Journal of Environmental Management 133 (2014) 388e396
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
A numerical flow analysis using the concept of inflow age for oxidation pond design Dong-kil Lee*, Young-wook Cheong Korea Institute Geoscience and Mineral Resources (KIGAM), 92 Gwahang-no, Yuseong-gu, Daejeon 305-325, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 July 2012 Received in revised form 21 October 2013 Accepted 28 October 2013 Available online 13 January 2014
A numerical flow analysis for the design of an oxidation pond was conducted to investigate the optimal flow characteristics. This analysis includes the inflow rate and the shape and depth of the oxidation pond. The total area and maximum depth of the pond were 500 m2 and 3 m, respectively. We defined the retention time, retention time ratio, homogeneity index, and inflow exchange efficiency in order to choose the optimal conditions. The optimum width to length ratio and depth of the pond were found to be 1:5 and 2 m, respectively. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Oxidation pond Acid mine drainage Inflow age Retention time
1. Introduction After mining minerals, the exposed rocks come into contact with water and oxygen. Acid mine drainage (AMD) is the result of the oxidation of minerals that contain reduced forms of sulfides [pyrite or its polymorph, marcasite (both FeS2), pyrrhotite (FeS), galena (PbS), sphalerite (ZnS), and chalcopyrite (FeCuS2)], which commonly occur when the sulfide bearing minerals in the rocks are exposed to air and water, transforming the sulfides into sulfuric acid. This means that the greater the surface area of the rock that is exposed, the greater the amount of acid (Silva et al., 2006). The acidic water leaches out the surrounding heavy metals and so contaminates the AMD (Tiwary, 2001). AMD, along with these contaminants, has been recognized as a major environmental pollution problem over the past few decades (Letterman and Mitsch, 1978; Kleinmann et al., 1981; Gray, 1997). An oxidation pond stores the AMD for a given period of time, allowing for the precipitation of the ferric hydroxide found within it, so that the remediation of the contaminated effluents can be accomplished. The key to the design of an oxidation pond is to ensure that the retention time of the AMD in a pond is sufficiently long for the iron precipitates to settle out effectively. In the past, the retention time recommendations for such purposes have ranged from as little as 8 h to more than 72 h. Additional guidance on
* Corresponding author. Tel.: þ82 42 868 3228. E-mail address: [email protected] (D.-k. Lee). 0301-4797/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.10.026
retention time of mine water can be drawn from studies conducted in the United States (Lin et al., 2003; Keefe et al., 2004), in U.K. reducing and alkalinity producing systems (Wolkersdorfer et al., 2005), and modeling studies (Goebes and Younger, 2004). Studies have revealed the existence of a relatively robust linear relationship between the percentage reduction in the influent iron concentration and nominal hydraulic retention time (Parker, 2003). In the UK Coal industry, the basic water treatment design guidance for engineers was set out by National Coal Board (1982). The total pond volume and flow rate are often designed on the basis of a 48 h retention time (Laine and Jarvis, 2003) and 100 m2 pond surface area for every l/s of drainage. The length to width ratio is designated to be within the range of 2:1e5:1 (National Coal Board, 1982). The depth of the pond is usually set at around 3 m to prevent the re-suspension of the settled particles due to the effects of the wind (PRAMID Consortium, 2003). The theoretical maximum concentration of ferrous iron that can be oxidized in a single aeration cascade is 50 mg/l. However, practical experience suggests that 30 mg/l is a more realistic figure (National Coal Board, 1982). This implies that for discharges in excess of 30 mg/l, it will be necessary to have a series of aeration cascades, with oxidation ponds in between them. Until now, the nominal retention time was calculated simply as the ratio of the volume of AMD stored in the pond to the flow rate. The flow distribution characteristics in the pond vary with its shape and depth and with the flow rate of the AMD, so that more sophisticated techniques, such as the application of the physics represented by the NaviereStokes equation, are needed.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
389
2. Computational method 2.1. Concept of inflow age The concept of “air age” in the field of mine ventilation was introduced to estimate the retention time of the airflow in a mine drift. Air age is defined as the time taken for the inflow air to move to point P, as shown in Fig. 1. The term “retention time” is defined as the time it takes for air to flow completely through the system. The fluid of this study is not air but mine drainage, thus the term air age was renamed “inflow age”. 2.2. Suggestions for estimating the indices for an oxidation pond design To optimize the design of the oxidation pond, the retention time, stagnant level, distribution level, and exchange level for the AMD in the oxidation pond need to be defined. We propose to solve this problem by defining the following concepts: the retention time, retention time ratio, homogeneity index, and inflow exchange efficiency. Retention time The retention time, ta, is generally defined as the time needed for the inflow AMD from the inlet to reach the outlet and is defined in this study as the average inflow age at the exit, as determined by computational analysis. Retention time ratio, R The retention time ratio, R, is a guideline, showing how long the inflow AMD stays in the pond and is defined as the ratio of the nominal retention time, tn, to the retention time, ta. A retention time ratio closer to 1 indicates that the flow of the inflow AMD is more stagnant.
tn R ¼ 1 t
(1)
a
Homogeneity index, H The homogeneity index H is defined as the value of the minimum areal retention time, taeral, divided by the volume average retention time, tvol. A homogeneity index closer to 1 indicates that the distribution of the inflow age is more homogeneous.
H ¼
taeral tvol
(2)
Inflow exchange efficiency, Ei The inflow exchange efficiency, Ei, is defined as the shortest time it takes for the pond’s contents to be exchanged with the fresh inflow AMD and is further defined as the ratio of the nominal retention time, tn, to the inflow exchange time, te.
Ei ¼
tn 100 ð%Þ te
(3)
2.3. The oxidation pond design cases Fig. 2 shows a schematic of a rectangular oxidation pond. The inlet and outlet size of the oxidation pond was set to 0.3 m 0.3 m in all cases with an area of 500 m2. The flow rate at the inlet was based on a pond surface area of 100 m2 for every l/s of AMD, as defined by National Coal Board (1982). Table 1 shows 19 cases with different flow rates, shapes, and depths of the oxidation ponds.
