Waste Management 34 (2014) 1667–1673
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Transient design of landfill liquid addition systems Pradeep Jain a,⇑, Timothy G. Townsend b,1, Thabet M. Tolaymat c,2 a
Innovative Waste Consulting Services, LLC, 6628 NW 9th Blvd. Suite 3, Gainesville, FL 32605, USA Department of Environmental Engineering Sciences, University of Florida, P.O. Box 116450, Gainesville, FL 32611-6450, USA c United States Department of Environmental Protection, National Risk Management Research Laboratory, 26 W. Martin Luther King St., Cincinnati, OH 45268, USA b
a r t i c l e
i n f o
Article history: Received 24 October 2013 Accepted 6 May 2014 Available online 25 June 2014 Keywords: Vertical well Horizontal trench Bioreactor landfill Leachate recirculation Liquids addition
a b s t r a c t This study presents the development of design charts that can be used to estimate lateral and vertical spacing of liquids addition devices (e.g., vertical well, horizontal trenches) and the operating duration needed for transient operating conditions (conditions until steady-state operating conditions are achieved). These design charts should be used in conjunction with steady-state design charts published earlier by Jain et al. (2010a, 2010b). The data suggest that the liquids addition system operating time can be significantly reduced by utilizing moderately closer spacing between liquids addition devices than the spacing needed for steady-state conditions. These design charts can be used by designers to readily estimate achievable flow rate and lateral and vertical extents of the zone of impact from liquid addition devices, and analyze the sensitivity of various input variables (e.g., hydraulic conductivity, anisotropy, well radius, screen length) to the design. The applicability of the design charts, which are developed based on simulations of a continuously operated system, was also evaluated for the design of a system that would be operated intermittently (e.g., systems only operated during facility operating hours). The design charts somewhat underestimates the flow rate achieved and overestimates the lateral extent of the zone of impact over an operating duration for an intermittently operated system. The associated estimation errors would be smaller than the margin of errors associated with measurement of other key design inputs such as waste properties (e.g., hydraulic conductivity) and wider variation of these properties at a given site due to heterogeneous nature of waste. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Liquids (leachate, groundwater, other liquids source such as industrial liquids and biosolids) addition to the conventional municipal solid waste (MSW) landfill (also commonly referred as a dry tomb landfill) is the most common approach to bioreactor operation. Vertical wells and horizontal sources (trenches, infiltration galleries) are frequently used to add liquids to landfilled waste. Like any other landfill components, (e.g., leachate collection system, gas collection system) the liquids addition system (also commonly referred as leachate recirculation system or liquids introduction system) needs to be designed for permitting, construction, and operation of bioreactor landfills. Based on the authors’ design and review experience, a common approach for ⇑ Corresponding author. Tel.: +1 352 2834742, +1 352 3314828; fax: +1 352 3314842. E-mail addresses: [email protected] (P. Jain), ttown@ufl.edu (T.G. Townsend), [email protected] (T.M. Tolaymat). 1 Tel.: +1 352 3920846; fax: +1 352 3923076. 2 Tel.: +1 513 4872860; fax: +1 513 5697879. http://dx.doi.org/10.1016/j.wasman.2014.05.008 0956-053X/Ó 2014 Elsevier Ltd. All rights reserved.
design of a liquids addition system is design for steady state conditions. Jain et al. (2010a, 2010b) presented charts a designer can use for steady-state design of a vertical well and horizontal source system for liquids introduction. These charts can be used to estimate the achievable flow rate, lateral and vertical extents of the zone of impact, and the associated total liquids volume and operating time based on device dimensions and waste properties (hydraulic conductivity, anisotropy, porosity) at steady state conditions. For a design based on steady-state conditions, liquids addition devices are spaced based on the maximum possible lateral extent of the zone of impact of a device, which requires the addition of the liquid volume needed to reach the steady state, which in turn dictates the operating duration to achieve the steady state. Conditions however, may exist such that the timeframe over which liquids can be added is shorter than necessary to reach steady state; the timeframes needed to reach steady state were almost 19 years for the design chart application examples presented in Jain et al. (2010a) for vertical well systems, which is significantly greater than the typical life of a cell at MSW landfills in the US. Therefore, there is a need for tools to design liquids addition systems for conditions
1668
P. Jain et al. / Waste Management 34 (2014) 1667–1673
Nomenclature
a aL g q h hr hs b0 a
C d k K Kr Kz l m, n mv p pI Qv
Van Genuchten parameter related to air entry pressure fluid compressibility (M1 L T2) dimensionless variable density of fluid (M L3) moisture content (dimensionless) residual moisture content (dimensionless) porosity of the media (dimensionless) constant equals to 1 when p P 0 and 0 when p 10 −4
0.4
0.4
0.2
0.2
η ≤ 10 −4
0.0
0.0 0.0
0.2
0.4
0.6
0.8
1.0
(a)
0.8
r /r
I Is
0.6
Decreasing η and pI 0.4
0.2
0.0 0.2
0.4
0.6
0.8
1.0
Vt/Vt,critical
(b) Fig. 2. Design Chart 1. Fraction of steady-state lateral extent as a function of fraction of steady-state pore volume for (a) horizontal source, and (b) vertical well.
No significant trend in the xI/xIs (or rI/rIs) ratio variation with Vt/Vt,critical ratio was observed with respect to g and injection pressure; as defined in Jain et al. (2010a, 2010b) g is a function of source dimensions and waste anisotropy (a) as defined below.
w2 For vertical well : g ¼ a 2 rw For horizontal source : g ¼ a
w2 l
0.2
0.4
0.6
0.8
1.0
Fig. 3. Design Chart 2. Fraction of steady-state vertical extent as a function of fraction of steady-state pore volume for horizontal source.
