versão On-line ISSN 2411-9717
versão impressa ISSN 0038-223X
J. S. Afr. Inst. Min. Metall. vol.114 no.5 Johannesburg Mai. 2014
BASE METALS PAPERS
R.J. ForzattiI; I. NathaII; L. RouxII; D.A. van den BergI
IHatch Associates, Perth, Western Australia
IIHatch Goba, Woodmead, South Africa
The production of sulphuric acid from sulphur generates heat. The majority of this heat is recovered as steam and is often used to generate electricity. Heat not recovered as steam is rejected to cooling water systems. The design of the turbine, condenser, and cooling water systems impacts the overall water, energy, and environmental footprint of the plant. This review considers a conceptual 2000 t/d sulphuric acid plant with several alternative cooling water systems. The review utilizes Hatch's 4 Quadrant sustainable design tool to compare the alternatives on both an economic and a weighted sustainability scale.
Keywords: sulphuric-burning acid plant, steam, power generation, cooling systems, sustainable design.
Conventional sulphuric acid plants require water for their cooling systems. Cooling is required to reject surplus heat not recovered as steam. Reducing the water consumption lowers the cost of sourcing reliable supplies of clean water as well as the cost associated with treating effluent streams. It also helps improve the sustainability of the acid plant operation by reducing the impact on surrounding communities.
This paper focuses on two broad categories of water-saving options:
► Pretreatment of the make-up water required for evaporative cooling systems
► Replacement of evaporative cooling with dry cooling.
The sustainability aspect of these options is analysed using Hatch's 4 Quadrant design tool.
Technologies integrated with the acid plant design that recover heat from acid will reduce the overall cooling water demand (and hence make-up water consumption). For the purposes of this evaluation, the following acid plant technology options are not considered:
► HRS (by MECS) and HEROS (by Outotec), which produce useful low-pressure steam
► Heat recovery acid coolers, which preheat boiler feed water.
A conventional 2000 t/d sulphur-burning sulphuric acid plant is considered in this paper.
Solid sulphur is delivered in bulk bags or containers and is stored and transferred to a melting and filtration circuit. The sulphur-melting system uses low- and medium-pressure (LP and MP) steam, usually provided from the acid plant steam system. Dirty molten sulphur is filtered to remove ash and other solid impurities. Molten sulphur is also sometimes received instead of solid sulphur if the sulphur source is nearby.
Clean sulphur is transferred to the acid plant where sulphuric acid is produced. The sulphur is burned in a furnace at approximately 1200°C in contact with dry air to produce SO2 gas (approx. 12 vol.%) (King, Davenport, and Moats, 2013). The SO2 gas is oxidized to SO3 gas in contact with a vanadium pentoxide-type catalyst. The SO3 is then absorbed and reacts with the aqueous component of strong sulphuric acid to produce H2SO4. Circulating and product acid cooling is achieved in heat exchangers supplied with cooling water, typically provided from an evaporative cooling tower.
The acid plant steam system is designed to recover the heat generated by the exothermic reactions within the acid plant. Heat is recovered from the sulphur burner off-gas via the production of saturated high-pressure (40 barg or 60 barg) steam in a waste heat boiler. Saturated steam from the boiler flows through a superheater to produce superheated steam, which is fed to a steam turbine generator to produce electricity.
Exhaust steam from the acid plant turbine is condensed, re-pressurized, and returned to the acid plant as boiler feed water. The turbine steam condenser ('surface' condenser) uses cooling water from an evaporative cooling tower. The evaporative cooling tower loses water through evaporation, drift (entrainment), and blowdown. A continuous supply of fresh water is required to make up for these losses.
Demineralized water is used as make-up for losses within the steam circuit (e.g. boiler blowdown and deaerator vent) and for dilution water within the acid plant.
Typical operating parameters for a 2000 t/d sulphur-burning acid plant are shown in Table I.
