Desalination techniques — A review of the opportunities for desalination in agriculture

DES-12457; No of Pages 15 Desalination xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Desalination journal homepage: www.elsevier.com/locate/desal

Desalination techniques — A review of the opportunities for desalination in agriculture Stewart Burn a,b,c,⁎, Manh Hoang a,b, Domingo Zarzo e, Frank Olewniak f, Elena Campos e, Brian Bolto a, Olga Barron d a

CSIRO Manufacturing, Clayton, Australia Victoria University, Werribee, Australia RMIT University, Melbourne, Australia d CSIRO Land & Water, Perth, Australia e Valoriza Water, Spain f Valoriza Water, Australia b c

H I G H L I G H T S • • • • •

Water from reverse osmosis/electrodialysis technologies is suitable for agriculture. Water from these technologies is more expensive than normal agricultural water. The use of desalination for agriculture may currently be justified for high value crops. Wider technology application is viable where fit for purpose water is limited. Technology development requires cost reduction and improved agricultural practices.

a r t i c l e

i n f o

Article history: Received 8 September 2014 Received in revised form 11 December 2014 Accepted 24 January 2015 Available online xxxx Keywords: Agriculture Irrigation Desalination Technologies Review

a b s t r a c t The adoption of desalination for agricultural purposes in countries such as Australia has been very limited, with only a small number of cases available to demonstrate its suitability. This can be compared to countries such as Spain where the uptake has been significant. A number of suitable technologies such as reverse osmosis and electrodialysis are available to provide desalinated water, but not at a cost comparable to that for water commonly utilised for agricultural purposes. The use of blended waters, where the quality of the water is tailored to the crop may go part way to addressing this cost differential. However, if the overall efficiency of the combined production of water and food, as well as opportunities for better soil management is considered, then desalination's applicability to agriculture becomes more viable. The use of state of the art technologies for the provision of desalinated water for agriculture is most likely to be cost effective in a tightly controlled environment, using agricultural practices with the most-effective water use and crops with high productivity. Such conditions are often associated with greenhouses and the production of high-value irrigated crops, where the cost of water is small compared to the infrastructure investment. Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.

Contents 1. 2. 3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . Integrated water and food production: desalination and agriculture . Desalination technologies . . . . . . . . . . . . . . . . . . . . 3.1. Established commercial technologies . . . . . . . . . . . 3.1.1. Reverse osmosis . . . . . . . . . . . . . . . . 3.1.2. Nanofiltration . . . . . . . . . . . . . . . . . 3.1.3. Electrodialysis . . . . . . . . . . . . . . . . . 3.1.4. Ion exchange resins . . . . . . . . . . . . . . . Other techniques applicable to desalination for agricultural purposes

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

. . . . . . . . .

⁎ Corresponding author at: CSIRO Manufacturing, Clayton, Australia. E-mail address: [email protected] (S. Burn).

http://dx.doi.org/10.1016/j.desal.2015.01.041 0011-9164/Crown Copyright © 2015 Published by Elsevier B.V. All rights reserved.

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

0 0 0 0 0 0 0 0 0

2

S. Burn et al. / Desalination xxx (2015) xxx–xxx

5.

Relevance of desalination technologies to agriculture 5.1. Feedwater quality . . . . . . . . . . . . . 5.2. Energy usage and costs of suitable technologies 5.3. Brine disposal . . . . . . . . . . . . . . . 6. Discussion . . . . . . . . . . . . . . . . . . . . 7. Conclusions . . . . . . . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

1. Introduction Population growth, food security concerns, climate change impacts on agriculture, freshwater resource overuse and land degradation worldwide are forcing international scientific communities to look for alternative approaches to our current resource management approach for agricultural purposes. This includes all aspects associated with water resources and their availability to support ever-growing demands for both agricultural and potable water demands. Desalination technologies may provide one opportunity for generating cost-effective and potentially climate-independent water resources of controlled quality for agriculture applications. As shown in Fig. 1, seawater desalination is the most used solution to address water shortage especially for potable water applications. In this respect the number of desalination plants around the world, both planned and under construction, has increased significantly in recent years, as shown in Fig. 2, especially in Australia where they have been targeted for providing additional sources of potable water [25]. It is estimated that about 69% of available water resources around the world are used for irrigation [110] and as water demands increase the number of desalination plants for irrigation for agriculture has also increased. Consequently there is increased emphasis on enabling cost effective desalination technologies to provide water of suitable quantity and quality for agricultural applications. Drier countries such as Australia and Spain have a long history with desalination technologies. In the past, the high capital and operating costs of desalination and the energy required have been major constraints to large-scale production of freshwater from brackish waters and seawater. However, desalinated water is becoming more competitive for urban use because desalinating costs are declining associated with increasing demand from population growth and reduced security of supply from surface water and usable groundwater and it is expected that these increases in efficiency will flow through to the agricultural sector. However, in spite of these developments, currently the cost of desalinated water is still too high for the use of this resource in broad-scale irrigated agriculture. An exception appears to be intensive horticulture for high-value cash crops, such as vegetables and flowers (mainly in greenhouses) grown in coastal areas where safe disposal of brines is easier than in inland areas [11]. For example Sundrop farms (Sundrop-Farms, Personal communication), uses 860,000 m 3 of fresh water yearly to irrigate 2000 m2 of greenhouses. If the costs for providing desalinated water continue to reduce, its use is expected to become more viable because desalination for agricultural purposes has a number of significant advantages including:

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

. . . . . . . .

2. Integrated water and food production: desalination and agriculture Desalination allows a widening in utilisation of available water resources by producing freshwater from saline or brackish natural water sources. Over the past decade conventional water production costs have been rising in many parts of the world and costs for desalination have been declining, consequently desalination has become more economically attractive and competitive. Lattemann et al. [58] estimated that by 2015 the costs of freshwater treatment, wastewater reuse and desalination are likely to be similar, at least in USA. However, currently desalinated water produced worldwide (77.4 million m3/day, IDA, 2012) still comprises less than 1% of total worldwide water use, with only 2% of total desalinated water production currently used for agriculture (Fig. 3). According to Desaldata [29] many countries are beginning to use desalinated water in agriculture, albeit at varying rates. The highest proportion of desalinated water use in agriculture occurs in Spain, where the current installed capacity is 1.4 million m3/day and 22% is used in agriculture for high value crops, such as vegetables, fruits including tomatoes and peppers, and vineyards for table grape production. In Kuwait, where the current installed capacity is in excess of 1 million m3/day, 13% is used for agriculture and in Saudi Arabia, the world's largest single producer of desalinated water; only 0.5% of its desalination capacity is used for agricultural purposes. Other countries which use desalinated water for food production are Italy (desalination capacity 64,700 m3/day — 1.5% for agriculture), Bahrain (620,000 m3/day — 0.4%), Qatar (0.1%), USA (1.3%) and Israel. The wider application of desalination technologies for agriculture is limited by its relatively higher cost, as well as by the need for agriculture to be close to saline and brackish feedwater resources as well as a safe and cost effective disposal option for brines. National assessments of the applicability of desalination technologies to support agricultural water supply are currently under way in Chile, China and Australia [49]. The overall efficiency of the combined production of water and food, energy use as well as an opportunity for better soil management, should be the basis for an assessment of desalination's applicability to

• • • • •

Tailored conductivity for irrigation water Assured supply Enables agricultural products of consistent quality Production may be increased compared to other water sources. The water may attain a higher resale price due to quality and supply assurance. • It allows recovery of saline soils by irrigation with high quality water.

0 0 0 0 0 0 0 0

Fig. 1. Total capacity installed in the world (IDA Desalination Yearbook 2013–2014).

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx

3

Fig. 2. Global cumulative contracted and commissioned desalination capacity, 1965–2011 [49,50].

agriculture. Relatively high desalination costs can be offset by the adaptation of more-efficient irrigation technologies (less water use per kg of crop production) and increases in agricultural productivity and crop quality (greater crop production per unit of water used). There are a number of benefits for desalinated water use in the agricultural sector. The most obvious one is that the technology produces an additional water resource. However, costs are a major limitation and less expensive options such as reverse osmosis are normally chosen, consequently, as shown in Fig. 4 the majority of the world's desalination capacity is supplied by RO [49]. Desalination plants used for agriculture are relatively small (apart from some multi-purpose plants in Spain and Israel) and according to ‘economy of scale’ principles they produce water at a high cost. For agricultural purposes, seawater desalination using RO, considered to be one of the most efficient desalination technologies, is not normally used as the process is expensive due to high energy demands. Brackish water desalting is typically a third of the cost of desalinating seawater; however, inland brackish water is more often associated with groundwater, which is not an infinitive resource (as seawater is considered) and a clear definition of a sustainable groundwater yield is required to avoid aquifer depletion. Inland desalination is also challenged with the required disposal of desalination by-products (brine) with all currently available options adding a significant cost to water production. However, higher desalinated water recovery rates available with new technologies such as

Fig. 3. Global desalination capacities by user type (IDA Desalination Yearbook 2013– 2014).

membrane distillation should lead to a reduction in disposal cost per unit of water produced. In addition, the proximity of feedwater sources to irrigated land can significantly reduce the cost of water supply as it minimises water distribution costs.

3. Desalination technologies There are a significant number of technologies available for desalination of both sea and brackish water supplies. Many of these are commercial, however; there are some that are either approaching commercialisation or are at an advanced stage of development such as membrane distillation; which are not assessed in this analysis. The choice of technology is influenced by the source water quality, energy demand, and most importantly the value placed upon the recovered water. For example the osmotic pressure for seawater of salinity 35,000 mg/L is 2800 kPa, versus 140 kPa for brackish water of salinity 1600 mg/L [55]. For reverse osmosis (RO) this means that when using seawater feed a significantly greater pressure must be applied to prevent osmotic transfer of water through the semi-permeable membrane and thus the energy requirements for the treatment of seawater are significantly higher than for brackish water. Whilst grid energy is the major energy source used for desalination, other alternative energy sources exist including solar [38,97], wind or wave power [10,27,43] and heat to provide the power necessary to drive the desalination process.

