环境工程专业英语文献中英双语版.docx

Treatment of geothermal waters for production of
industrial, agricultural or drinking water
Darrell L. Gallup ∗
Chevron Corporation, Energy Technology Company, 3901 Briarpark Dr., Houston, Texas 77042, USA
Received 14 March 2023; accepted 16 July 2023
Available online 12 September 2023
Abstract
A conceptual study has been carried out to convert geothermal water and condensate into a valuable industrial, agricultural or drinking water resource. Laboratory and field pilot test studies were used for the conceptual designs and preliminary cost estimates, referred to treatment facilities handling 750 kg/s of geothermal water and 350 kg/s of steam condensate. The experiments demonstrated that industrial, agricultural and drinking water standards could probably be met by adopting certain operating conditions. Six different treatments were examined. Unit processes for geothermal water/condensate treatment include desilication of the waters to produce marketable minerals, removal of dissolved solids by reverse osmosis or evaporation, removal of arsenic by oxidation/precipitation, and removal of boron by various methods including ion exchange. The total project cost estimates, with an accuracy of approximately ±25%, ranged from US$ 10 to 78 million in capital cost, with an operation and maintenance (or product) cost ranging from US$ 0.15 to 2.73m−3 of treated water.
© 2023 CNR. Published by Elsevier Ltd. All rights reserved.
Keywords: Geothermal water treatment; Water resources; Desilication; Arsenic; Boron
1. Introduction
With the world entering an age of water shortages and arid farming land, it is increasingly important that we find ways of recycling wastewater. The oil, gas and geothermal industries, for example, extract massive amounts of brine and water from the subsurface, most of which are injected back into underground formations. Holistic approaches to water management are being adopted ever more frequently, and produced water is now being considered as a potential resource. In the oil and gas arena, attempts have been made to convert produced water for drinking supply or other reuses (Doran et al., 1998). Turning oilfield-produced water into a valuable resource entails an understanding of the environmental and economic implications, and of the techniques required to remove dissolved organic and inorganic components from the waters. Treatments of geothermal water and condensate for beneficial use, on the other hand, involve the removal of inorganic components only.
We have explored the technical and economic feasibility of reusing waters and steam condensates from existing and future geothermal power plants. Produced geothermal fluids, especially in arid climates, should be viewed as valuable resources for industry and agriculture, as well as for drinking water supplies. This paper presents the results of laboratory and field pilot studies designed to convert geothermal-produced fluids into beneficially usable water. The preliminary economics of several water treatment strategies are also provided.
2. Design layout
The layout for the treatment strategies (units of operation) have been designed specifically for a nominal 50Mwe geothermal power plant located in an arid climate of the western hemisphere, hereafter referred to as the test plant. The average concentration of constituents in the produced water is shown in Table 1. The amount of spent water from the test flash plant is ∼750 kg/s. The potential amount of steam condensate that could be produced at the plant is ∼350 kg/s. Table 1 includes the composition of the steam condensate derived from well tests. The six treatment cases considered in the study are given in Table 2, together with product flows and unit operations of treatment. Fig. 1 provides simplified schematic layouts of the unit operations for each case.
3. Evaluation of treatment options
In this section the various operations considered for each case are described.
3.1. Arsenic removal
The techniques considered viable for removing traces of arsenic (As) from condensate or from water are ozone oxidation followed by iron co-precipitation or catalyzed photo-oxidation processes (Khoe et al., 1997). Other processes for extracting As from geothermal waters (e.g. Rothbaum and Anderton, 1975; Umeno and Iwanaga, 1998; Pascua et al., 2023) have not been considered in the present study. In the case of the test plant, ozone (O3) would be generated on-site using parasitic power, air and corona-discharge ultra-violet (UV) lamps, and iron in the form of ferric sulfate [Fe2(SO4)3] or ferric chloride (FeCl3) that would be delivered to the geothermal plant. The photo-oxidation processes consist of treating the condensate or water with Fe2+ in the form of ferrous sulfate (FeSO4) or ferrous chloride (FeCl2), or with SO2 photo absorbers. The latter is generated from the oxidation of H2S in turbine vent gas (Kitz and Gallup, 1997).
