Life cycle assessment of treatment and handling options for a highly saline brine extracted from a potential CO2 storage site

Water Research 122 (2017) 419e430

Contents lists available at ScienceDirect

Water Research journal homepage: www.elsevier.com/locate/watres

Life cycle assessment of treatment and handling options for a highly saline brine extracted from a potential CO2 storage site Hafiz H. Salih 1, Jiaxing Li 1, Ruth Kaplan, Seyed A. Dastgheib* Illinois State Geological Survey, University of Illinois, 615 East Peabody Drive, Champaign, IL 61820, United States

a r t i c l e i n f o

a b s t r a c t

Article history: Received 14 September 2016 Received in revised form 29 May 2017 Accepted 11 June 2017 Available online 12 June 2017

Carbon dioxide (CO2) injection in deep saline aquifers is a promising option for CO2 geological sequestration. However, brine extraction may be necessary to control the anticipated increase in reservoir pressure resulting from CO2 injection. The extracted brines usually have elevated concentrations of total dissolved solids (TDS) and other contaminants and require proper handling or treatment. Different options for the handling or treatment of a high-TDS brine extracted from a potential CO2 sequestration site (Mt. Simon Sandstone, Illinois, USA) are evaluated here through a life cycle assessment (LCA) study. The objective of this LCA study is to evaluate the environmental impact (EI) of various treatment or disposal options, namely, deep well disposal (Case 1); near-zero liquid discharge (ZLD) treatment followed by disposal of salt and brine by-products (Case 2); and near-ZLD treatment assuming beneficial use of the treatment by-products (Case 3). Results indicate that energy use is the dominant factor determining the overall EI. Because of the high energy consumption, desalination of the pretreated brine (Cases 2 and 3) results in the highest EI. Consequently, the overall EI of desalination cases falls mainly into two EI categories: global warming potential and resourcesefossil fuels. Deep well disposal has the least EI when the EI of brine injection into deep formations is not included. The overall freshwater consumption associated with different life cycle stages of the selected disposal or treatment options is 0.6e1.8 m3 of freshwater for every 1.0 m3 of brine input. The freshwater consumption balance is 0.6 m3 for every 1.0 m3 of brine input for Case 3 when desalination by-products are utilized for beneficial uses. © 2017 Elsevier Ltd. All rights reserved.

Keywords: Life cycle assessment Carbon dioxide sequestration Highly saline brine Deep well injection Evaporation

1. Introduction The accumulation of carbon dioxide (CO2) and other greenhouse gases in the atmosphere results in global warming. However, greenhouse gas emissions can be significantly reduced by capturing CO2 from emitting sources and storing it (Bachu and Adams, 2003). Among the various storage or sequestration options, geological sequestration by the dissolution of CO2 into deep saline formations (Finley, 2014) is one of the most attractive long-term storage methods because of the availability of high-volume saline formations (Birkholzer and Zhou, 2009). The U.S. Department of Energy has identified deep saline reservoirs as the largest potential sinks for CO2 storage in the United States. The CO2 sequestration capacity of saline formations is estimated to be in the range of 2379 to 21,633 billion metric tons of CO2 (U.S. Department of Energy, 2015);

* Corresponding author. E-mail address: [email protected] (S.A. Dastgheib). 1 Both authors contributed equally to this manuscript. http://dx.doi.org/10.1016/j.watres.2017.06.032 0043-1354/© 2017 Elsevier Ltd. All rights reserved.

however, the ultimate CO2 storage capacity in saline aquifers is dependent on several conditions, including pressure, temperature, and salinity of the formation water (Bachu and Adams, 2003). Large-scale industrial CO2 sequestration in deep saline reservoirs may cause the reservoir pressure to increase; however, continuous brine extraction is a potential strategy to manage pressure buildup and increase the CO2 storage capacity in the formation (Birkholzer and Zhou, 2009; Buscheck et al., 2011). Depending on the characteristics of the extracted brine and the availability of various brine management options, brine may be disposed of in other suitable geological formations (i.e., deep well disposal [DWD]) or considered for beneficial reuse after the required treatments. A large-scale carbon capture and storage (CCS) demonstration project located in Decatur, Illinois, has already stored 1 million tons of CO2 in the Mt. Simon Sandstone (Leetaru and Freiburg, 2014). The Mt. Simon Sandstone geological formation is 2600 ft thick and covered by the Eau Claire Formation (300 ft of low-permeability limestone). The Mt. Simon Sandstone, with its estimated storage capacity of 11e150 billion tons of CO2 (Leetaru and Freiburg, 2014),

420

H.H. Salih et al. / Water Research 122 (2017) 419e430

is considered one of the most suitable geological formations for carbon sequestration in the United States. Fig. 1 illustrates a scenario in which the supercritical CO2 is injected into the Mt. Simon Sandstone at a depth of ~7000 ft and brine is extracted at a different location to regulate the formation pressure. The extracted brine is then pretreated to remove the suspended solids before either disposal by DWD into the Potosi Dolomite (depth of ~4800 ft) or further treatment by desalination processes for beneficial reuse. The Potosi Dolomite is considered an excellent reservoir for wastewater disposal in the Illinois Basin because of its depth and porosity and because it has been used for disposal of brine wastewater from the oil industry for decades (Brower et al., 1989; Leetaru et al., 2014). Through a comparative life cycle assessment (LCA) analysis, we evaluated the environmental impact (EI) of various treatment or disposal options, including pretreatment of the extracted brine to remove suspended solids, DWD of the pretreated brine, a near-zero liquid discharge (ZLD) treatment followed by disposal of salt and concentrated brine by-products by DWD and solid waste disposal to landfill (SLD), and a near-ZLD treatment that assumes beneficial use of the treatment product (i.e., purified water) and by-products (i.e., dried salts and concentrated brine) (see Fig. 2 in the Materials and Methods Section). We focused on management options that are feasible with presently available technologies, namely, disposal or evaporation. Membrane processes such as reverse osmosis (RO) have salinity limitations that are exceeded by the Mt. Simon brine, and emerging high-total dissolved solids (TDS) desalination technologies are not yet advanced enough to accommodate the high volumes of brine that might be extracted (Kaplan et al., 2017). Selection of the best option for managing the extracted brine depends on various technical, economic, regulatory, and environmental factors. Some options might be more technoeconomically feasible but would have a greater negative impact on the environment. The main objective of this work was to evaluate and quantify the EI of each potential option for managing the extracted brine through a comparative LCA. Life cycle assessment results provide critical information for selecting the most environmentally friendly option. Among the published studies available on the LCA of water desalination processes (e.g., Vince et al., 2008; Zhou et al., 2014), many assume or show that the EI of the desalination processes