Fig. 4. Total capacity installed by technology (IDA Desalination Yearbook 2013–2014).

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

4

S. Burn et al. / Desalination xxx (2015) xxx–xxx

The technology for solar desalination already exists and has recently been applied for the provision of water, energy and heat in South Australia for the production of high value crops [93], whilst a number of photovoltaic RO demonstration plants have also been developed [78]. Wind power with direct conversion to mechanical energy is also under study in the Netherlands [70]. Wave energy is another possible direct mechanical energy source [102], whilst a system that employs hydrostatic pressure has been tested [5]. Water quality is a critical factor in determining the types of technologies that are suitable for desalination of different waters and salt concentration in particular will determine which process is appropriate, as will fouling and scaling which is a function of the raw water composition. The main practical difficulties with technologies such as RO are associated with membrane deterioration, especially by biofouling and the scaling of the membranes which is more pronounced with intermittent operation. 3.1. Established commercial technologies 3.1.1. Reverse osmosis Desalination via RO removes all the naturally occurring salts to give an un-buffered water that is deficient in calcium and other essential minerals, so for drinking purposes these must be added back into the water. This deficiency in divalent salts combined with the presence of CO2, causes a low pH (around 5.5) and a negative Langelier Index (LSI), which implies a corrosive and poorly balanced water. For agriculture this means that the SAR (sodium adsorption ratio) is not balanced, which can cause de-structuring and waterproofing of soils due to sodium and calcium exchange. The importance of these issues for agricultural purposes will need to be addressed, but it is expected that buffering would address the issue. RO is also relatively inefficient as all the input water must be chemically pre-treated and filtered even though a large proportion of the input is returned to the ocean or source as a concentrated waste stream. An example of a typical value, is the Southern Seawater Desalination, Binningup, RO plant in Perth, where this brine is equivalent to 55–60% of the input water stream [45]. Also one of the main issues in desalination for agriculture is the toxicity of boron for different crops and the high boron transfer through RO membranes. This is especially problematic for seawater and the reason why in many plants a second pass RO is required. 3.1.1.1. Reverse osmosis — desalination costs. As the number of installed plants world-wide has increased, to more than 15,000 in 125 countries

[89], there has been an overall decrease in the production cost of desalinated water obtained by RO, as seen from a survey of 20 plants over the period 1990 to 2005 (Fig. 5). However, there has been a recent upward trend in costs due to construction and power costs increasing. Between 1980 and 2000 the amount of energy needed for seawater desalination was halved because of improvements in pumps and other equipment, and has been further halved with new energy recovery systems that regain 97% of the energy used [89]. Currently the cost of producing desalinated seawater is estimated to be as low as US$0.50/m3 for large scale seawater reverse osmosis plants at specific locations and conditions (which may include local incentives and subsidies), whilst at other locations the cost is closer to US$1.0/m3 [46]. In comparison Tofigh and Najafpour [98], are predicting prices as low as US$0.35/m3 for water obtained from reverse osmosis. The cost to treat seawater or brackish waters to produce potable water is a function of numerous variables which are difficult to ascertain precisely from the literature because of the confidential nature of such costs. 3.1.1.2. Reverse osmosis — energy costs. The major costs in desalination are related to energy which can represent between 30 and 50% of the operating costs. For seawater feeds the energy requirement is high at 12 kWh/m3 if there is no energy recovery, and 4 kWh/m3 if there is energy recovery [19]. For seawater about 0.9 to 1.4 kWh/m3 is consumed for deep sea pumping (about 6 m below sea-level), for filtration devices and micro filtration finishers and for the disposal of the concentrate to the sea. The RO process itself, consumes between 2.2 and 2.8 kWh/m3, depending on the type of concentrate energy recovery used. When energy recovery turbines (ERT) are used to recover the energy stored in the concentrate; the RO energy is as low as 2.8 kWh/m3 of product; whilst devices like pressure exchangers may reduce energy consumption to 2.2 kWh/m3. It should be noted that based on thermodynamic principals the minimum separation energy for seawater (35,000 mg/L TDS, 25 °C) is 1.09 kWh/m3. Continuing advances in membrane development mean that the energy component for operation is continually being improved with a record low of 1.58 kWh/m3 being claimed [89]. Recent survey data for Australia shows that the average energy consumption is 3–3.7 kWh/m3 for sea water RO, 0.7–1 kWh/m3 for brackish water and 1.2 kWh/m3 for industrial effluent [52–54]. The highly efficient Kwinana plant uses between 2.7 and 3.1 kWh/m3 of water depending on temperature and membrane ageing [45]. Obviously the cost of desalinated water will depend upon the cost of energy. In Australia the cost of electricity varies depending on location,

US$/kL

RO Desalination Water Cost 2.0 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0

Fig. 5. RO desalination water cost in the year of bid [52–54].

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx

retailer and customer. For example, energy costs a maximum of AU$0.30 kWh in Victoria for domestic customers, but for a large customer varies from AU$0.20 kWh for small installations to AU$0.12 kWh for large sites using in the vicinity of 150 MWh/yr (Personal Communication Yarra Valley Water). Therefore RO water using the most efficient systems should cost between AU0.36 and AU$0.60/m3 depending on the size of the system. (Note: at the time of submission AU$1 ≈ US$0.8.) One option to reduce energy costs is to operate desalination plants during off-peak periods. An example of such a strategy is utilised at the Cuevas de Almanzora RO plant in Spain. As shown in Table 1, there are 6 periods of electricity price (P1 to P6) depending on the day, on the week or even on the month available to this plant. The main strategy of the Cuevas de Almanzora Plant is to operate as much as possible in the P6 periods and at a minimal level in the P1 and P2 periods. Only during months of maximum water production has there been significant consumption of electricity in region P5 as shown in Fig. 6. The utilisation of such a practice has seen water being produced by the plant at a cost of AU$0.308/m3. If this water were then blended with for example groundwater, then water could easily be produced at a cost suitable for agricultural purposes [44]. Alternatives to grid power are being explored to reduce energy costs. A hybrid system based on gravitational and wind energy has been proposed [39]. In this system a windmill pumps feed water to special pistons inside a complex series of desalination columns. Calculations suggest energy usage of 2.8 kWh/m3 for such a system. Submerging RO plants in boreholes to a depth of 500 m and pumping the product water to the surface has also been suggested [83]. Another alternative is to use the pressure contained within artesian bores as these can have a pressure of up to 1.3 MPa [81], which is enough to overcome the osmotic pressure of many brackish waters. For example, the osmotic pressure for water of salinity 1600 mg/L is only 0.14 MPa [55]. Therefore, it should be feasible to run a brackish water RO plant on artesian bore water pressure. The installation would be at the bore head and would provide suitable water for agricultural purposes provided that the pressure is retained and there is no diminution of pressure over a long time period.

3.1.1.3. Reverse osmosis — operational costs. Pre-treatment RO water is essential to prevent fouling of membranes for at least half of the major RO seawater desalination plants installed around the world and this adds an additional cost especially if the water source is contaminated. Inorganic salts, colloidal and particulate matter, organic compounds and microorganisms present in the feed water reduce membrane efficiency and lifespan. They are treated by chemical dosing with coagulants, acids, disinfectants (chlorine based or other biocides), antiscalants and sodium bisulphite for oxidant removal all of which increase the cost of treatment. However, coagulation only removes some pollutants and can produce small flocculants that penetrate and block membrane pores. New coagulants formulated for a number of water sources aim to greatly improve flocculent size, capture more pollutants, reduce membrane fouling, and can be easily washed from the membrane and are thus expected to reduce treatment costs [99]. Technologies are also being developed to allow membrane surfaces to be treated with compounds that have excellent anti-fouling properties. Table 1 Average prices of electricity in 2011 and 2012 by tariff periods (prices in AU$). Period

2011

2012

P1 P2 P3 P4 P5 P6

0.185 0.155 0.128 0.105 0.097 0.077

0.190 0.160 0.130 0.105 0.096 0.077

5

Fig. 6. Evolution of electricity consumption by tariff periods along 2011 and 2012 [44].

As shown in Fig. 7 other costs also contribute to the total cost of desalination with RO currently being cheaper than other common desalination techniques such as Multiple Effect Distillation (MED) and Multistage Flash Distillation (MSD) [58,59]. As detailed in Fig. 7 most cost analysis concentrates on the costs associated with the production of water for potable applications. A major question is which particular type of desalination plant is appropriate to produce water for use in agriculture, because there are a number of factors that can increase the costs of desalinated water over the ex-plant cost and thus the profitability of the crops to be grown. These include the following [69]: • the available area to be cultivated • the distance from that area to the desalination plant • the existing infrastructure for water distribution.

The costs of desalination vary significantly depending on the size and type of the desalination plant, the source and quality of the incoming feed water, pre-treatment requirements, automation and control, the plant location, site conditions, qualified labour, energy costs and plant lifetime. Lower salinity feed water requires less power consumption and dosing of anti-scaling chemicals. Larger plant capacities reduce the unit cost of water due to economies of scale, whilst, lower energy costs and longer plant maintenance periods also reduce the unit cost of water [114]. However, they do also require large energy plants nearby and concentrate large volumes of brine that needs to be discharged. The operation costs of desalination can be grouped into the following areas [100]: • • • • • • • • •

intake pre-treatment treatment (i.e., reverse osmosis) remineralisation pumping of product water post-treatment brine disposal energy utilisation civil works.

