- Email: [email protected]
The best screw shape for fine zinc oxide particles feeding
Advanced Powder Technology 23 (2012) 372–379
Contents lists available at ScienceDirect
Advanced Powder Technology journal homepage: www.elsevier.com/locate/apt
Original Research Paper
The best screw shape for fine zinc oxide particles feeding M. Barati Dalenjan ⇑, E. Jamshidi, H. Ale Ebrahim Department of Chemical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Petrochemical Center of Excellency, Tehran 15875-4413, Iran
a r t i c l e
i n f o
Article history: Received 8 January 2011 Received in revised form 1 May 2011 Accepted 9 May 2011 Available online 26 May 2011 Keywords: Screw feeder Zinc oxide powder Mass flow rate Particle size distribution
a b s t r a c t One of the most important requirements of particles feeding is a very homogeneous mass flow of fine particles. Screw feeders are among equipment used to feed particles in many industries. In this study, the performance of four single screw feeders with different pitch and blade shape screws has been studied experimentally for steady and stable feeding of fine zinc oxide particles at flow rates corresponding to laboratory scale range (under 10 g/min). The following results were obtained from this investigation: (I) the sticking problem of fine particles in angular pitch screw was solved by changing the pitch shape to circular; (II) the distribution of the fluctuations and their intensity in mass flow rates also lessen by using the circular pitch and a thin blade screw in place of the circular pitch and a thick blade screw. Also for both feeders, the feeding was often interrupted in low flow rates, but it will be disappeared by increasing the flow rate. Furthermore, the results of experiments show that the performance of the circular pitch and a thin blade screw feeder was better than other screw feeders and able to both swirl and mix the particles with different characteristics and reduce the mean aggregate size of the particle size distribution (PSD) when transmitting the zinc oxide particles. Ó 2011 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved.
1. Introduction Fine particles feeding are widely used, especially in polymer, chemical and food industries [1–3]. One of the most important applications is feeding of zinc oxide particles to a solar reactor for decomposition to zinc as an energy carrier and hydrogen producer by water splitting method [4–6]. In recent years there were considerable efforts to provide the uniform fine particles feeding to gas–solid two phase flow reactor [2,7]. On laboratory scale, it is particularly difficult to achieve steady and stable feeding because the fine particles are very compressible. Therefore, weighing and transmitting (metering and handling) in low flow rates is difficult [8,9]. The fine (less than 1 lm) ZnO particles are very cohesive and sticking, and classified as Geldart group C particles [10–13]. Group C particles tend to cling to each other as a consequence of the interparticle forces and form agglomerates [2,11,13]. Therefore, it was expected that we have coarse agglomerations in the initial ZnO powder. Furthermore, the fine particles feeding are very sensitive because a small variation in the gas–solid two phase flow will change the reaction process. There are many types of feeders available for feeding of particles in the literature. The screw feeders are among the most widely used types of volumetric feeders to particles feeding. They are a ⇑ Corresponding author. Tel.: +98 021 64543177; fax: +98 021 66405847. E-mail address: [email protected] (M. Barati Dalenjan).
popular choice for many applications because of that can easily be modified to fit the type of particles and process involved [10,14,15]. Some advantages of screw feeders include particles transmitting, stirring, mixing [3,14,16,17], and reliable to low and high feeding rates [10,14]. The drawbacks of screw feeders are as follows: Interlocking or cohesive arch of particles forming above the screw in the hopper [10,14]. ‘Clogging’ takes place when particles stick to the root corners of the screw flight. ‘Logging’ is a term used when the total volume of the screw is filled and rotates [14]. Instantaneous fluctuations of mass flow rate [18,19]. Promote the aggregation of particles that then drop off the end of the screw as particle aggregates [2]. In critical process-control applications, time intervals for sampling should be no longer than residence time in the actual process. This can be a problem with processes that have short residence times, particularly gas–solid reactors [10]. Complete conversion of gas–solid reactions may be achieved by increasing the operating temperature and residence time [20]. However, the residence time of gas–solid reactions is usually in the order of several seconds [21–23]. The novelty of the present work is using of a circular pitch and a thin blade screw for improving the performance of a screw feeder with respect to the above mentioned problems for steady and
0921-8831/$ - see front matter Ó 2011 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved. doi:10.1016/j.apt.2011.05.001
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
373
Table 1 Properties of zinc oxide and potassium dichromate powders. I. Properties of zinc oxide powder CAS number Bulk density True density Molar mass Particle size
1314-13-2 0.8 g/cm3 5.606 g/cm3 81.4 g/mol Less than 1 lm
II. Properties of potassium dichromate powder CAS number Bulk density True density Molar mass Mean particle size
7778-50-9 1.6 g/cm3 2.68 g/cm3 294.18 g/mol 80 lm Fig. 2. A schematic view of experimental setup.
stable feeding of fine particles, particularly for injection into gas– solid two phase flow reactors. To this end, tests were carried out on four single screw feeders with different pitch and blade shape screws, in the same condition, in order to find out the best shape of screw. Moreover, other important issues for a screw feeder consisting such as the mixing performance of particles having different characteristics, and reduction of mean aggregate size of the particle size distribution (PSD) have been considered for the best screw feeder.
on the screw. An agitation device, preventing bridging inside of the hopper, produces a rather uniform bulk density and supports the powder flow into the screw. Particles transmit from hopper into a fluid flow or process line by means of screw revolution. The rotational speed of screw is controlled electrically by means of the DC motor speed controlling unit, and the speed of screw could thus be changed. To determine the mass flow rate and the fluctuations of the screw feeding, a balance is put right underneath the screw outlet. The effect of feeders, especially for the production of homogeneous mass flow rate on laboratory scale was studied.
2. Experimental 2.1. Material Zinc oxide (ZnO) powder is used as the feed test powder. The powder is 99.7% pure and has a reported primary particle size of less than 1 lm. Table 1I shows the properties of ZnO powder. An Analysette 22 Micro Tec plus from Fritsch system (with wet dispersion unit by using a combination of a centrifugal pump in combination with ultrasonic) was used to measure the particle size distribution. The cumulative volume distribution of ZnO powder is shown in Fig. 1. The fresh powder is heated to 120 °C for 20 min in an oven before the tests. Therefore, the moisture of powder in all tests is the same. 2.2. Experimental method Fig. 2 shows a schematic of experimental setup. The ZnO powder is filled manually into the hopper, and the hopper is continuously refilled during the test to provide a uniform head pressure 110 100
Cumulative Volume (%)
90 80 70 60 50 40 30 20 10 0 0.5
5
50
Particle Size (µm) Fig. 1. The cumulative volume distribution of ZnO powder.