Fig. 1. The concept of air age.
Fig. 2. Schematic of oxidation pond.
2.4. Grid constructions In the case where both sides of the pond are symmetrical with the same shape and flow conditions, the analysis model was made for one side only. The meshes were denser near the inlet and outlet and the grid sizes were increased gradually as the model approached the middle of the pond. The total number of grids in selected cases was up to 1.2 million (Fig. 3). 2.5. The numerical analysis method Computational analyses were performed for the three governing equations: the continuity equation, the momentum equation, and the turbulent dissipation equation. These are used to predict Table 1 Cases of oxidation pond. Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Width:length
Flow rate (l/s)
Width (m)
Length (m)
Height (m)
1:3 1:3 1:3 1:3 1:3 1:2 1:4 1:5 Circle 1:2 1:3 1:4 1:5 Circle 1:2 1:3 1:4 1:5 Circle
1 5 10 15 20 5 5 5 5 5 5 5 5 5 5 5 5 5 5
12.90 12.90 12.90 12.90 12.90 15.81 11.18 10.00 25.23 15.81 12.90 11.18 10.00 25.23 15.81 12.90 11.18 10.00 25.23
38.73 38.73 38.73 38.73 38.73 31.62 44.72 50.00 e 31.62 38.73 44.72 50.00 e 31.62 38.73 44.72 50.00 e
3 3 3 3 3 3 3 3 3 2 2 2 2 2 1 1 1 1 1
390
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
the flow distribution, the retention time and the estimating indices in a three dimensional oxidation pond. In this study, the flow rate in the oxidation pond was very low, but we still used the ke3 turbulent model in order to take sudden changes in the velocity at the inflow part into consideration. To identify the flow characteristics in the oxidation pond, a commercial computational fluid dynamics program, FLUENT, using the finite volume method was used. The discretization method was used with a 2nd order upwind scheme. The Semi-Implicit Method for Pressure Linked Equation (SIMPLE) algorithm was used to find the solutions. The flow system was analyzed under steady-state conditions, and the inlet and outlet were set to constant flow rate conditions and constant atmospheric pressure conditions, respectively. The inflow age was calculated using a user defined scalar (UDS) to identify the distribution of the inflow age. 3. Results and discussion 3.1. Flow characteristics depending on flow rate
Fig. 3. Constructed grids of oxidation pond.
Fig. 4 shows the distribution of the velocity and inflow age in the pond depending on inflow rate of the AMD. In line with the increase in the flow rate, the velocity distribution was extended along the line from the inlet to the outlet. When it comes to the distribution of the inflow age, it appeared to be the lowest in the straight section linking the inlet to the outlet, while it was higher on the left and right sides of the pond. This flow pattern was attributed to the main flow of the AMD along the route between the inlet and outlet and the stagnant area caused by the big eddy, which was formed on both sides of the main flow, due to the high velocity at the center line of the pond. Thus, the AMD flows along a specific zone, not the whole area of the oxidation pond, and the flow area is categorized into the main flow area and stagnant area. In this study, the flow area linking the inlet and outlet is named the ‘main flow area’ and the area without flow is dubbed the ‘stagnant area’.
Fig. 4. Distributions of velocity and inflow age according to flow rate in oxidation ponds.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 7. Variation of retention time ratio as a function of flow rate.
Fig. 5. Variation of average retention time as a function of flow rate.
Fig. 5 shows the average retention time depending on the flow rate in cases 1e5 which adopted a depth of 3 m. As can be seen in this figure, the flow rate, Q, and average retention time, ta, are in inverse proportion to each other, as represented by Equation (4). The coefficient of determination was 1.0 and B and C are constants. The retention time can be predicted using this equation:
ta ¼
B C ðhÞ Q
(4)
Fig. 6 shows the relationship between the flow rate and volumetric average velocity. As indicated in this figure, the volumetric average velocity was in proportion to the flow rate, which is represented by Equation (5), in which the coefficient of determination was 1.0. Q (l/s) and u are the flow rate and volumetric average velocity, respectively, and the constant F has the same dimensions as the cross sectional area (m2).
Q ¼ F u
Fig. 6. Variation of volumetric average velocity as a function of flow rate.
391
(5)
Fig. 7 illustrates the retention time ratio as a function of the flow rate. As the flow rate increased, the retention time ratio slightly increased. Figs. 8 and 9 show the homogeneity index and inflow exchange efficiency depending on the flow rate. As can be seen in Fig. 8, the homogeneity index decreased till the flow rate reached 5 l/s and then remained constant. On the other hand, the inflow exchange efficiency increased as the inflow was increased from 5 l/s to 10 l/ s and remained constant thereafter. This variation was attributed to the fact that the homogeneity index decreases as the flow velocity is increased, because of the increase in the velocity gradient and, on the contrary, the time required for exchanging the AMD is reduced and, thus, the inflow exchange efficiency was increased. 3.2. Flow characteristics depending on shape of oxidation pond Fig. 10 shows the velocity distribution depending on the shape of the oxidation pond when the AMD flows into the oxidation pond with an area of 500 m2 at 5 l/s. As can be seen in this figure, when
Fig. 8. Variation of homogeneity index as a function of flow rate.
