Oil shale separation using a novel combined dry beneficiation process

Oil shale separation using a novel combined dry beneficiation process

Fuel 180 (2016) 148–156 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Full Length Article Oil shal...

2MB Sizes 2 Downloads 95 Views

Fuel 180 (2016) 148–156

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Oil shale separation using a novel combined dry beneficiation process Xiaodong Yu, Zhenfu Luo ⇑, Xuliang Yang, Haishen Jiang, Enhui Zhou, Bo Zhang, Liang Dong School of Chemical Engineering & Technology, China University of Mining and Technology, 221008 Xuzhou, China

h i g h l i g h t s  A combining dry separation process for oil shale of 100–0 mm was proposed.  The probable error (E) was 0.155–0.21 when oil shale was cleaned by compound dry cleaning.  The probable error (E) was 0.16–0.185 when oil shale was cleaned by a VADMFB separator.  Oil shale can be separated effectively by the combined dry beneficiation process.

a r t i c l e

i n f o

Article history: Received 15 February 2016 Received in revised form 19 March 2016 Accepted 7 April 2016 Available online 13 April 2016 Keywords: Compound dry separator Vibration air-dense medium fluidized bed separator Combined dry beneficiation process Oil shale

a b s t r a c t Significant mineral impurities are produced during oil shale mining. However, inorganic minerals can be partially discarded, with an improvement in oil shale quality, through physical separation methods. Based on analysis of physical properties of oil shale, this paper firstly proposes a separation process combining compound dry cleaning and vibrated air dense medium fluidized bed (VADMFB) separation, for separation of oil shale across the full size range of 100–0 mm. We systematically analyzed the effects of changes in vibration amplitude and frequency on separation results and oil ratios for both systems. When oil shale with a size range of 100–6 mm was separated using compound dry cleaning, using optimal vibrational parameters of f = 53 Hz and A = 3.6 mm, yields of concentrate was 20.49%, respectively, with corresponding oil content of 11.20%, The probable error (E) was 0.155–0.21 with the largest oil content segregation degree. Oil shale with a size fraction of 6–0 mm was separated in a VADMFB separator at optimal vibrational conditions (f = 36 Hz and A = 2.5 mm). Resultant yields of concentrate was 21.70%, respectively, with corresponding oil content of 10.1%. E was 0.16–0.185 with the largest oil ratio segregation intensity. This study thus confirms that the proposed combined dry beneficiation process can effectively separate oil shale along the full size range of 100–0 mm. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Energy resources provide one of the most important material foundations for social development. Oil and coal are still the main energy resources at present, but these fossil energy resources are non-renewable and will eventually be used up. A high rate of coal production is still presently maintained in China, but overall, coal production and consumption are progressively decreasing. As a result, it is becoming necessary to develop non-conventional energy sources. Oil shale is a particular type of sedimentary rock containing abundant organic matter. Shale oil and gas can be attained after pyrolysis of organic matter [1–3]. Oil shale is a combustible solid and is mainly composed (>70 wt% content) of

⇑ Corresponding author. E-mail address: [email protected] (Z. Luo). http://dx.doi.org/10.1016/j.fuel.2016.04.036 0016-2361/Ó 2016 Elsevier Ltd. All rights reserved.

inorganic minerals. Of the latter, clay minerals predominate, in addition to quartz, mica, carbonate rocks, and pyrite. Reserves of oil shale, a form of non-conventional energy, far exceed those of petroleum. Based on exploration to date, America has the most abundant oil shale resources in the world with ascertained geological resources of 33,400 billion tons. Geological reserves in China total 719.9 billion tons, equivalent to 47.6 billion tons of oil. In the latter country, oil shale is mainly used for power generation by combustion and to produce oil by pyrolysis. Gasoline, kerosene, diesel oil, paraffin, and other types of chemical products can be attained after hydro-cracking of shale oil [4–6], which therefore has significant value. Kerogen in oil shale is derived from macro-molecular organisms; it is dispersed within sedimentary rock and is not soluble in organic solvents. This is the most common form in which organics are present within rock strata [7–10].