For agricultural purposes away from the coast; energy, brine disposal and civil works costs would constitute the major costs for producing water. For coastal applications brine discharge to the ocean would still be the preferred pathway so the operational costs for this component would be minimal. 3.1.1.4. Reverse osmosis — installation and civil works costs. The costs associated with a desalination plant are usually expressed in two ways: the capital costs and total annual operating water costs per unit of

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

6

S. Burn et al. / Desalination xxx (2015) xxx–xxx

Relative Operation Costs 0.60 0.01

Relative Operation Costs (US$)

0.6

0.46 0.05

0.03

0.5

0.50 0.01

0.08

0.08

Parts

0.07

0.4

0.08

0.27

0.3

Chemicals Labour

0.1

Membranes

0.03 0.27

0.2

Thermal Energy

Electrical Energy 0.23

0.1

0.19

0.06

0

RO

MSF

MED

Fig. 7. Relative operation costs in US$ of the main desalination processes [58,59].

installed or process capacity. Table 2 presents construction and operating costs on this basis for RO, MED and EDR which have been updated for inflation from the 2002 dollars reported [3]. The distribution of these costs across different functions for a reverse osmosis plant can be seen in Fig. 8. It can be seen that energy and amortisation costs or return on investment costs comprise over three quarters of the total costs. Fig. 9 shows investment cost of seawater reverse osmosis (SWRO) plants up until 2005. As can be seen; after approximately 1995 the cost is clustered between 500 and 1000 US$ per m3/day capacity [84] with costs decreasing per unit capacity over time, which is associated with efficiencies as plant sizes increase as seen in Fig. 10 [37,41]. These costs vary significantly to those reported by [3], however more recent costs analysis in 2009 details costs above AU$5000 m3/day for Australian desalination plants as seen in Fig. 11 [51]. Because of lower osmotic pressures and thus lower operating pressures and higher yields, the cost of desalinating brackish water (BW) is considerably lower than the cost of seawater (SW) desalination, making it more suitable for agriculture. Fig. 12 shows the difference of capital cost between seawater and brackish water [107]. 3.1.2. Nanofiltration Nanofiltration (NF) is considered to be a most promising technique for the production of high quality water or a highly pre-treated feed water for RO, and many examples of its use exist, especially in the drinking water industry [9]. Impurities that are removed include dissolved solids such as inorganic ions, organic carbon, and regulated and unregulated organic compounds. NF membranes are mostly used for softening and the removal of organic compounds from surface and brackish water, and have received attention as a pre-treatment for seawater desalination. They are usually polyamide-based, thin-film composite structures, relatively close in chemical structure to RO membranes. Their pore size is 0.5–1.5 nm, ranging between that of Ultra Filtration

(UF) and RO membranes. The topic has been extensively covered in the literature [88]. When used for seawater desalination the application of a low pressure NF stage before RO, takes out the multivalent ions plus some sodium chloride and organics, leaving a feed for the following RO system that is of much lower ionic strength than the original raw water. The consequence is that there is a smaller osmotic pressure effect and therefore lower applied pressure needs, leading to lower energy use and higher water yield. Unfortunately, the total costs for an NF/RO system are usually about 10% more than for an RO only system. However, there are certain situations where the approach is justified on economic grounds, because of the organics removal by NF resulting in a marked decrease in RO membrane fouling and a significantly enhanced membrane life. A full scale plant in Saudi Arabia demonstrates the benefits that can be obtained via the use of NF [35]. The plant treats 8.6 ML/day, with the NF component • • • • •

reducing hardness from 7500 to 220 mg/L lowering TDS from 45,460 to 28,260 mg/L rejecting sulphate to N99% reducing divalent cations by 80–95% lowering the pressure used in the RO plant by 17%.

The NF stage operated at a 65% conversion rate and the RO stage at a 56% conversion rate, giving an overall conversion of 36%. This compared favourably with a parallel RO-only plant which had a conversion rate of 28%, thus showing a 30% increase in overall recovery for the NF/RO system, with an energy saving of 25–30% [94]. The application of NF in reuse of municipal wastewater for irrigation purposes has been explored recently, with the best performing membrane lowering the TDS from 3150 to 340 mg/L [18].

Table 2 Construction and operating costs for RO, MED and EDR. Parameter 3

Capital cost (AU$/m /day of product water) Operating cost (AU$/m3/day of product water)

Seawater RO

Brackish RO

MED

EDR

2130–3330 2.52–2.93

800–2400 0.87–2.00

3330–5200 With waste heat: 0.73–1.27 Without waste heat: 2.40–3734

760–4330 1.33–3.73

Note. For construction costs, only direct capital costs associated with process works, including pre-treatment and process treatment equipment, pumps, pipes and control systems have been incorporated, and not the costs for delivery of the water to and from the plant, or associated post treatment costs.

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx

Fig. 8. Distribution of construction and operation costs in a RO plant [1].

3.1.3. Electrodialysis Electrodialysis (ED) is a membrane separation process in which ions are separated through ion-exchange membranes under the influence of a potential gradient. In ED dissolved salts are transferred via a direct current electric field through ion-selective membranes arranged in multi-compartmented cell [91,95]. By applying a potential gradient across the electrodes, cationic species (Na+, K+, NH+ 4 ) tend to move towards the cathode passing through cation-exchange membranes (CEM), which allows only positive species to pass through whilst rejecting the negatively charged species. Conversely, anion species 3− (Cl−, SO2− 4 , PO4 ) move towards the anode passing through anionexchange membranes (AEM), which allow only negative species to pass through and reject ions of the positive species. Through this process, cations and anions are obtained separately in the concentrated solution. At an industrial scale, one stack of ED cells contains up to several hundred pairs of AEM and CEM arranged alternately between electrodes (Fig. 13). This technology whilst suitable for wastewater in that it removes and concentrates nutrients such as nitrogen, potassium, and phosphorus is also suitable for the treatment of brackish waters for salt removal. ED with polarity reversal (EDR) is employed to remove membrane foulants which, depending on the feed composition, are typically calcium and magnesium carbonates, sulphates or phosphates. There is periodic reversal of the direction of current through the membrane stack to maximise performance and reduce scaling and fouling [56,91]. EDR enables the brine stream to be operated under conditions of super-saturation with respect to solubility-limiting species like calcium carbonate and sulphate [71]. The process is usually limited to low salinity feeds with TDS up to 3000 mg/L, which gives high recovery (~85%), can cope with suspended solids and uses less chemicals. It also has a lower capital cost at 637 US$/m3/day, versus 925–2100 US$/m3/day for RO [40]. However, it is less flexible than RO, especially on feed water salinity. EDR works by progressive removal of ions, typically 50% per stage. More stages are added if further removal is required. Its main application is in brackish waters for which one or two stages are usually

Fig. 9. Investment cost of SWRO plant per m3/day capacity over years [84].

7

sufficient to produce potable water and the energy required is relatively low, being roughly proportional to the total dissolved solids in the feed. Because it does not involve filtration, EDR is more tolerant of feed water quality with respect to suspended material as indicated by silt density index and turbidity. On the down side there is no removal of pathogens, although the capacity to operate continuously with a free chlorine residual of 0.5 mg/L partially addresses this issue. In a pilot-plant trial recently performed by CSIRO at the Western Treatment Plant in Victoria [47,96], treated waste water was subjected to two-stage EDR after flocculation and media filtration. Using source water with an electrical conductivity (EC) of 2 dS/m, product at 1 dS/m was produced after blending with feedwater. The water produced was considered suitable for irrigation and was produced at an operation cost of approximately 72 kWh/ML or AU$100/ML ($0.1/m3) with electricity costed at $0.1/kWh. Fig. 14 shows a commercial wastewater EDR desalination plant at La Jolla, San Diego, USA. The largest EDR facility in the world (200,000 m3/day, Abrera, Barcelona, Spain), treats water from the Llobregat River with a variable salinity close to 3000 μS/cm, has an energy consumption of 0.6 kWh/m3 and produces water at a cost below €0.2/m3). Recent advances in EDR technology have resulted in improved performance, lowered cost and extended life of plant and materials. Monovalent selective membranes are now available to facilitate the preferential removal of sodium and potassium which allows a lowering of the sodium adsorption ratio (SAR), which is important when the water is being used for irrigation. The SAR level is an issue because if high SAR water is applied to a soil for extended periods of time, the sodium in the water can displace the calcium and magnesium in the soil. This decreases the ability of the soil to form stable aggregates and a loss of soil structure can occur. EDR has been installed in Spain mainly for drinking water and for reducing nitrates and trihalomethane precursors [117]. Comparisons of EDR with an RO plant of similar size show similar costs for the two processes. In another study, the reclamation of tertiary treated wastewater has proved to be 25% less costly with EDR than with RO [82]. Unlike RO, EDR can reliably remove selected ions such as nitrate, and its performance is not affected by silica. Water recovery is very high at 92% and it is not influenced by as many water constituents as RO. EDR has been applied to the brine concentrate from RO systems, reclaiming waters of 8000 mg/L salinity to give combined water recoveries of 96–98%. EDR for drinking water and irrigation from groundwater has also been applied in the Canary Islands, Spain [112]. In this case one of the main reasons for its use was the presence of silica in groundwater due to the volcanic origin of the islands, reducing the possible feasible recovery of RO plants. Photovoltaic ED systems have long been explored for brackish waters [2,6,87] and also for seawater as the feed [57]. The disadvantages of ED and EDR systems include the complexity of the system designs, the amount of scaling and fouling that occurs within the system, especially the membranes, and a low electrode life due to corrosion stemming from the reactions at the electrodes [48]. Specifically, the chlorine generated from the electrolysis of chloride ions in the salt water causes corrosion, particularly corrosion of membranes, lowering their effective life. Additionally, the gas evolution, oxygen at the anode and hydrogen at the cathode, requires the need for degasifiers, increasing the complexity and cost of desalination plants utilising this technology. Another additional issue is that EDR only removes charged substances and any substance not charged will go through to the product water. This is less worrying for agriculture purposes, but it is another issue to bear in mind. 3.1.4. Ion exchange resins Ion exchange (IX) is used for processes of purification, separation, and decontamination of aqueous and other ion-containing solutions with solid ion exchange resins. They are either cation exchangers or anion exchangers and can be regenerated by acid or alkali respectively,

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

8

S. Burn et al. / Desalination xxx (2015) xxx–xxx

Investment Costs of SWRO Plants 1,500

Investment (€m3/day)

1,400 1,300 1,200 1,100 1,000 900 800 700 600 500 0

20,000

40,000

60,000

80,000 100,000 120,000 140,000 160,000

Capactity (m3/day) Fig. 10. Investment costs of SWRO plant by size of plant.