392
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 9. Variation of inflow exchange efficiency as a function of flow rate.
the area of the oxidation pond is fixed, the velocity contour to the outlet tends to be reduced as the length/width ratio increases, which is attributable to the decrease in the stagnant area in line with the increase in the width to length ratio, resulting in an increase in the main flow area. Fig. 11 shows the inflow age distribution depending on the shape of the oxidation pond when the AMD flows into the oxidation pond with an area of 500 m2 at 5 l/s. As can be seen in this figure, the stagnation area and inflow age in the pond decreased as the
length to width ratio increased, and the length of the stagnation area formed by the eddy on the left and right sides of the pond tended to decrease as the length to width ratio increased, except in the case where the depth was 3 m. In the region around the outlet, the low inflow age zone expanded on both sides due to the low velocity of the AMD. Thus, as the length to width ratio was increased, the main flow area expanded and the low inflow age zone gradually became uniform near the outlet. This analysis result has the same uniform distribution as the existing experimental result. Practical experience suggests that the best sedimentation performance is obtained when the ratio of the pond length to its width is between 3:1 and 5:1. Higher length to width ratios may result in “streaming” across the pond, while lower length to width ratios may promote short-circuiting (PRAMID Consortium, 2003). In contrast, the stagnation area was widely distributed because of the large eddy formed on both sides of the round-shaped oxidation pond. Fig. 12 shows the relations between the shape of the oxidation pond and the volumetric average velocity. As can be seen in this figure, in line with the increase in the length to width ratio and depth of the oxidation pond, the volumetric average velocity decreased, due to the increase in the volume of the main flow area, where the kinetic energy is increasingly distributed with increasing the length/width ratio. The volumetric average velocity was affected more by the depth than by the length/width ratio, as illustrated in the figure. The volumetric average velocity in the round ponds with depths of 1 m and 3 m deep was higher than those in the rectangular ponds, but in the case of the 2 m deep round pond, it appeared to be similar to that of the rectangular oxidation pond with a length to width ratio of 2:1. Fig. 13 shows the variation of the retention time depending on the shape of the oxidation pond. As can be seen in this figure, the
Fig. 10. Distributions of the velocity according to the shape and the depth of the oxidation ponds.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
393
Fig. 11. Distributions of the inflow age according to the shape and the depth of the oxidation ponds.
retention time remained the same, irrespective of the length to width ratio when the depth was fixed in the rectangular oxidation pond. A very similar retention time to that of the rectangular pond was also observed in the round pond. Thus, the length to width ratio in the rectangular pond was independent of the retention time.
Fig. 14 shows the relations between the retention time ratio and the shape of the oxidation pond. As can be seen in this figure, the larger the length to width ratio, the lower the retention time ratio. Particularly, the retention time ratio was very large in the case of
Fig. 12. Variation of volumetric average velocity as a function of shape of the ponds.
Fig. 13. Variation of average retention time as a function of shape of the ponds.
394
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 14. Variation of the retention time ratio as a function of the shape of the pond. Fig. 16. Variation of the inflow exchange efficiency as a function of the shape of the pond.
the 3 m deep pool. Therefore, the stagnation level of the AMD in the pond was further increased when its depth and length to width ratio were smaller. In case of round ponds, the retention time ratio has a different pattern on the depth compared with rectangular ponds. This is attributable to a round pond having a different geometry from a rectangular pond. Fig. 15 shows the relations between the homogeneity index and the shape of the pond. The homogeneity index was the lowest in the case of the pool with a rectangular shape and length to width ratio of 2 and, as the length to width ratio of the rectangular pool increased, the homogeneity index rapidly increased at all depths. Particularly, it was the largest at a depth of 2 m. The homogeneity indices of round ponds are very low for all pond depths. Fig. 16 shows the relations between the shape of the oxidation pond and inflow exchange efficiency. As can be seen in this figure, the larger the width to length ratio of the rectangular pond, the higher the inflow exchange efficiency.
Fig. 15. Variation of the homogeneity index as a function of the shape of the pond.
The flow distribution characteristics and inflow exchange efficiency were the lowest when the length to width ratio was 1:2 for all of the shapes studied and, when the ratio was 1:4, the flow distribution characteristics were relatively uniform with the highest inflow exchange efficiency. At a depth of 1 m, the lowest inflow exchange efficiency was observed, while the highest exchange efficiency was observed at a depth of 2 m. Such characteristics also appeared in the case of the round shaped oxidation pond. 3.3. Flow characteristics depending on depth of oxidation pond Figs. 10 and 11 show the distributions of velocity and inflow age depending on the depth of the oxidation pond, respectively, when the AMD flows into the oxidation pond with an area of 500 m2 at 5 l/s. As can be seen in these figures, as the depth of the pond decreased, the relatively high velocity region in the pond expanded and the inflow age of the pond decreased for all shapes.
Fig. 17. Variation of the volumetric average as a function of the depth of the pond.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 18. Variation of the average retention time velocity as a function of the depth of the pond.
In the case where the length to width ratio is 2, as the depth decreased, the inflow age in the stagnation area was reduced, but in the oxidation pond with a length to width ratio of 3, the inflow age in the stagnation area appeared to be the highest at a depth of 2 m. On the contrary, the inflow age was the lowest at a depth of 2 m in the oxidation pond with length to width ratios of 4 and 5. Fig. 17 shows the volumetric average velocity depending on the depth for different shapes of the oxidation pond. As indicated in this figure, the volumetric average velocity tended to decrease as the depth increased in all cases, which was attributed to the increase in the fluid volume of the pond in line with the increase in depth. However, as can be seen from the round shaped pond, the pattern of decrease in the volumetric average velocity varied depending on the shape of the pond. Fig. 18 shows the retention time depending on the depth for different shapes of the oxidation pond. As indicated in this figure, the retention time at the same depth in the round shaped and
Fig. 19. Variation of the retention time ratio as a function of the depth of the pond.
395
Fig. 20. Variation of the homogeneity index as a function of the depth of the ponds.
rectangular shaped ponds appeared to be almost the same, indicating that the retention time was in proportion to the depth, irrespective of the shape. The relation between the depth D and the average retention time, ta, in the rectangular oxidation pond with an area of 500 m2 is represented by Equation (6), of which the determination coefficient is 0.996. In Equation (6), G and I are constants and D is the depth of the AMD in the pond.
ta ¼ G D I ðhÞ
(6)
The time when the sediments in the pond need to be removed can be calculated by using Equation (6). For example, when the retention time was set to 50 h, the minimum depth was (50 I)/G (m). In other words, the retention time could be sustained until the AMD depth in the pond is larger than (50 I)/G (m). Fig. 19 shows the retention time ratio depending on the depth for different shapes of the oxidation pond. In the cases of the
Fig. 21. Variation of the inflow exchange efficiency as a function of the depth of the ponds.
