
26
27
28
29
30
31 We have, for the first time, proposed an alternative method of estimating the power density for
32
33 the TSE when used as a continuous dissolution feed stream. This enables a fair comparison of
34
35 the dissolution rates between two devices, i.e. TSE and stirred tank. The faster dissolution rate in
36
37
38
the TSE is associated with higher power dissipation generated by the aggressive shear mixing
39
40 and thermal energy within the barrel. The dissolution rate constant per power density in the TSE
41
42 is slightly more favourable than that in the stirred tank. In addition, the variability in the output
43
44
concentration in the former (1%) is much less than that in the latter (up to 10%)39 . We are
45
46
47 carrying out further tests using a range of diverse materials and will report our results in a
48
49 separate communication.
50
51
52
53
54
55
56 AUTHOR INFORMATION
57
58
59
27
1
2
3 Corresponding Author
4
5
6 * Xiong-Wei Ni
7
8
9 tel: 00441314513781; fax: + 441314513129; email: x.ni@hw.ac.uk
10
11
12
13
Author Contributions
14
15 The manuscript was written through contributions of all authors. All authors have given approval
16
17 to the final version of the manuscript. ‡These authors contributed equally.
18
19
20
21 Funding Sources
22
23 UKRPIF (UK Research Partnership Institute Fund) capital award, SFC ref. H13054, from the
24 Higher Education Funding Council for England (HEFCE) and GSK.
25
26
27
28
29
30 ACKNOWLEDGMENT
31
32
33 Authors would like to thank EPSRC, the Centre for Innovative Manufacturing in Continuous
34
35 Manufacturing and Crystallisation, and GSK for their support and funding (EP/K503289/1).
36
37
38
Particular thanks and acknowledgement goes to the mentoring support provided by Gareth
39
40 Alford, GSK. The authors would like to acknowledge that this work was carried out in the
41
42 CMAC National Facility, housed within the University of Strathclyde’s Technology and
43
44
Innovation Centre, and funded with a UKRPIF (UK Research Partnership Institute Fund) capital
45
46
47 award, SFC ref. H13054, from the Higher Education Funding Council for England (HEFCE).
48
49
50
51
52
53 REFERENCES
54
55
56
57
58
59
28
1
2
3 1. Ni, X., Continuous Oscillatory Baffled Reactor Technology. Innovations in
4
Pharmaceutical Technology 2006, 20, 90-96.
5
6 2. McGlone, T.; Briggs, N. E. B.; Clark, C. A.; Brown, C. J.; Sefcik, J.; Florence, A. J.,
7 Oscillatory Flow Reactors (OFRs) for Continuous Manufacturing and Crystallization. Organic
8 Process Research & Development 2015, 19, 1186-1202.
9 3. O’Connor, T.; Lee, S., Chapter 37 - Emerging Technology for Modernizing
10 Pharmaceutical Production: Continuous Manufacturing A2 - Qiu, Yihong. In Developing Solid
11 Oral Dosage Forms (Second Edition), Chen, Y.; Zhang, G. G. Z.; Yu, L.; Mantri, R. V., Eds.
12
13
Academic Press: Boston, 2017; pp 1031-1046.
14 4. Simon Lawton, G. S., Phil Shering, Continuous Crystallization of Pharmaceuticals Using
15 a Continuous Oscillatory Baffled Crystallizer. Organic Process Research & Development 2009,,
16 13, 1357–1363.
17
18 5. Ferguson, S.; Morris, G.; Hao, H. X.; Barrett, M.; Glennon, B., Characterization of the
19 anti-solvent batch, plug flow and MSMPR crystallization of benzoic acid. Chemical Engineering
20 Science 2013, 104, 44-54.
21
6. Salvatore, M.; L., H. P.; Haitao, Z.; Richard, L.; Brahim, B.; I., B. P.; D., B. R.; L., C. C.;
22
23 B., E. J. M.; F., J. T.; F., J. K.; S., M. A.; L., T. B., End‐to‐End Continuous Manufacturing of
