- Email: [email protected]
Effect of filter layer thickness on the filtration characteristics of dual layer granular beds
Powder Technology 335 (2018) 344–353
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Effect of filter layer thickness on the filtration characteristics of dual layer granular beds Gui-Hui Xiao, Guo-Hua Yang ⁎, Qi Yang, Su-Rui Tian Faculty of Maritime and Transportation, Ningbo University, Zhejiang 315211, China
a r t i c l e
i n f o
Article history: Received 30 December 2017 Received in revised form 4 May 2018 Accepted 10 May 2018 Available online 26 May 2018 Keywords: Granular bed Filtering layer thickness Filtration
a b s t r a c t We herein report our investigation into the effect of the filter layer thickness on the filtration performance of dual layer granular bed filters using an experimental granular bed filter with an inner diameter of 100 mm and employing fly ash as the example dust. It was found that increasing the thickness of the upper filter layer from 180 to 280 mm reduced the average outlet dust concentration of the dual layer granular bed from 8.69 to 6.57 mg/m3, in addition to prolonging the time between regeneration cycles from 43.3 to 56 min, and increasing the pressure drop across the bed from 1873 to 1978 Pa. Furthermore, increasing the thickness of the lower filter layer from 45 to 85 mm reduced the average outlet dust concentration of the dual layer granular bed from 8.69 to 3.94 mg/m3 and extended the time between regeneration cycles from 43.3 to 59 min. However, this also resulted in an increased pressure drop across the filter from 1935 to 3077 Pa. These results indicated that an increase in the thickness of the upper filter layer reduced outlet dust concentrations and extended the regeneration cycles, without having a significant impact on the total pressure drop across the dual layer granular bed. Although an increase in the thickness of the lower filter layer significantly reduced outlet dust concentrations, the pressure drop across the dual layer granular bed filter was greatly increased in this case. © 2018 Published by Elsevier B.V.
1. Introduction Coal pyrolysis is a key intermediate process for thermochemical conversion procedures such as coal burning, gasification, and liquefaction, and it is also an important technique in the context of quality-based coal utilization [1]. In addition, the coal pyrolysis technique can be employed to obtain clean-burning gaseous fuels, high-value oil products, and high-quality solid fuels, thereby rendering it the primary technological pathway for the comprehensive utilization of coal [2]. However, a number of issues prevent the widespread adoption of coal pyrolysis techniques, including the large quantities of micron-grade dust particles generated along with the high temperature waste gases during coal pyrolysis [3], the carbon-forming tendencies of tar gases, and the lack of an effective solution for gas-solid separation. These issues subsequently lead to problems such as pipeline clogging, the production of tar with high dust contents, difficulties in subsequent coal gas applications, and severe wear in critical pieces of equipment [4,5]. However, as granular bed dust removers are characteristically tolerant of high temperatures, have a wide selection of filter media, and are relatively cheap [6,7], the development of a granular bed filter suitable for coal pyrolysis would represent a particularly significant achievement.
⁎ Corresponding author. E-mail address: [email protected] (G.-H. Yang).
https://doi.org/10.1016/j.powtec.2018.05.019 0032-5910/© 2018 Published by Elsevier B.V.
Granular beds are mainly divided into moving granular beds and stationary granular beds. Hsu et al. [8] used silica sand with particle size of 2–4 mm to study a granular bed filter. The granular bed model used a moving bed, and the inlet of the granular bed had a baffle arrangement. The influence of baffle angle and length on the filtration efficiency of the granular bed entrance was studied. The results showed that filtration efficiency of 98.55% could be achieved with baffle length of 170 mm and baffle angle of 50°, and the filtration efficiency of the granular bed filter could be improved by making the air distribution of the inlet uniform. Wenzel et al. [9] modified the model parameters of the moving granular bed filtration process and performed comparisons with the experimentally obtained pressure drop and filtration efficiency. El-Hedok et al. [10] studied the influence of the flow rate of filter particles on the filtration efficiency of a granular bed by using a moving bed with filter size of 1.6–3.2 mm. The results showed that the retention time of filter particles must reach the critical residence time for the filtration efficiency of the granular bed to exceed 99%. Chen et al. [11] examined the use of granular bed filters for filter power plant-generated fly ash with particle sizes ranging between 0.24 and 363.08 μm. For this purpose, they employed a moving bed granular filter measuring 1.07 m in length, 0.38 m in width, and 0.5 m in depth. The filter media of the granular bed consisted of sea sand with particle sizes ranging between 2 and 4 mm. Following filtration of the fly ash, it was found that the majority of dust particles N10 μm had been removed; however, the filtration efficiency of this granular
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
345
Fig. 1. Schematic representation of the experimental apparatus employed herein.
bed filter for fine dust was less than ideal. To address this issue, Rodon et al. [12] developed a granular bed filter using sand with particle sizes of 0.3–0.42 mm, and examined the filtration of fly ash at a filtration velocity of 0.111 m/s. In this experiment, the granular bed achieved filtration efficiencies of 99.99%, and the high filtration efficiency of the granular bed was found to be due to the formation of dust cakes on the surface of the filter media. However, these dust cakes also significantly increased the pressure drop across the filter bed. In addition, Wu et al. [13] employed three different types of filter media (i.e., 0.3–0.42 mm silicon carbide, 0.3–0.42 mm grit, and 0.15–0.42 mm copper particles) to study how the deposition of dust in the filter layer affected its filtration efficiency towards sub-micron grade particles in addition to its pressure drop. It was found that dust deposition led to gradual increases in the filtration efficiency for sub-micron grade particles. However, after the quantity of deposited dust reached an optimal level, any further increases in dust deposition led to decreases in the filtration efficiency. Furthermore, excessive quantities of dust led to enhanced pressure drops, which in turn led to perforations in the dust layers.
In view of the low filtration efficiencies and low pressure drops of coarse-particle granular layers, and the high filtration efficiencies, high pressure drops, and short regeneration cycles of fine-particle granular layers, Yang [14,15] developed a novel graded filtering technique using a dual layer granular bed that promised to combine the strengths of these filter layers while eliminating their weaknesses. This dual layer granular bed consists of lower and upper filter layers, where the upper filter layer contains low-density, coarse particles, and the lower filter layer contains highly dense, fine particles. During filtration, the dust-containing airflow enters from the top of the filter, and passes through the upper filter layer where the majority of dust is captured; this is known as “coarse filtering”. The fine particles that penetrate the upper filter layer are then captured by the lower filter layer, thus completing the “fine filtering” process. The combination of these layers therefore provides a cascaded filter.
XSVWUHDP
XSSHUOD\HU
ORZHUOD\HU
PLGVWUHDP
IORZVWDELOL]LQJOD\HU
GRZQVWUHDP
Fig. 2. Schematic representation of the dual layer granular bed filter employed herein.
Fig. 3. Particle size distribution of the fly ash dust employed herein.
346
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
Fig. 4. Variation in the filtration velocity with filter time for a range of upper filter layer thicknesses.
Although dual layer granular beds are known to characteristically exhibit high filtration efficiencies and low pressure drops, studies into the effects of the filter layer thickness on the filtration characteristics of filters of this type remain rather scarce. Thus, to address this gap in the literature, we herein report our investigation into how changes in the thicknesses of the upper and lower filter layers affect the filtration characteristics of a dual layer granular bed.
2. Materials and methods 2.1. Experimental apparatus The experimental apparatus employed herein consists of a granular bed, an RBG-1000 aerosol generator, a measurement system, and other auxiliary devices (see Fig. 1 and Section 2.1.3 for details).
100
Filtration efficiency(%)
Filtration efficiency凚 %凛
100
99
98
97
96
180
200
220
240
260
98 96 94
90 88 86 84
280
180mm 205mm 230mm 255mm 280mm
92
0
10
20
Thickness凚 mm凛
1000 800 600 400 200
20
30
40
50
60
Outlet dust concentration凚 mg/m3)
Outlet dust concentration凚 mg/m3凛
180mm 205mm 230mm 255mm 280mm
10
50
60
(b)
1200
0
40
Time 凚 min凛
(a)
0
30
250
200
150
100
180
205
230
255
Time凚 min凛
Thickness凚 mm凛
(c)
(d)
Fig. 5. Effect of the upper filter layer thickness on its filtration efficiency and outlet dust concentration.
280
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
2.1.2. The RBG-1000 aerosol generator A Palas RBG-1000 aerosol generator (Palas Gmbn, Germany) was used to disperse the dust. More specifically, the non-adhesive dust particles (0.1–100 μm) were loaded into a cylindrical canister prior to feeding through a feed piston and rotating precision brush at a precisely controlled rate. The dust particles were then ejected from a nozzle with airflows up to 180 m/s; the strong turbulent flows and shear forces ensure that the dust particles are optimally dispersed within the main airflow.