149

X. Yu et al. / Fuel 180 (2016) 148–156

Significant inorganic minerals are produced when oil shale is mined, leading to a decrease in the grade of oil shale and in oil yield, and increasing the cost of production during pyrolysis; this leads to the disadvantages in terms of comprehensive utilization of oil shale energy. Some comparisons can be drawn between coal and oil shale. The density of oil shale is around 2.35 g/cm3, generally higher than that of coal; the principle of coal beneficiation, however, provides insights that are relevant to oil shale separation processes. Traditional water separation processes mainly include the jigging washing process and heavy medium separation. Both of these process are barely able to reach the required oil shale separation density and the separation effects they produce are therefore poor and inadequate for upgrading of oil shale. The oil content is inversely proportional to the density of oil shale. The density of inorganic mineral impurities is higher, and these should be discarded to attain pure oil shale with low density. Oil shale can be separated effectively from inorganic mineral impurities through dry separation methods. Based on studies using compound dry cleaning apparatus [11], the lower separation size limit of this method is 6 mm; materials with a size range of 0–6 mm can, however, be separated after compound dry cleaning in a vibrating air dense medium fluidized bed (VADMFB) separator. Dry separation processes can therefore be used to separate oil shale across the full particle size range of 0–100 mm. In this paper, based on the laboratory experiments, we propose a novel dry beneficiation process combining the compound dry cleaning with VADMFB separation to beneficiate oil shale. The effects of main factors including vibration frequency and amplitude, on oil shale separation were systematically investigated and the optimal operational parameters were identified in the laboratory separation process, providing technological support for the industrialization promotion of dry oil shale separation technologies. 2. Experimental 2.1. Dry separation apparatus Fig. 1 shows the oil shale dry separation system. This is composed of a crude ore preparation system, a separator system, a medium purification, circulation and density control system, and an air supply and dust removal system. The crude ore preparation system contains a pre-classification screen, crusher, surge bin, and other related elements. Crude ore is pre-classified and crushed before it is conveyed to the surge bin for separation preparation. The separator system is composed of compound dry cleaning apparatus and a VADMFB separator. The compound dry separator and VADMFB separator deployed in this study are both laboratory equipment, and the bed form of compound dry separator is right-angled trapezoid, with an upper length of 0.2 m, a lower

length of 0.5 m, a height of 1 m, an effective separating area of 0.35 m2. The bed form of a VADMFB separator is rectangle, with a length of 1.876 m, a width of 0.4 m and an effective sorting area of 0.75 m2. Different particle sizes of oil shale can be separated effectively by the combined operation of a compound dry separator and a VADMFB separator. The medium purification, circulation, and density control system is used to purify and recycle the heavy medium discharged from the VADMFB separator through a scalping screen, magnetic separator, and diverter. The density and height of the vibrated fluidized bed can be adjusted by controlling heavy medium circulation content and the purification ratio in accordance with separation demand. The air supply and dust removal system contains an air blower, induced draft fan, dust collector, air bag, and valve, among other elements; it is used for air supply and dust collection. 2.2. Experimental process The raw ore is first delivered onto the classifying screen, which has a diameter of 100 mm. The screen overflow is crushed to a grain size <100 mm, and this is then mixed with the screen underflow, which is separated using the compound dry cleaning apparatus. Under the joint effects of vibration and wind, and due to the buoyancy effect of the autogenous medium, tailings are discharged and remaining middlings are concentrated. As noted above, minerals with a particle size of 6–0 mm cannot be separated using the compound dry cleaning apparatus [11], and concentrate, middlings, and tailings are therefore separated using a classifying screen with a diameter of 6 mm. The underflow is processed in the VADMFB separator under the influence of vibration and buoyancy of the air dense medium (ferro-silicon powder). The separation process is dependent on the density of the fluidized bed. After separation, dense media in the concentrate, middlings, and tailings are separated using the scalping screen, and final products can then be attained. One part of the dense medium is separated using a magnetic separator to discard non-magnetic material and attain ferro-silicon powder; the other part is circulated for separation. The dust captured by the dust separator is recycled and reused. 2.3. Material properties The physical properties of oil shale affect separation results; of particular importance are its density composition and oil ratio changes. Oil shale is a particular type of geo-material and its density, oil content, and other physical properties exhibit aeolotropism and nonlinearity in terms of the effects of geological structures [12]. As the temperature changes, the thermal stress generated by mismatch of thermal deformation in internal regions of oil shale

Feed stock

Dust Efflux air

Φ100mm Φ6mm

Φ6mm Φ6mm Φ0.5mm Φ0.5mm Φ0.5mm

Block Block Block concentrate middlings tailing

Forth tailing

Forth Forth middlings concentrate

Magnetic separation tailing Additional magnetite

Air

Fig. 1. Diagram of the oil shale dry separation system.1-classifying screen (¢100 mm), 2-crusher, 3-surge bunker, 4-feeder, 5-compound dry cleaning apparatus, 6/7/8classifying screen (¢6 mm), 9-VADMFB separator, 10-pressure transducer 11/12/13-scalping screen (¢0.5 mm), 14-diverter, 15-cycling medium bin, 16-feeder, 17-magnetic separator, 18-cycling medium bin, 19-feeder, 20-medium bin, 21-medium feeder, 22-dust collector, 23-induced draft fan, 24-valve, 25-air bag, 26-air blower.

