THE EFFECT OF SALINITY ON EVAPORATION RATES OF BRINES RESULTING FROM THE TREATMENT OF MINE WATER

Search for keyword THE EFFECT OF SALINITY ON EVAPORATION RATES OF BRINES RESULTING FROM THE TREATMENT OF MINE WATER Priyal Dama-Fakir, Annalien Toeri...
Author: Karen Morrison
25 downloads 0 Views 590KB Size
Search for keyword

THE EFFECT OF SALINITY ON EVAPORATION RATES OF BRINES RESULTING FROM THE TREATMENT OF MINE WATER Priyal Dama-Fakir, Annalien Toerien Golder Associates Africa, Tel.: 011 254 4967, email: [email protected] Abstract: Recent advancements in membrane technology has lead to an increase in the number of water treatment plants treating acid mine drainage water from coal mines to drinking water quality. Brine is one of the waste streams arising from these treatment processes. The most popular waste management option for the brine produced at the water treatment plants is brine evaporation ponds. The size of these evaporation ponds is dependent on the rate of evaporation from the pond. Salinity has a negative impact on evaporation rates. The purpose of this study is to determine the rate of evaporation for typical brines produced from the treatment of acid mine drainage, as opposed to municipal drinking water. The reduction in the rate of evaporation is often referred to as a salinity factor. A salinity factor of 70 % is recommended from literature. Preliminary field testing provided average salinity factors of between 80 and 87 %. These results have however been impacted upon by stratification. It has been decided that the methodology will be modified to include frequent mixing in the evaporation pans in order to reduce the effect of stratification. Introduction: Historically, mine water was considered a lost resource, but it is now considered a valuable source of water for augmenting current water demands. In several catchments, substantial water resources are stored in old and active mine workings. This water is typically continuously recharged by surface and underground water. If not utilised, the mining voids fill up, and the water, as it reaches the surface, decants into rivers and streams, resulting in surface water contamination. Improvements in membrane processes have made it possible to treat acid mine drainage (AMD) to a drinking water quality to be sold into water distribution systems. Membrane processes include Reverse Osmosis (RO), Nanofiltration (NF), electrodialysis (ED), ultrafiltration (UF) and microfiltration (MF) (American Water Works Association, 1999). The selection of the treatment type and appropriate membrane is dependent on the feed water quality and the desired end use of the water (Baker, 2000). Reverse Osmosis treatment is often selected for the treatment of AMD from coal mines due to the requirement to remove monovalent ions such as sodium and chloride, as well as divalents such as sulphate. This treatment results in a reject brine, or concentrated salt stream which has to be managed as a hazardous waste. The concentration of the brine is dependent on several factors including the feed water quality, pre-treatment processes

used, the membrane selected, water recovery achieved and post treatment applied (du Plessis et al, 2006). Current projects being investigated and implemented result in brine with a Total Dissolved Solids (TDS) level of between 17 500 mg/ℓ and 51 000 mg/ℓ. General disposal options for brine include; disposal to surface water resources, discharge to sewer, deep well injection, spray irrigation, mechanical or thermal evaporation or evaporation ponds, depending on the brine concentrations.   For the coal mining operations being considered under this project, evaporation ponds are the preferred option for the following reasons: •

The concentrates produced have high TDS concentrations with high sulphate and metal content and are not suitable for discharge to surrounding water sources, sewers or irrigation,



Obtaining permission for deep well injection in South Africa is unlikely due to the unqualified risk of groundwater contamination,



The mines are far away from the sea, thus discharge to sea is not practical,



The low rainfall, high temperature and windy climate makes evaporation ponds a viable option,



Operational costs of mechanical or thermal evaporation systems are exorbitant, and many projects will be financially unviable if this route is taken.