Filtration efficiency凚 %凛
99.96 99.95 99.94 99.93 99.92 99.91 99.90
180
200
220
240
260
280
Thickness凚 mm凛
(a) Outlet dust concentration 凚 mg/m3)
2.1.1. The granular bed The granular bed employed herein was constructed using transparent organic glass tubes with an inner diameter of 100 mm. As shown in Fig. 2 (a), the granular bed consists of three granular layers, namely the flowstabilizing layer, a lower filtering layer, and an upper filtering layer. The flow-stabilizing layer was set above an air distributing plate, and acts as a homogenizer for incoming airflows. The flow-stabilizing layer also supports the lower filter layer, which is situated on top of the flow-stabilizing layer. The upper filter layer was then placed above the lower filter layer, as implied by its name. To examine the individual contributions of the upper and lower filter layers, the upper filter layer was placed on a #100 metallic mesh to ensure that the upper and lower filter layers were clearly separated (mesh size of #100 metallic mesh is 165 μm). Furthermore, a row of pressure measurement holes and 3 dust sampling holes were punctured in the sides of the glass tube (see Fig. 2(b)) to facilitate sampling of the inlets and outlets of the two filter layers. The inlet of the upper filter layer is referred to as the “upstream” section, the outlet of the upper filter layer (i.e., the inlet of the lower filter layer) is referred to as the “midstream” section, and the outlet of the lower filter layer is referred to as the “downstream” section. The inlet/outlet dust concentrations and dust particle size distributions were analyzed to reveal the filtration characteristics of each filter layer. The upper filter medium consisted of expanded perlite with particle size of 1.0–2.0 mm and bulk density of 70 kg/m3, whereas the lower filter medium consisted of sea sand with particle size of 0.3–0.5 mm and bulk density of 1340 kg/m3. The flow-stabilizing layer comprised sea sand with particle size of 2.0–2.5 mm and bulk density of 1550 kg/m3. The bulk density of the filter material was measured using the mass volume method.
347
10 8 6 4 2 0 180
200
220
240
260
280
Thickness (mm)
(b) Fig. 6. Effect of the upper filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed.
2.2. Experimental outline 2.1.3. Measurement system and auxiliary devices The various measurements performed by the measurement system include flow measurements, pressure drop measurements, and dust measurements. The flow rate was measured using the rotor's flow rate monitor, while online pressure drop measurements were conducted using a PY differential pressure sensor. A Wales-3000 aerosol particle size spectrometer was used to perform dust sampling and analysis. This spectrometer was combined with two sensors to simultaneously measure the particle size distribution, number concentration, and volume concentration of the dust at the inlet and outlet of each filter layer to essentially provide the filtration efficiency of each filter layer. The full particle filtration efficiency is calculated using the following formula:
η ¼ 1−
cout cin
ð1Þ
Here, cin is the volume concentration of dust at the inlet; cout, the volume concentration of dust at the outlet; and η, the filtration efficiency. The auxiliary devices employed in the experimental setup include a digital scale, a 3L13WC Roots blower, a ball miller, an experimental sieve, and a Vernier caliper.
The fly ash generated by some power plant was used as the dust sample in this work. The particle size distribution of this dust is outlined in Fig. 3. All filtration tests used fresh identical test dust; the test dust was not recycled, and the dust particle size distribution (PSD) was unchanged. To investigate the effect of the upper filter layer thickness on the filtration characteristics of the dual layer granular filter, five filtration experiments were conducted using thicknesses of 180, 205, 230, 255, and 280 mm. All other experimental conditions were identical, and the thickness of the lower filter layer and flow-stabilizing layer was 45 and 55 mm, respectively. To investigate the effect of the lower filter layer thickness on the filtration characteristics of a dual layer granular filter, five filtration experiments were conducted using thicknesses of 45, 55, 65, 75, and 85 mm. All other experimental conditions were fixed throughout these experiments, and the thickness of the upper filter layer and flow-stabilizing layer was 180 and 75 mm, respectively. Here, the thickness of the flow-stabilizing layer was increased because the thickness of the lower filter material greatly increases the pressure drop of the lower filter layer. To ensure good fluidization effect, the thickness and pressure drop of the flow-stabilizing layer were increased correspondingly. In the filtration process using filtration layers with different thicknesses, after filter layer filling was finished, the filtering material was
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353 2000
2400
Pressure drop凚 Pa凛
Granular bed pressure drop凚 Pa凛
348
1950
1900
1850
180
205
230
255
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400 0
280
0
Thickness凚 mm凛
10
2000
2000
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
2400
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
800 400 0
0
10
20
30
40
1200 800 400 0
50
0
10
20
30
40
50
Time凚 min凛
(c) h = 205 mm
(d) h = 230 mm
2400
2400
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
50
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600
Time凚 min凛
0
40
(b) h = 180 mm
2400
1200
30
Time凚 min凛
(a) Variation in the pressure drop across the granular bed with changes in upper filtering layer thickness
1600
20
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400 0
0
10
20
30
40
50
0
10
20
30
40
50
60
Time凚 min凛
Time凚 min凛
(f) h = 280 mm
(e) h = 255 mm Fig. 7. Variations in the pressure drop during dust filtration.
first subjected to reverse blowing, after which it entered a natural loose accumulation state through velocity reduction to ensure that the porosity and pore structure of the filter layer remained unchanged in each filtration test. The filtration experiments conducted herein were
performed using a constant pressure and a variable filtration velocity, i.e., the total pressure drop across the granular bed was maintained constant throughout, while the filtration velocity was gradually reduced. Upon commencing the filtration experiment, the granular
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
349
3.1.1. Effect of the upper filter layer thickness on the filtration velocity As shown in Fig. 4, the filtration velocity gradually decreased during the filtration process from 0.25 m/s (i.e., the initial filtration velocity) to 0.15 m/s, and the rate at which the filtration velocity decreased was found to increase gradually with time. This decrease in the filtration velocity can be attributed to gradual increases in dust deposition over time, and thus to ensure a constant pressure drop across the dual layer granular bed, the filtration velocity must be gradually reduced. In addition, the gradual increase in the decrease of filtration velocity is caused by changes in the filtration mechanism. More specifically, during the early stages of filtration, deep bed filtration dominates, and dust penetration is significant. During this stage, the accumulation of dust has a minimal impact on the pressure drop of the granular bed. As the degree of dust deposition increases, the filtration mechanism transitions from deep bed filtration to surface filtration. In this stage, the majority of dust will accumulate on the surface of the filter layer, resulting in the formation of dust cakes, which have a significant impact on the pressure drop of the granular bed [16]. Furthermore, as shown in Fig. 4, the deceleration of the filtration velocity is reduced upon increasing the thickness of the upper filter layer, which effectively extends the time taken for the filtration velocity to decrease from 0.25 to 0.15 m/s (i.e., the filtration time). More specifically, an increase in the upper filter layer thickness from 180 to 280 mm resulted in an increase in the filtration time from 43.3 to 56 min, thereby indicating that a thicker upper filter layer lengthens the regeneration cycle of this dual layer granular bed.
for dust to penetrate the filter layer. In addition, Fig. 5(b) shows that the filtration efficiency increases rapidly with time in the initial stages of the filtration process, although this increase in efficiency tapers off after 10 min. This phenomenon can be accounted for by the accumulation of dust over time, which enhances the pore structure of the filter medium. The accumulated dust in the filter layer then acts as a filter and increases the effective filtration efficiency of the upper filter layer [17]. When the quantity of accumulated dust reaches an optimal level, dust cakes form on the surface of the filter layer and stabilize its filtration efficiency. However, beyond this optimal level, any further increase in dust deposition leads to a decrease in the filtration efficiency of the filter layer. This is due to excessive quantities of dust deposition greatly increasing the pressure drop of the filter, which in turn perforates the dust cakes formed on its surface [18]. In addition, Fig. 5(c) shows that the changes in outlet dust concentration of the upper filter layer exactly mirror the trends presented in Fig. 5(b). More specifically, at a constant upper filter layer thickness, the dust concentration rapidly decreases with filter time, and the rate of decline in the dust concentration drops after 10 min. These effects can be accounted for by the filtration mechanism of the granular bed filter. Indeed, Squires et al. noted that the accumulation of dust in a granular bed undergoes three distinct stages, namely a penetration period, a transitional period, and a surface period. During the penetration and transition periods, a rapid decrease in the dust concentration takes place, as the penetration of dust in the filter layer is significant during these periods. As the filtration process progresses, dust penetration into the inner parts of the filter layer decreases, leading to rapid increases in surface dust accumulation. In the later stages of the filtration process, a dust cake layer forms on the surface of the filter layer, and the optimal filtration efficiency is reached. Furthermore, as outlined in Fig. 5(d), the outlet dust concentration of the upper filter layer decreases in an approximately linear fashion upon increasing the thickness. More specifically, the outlet dust concentration decreased from 254.56 to 132.89 mg/m3 when the thickness of the upper filter layer was increased from 180 to 280 mm. However, this dust concentration does not meet the requirements outlined in China's national emission standards, and so additional filtration using a lower filter layer is therefore required. Thus, from Fig. 5, it may be inferred that increases in the upper filter layer thickness result in a significant enhancement in the filtration efficiency of this layer, in addition to decreasing its outlet dust concentration.