or in the case of softening resins, with brine. They are hence dependant on chemicals for their regeneration, which generally limits their application for desalination of low salinity waters and the polishing of industrial waters. One radical departure from the ordinary is the use of heat for regeneration, which is achievable with weak electrolyte resins [103]. It was demonstrated on a large-scale in both Adelaide and Perth, the latter on deep anaerobic ground waters. There are drawbacks with IX such as degradation of the anion exchanger by oxygen in surface waters under the hot regeneration conditions and build up of divalent metal ions on the cation exchanger, requiring a prior softening step. These extra steps made the approach uneconomic. IX is especially appropriate for the selective removal of organic [16] and inorganic [63] ions that may cause membrane fouling. Scaleforming compounds are mainly calcium, barium and strontium sulphates, calcium carbonate and silica. Iron fouling because of the negative charge on the membrane can occur if ferric chloride has been the coagulant added in a prior step. Metal cations can be removed by a high recovery softening with a weak acid resin of the carboxylic acid type, typically used before RO in treating brackish waters. Iminodiacetic acid chelating resins are used for Sr2+ and Ba2+ removal from RO concentrates ahead of further RO treatment, whilst to prevent other scaling problems an aminophosphonic acid resin may be employed. In continuous ion exchange (CIX) the service, regeneration and rinsing steps are carried out simultaneously, so the product flow is not interrupted. By using a moving bed of resin it is possible to achieve continuous operation, the main advantage of which is a high processing efficiency. Co-current and counter-current versions are in use, but cocurrent systems are of limited value as ions are not completely removed from the raw water and the efficiency of regeneration is poor. Countercurrent systems though have considerable industrial significance and

have found their place, especially in industrial wastewater treatment. The main features of counter-current CIX are that in the adsorption column the incoming raw water first contacts the most loaded resin, to ensure the highest degree of resin loading, whilst the treated water is last in contact with freshly regenerated resin, to give the highest possible amount of contaminant removal. Likewise, in the regeneration column the waste stream is last in contact with fully loaded resin, leading to the highest concentration of regenerant effluent, whilst the fully regenerated resin is last in contact with fresh regenerant, to give the maximum level of regeneration [13]. CIX is said to be the most efficient method of adsorption available, with resin inventories up to 25% less than fixed-bed systems, and capable of operating on solutions containing up to 100 mg/L of total suspended solids, thus minimising prefiltration requirements. The approach has been extended to the removal of insoluble particles with a continuous ionic filtration (CIF™) process, which can be likened to continuous sand filtration, where particles are physically removed by a sand filter bed [23]. In CIF, charged resins are used instead of sand and the process removes dissolved ions as well. Continuous downward movement of resin and upflow of water in a counter-current way allows the process to operate on a dirty water feed. It has a higher removal efficiency than conventional systems. The system consists of a series of stages, 1) ion exchange, 2) resin washing and 3) resin regeneration, each designed for a specific function. Each stage contains a moving packed bed of resin, where resin and solution have intimate contact. The resin is transported between columns using an air lift pump. Among the many applications that the technique has been used for are desalination, membrane pre-treatment and zero liquid discharge systems. A typical flow chart for CIF is shown in Fig. 15. The main advantages of CIF are as follows: • • • • • • •

low capital and operating costs high water recovery, up to 99% in some applications high quality product water low volume of reject wastewater simplicity of operation low energy requirement compact physical footprint and minimal civil works requirements.

4. Other techniques applicable to desalination for agricultural purposes

Fig. 11. Capital costs of SWRO in Australia [51].

There are a number of other commercial techniques available that may potentially be suitable for providing desalinated water for agricultural purposes depending on the specific location and water needs. These techniques include multistage flash distillation (MSF), multi effect distillation (MED), capacitive deionization (CDI), vapour compression (VC) and solar humidification and dehumidification (HDH).

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx

9

Fig. 12. Relative capital costs per cubic metre for seawater and brackish water RO desalination according to facility size. For this purpose brackish water is 1000 mg/L TDS [107].

Emerging technologies which may one day be suitable for providing desalinated water include Forward Osmosis (FO), Membrane distillation (direct, air gap and vacuum) (MD), pervaporation (PV), and solar desalination techniques such as multi effect humidification and greenhouse distillation. Whilst RO is a pressure-driven process, MD and PV are thermal processes involving a phase change, with the water flux being dependant on the vapour pressure difference across the membrane. Different types of membranes are required for the different applications. For MD, a porous hydrophobic membrane with a pore size of 1–10 μm is used, whereas for RO a dense hydrophilic membrane is required. For PV desalination, a dense hydrophilic membrane is needed for water transport [108,109]. MD and PV are similar processes but can be chiefly identified by the way in which the membrane functions. If the membrane is simply a support structure that allows a meniscus to form on the feed side and plays

no role in the separation, then the process is MD. If the membrane actively participates in the separation process then the process is PV [14,34]. Of the various approaches, FO has a potential advantage with regard to waste disposal as it can avoid reject brine, a major problem with RO. [113] FO relies on osmosis, the natural diffusion of water through a hydrophilic semipermeable membrane from a low concentration solution to a solution having a higher concentration of dissolved material. A neat conceptual feature appropriate to FO in agriculture is the use of a fertilizer in the draw solution, for the immediate practical application of the product water for irrigation [79,80]. As the product water is mixed with the draw solution, for agricultural uses the draw material must be of benefit in the final desalted water. Initial results suggest that 1 kg of fertilizer can extract 11–29 L of water from a seawater feed, with potassium chloride, sodium nitrate and potassium nitrate performing best of the nine compounds tested [8,15,20–22].

Fig. 13. Schematic of the ED process [from Ionics (now GE) [116]. “Electrodialysis & Electrodialysis Reversal Technology.” Ionics Incorporated, PS-4055 E-US 0201-208.].

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

10

S. Burn et al. / Desalination xxx (2015) xxx–xxx

economic advantage over RO for waters with total dissolved salts as high as 10,000 mg/L [92]. IX becomes too expensive for higher salinity brackish waters above 15,000 mg/L because of the greater cost of chemicals for regeneration of the resins as the salinity rises. It is not suitable for highly brackish waters and seawater. Distillation for these high salinity waters is certainly an option depending on the infrastructure costs.

5.2. Energy usage and costs of suitable technologies

Fig. 14. Electrodialysis plant being operated at the western treatment plant in San Diego. Source: R Taylor CSIRO.

5. Relevance of desalination technologies to agriculture 5.1. Feedwater quality A summary of the processes for treating a range of feedwater salinities is summarised in Table 3. For low salinity brackish waters of salinity up to 2500–3000 mg/L the processes that are in commercial scale use and that are the most economic for that salt concentration are RO, NF, ED, IX and HDH. RO and ED both have large-scale plants up to hundreds of ML/day capacity running. The rest have smaller but well-tested plants operating and CDI has a 1 ML/day plant at the commercial stage. Distillation may not be appropriate for these low salinity waters as it is energy intensive. For more concentrated brackish waters of up to 15,000 mg/L the same processes are usable, with the probable exclusion of CDI. In a more recent review it has been suggested that ED has an

For desalination, RO is the most commonly selected method at present, as it can tackle the entire range of saline waters up to seawater, even though it is very energy intensive. Unless the cost of water is not an issue, inexpensive energy sources are the key to the production of agricultural water from low quality raw waters that require desalination. There is a number of cheaper alternative energy sources emerging compared to conventional RO, although not at a large scale. For example, a number of solar RO demonstration plants have been built, and the main implementation obstacles identified. The major technical issues are increased biofouling and scaling of the membranes caused by intermittent operation due to the diurnal pattern of solar energy. To minimise biofouling it may become necessary to choose between either batteries as an energy storage system to allow continuous operation, or the frequent replacement of membranes. Apart from solar and wind powered electrical generation, there is the use of pressure produced by simple mechanical means. These also include utilising the pressure available at certain bore heads, and pressure that can be sourced hydrostatically from high terrain or deep bore immersion. NF alone for salinity feed waters or in conjunction with RO for high salinity waters has particular relevance as a lower energy system. Its concentration of nutrient ions containing N, K and P, and removal of divalent metal ions, however, means that the product water quality may need to be adjusted before its usage for irrigation. The nutrients can be added from the concentrate as can calcium and magnesium ions to avoid their displacement by sodium in the soil by correcting the SAR. However, except for highly fouling waters the RO/NF combination is rarely economical compared to RO alone.

Fig. 15. Flow sheet for CIF™ process. From [23]

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx

11

Table 3 Suitability of processes at various feed water salinities. Salinity, mg/L

Process in use now for general desalination

Potentially viable in the future

Not especially suitable for that level of salinity

Brackish, up to 2500–3000 Brackish, up to 3000–15,000 Brackish, above 15,000 Sea water, ±35,000

RO, NF, ED, IX, HDH, CDI RO, NF, ED, IX, HDH RO, NF/RO, ED, HDH, MSF, MED, VC RO, NF/RO, HDH, MSF, MED, VC

MD, PV, FO MD, PV, FO, CDI MD, PV MD, PV

CDI FO, IX, CDI FO, IX, CDI

For many of the processes described above, desalination needs to be followed by blending with raw water to maintain the appropriate nutrient levels for plant and vegetable growth or the appropriate nutrients need to be added. Post treatment to add essential nutrients can add another AU$ ~0.02–0.07/m3 to the production price. A list of energy use and costs for techniques other than distillation is given in Table 4. Only two methods can be regarded as being fully commercial: RO and ED with large plants in operation, whilst CDI has a smaller 1 ML/day plant at the commercial stage. The rest are still experimental, although pilot and demonstration systems have been operated successfully. All of the processes use moderately complex plants, with RO and ED being the most complicated, especially RO when energy recovery is installed. The pre- and post-treatment needs are similar, but have to be styled for the particular application. The disadvantages of ED systems include the complexity of the system designs, the amount of scaling and fouling that occurs within the system, especially the membranes, and a low electrode life due to corrosion arising from reactions at the electrodes. Also, in contrast with RO, there is no removal of microbes or non-charged toxic compounds. For ED capital costs are competitive or slightly higher than RO, except when RO needs additional treatment prior to the membranes. Operating costs though are low, with reduced pre-treatment/post-treatment and lower membrane replacement costs. For feed waters of b1500 mg/L the power consumption is less for ED. Overall cost savings often outweigh those of RO. In a more recent review it has been suggested that EDR has an economic advantage over RO for waters with total dissolved salts as high as 10,000 mg/L [92]. Nevertheless, RO is a favoured choice because of the limited need for chemicals. CDI offers high water recovery at lower energy cost, and is now operating at full scale in plants of a size appropriate for small towns. It merits closer attention as regards water for agriculture in remote communities such as mining, tourist and defence locations, where solar power production could be feasible. Long term studies of fouling, scaling and cleaning are yet to be done.