One of the key factors influencing the design of evaporation ponds is the evaporation rate. The evaporation rates for water are readily available for various catchments in South Africa. The high salinity of the brine however will reduce this evaporation rate. The aim of the study is to conduct a literature review to obtain reported evaporation rates on saline solutions, to carry out field investigations to measure the evaporation rate on brine solutions and to compare this against that of a standard water sample in order to develop a higher degree of confidence in the salinity factor applied to the design of evaporation ponds. Literature review Factors affecting evaporation The proper sizing of an evaporation pond depends on the accurate calculation of the evaporation rate. Evaporation ponds function by transferring water in the pond into water vapour in the atmosphere above the pond. (Ahmed et al, 2000). The larger the surface area, the greater the rate of evaporation from the pond. Climatic effects In order for evaporation to occur, sufficient energy is required to change the water from a liquid form to its vapour form. The air above the evaporation pond becomes saturated with water vapour and this moist air must be removed in order to allow the process to continue. The climatic factors affecting the rate of evaporation include:

• • • •

Temperature – heating the water molecules to the required temperature for vaporisation, Humidity – if the humidity levels are high, less evaporation occurs as saturation levels are quickly reached, Solar radiation – providing heat to enable evaporation, Wind – to replace the saturated air with unsaturated air to allow further evaporation to occur.

Properties of the liquid being evaporated Dissolved salts in the water result in a lower saturation vapour pressure due to the decreased chemical potential of the water and thus a lower evaporation rate. The second Law of thermodynamics implies that an increase in ion activity as a result of the presence of solute reduces the chemical potential of a liquid solvent and also the rate of spontaneous transformation of the liquid phase into the vapour phase (Kokya & Kokya, 2006). Procedures for calculating evaporation rates indicate that evaporation rate is directly proportional to vapour pressure. The vapour pressure of saline water is lower than that of fresh water resulting in a reduction in evaporation. (Mickley, 2001). Cohesive forces between the dissolved ions and water may also inhibit evaporation, as it will make it more difficult for the water molecules to vaporise (Miller, 1989) Leaney and Christen, (2000) indicated that the evaporation rate decreases exponentially with increasing salinity. Evaporation rates for fresh water in various areas are easily available from hydrological databases. The evaporation rates of saline solutions are calculated by multiplying the rate of evaporation of water by a salinity factor (Mickley, 2001).

 

Figure 1: Effect of salinity factor on the footprint area of a pond As shown in Figure 1 where a decrease in salinity factor, (indicating a more concentrated solution) leads to an increase in surface area required for evaporation. The pond footprint area increases exponentially with reduced evaporation rates. This leads to exorbitant capital costs to establish ponds arising from expensive liner systems and land

requirements. Estimates carried out show that the disposal cost of brine can be in the order of 15 % of the cost of desalination in the case of inland locations (Gilron et al, 2003). The dissolved salt composition will have an effect on the humidity levels at which evaporation will cease, for example for a water body saturated with sodium chloride, there will be no evaporation during periods when the humidity is above 70 %. For other salts, evaporation may cease at lower humidity levels. (Leaney & Christen, 2000) Surface area The surface area of the water body will have an effect on the evaporation rates for areas of similar climates. The evaporation rate from a water body is reduced from a maximum value for small water bodies, as determined for a standard evaporation pan, to a fraction of that value for larger bodies (Leaney & Christen, 2000). This fraction is often referred to as the pan factor. Equation 1 can be used for determining the pan factor.   [1]  

 

 

Where:

A = surface area of water body (ha)

Methods of determining evaporation rates Several equations and field testing methods have been developed for determining evaporation rates. One such equation, the Penman equation, was used by hydrologists to determine evaporation rates from open water sources. This equation is acceptable for natural water resources with relatively low TDS levels. The Penman equation was modified to reflect the reduced vapour pressure of a saline solution (Akridge, 2008). The modified Penman equation is presented as Equation 2.      

Where :

 

 

 

    [2] 

 

E = evaporation λ = latent heat of vaporisation γ = psychometric constant Rn = net solar radiation f(u) = function of wind speed es = saturation vapour pressure e = ambient water vapour pressure

As discussed, the rate of evaporation is negatively impacted upon by TDS. In South Africa, a data base of evaporation rates is available for various quaternary catchments. These rates are however measured and reported for low TDS waters. This rate is multiplied by a

salinity factor for solutions with a high TDS. Equation 3, know as the Turk and Byonthon equation, can be used for determining the salinity factor (Leaney & Christen, 2000): [3]

Where:

Fsalinity = salinity factor S = salinity in g/ℓ

Reported salinity factors Limited information was obtained from literature on typical salinity factors for brine produced from RO treatment. •

(Ahmed et al, 2000) cited a paper (USDI, 1970) stating that brine evaporation rates decreased by 1 % for each 0.01 increase in specific gravity compared to that of distilled water.