1.0
0.0
0.0
V /V t t,critical
V /V t t,critical
2
In general, a lower fraction of the lateral extents of the impact zone was achieved for a lower g and injection pressure (pI) for a given fraction of liquids volume for both type of liquids addition devices (horizontal source, vertical well). These design charts can be used to estimate transient lateral extents of the impact zone as a function of liquids volume once the steady-state lateral extents and the associated liquids volume are estimated using the design charts presented by Jain et al. (2010a, 2010b). More details on range of g values of the simulations conducted and rational of the range are presented elsewhere (Jain et al. (2010a, 2010b)). 3.2. Design Chart 2. Transient vertical zone of impact Vertical spacing is typically specified for horizontal systems. Fig. 3 presents the fraction of the steady-state vertical extents of the zone of impact (below a horizontal source) achieved as a func-
tion of the fraction of the added liquids volume. A hypothetical line segregating the simulation results into two groups (g 6 104 and g > 104) is shown; data from simulations with g > 104 fell above the line, whereas data from simulations with g 6 104 fell below the line. No significant trend with respect to g and injection pressure (pI) was observed within a group. Approximately 40–55% of the vertical zone of impact can be achieved with the addition of 40% of liquids volume needed to achieve the steady-state vertical extent of the zone of impact. As shown, a majority of simulation data were above the line depicted on the plot; data from 20 out of 27 simulations were above the line. Based on the data presented in Figs. 2(a) and 3, it can be seen that the reduction in vertical spacing of devices is significantly greater than the horizontal spacing for transient design. For example, for the scenario where only 40% of steady-state liquid volume would be added through a device, the lateral spacing of the device should be 80–90% of the spacing for steady-state design (a reduction of only 10–20%), whereas, the vertical spacing should be 40–55% of the spacing for steady-state design (a reduction of 45–60%). 3.3. Design Chart 3. Operating duration Fig. 4(a) and (b) can be used to estimate the time period needed to add the target liquid volume to a horizontal source and vertical well, respectively, for scenarios where the design volume is less than needed to achieve the steady-state condition. For example, 40% of the volume needed to achieve steady state can be added to a horizontal system in 30–38% of the time needed to reach the steady state. This design chart can also be used for estimating the ratio of the mean to steady-state flow rate; the ratio of Vt/Vt,critical to t/ts is equivalent to the mean to steady-state flow rate. For example, approximately 20–30% of steady-state liquids volume can be added to a horizontal source, operated 20% of the time needed to reach steady-state conditions. The average flow rate would be 100–150% of the steady-state flow rate, dependent on g and injection pressure. The use of the steady-state flow rate for estimation of operating duration to add designed liquid volume may result in significant overestimation of the operating time. 3.4. Impact of intermittent operation In order to evaluate the applicability of the developed charts for the design or analysis of a system operated on an intermittent basis, an additional 4 simulations were conducted. Two were conducted for simulation of continuous operation (one with Kz = 105 cm/s and other with Kz = 106 cm/s). For the other two
1671
P. Jain et al. / Waste Management 34 (2014) 1667–1673 80
1.0
pI =5 l=w=1
Volume of moisture added per unit trench length (m 3 m-1)
-4
0.8
t/t s
0.6
0.4
0.2
K x=10 cm/s K z=10-5 cm/s
60
Intermittent operation (8 hrs on and 16 hrs off) 40 Continuous operation 20
K x =10-5 cm/s K z =10-6 cm/s
Decreasing η and pI
Intermittent operation (8 hrs on and 16 hrs off) Continuous operation
0
0.0 0.0
0.2
0.4
0.6
0.8
0
1.0
10
20
30
40
50
60
70
80
90
100
110
120
Operational time (days)
V /V t t,critical
(a)
Fig. 5. Cumulative volume of liquids added as a function of operational time for intermittent versus continuous operations.
1.0
t/ts
0.6
0.4
0.2
Decreasing η and pI 0.0 0.0
0.2
0.4
0.6
0.8
1.0
Vt/Vt,critical
(b) Fig. 4. Design Chart 3. Fraction of time to reach steady state as a function of fraction of pore volume to reach steady state for (a) horizontal source, and (b) vertical well.
the pressure-specified boundary condition at the source-waste interface was applied intermittently (cycled at 8 h on and 16 h off, one with Kz = 105 cm/s and other with Kz = 106 cm/s). Based on values reported in literature, a waste anisotropy (a) of 10 was used for these simulations (Landva et al., 1998; Singh et al., 2014); anisotropy is defined as the ratio of hydraulic conductivity in lateral direction to hydraulic conductivity in vertical direction. Simulations representing continuous injection scenarios were run for a duration of approximately 125 days (approximately 3000 h) while those representing the intermittent scenario were run for a duration of approximately 380 days (approximately 9000 h of total time or 3000 h of active injection/operating time). The added liquids volume, flow rate and lateral extents of liquids movement were compared for these cases. In general, the flow rate for the intermittent scenario was greater than the corresponding continuous scenario flow rate; the flow rate was zero for the 16-h shut down period for intermittent cases. This is attributed to the potential dissipation in fluid pressure around the source while the system is non-operational. The liquids move away from the source under the influence of gravity and capillarity and consequently reduce fluid pressure around the source when the system is shut down. Fig. 5 presents the cumulative volume of liquids added as a function of operation time of the system. The flow rate for continuous operation was approximately 33% lower than that for intermittent operation (based on operation time) at the end of the 125 operating days. Based on actual time (including the daily
16 h shut down period for intermittent operation), the average daily and cumulative flow rate was approximately half of that for continuous operation. The use of the design charts presented by Jain et al. (2010a, 2010b) would result in underestimation of steady-state flow rate based on operating time for intermittent operation and thus implicates a conservative design. The error associated with the use of the steady-state flow rates presented by Jain et al. (2010a, 2010b) for designing a system that would be operated intermittently could be treated as insignificant in comparison to the range and errors associated with assumptions of hydraulic conductivity, anisotropy, etc. Multiple measurements of hydraulic conductivity from a single landfill cell are reported to vary over an order of magnitude (Jain et al., 2006). Moreover, the calculation of flow rate based on dimensionless variable estimation from design charts would require site-specific measurement of hydraulic conductivity and anisotropy values of the waste at site under consideration. It should be noted that these systems are installed as landfilling progresses and thus need to be designed before the waste is in place. The site-specific hydraulic conductivity of the waste, therefore, could not be measured at the design stage, even if desired, under most circumstances except where a liquids addition system is designed for a landfill cell at final (or near final) grades. Fig. 6 presents the variation of the lateral extent of liquids movement for all of the 4 simulations described in this section as a function of liquids volume added. It can be seen that the lateral
12
pI=5 l=w=1 a=10
10
8
xI (m)
0.8
6
4 Intermittent Operation (Kz=10-5 cm/s) Continuous Operation (Kz=10-5 cm/s)
2
Intermittent Operation (Kz=10-6 cm/s) Continuous Operation (Kz=10-6 cm/s)
0 0
20
40
60
80
Volume of moisture added per unit trench length (m 3 m-1) Fig. 6. Variation of lateral extent of the zone of impact as a function of liquids volume added for intermittent and continuous operation.