The photo-oxidation process consists of sparging air through the photo- adsorber-treated fluid, and then irradiating it with UV lamps or exposing it to sunlight to oxidize As3+ to As5+. In the Fe photo-oxidation mode, the Fe2+ is oxidized to Fe3+, which not only catalyzes the oxidation reaction, but also co-precipitates the As. In the SO2 photo-oxidation mode, after oxidizing the As, FeCl3 or Fe2(SO4)3 is added to the water to precipitate the As5+ as a scorodite-like mineral
Table 1
Approximate geothermal water and steam condensate compositions assumed in the study

a Total dissolved solids.
Table 2
Summary of the six cases of geothermal fluid treatment to produce marketable water

a On treatment of water, clays are produced at a rate of 7.4 ton/h.
(FeAsO4·2H2O). In the laboratory and field pilot tests, the photo-absorber and UV dosages were varied to decrease the As concentration in geothermal fluids to below the detection limit of 2 ppb (Simmons et al., 2023). Residual As in the precipitate may be slurry-injected into a water disposal well or fixed/stabilized for land disposal to meet United States Environmental Protection Agency (USEPA) Toxicity Characterization Leach Procedure (TCLP) limits using special cement formulations (Allen, 1996).
3.2. Ion exchange
Strong-base anion exchange resins have been shown to remove traces of As in geothermal fluids provided that the amorphous silica is decreased below its saturation point or the water stabilized against silica scaling by acidification. The ion exchange alternative to As removal by oxidation/precipitation has proven successful in reducing the concentrations of this element to below the limits set for drinking water standards. As part of the present study, laboratory and field columnar tests were successfully conducted with geothermal hot spring water containing 30 ppm As. Pre-oxidation of As3+ is required to achieve acceptable As removal by ion exchange. In these columnar tests, NaOCl and H2O2 were used to pre-treat the hot spring water to oxidize As3+ to As5+. Chloride-rich water, which had been treated with lime (CaOH2) and filtered to reduce amorphous silica to well below its saturation point, successfully regenerated the resin. In the field, and for simplicity of operation, we concluded that ozone/Fe co-precipitation or catalyzed photo-oxidation would be preferred for water treatment over ion exchange as this would eliminate the need to purchase and transport additional chemicals. On the other hand, ion exchange is an attractive option for extracting As from condensate.
Special ion-exchange resins have proven successful in removing boron (B) from geothermal fluids (Recepoglu and Beker, 1991; Gallup, 1995). Hot spring water from the geothermal field, containing 25 ppm B, had its B content decreased to <1 ppm in a laboratory columnar test. The resin was regenerated with sulfuric acid (H2SO4). No deterioration in resin performance was observed up to 10 loading and regenerationcycles.

Fig. 1. Flow chart of the basic unit operations involved in treatment cases 1–6.
3.3. pH adjustment
The majority of the cases considered in this study require adjustment to pH. Adding soda ash (Na2CO3) can increase the buffering capacity of the water and condensate. Soda ash or lime treatment can also be used to enhance precipitation of certain species. Purchased H2SO4, on-site generated sulfurous acid (H2SO3) or on-site generated hydrochloric acid (HCl) can be used to acidify waters to meet reuse requirements or to inhibit silica scaling (Hirowatari, 1996; Kitz and Gallup, 1997; Gallup, 2023). A number of geothermal power plants around the world utilize water acidification to inhibit silica scaling. Unocal Corporation commenced this practice of pH adjustment of hot and cold geothermal fluids in commercial operations in the early 1980s (Jost and Gallup, 1985; Gallup et al., 1993; Gallup, 1996). In water acidification the pH is reduced slightly so as to slow down the silica polymerization reaction kinetics without significantly increasing corrosion rates.
3.4. Cooling ponds
In this water processing option, the water is cooled in open, lined ponds prior to injection or treatment for beneficial use. The flashed water is allowed to flow into the pond where it “ages” for up to 3 days; this is a sufficient length of time to achieve amorphous silica saturation at ambient temperature, which is assumed to be below 20 ◦C most of the year. Adjustment of the water pH to 8.0±0.5 with soda ash or lime enhances water desilication, resulting in undersaturation with respect to amorphous silica (Gallup et al., 2023). At 15 ◦C, the solubility of amorphous silica in the water in our test field is predicted to be about 90 ppm (Fournier and Marshall, 1983). In a large bottle, field water was adjusted from pH 7.2 to 8.1 with soda ash and allowed to cool to 15 ◦C over a period of 90 min. The resultant dissolved silica [Si(OH)4] concentration in the supernatant fluid was 54 ppm (undersaturated by about 40%).