mainly depends on their energy use and that the impacts of chemical usage and infrastructure construction are less significant. For example, the EI of construction of the treatment infrastructure was estimated at 4%e10% of the total EI (Lundie et al., 2004; Zhou et al., 2014). The EI of pretreatments, including sludge disposal (i.e., sludge generated during coagulation or water softening) and management of the reject concentrated brine stream (i.e., that generated from the desalination process), is often not considered (Raluy et al., 2005a; Vince et al., 2008). Their impact might be significant, however, depending on the type of desalination process and feed water composition. The predominant impact of energy use in water desalination is evident in some LCA cases. For example, Raluy et al. (2005b) and Zhou et al. (2011a) found that ~80% of the overall EI of water desalination by RO was associated with electricity consumption (Raluy et al., 2006; Zhou et al., 2011a). However, this approach was challenged by several researchers who showed that when the pretreatment stage was considered, the construction of the infrastructure contributed 30%e50% of the total EI (Zhou et al., 2014). Furthermore, some studies indicated that chemical usage for water desalination had a significant impact on environmental acidification, global warming, eutrophication, and ozone depletion, especially when the chemical dosage was large (Tarnacki et al., 2012; Vince et al., 2008; Zhou et al., 2011a, 2014). Finally, the current literature on LCA lacks thorough investigations of the impact of desalination processes on water resources (i.e., the balance of freshwater withdrawal consumption versus the amount of freshwater produced throughout the desalination process). In this study, we investigated the EI of high-TDS brine management options. First, the current LCA literature provides limited information on brine disposal by DWD (Coday et al., 2015). Here, we consider the EIs of pretreatment, infrastructure, and disposal in addition to energy consumption. We also consider the EI of waste management through DWD or landfill disposal. Second, although a number of desalination processes have been examined by LCA, few studies have investigated the EI of high-TDS desalination processes. The majority of publications that include an LCA of desalination are related to brackish water or seawater desalination by conventional multistage flash distillation, multiple-effect evaporation (MEE), ndez-Torres forward osmosis, and RO (Coday et al., 2015; Ferna et al., 2012; Morton et al., 1997; Raluy et al., 2005a; Ras and Von

Fig. 1. Schematic diagram showing the injection of CO2 with the extraction of brine for pressure management, and handling of the pretreated brine by deep well disposal (DWD) or a combination of brine treatment and disposal. Formation information is taken from Leetaru and Freiburg (2014) and Leetaru et al. (2014). TDS, total dissolved solids.

H.H. Salih et al. / Water Research 122 (2017) 419e430

421

Fig. 2. Extracted brine handling or treatment flow diagram. The dashed boxes show the system boundary for each life cycle assessment case. SLD, solid landfill disposal; DWD, deep well disposal; near-ZLD, near-zero liquid discharge.

Blottnitz, 2012; Tarnacki et al., 2012; Zhou et al., 2011b). The third contribution of this work is to present a comparative LCA study of different management options (i.e., disposal and treatment options) for a high-TDS brine extracted from a potential CO2 sequestration site. 2. Materials and methods 2.1. Disposal or treatment options for extracted brine Fig. 2 shows a schematic diagram of handling or treatment options for extracted Mt. Simon brine. The pretreatment of extracted brine is the first stage for all options because it is required for pH adjustment and removal of suspended solids. Pretreatment begins with rapid mixing with lime and alum for 1 min (coagulation), followed by slow mixing for 30 min (flocculation), then 30 min of settling (sedimentation), and finally sand filtration, resulting in pretreated water and waste sludge. The design basis for each option was 8 million gallons per day (MGD) of brine inflow for a process lifetime of 20 years. A solids content of 80 wt% was assumed for the sludge generated from the pretreatment process. Sludge generated during the pretreatment process was sent for SLD with no further drying. Three options were considered for the disposal or treatment of pretreated brine, as described briefly in Table 1. For the disposal option (Case 1), we considered DWD into the Potosi Dolomite. For the treatment options, evaporation and crystallization by MEE were

selected as the best commercially available technologies for the near-ZLD treatment, followed by either the disposal of salt and concentrated brine by-products (Case 2) or the beneficial use of treatment products (i.e., purified water, and concentrated brine and salts by-products) (Case 3). Credit to the system was given for offsetting purified water production (Cases 2 and 3) and salt production (Case 3). Generated purified water can be used as cooling water in power plants with no further treatment, whereas produced salt and reject concentrated brine might be utilized for road deicing. Thus, the transportation of purified water to a power plant and produced salt and concentrated brine to a salt collection site located 5 miles from the treatment facility were included in the system boundary. 2.2. LCA methodology 2.2.1. Goal and scope definition The goal of this study was to determine and compare the EI of Mt. Simon brine pretreatment and three different management options. Four gate-to-grave LCA studies were within the scope of this study, including materials for construction of the main infrastructure, chemical inputs, and the energy required for each case. The thermal energy input (from natural gas) was specific to the United States. The electrical energy input was based on energy sources used for electrical energy generation in the state of Illinois in the United States. Other inputs were modeled by using global

422

H.H. Salih et al. / Water Research 122 (2017) 419e430

Table 1 Description of three cases for the handling or treatment of pretreated brine.a Case

Process description

1: DWD

Pretreated Mt. Simon brine is conveyed via pipeline to Class II injection wells, where it is injected without further treatment. The LCA explores the EI of drilling and construction of the injection wells and the energy consumption for transporting and injecting the pretreated brine. The impact of the injected brine on the receiving geological formation is not addressed in this study. Pretreated Mt. Simon brine is desalinated by an evaporation and crystallization process in which 88 wt% of water (from the water in the feed brine) is recovered. The concentrated reject brine is sent by pipeline to DWD and the by-product salt is transported by trucks to SLD. Purified water is sent by pipeline to a power plant to be used as cooling water. The LCA examines the EI of desalination of the Mt. Simon brine through the evaporation and crystallization process and disposal of the reject concentrated brine and salt by SLD and DWD in addition to the transport of purified water. This case is similar to Case 2 for water recovery, except that desalination by-products (i.e., the concentrated reject brine and dried salt crystals) are utilized for beneficial uses (e.g., road deicing). The LCA analyzes the EI of desalination of the pretreated Mt. Simon brine through the evaporation and crystallization process and assumes beneficial use of byproducts. The EI estimate includes energy consumption for drying the final salt by-product and transportation of the purified water and produced salt, but it does not include salt-drying equipment.

2: Near-ZLD, 88 wt% water recovery

3: Near-ZLD, 88 wt% water recovery with beneficial use of by-products

a

DWD, deep well disposal; LCA, life cycle assessment; EI, environmental impact; ZLD, zero-liquid discharge; SLD, solid waste disposal to landfill.

parameters in the GaBi databases, so this study may not be limited to one geographic location. For a consistent comparison basis among different brine-handling or treatment options, 1 m3 of the raw brine primary inlet stream was selected as the functional unit for all the cases.