24 Pharmaceuticals: Integrated Synthesis, Purification, and Final Dosage Formation. Angewandte
25 Chemie International Edition 2013, 52, 12359-12363.
26 7. Bernhard, G.; David, C.; Oliver, K. C., Continuous‐Flow Technology—A Tool for the
27 Safe Manufacturing of Active Pharmaceutical Ingredients. Angewandte Chemie International
28 Edition 2015, 54, 6688-6728.
29
30
8. Besenhard, M. O.; Neugebauer, P.; Scheibelhofer, O.; Khinast, J. G., Crystal Engineering
31 in Continuous Plug-Flow Crystallizers. Crystal Growth & Design 2017, 17, 6432-6444.
32 9. Wood, A., Generic Batch Procedures for Flexible Manufacturing. Control Engineering
33 2009, 56, P1-P5.
34 10. Hörmann, T.; Suzzi, D.; Khinast, J. G., Mixing and Dissolution Processes of
35 Pharmaceutical Bulk Materials in Stirred Tanks: Experimental and Numerical Investigations.
36
Ind. Eng. Chem. Res. 2011, 50, 12011-12025.
37
38
11. Paul, E. L.; Midler, M.; Sun, Y., Mixing in the Fine Chemicals and Pharmaceutical
39 Industries. In Handbook of Industrial Mixing, John Wiley & Sons, Inc.2004; pp 1027-1069.
40 12. Atiemo-Obeng, V. A.; Penney, W. R.; Armenante, P., Solid–Liquid Mixing. In
41 Handbook of Industrial Mixing, John Wiley & Sons, Inc.2004; pp 543-584.
42 13. Hemrajani, R. R.; Tatterson, G. B., Mechanically Stirred Vessels. In Handbook of
43 Industrial Mixing, John Wiley & Sons, Inc.2004; pp 345-390.
44
14. Thakur, R. K.; Vial, C.; Nigam, K. D. P.; Nauman, E. B.; Djelveh, G., Static Mixers in
45
46 the Process Industries—A Review. Chemical Engineering Research and Design 2003, 81, 787-
47 826.
48 15. Ghanem, A.; Lemenand, T.; Della Valle, D.; Peerhossaini, H., Static mixers:
49 Mechanisms, applications, and characterization methods – A review. Chemical Engineering
50 Research and Design 2014, 92, 205-228.
51 16. Baker, J. R., Motionless mixers stir up new uses. Chemical Engineering Progress 1991,
52
53
87, 32-38.
54 17. Crowley, M. M.; Zhang, F.; Repka, M. A.; Thumma, S.; Upadhye, S. B.; Battu, S. K.;
55 McGinity, J. W.; Martin, C., Pharmaceutical applications of hot-melt extrusion: part I. Drug
56 development and industrial pharmacy 2007, 33, 909-26.
57
58
59
29
1
2
3 18. Repka, M. A.; Battu, S. K.; Upadhye, S. B.; Thumma, S.; Crowley, M. M.; Zhang, F.;
4
Martin, C.; McGinity, J. W., Pharmaceutical applications of hot-melt extrusion: Part II. Drug
5
6 development and industrial pharmacy 2007, 33, 1043-57.
7 19. Patil, H.; Tiwari, R. V.; Repka, M. A., Hot-Melt Extrusion: from Theory to Application
8 in Pharmaceutical Formulation. AAPS PharmSciTech 2015.
9 20. Martinez-Marcos, L.; Lamprou, D. A.; McBurney, R. T.; Halbert, G. W., A novel hot-
10 melt extrusion formulation of albendazole for increasing dissolution properties. International
11 journal of pharmaceutics 2016, 499, 175-185.
12
13
21. Vercruysse, J.; Córdoba Díaz, D.; Peeters, E.; Fonteyne, M.; Delaet, U.; Van Assche, I.;
14 De Beer, T.; Remon, J. P.; Vervaet, C., Continuous twin screw granulation: Influence of process
15 variables on granule and tablet quality. European Journal of Pharmaceutics and
16 Biopharmaceutics 2012, 82, 205-211.
17 22. Mendez Torrecillas, C.; Halbert, G. W.; Lamprou, D. A., A novel methodology to study
18 polymodal particle size distributions produced during continuous wet granulation. International
19
journal of pharmaceutics 2017, 519, 230-239.
20
21 23. Dhenge, R. M.; Cartwright, J. J.; Doughty, D. G.; Hounslow, M. J.; Salman, A. D., Twin
22 screw wet granulation: Effect of powder feed rate. Advanced Powder Technology 2011, 22, 162-