Steam produced by the acid plant can be:
► Used to generate electricity in a steam turbine generator
► Supplied to other plant consumers (e.g. heating for hydrometallurgical equipment)
► Exported to other customers
Sulphur-burning acid plants produce more electricity than they consume when all of the steam is sent to a steam turbine generator. This excess electricity can be:
► Used to operate other facilities within the plant
► Sold to the market.
The water balance for a conventional 2 000 t/d sulphur-burning acid plant based on evaporative cooling is given in Table II.
The following is noted:
► The single largest water loss is due to water evaporation in the cooling tower
► Other losses include drift and blowdown. The blowdown indicated is calculated based on three cycles of concentration, assuming fresh water has a total dissolved solids (TDS) of 300-400 ppm.
Make-up water pretreatment
Treatment of make-up water to the cooling tower can be used to change the water chemistry to achieve higher cycles of concentration, thereby reducing blowdown.
Softening of the cooling water make-up can be used to remove several dissolved salts that cause scale formation such as calcium, magnesium, barium, strontium, and iron. Other scale-forming components, such as silica, are not removed.
A water softener consists of a vessel filled with cationic resin that exchanges (removes) the dissolved species from the water and replaces these with sodium. Cooling systems fed with high hardness water sources will benefit most from having the make-up water softened.
As an example, a water source with a feed hardness of approximately 300 mg/L (expressed as CaCO3), pH of approximately 8, and alkalinity of approximately 300 mg/L as CaCO3 might normally be concentrated three times; a cycles of concentration (COC) of 3. The scaling tendency of this water, at a COC of 3, is within the typical range that can be managed with a scale inhibitor. This same make-up water source could be concentrated more than seven times if first softened, representing a reduction in cooling water make-up of approximately 20%.
Filtration to remove suspended solids from water can be applied to either the entire fresh water make-up stream for the acid plant or to the cooling tower water recirculation stream. Filtration of the cooling circuit make-up water is generally considered where there is a high level of suspended solids in the feed. These solids, if not removed, can cause fouling within the cooling water circuit, which lowers cooling efficiency and increases the pressure drop through the piping system. The solids can also accelerate corrosion within the water circuit if they are abrasive. Unfiltered particles can serve as nucleation sites for biological growth. Filtration is required ahead of a cooling water softening system.
It should be noted that the majority of suspended solids in the cooling circuit are generated within the cooling circuit rather than introduced in the make-up water, as the cooling water is in contact with surrounding air in open-circuit evaporative cooling towers. Internal sources of solids include pipe corrosion products, biological growth material, and dust introduced from the air as it contacts the water in the cooling water tower. For this reason there is often more merit in filtering the cooling water itself rather than the make-up water alone.
Demineralized water treatment
Demineralized water is used as make-up for losses within the steam circuit (e.g. boiler blowdown and de-aerator vent) and for dilution water within the acid plant. Typical demineralized water system configurations include:
► Reverse osmosis (RO) only
► Ion exchange (IX) with a decarbonator tower
► RO followed by polishing IX.
Waste generation as a percentage of feed is typically 30%. The selected configuration is dependent on the site raw water quality, and can be optimized to provide water savings. These savings will, however, be small compared to potential savings in the cooling system.
The following cooling technologies are discussed:
► Evaporative cooling towers
► Dry cooling technologies
► Hybrid cooling towers.
Evaporative cooling towers
Evaporative cooling tower designers have identified many ways to reduce the overall water losses from these systems. Some of these include (EnduroSolv, 2012):
► Optimizing water chemistry to reduce scaling, corrosion, and biological growth, subsequently increasing the cycles of concentration and decreasing blowdown. This includes the use of automated chemical dosing systems
► Better operating procedures and equipment to monitor and control blowdown
► Use of high-efficiency drift eliminators and equipment to recapture drift
► Optimizing the selection and amount of fill inside the tower, which affects the heat transfer efficiency of the tower
► Automatic blowdown based on conductivity to avoid unnecessary blowdown in cases where the feed water quality is better than initially anticipated
► Minimizing unintentional water losses from leaks or overflow (i.e. faulty level control resulting in addition of excess make-up water)
► Special tower design considerations to reduce particulates, debris, and cooling water exposure to sunlight.