3.1.2. Effect of the upper filter layer thickness on the filtration efficiency and outlet dust concentration As outlined in Fig. 5(a), the average filtration efficiency of the upper filter layer increased in an approximately linear fashion with its thickness. This can be attributed to increases in thickness rendering it more difficult
3.1.3. Effect of the upper filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed filter As outlined in Fig. 6(a) and (b), an increase in the upper filter layer thickness increases the average filtration efficiency of the dual layer granular bed and decreases its average outlet dust concentration.
beds were clean and free from dust accumulation. The filtration velocity at this point is denoted as the initial filtration velocity. As the filtration process progresses and increasing quantities of ash are accumulated in the filter layers, the filtration velocity will gradually decrease while the total pressure drop across the granular bed will remain unchanged. During the filtration experiments, the inlet dust concentration of the dual layer granular bed was fixed at 1.37 g/min. Given an initial filtration velocity of 0.25 m/s, this corresponds to a dust concentration of 11.67 g/m3. 3. Results and discussion 3.1. Effect of the upper filter layer thickness on the filtration characteristics of the dual layer granular filter
Fig. 8. Variations in the filtration velocity with filter time and lower filter layer thickness.
350
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
100.00
Filtration efficiency凚 %凛
Filtration efficiency凚 %凛
99.0 98.4 97.8 97.2 96.6 96.0 95.4 40
50
60
70
80
99.98 99.96 99.94 99.92 99.90 99.88 99.86 40
90
50
(a)
98 97 45mm 55mm 65mm 75mm 85mm
96 95 10
20
30
40
50
60
Outlet dust concentration凚 mg/m3凛
45mm 55mm 65mm 75mm 85mm
30
20
10
20
10 8 6 4 2 0 40
50
60
70
80
90
Fig. 10. Effect of the lower filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed.
40
10
90
(b)
(b)
0
80
Thickness(mm)
7LPH˄PLQ˅
0
Outlet dust concentration(mg/m3)
Filtration efficiency(%)
99
0
70
(a)
100
94
60
Thickness凚 mm凛
Thickness凚 mm凛
30
40
50
60
Time凚 min凛
(c) Fig. 9. Effect of the lower filter layer thickness on its filtration efficiency and outlet dust concentration.
Upon increasing the thickness of the upper filter layer from 180 to 280 mm, the average filtration efficiency of the dual layer granular bed increased from 99.926 to 99.944%, while its average outlet dust concentration decreased from 8.69 to 6.57 mg/m3. 3.1.4. Effect of the upper filter layer thickness on the pressure drop across the granular bed As indicated in Fig. 7(a), the pressure drop across the dual layer granular bed at the same initial filtration velocity of 0.25 m/s increases
approximately linearly with the upper filter layer thickness. More specifically, upon increasing the thickness of the upper filter layer from 180 to 280 mm, the total pressure drop across the dual layer granular bed increased from 1873 to 1978 Pa. In addition, for granular beds with the same upper filter layer thickness, the pressure drop across this upper layer increased with filtration time, while the pressure drop of the lower filter layer and the flow-stabilizing layer decreased (see Fig. 7(b)–(f)). This can be attributed to the fact that the pressure drop in the granular bed is determined by two key factors [19]. The first factor is the filtration velocity [20], where the decrease in the pressure drops across the lower filter layer and the flow-stabilizing layer are caused by decreases in the filtration velocity. The second factor is the accumulation of dust in the granular bed [21], where the pressure drop in the granular bed increases with an increase in dust deposition. Given a fixed total pressure drop across the granular bed, the filtration velocity decreases as the filtration process progresses, leading to corresponding reductions in the pressure drop across the granular bed. However, the pressure drop across the upper filter layer is determined mainly by dust deposition, due to the rapid rate at which dust deposition increases in the upper filter layer. As such, the pressure drop of the upper filter layer increases with filtration time. In contrast, the lower filter layer exhibits a decrease in the pressure drop over time due to a decrease in the filtration velocity caused by relatively insignificant dust accumulation in this layer. Similarly, the flowstabilizing layer also exhibits filtration velocity-induced decreases in the pressure drop with time, since dust deposition in this layer is effectively negligible. Fig. 7 also shows that the pressure drop of the
351
2400 3000
Pressure drop凚 Pa凛
Granular bed pressure drop凚 Pa凛
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
2800
2600
2400
2200
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400
2000 40
45
50
55
60
65
70
75
80
85
0
90
0
Thickness凚 mm凛
10
20
(a) Variation in the pressure drop in the granular bed upon increasing the lower filter layer thickness
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1500
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
3000
2000
1000 500
0
10
20
30
40
2500
1500 1000 500 0
50
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
2000
0
10
Time凚 min凛
30
40
50
(d) h = 65 mm 3500
3000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
2500 2000 1500 1000 500
3000
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
20
Time凚 min凛
(c) h = 55mm
0
40
(b) h = 45 mm
2500
0
30
Time凚 min凛
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
2500 2000 1500 1000 500
0
10
20
30
40
50
0
0
10
20
30
40
50
60
Time凚 min凛
Time凚 min凛
(e) h = 75 mm
(f) h = 85 mm
Fig. 11. Variations in the pressure drop during dust filtration.
upper filter layer (as a proportion of the total pressure drop across the dual layer granular bed) increases from 10.67 to 15.87% upon increasing the layer thickness from 180 to 280 mm when the granular bed filter
was clean upon commencing the experiment. This indicates that the increase in the pressure drop induced by increases in the upper filter layer thickness are relatively mild.
352
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
3.2. Effect of the lower filter layer thickness on the filtration characteristics of the dual layer granular bed
bed filter by 54.7%, thereby confirming that increasing the thickness of this layer is beneficial for reducing the outlet dust concentration.
3.2.1. Effect of the lower filter layer thickness on the filtration velocity Fig. 8 outlines the effect of the lower filter layer thickness on the filtration velocity, where it is apparent that a decrease in filtration velocity from 0.25 to 0.15 m/s took place over time at a gradually accelerating rate given the same lower filter layer thickness. Upon increasing the thickness of this layer, the deceleration in the filtration velocity slowed down, with an increase in the lower filter layer thickness from 45 to 85 mm extending the filtration time from 43.3 to 59 min, thereby indicating that a thicker filter layer lengthens the regeneration cycle of the dual layer granular bed.
3.2.4. Effect of the lower filter layer thickness on the pressure drop across the granular bed As indicated in Fig. 11(a), the overall pressure drop of the dual layer granular bed increases linearly upon increasing the lower filter layer thickness. More specifically, upon increasing the thickness from 45 to 85 mm, the total pressure drop increased from 1935 to 3077 Pa. In addition, Fig. 11(b)–(f) show that at the same lower filter layer thickness, the pressure drop of the lower filter layer and the flow-stabilizing layer decrease with longer filter times, while the pressure drop across the upper filter layer increases. This can be accounted for by the combinatorial effects of the filtration velocity and dust accumulation in the granular bed layers. In the case of a clean granular bed, the pressure drop in the lower filter layer increased from 66.20 to 78.84% as a proportion of the total pressure drop across the dual layer granular bed when its thickness was increased from 45 to 85 mm. These results therefore indicate that an increase in thickness of the lower filter layer has a significant effect on the pressure drop of the dual layer granular bed filter.