For emerging technologies such as MD and PV, cheap heat sources need to be available because of the high energy need. For hot, arid climates, heat can be obtained intermittently by solar irradiation or continuously from thermal ground waters [12,24,28,42]. Both of these techniques have advantages over RO due to the higher recovery rates and the production of lower brine volumes. If zero liquid waste is a desired outcome MD certainly has potential, especially if combined with techniques such as Eutecic Freezing, which enables the selective removal of valuable salts. Capital costs are difficult to estimate as a large-scale MD plant has yet to be built, but capital and operating costs should be less than for RO as the process runs at very low pressures, so thinner piping is employed and fewer problems arise with leaks, pump failures and membrane fouling, so membranes should have longer lives. MD can only be competitive when low cost thermal energy is available and/or the source water is too difficult for RO treatment. Vacuum MD has an advantage over other MD modes, especially with solar energy for pumping and a cheap heat resource. A comparison with RO suggests that it should compete energetically with RO, but the flux is lower so that a higher membrane area needs to be installed. For FO the challenge of new draw solutions needs to be centred on compounds that are of small enough molecular size to generate high osmotic pressures, yet are easily recoverable [60–62]. Magnetic nanoparticles are one possibility for achieving simple recovery [62,73]. A better approach would seem to be the use of a draw material that is pertinent to the future application so that it does not have to be recovered, such as when fertilizer is used and the product water is then immediately appropriate for irrigation. HDH is of interest, but it is suitable only for very small drinking water units in arid, remote regions as it needs a large area of solar collectors, which makes it expensive to set up. However, it may be relevant for production that requires only relatively small amounts of water or where heat sources such as geothermal or waste heat are available. 5.3. Brine disposal

Table 4 Desalination systems possibly suitable for agricultural water production. Technologies

Energy use, kWh/kL

Total, US$/kL

Reference

RO

Brackish, 0.7–2.0 Sea, 1.6–12 Submarine, 2–2.5 Brackish, 0.25 Brackish, 1.6–2.3 Sea, 40

Brackish, 0.39–1.5

[66]

Sea, 0.55–1 Solar, 1.3 large plant, 2–6.5 small plant Not available 0.47

[77] [110,76,7]

Solar, 15–18 Geothermal, 13 Solar Pond, 0.4–1.3 Waste heat, 1.1–1.5 Solar, 18.3 Waste heat, 5.3 Solar, 16 Waste heat, 2 Solar, 3–6.4 Geothermal, 1.2 Not available

[85,86] [101] [115] [68] [115]

FO ED Direct contact MD

Air gap MD

Sea

Vacuum MD

Sea, 1.2–3.2

HDH

Brackish

CDI

Brackish, 0.13–0.59

One aspect that involves a significant environmental cost for desalination is brine discharge from desalination plants, this is especially so for inland applications where discharge to the ocean is not possible. The salinity of the discharge depends upon the salinity of the source water with discharge water from saltwater desalination being approximately twice as salty as the seawater, with typical concentrate figures being given in Table 5. Discharge from desalination plants generally consists of 98.5% rejection water with a significantly higher salt level than the input water and 1.5% from filter wash water and cleaning

[67] [66]

[85,86] [111] [36,17] [104,72]

Table 5 Typical salt concentration in RO concentrate and permeate with 45% recovery at 900 psi operating pressure (Dhawan [31]).

Sodium, Na Potassium, K Magnesium, Mg Calcium, Ca Bicarbonate, HCO3 Chloride, Cl Sulphate, SO4 TDS

Seawater (PPM)

Concentrate (PPM)

Permeate (PPM)

10,967 406 1306 419 109 19,682 2759 35,666

19,888 736 2372 761 194 35,771 5014 64,771

64 3 2 0.5 0.9 105 1.5 176

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

12

S. Burn et al. / Desalination xxx (2015) xxx–xxx

products [26]. In the case of seawater desalination, concentrated brines are negatively buoyant in seawater, giving them a tendency to sink and spread along the sea bottom, displacing normally saline water from hollows. This can have a devastating effect on sea bottom life which impacts more broadly on the entire shallow ecosystems, however, efficient diffuser design and placement can ensure that problems from brine discharge to the ocean are minimal [32]. Five main methods exist for the disposal of brines as detailed below. These methods are discussed in detail by [4], however, little detail is given regarding the feasibility of the individual techniques nor comparative costs and environmental impacts of the alternative techniques. Surface water/ocean discharge Discharge to the sewer Deep well injection Land application Evaporation/crystallisation.

Currently brine disposal from 28 Australian desalination plants is via discharge to ocean outfalls (6 plants), discharging to the sewer for smaller inland plants (11 plants), to evaporating basins (8 plants and ground infiltration (3 plants) [52–54]. An interesting development is in situ RO within boreholes, with the reject salt concentrate being pumped down to a lower level that is sealed off from the input side [30]. Fig. 16 shows some of the factors to consider when choosing brine-disposal methods. Environmentally safe brine disposal depends mainly on the site of the treatment plant. For plants situated near the sea or close to brackish environments, such as estuaries, brine disposal is comparatively easy compared to inland desalination facilities with sea discharge invariably being the preferred option. The most common methods of brine disposal for all RO installations in Australia are shown in Fig. 17 [75]. As seen in Fig. 17 the most common disposal options for inland plants is discharge to the sewerage network or evaporation in ponds, however, neither will be sustainable in the long run, as larger quantities of brine are produced. Large evaporation ponds are expensive and do not reuse the water and based upon costs for the building of lined earthen dam impoundments, they are estimated to have a capital cost of AU$12,000/ML and require a land surface of 600 m2/ML to give a cost of AU$20/m2 [33]. Land application also has significant issues associated with the effects of salt on plant growth and issues associated with sodicity and deep well injection whilst attractive can be expensive. An alternative to these techniques is direct injection of the brines back into the aquifer from which they came. Significant studies exist for the injection of CO2 into aquifers for carbon sequestration and

Re egullato ory envvironment

Publicc pe erceepon

although some include costs studies [64] the costs were not deemed relevant to brine injection due to the high depth of the wells at 1500– 2000 m. Few studies exist regarding the injection of brine following desalination and details regarding the cost of injection are limited. Nassar, El-Damak, and Ghanem [74] for example used computational models to simulate the disposal of brine into subsea aquifers and they developed a range of design procedures for the management of brine injection based upon an understanding of dense brine flows in a salty aquifer, unfortunately no cost analysis was provided. Woolley and Kalf [106] investigated the injection of bitterns into an aquifer at a depth of approximately 300 m in the Tullakool Irrigation area west of Deniliquin, but again no cost analysis was provided. Weller [105] carried out a detailed analysis of the costs for disposing brine underground, developing algorithms for both capital and operating costs utilising parameters with the only information available being disposal costs v's half fracture length. Some studies have collected data on the costs associated with treated stormwater injection. For example experience from the Salisbury, Playford and “Waterproofing the West” projects (personal communication Peter Dillon CSIRO) indicates that drilling for aquifer recharge; including drilling costs, casing and cementing is in the order of AU$330/m for a completed well. Equipping the well with a pump (1–2000 m3 L/d) and rising main costs an additional AU$30,000. These costs exclude the costs of bringing power to the site and the costs of pipelines to and from the site and can be seen in Fig. 18. This can be compared with the costs for drilling a 200 mm diameter well 407 m deep in the Chowilla floodplain which cost AU$660/m including drilling, casing and cementing costs [65]. If brine injection into the aquifer is to be considered seriously then the environmental impacts and the necessary approvals at a local level would need to be addressed. Traditionally brine has been seen as having significant costs associated with disposal, however, more recently brine is being seen as a potential resource that when exploited, is able to reduce the net cost of providing fresh water [3]. As discussed in the AFFA document options for resource recovery of brines include salt harvesting as a high valued product for agriculture at a value of AU$25–$250/tonne, irrigation for salt tolerant crops such as Pistachios, aquaculture, solar ponds which

Kn now wled dge and d de egre ee of soph hiscation

160,000 140,000

Lo ocal avaailab bilityy, prroxim mityy,geo ologgy, geog g graphy, PEW P W disch d hargge lim mitss

C Concentratte ch haraacteeristtics

Faacto orss Teechn nolo ogy

Corrrosion and pipeline inte egritty p

So oil co ond dio ons, agri a icultture e, livvesto ock

o A Availability of nerggy en

C Costt

120,000 100,000 Cost $

• • • • •

Fig. 17. Common approaches to brine water management in Australia [75].

80,000

Fixed Costs

60,000

Well Costs

40,000

Total Costs

20,000 0 50

100

150

200

250

300

350

Bore Depth (m) Fig. 16. Factors affecting brine disposal options, [75].

Fig. 18. Costs associated with drilling a 200 mm well.