396
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
rectangular oxidation ponds with length to width ratios of 4 and 5, the retention time ratios were almost the same, irrespective of the depth. But, in the case of the rectangular oxidation ponds with length to width ratios of 2 and 3, the retention time ratio tended to increase as the depth increased. However, as regards the pools with a round shape, the retention time ratio was large at depths of 1 m and 2 m and decreased at a depth of 3 m. Fig. 20 illustrates the homogeneity index depending on the depth for different shapes of the oxidation pond. In the case of the rectangular shape with a length to width ratio of 2, it showed a constant value, irrespective of the depth, but when the length to width ratio was in the range from 3 to 5, the homogeneity index was the largest at a depth of 2 m as the length to width ratio increased. Regarding the round-shaped pond, the homogeneity index was quite similar to that of the rectangular pond with a length to width ratio of 2. Fig. 21 shows the relation between the depth and inflow exchange efficiency for different shapes of the oxidation pond. In regard to the rectangular pond, the largest inflow exchange efficiency was observed at a depth of 2 m, indicating that the larger the length to width ratio, the higher the inflow exchange efficiency. Such a variation also appeared in the case of the round pond, which showed relatively higher inflow exchange efficiency than the rectangular pond with a length to width ratio of 2. Thus, based on the analysis of the retention time, retention time ratio, homogeneity index and inflow exchange efficiency, the optimal depth for a rectangular oxidation pond with an area of 500 m2 was found to be 2 m. 4. Conclusions 1) In order to estimate the retention time for an oxidation pond design, a computational analysis method was proposed using the concepts of the inflow age and retention time, along with the stagnant level, the distribution level and the exchange level of the mine drainage in the oxidation pond. This was accomplished by defining the concepts of the retention time, the retention time ratio, the homogeneity index and the inflow exchange efficiency. Through these concepts, a method of determining the optimum conditions needed for designing an oxidation pond was established. 2) The flow characteristics depending on the flow rate in the rectangular pond with an area of 500 m2, a width to length of 1:3 and a depth of 3 m were analyzed. It was confirmed that the optimum inflow rate is more than 10 l/s by analyzing the indices suggested for the oxidation pond design. 3) The retention time is constant with the same volume of the pond and independent of the shape of the oxidation pond. The
more elongated the pond, the lower the retention time, the more uniform the distribution of the inflow age and the higher the exchange efficiency. In regard to the round oxidation pond, it showed a more uniform distribution and higher exchange efficiency than the rectangular oxidation pond with a length to width ratio of 2, which appeared to be the most ineffective structure. 4) The relation between the retention time and depth in the rectangular oxidation pond was presented to help predict the optimal time for removing the deposit. As a result of the analysis of the retention time, retention time ratio, homogeneity index and inflow exchange efficiency, the optimum depth of the rectangular oxidation pond with an area of 500 m2 was found to be 2 m. Acknowledgments This research was supported by the Research Project of the Korea Institute of Geoscience and Mineral Resources (KIGAM) and Mine Reclamation Corporation (MIRECO) funded by the Ministry of Knowledge Economy of Korea. References Goebes, M., Younger, P.L., 2004. A simple analytical model for interpretation of tracer tests in two-domain subsurface flow systems. Mine Water Environ. 23, 138e143. Gray, N.F., 1997. Environmental impact and remediation of acid-mine drainage: a management problem. Environ. Geol. 30, 62e71. Keefe, S.H., Barber, L.B., Runkel, R.L., Ryan, J.N., McKnight, D.M., Wass, R.D., 2004. Conservative and reactive solute transport in constructed wetlands. Water Resour. Res. 40, W01201. Kleinmann, R.L.P., Crerar, D.A., Pacelli, R.R., 1981. Biogeochemistry of acid mine drainage and a method to control acid formation. J. Min. Eng. 33, 300e304. Laine, D.M., Jarvis, A.P., 2003. Engineering design aspects of passive in situ remediation of mining effluents. Land Contam. Reclam. 11, 113e125. Letterman, R.D., Mitsch, W.J., 1978. Impact of mine drainage on a mountain in Pennsylvania. Environ. Pollut. 17, 53e73. Lin, A., Debroux, J., Cunningham, J., Reinhard, M., 2003. Comparison of Rhodamine WT and bromide in the determination of hydraulic characteristics of constructed wetlands. Ecol. Eng. 20, 75e88. National Coal Board, 1982. Technical Management of Water in the Coal Mining Industry. Mining Department, London. Parker, K., 2003. Mine water management on a national scale e experiences from the coal authority. Land Contam. Reclam. 21, 181e190. PRAMID Consortium, 2003. Engineering Guidelines for the Passive Remediation of Acid and/or Metalliferous Mine Drainage and Similar Wastewaters. Newcastle. Silva, E.F., Patinha, C., Reis, P., Fonseca, E.C., Matos, J.X., Barrosinho, J., Oliveira, J.M.S., 2006. Interaction of acid mine drainage with waters and sediments at the Corona stream, Lousal mine. Environ. Geol. 50, 1001e1013. Tiwary, R.K., 2001. Environmental impact of coal mining on water regime and its management. Water Air Soil Pollut. 132, 185e199. Wolkersdorfer, C., Hasche, A., Gobel, A., Younger, P.L., 2005. Tracer test in the Bowden Close passive treatment system. Wiss. Mitt. 28, 87e92.