Fig. 3. Screws employed to feeding tests: (a) rectangular pitch screw, (b) trapezoidal pitch screw, (c) circular pitch and a thick blade screw, (d) circular pitch and a thin blade screw.
374
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
Fig. 3 shows four single screws which were tested in feeding experiments. There are two types of screws, one type is the angular pitch screws and another one is the circular pitch screws. The angular pitch screws include a rectangular pitch screw (Fig. 3a), and a trapezoidal pitch screw (Fig. 3b). The circular pitch screws comprise a circular pitch and a thick blade screw (Fig. 3c), and a circular pitch and a thin blade screw (Fig. 3d). Table 2 shows the dimensions of main parts for the four screws in Fig. 3. The main differences among these screws were in both pitch forming and thickness of blade. The screws were tested at rotational speeds of 2 1/2 to 17 3/4 rpm. The tests were done to measure the particles mass flow rate and its fluctuations to find out the best screw which tends to reduction of feeding fluctuations. The performance of a screw feeder is normally expressed in statistical terms based on the uniformity of the delivery rate as a function of time. The mass delivered in a given time interval is measured for a number of time intervals. The mean and standard deviation of the masses is calculated, and the value of standard deviation divided by the mean provides a percentage used in describing feeder performance. Because of the residence time of gas–solid reactions is about several seconds, time interval was selected as 1 s. The balance used for weighing the particles mass gave five signals of measurements per second which sent to the serial port of a PC and stored. For each screw the mass of particles fed in each 1 s interval was measured and compared.
3. Results and discussion To examine the performance of each screw, especially with respect to production of a uniform mass flow rate on laboratory scale, two different test series were conducted. Firstly, the performance of each screw to particles feeding was studied to measure the mass flow rate for 30 min at different rotational speeds, and secondly, feeding fluctuations of the best screws were examined. 3.1. Rectangular pitch screw feeder tests Fig. 4 shows the feeding curves for rectangular pitch screw feeder when the rotational speed of screw increases from 4 1/4 to 17 3/ 4 rpm. It shows that the shapes of all curves are similar and can divide feeding curves into three regions: feed region 1, is occurring in initial 5 min that the slopes of the curves are high, and there is no variation in feeding rates. Then feed region two occurs until about
Table 2 The dimensions of main parts for the four screws. Type of screw (mm)
Rectangular pitch screw
Trapezoidal pitch screw
Circular pitch and a thick blade screw
Circular pitch and a thin blade screw
The screw diameter The shaft diameter The pitch length The blade thickness
26
22
28
26
21
18
20
19
3
4
7.5
7
2.5
4
7.5
1
Powder Mass (grams)
16
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/1 rpm
12
8
4
0 0
5
10
15
20
25
30
Time (min) Fig. 4. Feeding curves for rectangular pitch screw feeder.
35 Fig. 5. Screws after the feeding tests: (a) rectangular pitch screw, (b) trapezoidal pitch screw, (c) circular pitch and a thick blade screw, (d) circular pitch and a thin blade screw.
375
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
25 min, where the downward slope is distinct because of cylinder obstruction or logging. On the one hand, length of cylinders was selected to be short in order to prevent probable obstructions. Accordingly, this decreasing of the mass feeding rate is for sticking the particles into the pitch and reduction of the screw volume. Feed region three is after 25 min until 30 min, where the slopes of the curves are almost zero. It is shown that the mass feeding rate has plunged and the pitch of screw is approximately filled with impacted particles. The particles rotate with screw because the forces promoting the rotation exceed the forces restraining their movement. Fig. 5a shows the rectangular pitch screw after the test.
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/4 rpm
3.2. Trapezoidal pitch screw feeder tests Fig. 6 shows the feeding curves for trapezoidal pitch screw feeder. Unlike rectangular pitch screw feeder, it is clear in Fig. 6 that the curves can be split into two feed regions. An initial fast rate of mass flow occurs in first 10 min, feed region 1, where there is no variation in feeding rates observed. Then, the curves switched to slower mass flow rate, but the slopes of the curves are not zero, feed region 2. This drop to feeding rate arose from cohering of particles on the screw shaft or sticking of particles in the angle of shaft and blade. It is clear that, by increasing the time of the tests, the feed region 3 of rectangular pitch screw feeder will occur and feeding will be stopped. Clogging and logging for the trapezoidal pitch screw after the test is shown in Fig. 5b.
12
2 1/2 rpm
16
Powder Mass Flow (g/min)
Powder Mass (grams)
10 12
8
4
8
6
4
2
0
0 0
5
10
15
20
25
30
0
35
4
12
16
20
Fig. 8. Relation between rotational screw speed and particles mass flow for circular pitch and a thick blade screw feeder.
Fig. 6. Feeding curves for trapezoidal pitch screw feeder.
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/4 rpm
8
2 1/2 rpm
350
Deviation from The Mean Mass Flow Rate (g/s) (%)
6
300
Powder Mass (grams)
8
Rotational Speed (rpm)
Time (min)
250 200 150 100
4
2
0
-2 50 -4
0
0 0
5
10
15
20
25
30
35
10
20
30
40
50
60
Time (sec)
Time (min) Fig. 7. Feeding curves for circular pitch and a thick blade screw feeder.
Fig. 9. Fluctuations at rotational speed of 2 1/2 rpm for circular pitch and a thick blade screw feeder.