3.2.2. Effect of the lower filter layer thickness on its filtration efficiency and outlet dust concentration As outlined in Fig. 9(a), the average filtration efficiency of the lower filter layer increases upon increasing its thickness. More specifically, an increase in the thickness of this layer from 45 to 85 mm resulted in an increase in its average filtration efficiency from 96.59 to 98.56%. In addition, the average filtration efficiency was significantly improved upon increasing the thickness of the lower filter layer from 45 to 65 mm, although further increases (i.e., to 85 mm) resulted in only slight improvements. This can be attributed to the diffusion, blockage, and inertial effects of the lower filter layer increasing with increasing thickness [22,23], albeit in a non-linear fashion. In addition, the results presented in Fig. 9(b) demonstrate that at a critical thickness, a slight dip in the filtration efficiency is observed prior to a subsequent increase with filtration time. This can be attributed firstly to the filtration efficiency of the upper filter layer being relatively low at the beginning of the filtration process, which results in higher dust levels at the inlet of the lower filter layer. This gives rise to higher filtration efficiencies in the lower filter layer. Indeed, the inlet dust concentration of the lower filter layer gradually decreases as the filtration process continues, thereby resulting in a decrease in its filtration efficiency. When the dust concentration at the inlet of the lower filter layer is low, the probability of dust particle collision and reunion is small. Therefore, the probability of dust being trapped by the lower filter material is also small, resulting in lower filtration efficiency. Secondly, the accumulation of dust in the lower filter layer improves its pore structure and its filtration efficiency. The combination of these effects ultimately results in an initial decrease in the filtration efficiency, followed by a subsequent increase over time. Furthermore, Fig. 9(c) shows that the outlet dust concentration of the lower filter layer initially exhibits a rapid decrease upon increasing the filtration time, which is followed by a significantly slower decrease. This can be attributed to the small particle sizes present in the lower filter layer, and the resulting penetration of dust particulates in the filter layer during the initial stages of the filtration process, followed by a rapid accumulation on the surface with increasing filtration time. The dust cakes formed on the filter layer then stabilize its filtration efficiency [24]. 3.2.3. Effect of the lower filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed We then moved on to investigate the effect of the lower filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed. As shown in Fig. 10(a) and (b), for thicknesses ranging from 45 to 85 mm, an increase in the lower filter layer thickness led to a corresponding increase in the average filtration efficiency of the dual granular bed and a decrease in its average outlet dust concentration. In addition, an increase in thickness from 45 to 85 mm resulted in an increase in the average filtration efficiency of the dual layer granular bed from 99.926 to 99.967%, in addition to a decrease in its average outlet dust concentration from 8.69 to 3.94 mg/m3. A 40 mm increase in the thickness of the lower filter layer therefore decreased the outlet dust concentration of the granular
4. Conclusions We herein presented our investigation into the effect of upper and lower filter layer thicknesses on the filtration performance of dual layer granular bed filters using an experimental granular bed filter and fly ash dust. It was found that upon increasing the thickness of the upper filter layer from 180 to 280 mm, its average filtration efficiency increased from 97.81 to 98.60%, while its average outlet dust concentration decreased from 254.54 to 132.89 mg/m3. In addition, the average outlet dust concentration of the dual layer granular bed was reduced from 8.69 to 6.57 mg/m3, and its overall pressure drop increased from 1873 to 1978 Pa at the same initial filtration velocity of 0.25 m/s. Furthermore, increasing the thickness of the upper filter layer also lengthened the granular bed regeneration cycle, as the filtration time was increased from 43.3 to 56 min. We also found that when the thickness of the lower filter layer was increased from 45 to 85 mm, the average filtration efficiency of the dual layer granular bed increased from 99.926 to 99.967%, while the average outlet dust concentration decreased from 8.69 to 3.94 mg/m3, which corresponds to a 54.7% reduction in the average outlet dust concentration. These results indicate that an increase in the thickness of the lower filter layer had a significant impact in reducing the outlet dust concentration of the dual layer granular bed. However, it also resulted in the pressure drop of the dual layer granular bed increasing greatly from 1935 to 3307 Pa at the same initial filtration velocity of 0.25 m/s when the lower filter layer thickness was increased from 45 to 85 mm. This represents a 70.9% increase in the pressure drop. The matching of the upper and lower layer thicknesses is related to the inlet dust characteristics, allowable pressure drop and the dust discharge concentration. For the test dust in this paper, it is suitable to select a thickness of 280 mm for the upper layer, a thickness of 45 mm for the lower layer, and a thickness of 55 mm for the flow-stabilizing layer. At this time, the outlet dust concentration is 6.57 mg/m3, and the total pressure drop is 1978 pa. Acknowledgments This work was sponsored by K.C. Wong Magna Fund in Ningbo University, and supported by the National 863 High Technology Research Program of China (2008AA05Z205) and the Student Research and Innovation Program of Ningbo University (G17102). References [1] W. Zhu, W. Song, W. Lin, Catalytic gasification of char from co-pyrolysis of coal and biomass, Fuel Process. Technol. 89 (2008) 890–896.
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353 [2] S. Krerkkaiwan, C. Fushimi, A. Tsutsumi, P. Kuchonthara, Synergetic effect during co-pyrolysis/gasification of biomass and sub-bituminous coal, Fuel Process. Technol. 115 (2013) 11–18. [3] Y.S. Chen, Y.P. Chyou, S.C. Li, Hot gas clean-up technology of dust particulates with a moving granular bed filter, Appl. Therm. Eng. 74 (2015) 146–155. [4] G.Q. Liu, S.P. Deng, X.R. Qian, Research progress of coal pyrolysis technology in China, Mod. Chem. Ind. 27 (2007) 37–43 (in Chinese). [5] K.Y. Shi, G.H. Yang, S. Huang, Study on filtering characteristics of aerosol particulates in a powder-grain dual-layer granular bed, Powder Technol. 272 (2015) 54–63. [6] G. Xiao, X. Wang, J. Zhang, Granular bed filter: a promising technology for hot gas clean-up, Powder Technol. 244 (2013) 93–99. [7] S.R. Tian, G.H. Yang, Z. Li, Cascade filtration properties of a dual-layer granular bed filter, Powder Technol. 301 (2016) 545–556. [8] C.J. Hsu, S.S. Hsiau, A study of filtration performance in a cross-flow moving granular bed filter: the influence of gas flow uniformity, Powder Technol. 274 (2015) 20–27. [9] B.M. Wenzel, C.B. Porciúncula, N.R. Marcilio, Filtration of dust in an intermittent moving granular bed filter: performance and modeling, Sep. Purif. Technol. 133 (2014) 108–119. [10] I.A. El-Hedok, L. Whitmer, R.C. Brown, The influence of granular flow rate on the performance of a moving bed granular filter, Powder Technol. 214 (2011) 69–76. [11] Y.S. Chen, S.S. Hsiau, S.C. Lai, Filtration of dust particulates with a moving granular bed filter, J. Hazard. Mater. 171 (2009) 987–994. [12] I. Rodon, K.C. Lee, R. Pfeffer, Granular-bed filtration assisted by filter-cake formation: 2. The panel bed gas filter with puffback renewal of gas-entry surfaces, Powder Technol. 155 (2005) 5–16. [13] M.S. Wu, K.C. Lee, R. Pfeffer, Granular-bed filtration assisted by filter cake formation 3. Penetration of filter cakes by a monodisperse aerosol, Powder Technol. 155 (2005) 62–73.
353
[14] G.H. Yang, J.H. Zhou, H.P. Shu, A new dual-layered granule bed filter for cleaning-up gas, J. Power Eng. 25 (2005) 891–894 (in Chinese). [15] G.H. Yang, J.H. Zhou, A study of the pressure-drop characteristics of a dual-layer filter-material filtering bed, J. Eng. Therm. Energy Power 22 (2007) 267–269 (in Chinese). [16] Y.S. Chen, S.S. Hsiau, Cake formation and growth in cake filtration, Powder Technol. 192 (2010) 217–224. [17] K.Y. Shi, G.H. Yang, S.R. Tian, Study of filtration characteristics of PM2.5 by a powdergrain dual-layer granular bed, J. China Univ. Min. Technol. 44 (2015) 138–143 (in Chinese). [18] A.M. Squires, Granular-bed filtration assisted by filter-cake formation: 4. Advanced designs for panel-bed filtration and gas treating, Powder Technol. 155 (2005) 74–84. [19] D. Bémer, I. Subra, Y. Morele, Experimental study of granular bed filtration of ultrafine particles emitted by a thermal spraying process, J. Aerosol Sci. 63 (2013) 25–37. [20] Y.Q. Zhang, P. Liang, J. Yu, Studies of granular bed filter for dust removal in the process of coal pyrolysis by solid heat carrier, RSC Adv. 7 (2017) 20266–20272. [21] Y.S. Chen, S.S. Hsiau, H.Y. Lee, Filtration of dust particulates using a new filter system with louvers and sublouvers, Fuel 99 (2012) 118–128. [22] A. Zamani, B. Maini, Flow of dispersed particles through porous media–deep bed filtration, J. Pet. Sci. Eng. 69 (2009) 71–88. [23] C. Tien, B.V. Ramarao, Granular Filtration of Aerosols and Hydrosols, Elsevier/ Butterworths, second edition, 2007. [24] D. Stanghelle, T. Slungaard, O.K. Sonju, Granular bed filtration of high temperature biomass gasification gas, J. Hazard. Mater. 144 (2007) 668–672.