Contents lists available at ScienceDirect
Journal of Environmental Management journal homepage: www.elsevier.com/locate/jenvman
A numerical flow analysis using the concept of inflow age for oxidation pond design Dong-kil Lee*, Young-wook Cheong Korea Institute Geoscience and Mineral Resources (KIGAM), 92 Gwahang-no, Yuseong-gu, Daejeon 305-325, South Korea
a r t i c l e i n f o
a b s t r a c t
Article history: Received 24 July 2012 Received in revised form 21 October 2013 Accepted 28 October 2013 Available online 13 January 2014
A numerical flow analysis for the design of an oxidation pond was conducted to investigate the optimal flow characteristics. This analysis includes the inflow rate and the shape and depth of the oxidation pond. The total area and maximum depth of the pond were 500 m2 and 3 m, respectively. We defined the retention time, retention time ratio, homogeneity index, and inflow exchange efficiency in order to choose the optimal conditions. The optimum width to length ratio and depth of the pond were found to be 1:5 and 2 m, respectively. Ó 2013 Elsevier Ltd. All rights reserved.
Keywords: Oxidation pond Acid mine drainage Inflow age Retention time
1. Introduction After mining minerals, the exposed rocks come into contact with water and oxygen. Acid mine drainage (AMD) is the result of the oxidation of minerals that contain reduced forms of sulfides [pyrite or its polymorph, marcasite (both FeS2), pyrrhotite (FeS), galena (PbS), sphalerite (ZnS), and chalcopyrite (FeCuS2)], which commonly occur when the sulfide bearing minerals in the rocks are exposed to air and water, transforming the sulfides into sulfuric acid. This means that the greater the surface area of the rock that is exposed, the greater the amount of acid (Silva et al., 2006). The acidic water leaches out the surrounding heavy metals and so contaminates the AMD (Tiwary, 2001). AMD, along with these contaminants, has been recognized as a major environmental pollution problem over the past few decades (Letterman and Mitsch, 1978; Kleinmann et al., 1981; Gray, 1997). An oxidation pond stores the AMD for a given period of time, allowing for the precipitation of the ferric hydroxide found within it, so that the remediation of the contaminated effluents can be accomplished. The key to the design of an oxidation pond is to ensure that the retention time of the AMD in a pond is sufficiently long for the iron precipitates to settle out effectively. In the past, the retention time recommendations for such purposes have ranged from as little as 8 h to more than 72 h. Additional guidance on
* Corresponding author. Tel.: þ82 42 868 3228. E-mail address: [email protected] (D.-k. Lee). 0301-4797/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jenvman.2013.10.026
retention time of mine water can be drawn from studies conducted in the United States (Lin et al., 2003; Keefe et al., 2004), in U.K. reducing and alkalinity producing systems (Wolkersdorfer et al., 2005), and modeling studies (Goebes and Younger, 2004). Studies have revealed the existence of a relatively robust linear relationship between the percentage reduction in the influent iron concentration and nominal hydraulic retention time (Parker, 2003). In the UK Coal industry, the basic water treatment design guidance for engineers was set out by National Coal Board (1982). The total pond volume and flow rate are often designed on the basis of a 48 h retention time (Laine and Jarvis, 2003) and 100 m2 pond surface area for every l/s of drainage. The length to width ratio is designated to be within the range of 2:1e5:1 (National Coal Board, 1982). The depth of the pond is usually set at around 3 m to prevent the re-suspension of the settled particles due to the effects of the wind (PRAMID Consortium, 2003). The theoretical maximum concentration of ferrous iron that can be oxidized in a single aeration cascade is 50 mg/l. However, practical experience suggests that 30 mg/l is a more realistic figure (National Coal Board, 1982). This implies that for discharges in excess of 30 mg/l, it will be necessary to have a series of aeration cascades, with oxidation ponds in between them. Until now, the nominal retention time was calculated simply as the ratio of the volume of AMD stored in the pond to the flow rate. The flow distribution characteristics in the pond vary with its shape and depth and with the flow rate of the AMD, so that more sophisticated techniques, such as the application of the physics represented by the NaviereStokes equation, are needed.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
389
2. Computational method 2.1. Concept of inflow age The concept of “air age” in the field of mine ventilation was introduced to estimate the retention time of the airflow in a mine drift. Air age is defined as the time taken for the inflow air to move to point P, as shown in Fig. 1. The term “retention time” is defined as the time it takes for air to flow completely through the system. The fluid of this study is not air but mine drainage, thus the term air age was renamed “inflow age”. 2.2. Suggestions for estimating the indices for an oxidation pond design To optimize the design of the oxidation pond, the retention time, stagnant level, distribution level, and exchange level for the AMD in the oxidation pond need to be defined. We propose to solve this problem by defining the following concepts: the retention time, retention time ratio, homogeneity index, and inflow exchange efficiency. Retention time The retention time, ta, is generally defined as the time needed for the inflow AMD from the inlet to reach the outlet and is defined in this study as the average inflow age at the exit, as determined by computational analysis. Retention time ratio, R The retention time ratio, R, is a guideline, showing how long the inflow AMD stays in the pond and is defined as the ratio of the nominal retention time, tn, to the retention time, ta. A retention time ratio closer to 1 indicates that the flow of the inflow AMD is more stagnant.
tn R ¼ 1 t
(1)
a
Homogeneity index, H The homogeneity index H is defined as the value of the minimum areal retention time, taeral, divided by the volume average retention time, tvol. A homogeneity index closer to 1 indicates that the distribution of the inflow age is more homogeneous.
H ¼
taeral tvol
(2)
Inflow exchange efficiency, Ei The inflow exchange efficiency, Ei, is defined as the shortest time it takes for the pond’s contents to be exchanged with the fresh inflow AMD and is further defined as the ratio of the nominal retention time, tn, to the inflow exchange time, te.
Ei ¼
tn 100 ð%Þ te
(3)
2.3. The oxidation pond design cases Fig. 2 shows a schematic of a rectangular oxidation pond. The inlet and outlet size of the oxidation pond was set to 0.3 m 0.3 m in all cases with an area of 500 m2. The flow rate at the inlet was based on a pond surface area of 100 m2 for every l/s of AMD, as defined by National Coal Board (1982). Table 1 shows 19 cases with different flow rates, shapes, and depths of the oxidation ponds.