23 166.
24 24. Dhenge, R. M.; Cartwright, J. J.; Hounslow, M. J.; Salman, A. D., Twin screw wet
25 granulation: Effects of properties of granulation liquid. Powder Technology 2012, 229, 126-136.
26 25. Saleh, M. F.; Dhenge, R. M.; Cartwright, J. J.; Hounslow, M. J.; Salman, A. D., Twin
27
screw wet granulation: Binder delivery. International journal of pharmaceutics 2015, 487, 124-
28
29 134.
30 26. Cartwright, J. J.; Robertson, J.; D'Haene, D.; Burke, M. D.; Hennenkamp, J. R., Twin
31 screw wet granulation: Loss in weight feeding of a poorly flowing active pharmaceutical
32 ingredient. Powder Technology 2013, 238, 116-121.
33 27. Wang, Y.; Li, T.; Muzzio, F. J.; Glasser, B. J., Predicting feeder performance based on
34 material flow properties. Powder Technology 2017, 308, 135-148.
35
36
28. William E. Engisch, F. J. M., Method for characterization of loss-in-weight feeder
37 equipment. Powder Technology 2012, 228, 395–403.
38 29. Rohani, H. H. a. S., Measurement and Prediction of Solubility of Paracetamol in Water-
39 Isopropanol Solution. Part 1. Measurement and Data Analysis. Organic Process Research &
40 Development 2006, 10, 1101-1109.
41 30. McLaughlin, A. M.; Robertson, J.; Ni, X., Material characterisation and parameter effects
42
on bulk solid dissolution rate of paracetamol in a stirred tank vessel using an in situ UV-ATR
43
44 probe. International Journal of Engineering Research & Science 2018, 4, 10-20.
45 31. Tingstad, J. E.; Riegelman, S., Dissolution Rate Studies I: Design and Evaluation of a
46 Continuous Flow Apparatus. Journal of Pharmaceutical Sciences 1970, 59, 692-696.
47 32. Thiry, J.; Krier, F.; Evrard, B., A review of pharmaceutical extrusion: critical process
48 parameters and scaling-up. International journal of pharmaceutics 2015, 479, 227-40.
49 33. Sinnott, R. K., Equipment Selection, Specification and Design. In Coulson and
50
Richardson's Chemical Engineering Volume 6 - Chemical Engineering Design, Third ed.;
51
52 Sinnott, R. K., Ed. Butterworth-Heinemann: Oxford, 1999; pp 471-473.
53 34. Garside, J.; Davey, R. J., Invited review secondary contact nucleation: kinetics, growth
54 and scale-up. Chemical Engineering Communications 1980, 4, 393-424.
55
56
57
58
59
30
1
2
3 35. Nienow, A.; Miles, D., Impellor power numbers in closed vessels. Ind. Eng. Chem.
4
Process Des. Dev. 1971, 10, 41-43.
5
6 36. Meijer, H. E. H.; Elemans, P. H. M., The modeling of continuous mixers. Part I: The
7 corotating twin‐screw extruder. Polymer Engineering & Science 1988, 28, 275-290.
8 37. Kruijt, P. G. M.; Galaktionov, O. S.; Peters, G. W. M.; Meijer, H. E. H., The Mapping
9 Method for Mixing Optimization. International Polymer Processing 2001, 16, 161-171.
10 38. Martin, C., Melt Extrusion. In Materials, Technology and Drug Product Design, Repka,
11 M. A. L., N. ; DiNunzio, J., Ed. AAPS Advances in the Pharmaceutical Sciences Series 9:
12
13
Springer, 2013; pp 47-79.
14 39. Faanes, A.; Skogestad, S., Buffer Tank Design for Acceptable Control Performance. Ind.
15 Eng. Chem. Res. 2003, 42, 2198-2208.
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
31

Twin Screw Reactor

Uploaded by

Copyright:

Available Formats

You might also like

Twin Screw Reactor

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Twin Screw Reactor

Uploaded by

Copyright:

Available Formats

Subscriber access provided by UNIV OF WESTERN ONTARIO

is published by the American Chemical Society. 1155 Sixteenth Street N.W.,

You might also like