2.2.2. Inventory analysis Table 2 lists the main inventory inputs for the pretreatment process, DWD, and the two desalination options. The inventory data in Table 2 are based on experimental data, modeling values, and bibliographic data obtained from the relevant literature. The library of processes in the GaBi Professional and Ecoinvent 3.2 (Frischknecht et al., 2005) databases were also used. Libraries in the databases contain all the related EI information (i.e., emission data, energy consumption, and material production data), and many of

the inputs and outputs identified in the life cycle inventory (LCI) could be linked to that information. Detailed input data for the four LCA studies are presented in Tables S1eS5 in the supplementary material. The engineering design of the pretreatment facility, including the volume of reinforced concrete, steel, and pipes along with the energy consumption (for pumping, mixing, and sludge removal) during the pretreatment process, was estimated based on data provided by the U.S. Environmental Protection Agency (Gumerman et al., 1979), as explained in detail in the supplementary material. The pretreatment design parameters were obtained experimentally in our laboratory by using a brine sample collected from the Mt. Simon Sandstone. An alum dose of 145 mg/L and a lime dose of 100 mg/L (to increase the pH from ~5 to ~7.5) were determined based on jar tests for reducing the turbidity of the extracted brine

Table 2 Life cycle inventory data of the pretreatment and three handling or treatment cases. The source of information for each item is included in the supplementary material.a Casea

Product flow

Material

Amount per m3 of raw brine

Uncertainty (%)

Unit

Pretreatment

Infrastructure

Concrete PVC pipe Steel Pump station Sand Lime Alum Truck distanceb Electricity Deep well (4800 ft) Pump station Electricity Pipeline Hot water tank model Pump station Deep well (4800 ft) Pump station Truck distanceb Thermal energy from natural gas Electricity Pipeline (water) Pipeline (concentrated brine) Hot water tank model Pump station Thermal energy from natural gas Electricity Truck distanceb Pipeline (water) Pipeline (concentrated brine)

9.36E-06 1.65E-03 5.35E-06 1.72E-07 6.71E-07 1.50E-01 1.00E-01 1.15E-03 3.07E-02 4.52E-08 1.63E-07 3.39Eþ00 7.23E-09 4.98E-05 1.63E-07 9.04E-09 3.61E-08 3.66E-02 1.73Eþ02 2.07Eþ00 7.23E-09 7.23E-09 4.98E-05 1.63E-07 2.00Eþ02 1.63Eþ00 3.66E-02 7.23E-09 7.23E-09

e e e e e ±25 ±25 ±50 ±50 e e ±50 ±50 e e e e ±50 ±50 ±50 ±50 ±50 e e ±50 ±50 ±50 ±50 ±50

metric ton kg metric ton pcs metric ton kg kg km kWh pcs pcs kWh km pcs pcs pcs pcs km kWh kWh km km pcs pcs kWh kWh km km km

Chemicals Energy/transportation Case 1: DWD

Infrastructure Energy/transportation

Case 2: Near-ZLD (disposal)

Infrastructure

Energy/transportation

Case 3: Near-ZLD (beneficial use)

Infrastructure Energy

a b

DWD, deep well disposal; ZLD, near-zero liquid discharge. Based on the fraction of a 27 metric ton truck occupied by an equivalent solid waste or salt amount corresponding to 1 m3 of brine for an 8 km trip.

H.H. Salih et al. / Water Research 122 (2017) 419e430

sample from 598 nephelometric turbidity units (NTU) to less than 5 NTU. The amount of sludge generated during the pretreatment process was estimated based on the brine total suspended solids (TSS) value of 2850 mg/L (measured in house) in addition to the added mass of alum and lime. The wet sludge (80 wt% solids) generated during the pretreatment process was sent to SLD with no additional drying. The energy required for brine injection in the DWD option was obtained from the literature (Coday et al., 2015). The thermal energy required for the evaporation and crystallization process and the electrical energy demand of the desalination process were estimated through a process simulation conducted by the Trimeric Corporation (Buda, TX) using Aspen Plus software (version 8.6). The estimated energy values were calculated using a TDS concentration of approximately 200 g/L for the Mt. Simon brine. For Case 2, the thermal energy for the evaporative equipment was estimated at 212 kWh/m3 of distilled water. In Case 3, the thermal energy consumption was expected to be slightly higher, ~246 kWh/m3, because of the additional energy demand for salt-drying equipment before the salt could be used as a valuable resource. The thermal energy was assumed to be generated from natural gas. The selected electrical energy source was the Illinois electricity mix. The electrical energy profile of the state of Illinois was obtained from the U.S. Energy Information Administration and consists of 50% nuclear, 38% coal, 5.6% natural gas, 5.5% wind, and 0.9% other sources (U.S. Energy Information Administration, 2017). The EI associated with disposal well drilling was estimated from the Ecoinvent 3.2 database of onshore well production. This data set covers the drilling operations of all types of onshore wells, including the energy uses, materials, land use, and emissions for well drilling and finishing. The EIs of brine transportation from the pretreatment unit to the disposal well and disposal by injection (pumping) into the appropriate geological formation were modeled using the pipeline transportation data set provided in the GaBi Professional database and the pump station construction data set available in Ecoinvent 3.2. The EI of the MEE (evaporation) equipment was assumed to be similar to that of the hot water tank data set provided in Ecoinvent. The hot water tank data set covers the production of a chrome steel tank, including a heat exchanger and boiler. For SLD in the pretreatment and Case 2, the inert waste/glass data set provided by GaBi Professional database was used. The inert waste data set includes main components of hauling, lining, compacting, and other components to control the leachate. The inert data set was selected because the solid waste materials sent for land filling are mainly inert inorganic materials. 2.2.3. Life cycle impact assessment methodology GaBi LCA software version 6.0, developed by Thinkstep (Leinfelden-Echterdingen, Germany), was used to determine the EI of eight impact categories for each LCA process and calculate a total EI score. GaBi software was also utilized for the interpretation of results, including data normalization and comparison. Environmental impacts for eight baseline impact categories, namely, the impacts on resourcesefossil fuels, global warming, smog potential, ozone depletion, human toxicityecancer, eutrophication potential, ecotoxicity potential, and acidification potential, were characterized with TRACI 2.1, the U.S. Environmental Protection Agency impact assessment methodology. The impact categories in the TRACI methodology are characterized at the midpoint level and were specifically developed for U.S. environmental conditions. Because each impact category has different units, further interpretation steps were needed to evaluate the results. Within each impact category, results were normalized and weighted, and the sum of the eight TRACI categories was used to determine an overall impact score. The “TRACI 2.1, USA 2008, including biogenic carbon (person