Fig. 1. The concept of air age.
Fig. 2. Schematic of oxidation pond.
2.4. Grid constructions In the case where both sides of the pond are symmetrical with the same shape and flow conditions, the analysis model was made for one side only. The meshes were denser near the inlet and outlet and the grid sizes were increased gradually as the model approached the middle of the pond. The total number of grids in selected cases was up to 1.2 million (Fig. 3). 2.5. The numerical analysis method Computational analyses were performed for the three governing equations: the continuity equation, the momentum equation, and the turbulent dissipation equation. These are used to predict Table 1 Cases of oxidation pond. Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case Case
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
Width:length
Flow rate (l/s)
Width (m)
Length (m)
Height (m)
1:3 1:3 1:3 1:3 1:3 1:2 1:4 1:5 Circle 1:2 1:3 1:4 1:5 Circle 1:2 1:3 1:4 1:5 Circle
1 5 10 15 20 5 5 5 5 5 5 5 5 5 5 5 5 5 5
12.90 12.90 12.90 12.90 12.90 15.81 11.18 10.00 25.23 15.81 12.90 11.18 10.00 25.23 15.81 12.90 11.18 10.00 25.23
38.73 38.73 38.73 38.73 38.73 31.62 44.72 50.00 e 31.62 38.73 44.72 50.00 e 31.62 38.73 44.72 50.00 e
3 3 3 3 3 3 3 3 3 2 2 2 2 2 1 1 1 1 1
390
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
the flow distribution, the retention time and the estimating indices in a three dimensional oxidation pond. In this study, the flow rate in the oxidation pond was very low, but we still used the ke3 turbulent model in order to take sudden changes in the velocity at the inflow part into consideration. To identify the flow characteristics in the oxidation pond, a commercial computational fluid dynamics program, FLUENT, using the finite volume method was used. The discretization method was used with a 2nd order upwind scheme. The Semi-Implicit Method for Pressure Linked Equation (SIMPLE) algorithm was used to find the solutions. The flow system was analyzed under steady-state conditions, and the inlet and outlet were set to constant flow rate conditions and constant atmospheric pressure conditions, respectively. The inflow age was calculated using a user defined scalar (UDS) to identify the distribution of the inflow age. 3. Results and discussion 3.1. Flow characteristics depending on flow rate
Fig. 3. Constructed grids of oxidation pond.
Fig. 4 shows the distribution of the velocity and inflow age in the pond depending on inflow rate of the AMD. In line with the increase in the flow rate, the velocity distribution was extended along the line from the inlet to the outlet. When it comes to the distribution of the inflow age, it appeared to be the lowest in the straight section linking the inlet to the outlet, while it was higher on the left and right sides of the pond. This flow pattern was attributed to the main flow of the AMD along the route between the inlet and outlet and the stagnant area caused by the big eddy, which was formed on both sides of the main flow, due to the high velocity at the center line of the pond. Thus, the AMD flows along a specific zone, not the whole area of the oxidation pond, and the flow area is categorized into the main flow area and stagnant area. In this study, the flow area linking the inlet and outlet is named the ‘main flow area’ and the area without flow is dubbed the ‘stagnant area’.
Fig. 4. Distributions of velocity and inflow age according to flow rate in oxidation ponds.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 7. Variation of retention time ratio as a function of flow rate.
Fig. 5. Variation of average retention time as a function of flow rate.
Fig. 5 shows the average retention time depending on the flow rate in cases 1e5 which adopted a depth of 3 m. As can be seen in this figure, the flow rate, Q, and average retention time, ta, are in inverse proportion to each other, as represented by Equation (4). The coefficient of determination was 1.0 and B and C are constants. The retention time can be predicted using this equation:
ta ¼
B C ðhÞ Q
(4)
Fig. 6 shows the relationship between the flow rate and volumetric average velocity. As indicated in this figure, the volumetric average velocity was in proportion to the flow rate, which is represented by Equation (5), in which the coefficient of determination was 1.0. Q (l/s) and u are the flow rate and volumetric average velocity, respectively, and the constant F has the same dimensions as the cross sectional area (m2).
Q ¼ F u
Fig. 6. Variation of volumetric average velocity as a function of flow rate.
391
(5)
Fig. 7 illustrates the retention time ratio as a function of the flow rate. As the flow rate increased, the retention time ratio slightly increased. Figs. 8 and 9 show the homogeneity index and inflow exchange efficiency depending on the flow rate. As can be seen in Fig. 8, the homogeneity index decreased till the flow rate reached 5 l/s and then remained constant. On the other hand, the inflow exchange efficiency increased as the inflow was increased from 5 l/s to 10 l/ s and remained constant thereafter. This variation was attributed to the fact that the homogeneity index decreases as the flow velocity is increased, because of the increase in the velocity gradient and, on the contrary, the time required for exchanging the AMD is reduced and, thus, the inflow exchange efficiency was increased. 3.2. Flow characteristics depending on shape of oxidation pond Fig. 10 shows the velocity distribution depending on the shape of the oxidation pond when the AMD flows into the oxidation pond with an area of 500 m2 at 5 l/s. As can be seen in this figure, when
Fig. 8. Variation of homogeneity index as a function of flow rate.