These advances in cooling tower design and control have resulted in minor water savings. Fundamentally, the cooling is provided through the evaporation of water, and hence there is an inherent loss of water when adopting this technology.
Dry cooling technologies
Dry cooling technologies work by heat exchange to air and do not rely on the evaporation of water to provide cooling. Applicable technologies for a sulphur-burning acid plant include:
► Air-cooled condenser (ACC) on the steam turbine exhaust,
► Fin-fan coolers to supply cooling water to the absorbing acid coolers.
Dry cooling technologies are dependent on the difference between ambient temperature and the cooling water temperature. In locations with high ambient temperatures, the temperature difference will be lower, leading to significantly increased dry cooling unit size.
Air-cooled condenser (ACC)
An ACC (Figure 1) is comprised of finned tube bundles grouped together into modules and mounted in an A-frame configuration on a concrete or steel support structure. Steam from the turbine exhaust enters the top of the condenser via a steam duct and manifold. Steam flows downward through two or three rows of finned tubes. Condensate is recovered inside bonnet header boxers connected to a hot water tank. The axial-flow forced-draught fan is fixed in the module and forces the atmospheric cooling air across the condensate area of the fin tubes.
Dry cooling in an ACC requires a significant temperature difference to provide adequate heat exchange to the surrounding air. Typically, the cooling water supply temperature will be 25°C to 30°C higher than the ambient air temperature. This results in a higher condenser outlet temperature which in turn raises the condenser pressure, causing the steam turbine to operate less efficiently. An ACC can also be impacted by wind direction and speed as well as proximity to large buildings. More recent advancements in ACC technology include (Mortensen, 2011; Maulbetsch, DiFilippo, and Zammit, 2008):
► Wind guide vane technologies to mitigate wind impacts including walls, screens, lips, and louvers. CFD wind flow modelling is also used to optimize the location and arrangement of the ACCs
► Improved finned tube bundle designs for higher heat transfer efficiency and lower pressure drop
► Pre-cooled ACC, which uses the evaporative cooling effect of a fog spray into the upstream side of the ACC fans. The expected water consumption is approximately 75% less than equivalent evaporative-only cooling. The advantage of this system is that it reduces air temperatures to the fans on very hot days.
The application of an ACC for cooling of turbine exhaust is widely adopted on many steam turbine systems. Eskom, the South African power utility, has adopted the largest ACCs currently in operation in the world for the Matimba, Kendal, and Majuba power stations (Eskom, 2010).
Fin-fan coolers (as depicted in Figure 2) include one or more bundles of finned tubes connected by headers with an air-moving device such as an axial fan located above (induced draught) or below (forced draught) the tube bundle. Cooling water flows through the tubes and heat is exchanged to ambient air. The fin-fan circuit uses demineralized quality water and is closed-loop (not open to atmosphere), eliminating the need for a continuous water supply.
Fin-fan coolers require a significant temperature difference to provide adequate heat exchange to the surrounding air. Typically, the cooling water supply temperature will be 15°C higher than the ambient air temperature. A fin-fan cooler can be used to provide cooling for the drying and absorption sections of an acid plant because the absorbing acid heat exchangers target approximately 70°C. The product acid heat exchangers target 35-40°C, which cannot be consistently achieved in most locations using fin-fan coolers.
Hybrid cooling towers
Hybrid cooling towers have an air-to-air (dry cooling) section and evaporative cooling section operating in series. As shown in Figure 3, heated cooling water first passes through the dry section, where part of the heat load is removed by an air current, typically induced via fans. After passing the dry section, water is further cooled in the wet section of the tower, which can be cooled in a conventional open evaporative circuit or closed circuit (tubes are cooled with water on the outside).