3.5. Filtration
Sand and plate/frame filters were adopted in this study to polish water and dewater sludges, respectively. This does not mean that other filters could not be used in the water treatment project. At the Salton Sea (California, USA) geothermal field, for example, flocculated secondary clarifiers and pressure or vacuum filters have been adopted with success for many years as alternatives to media and plate/frame filters, respectively (Featherstone et al., 1989).
3.6. Multi-stage vacuum-assisted evaporator
In this unit of operation, cool, ponded water is combined with cooled and re-circulated water (from the evaporator heat rejection stages), and pumped to the heat recovery portion of the evaporator system. The cool water provides the thermal sink for the vapors from the final stages of the evaporator concentrate. The inlet water and concentrate flow countercurrent in the evaporator. After flowing through the heat recovery stages, the water temperature has increased somewhat. Most of this heated water is sent to a separate cooling pond before returning to the heat recovery stages. A portion of the heated water continues on through the heat recovery stages; the water also functions as the heat sink for this portion of the process.
After the heat recovery stages, the water is heated with steam and returned to the heat recovery stages for flashing. The water proceeds through the heat recovery and rejection stages until it is fully concentrated. The concentrate is sent to an injection well, while the distillate is collected and re-routed for pH adjustment, as required, before passing to other treatments discussed here. The evaporator has not yet been tested at the field; the present discussion is provided for conceptualization only.
3.7. Reverse osmosis
The reverse osmosis (RO) process removes dissolved salts through fine filtration at the molecular level of water. The RO membrane allows water to pass through but blocks 98% of the salts. The typical RO operating pressure is 2760–3100 kPa, which is achieved by gravity flow from the power plant to the RO unit located 300m downhill. The RO feed is pre-treated with a 2 _m cartridge filter. The rejected fluid is injected into a disposal well, while the permeate can be sent to other treatment units for polishing.The RO unit has not yet been tested at the field; the present discussion is again provided for conceptualization only. However, RO has been successfully tested at the Mammoth Lakes, California, USA, field to recover useable silica (Bourcier et al., 2023).
3.8. Desilication and production of clays
Silica can be eliminated from the water by holding the latter in cooling ponds for up to 3 days. Soda ash or lime can be added to the water to enhance silica precipitation. Laboratory and field jar test experiments showed that desilication of the water can also be achieved by treating with various metal cations at elevated pH to precipitate metal silicates. Below ∼90 ◦Cand at elevated pH (typically 9–10) treatments with caustic soda (NaOH), magnesium hydroxide [Mg(OH)2], lime, strontium hydroxide [Sr(OH)2], barium hydroxide [Ba(OH)2], ferric hydroxide [Fe(OH)3], birnessite [(Na,Ca)0.5(Mn4+,Mn3+)2O4·1.5H2O], copper hydroxide, [Cu(OH)2] and zinc hydroxide [Zn(OH)2] precipitated only amorphous or poorly crystalline metal-rich silicates of little commercial value. Treatment of water with alkaline-earth metals below ∼90 ◦C, except magnesium, tended to co-precipitate metal carbonates. Laboratory reactions conducted at ∼130 ◦C demonstrated that certain metal ions may react with the silica in the water to precipitate crystalline compounds of commercial value. For example, kerolite1 clay was precipitated upon treating synthetic and field waters with magnesium at 130 ◦C, whereas, under similar conditions, sodalite (Na4Al3 Si3O12Cl) and Zeolite P2 were precipitated upon treatment with aluminum hydroxide or sodium aluminate (Gallup et al., 2023; Gallup and Glanzman, 2023). Treatment of waters with a combination of magnesium and iron precipitated hectorite (i.e. a lithium-rich clay mineral of the montmorillonite group).The desilication process designed for the field consists of a crystallizer-clarifier similar to those used at the Salton Sea field (Newell et al., 1989). For kerolite production, magnesium chloride (MgCl2) is added at slightly above stoichiometric proportions (3Mg:4Si) and the pH is increased to ∼10.0 with caustic soda or lime. The crystallizer and clarifier include sludge recirculation to maximize the “seed crystal” effect, thus providing a high surface area for precipitation. After precipitation, the water is clarified, possibly treated further to meet industrial water specifications, cooled to pipeline specifications, and finally sent to a pipeline for transport to the industrial site. The kerolite sludge is dewatered using a filter, as discussed earlier. The dewatered sludge can be dried in a steam-heated kiln or in an arid, but cool environment at the power plant. D