392
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 9. Variation of inflow exchange efficiency as a function of flow rate.
the area of the oxidation pond is fixed, the velocity contour to the outlet tends to be reduced as the length/width ratio increases, which is attributable to the decrease in the stagnant area in line with the increase in the width to length ratio, resulting in an increase in the main flow area. Fig. 11 shows the inflow age distribution depending on the shape of the oxidation pond when the AMD flows into the oxidation pond with an area of 500 m2 at 5 l/s. As can be seen in this figure, the stagnation area and inflow age in the pond decreased as the
length to width ratio increased, and the length of the stagnation area formed by the eddy on the left and right sides of the pond tended to decrease as the length to width ratio increased, except in the case where the depth was 3 m. In the region around the outlet, the low inflow age zone expanded on both sides due to the low velocity of the AMD. Thus, as the length to width ratio was increased, the main flow area expanded and the low inflow age zone gradually became uniform near the outlet. This analysis result has the same uniform distribution as the existing experimental result. Practical experience suggests that the best sedimentation performance is obtained when the ratio of the pond length to its width is between 3:1 and 5:1. Higher length to width ratios may result in “streaming” across the pond, while lower length to width ratios may promote short-circuiting (PRAMID Consortium, 2003). In contrast, the stagnation area was widely distributed because of the large eddy formed on both sides of the round-shaped oxidation pond. Fig. 12 shows the relations between the shape of the oxidation pond and the volumetric average velocity. As can be seen in this figure, in line with the increase in the length to width ratio and depth of the oxidation pond, the volumetric average velocity decreased, due to the increase in the volume of the main flow area, where the kinetic energy is increasingly distributed with increasing the length/width ratio. The volumetric average velocity was affected more by the depth than by the length/width ratio, as illustrated in the figure. The volumetric average velocity in the round ponds with depths of 1 m and 3 m deep was higher than those in the rectangular ponds, but in the case of the 2 m deep round pond, it appeared to be similar to that of the rectangular oxidation pond with a length to width ratio of 2:1. Fig. 13 shows the variation of the retention time depending on the shape of the oxidation pond. As can be seen in this figure, the
Fig. 10. Distributions of the velocity according to the shape and the depth of the oxidation ponds.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
393
Fig. 11. Distributions of the inflow age according to the shape and the depth of the oxidation ponds.
retention time remained the same, irrespective of the length to width ratio when the depth was fixed in the rectangular oxidation pond. A very similar retention time to that of the rectangular pond was also observed in the round pond. Thus, the length to width ratio in the rectangular pond was independent of the retention time.
Fig. 14 shows the relations between the retention time ratio and the shape of the oxidation pond. As can be seen in this figure, the larger the length to width ratio, the lower the retention time ratio. Particularly, the retention time ratio was very large in the case of
Fig. 12. Variation of volumetric average velocity as a function of shape of the ponds.
Fig. 13. Variation of average retention time as a function of shape of the ponds.
394
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 14. Variation of the retention time ratio as a function of the shape of the pond. Fig. 16. Variation of the inflow exchange efficiency as a function of the shape of the pond.
the 3 m deep pool. Therefore, the stagnation level of the AMD in the pond was further increased when its depth and length to width ratio were smaller. In case of round ponds, the retention time ratio has a different pattern on the depth compared with rectangular ponds. This is attributable to a round pond having a different geometry from a rectangular pond. Fig. 15 shows the relations between the homogeneity index and the shape of the pond. The homogeneity index was the lowest in the case of the pool with a rectangular shape and length to width ratio of 2 and, as the length to width ratio of the rectangular pool increased, the homogeneity index rapidly increased at all depths. Particularly, it was the largest at a depth of 2 m. The homogeneity indices of round ponds are very low for all pond depths. Fig. 16 shows the relations between the shape of the oxidation pond and inflow exchange efficiency. As can be seen in this figure, the larger the width to length ratio of the rectangular pond, the higher the inflow exchange efficiency.
Fig. 15. Variation of the homogeneity index as a function of the shape of the pond.
The flow distribution characteristics and inflow exchange efficiency were the lowest when the length to width ratio was 1:2 for all of the shapes studied and, when the ratio was 1:4, the flow distribution characteristics were relatively uniform with the highest inflow exchange efficiency. At a depth of 1 m, the lowest inflow exchange efficiency was observed, while the highest exchange efficiency was observed at a depth of 2 m. Such characteristics also appeared in the case of the round shaped oxidation pond. 3.3. Flow characteristics depending on depth of oxidation pond Figs. 10 and 11 show the distributions of velocity and inflow age depending on the depth of the oxidation pond, respectively, when the AMD flows into the oxidation pond with an area of 500 m2 at 5 l/s. As can be seen in these figures, as the depth of the pond decreased, the relatively high velocity region in the pond expanded and the inflow age of the pond decreased for all shapes.
Fig. 17. Variation of the volumetric average as a function of the depth of the pond.
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
Fig. 18. Variation of the average retention time velocity as a function of the depth of the pond.
In the case where the length to width ratio is 2, as the depth decreased, the inflow age in the stagnation area was reduced, but in the oxidation pond with a length to width ratio of 3, the inflow age in the stagnation area appeared to be the highest at a depth of 2 m. On the contrary, the inflow age was the lowest at a depth of 2 m in the oxidation pond with length to width ratios of 4 and 5. Fig. 17 shows the volumetric average velocity depending on the depth for different shapes of the oxidation pond. As indicated in this figure, the volumetric average velocity tended to decrease as the depth increased in all cases, which was attributed to the increase in the fluid volume of the pond in line with the increase in depth. However, as can be seen from the round shaped pond, the pattern of decrease in the volumetric average velocity varied depending on the shape of the pond. Fig. 18 shows the retention time depending on the depth for different shapes of the oxidation pond. As indicated in this figure, the retention time at the same depth in the round shaped and
Fig. 19. Variation of the retention time ratio as a function of the depth of the pond.