150

X. Yu et al. / Fuel 180 (2016) 148–156

damages the minerals less strength or around the micro cracks and pores which vary in sizes and structures. Changes of the thermal stress lead to aeolotropism of stress of oil shale. Simultaneously, extreme heterogeneity of oil shale media consisting of inorganic minerals and organics may result in non-linearity of oil shale deformation when the load damages the internal structure of oil shale. Its physical properties generally depend on its microstructures and compositional distribution. It is consequently necessary to analyze the latter aspects of oil shale to identify its physical characteristics and to provide technological support for the oil shale separation process. The oil content is an important parameter to evaluate oil shale grade and is expressed by ‘‘x”. A scanning electron microscope (SEM) was also used to analyze surface morphological structure, density, and oil content characteristics of crude ore (x 6 5%), concentrate (x P 10%), middling (3% 6 x 6 10%) and tailings (0 6 x 6 3%).The oil content of oil shale is 3.82%. Table 1 shows oil content for raw ores with different particle sizes. Table 2 shows oil shale sink-float test results, upon which are based the washability curves shown in Fig. 2. The true density and bulk density of the dense medium (ferro-silicon powder) are 6.9 g/cm3 and 3.17 g/cm3 respectively, while its average particle diameter and dominant size range are 0.06 mm and 0.074–0.045 mm, respectively. The ferro-silicon powder has a silicon content of 75% and an iron content of 25%, the size range of the ferro-silicon used is shown in Fig. 3. 2.4. Evaluation The purpose of oil shale separation is to raise its grade and increase oil content. The combined oil shale separation process described here, using compound dry cleaning apparatus and VADMFB separation, takes into account density differences of impurities. The oil content in oil shale decreases with increasing oil shale density but the combined separation process leads to enrichment of oil content. It can be noted from Fig. 4 that the discharge baffle of the compound dry cleaning separator is divided into 5 parts, from feed end to tailing discharge end. The VADMFB separator is also divided into 5 parts along the height of the bed. This allows for the derivation of different products, the yields and oil content of which can be measured. Oil segregation intensity is defined as follows (1), using ash segregation intensity as a Refs. [12,13].

Soil

rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi 1 Xn ¼ ðxi =xo  1Þ2 i¼1 n1

ð1Þ

where xi is the oil content of the i layer (%), xo is the oil content of the feed oil shale (%), and n is the sampling number. This parameter illustrates the deviation intensity of the oil content of layer i from the oil content of feed ore. Better separation results will be obtained when Soil is higher. In the case of the compound dry cleaning apparatus, concentrate with higher oil content is found near the feed end, gangue with lower oil content is found close to the tailing discharge end, while middlings are distributed between the two ends.

Table 1 Oil ratios for raw ores with different particle sizes. Particle size (mm)

Yield (%)

Oil content (%)