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx

have the potential to produce electrical energy at an economical cost and finally controlled drainage integrated systems that concentrate the salt via a range of agricultural steps such as Lucerne fodder, barley fodder, dates and saltbush, saltwater fish and crustaceans, solar pond and finally evaporated solar salts. 6. Discussion Opportunities for applying desalination technologies to agriculture go hand-in-hand with requirements to improve irrigation practices and efficiency, which can achieve significant reductions in water demand and therefore make the application of desalinated technologies more viable, because the cost of the water is not as critical a component. Gregory [118] summarised the challenges and opportunities in water-use efficiency in irrigated agriculture as “Irrigated crops occupy about 15–20% of the total cropped area but contribute 33–40% of the production, so they are crucial to the world's food supply. Most irrigation is applied on the surface (84.5% of the total), with smaller amounts via sprinklers (13.5%) and localised systems (2%).” Globally, almost twothirds of the water delivered to irrigated crops is lost as drainage or runoff, or both, with those losses being associated with storage and conveyancing efficiencies (30%) as well as farm irrigation efficiency (37%). [118] continues, “Until recently, most irrigation was scheduled on the basis of fully meeting crop water requirements, but with sprinkler and localised drip systems it has been possible to demonstrate that deficit irrigation strategies can not only sustain yields and profitability, but also reduce water use.” This suggests that it is possible to significantly reduce water use in irrigated agriculture, which also means that the viability of more expensive sources of water is possible as less water is required per unit of crop production. Whilst the use of alternate technologies such as desalination should be explored, they should be done in conjunction with implementation of more efficient water use methods. The majority of irrigated crops can tolerate water of relatively high salt content with crops such as broccoli, tomatoes and beetroot being much more tolerant than okra and peas for example as shown in Table 6. This provides opportunities to mix permeate water from desalination with water of marginal quality to increase the volume of water available for agriculture. If blending can occur, then water production costs would be lower when estimated per water unit supplied. This means that desalination technologies if combined with water blending and tailored to a particular crop can be used to provide water at a significantly reduced cost to providing desalinated water alone. Mixing may also reduce the need for post-treatment which adds cost to produced water as permeate is commonly required to be remineralised and ionically balanced (as permeate has reduced levels of calcium and magnesium as well as being slightly acidic). No detailed analysis of the use of blended water was found in the literature and an in depth analysis of the use of such waters for different crops would be a valuable addition to the literature regarding the use of desalinated water for agricultural purposes. It should be noted however that sometimes the blending of groundwater with desalinated water would not be enough for meet Table 6 Water quality required for different crops [90]. Vegetable

Threshold level (dS m−1)

Pea Okra Tomato Eggplant Pepper Carrot Broccoli Cauliflower Potato

1.5 1.2 2.5 1.1 1.5 1.0 2.8 1.8 1.7

13

the recommendable water quality, for example if the resulting water does not have an equilibrated SAR. In these cases, the only way to meet the requirements would be additional remineralisation. Other aspects such as the presence of organic matter or other chemical compounds in water to be used for blending will also have to be considered. One of the identified benefits for desalinated water use in the agricultural sector is that the use of desalinated water increases the productivity [112] and quality of some agriculture products and at the same time leads to lower water consumption and recovery of salinity-affected soils. As discussed by Zarzo et al. [112] irrigation of citric fruit plantations with desalinated water led to increases in production by 10 to 50% (depending on the water quality used prior to introduction of desalinated water), whilst water needs reduced by 20%. For a case of greenhouse production of bananas irrigated with desalinated wastewater, fertilizers and water use were reduced by 50 and 30% respectively, leading to an increase in banana production and the earlier maturation of plants. An increase in productivity may be related to the leaching of accumulated salts in the soil profile by the high quality desalinated water. In Spain, greenhouse products (horticultural, flowers and ornamental plants) provide greater added value per unit of irrigated water (5.79 €/m3 on average), followed by vineyards and fruit trees (1.08 and 0.68 €/m3 respectively), and cereal grains (0.06 €/m3). An average of 0.41 €/m3 was estimated for all products. These figures relate to high-value crops, for which the overall water cost may be marginal compared to total costs. At this stage it is unlikely that the production of cotton, rice or sugar can be effectively supported by water supplied from desalination plants. The above leads to the conclusion that the use of desalinated water for agriculture will be viable where there is limited access to water that is fit for purpose and it is most likely to be cost effective in a tightly controlled environment, using agricultural practices with the mosteffective water use and crops with high productivity. Such conditions are often associated with greenhouses and the production of highvalue irrigated crops. This is the case in Spain; however it is also important to mention that the high level of financial support and subsidies provided to the agricultural sector in Spain and other EU countries make this option more viable. A national analysis undertaken by the National Centre of Excellence for Desalination (NCED has shown that existing farmers are unwilling to pay more than AU$1.00/m3 for water and in many regions even this cost is considered to be unacceptably high (e.g., common water costs in southwest Western Australia are AU$0.18–0.50/m3). Comparing this willingness-to-pay with the unit cost of water production in large desalination plants means that seawater desalination is an unlikely option for traditional agricultural practice where subsidies are not available. However for applications where the cost of water is small compared to the infrastructure investment, i.e., glasshouses and hydroponics, the application of seawater desalination technologies is currently viable. Where large volumes of brackish water are available, cost-effective desalination for agriculture is likely to be related to circumstances where high yielding brackish groundwater resources are available close to the ocean and an existing irrigational water demand exists. Such conditions are likely to minimise the cost of water production and distribution. Low-salinity groundwater is likely to allow higher recovery rates and provide the possibility of mixing feedwater with the high-purity desalinated water. Proximity to the ocean enables disposal of brine to the ocean, reducing disposal costs compared to inland disposal via evaporation ponds or deep-well injection. 7. Conclusions Suitable technologies such as reverse osmosis and electrodialysis to provide desalinated water for agriculture are currently available and can provide water for agriculture, but at a cost that is currently more expensive than that generally used for agricultural purposes. The

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

14

S. Burn et al. / Desalination xxx (2015) xxx–xxx

adaptation of desalination to supply water for agriculture may be cost effective; especially when applied to high value crops where the cost of the water is not a critical issue. For these applications established technologies such as reverse osmosis and electrodialysis are being commonly applied in a number of countries around the world. Wider application of technologies, especially for broader scale allocations will only occur where there is limited availability of fit for purpose water for irrigation and will depend on further developments in desalination technologies, reduction in costs, and advancements in agricultural water-use practices that reduce water losses and increase water-use efficiency. In this context, the cost of water is always relative and depends on the availability of supply and its further use. In addition we must bear in mind that food supply and water security are issues of national importance and having an inexhaustible water resource, such as that which can be supplied by desalination, could be a factor that may make the cost of water insignificant. Acknowledgements The support of the National Centre of Excellence for Desalination and the CSIRO Water for a Healthy Country Flagship is gratefully acknowledged. References [1] AcuaMed, La desalación en España, Sostenibilidad para zonas vulnerables (Desalination in Spain. Sustainability for vulnerable areas), Ministry of Environment, Rural and Marine. Spain Government, 2011. [2] M.R. Adiga, S.K. Adhikary, P.K. Narayanan, W.P. Harkare, S.D. Gomkale, K.P. Govindan, Performance analysis of photovoltaic electrodialysis desalination plant at Tanote in Thar desert, Desalination 67 (1987) 59–66. [3] AFFA, Introduction to Desalination Technologies in Australia, Agriculture, Fisheries & Forestry, Australia, 2002. (Retrieved from http://www.environment.gov.au/ water/publications/urban/desalination-summary.html). [4] N. Afrasiabi, E. Shahbazali, RO brine treatment and disposal methods, Desalin. Water Treat. 35 (1–3) (2011) 39. [5] S. Al-Kharabsheh, An innovative reverse osmosis desalination system using hydrostatic pressure, Desalination 196 (2006) 210–214. [6] H.M.N. Al Madani, Water desalination by solar powered electrodialysis process, Renew. Energy 28 (2003) 1915–1924 (membranes: a review. Desal. 259, 1-10). [7] S. Al-Hallaj, S. Parekh, M.M. Farid, J.R. Selman, Solar desalination with humidification–dehumidification cycle: review of economics, Desalination 195 (2006) 169–186. [8] A. Alkhudhiri, N. Darwish, N. Hilal, Membrane distillation: a comprehensive review, Desalination 287 (2012) 2–18. [9] A.S. Al-Amoudi, Factors Affecting Natural Organic Matter Removal (NOM) and Scaling Fouling in NF, 2010. [10] M. Beccali, M. Sorce, J. Galletto, The potential of renewable energies in Sicily for water desalination applications, in: L. Rizzuti, H.M. Ettouney, A. Cipollina (Eds.), Solar Desalination for the 21st Century: A Review of Modern Technologies and Researches on Desalination Coupled to Renewable Energies, Springer, Dordrecht, 2007, pp. 179–194. [11] Beltrán, Koo-Oshima, Water desalination for agricultural applications, Proceedings of the FAO Expert Consultation on Water Desalination for Agricultural Applications, Taylor & Francis Group, Rome, April 26–27, 2004. [12] C. Bier, U. Plantikow, Solar-powered desalination by membrane distillation, Proc. IDA World Congress, Abu Dhabi1995. (http://www2.hawaii.edu/~nabil/solar.htm). [13] B.A. Bolto, L. Pawlowski, Wastewater Treatment by Ion Exchange, Spon, London, 1987. 179–194. [14] B.A. Bolto, M. Hoang, Z. Xie, A review of organic polymeric membranes for the dehydration of aqueous ethanol by pervaporation, Chem. Eng. Process. Process Intensif. 50 (2011) 227–235. [15] B. Bolto, T. Tran, M. Hoang, Membrane distillation — a low energy desalting technique? Water 34 (4) (2007) 59–62. [16] M. Bourke, M. Slunjski, MIEX® DOC process launched in Western Australia, Water J. 26 (6) (1999) 17–20. [17] K. Bourouni, M.T. Chaibi, L. Tadrist, Water desalination by humidification and dehumidification of air: state of the art, Desalination 137 (2001) 167–176. [18] S. Bunani, E. Yörükoğlu, G. Sert, Ü. Yüksel, M. Yüksel, N. Kabay, Application of nanofiltration for reuse of municipal wastewater and quality analysis of product water, Desalination 315 (2013) 33–36. [19] C. Cabassud, D. Wirth, Membrane distillation for water desalination: how to choose an appropriate membrane? Desalination 157 (1–3) (2003) 307–314. [20] CADETT, A solar desalination system using the membrane distillation process, Technical Brochure No. 461996. (http://gasunie.eldoc.ub.rug.nl/FILES/root/1996/ 2037986/2037986.tif).