Contents lists available at ScienceDirect
Powder Technology journal homepage: www.elsevier.com/locate/powtec
Effect of filter layer thickness on the filtration characteristics of dual layer granular beds Gui-Hui Xiao, Guo-Hua Yang ⁎, Qi Yang, Su-Rui Tian Faculty of Maritime and Transportation, Ningbo University, Zhejiang 315211, China
a r t i c l e
i n f o
Article history: Received 30 December 2017 Received in revised form 4 May 2018 Accepted 10 May 2018 Available online 26 May 2018 Keywords: Granular bed Filtering layer thickness Filtration
a b s t r a c t We herein report our investigation into the effect of the filter layer thickness on the filtration performance of dual layer granular bed filters using an experimental granular bed filter with an inner diameter of 100 mm and employing fly ash as the example dust. It was found that increasing the thickness of the upper filter layer from 180 to 280 mm reduced the average outlet dust concentration of the dual layer granular bed from 8.69 to 6.57 mg/m3, in addition to prolonging the time between regeneration cycles from 43.3 to 56 min, and increasing the pressure drop across the bed from 1873 to 1978 Pa. Furthermore, increasing the thickness of the lower filter layer from 45 to 85 mm reduced the average outlet dust concentration of the dual layer granular bed from 8.69 to 3.94 mg/m3 and extended the time between regeneration cycles from 43.3 to 59 min. However, this also resulted in an increased pressure drop across the filter from 1935 to 3077 Pa. These results indicated that an increase in the thickness of the upper filter layer reduced outlet dust concentrations and extended the regeneration cycles, without having a significant impact on the total pressure drop across the dual layer granular bed. Although an increase in the thickness of the lower filter layer significantly reduced outlet dust concentrations, the pressure drop across the dual layer granular bed filter was greatly increased in this case. © 2018 Published by Elsevier B.V.
1. Introduction Coal pyrolysis is a key intermediate process for thermochemical conversion procedures such as coal burning, gasification, and liquefaction, and it is also an important technique in the context of quality-based coal utilization [1]. In addition, the coal pyrolysis technique can be employed to obtain clean-burning gaseous fuels, high-value oil products, and high-quality solid fuels, thereby rendering it the primary technological pathway for the comprehensive utilization of coal [2]. However, a number of issues prevent the widespread adoption of coal pyrolysis techniques, including the large quantities of micron-grade dust particles generated along with the high temperature waste gases during coal pyrolysis [3], the carbon-forming tendencies of tar gases, and the lack of an effective solution for gas-solid separation. These issues subsequently lead to problems such as pipeline clogging, the production of tar with high dust contents, difficulties in subsequent coal gas applications, and severe wear in critical pieces of equipment [4,5]. However, as granular bed dust removers are characteristically tolerant of high temperatures, have a wide selection of filter media, and are relatively cheap [6,7], the development of a granular bed filter suitable for coal pyrolysis would represent a particularly significant achievement.
⁎ Corresponding author. E-mail address: [email protected] (G.-H. Yang).
https://doi.org/10.1016/j.powtec.2018.05.019 0032-5910/© 2018 Published by Elsevier B.V.
Granular beds are mainly divided into moving granular beds and stationary granular beds. Hsu et al. [8] used silica sand with particle size of 2–4 mm to study a granular bed filter. The granular bed model used a moving bed, and the inlet of the granular bed had a baffle arrangement. The influence of baffle angle and length on the filtration efficiency of the granular bed entrance was studied. The results showed that filtration efficiency of 98.55% could be achieved with baffle length of 170 mm and baffle angle of 50°, and the filtration efficiency of the granular bed filter could be improved by making the air distribution of the inlet uniform. Wenzel et al. [9] modified the model parameters of the moving granular bed filtration process and performed comparisons with the experimentally obtained pressure drop and filtration efficiency. El-Hedok et al. [10] studied the influence of the flow rate of filter particles on the filtration efficiency of a granular bed by using a moving bed with filter size of 1.6–3.2 mm. The results showed that the retention time of filter particles must reach the critical residence time for the filtration efficiency of the granular bed to exceed 99%. Chen et al. [11] examined the use of granular bed filters for filter power plant-generated fly ash with particle sizes ranging between 0.24 and 363.08 μm. For this purpose, they employed a moving bed granular filter measuring 1.07 m in length, 0.38 m in width, and 0.5 m in depth. The filter media of the granular bed consisted of sea sand with particle sizes ranging between 2 and 4 mm. Following filtration of the fly ash, it was found that the majority of dust particles N10 μm had been removed; however, the filtration efficiency of this granular
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
345
Fig. 1. Schematic representation of the experimental apparatus employed herein.
bed filter for fine dust was less than ideal. To address this issue, Rodon et al. [12] developed a granular bed filter using sand with particle sizes of 0.3–0.42 mm, and examined the filtration of fly ash at a filtration velocity of 0.111 m/s. In this experiment, the granular bed achieved filtration efficiencies of 99.99%, and the high filtration efficiency of the granular bed was found to be due to the formation of dust cakes on the surface of the filter media. However, these dust cakes also significantly increased the pressure drop across the filter bed. In addition, Wu et al. [13] employed three different types of filter media (i.e., 0.3–0.42 mm silicon carbide, 0.3–0.42 mm grit, and 0.15–0.42 mm copper particles) to study how the deposition of dust in the filter layer affected its filtration efficiency towards sub-micron grade particles in addition to its pressure drop. It was found that dust deposition led to gradual increases in the filtration efficiency for sub-micron grade particles. However, after the quantity of deposited dust reached an optimal level, any further increases in dust deposition led to decreases in the filtration efficiency. Furthermore, excessive quantities of dust led to enhanced pressure drops, which in turn led to perforations in the dust layers.
In view of the low filtration efficiencies and low pressure drops of coarse-particle granular layers, and the high filtration efficiencies, high pressure drops, and short regeneration cycles of fine-particle granular layers, Yang [14,15] developed a novel graded filtering technique using a dual layer granular bed that promised to combine the strengths of these filter layers while eliminating their weaknesses. This dual layer granular bed consists of lower and upper filter layers, where the upper filter layer contains low-density, coarse particles, and the lower filter layer contains highly dense, fine particles. During filtration, the dust-containing airflow enters from the top of the filter, and passes through the upper filter layer where the majority of dust is captured; this is known as “coarse filtering”. The fine particles that penetrate the upper filter layer are then captured by the lower filter layer, thus completing the “fine filtering” process. The combination of these layers therefore provides a cascaded filter.
XSVWUHDP
XSSHUOD\HU
ORZHUOD\HU
PLGVWUHDP
IORZVWDELOL]LQJOD\HU
GRZQVWUHDP
Fig. 2. Schematic representation of the dual layer granular bed filter employed herein.
Fig. 3. Particle size distribution of the fly ash dust employed herein.
346
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
Fig. 4. Variation in the filtration velocity with filter time for a range of upper filter layer thicknesses.
Although dual layer granular beds are known to characteristically exhibit high filtration efficiencies and low pressure drops, studies into the effects of the filter layer thickness on the filtration characteristics of filters of this type remain rather scarce. Thus, to address this gap in the literature, we herein report our investigation into how changes in the thicknesses of the upper and lower filter layers affect the filtration characteristics of a dual layer granular bed.
2. Materials and methods 2.1. Experimental apparatus The experimental apparatus employed herein consists of a granular bed, an RBG-1000 aerosol generator, a measurement system, and other auxiliary devices (see Fig. 1 and Section 2.1.3 for details).
100
Filtration efficiency(%)
Filtration efficiency凚 %凛
100
99
98
97
96
180
200
220
240
260
98 96 94
90 88 86 84
280
180mm 205mm 230mm 255mm 280mm
92
0
10
20
Thickness凚 mm凛
1000 800 600 400 200
20
30
40
50
60
Outlet dust concentration凚 mg/m3)
Outlet dust concentration凚 mg/m3凛
180mm 205mm 230mm 255mm 280mm
10
50
60
(b)
1200
0
40
Time 凚 min凛
(a)
0
30
250
200
150
100
180
205
230
255
Time凚 min凛
Thickness凚 mm凛
(c)
(d)
Fig. 5. Effect of the upper filter layer thickness on its filtration efficiency and outlet dust concentration.
280
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
2.1.2. The RBG-1000 aerosol generator A Palas RBG-1000 aerosol generator (Palas Gmbn, Germany) was used to disperse the dust. More specifically, the non-adhesive dust particles (0.1–100 μm) were loaded into a cylindrical canister prior to feeding through a feed piston and rotating precision brush at a precisely controlled rate. The dust particles were then ejected from a nozzle with airflows up to 180 m/s; the strong turbulent flows and shear forces ensure that the dust particles are optimally dispersed within the main airflow.