+100 100 + 50 50 + 25 25 + 13 13 + 6 6 + 0.5 0.5

19.93 13.64 13.33 15.19 12.69 18.92 6.31

2.21 3.15 5.13 4.56 5.57 3.81 2.04

Total

100

3.80

In the case of the VADMFB separator, concentrate with higher oil content is found in the upper bed and gangue with lower oil content is found in the bottom bed. Fig. 4 shows the sampling method used in both separators. The separation process of the compound dry cleaning apparatus depends on density and also on the oil content. Stability of the oil content of material in different parts of the bed is key to maintaining separation. The oil content distribution of material on the bed can directly indicate material movement and separation effects at a particular point in time. Measurement of material oil content on the bed can therefore render separation results clearer, enabling optimization of separation conditions and results. To simplify the study of oil content distribution on the bed, a coordinate system based was setup, as shown in Fig. 5. The X-axis extends from the feed end to the tailings end and the bed in this direction is divided into 10 equal parts. The Y-axis extends from the backboard to the discharge end. Material was sampled at the cross region, where oil content was measured. 3. Results and discussion 3.1. Physical properties of oil shale As can be seen in Fig. 6(a), crude ore is a fine-grained rock with an irregular surface and a thin layer laminated structure. Color changes with organic matter content. White materials are inorganic minerals, while black material is organic. The density of oil shale mainly depends on the content, composition, and density of inorganic minerals that infiltrated during the deposition process [14]. Oil content depends on the degree of porosity development of oil shale [15]. In Fig. 6(a), inorganic minerals are apparently more frequent than organic materials; their density is also greater and the density of crude ore is therefore comparatively larger. Its porosity is low, and this is mainly residual porosity of organic cells. There is low organic matter and, as a result, the oil content is relatively low. In Fig. 6(b), black material areas are larger but that there are still some white material areas surrounded by black. This indicates a comparative increase in organic matter content in concentrate; the density of organic material is lower, so the density of oil shale concentrate is also lower. In Fig. 6(c), it can be noted that areas of white material are more extensive than areas of black material, and the latter are partially surrounded by white areas; this indicates an apparent increase in inorganic minerals. As a result, the density and oil content of middlings are comparable to those of crude ore. In Fig. 6(d), it can be noted that white material areas increase significantly, with negligible black areas; this indicates a sharp increase in the content of inorganic minerals in tailings, with an increase in density. In this case, the oil content is extremely low. 3.2. Analysis of factors affecting the separation efficiency of the compound dry cleaning apparatus Factors that affect the separation efficiency of the compound dry separator include raw material properties and operational parameters. There have already been preliminary investigations related to these factors [11]; these have shown that the main influential raw material properties are external moisture and size distribution. Specifically, external moisture content and size limit should be <8% and 100 mm, respectively. Based on these considerations, this study used oil shale with a size fraction of 100–0 mm. Relevant operational parameters mainly include vibration frequency and amplitude, air flow rates, and the horizontal and vertical angles of the bed surface. The influence of air flow on separation is only to loosen material. Vibration, meanwhile, contributes to

151

X. Yu et al. / Fuel 180 (2016) 148–156 Table 2 Sink-float test results for raw ores of 100–0 mm. Density (g/cm3)

1.8 1.8–1.9 1.9–2.0 2.0–2.1 2.1–2.2 2.2–2.3 2.3–2.4 2.4–2.5 +2.5 Total

Yield (%)

Oil content (%)

5.89 1.39 1.39 2.12 1.16 3.59 3.95 8.76 71.75

25.86 15.19 12.34 8.12 6.44 5.47 3.69 2.82 1.50

100.00

3.82

Float products

Sink products

dp ± 0.1

Yield (%)

Oil content (%)

Yield (%)

Oil content (%)

Density (g/cm3)

Yield (%)

5.89 7.28 8.67 10.79 11.95 15.54 19.49 28.25 100.00

25.86 23.82 21.98 19.26 18.01 15.12 12.80 9.70 3.82

100.00 94.11 92.72 91.33 89.21 88.05 84.46 80.51 71.75

3.82 2.44 2.25 2.10 1.95 1.89 1.74 1.65 1.50

1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5

7.28 2.78 3.51 3.28 4.75 7.54 12.71 71.75

Fig. 2. Washability curves of crude ore with a grain size of 100–0 mm.

Fig. 3. Grain size composition of dense medium in VADMFB separator.

lamination of materials on the bed. The horizontal and vertical angle of the bed plays a role in material transport but has little influence on sorting. Based on several experiments, the best operating parameters for a compound dry cleaning separator have been determined to be as follows: air flow U is 1.93 m/s, horizontal angle of 7°, and vertical angle of 2° [16].

3.2.1. Effect of vibration frequency on performance of compound dry cleaning apparatus Based on the above optimal operating parameters [16], the effects of vibration frequency were investigated in this study when the amplitude is 3.0 mm. The results were as shown in Figs. 7 and 8. Oil content dispersion on the bed and Soil were also analyzed, as shown in the same figures. The figure visually shows the distribution of bed oil content at different frequencies. The (0, 0) point represents the origin point of feeding, with higher oil content (red part) distributed in the discharge area along the Y-axis. Oil content tended to increase along the Y-axis and to simultaneously decrease along the X-axis. However, the distribution of oil content was uneven, especially in the 100–300 mm region of the Y-axis; this indicates that the segregation of bed material layer is not obvious at this time. The particles at the bottom were too active which lead to significant back mixing of particles on the bed surface and then a poor separation performance. With increasing frequency of vibration, the distribution of bed oil content gradually increased along the Y-axis and decreased along the X-axis. Oil content distribution was most notable at 53 Hz; at this frequency, the oil content of bed material gradually decreased along the X-axis while increasing along the Y-axis, indicating optimal separation, the bed material layer could achieve fully loose, with materials entirely divided according to density. With continued increase in frequency to 55 Hz, there were clearly different degrees of staggered distribution of the oil content along X and Y-axes. The oil content increased along the Y-axis, while it decreased along the X-axis, indicating that some high oil content concentrate is gradually mixed with gangue, while low oil content gangue is mixed with concentrate with increasing frequency; this leads to a mismatch of bed materials and significant deterioration of separation efficiency. It can be seen from Fig. 8 that the intensity of segregation (IOS) of oil content presented an inverted V trend with increasing frequency. Although low vibration frequency activates materials at the bottom of the bed, these cannot be stratified sufficiently due to poor activity of the whole bed, resulting in low IOS. A higher vibration frequency renders materials at the bottom of the bed more active and the looseness of the entire bed gradually increases as a result of the transfer of vibration energy to inter-particles. The bed became fully loose at a frequency of 53 Hz and, at this frequency, materials were sufficiently stratified by density, with the highest IOS. With further increases in frequency, excessive vibration energy led to undesirable activity of materials within the bed. Additionally, the distance limitation of material movement significantly enhanced back mixing, leading to ore particle mixing becoming dominant. As a consequence, materials could not be stratified by density and there was a corresponding decrease in IOS.