376
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
3.3. Circular pitch and a thick blade screw feeder tests Feeding curves and the average particles mass flow rate versus the rotational speed for circular pitch and a thick blade screw feeder are shown in Figs. 7 and 8, respectively. Results indicate that, the performance of this feeder compared with two preceding screw feeders is very different. Fig. 7 shows that the rotational speeds of 2 1/2 up to 11 rpm can provide continuous feeding of particles from 1.3 to 7 g/min, respectively. But there are some dispersal in feeding at rotational speeds of 13 1/2 and 17 3/7 rpm. Also Fig. 8 confirms the linearity of particles mass flow rate from 2 1/2 to 11 rpm, not in 13 1/2 and 17 3/4 rpm. This problem in relatively high rotational speeds occurs because the evolution and transmitting speeds of screw is more than the entering speed of particles into the screw volume. So, particles do not have enough
time to fill the volume of screw. Moreover, the thick blade of screw excludes particles from entering into the screw volume. Fig. 5c shows the circular pitch and a thick blade screw after the test. Also it shows that the sticking problem of particles in the rectangular pitch screw and the trapezoidal pitch screw is solved by changing the pitch form of screw from angular to circular. The minimum mass flow rate (1.3 g/min) is corresponding to rotational speed of approximately 2 1/2 rpm and the maximum mass flow rate (9.5 g/min) is at 17 3/4 rpm. Figs. 9 and 10 present the fluctuations of the flow rates at rotational speeds of 2 1/2 and 17 3/4 rpm, respectively. The deviations from the mean flow rate are plotted versus time. In accordance with the 5 weight measurements per second, the arithmetic mean of five samples is taken to represent the actual deviation of the mass flow rate during the time period of 1 s. Both figures show
40
Powder Mass Flow (g/min)
Deviation from The Mean Mass Flow Rate (g/s) (%)
12 30 20 10 0 -10
10
8
6
4
-20 2 -30 0
20
40
60
0
Time (sec)
0
4
8
12
16
20
Rotational Speed (rpm)
Fig. 10. Fluctuations at rotational speed of 17 3/4 rpm for circular pitch and a thick blade screw feeder.
Fig. 12. Relation between rotational screw speed and particles mass flow for circular pitch and a thin blade screw feeder.
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/4 rpm
8
2 1/2 rpm
350
R² = 0.9998
250
R² = 0.9998
200
R² = 0.9998 R² = 0.9999
150
R² = 0.9995
100
R² = 0.9998
50
6
Deviation from The Mean Mass Flow Rate (g/s) (%)
Powder Mass (grams)
300
4
2
0
-2
R² = 0.9998
-4 0
0 0
5
10
15
20
25
30
35
10
20
30
40
50
60
Time (sec)
Time (min) Fig. 11. Feeding curves for circular pitch and a thin blade screw feeder.
Fig. 13. Fluctuations at rotational speed of 2 1/2 rpm for circular pitch and a thin blade screw feeder.
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
where x is the arithmetic mean of the values xi, defined as:
30
x ¼
20
Deviation fromThe Mean Mass Flow Rate (g/s) (%)
377
N x1 þ x2 þ þ xN 1 X xi : ¼ N i¼1 N
The standard deviation can be used to provide a quantitative measure of the variations of particles feeding rate. The standard deviation of ZnO particles flow rate is calculated from the experimental data, which has increased from 0.058 (rotational speed of 2 1/2 rpm) to 0.164 g/s (rotational speed of 17 3/4 rpm).
10
0
3.4. Circular pitch and a thin blade screw feeder tests -10
-20
-30 0
10
20
30
40
50
60
Time (sec) Fig. 14. Fluctuations at rotational speed of 17 3/4 rpm for circular pitch and a thin blade screw feeder.
typical characteristics of screw feeding: counting the peaks during 1 min gives a number equal to the rotational speed of the screw. This implies that there is a direct relation between the screw speed and the frequency of fluctuations. Moreover, the magnitude of the deviations must increase by increasing the screw speed, which is seen when Figs. 9 and 10 are compared. In Fig. 9, there are some interruptions in feeding which have displayed by circles, but Fig. 10 shows that the problem of pulsing is solved by increasing the screw speed. The standard deviation is defined as follows:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u N u1 X r¼t ðx xÞ2 ; N i¼1
Feeding curves and the average particles mass flow rate versus the rotational speed for circular pitch and a thin blade screw feeder are shown in Figs. 11 and 12, respectively. The results are surprising because of the R2 value for all curves in Fig. 11 is more than 0.99, that shows the precision of experimental results. It means that, this feeder can continuously provide feeding of fine particles from 1.2 g/min (at minimum rotational speed) to 10.5 g/min (at maximum rotational speed). Fig. 12 shows that the problem of circular pitch and a thick blade screw feeder at high rotational speeds is solved by means of narrowing the blade. Since the delivery rate of circular pitch and a thin blade screw feeder responds to changes in rotational speed in a linear and predictable fashion, the process control will be simplified. Fig. 5d shows the circular pitch and a thin blade screw after the test and it is obvious that no particle sticks to the screw. Figs. 13 and 14 present the fluctuations of flow rates at rotational speeds of 2 1/2 and 17 3/4 rpm, respectively. A significant reduction of fluctuations can be noticed when comparing them with Figs. 9 and 10. The standard deviation at rotational speed of 2 1/2 rpm is reduced from 0.058 to 0.048 g/s, and at rotational speed of 17 3/4 rpm from 0.164 to 0.146 g/s. Moreover, the feeding interruption can be improved. The circles in Fig. 13 show that the interruptions in feeding are less than Fig. 9. Fig. 14 shows that the problem of pulsing in circular pitch and a thin blade screw feeder is solved by increasing the rotational speed of screw. It indicates that increasing the screw revolution speed causes more uniformity in the mass flow.
Fig. 15. Swirl and mix the particles with different characteristics, (a) under part of the screw was filled with zinc oxide particles, (b) upper part of the screw was filled with potassium dichromate particles, (c) and (d) are shown swirling and mixing of particles, (e) output of the screw feeder.
378
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
Fig. 16. SEM observations of ZnO particles, (a) particles before the test, (b) particles after shearing mechanically by the screw.
(scanning electron microscopy) tests. Fig. 16a and b shows the experimental micrographs of the particle aggregates before the test and after shearing by the screw, respectively. Fig. 17 was produced with SPIP software, and shows the results of the PSD data. When analyzing, it is apparent that the screw feeder leads to cause a mechanical shear of the ZnO particle aggregates, and the PSDs are shifted to a lower mean particle aggregate size. The mean aggregate sizes for the particle aggregates before and after the test are 65 and 43 lm, respectively. These values were determined from the curves area of Fig. 17 (by trapezoidal method on the y axis). Therefore, the calculated mean aggregate sizes are volumetric averages from SEM tests.