Filtration efficiency凚 %凛
99.96 99.95 99.94 99.93 99.92 99.91 99.90
180
200
220
240
260
280
Thickness凚 mm凛
(a) Outlet dust concentration 凚 mg/m3)
2.1.1. The granular bed The granular bed employed herein was constructed using transparent organic glass tubes with an inner diameter of 100 mm. As shown in Fig. 2 (a), the granular bed consists of three granular layers, namely the flowstabilizing layer, a lower filtering layer, and an upper filtering layer. The flow-stabilizing layer was set above an air distributing plate, and acts as a homogenizer for incoming airflows. The flow-stabilizing layer also supports the lower filter layer, which is situated on top of the flow-stabilizing layer. The upper filter layer was then placed above the lower filter layer, as implied by its name. To examine the individual contributions of the upper and lower filter layers, the upper filter layer was placed on a #100 metallic mesh to ensure that the upper and lower filter layers were clearly separated (mesh size of #100 metallic mesh is 165 μm). Furthermore, a row of pressure measurement holes and 3 dust sampling holes were punctured in the sides of the glass tube (see Fig. 2(b)) to facilitate sampling of the inlets and outlets of the two filter layers. The inlet of the upper filter layer is referred to as the “upstream” section, the outlet of the upper filter layer (i.e., the inlet of the lower filter layer) is referred to as the “midstream” section, and the outlet of the lower filter layer is referred to as the “downstream” section. The inlet/outlet dust concentrations and dust particle size distributions were analyzed to reveal the filtration characteristics of each filter layer. The upper filter medium consisted of expanded perlite with particle size of 1.0–2.0 mm and bulk density of 70 kg/m3, whereas the lower filter medium consisted of sea sand with particle size of 0.3–0.5 mm and bulk density of 1340 kg/m3. The flow-stabilizing layer comprised sea sand with particle size of 2.0–2.5 mm and bulk density of 1550 kg/m3. The bulk density of the filter material was measured using the mass volume method.
347
10 8 6 4 2 0 180
200
220
240
260
280
Thickness (mm)
(b) Fig. 6. Effect of the upper filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed.
2.2. Experimental outline 2.1.3. Measurement system and auxiliary devices The various measurements performed by the measurement system include flow measurements, pressure drop measurements, and dust measurements. The flow rate was measured using the rotor's flow rate monitor, while online pressure drop measurements were conducted using a PY differential pressure sensor. A Wales-3000 aerosol particle size spectrometer was used to perform dust sampling and analysis. This spectrometer was combined with two sensors to simultaneously measure the particle size distribution, number concentration, and volume concentration of the dust at the inlet and outlet of each filter layer to essentially provide the filtration efficiency of each filter layer. The full particle filtration efficiency is calculated using the following formula:
η ¼ 1−
cout cin
ð1Þ
Here, cin is the volume concentration of dust at the inlet; cout, the volume concentration of dust at the outlet; and η, the filtration efficiency. The auxiliary devices employed in the experimental setup include a digital scale, a 3L13WC Roots blower, a ball miller, an experimental sieve, and a Vernier caliper.
The fly ash generated by some power plant was used as the dust sample in this work. The particle size distribution of this dust is outlined in Fig. 3. All filtration tests used fresh identical test dust; the test dust was not recycled, and the dust particle size distribution (PSD) was unchanged. To investigate the effect of the upper filter layer thickness on the filtration characteristics of the dual layer granular filter, five filtration experiments were conducted using thicknesses of 180, 205, 230, 255, and 280 mm. All other experimental conditions were identical, and the thickness of the lower filter layer and flow-stabilizing layer was 45 and 55 mm, respectively. To investigate the effect of the lower filter layer thickness on the filtration characteristics of a dual layer granular filter, five filtration experiments were conducted using thicknesses of 45, 55, 65, 75, and 85 mm. All other experimental conditions were fixed throughout these experiments, and the thickness of the upper filter layer and flow-stabilizing layer was 180 and 75 mm, respectively. Here, the thickness of the flow-stabilizing layer was increased because the thickness of the lower filter material greatly increases the pressure drop of the lower filter layer. To ensure good fluidization effect, the thickness and pressure drop of the flow-stabilizing layer were increased correspondingly. In the filtration process using filtration layers with different thicknesses, after filter layer filling was finished, the filtering material was
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353 2000
2400
Pressure drop凚 Pa凛
Granular bed pressure drop凚 Pa凛
348
1950
1900
1850
180
205
230
255
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400 0
280
0
Thickness凚 mm凛
10
2000
2000
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
2400
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
800 400 0
0
10
20
30
40
1200 800 400 0
50
0
10
20
30
40
50
Time凚 min凛
(c) h = 205 mm
(d) h = 230 mm
2400
2400
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
50
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600
Time凚 min凛
0
40
(b) h = 180 mm
2400
1200
30
Time凚 min凛
(a) Variation in the pressure drop across the granular bed with changes in upper filtering layer thickness
1600
20
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400 0
0
10
20
30
40
50
0
10
20
30
40
50
60
Time凚 min凛
Time凚 min凛
(f) h = 280 mm
(e) h = 255 mm Fig. 7. Variations in the pressure drop during dust filtration.
first subjected to reverse blowing, after which it entered a natural loose accumulation state through velocity reduction to ensure that the porosity and pore structure of the filter layer remained unchanged in each filtration test. The filtration experiments conducted herein were
performed using a constant pressure and a variable filtration velocity, i.e., the total pressure drop across the granular bed was maintained constant throughout, while the filtration velocity was gradually reduced. Upon commencing the filtration experiment, the granular
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
349
3.1.1. Effect of the upper filter layer thickness on the filtration velocity As shown in Fig. 4, the filtration velocity gradually decreased during the filtration process from 0.25 m/s (i.e., the initial filtration velocity) to 0.15 m/s, and the rate at which the filtration velocity decreased was found to increase gradually with time. This decrease in the filtration velocity can be attributed to gradual increases in dust deposition over time, and thus to ensure a constant pressure drop across the dual layer granular bed, the filtration velocity must be gradually reduced. In addition, the gradual increase in the decrease of filtration velocity is caused by changes in the filtration mechanism. More specifically, during the early stages of filtration, deep bed filtration dominates, and dust penetration is significant. During this stage, the accumulation of dust has a minimal impact on the pressure drop of the granular bed. As the degree of dust deposition increases, the filtration mechanism transitions from deep bed filtration to surface filtration. In this stage, the majority of dust will accumulate on the surface of the filter layer, resulting in the formation of dust cakes, which have a significant impact on the pressure drop of the granular bed [16]. Furthermore, as shown in Fig. 4, the deceleration of the filtration velocity is reduced upon increasing the thickness of the upper filter layer, which effectively extends the time taken for the filtration velocity to decrease from 0.25 to 0.15 m/s (i.e., the filtration time). More specifically, an increase in the upper filter layer thickness from 180 to 280 mm resulted in an increase in the filtration time from 43.3 to 56 min, thereby indicating that a thicker upper filter layer lengthens the regeneration cycle of this dual layer granular bed.
for dust to penetrate the filter layer. In addition, Fig. 5(b) shows that the filtration efficiency increases rapidly with time in the initial stages of the filtration process, although this increase in efficiency tapers off after 10 min. This phenomenon can be accounted for by the accumulation of dust over time, which enhances the pore structure of the filter medium. The accumulated dust in the filter layer then acts as a filter and increases the effective filtration efficiency of the upper filter layer [17]. When the quantity of accumulated dust reaches an optimal level, dust cakes form on the surface of the filter layer and stabilize its filtration efficiency. However, beyond this optimal level, any further increase in dust deposition leads to a decrease in the filtration efficiency of the filter layer. This is due to excessive quantities of dust deposition greatly increasing the pressure drop of the filter, which in turn perforates the dust cakes formed on its surface [18]. In addition, Fig. 5(c) shows that the changes in outlet dust concentration of the upper filter layer exactly mirror the trends presented in Fig. 5(b). More specifically, at a constant upper filter layer thickness, the dust concentration rapidly decreases with filter time, and the rate of decline in the dust concentration drops after 10 min. These effects can be accounted for by the filtration mechanism of the granular bed filter. Indeed, Squires et al. noted that the accumulation of dust in a granular bed undergoes three distinct stages, namely a penetration period, a transitional period, and a surface period. During the penetration and transition periods, a rapid decrease in the dust concentration takes place, as the penetration of dust in the filter layer is significant during these periods. As the filtration process progresses, dust penetration into the inner parts of the filter layer decreases, leading to rapid increases in surface dust accumulation. In the later stages of the filtration process, a dust cake layer forms on the surface of the filter layer, and the optimal filtration efficiency is reached. Furthermore, as outlined in Fig. 5(d), the outlet dust concentration of the upper filter layer decreases in an approximately linear fashion upon increasing the thickness. More specifically, the outlet dust concentration decreased from 254.56 to 132.89 mg/m3 when the thickness of the upper filter layer was increased from 180 to 280 mm. However, this dust concentration does not meet the requirements outlined in China's national emission standards, and so additional filtration using a lower filter layer is therefore required. Thus, from Fig. 5, it may be inferred that increases in the upper filter layer thickness result in a significant enhancement in the filtration efficiency of this layer, in addition to decreasing its outlet dust concentration.