423

equivalents)” normalization factors and the “Thinkstep LCIA Survey 2012, North-America, TRACI 2.1, including biogenic carbon (person equiv. weighted)” weighting factors were applied with the software. It is worth mentioning that the normalization factors based on person equivalents constitute a validated method that is frequently used in the literature. However, weighting methods embedded in the LCA have some drawbacks that may result in an unavoidable bias (Reap et al., 2008). Normalization according to person equivalents may suffer from bias because of the embedded assumption, and inherent weightings may contain incomplete information that results in under- or overestimations of the EI (Heijungs et al., 2006). An additional impact category, water depletion, was obtained by  Conusing the ReCiPe methodology created by RIVM, CML, PRe sultants, Radboud Universiteit Nijmegen, and CE Delf (Nijmegen, Netherlands). ReCiPe was developed as a combination of the CML and Eco-indicator impact assessment methodologies; thus, it utilizes both the midpoint indicators and the endpoint indicators. 2.2.4. Uncertainty analyses Results of any LCA study are subject to uncertainty because of the uncertainties in LCI data that emerge from different sources, such as errors in measurements, incorrect estimation assumptions, and a lack of data. Furthermore, errors in characterization factors in the impact assessment phase of the LCA is another source of uncertainty. Major input parameters that can significantly affect the outcome of an LCA for the selected cases include the TDS and TSS contents of the brine, the chemical dosage, energy requirements of the desalination processes, and the transportation distance required to pump or haul waste materials to disposal sites or to transport treatment products for beneficial use. Chemical dosage in the pretreatment stage depends on TSS concentration, pH, and other properties of the brine. The TDS and TSS concentrations of the brine may vary by geographical location or depth. For instance, the TDS concentration of the Mt. Simon Sandstone varies based on sampling depth and location (Labotka et al., 2015). Energy consumption depends on many factors, including the TDS concentration and desalination technology. In this study, an uncertainty analysis was conducted to assess the variations in the LCA outcomes. The uncertainty associated with our LCI was estimated using a statistical sampling method and by parameter variation or scenario analysis (Heijungs and Huijbregts, 2004). Statistical sampling was conducted by using the Monte Carlo analysis in Microsoft Excel 2016, assuming continuous uniform distribution for the probability distributions of uncertainty. The Monte Carlo analysis calculates uncertainty by the random variation of uncertain parameters thousands of times. For each run, a random value in the specified uncertainty range (Table 2) is generated for each selected parameter in the inventory. The same uncertainty parameters (inputs) were adjusted for the parameter variation or scenario analysis of the Mt. Simon brine-handling or brine-treatment options. For the pretreatment, we considered a sensitivity analysis in which the impact of transportation distance of the sludge was varied by changing the quantity of sludge generated and the transportation distance. We varied the TSS concentration by ±25% from the measured baseline value of 2850 mg/L, and we varied the transportation distance to the landfill by ±50%. The amounts of chemicals (i.e., alum and lime) for pretreatment were also varied by ±25%. For Cases 2 and 3, in which energy consumption and the TDS concentration are the main sources of uncertainty, the TDS concentration and the thermal energy required for the evaporation and crystallization process were varied by ±50%. Disposal distance was also varied by ±50%. In addition, electrical energy consumption was varied by ±50% for all cases (Table 2).

424

H.H. Salih et al. / Water Research 122 (2017) 419e430

The composition of the electricity mix has a direct impact on the emission of greenhouse gases and the overall EI. Therefore, a sensitivity analysis was also conducted by using the U.S. electricity grid mix along with those of selected countries (i.e., Australia, Brazil, Germany, Spain, Finland, and India). The selected countries were chosen based on their geographical zone, main sources for electrical power generation, and availability in the GaBi Professional database. 3. Results and discussion The main objective of this study was to evaluate and compare the EIs of selected handling or treatment options for the high-TDS brine extracted from the Mt. Simon Sandstone. A comparative LCA was conducted in a case-by-case manner, with similar system boundaries for each case, as shown in Fig. 2 by the four dashed boxes. The pretreatment section was analyzed separately because it represents the first stage of all handling and treatment options. Life cycle assessment results for the pretreatment, DWD, and two treatment options for handling the extracted pretreated brine were compared side-by-side based on their overall EI scores (Fig. 3a). In addition, the contribution of each impact category for each case was evaluated (Fig. 3b). Similarly, for the pretreatment process and each handling or treatment case, a more detailed analysis of the LCA results was performed (Figs. 4 and 5), as discussed below. A higher impact score was indicative of a greater impact on the environment; therefore, low impact scores were desirable. More detailed LCA results are provided in the supplementary material (Tables S6eS16). 3.1. Pretreatment Extracted brine requires pretreatment before disposal into deep wells. The properties of disposed brine, such as TDS, composition, and TSS, can determine the injection well half-life by reducing the injectivity of the injection well (Barkman and Davidson, 1972; Eylander, 1988). Depending on the formation porosity, the presence of TSS, even at a low concentration of 5 ppm, may induce plugging of the injection well in a relatively short time (Pang and Sharma, 1994). Pretreatment is also required as the first stage for desalination options. One challenge associated with the extracted Mt. Simon brine is its high levels of TDS (~200,000 ppm) and TSS (2850 ppm). The pretreatment process included coagulation, flocculation, sedimentation, filtration, and sludge disposal (landfill)

units. The overall EI of pretreatment was broken down by these units (Fig. 4a1). The coagulation unit accounted for 95% of the overall EI of the pretreatment process. Sand filtration and the sludge disposal by SLD were responsible for ~3% and ~2% of the overall EI, respectively, whereas the EIs of flocculation and clarification were negligible (Fig. 4a1). The contribution of each impact category to the overall EI of the pretreatment process was as follows: 85% human toxicity potential, 9% ecotoxicity potential, 2% eutrophication potential, 1.5% global warming potential, and ~1% each related to impact on resourcesefossil fuels, smog, and acidification potential (Fig. 3b). The high impacts on human toxicity and ecotoxicity potential were mainly related to the coagulation unit. These results were mainly attributable to chemical production of the alum and lime additives that were used in the coagulation process. The production of lime, for instance, is known to generate a large volume of emissions, and during alum production, various pollutants are generated. Studies by Vince et al. (2008) and Zhou et al. (2014) also confirmed the high EI of the chemicals used in water treatment operations. For the sand filtration unit, the main parameter affecting the overall EI was the high volume of concrete required for construction. The amount of concrete required to build different sections of the filtration unit was 32 times greater than the amount needed for construction of the coagulation basin (Tables S1 in the supplementary material). The contribution of each process unit of the pretreatment to each EI category is shown in Fig. 4a2. Coagulation accounted for at least 91% of the EIs from ozone depletion, human toxicity, eutrophication, and ecotoxicity, as well as 58%e73% of the EIs from resourcesefossil fuels, smog, and acidification. Landfill disposal accounted for 37% of the EI from smog, 32% of the EI from resourcesefossil fuels and 6%e21% of the EIs from global warming potential, eutrophication, and acidification. Other components of the pretreatment process (diesel fuel, transportation, sand filtration, flocculation, and clarification) accounted for 5% or less of the EI in each impact category. 3.2. Disposal of extracted brine by deep well disposal Deep well injection is considered the most common means of disposing of hydraulic fracturing flowback and produced water (Green et al., 2016). The feasibility of deep well injection depends primarily on the geological formation of the disposal site. An injection well must consist of a permeable injection zone to hold the liquid waste confined by impermeable rock layers above to prevent contamination of freshwater resources or the environment.