152

X. Yu et al. / Fuel 180 (2016) 148–156

Feed

Feed

Clean mineral Middling

Tailings Clean mineral Middling Tailings

(a) Compound dry cleaning apparatus

(b) VADMFB seoarator

Fig. 4. Diagram showing stratified sampling for separation. (a) Compound dry cleaning apparatus, (b) VADMFB separator.

Fig. 5. Distribution of sampled material on the compound dry cleaning separator bed for oil content measurement.

3.2.2. Effect of vibration amplitude on performance of compound dry cleaning The effects of vibration amplitude on the compound dry cleaning separator were also investigated when the frequency was 53 Hz, based on the above optimal operating parameters, with results shown in Figs. 9 and 10. Oil content dispersion on the separation bed and Soil were also analyzed, as shown in the same figures. Fig. 9 shows the distribution of oil content along the bed surface of the compound dry separator at various amplitudes. Materials were fed at point (0, 0). At an amplitude of 3.2 mm, materials containing relatively high oil content were distributed in the discharge region along the Y-axis, and oil content decreased along the nega-

tive Y-axis. However, the distribution was uneven, particularly within the 50–350 mm range along the Y-axis, indicating that small amplitudes fail to sufficiently loosen materials. These particles on the surface of the bed could not obtain enough energy to become activated. Consequently, the particles bed could not be efficiently stratified, producing poor separation results. As the amplitude increased, the oil content on the bed surface increased in the Y-direction while decreasing in the X-direction. Oil content on the bed surface dispersed effectively when A was 3.6 mm. Furthermore, this increased from the back plane to the discharge area along the Y-direction, while decreasing from the feed point along the X-axis, reaching its lowest value at the tailings end; this indicates optimal separation. With further increases in amplitude to 4.0 mm, the distribution of oil content on the bed appeared to extend to different degrees either along the X or Y directions, indicating drastic separation of particles in the bed due to excessive amplitude. Therefore, the activity of the bottom and top layer particles were both enhanced too much, leading to disorder motion of bed particles. Specifically, the transfer of vibration energy affects the forces acting on particles, leading to irregular particle movement, as well as to material back mixing throughout the whole bed. Thus, poor separation results were therefore obtained. Fig. 10 shows variations in IOS at different amplitudes, with an inverted V trend. Particles at the bottom of bed were strongly activated at low amplitudes, but vibration energy gradually weakened in the transmission process. Oil shale on the separation bed exhibited low activity at low amplitudes and could not be fully loosened, and stratification was therefore difficult. As a result, IOS was low. With increasing amplitude, the activity of particles at the bottom

Fig. 6. Microstructures of oil shale with different oil content. (a) Crude ore (x = 3.82%), (b) concentrate (x = 25.6%), (c) middling (x = 5.28%), (d) tailing (x = 0.86%).

X. Yu et al. / Fuel 180 (2016) 148–156

153

Fig. 7. Effects of vibration frequency on oil content dispersion in the case of the compound dry cleaning apparatus.

Fig. 8. Effects of frequency on the IOS of oil content in the compound dry cleaning apparatus.

Fig. 10. Effects of amplitude on IOS of oil content in the compound dry cleaning apparatus.

Fig. 9. Effects of amplitude on oil content dispersion in the case of the compound dry cleaning apparatus.