395
Fig. 20. Variation of the homogeneity index as a function of the depth of the ponds.
rectangular shaped ponds appeared to be almost the same, indicating that the retention time was in proportion to the depth, irrespective of the shape. The relation between the depth D and the average retention time, ta, in the rectangular oxidation pond with an area of 500 m2 is represented by Equation (6), of which the determination coefficient is 0.996. In Equation (6), G and I are constants and D is the depth of the AMD in the pond.
ta ¼ G D I ðhÞ
(6)
The time when the sediments in the pond need to be removed can be calculated by using Equation (6). For example, when the retention time was set to 50 h, the minimum depth was (50 I)/G (m). In other words, the retention time could be sustained until the AMD depth in the pond is larger than (50 I)/G (m). Fig. 19 shows the retention time ratio depending on the depth for different shapes of the oxidation pond. In the cases of the
Fig. 21. Variation of the inflow exchange efficiency as a function of the depth of the ponds.
396
D. Lee, Y.-w. Cheong / Journal of Environmental Management 133 (2014) 388e396
rectangular oxidation ponds with length to width ratios of 4 and 5, the retention time ratios were almost the same, irrespective of the depth. But, in the case of the rectangular oxidation ponds with length to width ratios of 2 and 3, the retention time ratio tended to increase as the depth increased. However, as regards the pools with a round shape, the retention time ratio was large at depths of 1 m and 2 m and decreased at a depth of 3 m. Fig. 20 illustrates the homogeneity index depending on the depth for different shapes of the oxidation pond. In the case of the rectangular shape with a length to width ratio of 2, it showed a constant value, irrespective of the depth, but when the length to width ratio was in the range from 3 to 5, the homogeneity index was the largest at a depth of 2 m as the length to width ratio increased. Regarding the round-shaped pond, the homogeneity index was quite similar to that of the rectangular pond with a length to width ratio of 2. Fig. 21 shows the relation between the depth and inflow exchange efficiency for different shapes of the oxidation pond. In regard to the rectangular pond, the largest inflow exchange efficiency was observed at a depth of 2 m, indicating that the larger the length to width ratio, the higher the inflow exchange efficiency. Such a variation also appeared in the case of the round pond, which showed relatively higher inflow exchange efficiency than the rectangular pond with a length to width ratio of 2. Thus, based on the analysis of the retention time, retention time ratio, homogeneity index and inflow exchange efficiency, the optimal depth for a rectangular oxidation pond with an area of 500 m2 was found to be 2 m. 4. Conclusions 1) In order to estimate the retention time for an oxidation pond design, a computational analysis method was proposed using the concepts of the inflow age and retention time, along with the stagnant level, the distribution level and the exchange level of the mine drainage in the oxidation pond. This was accomplished by defining the concepts of the retention time, the retention time ratio, the homogeneity index and the inflow exchange efficiency. Through these concepts, a method of determining the optimum conditions needed for designing an oxidation pond was established. 2) The flow characteristics depending on the flow rate in the rectangular pond with an area of 500 m2, a width to length of 1:3 and a depth of 3 m were analyzed. It was confirmed that the optimum inflow rate is more than 10 l/s by analyzing the indices suggested for the oxidation pond design. 3) The retention time is constant with the same volume of the pond and independent of the shape of the oxidation pond. The
more elongated the pond, the lower the retention time, the more uniform the distribution of the inflow age and the higher the exchange efficiency. In regard to the round oxidation pond, it showed a more uniform distribution and higher exchange efficiency than the rectangular oxidation pond with a length to width ratio of 2, which appeared to be the most ineffective structure. 4) The relation between the retention time and depth in the rectangular oxidation pond was presented to help predict the optimal time for removing the deposit. As a result of the analysis of the retention time, retention time ratio, homogeneity index and inflow exchange efficiency, the optimum depth of the rectangular oxidation pond with an area of 500 m2 was found to be 2 m. Acknowledgments This research was supported by the Research Project of the Korea Institute of Geoscience and Mineral Resources (KIGAM) and Mine Reclamation Corporation (MIRECO) funded by the Ministry of Knowledge Economy of Korea. References Goebes, M., Younger, P.L., 2004. A simple analytical model for interpretation of tracer tests in two-domain subsurface flow systems. Mine Water Environ. 23, 138e143. Gray, N.F., 1997. Environmental impact and remediation of acid-mine drainage: a management problem. Environ. Geol. 30, 62e71. Keefe, S.H., Barber, L.B., Runkel, R.L., Ryan, J.N., McKnight, D.M., Wass, R.D., 2004. Conservative and reactive solute transport in constructed wetlands. Water Resour. Res. 40, W01201. Kleinmann, R.L.P., Crerar, D.A., Pacelli, R.R., 1981. Biogeochemistry of acid mine drainage and a method to control acid formation. J. Min. Eng. 33, 300e304. Laine, D.M., Jarvis, A.P., 2003. Engineering design aspects of passive in situ remediation of mining effluents. Land Contam. Reclam. 11, 113e125. Letterman, R.D., Mitsch, W.J., 1978. Impact of mine drainage on a mountain in Pennsylvania. Environ. Pollut. 17, 53e73. Lin, A., Debroux, J., Cunningham, J., Reinhard, M., 2003. Comparison of Rhodamine WT and bromide in the determination of hydraulic characteristics of constructed wetlands. Ecol. Eng. 20, 75e88. National Coal Board, 1982. Technical Management of Water in the Coal Mining Industry. Mining Department, London. Parker, K., 2003. Mine water management on a national scale e experiences from the coal authority. Land Contam. Reclam. 21, 181e190. PRAMID Consortium, 2003. Engineering Guidelines for the Passive Remediation of Acid and/or Metalliferous Mine Drainage and Similar Wastewaters. Newcastle. Silva, E.F., Patinha, C., Reis, P., Fonseca, E.C., Matos, J.X., Barrosinho, J., Oliveira, J.M.S., 2006. Interaction of acid mine drainage with waters and sediments at the Corona stream, Lousal mine. Environ. Geol. 50, 1001e1013. Tiwary, R.K., 2001. Environmental impact of coal mining on water regime and its management. Water Air Soil Pollut. 132, 185e199. Wolkersdorfer, C., Hasche, A., Gobel, A., Younger, P.L., 2005. Tracer test in the Bowden Close passive treatment system. Wiss. Mitt. 28, 87e92.