_
Normalize Volume Fraction (µm ¹)
0.005
0.004
b a
0.003
0.002
4. Conclusions
0.001
0 0
200
400
600
Particle Diameter (µm) Fig. 17. Volume PSDs for ZnO particles taken from two locations, particles before the test (curve a) and after shearing mechanically by the screw (curve b).
3.5. Swirl and mix the particles test The screw revolves to transmit particles in a feeder. Moreover, revolution of screw causes swirling and mixing of particles. Hence, zinc oxide and potassium dichromate particles were applied to investigate these two characteristics. Table 1II shows the properties of potassium dichromate powder. A circular pitch and a thin blade screw feeder was used, and at first the under part of the screw was filled with zinc oxide which is shown in Fig. 15a. Then the upper part of the screw was filled with potassium dichromate which is shown in Fig. 15b. Fig. 15c and d shows the swirling of zinc oxide and potassium dichromate particles and their mixing together with the screw revolution. Fig. 15e shows the output of the screw that testifies to the accuracy of the test. 3.6. Particle size distribution test The screw feeder is able to decrease the mean aggregate size of PSD. Hence, two locations on circular pitch and a thin blade screw feeder are identified to collect the PSD data to characterize the system. The first one is where the particles falls into the hopper before the test. The next PSD data is measured at the exit of the screw. At this point, the PSD data is obtained just after the particle aggregates are sheared mechanically by the screw. The scanning probe image processor (SPIP) software was used for obtaining of the volume fraction distribution of the particle aggregates from SEM
The steady and stable feeding of fine particles is very difficult because of the sticking property and compressibility of them. The present study examines four single screw feeders with different screws. The studies on the performance of the screw feeders showed that clogging and logging problems in screws are solved by selecting the circular pitch screw instead of angular pitch one. In addition, the circular pitch and a thin blade screw feeder can be used to feed fine particles more reliably and more steadily than other screw feeders. The circular pitch and a thin blade screw feeder are solely able to consistently shear fine agglomerates of particles down where the particle size is submicron. Moreover, this feeder can mix together particles with different characteristics. It can be finally suggested that the circular pitch and a thin blade screw feeder could improve the uniformity of the feeding process. References [1] M.Y. Gundogdu, Design improvements on rotary valve particle feeders used for obtaining suspended airflows, Powder Technology 139 (1) (2004) 76–80. [2] T.M. Francis, C.J. Gump, A.W. Weimer, Spinning wheel powder feeding device– fundamentals and applications, Powder Technology 170 (2006) 36–44. [3] K. Uchida, K. Okamoto, Measurement technique on the diffusion coefficient of powder flow in a screw feeder by X-ray visualization, Powder Technology 187 (2008) 138–145. [4] R. Müller, A. Steinfeld, H2O-splitting thermochemical cycle based on ZnO/Zn redox: quenching the effluents from the ZnO dissociation, Chemical Engineering Science 63 (2008) 217–227. [5] H.H. Funkea, H. Diaza, X. Lianga, C.S. Carneya, A.W. Weimer, P. Li, Hydrogen generation by hydrolysis of zinc powder aerosol, International Journal of Hydrogen Energy 33 (2008) 1127–1134. [6] C. Perkins, P.R. Lichty, A.W. Weimer, Thermal ZnO dissociation in a rapid aerosol reactor as part of a solar hydrogen production cycle, International Journal of Hydrogen Energy 33 (2008) 499–510. [7] T.M. Francis, P.B. Kreider, P.R. Lichty, A.W. Weimer, An investigation of a fluidized bed solids feeder for an aerosol flow reactor, Powder Technology (2009). doi:10.1016/j.powtec.2009.04.020. [8] T.W. Davies, The Production of concentrated powder suspensions at low flow rates, Powder Technology 42 (1985) 249–253. [9] L. Tang, W.Y. Chen, Improvements on a particle feeder for experiments requiring low feed rates, Review of Scientific Instruments 70 (7) (1999) 3143– 3144.
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379 [10] T.A. Bell, S.W. Couch, T.L. Kreiger, H.J. Feise, Screw feeders: a guide to selection use, Chemical Engineering progress 99 (2) (2003) 44–53. [11] C. Xu, J. Zhu, Parametric study of fine particle fluidization under mechanical vibration, Powder Technology 161 (2006) 135–144. [12] D. Geldart, Types of gas fluidization, Powder Technology 7 (1973) 285–297. [13] Z. Wang, M. Kwauk, H. Li, Fluidization of fine particles, Chemical Engineering Science 53 (3) (1998) 377–395. [14] L. Bates, Guide to the design, selection, application of screw feeders, First Published, Wiltshire, UK, 2000. [15] Y. Yu, P.C. Arnold, The influence of screw feeders on bin flow patterns, Powder Technology 88 (1) (1996) 81–87. [16] K. Uchida, K. Okamoto, Measurement of powder flow in a screw feeder by xray penetration image analysis, Measurement Science and Technology 17 (2006) 419–426. [17] Wei-Ren, Tsai, Chun-I Lin, On the mixing of granular materials in a screw feeder, Powder Technology 80 (1994) 119–126.
379
[18] S.J. Wiche, A.W. Roberts, Investigations into obtaining constant feed rates from screw feeders, Fuel and Energy Abstracts 40 (3) (1999) 188. [19] A. Joppicha, H. Salmanb, Wood powder feeding, diculties and solutions, Biomass and Bioenergy 16 (3) (1999) 191–198. [20] A. Steinfeld, M. Brack, A. Meier, A. Weidenkaff, D. Wuillemin, A solar chemical reactor for co-production of zinc and synthesis gas, Energy 23 (10) (1998) 803–814. [21] S. Kräupl, A. Steinfeld, Experimental investigation of a vortex-flow solar chemical reactor for the combined ZnO-reduction and CH4-reforming, Journal of Solar Energy Engineering 123 (2001) 237–243. [22] S. Kräupl, A. Steinfeld, Operational performance of a 5-kW solar chemical reactor for the co-production of zinc and syngas, Journal of Solar Energy Engineering 125 (2003) 124–126. [23] S. Kräupl, A. Steinfeld, Monte carlo radiative transfer modeling of a solar chemical reactor for the co-production of zinc and syngas, Journal of Solar Energy Engineering 127 (2005) 102–108.