3.1.2. Effect of the upper filter layer thickness on the filtration efficiency and outlet dust concentration As outlined in Fig. 5(a), the average filtration efficiency of the upper filter layer increased in an approximately linear fashion with its thickness. This can be attributed to increases in thickness rendering it more difficult
3.1.3. Effect of the upper filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed filter As outlined in Fig. 6(a) and (b), an increase in the upper filter layer thickness increases the average filtration efficiency of the dual layer granular bed and decreases its average outlet dust concentration.
beds were clean and free from dust accumulation. The filtration velocity at this point is denoted as the initial filtration velocity. As the filtration process progresses and increasing quantities of ash are accumulated in the filter layers, the filtration velocity will gradually decrease while the total pressure drop across the granular bed will remain unchanged. During the filtration experiments, the inlet dust concentration of the dual layer granular bed was fixed at 1.37 g/min. Given an initial filtration velocity of 0.25 m/s, this corresponds to a dust concentration of 11.67 g/m3. 3. Results and discussion 3.1. Effect of the upper filter layer thickness on the filtration characteristics of the dual layer granular filter
Fig. 8. Variations in the filtration velocity with filter time and lower filter layer thickness.
350
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
100.00
Filtration efficiency凚 %凛
Filtration efficiency凚 %凛
99.0 98.4 97.8 97.2 96.6 96.0 95.4 40
50
60
70
80
99.98 99.96 99.94 99.92 99.90 99.88 99.86 40
90
50
(a)
98 97 45mm 55mm 65mm 75mm 85mm
96 95 10
20
30
40
50
60
Outlet dust concentration凚 mg/m3凛
45mm 55mm 65mm 75mm 85mm
30
20
10
20
10 8 6 4 2 0 40
50
60
70
80
90
Fig. 10. Effect of the lower filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed.
40
10
90
(b)
(b)
0
80
Thickness(mm)
7LPH˄PLQ˅
0
Outlet dust concentration(mg/m3)
Filtration efficiency(%)
99
0
70
(a)
100
94
60
Thickness凚 mm凛
Thickness凚 mm凛
30
40
50
60
Time凚 min凛
(c) Fig. 9. Effect of the lower filter layer thickness on its filtration efficiency and outlet dust concentration.
Upon increasing the thickness of the upper filter layer from 180 to 280 mm, the average filtration efficiency of the dual layer granular bed increased from 99.926 to 99.944%, while its average outlet dust concentration decreased from 8.69 to 6.57 mg/m3. 3.1.4. Effect of the upper filter layer thickness on the pressure drop across the granular bed As indicated in Fig. 7(a), the pressure drop across the dual layer granular bed at the same initial filtration velocity of 0.25 m/s increases
approximately linearly with the upper filter layer thickness. More specifically, upon increasing the thickness of the upper filter layer from 180 to 280 mm, the total pressure drop across the dual layer granular bed increased from 1873 to 1978 Pa. In addition, for granular beds with the same upper filter layer thickness, the pressure drop across this upper layer increased with filtration time, while the pressure drop of the lower filter layer and the flow-stabilizing layer decreased (see Fig. 7(b)–(f)). This can be attributed to the fact that the pressure drop in the granular bed is determined by two key factors [19]. The first factor is the filtration velocity [20], where the decrease in the pressure drops across the lower filter layer and the flow-stabilizing layer are caused by decreases in the filtration velocity. The second factor is the accumulation of dust in the granular bed [21], where the pressure drop in the granular bed increases with an increase in dust deposition. Given a fixed total pressure drop across the granular bed, the filtration velocity decreases as the filtration process progresses, leading to corresponding reductions in the pressure drop across the granular bed. However, the pressure drop across the upper filter layer is determined mainly by dust deposition, due to the rapid rate at which dust deposition increases in the upper filter layer. As such, the pressure drop of the upper filter layer increases with filtration time. In contrast, the lower filter layer exhibits a decrease in the pressure drop over time due to a decrease in the filtration velocity caused by relatively insignificant dust accumulation in this layer. Similarly, the flowstabilizing layer also exhibits filtration velocity-induced decreases in the pressure drop with time, since dust deposition in this layer is effectively negligible. Fig. 7 also shows that the pressure drop of the
351
2400 3000
Pressure drop凚 Pa凛
Granular bed pressure drop凚 Pa凛
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
2800
2600
2400
2200
2000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1600 1200 800 400
2000 40
45
50
55
60
65
70
75
80
85
0
90
0
Thickness凚 mm凛
10
20
(a) Variation in the pressure drop in the granular bed upon increasing the lower filter layer thickness
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
1500
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
3000
2000
1000 500
0
10
20
30
40
2500
1500 1000 500 0
50
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
2000
0
10
Time凚 min凛
30
40
50
(d) h = 65 mm 3500
3000 total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
2500 2000 1500 1000 500
3000
Pressure drop凚 Pa凛
Pressure drop凚 Pa凛
20
Time凚 min凛
(c) h = 55mm
0
40
(b) h = 45 mm
2500
0
30
Time凚 min凛
total pressure drop pressure drop across the lower layer pressure drop across the upper layer pressure drop across the flow-stabilizing layer
2500 2000 1500 1000 500
0
10
20
30
40
50
0
0
10
20
30
40
50
60
Time凚 min凛
Time凚 min凛
(e) h = 75 mm
(f) h = 85 mm
Fig. 11. Variations in the pressure drop during dust filtration.
upper filter layer (as a proportion of the total pressure drop across the dual layer granular bed) increases from 10.67 to 15.87% upon increasing the layer thickness from 180 to 280 mm when the granular bed filter
was clean upon commencing the experiment. This indicates that the increase in the pressure drop induced by increases in the upper filter layer thickness are relatively mild.
352
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353
3.2. Effect of the lower filter layer thickness on the filtration characteristics of the dual layer granular bed
bed filter by 54.7%, thereby confirming that increasing the thickness of this layer is beneficial for reducing the outlet dust concentration.
3.2.1. Effect of the lower filter layer thickness on the filtration velocity Fig. 8 outlines the effect of the lower filter layer thickness on the filtration velocity, where it is apparent that a decrease in filtration velocity from 0.25 to 0.15 m/s took place over time at a gradually accelerating rate given the same lower filter layer thickness. Upon increasing the thickness of this layer, the deceleration in the filtration velocity slowed down, with an increase in the lower filter layer thickness from 45 to 85 mm extending the filtration time from 43.3 to 59 min, thereby indicating that a thicker filter layer lengthens the regeneration cycle of the dual layer granular bed.
3.2.4. Effect of the lower filter layer thickness on the pressure drop across the granular bed As indicated in Fig. 11(a), the overall pressure drop of the dual layer granular bed increases linearly upon increasing the lower filter layer thickness. More specifically, upon increasing the thickness from 45 to 85 mm, the total pressure drop increased from 1935 to 3077 Pa. In addition, Fig. 11(b)–(f) show that at the same lower filter layer thickness, the pressure drop of the lower filter layer and the flow-stabilizing layer decrease with longer filter times, while the pressure drop across the upper filter layer increases. This can be accounted for by the combinatorial effects of the filtration velocity and dust accumulation in the granular bed layers. In the case of a clean granular bed, the pressure drop in the lower filter layer increased from 66.20 to 78.84% as a proportion of the total pressure drop across the dual layer granular bed when its thickness was increased from 45 to 85 mm. These results therefore indicate that an increase in thickness of the lower filter layer has a significant effect on the pressure drop of the dual layer granular bed filter.