1672
P. Jain et al. / Waste Management 34 (2014) 1667–1673
extents for both continuous and intermittent operation for Kz = 106 cm/s and continuous operation for Kz = 105 cm/s did not reach steady-state and compared very well with the values predicted based on the results presented earlier. The lateral extent for the simulation modeling intermittent operation at Kz = 105 cm/s reached steady-state and the steady-state value was almost 45% lower than the steady-state value predicted based on the results presented in previous section. It should be noted that steady-state lateral extents of the zone of impact is strongly correlated with flow rate (Fig. 5 in Jain et al. (2010a, 2010b)). As described earlier, the flow rate for intermittent operation based on operating time was 33% more and that based on the total time was 50% less than the steady-state flow rate estimated based on design charts presented by Jain et al. (2010a, 2010b). Since the flow rate in the intermittent scenario was not the same as for the continuous scenario modeled previously, two additional simulations were conducted to compare the lateral extents of the zone of impact between the continuous and intermittent operating strategies for a 1 m 1 m trench installed in waste with Kz = 106 cm/s and a = 10. Flux instead of injection pressure at the source-waste interface was specified as the boundary condition and the same liquids injection volume was employed for these simulations. A flux of 60 L (0.06 m3) per m trench length per day was used for these simulations. For the continuous scenario this volume was added over 24 h per day whereas for the modeled intermittent case the 60 L/m injection rate was added in 2 h and the system was shut down for 22 h to simulate the impact of an extreme end of intermittent operation. As expected, the injection pressure (during operating hours) needed to add the liquid volume over much shorter duration in intermittent scenario was significantly greater than that for the continuous injection scenario. The lateral extent for intermittent case was expected to be greater than the continuous injection scenario as the same volume of liquids is injected under much greater pressure. However, surprisingly, the lateral extent for the intermittent scenario was lower than that for the continuous injection scenario (Fig. 7). The results suggest that the duration of the applied pressure at the well has a greater impact on the lateral extent of the zone of impact than the magnitude of pressure for a given volume of liquids added. The liquids migrate away from the well laterally and vertically, due to the positive pressure at the source during operating hours, gravity and capillarity. While the system is shut down, liquids in the well vicinity predominantly migrate downwards under the influence of gravity and the lateral migration is slowed due to lower remnant pressure at the source. The
7
Kx=10-5 cm/s Kz=10-6 cm/s l=w=1 m
6
xI (m)
5
Continuous opreation
4
intermittent operation, therefore, results in greater vertical extent and lower lateral extent of the zone of impact than that of the continuous operation case. The magnitude of the error introduced by the use of design charts for estimation of the lateral extent of the scenario modeled was less than 10%, which is insignificant compared to the margins of error associated with measurement of the key waste properties such as hydraulic conductivity, which are a critical input to the liquids addition system design.
4. Limitations A constant waste permeability was assumed for all simulations. However, Kz is a function of overburden pressure, among other things such as liquids content and age of waste, and it is the expected to change with time and as landfilling progresses (Jain et al., 2006). These factors that impact Kz values are not well understood to date and were not considered in this study. Waste was modeled as a homogenous media, however, waste is highly heterogeneous in nature. In the absence of availability of reliable unsaturated media properties, unsaturated properties typical of sandy soil were used for modeling. Jain et al. (2010b) documented that unsaturated waste properties did not have a significant impact on steady-state results. The impact of landfill gas on liquid flow was not taken into consideration and waste was modeled as a single-phase flow system. Future studies should evaluate the impact of unsaturated media properties and landfill gas on the extents of the zone of impact, and collect and compare the actual measurements of the lateral and vertical extents of the zone of impact as a function of liquids volume added with the modeling results
5. Summary and conclusions Design charts were developed to estimate spacing of liquids addition devices under transient conditions. These design charts complement steady-state liquids addition system design procedure presented by Jain et al. (2010a, 2010b). The results suggest that by adding approximately 50% of the liquid volume needed to achieve the steady-state conditions, greater than 80% of the maximum lateral extents of the zone of impact can be achieved for both a horizontal source and vertical well. The results also suggest that design charts presented by Jain et al. (2010a, 2010b) underestimate the flow rate (on operating time basis) for intermittent operating conditions. The lateral extent of the zone of impact for a give volume of liquids added was found to be lower for intermittent operation even with greater operating injection pressure that for the continuous operation case. The design charts slightly overestimate the lateral extent of the zone of impact for an intermittent flow scenario for a given volume of liquids addition. Overall, the error associated with design of an intermittent operating system using charts provided by Jain et al. (2010a, 2010b) would be lower than the margin of error associated with measurement of the waste properties such as hydraulic conductivity that are key inputs in the liquid addition design procedure.
Intermittent operation (2 hrs on and 22 hrs off)
3
6. Disclaimer
2 1 0 0
5
10
15
20
25 3
30 -1
Volume of moisture added per unit trench length (m m ) Fig. 7. Variation of lateral extent of the zone of impact as a function of liquids volume added for intermittent and continuous operation for constant flow rate condition.
The US Environmental Protection Agency (US EPA) through its Office of Research and Development funded and managed the research described here under purchase order number: EP05C000550 to Innovative Waste Consulting Services, LLC. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. The scientific views expressed are solely those of the authors and do not necessarily reflect those of the US EPA.