(Kokya & Kokya, 2006) observed a reduction in evaporation rate for sea water at a TDS of approximately 40 g/ℓ to 0.94 times that of fresh water.



Mickley, (2006) cited various studies which indicate that the salinity factors of 2, 5, 10 and 20 % NaCl solutions are 98, 97, 93 and 78 % respectively.

The above studies were not carried out on brines produced in the treatment of mine water. Mine water emanating from South African coal fields and desalinated with membrane based technologies produce a waste brine typically containing metals, including iron, aluminium, manganese, calcium and magnesium sulphates and other compounds (Van Niekerk, 2009). Past research has indicated that the composition of the waste will have an effect on the evaporation rate (Leaney & Christen, 2000). This highlighted the need for a field study to better quantify the effect of various factors on the evaporation rates of brine solutions produced in the treatment of mine water. Mickley (2006) concluded that there is no simple relationship between salinity and evaporation, as there are always complex interactions among site-specific variables such as temperature, wind speed, relative humidity, atmospheric pressure, water surface temperature, heat exchange rate with atmosphere, incident solar absorption and reflection, thermal currents and the depth of ponds. A salinity factor of 0.7 was recommended for use in the design of evaporation ponds. This factor needs to be confirmed and further quantified. Methodology for field investigations The need for a better understanding of evaporation rates for brine solutions was identified in the literature review. Important factors influencing evaporation rates included climatic conditions, brine composition and concentration. Climatic conditions

In order to understand the impact of climatic conditions on evaporation rates, a Davis Pro2 Weather Station was purchased. The weather station has the ability to monitor temperature, humidity, rainfall, wind speed, wind direction, and solar intensity. A data logger and software that enable data collection and reporting from the Davis Pro Weather Station was purchased for the project. A photograph of the weather station is presented in Figure 2.

 

Figure 2: Davis Pro 2 weather station at the Golder offices in Midrand Evaporation pans Standard A-Pans were installed to measure evaporation rates. In order to replicate data collected at different locations, it was necessary to use standard equipment. Due to the corrosive nature of the brine, the pans were coated with a grey corrosion proofing paint. Four pans were set-up in total. One was used as a standard and the other three represented three different brine compositions and concentrations. Standard brine solutions For practical purposes, standard solutions were prepared for the study. The brine produced at the eMalahleni Water Reclamation Facility was analysed and used to prepare the standard solution. Brine from the proposed Optimum and Kilbarchan mine water treatment plants was not available as these plants were not constructed at the time of the study. The brine produced during bench scale testing of the Kilbarchan Colliery water was analysed and used for the study. The modelled brine composition expected from the treatment of Optimum mine water was used. The following chemicals used to produce solutions containing the major cations and anions found in the brines being investigated. •

sodium chloride,



sodium sulphate,



calcium sulphate,



potassium chloride,



sodium nitrate,



magnesium sulphate.

The mine water treated at the eMalahleni Water Reclamation Facility water is largely a high calcium and magnesium sulphate based water. High concentrations of potassium, sodium and chloride are also present. The brine produced from membrane treatment of this water therefore contains concentrated levels of sodium, calcium, potassium, sulphate and chloride. The relative salt concentrations in this brine is shown in Figure 3.

 

Figure 3: Relative salt concentrations: eMalahleni Water Reclamation Facility The brine produced during the treatment of mine water from Optimum Mine typically has high concentrations of sodium, potassium and sulphate, with calcium, magnesium and chloride also present in significant concentrations. The relative salt concentrations of this brine are shown in Figure 4.

 

Figure 4: Relative salt concentrations: Optimum mine water Mine water from the Kilbarchan Colliery is a sodium sulphate type water, with calcium, magnesium and chloride also present in high concentrations. The brine resulting from the treatment of this water is of a much higher concentration compared to brine from the Optimum and eMalahleni treatment processes. The relative salt concentrations is shown in Figure 5.

 

Figure 5: Relative salt concentrations: Kilbarchan water A comparison of the different brines being investigated in the study is shown in Figure 6. Standard solutions for all three brines are being investigated in order to investigate the effect of composition on evaporation rate.