154

X. Yu et al. / Fuel 180 (2016) 148–156

Table 3 Sink-float test results for 6–0 mm oil shale after compound dry cleaning. Density (g/cm3)

1.8 1.8–1.9 1.9–2.0 2.0–2.1 2.1–2.2 2.2–2.3 2.3–2.4 2.4–2.5 +2.5 Total

Yield (%)

Oil content (%)

1.65 5.86 0.85 4.25 4.18 2.23 7.86 13.42 59.70

28.89 17.89 10.30 8.35 5.43 3.80 2.46 1.38 0.95

100.00

3.22

Float accumulation

Sink accumulation

dp ± 0.1

Yield (%)

Oil content (%)

Yield (%)

Oil content (%)

Density (g/cm3)

Yield (%)

1.65 7.51 8.36 12.61 16.79 19.02 26.88 40.30 100

28.89 20.31 19.29 15.60 13.07 11.98 9.20 6.59 3.22

100 98.35 92.49 91.64 87.39 83.21 80.98 73.12 59.70

3.22 2.79 1.84 1.76 1.44 1.24 1.17 1.03 0.95

1.8 1.9 2.0 2.1 2.2 2.3 2.4 2.5

7.51 6.71 5.10 8.43 6.41 10.09 21.28 59.70

Fig. 13. Effects of amplitude on the IOS of oil content in the VADMFB separator.

Fig. 11. Washability curves of 6–0 mm oil shale after compound dry cleaning.

bed particle mismatch phenomenon. As a result, the material was not stratified by density and the degree of oil segregation of bed material gradually decreased. 3.3. Analysis of factors affecting the VADMFB separator

Fig. 12. Effect of frequency on IOS of oil content in the VADMFB separator.

of bed strengthened, and the activity of the whole bed was gradually enhanced by transfer of vibration energy between particles. Materials were thus sufficiently loosened. The degree of looseness of the bed was optimal when A was 3.6 mm. At this point, there was clear stratification and the oil segregation degree of bed material reached its maximum value. As amplitude continued to increase, particle motion at the top and bottom of the bed became too intense, with bed thickness being thinner, and the bed movement distance being shorter; this led to a more significant

In order to investigate the properties and separation efficiency of 6–0 mm oil shale, sink-float tests were carried out, as shown in Table 3 and Fig. 11. The main factors influencing the separation efficiency of the VADMFB separator include material properties and operational parameters. Material properties have already been outlined and these met the separation conditions; our investigation thus focused mainly on the influence of operational parameters on separation efficiency. Related research [17] has shown that the stability of the separation effect of the VADMFB separator is influenced by the uniformity and stability of sorting bed density. However, under conditions of particular vibration intensity and bed height, changes in gas velocity within the proper operating range did not significantly impact the pressure and density distribution of each point in the bed. Bed pressure remained linearly distributed along the bed height under different conditions. The density of the bed was evenly distributed while changes in amplitude and vibration frequency had a significant impact on separation efficiency. Based on the above findings, this study mainly analyzed the effects of vibration amplitude and frequency on the performance of the VADMFB separator. 3.3.1. Effects of vibration frequency on VADMFB separation performance To investigate the effects of frequency on the separation performance of the VADMFBS for 6–0 mm oil shale, separation tests were

155

X. Yu et al. / Fuel 180 (2016) 148–156 Table 4 Partition coefficient results of 100–0 mm oil shale for compound dry cleaning. Density (g/cm3)

1.9 1.9–2.0 2.0–2.1 2.1–2.2 2.2–2.3 2.3–2.5 +2.5

Average density (g/cm3)

1.75 1.95 2.05 2.15 2.25 2.40 2.60

Total

Sink-float results (accounting for the feed) (%) Tailings

Middlings

0.15 0.08 0.09 0.08 0.65 5.26 51.12

1.52 0.55 0.45 0.35 2.05 6.16 11.00

7.84 2.35 1.35 0.80 2.15 3.00 3.00

57.43

22.08

20.49

Calculated feedstock Sink-float results (%)

Concentrate

9.61 3.13 1.95 1.27 5.04 14.19 64.81 100

Calculated middlings tailings results (%)

1.67 0.63 0.54 0.43 2.70 11.42 62.12

Partition coefficient High density separation (%)

Low density separation (%)

17.56 21.14 28.57 34.96 55.67 79.20 95.39

8.98 12.70 16.67 18.60 24.07 46.06 82.29

79.51

density. At a frequency of 36 Hz, Soil was 1.83, representing optimal separation effects. As the vibrating frequency increased further, the energy input to the bed became excessive. In the process of energy transfer, the suffering force characteristics of grains changed, leading to unfavorable disorderly movement of bed particles; as a result, the bed could not be layered according to density. Separation effects therefore weakened.