P. Jain et al. / Waste Management 34 (2014) 1667–1673
Appendix A. Example problems Example 1 Leachate recirculation at a landfill is planned using 0.5 m 0.5 m horizontal trenches. The liquids injection would occur at 5 m water column (w.c.) measured at the trench-waste interface. The hydraulic conductivity of the waste in the horizontal and vertical directions are 107 m s1 and 108 m s1 (anisotropy = 10), respectively, and the drainable porosity is 0.4. The values of other waste parameters were assumed to be the same as the ones presented in Table 1. The design engineer wishes to estimate the (a) vertical and (b) lateral spacing between devices, and (c) operating duration for adding 40 m3 of liquids per m trench. The steady-state lateral and vertical zones, liquid volume added, and operating time to reach the steady-state conditions were estimated using design charts presented by Jain et al. (2010b). The estimated steady-state values are presented as follows: 1. Steady-state lateral extent of the zone of impact, xIs = 11 m. 2. Steady-state vertical extent of the zone of impact, zIs = 12.5 m. 3. Volume needed to achieve steady state, Vt,critical = 100 m3 per m trench length. 4. Time taken to reach steady state, ts = 9.8 years of continuous operation. (a) As only 40 m3 of liquid will be added per m trench, the system would not reach the steady-state conditions. The design charts presented in this paper are used to estimate the lateral and vertical zones of impact and operating time for the design liquids addition volume. The fraction of steady-state volume (Vt/Vt,critical) added in this case is 0.4. Design Chart 1 is used to estimate the lateral extent of the zone of impact. For Vt/Vt,critical = 0.4, xI/xIs ranges from 0.79 to 0.92. The lateral extent of the zone of impact is estimated to range from 8.7 to 10.1 m. Therefore, lateral trench spacing is estimated to range from 17.4 to 20.2 m. (b) Using Eq. (8) in Jain et al. (2010b), g is estimated as follows:
g¼a
w2 2
l
¼ 10
ð0:5 mÞ2 ð0:5 mÞ2
¼ 10
Using Design Chart 2, for Vt/Vt,critical = 0.4, zI/zIs ranges from 0.5 to 0.55. Using the value of zIs presented previously, the vertical extent of the zone of impact is estimated to range from 6.3 m to 6.9 m and vertical trench spacing is estimated to range from 12.6 to 13.8 m. (c) Using Design Chart 3, t/ts is estimated to range from 0.33 to 0.39 for Vt/Vt,critical = 0.4. Using ts = 9.8 years presented above, the operating duration needed to add 40 m3 of liquids per m trench length ranges from 3.2 to 3.8 years. Example 2 A leachate recirculation system in a 40-m deep landfill consists of 0.6-m diameter and 18-m deep vertical wells each with
1673
a 15-m screen. The pressure at the bottom of the well is 17 m w.c. The hydraulic conductivity of the waste in the horizontal and vertical direction is 107 m s1 and 108 m s1 (anisotropy = 10), respectively, and the drainable porosity is 0.4. The values of other waste parameters were assumed to be the same those presented in Table 1. The design engineer wishes to estimate the (a) well spacing and (b) operating duration for addition of 7500 m3 to a well. The steady-state lateral and vertical zones, liquid volume added, and operating time to reach steady-state conditions are estimated used design charts presented by Jain et al. (2010a). The estimated steady-state values are presented as follows: 1. Steady-state lateral extent of the zone of impact, rI = 28.5 m. 2. Volume needed to achieve steady state, Vt,critical = 15,260 m3. 3. Time taken to reach steady state, ts = 18.9 years of continuous operation. (a) As only 7500 m3 of liquid will be added to a well, the system would not reach the steady-state conditions within the operating period. The design charts presented in this paper are used to estimate the lateral zone of impact and operating time for the design liquids addition volume. The fraction of steady-state volume (Vt/Vt,critical) added in this case is 0.49. Design Chart 1 is used to estimate the lateral extent of the zone of impact. For Vt/Vt,critical = 0.49, rI/rIs ranges from 0.78 to 0.93. The lateral extent of the zone of impact is, therefore, estimated to range from 22.2 to 26.5 m. The well spacing is, therefore, estimated to range from 44 to 53 m. (b) Using Design Chart 3, t/ts is estimated to range from 0.4 to 0.46 for Vt/Vt,critical = 0.49. Using ts = 18.9 years presented above, the operating duration needed to add 7500 m3 of liquids per well ranges from 7.6 to 8.7 years.
References Jain, P., Powell, J., Townsend, T., Reinhart, D., 2006. Estimating the hydraulic conductivity of landfilled municipal solid waste using the borehole permeameter test. J. Environ. Eng., ASCE 132 (6), 645–652. Jain, P., Tolaymat, T., Townsend, T., 2010a. Steady state design of vertical wells for liquids addition at bioreactor landfills. Waste Manage. 30 (11), 2022–2029. Jain, P., Tolaymat, T., Townsend, T., 2010b. Steady state design of horizontal systems for liquids addition at bioreactor landfills. Waste Manage. 30 (12), 2560–2569. Krahn, J., 2004. Seepage Modeling with SEEP/W: An Engineering Methodology. GeoSlope International, Calgary, Canada. Landva, A.O., Pelkey, S.G., Valsangkar, A.J., 1998. Coefficient of permeability of municipal refuse. In: Proceedings of the Third International Congress on Environmental Geotechnics, Lisbon, Portugal, pp. 163–167. Singh, K., Kadambala, R., Jain, P., Xu, Q., Townsend, T., 2014. Anisotropy estimation of compacted municipal solid waste using pressurized vertical well liquids injection. Waste Manage. Res.. http://dx.doi.org/10.1177/0734242X14532003. Stephens, D.B., 1995. Vadose Zone Hydrology. Lewis Publishers, Boca Raton, FL, USA. Tchobanoglous, G., Theisen, H., Vigil, S., 1993. Integrated Solid Waste Management. McGraw-Hill, New York, USA. Van Genuchten, M.Th., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892–898.