 

Figure 6: Composition of brine resulting from membrane treatment of different mine waters Only the major constituents were taken into account when preparing the standard solution. The standard solutions were made-up as presented in Table 1. Analytical grade chemicals were used. Table 1: Mass of salts added for standard solutions Salt Added (mg/ℓ) eMalahleni Optimum Kilbarchan Sodium Sulphate 14 204 18 457 14 204 Calcium sulphate 3 145 3 989 1 225 Potassium Chloride 2 177 5 218 373 Magnesium Sulphate 325 7 041 3 972 Sodium Nitrate 136 425 0 Sodium Chloride 584 2 922 21 917 Salt

 

Experimental set-up Field studies are being carried out at the Golder Offices in Midrand. The following aspects were taken into consideration in the selection of the location for the evaporation pans and weather station: •

Ease of access – water levels in the pans will be read manually on a daily basis.



A reasonable distance away from buildings, trees or other structures that will impact on the results by obstructing rainfall, airflow, solar intensity etc.



A level surface in required.



No interference from irrigation systems.

The weather station with data logger and four sets of evaporation pans were installed. The evaporation pans contain the following solutions: •

Pan 1 – reference sample – municipal water



Pan 2 – standard eMalahleni type brine



Pan 3 – standard Kilbarchan type brine



Pan 4 – standard Optimum type brine

A paved area in front of the office block, as indicated in Figure 7 was selected. The pans were placed on wooden sheets to limit heat transfer between the paving and the evaporation pan. Chicken mesh was fitted over the pans to protect against interference. Since the pans would not have a constant inflow, the methodology allowed for top-up the pans when the level fell below 135 mm. The pan was designed to enable level readings up to 138 mm.

Figure 7: Evaporation pans installed at Golder offices in Midrand Data recorded The following information is recorded: •

Hourly climate data,



Daily levels (Monday to Friday) in the evaporation pans,



TDS of the brine: o Standard solution. o Before top-up. o After top-up.

Results Evaporation rates Daily evaporation rates were calculated using the level readings in the pans and the daily rainfall figures. The rainfall was subtracted from the level difference to determine the actual evaporation. The recorded evaporation rates are presented in Figure 8. Note that since levels are not measured on Saturdays and Sundays, these are not necessarily daily evaporation rates. As time progressed, it was noted that the evaporation rate in the pans containing brine tended towards the evaporation rate in the pan containing water.

Figure 8: Recorded evaporation levels Salinity factor Pan 1 which contained municipal water was used as the reference. Salinity factors were calculated by dividing the evaporation rates of the brine containing pans by the evaporation rate of the reference pan. The 5, 25, 50, 75, 95 percentile and average values were calculated. These are presented in Table 2. Table 2: Statistical analyses of calculated salinity factors (original database) Date    5 %tile  25%tile  50%tile  75%tile 

Salinity factor (%)  eMalahleni  Kilbarchan  Optimum  40.26  37.09  3.23  75.37  68.24  62.55  91.08  86.27  95.14  100  100  105.85 

Date    95%tile  Average 

Salinity factor (%)  eMalahleni  Kilbarchan  Optimum  119.59  130.38  137.41  91.10  86.08  84.88 

Several salinity factors above 100 % were observed. It was also observed that some pans displayed negative evaporation. This may be due to errors when reading the levels in the pans or the effect of nearby buildings or trees. A decision was made to exclude all results below 20 % and above 100 %. The 5, 25, 50, 75 and 95 percentile and averages were calculated for the adjusted data and are presented in Table 3. Table 3: Statistical analyses of salinity factors on adjusted dataset Date    5 %tile  25%tile  50%tile  75%tile  95%tile  Average 

Salinity factor (%)  eMalahleni  Kilbarchan  Optimum  37.33  36.75  41.53  71.43  63.04  61.52  86.84  84.90  80.00  95.97  100.00  100.00  100.00  100.00  100.00  79.15  78.82  77.04 

 

The results were also presented graphically in Figure 9.