Contents lists available at ScienceDirect
Advanced Powder Technology journal homepage: www.elsevier.com/locate/apt
Original Research Paper
The best screw shape for fine zinc oxide particles feeding M. Barati Dalenjan ⇑, E. Jamshidi, H. Ale Ebrahim Department of Chemical Engineering, Amirkabir University of Technology (Tehran Polytechnic), Petrochemical Center of Excellency, Tehran 15875-4413, Iran
a r t i c l e
i n f o
Article history: Received 8 January 2011 Received in revised form 1 May 2011 Accepted 9 May 2011 Available online 26 May 2011 Keywords: Screw feeder Zinc oxide powder Mass flow rate Particle size distribution
a b s t r a c t One of the most important requirements of particles feeding is a very homogeneous mass flow of fine particles. Screw feeders are among equipment used to feed particles in many industries. In this study, the performance of four single screw feeders with different pitch and blade shape screws has been studied experimentally for steady and stable feeding of fine zinc oxide particles at flow rates corresponding to laboratory scale range (under 10 g/min). The following results were obtained from this investigation: (I) the sticking problem of fine particles in angular pitch screw was solved by changing the pitch shape to circular; (II) the distribution of the fluctuations and their intensity in mass flow rates also lessen by using the circular pitch and a thin blade screw in place of the circular pitch and a thick blade screw. Also for both feeders, the feeding was often interrupted in low flow rates, but it will be disappeared by increasing the flow rate. Furthermore, the results of experiments show that the performance of the circular pitch and a thin blade screw feeder was better than other screw feeders and able to both swirl and mix the particles with different characteristics and reduce the mean aggregate size of the particle size distribution (PSD) when transmitting the zinc oxide particles. Ó 2011 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved.
1. Introduction Fine particles feeding are widely used, especially in polymer, chemical and food industries [1–3]. One of the most important applications is feeding of zinc oxide particles to a solar reactor for decomposition to zinc as an energy carrier and hydrogen producer by water splitting method [4–6]. In recent years there were considerable efforts to provide the uniform fine particles feeding to gas–solid two phase flow reactor [2,7]. On laboratory scale, it is particularly difficult to achieve steady and stable feeding because the fine particles are very compressible. Therefore, weighing and transmitting (metering and handling) in low flow rates is difficult [8,9]. The fine (less than 1 lm) ZnO particles are very cohesive and sticking, and classified as Geldart group C particles [10–13]. Group C particles tend to cling to each other as a consequence of the interparticle forces and form agglomerates [2,11,13]. Therefore, it was expected that we have coarse agglomerations in the initial ZnO powder. Furthermore, the fine particles feeding are very sensitive because a small variation in the gas–solid two phase flow will change the reaction process. There are many types of feeders available for feeding of particles in the literature. The screw feeders are among the most widely used types of volumetric feeders to particles feeding. They are a ⇑ Corresponding author. Tel.: +98 021 64543177; fax: +98 021 66405847. E-mail address: [email protected] (M. Barati Dalenjan).
popular choice for many applications because of that can easily be modified to fit the type of particles and process involved [10,14,15]. Some advantages of screw feeders include particles transmitting, stirring, mixing [3,14,16,17], and reliable to low and high feeding rates [10,14]. The drawbacks of screw feeders are as follows: Interlocking or cohesive arch of particles forming above the screw in the hopper [10,14]. ‘Clogging’ takes place when particles stick to the root corners of the screw flight. ‘Logging’ is a term used when the total volume of the screw is filled and rotates [14]. Instantaneous fluctuations of mass flow rate [18,19]. Promote the aggregation of particles that then drop off the end of the screw as particle aggregates [2]. In critical process-control applications, time intervals for sampling should be no longer than residence time in the actual process. This can be a problem with processes that have short residence times, particularly gas–solid reactors [10]. Complete conversion of gas–solid reactions may be achieved by increasing the operating temperature and residence time [20]. However, the residence time of gas–solid reactions is usually in the order of several seconds [21–23]. The novelty of the present work is using of a circular pitch and a thin blade screw for improving the performance of a screw feeder with respect to the above mentioned problems for steady and
0921-8831/$ - see front matter Ó 2011 Published by Elsevier B.V. on behalf of The Society of Powder Technology Japan. All rights reserved. doi:10.1016/j.apt.2011.05.001
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
373
Table 1 Properties of zinc oxide and potassium dichromate powders. I. Properties of zinc oxide powder CAS number Bulk density True density Molar mass Particle size
1314-13-2 0.8 g/cm3 5.606 g/cm3 81.4 g/mol Less than 1 lm
II. Properties of potassium dichromate powder CAS number Bulk density True density Molar mass Mean particle size
7778-50-9 1.6 g/cm3 2.68 g/cm3 294.18 g/mol 80 lm Fig. 2. A schematic view of experimental setup.
stable feeding of fine particles, particularly for injection into gas– solid two phase flow reactors. To this end, tests were carried out on four single screw feeders with different pitch and blade shape screws, in the same condition, in order to find out the best shape of screw. Moreover, other important issues for a screw feeder consisting such as the mixing performance of particles having different characteristics, and reduction of mean aggregate size of the particle size distribution (PSD) have been considered for the best screw feeder.
on the screw. An agitation device, preventing bridging inside of the hopper, produces a rather uniform bulk density and supports the powder flow into the screw. Particles transmit from hopper into a fluid flow or process line by means of screw revolution. The rotational speed of screw is controlled electrically by means of the DC motor speed controlling unit, and the speed of screw could thus be changed. To determine the mass flow rate and the fluctuations of the screw feeding, a balance is put right underneath the screw outlet. The effect of feeders, especially for the production of homogeneous mass flow rate on laboratory scale was studied.
2. Experimental 2.1. Material Zinc oxide (ZnO) powder is used as the feed test powder. The powder is 99.7% pure and has a reported primary particle size of less than 1 lm. Table 1I shows the properties of ZnO powder. An Analysette 22 Micro Tec plus from Fritsch system (with wet dispersion unit by using a combination of a centrifugal pump in combination with ultrasonic) was used to measure the particle size distribution. The cumulative volume distribution of ZnO powder is shown in Fig. 1. The fresh powder is heated to 120 °C for 20 min in an oven before the tests. Therefore, the moisture of powder in all tests is the same. 2.2. Experimental method Fig. 2 shows a schematic of experimental setup. The ZnO powder is filled manually into the hopper, and the hopper is continuously refilled during the test to provide a uniform head pressure 110 100
Cumulative Volume (%)
90 80 70 60 50 40 30 20 10 0 0.5
5
50
Particle Size (µm) Fig. 1. The cumulative volume distribution of ZnO powder.