Contents lists available at ScienceDirect
Waste Management journal homepage: www.elsevier.com/locate/wasman
Transient design of landfill liquid addition systems Pradeep Jain a,⇑, Timothy G. Townsend b,1, Thabet M. Tolaymat c,2 a
Innovative Waste Consulting Services, LLC, 6628 NW 9th Blvd. Suite 3, Gainesville, FL 32605, USA Department of Environmental Engineering Sciences, University of Florida, P.O. Box 116450, Gainesville, FL 32611-6450, USA c United States Department of Environmental Protection, National Risk Management Research Laboratory, 26 W. Martin Luther King St., Cincinnati, OH 45268, USA b
a r t i c l e
i n f o
Article history: Received 24 October 2013 Accepted 6 May 2014 Available online 25 June 2014 Keywords: Vertical well Horizontal trench Bioreactor landfill Leachate recirculation Liquids addition
a b s t r a c t This study presents the development of design charts that can be used to estimate lateral and vertical spacing of liquids addition devices (e.g., vertical well, horizontal trenches) and the operating duration needed for transient operating conditions (conditions until steady-state operating conditions are achieved). These design charts should be used in conjunction with steady-state design charts published earlier by Jain et al. (2010a, 2010b). The data suggest that the liquids addition system operating time can be significantly reduced by utilizing moderately closer spacing between liquids addition devices than the spacing needed for steady-state conditions. These design charts can be used by designers to readily estimate achievable flow rate and lateral and vertical extents of the zone of impact from liquid addition devices, and analyze the sensitivity of various input variables (e.g., hydraulic conductivity, anisotropy, well radius, screen length) to the design. The applicability of the design charts, which are developed based on simulations of a continuously operated system, was also evaluated for the design of a system that would be operated intermittently (e.g., systems only operated during facility operating hours). The design charts somewhat underestimates the flow rate achieved and overestimates the lateral extent of the zone of impact over an operating duration for an intermittently operated system. The associated estimation errors would be smaller than the margin of errors associated with measurement of other key design inputs such as waste properties (e.g., hydraulic conductivity) and wider variation of these properties at a given site due to heterogeneous nature of waste. Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction Liquids (leachate, groundwater, other liquids source such as industrial liquids and biosolids) addition to the conventional municipal solid waste (MSW) landfill (also commonly referred as a dry tomb landfill) is the most common approach to bioreactor operation. Vertical wells and horizontal sources (trenches, infiltration galleries) are frequently used to add liquids to landfilled waste. Like any other landfill components, (e.g., leachate collection system, gas collection system) the liquids addition system (also commonly referred as leachate recirculation system or liquids introduction system) needs to be designed for permitting, construction, and operation of bioreactor landfills. Based on the authors’ design and review experience, a common approach for ⇑ Corresponding author. Tel.: +1 352 2834742, +1 352 3314828; fax: +1 352 3314842. E-mail addresses: [email protected] (P. Jain), ttown@ufl.edu (T.G. Townsend), [email protected] (T.M. Tolaymat). 1 Tel.: +1 352 3920846; fax: +1 352 3923076. 2 Tel.: +1 513 4872860; fax: +1 513 5697879. http://dx.doi.org/10.1016/j.wasman.2014.05.008 0956-053X/Ó 2014 Elsevier Ltd. All rights reserved.
design of a liquids addition system is design for steady state conditions. Jain et al. (2010a, 2010b) presented charts a designer can use for steady-state design of a vertical well and horizontal source system for liquids introduction. These charts can be used to estimate the achievable flow rate, lateral and vertical extents of the zone of impact, and the associated total liquids volume and operating time based on device dimensions and waste properties (hydraulic conductivity, anisotropy, porosity) at steady state conditions. For a design based on steady-state conditions, liquids addition devices are spaced based on the maximum possible lateral extent of the zone of impact of a device, which requires the addition of the liquid volume needed to reach the steady state, which in turn dictates the operating duration to achieve the steady state. Conditions however, may exist such that the timeframe over which liquids can be added is shorter than necessary to reach steady state; the timeframes needed to reach steady state were almost 19 years for the design chart application examples presented in Jain et al. (2010a) for vertical well systems, which is significantly greater than the typical life of a cell at MSW landfills in the US. Therefore, there is a need for tools to design liquids addition systems for conditions
1668
P. Jain et al. / Waste Management 34 (2014) 1667–1673
Nomenclature
a aL g q h hr hs b0 a
C d k K Kr Kz l m, n mv p pI Qv
Van Genuchten parameter related to air entry pressure fluid compressibility (M1 L T2) dimensionless variable density of fluid (M L3) moisture content (dimensionless) residual moisture content (dimensionless) porosity of the media (dimensionless) constant equals to 1 when p P 0 and 0 when p 10 −4
0.4
0.4
0.2
0.2
η ≤ 10 −4
0.0
0.0 0.0
0.2
0.4
0.6
0.8
1.0
(a)
0.8
r /r
I Is
0.6
Decreasing η and pI 0.4
0.2
0.0 0.2
0.4
0.6
0.8
1.0
Vt/Vt,critical
(b) Fig. 2. Design Chart 1. Fraction of steady-state lateral extent as a function of fraction of steady-state pore volume for (a) horizontal source, and (b) vertical well.
No significant trend in the xI/xIs (or rI/rIs) ratio variation with Vt/Vt,critical ratio was observed with respect to g and injection pressure; as defined in Jain et al. (2010a, 2010b) g is a function of source dimensions and waste anisotropy (a) as defined below.
w2 For vertical well : g ¼ a 2 rw For horizontal source : g ¼ a
w2 l
0.2
0.4
0.6
0.8
1.0
Fig. 3. Design Chart 2. Fraction of steady-state vertical extent as a function of fraction of steady-state pore volume for horizontal source.