Fig. 14. Partition curves for compound dry separation.

carried out, with bed height of 75 mm, amplitude of 1.4 mm, and air flow of 1.15 cm/s. Results are shown in Fig. 12. This research assumes that uniformity and stability of bed density can be maintained when vibration intensity K is 1.21–1.60, air flow U is 1.10–1.28 cm/s, and bed height is 75–125 mm [18,19]. As shown in Fig. 12, the IOS of oil content in bed materials first increased and then decreased as frequency increased, with constant amplitude and air flow velocity. This was because vibration frequency determined the periodic excitation frequency acting on surplus air flow, which directly determined the number of bubble groups produced per unit time. As vibrating frequency increased, the number of bubble groups produced per unit time increased and the number of merging bubble groups within a certain area increased, resulting in the layering of bed particles according to

3.3.2. Effects of vibration amplitude on VADMFB separation performance Effects of amplitude on VADMFB separator performance were investigated through the experimental setup shown in Fig. 13; bed height was 75 mm, frequency was 36 Hz, and air flow was 1.15 cm/s. As shown in Fig. 13, the IOS of oil content in bed materials first increased and then decreased as amplitude increased. The extent of variation was limited when frequency and air flow velocity were constant. Bed activity was low and viscosity was high when amplitude was small. Under these conditions, the resistance acting on the particles was high and settling velocity was low, and as a result, the particles could not be completely layered. As amplitude increased, bed activity was enhanced and viscosity was weakened, and particles in the bed showed an apparent layering trend. Optimal separation effects (Soil = 1.79) were obtained at an amplitude of 2.5 mm. As the amplitude increased further, bed activity was strengthened and viscosity weakened; however, the dense medium particles were acted on by vibrating forces with large amplitude, causing significant back mixing. The buoyancy of the bed could not overcome forces, with significant back-mixing of layered particles resulting in poor separation. 3.4. Separation performance When oil shale of 100–6 mm was separated using the compound dry separator, yields of concentrate, middlings, and tailings

Table 5 Partition coefficient results of 6–0 mm oil shale for VADMFB separator. Density (cm3)

1.9 1.9–2.0 2.0–2.1 2.1–2.2 2.2–2.3 2.3–2.5 +2.5 Total

Average density g (cm3)

1.75 1.95 2.05 2.15 2.25 2.40 2.60

Sink-float results (accounting for the feed) (%) Tailings

Middlings

0.10 0.03 0.30 0.50 0.50 8.86 47.39

0.55 0.08 0.85 1.05 0.66 6.36 11.07

6.86 0.68 3.8 3.63 0.46 3.61 2.66

57.68

20.62

21.70

Calculated feedstock Sink-float results (%)

Calculated concentrate middlings results (%)

Concentrate 7.51 0.79 4.95 5.18 1.62 18.83 61.12 100

7.41 0.76 4.65 4.68 1.12 9.97 13.73 42.32

Partition coefficient High density separation (%)

Low density separation (%)

1.33 3.80 6.06 9.65 30.86 47.05 77.54

7.42 10.53 18.28 22.44 58.93 63.79 80.63

156

X. Yu et al. / Fuel 180 (2016) 148–156

middlings, and tailings were 20.49%, 22.08%, and 57.43%, with corresponding oil contents of 11.20%, 4.12%, and 1.50%, respectively. When oil shale of 6–0 mm was separated using a VADMFB separator, the yields of concentrate, middlings, and tailing were 21.7%, 20.62%, and 57.68%, with corresponding oil contents of 10.1%, 4.21%, and 0.98%, respectively. (4) Oil shale of 100–0 mm can be effectively and efficiently separated via a combined dry beneficiation process of compound dry cleaning and VADMFB separation.

Acknowledgements

Fig. 15. Partition curves for VADMFB separation.