Fig. 3. Screws employed to feeding tests: (a) rectangular pitch screw, (b) trapezoidal pitch screw, (c) circular pitch and a thick blade screw, (d) circular pitch and a thin blade screw.
374
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
Fig. 3 shows four single screws which were tested in feeding experiments. There are two types of screws, one type is the angular pitch screws and another one is the circular pitch screws. The angular pitch screws include a rectangular pitch screw (Fig. 3a), and a trapezoidal pitch screw (Fig. 3b). The circular pitch screws comprise a circular pitch and a thick blade screw (Fig. 3c), and a circular pitch and a thin blade screw (Fig. 3d). Table 2 shows the dimensions of main parts for the four screws in Fig. 3. The main differences among these screws were in both pitch forming and thickness of blade. The screws were tested at rotational speeds of 2 1/2 to 17 3/4 rpm. The tests were done to measure the particles mass flow rate and its fluctuations to find out the best screw which tends to reduction of feeding fluctuations. The performance of a screw feeder is normally expressed in statistical terms based on the uniformity of the delivery rate as a function of time. The mass delivered in a given time interval is measured for a number of time intervals. The mean and standard deviation of the masses is calculated, and the value of standard deviation divided by the mean provides a percentage used in describing feeder performance. Because of the residence time of gas–solid reactions is about several seconds, time interval was selected as 1 s. The balance used for weighing the particles mass gave five signals of measurements per second which sent to the serial port of a PC and stored. For each screw the mass of particles fed in each 1 s interval was measured and compared.
3. Results and discussion To examine the performance of each screw, especially with respect to production of a uniform mass flow rate on laboratory scale, two different test series were conducted. Firstly, the performance of each screw to particles feeding was studied to measure the mass flow rate for 30 min at different rotational speeds, and secondly, feeding fluctuations of the best screws were examined. 3.1. Rectangular pitch screw feeder tests Fig. 4 shows the feeding curves for rectangular pitch screw feeder when the rotational speed of screw increases from 4 1/4 to 17 3/ 4 rpm. It shows that the shapes of all curves are similar and can divide feeding curves into three regions: feed region 1, is occurring in initial 5 min that the slopes of the curves are high, and there is no variation in feeding rates. Then feed region two occurs until about
Table 2 The dimensions of main parts for the four screws. Type of screw (mm)
Rectangular pitch screw
Trapezoidal pitch screw
Circular pitch and a thick blade screw
Circular pitch and a thin blade screw
The screw diameter The shaft diameter The pitch length The blade thickness
26
22
28
26
21
18
20
19
3
4
7.5
7
2.5
4
7.5
1
Powder Mass (grams)
16
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/1 rpm
12
8
4
0 0
5
10
15
20
25
30
Time (min) Fig. 4. Feeding curves for rectangular pitch screw feeder.
35 Fig. 5. Screws after the feeding tests: (a) rectangular pitch screw, (b) trapezoidal pitch screw, (c) circular pitch and a thick blade screw, (d) circular pitch and a thin blade screw.
375
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
25 min, where the downward slope is distinct because of cylinder obstruction or logging. On the one hand, length of cylinders was selected to be short in order to prevent probable obstructions. Accordingly, this decreasing of the mass feeding rate is for sticking the particles into the pitch and reduction of the screw volume. Feed region three is after 25 min until 30 min, where the slopes of the curves are almost zero. It is shown that the mass feeding rate has plunged and the pitch of screw is approximately filled with impacted particles. The particles rotate with screw because the forces promoting the rotation exceed the forces restraining their movement. Fig. 5a shows the rectangular pitch screw after the test.
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/4 rpm
3.2. Trapezoidal pitch screw feeder tests Fig. 6 shows the feeding curves for trapezoidal pitch screw feeder. Unlike rectangular pitch screw feeder, it is clear in Fig. 6 that the curves can be split into two feed regions. An initial fast rate of mass flow occurs in first 10 min, feed region 1, where there is no variation in feeding rates observed. Then, the curves switched to slower mass flow rate, but the slopes of the curves are not zero, feed region 2. This drop to feeding rate arose from cohering of particles on the screw shaft or sticking of particles in the angle of shaft and blade. It is clear that, by increasing the time of the tests, the feed region 3 of rectangular pitch screw feeder will occur and feeding will be stopped. Clogging and logging for the trapezoidal pitch screw after the test is shown in Fig. 5b.
12
2 1/2 rpm
16
Powder Mass Flow (g/min)
Powder Mass (grams)
10 12
8
4
8
6
4
2
0
0 0
5
10
15
20
25
30
0
35
4
12
16
20
Fig. 8. Relation between rotational screw speed and particles mass flow for circular pitch and a thick blade screw feeder.
Fig. 6. Feeding curves for trapezoidal pitch screw feeder.
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/4 rpm
8
2 1/2 rpm
350
Deviation from The Mean Mass Flow Rate (g/s) (%)
6
300
Powder Mass (grams)
8
Rotational Speed (rpm)
Time (min)
250 200 150 100
4
2
0
-2 50 -4
0
0 0
5
10
15
20
25
30
35
10
20
30
40
50
60
Time (sec)
Time (min) Fig. 7. Feeding curves for circular pitch and a thick blade screw feeder.
Fig. 9. Fluctuations at rotational speed of 2 1/2 rpm for circular pitch and a thick blade screw feeder.