1.0
0.0
0.0
V /V t t,critical
V /V t t,critical
2
In general, a lower fraction of the lateral extents of the impact zone was achieved for a lower g and injection pressure (pI) for a given fraction of liquids volume for both type of liquids addition devices (horizontal source, vertical well). These design charts can be used to estimate transient lateral extents of the impact zone as a function of liquids volume once the steady-state lateral extents and the associated liquids volume are estimated using the design charts presented by Jain et al. (2010a, 2010b). More details on range of g values of the simulations conducted and rational of the range are presented elsewhere (Jain et al. (2010a, 2010b)). 3.2. Design Chart 2. Transient vertical zone of impact Vertical spacing is typically specified for horizontal systems. Fig. 3 presents the fraction of the steady-state vertical extents of the zone of impact (below a horizontal source) achieved as a func-
tion of the fraction of the added liquids volume. A hypothetical line segregating the simulation results into two groups (g 6 104 and g > 104) is shown; data from simulations with g > 104 fell above the line, whereas data from simulations with g 6 104 fell below the line. No significant trend with respect to g and injection pressure (pI) was observed within a group. Approximately 40–55% of the vertical zone of impact can be achieved with the addition of 40% of liquids volume needed to achieve the steady-state vertical extent of the zone of impact. As shown, a majority of simulation data were above the line depicted on the plot; data from 20 out of 27 simulations were above the line. Based on the data presented in Figs. 2(a) and 3, it can be seen that the reduction in vertical spacing of devices is significantly greater than the horizontal spacing for transient design. For example, for the scenario where only 40% of steady-state liquid volume would be added through a device, the lateral spacing of the device should be 80–90% of the spacing for steady-state design (a reduction of only 10–20%), whereas, the vertical spacing should be 40–55% of the spacing for steady-state design (a reduction of 45–60%). 3.3. Design Chart 3. Operating duration Fig. 4(a) and (b) can be used to estimate the time period needed to add the target liquid volume to a horizontal source and vertical well, respectively, for scenarios where the design volume is less than needed to achieve the steady-state condition. For example, 40% of the volume needed to achieve steady state can be added to a horizontal system in 30–38% of the time needed to reach the steady state. This design chart can also be used for estimating the ratio of the mean to steady-state flow rate; the ratio of Vt/Vt,critical to t/ts is equivalent to the mean to steady-state flow rate. For example, approximately 20–30% of steady-state liquids volume can be added to a horizontal source, operated 20% of the time needed to reach steady-state conditions. The average flow rate would be 100–150% of the steady-state flow rate, dependent on g and injection pressure. The use of the steady-state flow rate for estimation of operating duration to add designed liquid volume may result in significant overestimation of the operating time. 3.4. Impact of intermittent operation In order to evaluate the applicability of the developed charts for the design or analysis of a system operated on an intermittent basis, an additional 4 simulations were conducted. Two were conducted for simulation of continuous operation (one with Kz = 105 cm/s and other with Kz = 106 cm/s). For the other two
1671
P. Jain et al. / Waste Management 34 (2014) 1667–1673 80
1.0
pI =5 l=w=1
Volume of moisture added per unit trench length (m 3 m-1)
-4
0.8
t/t s
0.6
0.4
0.2
K x=10 cm/s K z=10-5 cm/s
60
Intermittent operation (8 hrs on and 16 hrs off) 40 Continuous operation 20
K x =10-5 cm/s K z =10-6 cm/s
Decreasing η and pI
Intermittent operation (8 hrs on and 16 hrs off) Continuous operation
0
0.0 0.0
0.2
0.4
0.6
0.8
0
1.0
10
20
30
40
50
60
70
80
90
100
110
120
Operational time (days)
V /V t t,critical
(a)
Fig. 5. Cumulative volume of liquids added as a function of operational time for intermittent versus continuous operations.
1.0
t/ts
0.6
0.4
0.2
Decreasing η and pI 0.0 0.0
0.2
0.4
0.6
0.8
1.0
Vt/Vt,critical
(b) Fig. 4. Design Chart 3. Fraction of time to reach steady state as a function of fraction of pore volume to reach steady state for (a) horizontal source, and (b) vertical well.
the pressure-specified boundary condition at the source-waste interface was applied intermittently (cycled at 8 h on and 16 h off, one with Kz = 105 cm/s and other with Kz = 106 cm/s). Based on values reported in literature, a waste anisotropy (a) of 10 was used for these simulations (Landva et al., 1998; Singh et al., 2014); anisotropy is defined as the ratio of hydraulic conductivity in lateral direction to hydraulic conductivity in vertical direction. Simulations representing continuous injection scenarios were run for a duration of approximately 125 days (approximately 3000 h) while those representing the intermittent scenario were run for a duration of approximately 380 days (approximately 9000 h of total time or 3000 h of active injection/operating time). The added liquids volume, flow rate and lateral extents of liquids movement were compared for these cases. In general, the flow rate for the intermittent scenario was greater than the corresponding continuous scenario flow rate; the flow rate was zero for the 16-h shut down period for intermittent cases. This is attributed to the potential dissipation in fluid pressure around the source while the system is non-operational. The liquids move away from the source under the influence of gravity and capillarity and consequently reduce fluid pressure around the source when the system is shut down. Fig. 5 presents the cumulative volume of liquids added as a function of operation time of the system. The flow rate for continuous operation was approximately 33% lower than that for intermittent operation (based on operation time) at the end of the 125 operating days. Based on actual time (including the daily
16 h shut down period for intermittent operation), the average daily and cumulative flow rate was approximately half of that for continuous operation. The use of the design charts presented by Jain et al. (2010a, 2010b) would result in underestimation of steady-state flow rate based on operating time for intermittent operation and thus implicates a conservative design. The error associated with the use of the steady-state flow rates presented by Jain et al. (2010a, 2010b) for designing a system that would be operated intermittently could be treated as insignificant in comparison to the range and errors associated with assumptions of hydraulic conductivity, anisotropy, etc. Multiple measurements of hydraulic conductivity from a single landfill cell are reported to vary over an order of magnitude (Jain et al., 2006). Moreover, the calculation of flow rate based on dimensionless variable estimation from design charts would require site-specific measurement of hydraulic conductivity and anisotropy values of the waste at site under consideration. It should be noted that these systems are installed as landfilling progresses and thus need to be designed before the waste is in place. The site-specific hydraulic conductivity of the waste, therefore, could not be measured at the design stage, even if desired, under most circumstances except where a liquids addition system is designed for a landfill cell at final (or near final) grades. Fig. 6 presents the variation of the lateral extent of liquids movement for all of the 4 simulations described in this section as a function of liquids volume added. It can be seen that the lateral
12
pI=5 l=w=1 a=10
10
8
xI (m)
0.8
6
4 Intermittent Operation (Kz=10-5 cm/s) Continuous Operation (Kz=10-5 cm/s)
2
Intermittent Operation (Kz=10-6 cm/s) Continuous Operation (Kz=10-6 cm/s)
0 0
20
40
60
80
Volume of moisture added per unit trench length (m 3 m-1) Fig. 6. Variation of lateral extent of the zone of impact as a function of liquids volume added for intermittent and continuous operation.