Fig. 3. A side-by-side comparison of the environmental impact (EI) of the pretreatment and three handling or treatment options for Mt. Simon brine: (a) overall EI scores, (b) contribution of each EI category to each process unit. RFF, resources-fossil fuels; GW, global warming; SP, smog potential; OD, ozone depletion; HTC, human toxicityecancer; EP, eutrophication potential; ETP, ecotoxicity potential; AP, acidification potential; near-ZLD, near-zero liquid discharge.

H.H. Salih et al. / Water Research 122 (2017) 419e430

425

Fig. 4. Environmental impact (EI) of pretreatment and deep well disposal (Case 1): (a1) overall EI scores of the pretreatment units; (a2) categorized contribution of each EI category to each process unit of the pretreatment; (b1) overall EI scores of deep well disposal units; (b2) categorized contribution of each EI category to each process unit of the deep well disposal. RFF, resourcesefossil fuels; GW, global warming; SP, smog potential; OD, ozone depletion; HTC, human toxicityecancer; EP, eutrophication potential; ETP, ecotoxicity potential; AP, acidification potential.

Pretreated Mt. Simon brine was assumed to be disposed of into the Potosi Formation in a manner similar to the disposal of brine wastewater generated from the oil and gas production industry (Brower et al., 1989), and disposal included pipeline transportation to the wellhead followed by injection at an appropriate depth. Major components that affected the overall EI of this process included well construction for brine disposal and electrical energy for brine injection and pump station construction; however, the EI of brine transportation was relatively small (Fig. 4b1). Our LCA results indicated that the EI of the energy for pipeline transportation was negligible (0.2% of the overall EI); therefore, it is not discussed further. The largest portion (i.e., 56%) of the overall EI was associated with construction of the disposal wells (Fig. 4b1). The drilling and construction of disposal wells consists of several activities that could result in the release of various pollutants during the manufacture of materials (e.g., cement, chemicals) used for well construction. The electrical energy consumed for brine injection was responsible for 20% of the overall EI. The rest of the overall EI (24%) was associated with the pump station infrastructure. The overall EI of the DWD was categorized as follows: 63% was attributed to human toxicity; 11% to ecotoxicity; 9% to an impact on global warming; and 2%e6% each to resourcesefossil fuels, smog, eutrophication, and acidification (Fig. 3b). The contribution of each process unit to each EI category is shown in Fig. 4b2. Approximately 70% of the EI of human toxicity was associated with well production. The well infrastructure was also associated with 83% of the EI from ozone depletion; 37%e51%

of the EI was from smog, resourcesefossil fuels, eutrophication, and ecotoxicity; and 8%e17% was from the remaining impact categories, including global warming and acidification. Pump station construction contributed significantly to the impacts from ecotoxicity (46%), human toxicity (29%), eutrophication (21%), and ozone depletion (9%). Electrical energy was responsible for more than 53% of the resourcesefossil fuel, global warming, smog potential, and acidification impact categories. The impact of the injected brine on the receiving geological formation was not considered in this study because the required inventory data were lacking. However, this impact cannot be ignored. In spite of its long history as a liquid waste disposal method, DWD might be a source of concern worldwide because it could impose threats to the environment in general and to groundwater aquifers in particular if the impermeable rock layers above the injection zone fail. Furthermore, the fate of injected brine and its interaction with the fluid or rock in the disposal zone were not evaluated. 3.3. Desalination of pretreated brine The EI score and categorized impact of the two near-ZLD desalination options with 88% water recovery are presented in Fig. 5. In Case 2, in which the desalination by-products were disposed of, the highest impact factor was associated with thermal energy (83%; Fig. 5a1) followed by salt disposal by land filling (~7%). Brine pumping and the evaporation equipment were associated with only ~2 and ~6% of the overall EI of Case 2, respectively. The EI of

426

H.H. Salih et al. / Water Research 122 (2017) 419e430

Fig. 5. Environmental impact (EI) of two cases for near-zero liquid discharge (ZLD) desalination of pretreated Mt. Simon brine for 88 wt% water recovery (Case 2: disposal of desalination by-products; Case 3: beneficial use of by-products): (a1) overall EI scores of Case 2 units; (a2) contribution of each EI category to each process unit of Case 2; (b1) overall EI scores of Case 3 units; (b2) contribution of each EI category to each process unit of Case 3. RFF, resourcesefossil fuels; GW, global warming; SP, smog potential; OD, ozone depletion; HTC, human toxicityecancer; EP, eutrophication potential; ETP, ecotoxicity potential; AP, acidification potential.

land filling of inert salt materials is significantly lower than the EI of thermal energy used for the desalination process. However, landfills, if not properly managed, can create significant environmental and human health adverse impacts. Landfills can potentially affect the surrounding environment through the leaching or emission of pollutants to the air, soil, and water. Additionally, land use by the landfill infrastructure occupies ecologically productive land and may lead to reductions in biodiversity and biotic production, adversely affecting soil quality. In Case 3, which required no waste disposal and assumed beneficial use of salt crystals and concentrated brine by-products, the EI of the thermal energy is more pronounced, which was associated with 92% of the overall EI of the near-ZLD case (Fig. 5b1). The thermal energy required for evaporation of the recovered water for Cases 2 and 3 was the same, but Case 3 required additional energy for drying the salt by-product. The EIs of the evaporation equipment infrastructure for both cases were similar. The impact of the evaporation equipment (modeled by the hot water tanks) was ~5.8% of the total EI for Case 2 and approximately 5.6% of the total EI for Case 3. The thermal evaporation process requires a relatively low amount of electrical energy; therefore, the impact of electricity use was small, less than 1% in each case. Results also showed that the impact of DWD of concentrated brine in Case 2 was insignificant (less than 1% of the total) because only ~15 wt% of the pretreated brine was disposed of by DWD. Case 3 had no DWD of brine because every by-product of desalination was recovered as a valuable resource. The overall EI of Case 2, fell into mainly two EI categories, with 52% as resourcesefossil fuels and 22% as global warming potential.