376
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
3.3. Circular pitch and a thick blade screw feeder tests Feeding curves and the average particles mass flow rate versus the rotational speed for circular pitch and a thick blade screw feeder are shown in Figs. 7 and 8, respectively. Results indicate that, the performance of this feeder compared with two preceding screw feeders is very different. Fig. 7 shows that the rotational speeds of 2 1/2 up to 11 rpm can provide continuous feeding of particles from 1.3 to 7 g/min, respectively. But there are some dispersal in feeding at rotational speeds of 13 1/2 and 17 3/7 rpm. Also Fig. 8 confirms the linearity of particles mass flow rate from 2 1/2 to 11 rpm, not in 13 1/2 and 17 3/4 rpm. This problem in relatively high rotational speeds occurs because the evolution and transmitting speeds of screw is more than the entering speed of particles into the screw volume. So, particles do not have enough
time to fill the volume of screw. Moreover, the thick blade of screw excludes particles from entering into the screw volume. Fig. 5c shows the circular pitch and a thick blade screw after the test. Also it shows that the sticking problem of particles in the rectangular pitch screw and the trapezoidal pitch screw is solved by changing the pitch form of screw from angular to circular. The minimum mass flow rate (1.3 g/min) is corresponding to rotational speed of approximately 2 1/2 rpm and the maximum mass flow rate (9.5 g/min) is at 17 3/4 rpm. Figs. 9 and 10 present the fluctuations of the flow rates at rotational speeds of 2 1/2 and 17 3/4 rpm, respectively. The deviations from the mean flow rate are plotted versus time. In accordance with the 5 weight measurements per second, the arithmetic mean of five samples is taken to represent the actual deviation of the mass flow rate during the time period of 1 s. Both figures show
40
Powder Mass Flow (g/min)
Deviation from The Mean Mass Flow Rate (g/s) (%)
12 30 20 10 0 -10
10
8
6
4
-20 2 -30 0
20
40
60
0
Time (sec)
0
4
8
12
16
20
Rotational Speed (rpm)
Fig. 10. Fluctuations at rotational speed of 17 3/4 rpm for circular pitch and a thick blade screw feeder.
Fig. 12. Relation between rotational screw speed and particles mass flow for circular pitch and a thin blade screw feeder.
17 3/4 rpm
13 1/2 rpm
11 rpm
9 rpm
6 3/4 rpm
4 1/4 rpm
8
2 1/2 rpm
350
R² = 0.9998
250
R² = 0.9998
200
R² = 0.9998 R² = 0.9999
150
R² = 0.9995
100
R² = 0.9998
50
6
Deviation from The Mean Mass Flow Rate (g/s) (%)
Powder Mass (grams)
300
4
2
0
-2
R² = 0.9998
-4 0
0 0
5
10
15
20
25
30
35
10
20
30
40
50
60
Time (sec)
Time (min) Fig. 11. Feeding curves for circular pitch and a thin blade screw feeder.
Fig. 13. Fluctuations at rotational speed of 2 1/2 rpm for circular pitch and a thin blade screw feeder.
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
where x is the arithmetic mean of the values xi, defined as:
30
x ¼
20
Deviation fromThe Mean Mass Flow Rate (g/s) (%)
377
N x1 þ x2 þ þ xN 1 X xi : ¼ N i¼1 N
The standard deviation can be used to provide a quantitative measure of the variations of particles feeding rate. The standard deviation of ZnO particles flow rate is calculated from the experimental data, which has increased from 0.058 (rotational speed of 2 1/2 rpm) to 0.164 g/s (rotational speed of 17 3/4 rpm).
10
0
3.4. Circular pitch and a thin blade screw feeder tests -10
-20
-30 0
10
20
30
40
50
60
Time (sec) Fig. 14. Fluctuations at rotational speed of 17 3/4 rpm for circular pitch and a thin blade screw feeder.
typical characteristics of screw feeding: counting the peaks during 1 min gives a number equal to the rotational speed of the screw. This implies that there is a direct relation between the screw speed and the frequency of fluctuations. Moreover, the magnitude of the deviations must increase by increasing the screw speed, which is seen when Figs. 9 and 10 are compared. In Fig. 9, there are some interruptions in feeding which have displayed by circles, but Fig. 10 shows that the problem of pulsing is solved by increasing the screw speed. The standard deviation is defined as follows:
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi u N u1 X r¼t ðx xÞ2 ; N i¼1
Feeding curves and the average particles mass flow rate versus the rotational speed for circular pitch and a thin blade screw feeder are shown in Figs. 11 and 12, respectively. The results are surprising because of the R2 value for all curves in Fig. 11 is more than 0.99, that shows the precision of experimental results. It means that, this feeder can continuously provide feeding of fine particles from 1.2 g/min (at minimum rotational speed) to 10.5 g/min (at maximum rotational speed). Fig. 12 shows that the problem of circular pitch and a thick blade screw feeder at high rotational speeds is solved by means of narrowing the blade. Since the delivery rate of circular pitch and a thin blade screw feeder responds to changes in rotational speed in a linear and predictable fashion, the process control will be simplified. Fig. 5d shows the circular pitch and a thin blade screw after the test and it is obvious that no particle sticks to the screw. Figs. 13 and 14 present the fluctuations of flow rates at rotational speeds of 2 1/2 and 17 3/4 rpm, respectively. A significant reduction of fluctuations can be noticed when comparing them with Figs. 9 and 10. The standard deviation at rotational speed of 2 1/2 rpm is reduced from 0.058 to 0.048 g/s, and at rotational speed of 17 3/4 rpm from 0.164 to 0.146 g/s. Moreover, the feeding interruption can be improved. The circles in Fig. 13 show that the interruptions in feeding are less than Fig. 9. Fig. 14 shows that the problem of pulsing in circular pitch and a thin blade screw feeder is solved by increasing the rotational speed of screw. It indicates that increasing the screw revolution speed causes more uniformity in the mass flow.
Fig. 15. Swirl and mix the particles with different characteristics, (a) under part of the screw was filled with zinc oxide particles, (b) upper part of the screw was filled with potassium dichromate particles, (c) and (d) are shown swirling and mixing of particles, (e) output of the screw feeder.
378
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379
Fig. 16. SEM observations of ZnO particles, (a) particles before the test, (b) particles after shearing mechanically by the screw.
(scanning electron microscopy) tests. Fig. 16a and b shows the experimental micrographs of the particle aggregates before the test and after shearing by the screw, respectively. Fig. 17 was produced with SPIP software, and shows the results of the PSD data. When analyzing, it is apparent that the screw feeder leads to cause a mechanical shear of the ZnO particle aggregates, and the PSDs are shifted to a lower mean particle aggregate size. The mean aggregate sizes for the particle aggregates before and after the test are 65 and 43 lm, respectively. These values were determined from the curves area of Fig. 17 (by trapezoidal method on the y axis). Therefore, the calculated mean aggregate sizes are volumetric averages from SEM tests.