were 20.49%, 22.08%, and 57.43%, with respective oil contents of 11.20%, 4.12%, and 1.50%, respectively. The partition coefficient results and partition curves are shown in Table 4 and Fig. 14, respectively. Following separation of oil shale of 6–0 mm by the VADMFB separator, yields of concentrate, middlings, and tailings were 21.70%, 20.62%, and 57.68%, respectively, with corresponding oil contents of 10.1%, 4.21%, and 0.98%. The partition coefficient results and partition curves are shown in Table 5 and Fig. 15, respectively. The cleaning oil shale (x > 10%) separated by dry combined process was carbonized at high temperatures (500 °C), the kerogen was decomposed to generate shale oil, which could be further refined into cheaper and higher quality gasoline and diesel than those refined oil shale (x < 10%). 4. Conclusions A combined dry beneficiation process, consisting of compound dry cleaning and VADMFB separation, was used to separate 100–0 mm oil shale. Effects of vibration frequency and amplitude on separation were investigated, with the following observations: (1) Pore structure characteristics of ore, concentrate, middlings, and tailings were analyzed by SEM; results showed that oil contents were higher with high organic matter content and when the pore structure was well developed. (2) The oil content segregation degree (Soil) of compound dry cleaning first increased and then decreased with an increase in vibration frequency and amplitude. The maximum value of Soil was obtained with f and A values of 53 Hz and 3.6 mm, respectively. Oil content decreased from the feed end to the tailing end along the X-axis direction and increased from the feed end to the discharge end along the Y-axis direction. In the case of the VADMFB separator, Soil values gradually increased and then decreased with increasing frequency and amplitude. The optimal value of Soil was obtained with f and A values of 36 Hz and 2.5 mm, respectively. (3) When oil shale of 100–6 mm was separated using a compound dry cleaning apparatus, the yields of concentrate,

The financial support by the National Natural Science Foundation (51404244), the Natural Science Foundation of Jiangsu Province (BK20140209), Jiangsu Province postdoctoral research funding schemes (1501058A) and A Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions is gratefully acknowledged. References [1] Hughes JD. Energy: a reality check on the shale revolution. Nature 2013;494 (7437):307–8. [2] Jiang XM, Han XX, Cui ZG. New technology for the comprehensive utilization of Chinese oil shale resources. Energy 2007;32:772–7. [3] Qian JL, Wang JQ, Li SY. World oil shale utilization and its future. J Jilin University (Earth Sci Ed) 2006;36:877–87 [in Chinese]. [4] Na JG, Im CH, Chung SH, Lee KB. Effect of oil shale retorting temperature on shale oil yield and properties. Fuel 2012;95:131–5. [5] Dyni JR. Geology and resources of some world oil shale deposits. Oil Shale 2003;20:193–252. [6] Lai DG, Chen ZH, Lin LX, Zhang YM, Gao SQ, Xu GW. Secondary cracking and upgrading of shale oil from pyrolyzing oil shale over shale ash. Energy Fuel 2015;29:2219–26. [7] Oja V, Elenurm A, Rohtla I, Tali E, Tearo E, Yanchilin A. Comparison of oil shales from different deposits: oil shale pyrolysis and co-pyrolysis with ash. Oil Shale 2007;24:101–8. [8] Wang S, Liu JX, Jiang XM, Han XX, Tong JH. Effect of heating rate on products yield and characteristics of non-condensable gases and shale oil obtained by retorting Dachengzi oil shale. Oil Shale 2013;30:27–47. [9] Subasinghe ND, Awaja F, Bhargava SK. Variation of kerogen content and mineralogy in some australian tertiary oil shales. Fuel 2009;88:335–9. [10] Niu M, Wang S, Han XX, Jiang XM. Yield and characteristics of shale oil from the retorting of oil shale and fine oil-shale ash mixtures. Appl Energy 2013; 111:234–9. [11] Shen LJ. Study of the law governing the separation of compound dry separator. Xuzhou: China University of Mining and Technology; 1996. [12] Yang X, Zhao Y, Luo Z, Song S, Duan C, Dong L. Fine coal dry cleaning using a vibrated gas-fluidized bed. Fuel Process Technol 2013;106:338–43. [13] Yang X, Zhao Y, Luo Z, Song S, Chen Z. Fine coal dry beneficiation using autogenous medium in a vibrated fluidized bed. Int J Miner Process 2013; 125:86–91. [14] Wang Q, Xu F, Bo J. Study on the basic physicochemical characteristics of the Huadian oil shale. J Jilin University (Earth Sci Ed) 2006;36:1006–11 [in Chinese]. [15] Li Q. Simulation of temperature field and experiment of in-situ oil shale pyrolysis. Changchun: Jilin University; 2012. [16] Wu WC. Study on compound dry separation characteristics. Xuzhou: China University of Mining and Technology; 2010. [17] Luo Z, Fan M, Zhao Y, Tao X, Chen Q, Chen Z. Density-dependent separation of dry fine coal in a vibrated fluidized bed. Powder Technol 2008;187:119–23. [18] Luo Z, Zuo W, Tang L, Zhao Y, Fan M. Preparation of solid medium for use in separation with gas-solid fluidized beds. Int J Min Sci Technol 2010;20:743–6. [19] Lv B, Luo Z, Zhang B, Zhao Y, Zhou C, Yuan W. Effect of the secondary air distribution layer on separation density in a dense-phase gas-solid fluidized bed. Int J Min Sci Technol 2015;25:969–73.