3.2.2. Effect of the lower filter layer thickness on its filtration efficiency and outlet dust concentration As outlined in Fig. 9(a), the average filtration efficiency of the lower filter layer increases upon increasing its thickness. More specifically, an increase in the thickness of this layer from 45 to 85 mm resulted in an increase in its average filtration efficiency from 96.59 to 98.56%. In addition, the average filtration efficiency was significantly improved upon increasing the thickness of the lower filter layer from 45 to 65 mm, although further increases (i.e., to 85 mm) resulted in only slight improvements. This can be attributed to the diffusion, blockage, and inertial effects of the lower filter layer increasing with increasing thickness [22,23], albeit in a non-linear fashion. In addition, the results presented in Fig. 9(b) demonstrate that at a critical thickness, a slight dip in the filtration efficiency is observed prior to a subsequent increase with filtration time. This can be attributed firstly to the filtration efficiency of the upper filter layer being relatively low at the beginning of the filtration process, which results in higher dust levels at the inlet of the lower filter layer. This gives rise to higher filtration efficiencies in the lower filter layer. Indeed, the inlet dust concentration of the lower filter layer gradually decreases as the filtration process continues, thereby resulting in a decrease in its filtration efficiency. When the dust concentration at the inlet of the lower filter layer is low, the probability of dust particle collision and reunion is small. Therefore, the probability of dust being trapped by the lower filter material is also small, resulting in lower filtration efficiency. Secondly, the accumulation of dust in the lower filter layer improves its pore structure and its filtration efficiency. The combination of these effects ultimately results in an initial decrease in the filtration efficiency, followed by a subsequent increase over time. Furthermore, Fig. 9(c) shows that the outlet dust concentration of the lower filter layer initially exhibits a rapid decrease upon increasing the filtration time, which is followed by a significantly slower decrease. This can be attributed to the small particle sizes present in the lower filter layer, and the resulting penetration of dust particulates in the filter layer during the initial stages of the filtration process, followed by a rapid accumulation on the surface with increasing filtration time. The dust cakes formed on the filter layer then stabilize its filtration efficiency [24]. 3.2.3. Effect of the lower filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed We then moved on to investigate the effect of the lower filter layer thickness on the filtration efficiency and outlet dust concentration of the dual layer granular bed. As shown in Fig. 10(a) and (b), for thicknesses ranging from 45 to 85 mm, an increase in the lower filter layer thickness led to a corresponding increase in the average filtration efficiency of the dual granular bed and a decrease in its average outlet dust concentration. In addition, an increase in thickness from 45 to 85 mm resulted in an increase in the average filtration efficiency of the dual layer granular bed from 99.926 to 99.967%, in addition to a decrease in its average outlet dust concentration from 8.69 to 3.94 mg/m3. A 40 mm increase in the thickness of the lower filter layer therefore decreased the outlet dust concentration of the granular
4. Conclusions We herein presented our investigation into the effect of upper and lower filter layer thicknesses on the filtration performance of dual layer granular bed filters using an experimental granular bed filter and fly ash dust. It was found that upon increasing the thickness of the upper filter layer from 180 to 280 mm, its average filtration efficiency increased from 97.81 to 98.60%, while its average outlet dust concentration decreased from 254.54 to 132.89 mg/m3. In addition, the average outlet dust concentration of the dual layer granular bed was reduced from 8.69 to 6.57 mg/m3, and its overall pressure drop increased from 1873 to 1978 Pa at the same initial filtration velocity of 0.25 m/s. Furthermore, increasing the thickness of the upper filter layer also lengthened the granular bed regeneration cycle, as the filtration time was increased from 43.3 to 56 min. We also found that when the thickness of the lower filter layer was increased from 45 to 85 mm, the average filtration efficiency of the dual layer granular bed increased from 99.926 to 99.967%, while the average outlet dust concentration decreased from 8.69 to 3.94 mg/m3, which corresponds to a 54.7% reduction in the average outlet dust concentration. These results indicate that an increase in the thickness of the lower filter layer had a significant impact in reducing the outlet dust concentration of the dual layer granular bed. However, it also resulted in the pressure drop of the dual layer granular bed increasing greatly from 1935 to 3307 Pa at the same initial filtration velocity of 0.25 m/s when the lower filter layer thickness was increased from 45 to 85 mm. This represents a 70.9% increase in the pressure drop. The matching of the upper and lower layer thicknesses is related to the inlet dust characteristics, allowable pressure drop and the dust discharge concentration. For the test dust in this paper, it is suitable to select a thickness of 280 mm for the upper layer, a thickness of 45 mm for the lower layer, and a thickness of 55 mm for the flow-stabilizing layer. At this time, the outlet dust concentration is 6.57 mg/m3, and the total pressure drop is 1978 pa. Acknowledgments This work was sponsored by K.C. Wong Magna Fund in Ningbo University, and supported by the National 863 High Technology Research Program of China (2008AA05Z205) and the Student Research and Innovation Program of Ningbo University (G17102). References [1] W. Zhu, W. Song, W. Lin, Catalytic gasification of char from co-pyrolysis of coal and biomass, Fuel Process. Technol. 89 (2008) 890–896.
G.-H. Xiao et al. / Powder Technology 335 (2018) 344–353 [2] S. Krerkkaiwan, C. Fushimi, A. Tsutsumi, P. Kuchonthara, Synergetic effect during co-pyrolysis/gasification of biomass and sub-bituminous coal, Fuel Process. Technol. 115 (2013) 11–18. [3] Y.S. Chen, Y.P. Chyou, S.C. Li, Hot gas clean-up technology of dust particulates with a moving granular bed filter, Appl. Therm. Eng. 74 (2015) 146–155. [4] G.Q. Liu, S.P. Deng, X.R. Qian, Research progress of coal pyrolysis technology in China, Mod. Chem. Ind. 27 (2007) 37–43 (in Chinese). [5] K.Y. Shi, G.H. Yang, S. Huang, Study on filtering characteristics of aerosol particulates in a powder-grain dual-layer granular bed, Powder Technol. 272 (2015) 54–63. [6] G. Xiao, X. Wang, J. Zhang, Granular bed filter: a promising technology for hot gas clean-up, Powder Technol. 244 (2013) 93–99. [7] S.R. Tian, G.H. Yang, Z. Li, Cascade filtration properties of a dual-layer granular bed filter, Powder Technol. 301 (2016) 545–556. [8] C.J. Hsu, S.S. Hsiau, A study of filtration performance in a cross-flow moving granular bed filter: the influence of gas flow uniformity, Powder Technol. 274 (2015) 20–27. [9] B.M. Wenzel, C.B. Porciúncula, N.R. Marcilio, Filtration of dust in an intermittent moving granular bed filter: performance and modeling, Sep. Purif. Technol. 133 (2014) 108–119. [10] I.A. El-Hedok, L. Whitmer, R.C. Brown, The influence of granular flow rate on the performance of a moving bed granular filter, Powder Technol. 214 (2011) 69–76. [11] Y.S. Chen, S.S. Hsiau, S.C. Lai, Filtration of dust particulates with a moving granular bed filter, J. Hazard. Mater. 171 (2009) 987–994. [12] I. Rodon, K.C. Lee, R. Pfeffer, Granular-bed filtration assisted by filter-cake formation: 2. The panel bed gas filter with puffback renewal of gas-entry surfaces, Powder Technol. 155 (2005) 5–16. [13] M.S. Wu, K.C. Lee, R. Pfeffer, Granular-bed filtration assisted by filter cake formation 3. Penetration of filter cakes by a monodisperse aerosol, Powder Technol. 155 (2005) 62–73.
353
[14] G.H. Yang, J.H. Zhou, H.P. Shu, A new dual-layered granule bed filter for cleaning-up gas, J. Power Eng. 25 (2005) 891–894 (in Chinese). [15] G.H. Yang, J.H. Zhou, A study of the pressure-drop characteristics of a dual-layer filter-material filtering bed, J. Eng. Therm. Energy Power 22 (2007) 267–269 (in Chinese). [16] Y.S. Chen, S.S. Hsiau, Cake formation and growth in cake filtration, Powder Technol. 192 (2010) 217–224. [17] K.Y. Shi, G.H. Yang, S.R. Tian, Study of filtration characteristics of PM2.5 by a powdergrain dual-layer granular bed, J. China Univ. Min. Technol. 44 (2015) 138–143 (in Chinese). [18] A.M. Squires, Granular-bed filtration assisted by filter-cake formation: 4. Advanced designs for panel-bed filtration and gas treating, Powder Technol. 155 (2005) 74–84. [19] D. Bémer, I. Subra, Y. Morele, Experimental study of granular bed filtration of ultrafine particles emitted by a thermal spraying process, J. Aerosol Sci. 63 (2013) 25–37. [20] Y.Q. Zhang, P. Liang, J. Yu, Studies of granular bed filter for dust removal in the process of coal pyrolysis by solid heat carrier, RSC Adv. 7 (2017) 20266–20272. [21] Y.S. Chen, S.S. Hsiau, H.Y. Lee, Filtration of dust particulates using a new filter system with louvers and sublouvers, Fuel 99 (2012) 118–128. [22] A. Zamani, B. Maini, Flow of dispersed particles through porous media–deep bed filtration, J. Pet. Sci. Eng. 69 (2009) 71–88. [23] C. Tien, B.V. Ramarao, Granular Filtration of Aerosols and Hydrosols, Elsevier/ Butterworths, second edition, 2007. [24] D. Stanghelle, T. Slungaard, O.K. Sonju, Granular bed filtration of high temperature biomass gasification gas, J. Hazard. Mater. 144 (2007) 668–672.