Figure 9: Adjusted salinity factors A linear trend analyses as presented in Figure 9 indicated that the salinity factor increased with time and approached 100 %. The Midrand area has had a total of 394 mm of rain between the 14 October 2009 and 25 January 2010. The increases salinity factor was initially assumed to be due to the dilution of the saline solutions in the pans as a result of

the rainfall. Samples were sent to an analytical laboratory for analyses. Results obtained are presented in Table 4. Sample ID pH GAA Pan GAA Pan GAA Pan GAA Pan

1 2 3 4

7.7 7.9 8.4 8.2

Table 4: Brine analyses results October 2009 January 2010 EC TDS pH EC (mS/m) (mg/ℓ) (mS/m) 24.4 104 7.8 32.4 2 130 18 930 8.9 2 718 4 950 40 284 8.6 8 960 3 456 32 536 8.4 4 400

TDS (mg/ℓ) 226 25 638 55 366 44 674

The TDS levels obtained from the laboratory analyses of samples indicate that the solutions in the pans concentrated with time. A possibility explanation for the improved evaporation rates of the brine solutions is that stratification was taking place in the pans with the higher concentrations of salts at the bottom of the pans. A conductivity profile of the pans confirmed that stratification took place in the pans and that the surface layer had a lower TDS concentration similar to water. The conductivity profile is shown in Table 5 and Figure 10. It was not possible to obtain a conductivity reading from the upper 4 cm in the pan without taking a sample. It is however suspected that the upper layer is mainly rain water, and hence evaporating at a similar rate to that in the reference pan. Table 5: Conductivity profile in the pans Conductivity (mS/m) Depth (cm) Pan1 Pan2 Pan3 Pan4 4 14.6 87.4 205.8 142 6 6 28 1 751 3 151 2 416 8 27 1 758 3 149 2 699 10 26.7 1 788 3 301 2 670 12 26.8 1 946 4 461 3 358 14 26.8 2 001 5 419 3 983 16 26.7 2 045 5 463 4 004 18 26.6 2 007 5 405 3 969  

 

Figure 10: Conductivity profile in the evaporation pans

Recommendations The following recommendations were made to improve on the findings of the study: •





Repeat the study at the eMalahleni Water Reclamation Facility where brine is produced and compare the evaporation rates on actual brine, as opposed to the standard solutions prepared. Ensure regular mixing of the pan contents to reduce the effect of stratification in order to compare the evaporation rate of the saline solutions to the standard reference (clean water) Continue the experimental work during the dry season.

Acknowledgements The project team would like to thank the Water Research Commission for funding the study. We would also like to thank the eMalahleni Water Reclamation Facility, Optimum Mine and Kilbarchan Colliery for their involvement in the study. References Ahmed, M., Shayya, W. H., Hoey, D., Mahendran, A., Morris, R., & Al‐Handaly, J. (2000). Use of evporation  ponds for brine disposal in deslaination plants. Desalination , 155 ‐ 168.  Akridge, D. G. (2008). Methods of calculating brine evaporation rates during salt production. Journal of  Archaeological Science , 1453 ‐ 1462.  American Water Works Association. (1999). Water Quality and Treatment ‐ A handbook of community  water suplies (5th Edition ed.). (R. D. Letterman, Ed.) McGraw‐Hill Handbooks.  Baker, R. W. (2000). Membrane technology and applications. New York: McGraw‐Hill ‐ Professional  Engineering.  du Plessis, J., Burger, J., Swart, C., & Museev, N. (2006). A Desalination Guide for South African Municipal  Engineers. Pretoria: Water Research Commission.  Gilron, J., Folkman, Y., Savliev, R., Waisman, M., & Kedem, O. (2003). WAIV ‐ wind aided intensified  evaporation for the reduction of desalination brine volume. Desalination 158 , 205 ‐ 214.  Kokya, B. A., & Kokya, T. A. (2006). Proposing a formula for the evaporation measurement from salt water  resources. Hydrological processes , 22, 2005 ‐ 2012.  Leaney, F., & Christen, E. (2000). On‐Farm and community‐scale salt disposal basins on the riveine plain:  Evaluating the leakage rate, disposal capacity and plume development. CRC for catchment hydrology.  Mickley, M. (2001). Membrane concentrate disposal: Practices and regulation. US Department of Interior.  Sartori, E. (2000). A critical review on equations employed for the calculation of evaporation rate from free  saline surfaces. Solar energy 68‐1 , 77 ‐ 89.  Schutte, F. (2009). Chemical Water Treatment notes from Water Utilisation Engineering. April 2009 . 

Schutte, F. (2006). Handbook for the operation of Water Treatment Works. Pretoria: Water Institute of  South Africa.  Van Niekerk, A. M. (2009). Mine Water Reclamation and Re‐use ‐ South African Experience. Mine Water  Conference 2009.    

Suggest Documents