_
Normalize Volume Fraction (µm ¹)
0.005
0.004
b a
0.003
0.002
4. Conclusions
0.001
0 0
200
400
600
Particle Diameter (µm) Fig. 17. Volume PSDs for ZnO particles taken from two locations, particles before the test (curve a) and after shearing mechanically by the screw (curve b).
3.5. Swirl and mix the particles test The screw revolves to transmit particles in a feeder. Moreover, revolution of screw causes swirling and mixing of particles. Hence, zinc oxide and potassium dichromate particles were applied to investigate these two characteristics. Table 1II shows the properties of potassium dichromate powder. A circular pitch and a thin blade screw feeder was used, and at first the under part of the screw was filled with zinc oxide which is shown in Fig. 15a. Then the upper part of the screw was filled with potassium dichromate which is shown in Fig. 15b. Fig. 15c and d shows the swirling of zinc oxide and potassium dichromate particles and their mixing together with the screw revolution. Fig. 15e shows the output of the screw that testifies to the accuracy of the test. 3.6. Particle size distribution test The screw feeder is able to decrease the mean aggregate size of PSD. Hence, two locations on circular pitch and a thin blade screw feeder are identified to collect the PSD data to characterize the system. The first one is where the particles falls into the hopper before the test. The next PSD data is measured at the exit of the screw. At this point, the PSD data is obtained just after the particle aggregates are sheared mechanically by the screw. The scanning probe image processor (SPIP) software was used for obtaining of the volume fraction distribution of the particle aggregates from SEM
The steady and stable feeding of fine particles is very difficult because of the sticking property and compressibility of them. The present study examines four single screw feeders with different screws. The studies on the performance of the screw feeders showed that clogging and logging problems in screws are solved by selecting the circular pitch screw instead of angular pitch one. In addition, the circular pitch and a thin blade screw feeder can be used to feed fine particles more reliably and more steadily than other screw feeders. The circular pitch and a thin blade screw feeder are solely able to consistently shear fine agglomerates of particles down where the particle size is submicron. Moreover, this feeder can mix together particles with different characteristics. It can be finally suggested that the circular pitch and a thin blade screw feeder could improve the uniformity of the feeding process. References [1] M.Y. Gundogdu, Design improvements on rotary valve particle feeders used for obtaining suspended airflows, Powder Technology 139 (1) (2004) 76–80. [2] T.M. Francis, C.J. Gump, A.W. Weimer, Spinning wheel powder feeding device– fundamentals and applications, Powder Technology 170 (2006) 36–44. [3] K. Uchida, K. Okamoto, Measurement technique on the diffusion coefficient of powder flow in a screw feeder by X-ray visualization, Powder Technology 187 (2008) 138–145. [4] R. Müller, A. Steinfeld, H2O-splitting thermochemical cycle based on ZnO/Zn redox: quenching the effluents from the ZnO dissociation, Chemical Engineering Science 63 (2008) 217–227. [5] H.H. Funkea, H. Diaza, X. Lianga, C.S. Carneya, A.W. Weimer, P. Li, Hydrogen generation by hydrolysis of zinc powder aerosol, International Journal of Hydrogen Energy 33 (2008) 1127–1134. [6] C. Perkins, P.R. Lichty, A.W. Weimer, Thermal ZnO dissociation in a rapid aerosol reactor as part of a solar hydrogen production cycle, International Journal of Hydrogen Energy 33 (2008) 499–510. [7] T.M. Francis, P.B. Kreider, P.R. Lichty, A.W. Weimer, An investigation of a fluidized bed solids feeder for an aerosol flow reactor, Powder Technology (2009). doi:10.1016/j.powtec.2009.04.020. [8] T.W. Davies, The Production of concentrated powder suspensions at low flow rates, Powder Technology 42 (1985) 249–253. [9] L. Tang, W.Y. Chen, Improvements on a particle feeder for experiments requiring low feed rates, Review of Scientific Instruments 70 (7) (1999) 3143– 3144.
M. Barati Dalenjan et al. / Advanced Powder Technology 23 (2012) 372–379 [10] T.A. Bell, S.W. Couch, T.L. Kreiger, H.J. Feise, Screw feeders: a guide to selection use, Chemical Engineering progress 99 (2) (2003) 44–53. [11] C. Xu, J. Zhu, Parametric study of fine particle fluidization under mechanical vibration, Powder Technology 161 (2006) 135–144. [12] D. Geldart, Types of gas fluidization, Powder Technology 7 (1973) 285–297. [13] Z. Wang, M. Kwauk, H. Li, Fluidization of fine particles, Chemical Engineering Science 53 (3) (1998) 377–395. [14] L. Bates, Guide to the design, selection, application of screw feeders, First Published, Wiltshire, UK, 2000. [15] Y. Yu, P.C. Arnold, The influence of screw feeders on bin flow patterns, Powder Technology 88 (1) (1996) 81–87. [16] K. Uchida, K. Okamoto, Measurement of powder flow in a screw feeder by xray penetration image analysis, Measurement Science and Technology 17 (2006) 419–426. [17] Wei-Ren, Tsai, Chun-I Lin, On the mixing of granular materials in a screw feeder, Powder Technology 80 (1994) 119–126.
379
[18] S.J. Wiche, A.W. Roberts, Investigations into obtaining constant feed rates from screw feeders, Fuel and Energy Abstracts 40 (3) (1999) 188. [19] A. Joppicha, H. Salmanb, Wood powder feeding, diculties and solutions, Biomass and Bioenergy 16 (3) (1999) 191–198. [20] A. Steinfeld, M. Brack, A. Meier, A. Weidenkaff, D. Wuillemin, A solar chemical reactor for co-production of zinc and synthesis gas, Energy 23 (10) (1998) 803–814. [21] S. Kräupl, A. Steinfeld, Experimental investigation of a vortex-flow solar chemical reactor for the combined ZnO-reduction and CH4-reforming, Journal of Solar Energy Engineering 123 (2001) 237–243. [22] S. Kräupl, A. Steinfeld, Operational performance of a 5-kW solar chemical reactor for the co-production of zinc and syngas, Journal of Solar Energy Engineering 125 (2003) 124–126. [23] S. Kräupl, A. Steinfeld, Monte carlo radiative transfer modeling of a solar chemical reactor for the co-production of zinc and syngas, Journal of Solar Energy Engineering 127 (2005) 102–108.