Smog, human toxicity, eutrophication, ecotoxicity, and acidification were each associated with 1%e11% of the total EI (Fig. 3b). Case 3, without landfill disposal, exhibited a similar distribution: 56% of the EI was from resourcesefossil fuels, 24% from global warming potential, and less than 10% each from other impact categories (Fig. 3b). The high impacts on resourcesefossil fuels and global warming were due to the high energy demand of fossil fuels to provide thermal energy for the evaporation and drying of salt (Fig. 3b). In Case 2, the thermal energy was responsible for 97% of the impact on resourcesefossil fuels, 94% of the impact on global warming, and 78% of the impact on smog (Fig. 5a2). Thermal energy was also responsible for 48%e68% of the impact on ecotoxicity, eutrophication and acidification. In Case 3, the thermal energy used for desalination and salt drying was responsible for more than 97%e99% of the impact on resourcesefossil fuels, global warming potential, and smog potential, as well as 59% of the ecotoxicity potential and ~83%e91% of the eutrophication and acidification impact categories (Fig. 5b2). The impact on human toxicity was primarily due to production of the evaporation equipment (68% of the human toxicity EI; Fig. 5b2). According to the overall comparison of different handling or treatment methods shown in Fig. 3a, the near-ZLD process created the greatest EI (Cases 2 and 3), whereas disposal of the pretreated extracted brine by DWD had the lowest EI (Case 1). Deep well disposal of the extracted pretreated brine could be seen as the best among all options when the goal was to dispose of the brine waste. However, the DWD option would result in the largest brine discharge (~8 MGD) to the environment compared with the

H.H. Salih et al. / Water Research 122 (2017) 419e430

treatment options. Case 2 requires only 1.36 MGD of reject brine to be disposed of by DWD, whereas about 4000 tons per day of salts are sent for landfilling. Moreover, brine disposal by deep well injection is controlled or affected by several factors, including brine volume, brine quality, geological specifications, geographical location, and environmental regulations, in addition to public acceptance and capital or operating costs. Thus, disposal of the extracted brine may not be an option for many CO2 sequestration sites, and treatment for production of usable water may be a more desirable option. The overall amount of salt released to the environment (i.e., by landfilling or DWD) was the same for both Cases 1 and 2. Assuming that the desalination by-products could be utilized for beneficial uses, the near-ZLD process might represent a better option, especially if purified water is needed. It should be noted that freshwater production from the highly saline Mt. Simon brine using the conventional evaporators required a high thermal energy of 212 kWh/m3 of distilled water, compared with the significantly lower energy of seawater desalination (about 3e4 kWh of electrical energy for RO or 41.67 to 61.11 kWh of thermal energy for thermal desalination per cubic meter of freshwater produced) (Mezher et al., 2011). The high energy requirement for desalination of highly saline brine resulted in EI scores for Mt. Simon brine treatment (Fig. 5) that were significantly higher than the EI of conventional desalination processes reported in other studies (e.g., Coday et al., 2015; Raluy et al., 2005a; Tarnacki et al., 2012). 3.4. Depletion of water resources Depending on the life cycle of different scenarios, each case might have a different impact on the depletion of freshwater resources that are required mainly for electricity generation but also for fuel and material production for the processes. Water depletion analysis is the evaluation of freshwater use and consumption, both directly and indirectly, over the full life cycle of a product or process and is indicative of the overall freshwater use. Water consumption also includes the volumes of water required to dilute the polluted water to dischargeable levels. The ReCiPe method assigns a water depletion score of approximately 0.35 m3 for every kilowatt-hour of electricity consumed, based on the Illinois electricity mix data. The majority of water withdrawal for thermoelectric power production is used for cooling water. After exiting the power plant, the cooling water is discharged at a higher temperature and might be harmful to some aquatic species because of its relatively high temperature and low dissolved oxygen content. The negative environmental impact of chemicals and contaminants added to the water in the

Fig. 6. The impact of each handling or treatment option for Mt. Simon brine on the depletion of freshwater resources, based on the GaBi estimation. Water depletion was quantified as cubic meters of freshwater depleted per cubic meter of Mt. Simon brine that is disposed of or treated. Near-ZLD, near-zero liquid discharge.

427

power plant cooling process might also be significant. Fig. 6 presents the water depletion values based on the GaBi software estimation. The GaBi results suggest that in Cases 2 and 3, all the selected options will result in overall freshwater consumption even when distilled water is produced. In most scenarios, the majority of the water depletion predicted by GaBi was due to water used for electricity generation, deep well injection, evaporation equipment production, or landfill disposal. A closer examination revealed that 82% of the contribution to water depletion from deep well injection (Case 1) was due to electrical energy for the pipeline and injection. Case 3 shows the lowest water depletion impact, mainly because desalination by-products are utilized and solid and liquid waste disposal is eliminated. 3.5. Uncertainty analysis The issue of uncertainty in the LCA of desalination processes is commonly overlooked in the published literature (Zhou et al., 2014). Major input parameters that can significantly affect the outcome of our LCA study for selected scenarios include the TDS and TSS contents of the brine, the chemical dosage, the energy requirements of desalination processes, and the transportation distance required to pump pretreated brine to disposal wells or haul solid by-products to landfills in addition to the distribution distance of the purified water, concentrated brine by-product, and solid salt for beneficial uses. The measured TSS value was based on analysis of a brine sample collected from a new well; thus, this value might overestimate the TSS of the extracted brine after continuous operation of the well. The TSS content of the extracted brine might change depending on the extraction conditions and the maturity of the well; therefore, it was increased or decreased by 25% of the baseline value. The dose of chemicals for coagulation might also vary based on brine quality; therefore, it was varied by ±25%. The TDS value was varied by ±50% to include the salinity of other high-TDS brines that might be extracted from the same formation at a different depth or from other formations. The energy requirement of the desalination process was based on the simulation results of conventional evaporators and might vary depending on the desalination technology used; therefore, it was increased or decreased by 50% of the baseline value. The transportation distance for the liquid or solid waste materials and the purified water and treatment by-products was also varied by ±50%. The Monte Carlo analysis results are shown as 5th and 95th percentiles in the error bars of EI values in Figs. 3e5. A relatively low variation of 9% from the baseline EI was observed for Case 1, whereas Case 3 had the highest EI variability, at 41%. The pretreatment and Case 2 variations from the baseline EI were 18% and 37%, respectively. The higher observed variations for Cases 2 and 3 were mainly due to the impact of energy variation for these cases. Therefore, variation in energy consumption of the desalination process had the most pronounced impact on the LCA results. Thermal energy from natural gas combustion was the dominant source of energy for brine desalination in Cases 2 and 3. Results indicated that the change in energy consumption could result in EIs that might manifest mostly as impacts on global warming and resourcesefossil fuels. Thus, utilization of renewable energy sources (e.g., solar energy) would greatly reduce the overall EI. The composition of the electricity mix has a direct impact on the emission of greenhouse gases; therefore, a sensitivity analysis was also conducted using an electricity grid mix for the United States, along with other selected countries (Table 3). Significant variation was found in the source of electricity generation for the listed countries. For example, more than 78% of electrical power in India is generated from fossil fuels, whereas hydropower accounts for more than 80% of the electricity generation in Brazil. For these

428

H.H. Salih et al. / Water Research 122 (2017) 419e430

Table 3 Source of electrical power generation in selected countries based on information provided in the GaBi Professional database. Power source

Australia

Brazil

Germany

Spain

Finland

India

U.S.