[21] T.Y. Cath, A.E. Childress, M. Elimelech, Forward osmosis: principles, applications and recent developments, J. Membr. Sci. 281 (2006) 70–87. [22] M. Chaibi, An Overview of Solar Desalination for Domestic and Agriculture Water Needs in Remote Arid Areas, 1999. [23] CleanTeQ, Continuous ionic filtration, http://www.cleanteq.com/index.php? sectionID=10709&pageID=10729 2012. [24] D. Clery, Greenhouse-power plant hybrid set to make Jordan's desert bloom, Science 331 (6014) (2011) 136. [25] CSIRO, Water, Chapter 7, http://www.csiro.au/Outcomes/Water/Water-Book. aspx 2011. [26] CONAMA, 9th Congreso nacional de medio ambiente (9th National Environment Congress), Nuevas fuentes de agua (New Water Sources)2008. [27] A.T. De Almeida, P.S. Moura, Desalination with wind and wave power, in: L. Rizzuti, H.M. Ettouney, A. Cipollina (Eds.), Solar Desalination for the 21st Century: A Review of Modern Technologies and Researches on Desalination Coupled to Renewable Energies, Springer, Dordrecht, 2007, pp. 305–325. [28] S. Deng, Solar powered membrane distillation, Mem. & Sepn. Technol. NewsApril 2006. (http://goliath.ecnext.com/coms2/gi_0199-5534673/Solar-Power-MakesLow-Cost.htm). [29] Desaldata, Global Desalination Market GWIRetrieved October 30rd, 2012, from Desaldata: www.desaldata.com 2012. [30] Desalin8, from http://www.desaln8.com/index.php 2013. [31] G.K. Dhawan, Water treatment guide — technical database for the water treatment industryfrom http://www.watertreatmentguide.com/seawater_desalination. htm 2007. [32] P. Dickie, Making water — desalination: option or distraction for a thirsty world? Retrieved from assets.panda.org/downloads/desalinationreportjune2007.pdf 2007. [33] P.J. Dillon, P. Pavelic, D.W. Page, H. Beringen, J. Ward, Managed aquifer recharge: an introduction, Waterlines Report No 13Feb 2009. [34] M. Elma, C. Yacou, D.K. Wang, Microporous silica based membranes for desalination, Water 4 (2012) 629–649. [35] P. Eriksson, M. Kyburz, W. Pergande, NF membrane characteristics and evaluation for sea water processing applications, Desalination 184 (2005) 281–294. [36] A. Eslamimanesh, M.S. Hatamipour, Economical study of a small-scale direct contact humidification–dehumidification desalination plant, Desalination 250 (2010) 203–207. [37] A. Estevan, Desalación, energía y medio ambiente, Desalination, Energy and Environment, University of Sevilla—Ministry of Environment, 2007. [38] H. Ettouney, L. Rizzuti, Solar desalination: a challenge for sustainable fresh water in the 21st century, 2007. 1–18. [39] E. Fadigas, J.R. Dias, Desalination of water by reverse osmosis using gravitational potential energy and wind energy, Desalination 237 (1–3) (2009) 140–146. [40] FAO, Water desalination for agricultural applications, in: J.M. Baltrán, S. KooOshima (Eds.), Proc. FAO Expert Consultation on Water Desalination for Agricultural Applications, 26–27 April 2004, Food and Agriculture Organisation of the United Nations, Rome, 2006. [41] M. Farinas, El coste del agua producida por las grandes desaladoras de agua de mar en España (The cost of produced water by big seawater desalination plants in Spain), Ingeniería y territorio (Engineering and Territory), 2007. 72. [42] H.E.S. Fath, S.M. Elsherbiny, A.A. Hassan, M. Rommel, M. Wieghaus, J. Koschikowski, M. Vatansever, PV and thermally driven small-scale, stand-alone solar desalination systems with very low maintenance needs, Desalination 225 (2008) 58–69. [43] M. Forstmeier, F. Mannerheim, F. D'Amato, M. Shah, Y. Liu, M. Baldea, A. Stella, Feasibility study on wind-powered desalination, Desalination 203 (1–3) (2007) 463–470. [44] C. Garcia, F. Molina, D. Zarzo, 7 year operation of a BWRO plant with raw water from a coastal aquifer for agricultural irrigation, Desalin. Water Treat. 31 (2011) 331–338. [45] V. Garcia, M.A. Marzal, Designing membrane systems for the coming future: Perth II desalination plant, Proceedings from the IDA World Congress on Desalination and Water reuse, Dubai, 2009, 2009. [46] N.T.M. Ghaffour, G. Amy, Technical review and evaluation of the economics of water desalination: current and future challenges for better water supply sustainability, Desalination 309 (2013) 197–207. . http://dx.doi.org/10.1016/j.desal.2012.10.015. [47] N.B. Goodman, R.J. Taylor, Z. Xie, Y. Gozukara, A. Clements, A feasibility study of municipal wastewater desalination using electrodialysis reversal to provide recycled water for horticultural irrigation, Desalination 317 (2013) 77–83. [48] B.K. Gurtler, T.A. Vetter, E.M. Perdue, E. Ingall, J.-F. Koprivnjak, P.H. Pfromm, Combining reverse osmosis and pulsed electrical current electrodialysis for improved recovery of dissolved organic matter from seawater, J. Membr. Sci. 323 (2008) 328–336. [49] GWI, Market-leading Analysis of the International Water Industry, Global Water Intelligence, 2012. [50] GWI, GWI Desaldata, 2012. [51] GWI, Capital costs of Australian desalination plants are rising – CHART – Vol 10, Issue 9 (September 2009)from http://www.globalwaterintel.com/archive/10/9/ analysis/capital-costs-australian-desalination-plants-chart.html 2013. [52] M. Hoang, B.A. Bolto, T. Tran, Desalination by capacitive deionisation, Water 36 (1) (2009) 132–135. [53] M. Hoang, B. Bolto, C. Haskard, O. Barron, S. Gray, G. Leslie, Desalination plants: an Australia survey, Water 36 (2) (2009) 67–73. [54] M. Hoang, B. Bolto, C. Haskard, O. Barron, S. Gray, G. Leslie, Desalination in Australia, 2009. [55] Hydration Technologies, Osmatic Water Purification Devices — Osmotic White Paperfrom www.Hydrationtech.com 2003.

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041

S. Burn et al. / Desalination xxx (2015) xxx–xxx [56] W.E. Katz, State-of-the-art of the electrodialysis reversal (EDR) process in 1983, Proc. 44th Annual Meeting International Water Conference, Pittsburgh, Engineering Society of Western Pennsylvania, Bulletin TP 329, 1983. [57] O. Kuroda, S. Takahashi, K. Wakamatsu, S. Itoh, S. Kubota, K. Kikuchi, Y. Eguchi, Y. Ikenaga, N. Sohma, K. Nishinoiri, An electrodialysis sea water desalination system powered by photovoltaic cells, Desalination 65 (1987) 161–169. [58] S. Lattemann, M.D. Kennedy, J.C. Schippers, G. Amy, Sustainable water for the future: water recycling versus desalination chapter 2 global desalination situation, Sustain. Water Future 2 (2010) 7–39 (c). [59] S. Lattemann, M.D. Kennedy, J.C. Shippers, G. Amy, Sustainability Science and Engineering, Elsevier B.V., 2010 [60] M.M. Ling, T.-S. Chung, Desalination process using super hydrophilic nanoparticles via forward osmosis integrated with ultrafiltration regeneration, Desalination 278 (2011) 194–202. [61] M.M. Ling, T.-S. Chung, Surface-dissociated nanoparticle draw solutions in forward osmosis and the regeneration in an integrated electric field and nanofiltration system, Ind. Eng. Chem. Res. 51 (2012) 15463–15471. [62] M.M. Ling, K.Y. Wang, T.-S. Chung, Highly water-soluble magnetic nanoparticles as novel draw solutes in forward osmosis for water reuse, Ind. Eng. Chem. Res. 49 (12) (2010) 5869–5876. [63] J. Lipnizki, B. Adams, M. Okazaki, A. Sharpe, Combining reverse osmosis and ion exchange, Filtr. Sep. 49 (2012) 30–33. [64] A. Lucier, M. Zoback, Assessing the economic feasibility of regional deep saline aquifer CO2 injection and storage: a geomechanics-based workflow applied to the Rose Run sandstone in Eastern Ohio, USA, Int. J. Greenh. Gas Control 2 (2) (2008) 230–247. . http://dx.doi.org/10.1016/j.ijggc.2007.12.002. [65] P. Magarey, O. Kwadwo, Regional Disposal Strategy, Renmark Group Deep Injection: Phase II Feasibility Study: DWLBC, 2008. [66] D.Z. Martinez, C.G.G. Soto, R.B. Candel, Experiences on desalination of different brackish water, Proc. IDA World Congress, Dubai. IDA WC/DB09-135, 2009. [67] R.L. McGinnis, M. Elimelech, Energy requirements of ammonia-carbon dioxide forward osmosis desalination, Desalination 207 (2007) 370–382. [68] Medesol, Seawater desalination by innovative solar-powered membrane distillation system, http://www.psa.es/webeng/projects/medesol/overview.html2007. [69] J.A. Medina, Feasibility of water desalination for agriculture, Paper presented at the Proceedings of the FAO Expert Consultation on Water Desalination for Agricultural Applications Rome, 2004. [70] R.E.T. Meijer, TU Delft tests wind, ill for seawater desalinationfrom www.tudelft.nl/ live/pagina.jsp?id=afl186f83-e60f-4d2e-b0e5-dcac3423fcd5&lang=en 2008. [71] F.H. Meller, Electrodialysis (ED) and Electrodialysis Reversal (EDR) Technology, Ionics Incorporated, 1984. [72] M. Mossad, W. Zhang, L. Zou, Using capacitive deionisation for inland brackish groundwater desalination in a remote location, Desalination 308 (2013) 154–160. [73] NanoMagnetics, Water filters rely on nanotech, http://www.wired.com/science/discoveries/news/2004/10/65287?currentPage=all 2004 (http://www.nanomagnetics. com). [74] M.K.K. Nassar, R.M. El-Damak, A.H.M. Ghanem, Impact of desalination plants brine injection wells on coastal aquifers, Environ. Geol. 54 (3) (2008) 445–454. . http:// dx.doi.org/10.1007/s00254-007-0849-9. [75] NSW Public Works, Brackish groundwater: a viable community water supply option? Waterlines Report Series No. 66, NSW-Public-Works, 2011. [76] P. Pacenti, M. de Gerloni, M. Reali, D. Chiaramonti, S.O. Gärtner, P. Helm, M. Stöhr, Submarine seawater reverse osmosis desalination system, Desalination 126 (1999) 213–218. [77] G. Papadakis, E.S. Mohamed, D. Manolakos, Small autonomous RO desalination systems powered by renewable energies, in: L. Rizzuti, H.M. Ettouney, A. Cipolllina (Eds.),Technological advances and economics. Solar desalination for the 21st century, Proc. NATO Advanced Research Workshop, 2007, pp. 293–304. [78] M. Papapetrou, C. Epp, E. Tzen, Autonomous desalination units based on renewable energy systems — a review of representative installations worldwide, in: L. Rizzuti, H.M. Ettouney, A. Cipollina (Eds.), Solar Desalination for the 21st Century: A Review of Modern Technologies and Researches on Desalination Coupled to Renewable Energies, Springer, Dordrecht, 2007, pp. 343–353. [79] S. Phuntsho, H.K. Shon, S. Hong, S. Lee, S. Vigneswaran, A novel low energy fertilizer driven forward osmosis desalination for direct fertigation: evaluating the performance of fertilizer draw solutions, J. Membr. Sci. 375 (2011) 172–181. [80] S. Phuntsho, S. Hong, M. Elimelech, H.K. Shon, Forward osmosis desalination of brackish groundwater: meeting water quality requirements for fertigation by integrating nanofiltration, J. Membr. Sci. 436 (2013) 1–15. [81] Queensland-Government, Water Management: Facts and Figuresfrom www.nrw. qld.gov.au/water/gab/facts.html 2008. [82] E.R. Reahl, Half a century of desalination with electrodialysis, in: G.P.A. Water (Ed.), Technical Paper, General Electric Company, Trevose, Pennsylvania, 2006. [83] M. Reali, M. deGerloni, A. Sampaolo, Submarine and underground reverse osmosis schemes for energy-efficient seawater desalination, Desalination 109 (3) (1997) 269–275. . http://dx.doi.org/10.1016/s0011-9164(97)00073-8. [84] K. Reddy, N. Gaffour, Overview of the cost of desalinated water and costing methodologies, Desalination 205 (1–3) (2007) 340. [85] R.B. Saffarini, E.K. Summers, H.A. Arafat, V.J.H. Lienhard, Technical evaluation of stand-alone solar powered membrane distillation systems, Desalination 286 (2012) 332–341.