1672
P. Jain et al. / Waste Management 34 (2014) 1667–1673
extents for both continuous and intermittent operation for Kz = 106 cm/s and continuous operation for Kz = 105 cm/s did not reach steady-state and compared very well with the values predicted based on the results presented earlier. The lateral extent for the simulation modeling intermittent operation at Kz = 105 cm/s reached steady-state and the steady-state value was almost 45% lower than the steady-state value predicted based on the results presented in previous section. It should be noted that steady-state lateral extents of the zone of impact is strongly correlated with flow rate (Fig. 5 in Jain et al. (2010a, 2010b)). As described earlier, the flow rate for intermittent operation based on operating time was 33% more and that based on the total time was 50% less than the steady-state flow rate estimated based on design charts presented by Jain et al. (2010a, 2010b). Since the flow rate in the intermittent scenario was not the same as for the continuous scenario modeled previously, two additional simulations were conducted to compare the lateral extents of the zone of impact between the continuous and intermittent operating strategies for a 1 m 1 m trench installed in waste with Kz = 106 cm/s and a = 10. Flux instead of injection pressure at the source-waste interface was specified as the boundary condition and the same liquids injection volume was employed for these simulations. A flux of 60 L (0.06 m3) per m trench length per day was used for these simulations. For the continuous scenario this volume was added over 24 h per day whereas for the modeled intermittent case the 60 L/m injection rate was added in 2 h and the system was shut down for 22 h to simulate the impact of an extreme end of intermittent operation. As expected, the injection pressure (during operating hours) needed to add the liquid volume over much shorter duration in intermittent scenario was significantly greater than that for the continuous injection scenario. The lateral extent for intermittent case was expected to be greater than the continuous injection scenario as the same volume of liquids is injected under much greater pressure. However, surprisingly, the lateral extent for the intermittent scenario was lower than that for the continuous injection scenario (Fig. 7). The results suggest that the duration of the applied pressure at the well has a greater impact on the lateral extent of the zone of impact than the magnitude of pressure for a given volume of liquids added. The liquids migrate away from the well laterally and vertically, due to the positive pressure at the source during operating hours, gravity and capillarity. While the system is shut down, liquids in the well vicinity predominantly migrate downwards under the influence of gravity and the lateral migration is slowed due to lower remnant pressure at the source. The
7
Kx=10-5 cm/s Kz=10-6 cm/s l=w=1 m
6
xI (m)
5
Continuous opreation
4
intermittent operation, therefore, results in greater vertical extent and lower lateral extent of the zone of impact than that of the continuous operation case. The magnitude of the error introduced by the use of design charts for estimation of the lateral extent of the scenario modeled was less than 10%, which is insignificant compared to the margins of error associated with measurement of the key waste properties such as hydraulic conductivity, which are a critical input to the liquids addition system design.
4. Limitations A constant waste permeability was assumed for all simulations. However, Kz is a function of overburden pressure, among other things such as liquids content and age of waste, and it is the expected to change with time and as landfilling progresses (Jain et al., 2006). These factors that impact Kz values are not well understood to date and were not considered in this study. Waste was modeled as a homogenous media, however, waste is highly heterogeneous in nature. In the absence of availability of reliable unsaturated media properties, unsaturated properties typical of sandy soil were used for modeling. Jain et al. (2010b) documented that unsaturated waste properties did not have a significant impact on steady-state results. The impact of landfill gas on liquid flow was not taken into consideration and waste was modeled as a single-phase flow system. Future studies should evaluate the impact of unsaturated media properties and landfill gas on the extents of the zone of impact, and collect and compare the actual measurements of the lateral and vertical extents of the zone of impact as a function of liquids volume added with the modeling results
5. Summary and conclusions Design charts were developed to estimate spacing of liquids addition devices under transient conditions. These design charts complement steady-state liquids addition system design procedure presented by Jain et al. (2010a, 2010b). The results suggest that by adding approximately 50% of the liquid volume needed to achieve the steady-state conditions, greater than 80% of the maximum lateral extents of the zone of impact can be achieved for both a horizontal source and vertical well. The results also suggest that design charts presented by Jain et al. (2010a, 2010b) underestimate the flow rate (on operating time basis) for intermittent operating conditions. The lateral extent of the zone of impact for a give volume of liquids added was found to be lower for intermittent operation even with greater operating injection pressure that for the continuous operation case. The design charts slightly overestimate the lateral extent of the zone of impact for an intermittent flow scenario for a given volume of liquids addition. Overall, the error associated with design of an intermittent operating system using charts provided by Jain et al. (2010a, 2010b) would be lower than the margin of error associated with measurement of the waste properties such as hydraulic conductivity that are key inputs in the liquid addition design procedure.
Intermittent operation (2 hrs on and 22 hrs off)
3
6. Disclaimer
2 1 0 0
5
10
15
20
25 3
30 -1
Volume of moisture added per unit trench length (m m ) Fig. 7. Variation of lateral extent of the zone of impact as a function of liquids volume added for intermittent and continuous operation for constant flow rate condition.
The US Environmental Protection Agency (US EPA) through its Office of Research and Development funded and managed the research described here under purchase order number: EP05C000550 to Innovative Waste Consulting Services, LLC. Mention of trade names or commercial products does not constitute endorsement or recommendation for use. The scientific views expressed are solely those of the authors and do not necessarily reflect those of the US EPA.