Biomass Hard coal Hydropower Lignite Natural gas Nuclear power Wind power Photovoltaics Others

0.43 46.55 6.65 21.89 19.66 e 2.30 0.34 2.18

5.99 0.15 80.63 1.08 4.72 2.95 0.51 e 3.97

1.90 18.56 3.88 24.78 13.80 17.82 8.07 3.19 8.00

1.01 13.66 11.30 1.36 29.03 19.84 14.57 2.98 6.25

14.78 13.36 17.00 0.01 12.91 31.68 0.66 0.01 9.59

2.73 61.19 12.42 6.63 10.31 3.16 2.28 e 1.28

0.98 42.51 6.81 2.00 23.45 20.24 1.86 0.02 2.13

countries, we reevaluated the EI of the pretreatment and three handling or treatment cases by considering their electrical mix source to partially evaluate the transferability of the results of this work to other countries. Fig. 7 presents a comparison of the total EI of the pretreatment and three handling or treatment cases for the selected countries. The electricity consumption in Case 1 was the highest among the pretreatment and three other cases. This result was due to electricity being used to pump all the pretreated brine into the disposal wells in Case 1, whereas less electrical energy was used for the pretreatment and treatment cases (i.e., Cases 2 and 3). As a result, the most significant difference in the EI values for different countries was observed for Case 1, in which India had the highest EI because it had the highest use of fossil fuels for electrical power generation. It should be noted that the thermal energy source for all LCA cases was the same (natural gas in the United States). The thermal energy source had a high impact on the overall EI values of Cases 2 and 3; this source required large amounts of energy for water evaporation in the desalination process. A dramatic difference in EI values of different countries for Cases 2 and 3 was expected if the source of thermal energy was changed from natural gas to fuels with a higher EI (e.g., coal) or to renewable heat sources (i.e., solar or geothermal) that had significantly lower EIs. 4. Conclusions Large-scale industrial CO2 sequestration in deep saline formations may require the extraction of large volumes of highly saline brine that must be safely disposed of or properly treated. Through a

comparative LCA study, we investigated the EI of different options for the disposal or treatment of Mt. Simon brine from a potential CO2 geological sequestration site. Extracted brine was assumed to first be pretreated for TSS removal and then disposed of by DWD (Case 1), or treated by a near-ZLD treatment followed by disposal of the salt and brine by-products (Case 2), or treated by a near-ZLD treatment assuming beneficial use of the treatment by-products (Case 3). A comprehensive LCI was built that contained the main design and operation parameters, such as input materials along with chemical usage and energy consumption values. The main findings of the study are as follows:  The pretreatment process may generate a higher overall EI than the DWD (Case 1). This is mainly due to the impacts of chemical usage, sludge disposal, and infrastructure construction. Furthermore, the coagulation process accounts for the largest portion of the overall EI of the pretreatment process.  Disposal of the pretreated brine by DWD (Case 1) may result in the smallest EI compared with other investigated treatment options if the EI of brine injection into deep formations is not included. Because of the high energy consumption, desalination of the pretreated brine (Cases 2 and 3) results in the highest EI. The overall EI of the desalination cases falls into mainly two EI categories: global warming potential and resourcesefossil fuels.  The impact of the injected brine on the receiving geological formation was not considered because the required inventory data and characterization factors are lacking. However, this impact cannot be ignored because of environmental concerns (e.g., contamination of freshwater resources or soil contamination) in the event that the impermeable rock layers above the injection zone fail.  The water depletion analyses for Cases 1 to 3 suggest that for different disposal or treatment cases, 0.6e1.8 m3 of freshwater may be used for every 1.0 m3 of raw brine input. For Cases 2 and 3, in spite of producing ~0.8 m3 of purified water from each 1 m3 of raw brine, no global net water gain is attained. The overall freshwater consumption balance is at the lowest level (i.e., 0.6 m3) when brine is desalinated and desalination by-products are utilized for beneficial uses (i.e., Case 3).  In a sensitivity analysis, the values of TSS and chemicals were varied by ±25%, and values of TDS, transportation distance, and energy use were varied by ±50%. Compared with the baseline

Fig. 7. Effect of the change in electrical power source in selected countries on the overall environmental impact.

H.H. Salih et al. / Water Research 122 (2017) 419e430

values, these variations resulted in EI variations of 9%e41%. Cases 2 and 3 had the highest EI variability of 37% and 41%, respectively, mainly because of the high impact of energy variation.  To investigate the global transferability of these results to other countries, a sensitivity analysis was also conducted using a U.S. electricity grid mix and compared with other selected countries with a large variation in their electrical power source. The electricity consumption in Case 1 was the highest among the pretreatment and two other cases. As a result, the most significant difference in the EI values for different countries was observed for Case 1, which required the highest amount of electricity for brine pumping and injection. The thermal energy source for all LCA cases was assumed to be the same (natural gas in the United States). A dramatic difference in EI values in different countries for Cases 2 and 3 is to be expected if the source of thermal energy is changed from natural gas to fuels with higher EI (e.g., coal) or to renewable heat sources (i.e., solar or geothermal) that have significantly lower EI.  The LCA methodology and software are continually evolving. Available data sets were not specifically tailored to our study, so some challenges and uncertainties were inherent when assessing the exact EI of each process. For example, (1) the infrastructure for the evaporation equipment was modeled with simple hot water tanks, (2) the available landfill data set might not be the ideal predictive model for the disposal of salt in Case 2, and (3) the EI of brine release into deep geological formations was not included because no data were available.

Acknowledgments This research was funded by the National Energy Technology Laboratory of the U.S. Department of Energy under Cooperative Agreement DE-FE0026136. Any opinions expressed in this paper are those of the authors and do not necessarily reflect the views of the funding agency; therefore, no official endorsement should be inferred. Any mention of trade names or commercial products does not constitute endorsement or recommendation for use.

List of Abbreviations AP CCS CO2 DWD EI EP ETP GW HTC LCA LCI MEE MGD NTU OD RFF RO SLD SP TDS TSS ZLD

acidification potential carbon capture and storage carbon dioxide deep well disposal environmental impact eutrophication potential ecotoxicity potential global warming human toxicityecancer life cycle assessment life cycle inventory multiple-effect evaporation million gallons per day nephelometric turbidity unit ozone depletion resourcesefossil fuels reverse osmosis solid waste disposal to landfill smog potential total dissolved solids total suspended solids zero-liquid discharge