15

[86] R.B. Saffarini, E.K. Summers, H.A. Arafat, V.J.H. Lienhard, Economic evaluation of stand-alone solar powered membrane distillation systems, Desalination 299 (2012) 55–62. [87] G. Savchenko, B.V. Tomizhevskio, T.M. Kotina, B.I. Lemasov, Solar photoelectric electrodialysis desalination plant, Appl. Solar Energy 14 (1978) 34–38. [88] A.I. Schäfer, A.G. Fane, T.D. Waite, Nanofiltration — Principles and Applications, in: A.I. Schäfer, A.G. Fane, T.D. Waite (Eds.), Elsevier, Oxford, 2005. [89] R.F. Service, Desalination freshens up, Science 313 (Aug 26 2006) 1088–1090. [90] M. Shahbaz, M. Ashraf, F. Ashraf, P.J.C. Al-Qurainy, P.J.C. Harris, Salt tolerance in selected vegetable crops, Crit. Rev. Plant Sci. 31 (4) (July 2012) 303–320. [91] H. Strathmann, Operating principle of electrodialysis and related processes, in: H. Strathmann (Ed.), Ion-Exchange Membrane Separation Processes, Elsevier, Amsterdam, 2004, pp. 147–233. [92] H. Strathmann, Electrodialysis, a mature technology with a multitude of new applications, Desalination 264 (2010) 268–288. [93] Sundrop-Farms, Sundrop Farmsfrom http://www.sundropfarms.com/ 2012. [94] J. Tanninen, L. Kamppinen, M. Nyström, Pretreatment and hybrid processes, in: A.I. Schäfer, A.G. Fane, T.D. Waite (Eds.), Nanofiltration — Principles and Applications, Elsevier, Oxford, 2005, pp. 253–254. [95] R. Taylor, B. Bolto, Electrodialysis — a mature membrane desalting process with a bright future? Water J. 38 (2) (2011) 96–99. [96] R. Taylor, N.B. Goodman, Pilot plant demonstration of effluent desalination by electrodialysis at the Western Treatment Plant, Werribee, CSIRO Water for a Healthy Country Flagship Report2007. (ISSN: 1835-095X). [97] M. Thomson, D. Infield, A photovoltaic-powered seawater reverse-osmosis system without batteries, Desalination 153 (1–3) (2003) 1–8. [98] A.A. Tofigh, G.D. Najafpour, Technical and economical evaluation of desalination processes for potable water from seawater, Middle East J. Sci. Res. 12 (1) (2012). [99] T. Tran, M. Hoang, T. Duong, B. Bolto, Improving pollutant removal and membrane performance via pre-treatment with a specific formulation of polysilicato-iron, Desalin. Water Treat. 34 (1–3) (2011) 106–111. [100] A.A. Urcelay, Tecnologías Sostenibles sobre la gestión del agua: La desalación, Estudio del caso en la zona costera del Mediterráneo, 2007. (from http:// upcommons.upc.edu/pfc/bitstream/2099.1/4248/1/Ainhoa%20Albaina.pdf). [101] J. Walton, H. Lu, C. Turner, S. Solis, H. Hein, Solar and waste heat desalination by membrane distillation, Desalination and Water Purification Research and Development Program Report No. 81, U.S. Department of the Interior, Bureau of Reclamation, Denver, 2004. [102] WaterTech, Proposed seawater-to-bottled-water plan opposedfrom http://www. watertechonline.com/news.asp?N_ID=71809 2008. [103] D.E. Weiss, B.A. Bolto (1968) Improvements in or relating to water demineralisation (basic SIROTHERM process). Australian Patent 274,029. [104] T.J. Welgemoed, C.F. Schutte, Capacitive deionisation technology: an alternative desalination solution, Desalination 183 (2005) 327–340. [105] T.F. Weller, Produced Brine Underground Disposal Injection Cost Model, 1993. 25–34. [106] D.R. Woolley, F.R. Kalf, Feasibility of disposal of bitterns or brine by injection into an aquifer, Tullakool Irrigation Area, New South Wales, Proceedings of the Groundwater Pollution Conference, 1981. [107] WSTB, Desalination: a national perspectiveRetrieved from http://www.nap.edu/ openbook.php?record_id=12184&page=R1 2008. [108] Z. Xie, M. Hoang, T. Duong, D. Ng, B. Dao, S. Gray, Sol–gel derived poly(vinyl alcohol)/maleic acid/silica hybrid membrane for desalination by pervaporation, J. Membr. Sci. 383 (2011) 96–103. [109] Z. Xie, D. Ng, M. Hoang, T. Duong, S. Gray, Separation of aqueous salt solution by pervaporation through hybrid organic–inorganic membrane: effect of operating conditions, Desalination 273 (2011) 220–225. [110] U. Yermiyahu, A. Tal, A. Ben-Gal, A. Bar-Tal, J. Tarchitzky, O. Lahav, Environmental science — rethinking desalinated water quality and agriculture, Science 318 (5852) (2007) 920–921. . http://dx.doi.org/10.1126/science.1146339. [111] G. Yuan, Z. Wang, H. Li, X. Li, Experimental study of a solar desalination system based on humidification–dehumidification process, Desalination 277 (2011) 92–98. [112] D. Zarzo, E. Campos, P. Terrero, Spanish experience in desalination for agriculture, Desalin. Water Treat. 51 (1–3) (2013) 53–56. [113] S. Zhao, L. Zou, C.Y. Tang, D. Mulcahy, Recent developments in forward osmosis: opportunities and challenges, J. Membr. Sci. 396 (2012) 1–21. [114] Y. Zhou, R.S.J. Tol, Evaluating the costs of desalination and water transport, Water Resour. Res. 41 (3) (2005) 3003. [115] G. Zuo, R. Wang, R. Field, A.G. Fane, Energy efficiency evaluation and economic analyses of direct contact membrane distillation system using Aspen Plus, Desalination 283 (2011) 237–244. [116] F.H. Meller, Electrodialysis (ED) and Electrodialysis Reversal (EDR) Technology, Ionics Incorporated, 1984. [117] M.A. Ferrer, D.M. Zarzo, R. Lemes, Trihalomethanes (THM) Precursor Reduction of Surface Water by EDR Technology, 2007. (Retrieve from https://www.gewater. com/kcpguest/salesedge/documents/Technical%20Papers_Cust/Americas/English/ TP1179EN.pdf). [118] P.J. Gregory, Soils and food security: challenges and opportunities, in: R.E. Hester, R.M. Harrison (Eds.), Soils and Food Security, RSC Publishing, 2012. http:// dx.doi.org/10.1039/9781849735438.

Please cite this article as: S. Burn, et al., Desalination techniques — A review of the opportunities for desalination in agriculture, Desalination (2015), http://dx.doi.org/10.1016/j.desal.2015.01.041