P. Jain et al. / Waste Management 34 (2014) 1667–1673
Appendix A. Example problems Example 1 Leachate recirculation at a landfill is planned using 0.5 m 0.5 m horizontal trenches. The liquids injection would occur at 5 m water column (w.c.) measured at the trench-waste interface. The hydraulic conductivity of the waste in the horizontal and vertical directions are 107 m s1 and 108 m s1 (anisotropy = 10), respectively, and the drainable porosity is 0.4. The values of other waste parameters were assumed to be the same as the ones presented in Table 1. The design engineer wishes to estimate the (a) vertical and (b) lateral spacing between devices, and (c) operating duration for adding 40 m3 of liquids per m trench. The steady-state lateral and vertical zones, liquid volume added, and operating time to reach the steady-state conditions were estimated using design charts presented by Jain et al. (2010b). The estimated steady-state values are presented as follows: 1. Steady-state lateral extent of the zone of impact, xIs = 11 m. 2. Steady-state vertical extent of the zone of impact, zIs = 12.5 m. 3. Volume needed to achieve steady state, Vt,critical = 100 m3 per m trench length. 4. Time taken to reach steady state, ts = 9.8 years of continuous operation. (a) As only 40 m3 of liquid will be added per m trench, the system would not reach the steady-state conditions. The design charts presented in this paper are used to estimate the lateral and vertical zones of impact and operating time for the design liquids addition volume. The fraction of steady-state volume (Vt/Vt,critical) added in this case is 0.4. Design Chart 1 is used to estimate the lateral extent of the zone of impact. For Vt/Vt,critical = 0.4, xI/xIs ranges from 0.79 to 0.92. The lateral extent of the zone of impact is estimated to range from 8.7 to 10.1 m. Therefore, lateral trench spacing is estimated to range from 17.4 to 20.2 m. (b) Using Eq. (8) in Jain et al. (2010b), g is estimated as follows:
g¼a
w2 2
l
¼ 10
ð0:5 mÞ2 ð0:5 mÞ2
¼ 10
Using Design Chart 2, for Vt/Vt,critical = 0.4, zI/zIs ranges from 0.5 to 0.55. Using the value of zIs presented previously, the vertical extent of the zone of impact is estimated to range from 6.3 m to 6.9 m and vertical trench spacing is estimated to range from 12.6 to 13.8 m. (c) Using Design Chart 3, t/ts is estimated to range from 0.33 to 0.39 for Vt/Vt,critical = 0.4. Using ts = 9.8 years presented above, the operating duration needed to add 40 m3 of liquids per m trench length ranges from 3.2 to 3.8 years. Example 2 A leachate recirculation system in a 40-m deep landfill consists of 0.6-m diameter and 18-m deep vertical wells each with
1673
a 15-m screen. The pressure at the bottom of the well is 17 m w.c. The hydraulic conductivity of the waste in the horizontal and vertical direction is 107 m s1 and 108 m s1 (anisotropy = 10), respectively, and the drainable porosity is 0.4. The values of other waste parameters were assumed to be the same those presented in Table 1. The design engineer wishes to estimate the (a) well spacing and (b) operating duration for addition of 7500 m3 to a well. The steady-state lateral and vertical zones, liquid volume added, and operating time to reach steady-state conditions are estimated used design charts presented by Jain et al. (2010a). The estimated steady-state values are presented as follows: 1. Steady-state lateral extent of the zone of impact, rI = 28.5 m. 2. Volume needed to achieve steady state, Vt,critical = 15,260 m3. 3. Time taken to reach steady state, ts = 18.9 years of continuous operation. (a) As only 7500 m3 of liquid will be added to a well, the system would not reach the steady-state conditions within the operating period. The design charts presented in this paper are used to estimate the lateral zone of impact and operating time for the design liquids addition volume. The fraction of steady-state volume (Vt/Vt,critical) added in this case is 0.49. Design Chart 1 is used to estimate the lateral extent of the zone of impact. For Vt/Vt,critical = 0.49, rI/rIs ranges from 0.78 to 0.93. The lateral extent of the zone of impact is, therefore, estimated to range from 22.2 to 26.5 m. The well spacing is, therefore, estimated to range from 44 to 53 m. (b) Using Design Chart 3, t/ts is estimated to range from 0.4 to 0.46 for Vt/Vt,critical = 0.49. Using ts = 18.9 years presented above, the operating duration needed to add 7500 m3 of liquids per well ranges from 7.6 to 8.7 years.
References Jain, P., Powell, J., Townsend, T., Reinhart, D., 2006. Estimating the hydraulic conductivity of landfilled municipal solid waste using the borehole permeameter test. J. Environ. Eng., ASCE 132 (6), 645–652. Jain, P., Tolaymat, T., Townsend, T., 2010a. Steady state design of vertical wells for liquids addition at bioreactor landfills. Waste Manage. 30 (11), 2022–2029. Jain, P., Tolaymat, T., Townsend, T., 2010b. Steady state design of horizontal systems for liquids addition at bioreactor landfills. Waste Manage. 30 (12), 2560–2569. Krahn, J., 2004. Seepage Modeling with SEEP/W: An Engineering Methodology. GeoSlope International, Calgary, Canada. Landva, A.O., Pelkey, S.G., Valsangkar, A.J., 1998. Coefficient of permeability of municipal refuse. In: Proceedings of the Third International Congress on Environmental Geotechnics, Lisbon, Portugal, pp. 163–167. Singh, K., Kadambala, R., Jain, P., Xu, Q., Townsend, T., 2014. Anisotropy estimation of compacted municipal solid waste using pressurized vertical well liquids injection. Waste Manage. Res.. http://dx.doi.org/10.1177/0734242X14532003. Stephens, D.B., 1995. Vadose Zone Hydrology. Lewis Publishers, Boca Raton, FL, USA. Tchobanoglous, G., Theisen, H., Vigil, S., 1993. Integrated Solid Waste Management. McGraw-Hill, New York, USA. Van Genuchten, M.Th., 1980. A closed-form equation for predicting the hydraulic conductivity of unsaturated soils. Soil Sci. Soc. Am. J. 44, 892–898.