429

Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.watres.2017.06.032. References Bachu, S., Adams, J.J., 2003. Sequestration of CO2 in geological media in response to climate change: capacity of deep saline aquifers to sequester CO2 in solution. Energy Convers. Manag. 44, 3151e3175. Barkman, J.H., Davidson, D.H., 1972. Measuring water quality and predicting well impairment. J. Pet. Technol. 24, 865e873. Birkholzer, J.T., Zhou, Q., 2009. Basin-scale hydrogeologic impacts of CO2 storage: capacity and regulatory implications. Int. J. Greenh. Gas. Control 3, 745e756. Brower, R., Visocky, A., Krapac, I., Hensel, B., Peyton, G., Nealon, J., Guthrie, M., 1989. Evaluation of Underground Injection of Industrial Waste in Illinois. Final Report for ENR Contracts AD-94 and UI-8501. Illinois State Geological Survey, Champaign. Buscheck, T.A., Sun, Y., Hao, Y., Wolery, T.J., Bourcier, W., Tompson, A.F.B., Jones, E.D., Friedmann, S.J., Aines, R.D., 2011. Combining brine extraction, desalination, and residual-brine reinjection with CO2 storage in saline formations: implications for pressure management, capacity, and risk mitigation. Energy Procedia 4, 4283e4290. Coday, B.D., Miller-Robbie, L., Beaudry, E.G., Munakata-Marr, J., Cath, T.Y., 2015. Life cycle and economic assessments of engineered osmosis and osmotic dilution for desalination of Haynesville shale pit water. Desalination 369, 188e200. Eylander, J.G.R., 1988. Suspended solids specifications for water injection from coreflood tests. Soc. Pet. Eng. Reserv. Eng. 3, 1287e1294. ndez-Torres, M.J., Randall, D.G., Melamu, R., von Blottnitz, H., 2012. Ferna A comparative life cycle assessment of eutectic freeze crystallisation and evaporative crystallisation for the treatment of saline wastewater. Desalination 306, 17e23. Finley, R.J., 2014. An overview of the Illinois basin - Decatur project. Greenh. Gas. Sci. Technol. 4, 571e579. Frischknecht, R., Jungbluth, N., Althaus, H., Doka, G., Dones, R., Heck, T., Hellweg, S., Hischier, R., Nemecek, T., Rebitzer, G., Spielmann, M., 2005. The ecoinvent database: overview and methodological framework. Int. J. Life Cycle Assess. 10, 3e9. Green, R., Bertetti, F.P., Toll, N., Hill, N., 2016. Environmental impact of deep well disposal of liquid wastes from hydraulic fracturing. In: GSA Annual Meeting. Denver, Colorado, September 25e28. Gumerman, R., Culp, R., Hansen, S., 1979. Estimating Water Treatment Costs. Volume 2: Cost Curves Applicable to 1 to 200 mgd Treatment Plants. U.S. Environmental Protection Agency, Cincinnati, Ohio. Heijungs, R., Huijbregts, M.A., 2004. A review of approaches to treat uncertainty in LCA. In: Proceedings of the 2nd Biennial Meeting of IEMSs. Complexity and Integrated Resources Management, Osnabrück, Germany, pp. 332e339. June 14e17. e, J., Kleijn, R., Rovers, V., 2006. Bias in normalization: causes, Heijungs, R., Guine consequences, detection and remedies. Int. J. Life Cycle Assess. 12, 211. Kaplan, R., Mamrosh, D., Salih, H.H., Dastgheib, S.A., 2017. Assessment of desalination technologies for treatment of a highly saline brine from a potential CO2 storage site. Desalination 404, 87e101. Labotka, D.M., Panno, S.V., Locke, R.A., Freiburg, J.T., 2015. Isotopic and geochemical characterization of fossil brines of the Cambrian Mt. Simon Sandstone and IrontoneGalesville formation from the Illinois basin, USA. Geochim. Cosmochim. Acta 165, 342e360. Leetaru, H.E., Freiburg, J.T., 2014. Litho-facies and reservoir characterization of the Mt Simon Sandstone at the Illinois basin - Decatur project. Greenh. Gas. Sci. Technol. 4, 580e595. Leetaru, H.E., Smith, V., Adushita, Y., Freiburg, J.T., 2014. An integrated approach to evaluating the suitability of the Potosi Dolomite as a carbon sequestration target. Interpretation 2. SF125eSF133. Lundie, S., Peters, G., Beavis, P., 2004. Life cycle assessment for sustainable metropolitan water systems planning. Environ. Sci. Technol. 38, 3465e3473. Mezher, T., Fath, H., Abbas, Z., Khaled, A., 2011. Techno-economic assessment and environmental impacts of desalination technologies. Desalination 266, 263e273. Morton, A.J., Callister, I.K., Wade, N.M., 1997. Environmental impacts of seawater distillation and reverse osmosis processes. Desalination 108, 1e10. Pang, S., Sharma, M.M., 1994. Model for predicting injectivity decline in water injection wells. In: Production Operations and Engineering. Society of Petroleum Engineers (SPE), Richardson, Texas, pp. 275e284. Raluy, R.G., Serra, L., Uche, J., 2005a. Life cycle assessment of water production technologiesdPart 1: life cycle assessment of different commercial desalination technologies (MSF, MED, RO). Int. J. Life Cycle Assess. 10, 285e293. Raluy, R.G., Serra, L., Uche, J., Valero, A., 2005b. Life cycle assessment of water production technologiesdPart 2: reverse osmosis desalination versus the Ebro River water transfer. Int. J. Life Cycle Assess. 10, 346e354. Raluy, R.G., Serra, L., Uche, J., 2006. Life cycle assessment of MSF, MED and RO desalination technologies. Energy 31, 2025e2036. Ras, C., Von Blottnitz, H., 2012. A comparative life cycle assessment of process water treatment technologies at the Secunda industrial complex, South Africa. Water

430

H.H. Salih et al. / Water Research 122 (2017) 419e430

SA. 38, 549e554. Reap, J., Roman, F., Duncan, S., Bras, B., 2008. A survey of unresolved problems in life cycle assessment. Int. J. Life Cycle Assess. 13, 374. Tarnacki, K., Meneses, M., Melin, T., van Medevoort, J., Jansen, A., 2012. Environmental assessment of desalination processes: reverse osmosis and Memstill®. Desalination 296, 69e80. U.S. Department of Energy, 2015. The United States 2012 carbon utilization and storage atlas, fourth ed. U.S. Department of Energy, National Energy Technology Laboratory, Office of Fossil Energy, Pittsburgh, Pennsylvania. U.S. Energy Information Administration, 2017. Form EIA-923, Power Plant Operations Report. https://www.eia.gov/electricity/state/Illinois/ (Accessed 30 March 2017).

ant, P., Marechal, F., 2008. LCA tool for the environmental Vince, F., Aoustin, E., Bre evaluation of potable water production. Desalination 220, 37e56. Zhou, J., Chang, V.W.C., Fane, A.G., 2011a. Environmental life cycle assessment of brackish water reverse osmosis desalination for different electricity production models. Energy Environ. Sci. 4 (6), 2267e2278. Zhou, J., Chang, V.W.C., Fane, A.G., 2011b. Environmental life cycle assessment of reverse osmosis desalination: the influence of different life cycle impact assessment methods on the characterization results. Desalination 283, 227e236. Zhou, J., Chang, V.W.C., Fane, A.G., 2014. Life cycle assessment for desalination: a review on methodology feasibility and reliability. Water Res. 61, 210e223.