The resulting heat transfer performance is similar to a wet cooling tower, with the dry cooler providing the advantage of protecting the working fluid from environmental exposure and contamination. Depending on the hybrid tower configuration, the water consumption lies between the wet and dry circuit options reported in this evaluation.
Cooling technology and steam turbine electricity generation
The turbine exhaust cooling system performance directly affects the amount of electricity produced by the steam turbine generator. The lower the condenser outlet temperature, the lower the condenser outlet pressure and turbine exhaust back-pressure. A lower turbine exhaust back-pressure increases turbine output.
The impact of increasing ambient temperature (dry-bulb temperature) on the turbine output is shown in Figure 4, similar to the loss of electricity generation reported by others (US Department of Energy. 2009). It compares the base case (turbine surface condenser on evaporative cooling) to the turbine ACC option. For both options, the equipment is sized to remove the full heat load over the full ambient temperature range. As can be seen, with an ACC, the power generation is lower because an ACC runs at a higher temperature than a surface condenser.
The following cooling circuit water-saving options are compared in this paper:
► Base case - evaporative cooling (no pretreatment)
► Evaporative cooling (with pretreatment)
► Dry cooling (no pretreatment) with the following variants:
- ACC on turbine exhaust
- ACC on turbine exhaust, and fin-fans on absorbing acid circuit.
These options are shown in the schematic in Figure 5.
All options assume steam is generated in the sulphur burner waste heat boiler and electricity generated in a steam turbine generator. Where dry cooling options are considered, the balance of cooling is provided by evaporative cooling.
Table III summarizes the make-up water consumption and heat removal duties of the cooling circuit options.
Evaluation of options
The Hatch 4QA approach compares the economic and sustainability impacts of alternative project options.
Order-of-magnitude capital and operating cost estimates were developed for each cooling system option and compared to the base case (evaporative cooling with no pretreatment). A summary of the comparison is presented in Table IV.
Sulphur-burning acid plant emissions include:
► Gaseous emissions-sulphur dioxide, nitrogen oxides, and acid mist in tail gas
► Liquid effluents-waste heat boiler and cooling circuit blowdown, demineralized water treatment plant waste, plant washings, spillages and leakages
► Solid effluent-sulphur filter cake residues and spent converter catalyst
► Noise pollution-main blower and turbine.
Commitments by major corporations as well as government regulatory requirements have resulted in the development of several new and cost-effective technologies to efficiently reduce gaseous emissions from an acid plant. Noise pollution has been addressed with suitable sound reduction measures such as acoustic insulation, enclosures, and silencers.
The treatment and disposal of liquid and solid effluents is facing more stringent controls through public awareness and government regulations. Reduction in water consumption is potentially the greatest beneficial impact on local communities and the environment. Furthermore, a reduction in the cooling water blowdown will reduce the plant effluent and ultimately reduce the impact on the overall plant effluent catchment area.
Water, footprint, power, and waste
Four sustainability criteria were identified to quantitatively compare the different water saving options, namely: water intensity, power intensity, waste intensity, and footprint. A comparison of these criteria is given in Table V. Water and waste intensity are calculated from the mass balance; power and footprint values are estimated from recent project experience. The values shown are for the cooling system only and exclude the criteria associated with the remainder of the acid plant.
The following is noted with respect to each of the intensity factors:
► Water intensity
- Pretreatment provides water savings when applied to the base case
- Dry cooling options provide the lowest overall water consumption.
► Power intensity
- Evaporative cooling includes power to operate the cooling tower fans and the cooling water supply pumps
- Base case with pretreatment has a slightly higher power usage due to more pumping required between upfront unit operations
- Dry cooling includes power to operate the fans only. Although electricity consumption is lower for dry cooling options, the power intensity shown in Table V is higher because it has been calculated taking into account a reduction in turbine electricity generation of 1.5 MWe (see previously).
► Footprint intensity
- The footprint of the base case with pretreatment is comparable with the base case
- Dry cooling options require more footprint for the same cooling duty.
► Waste intensity
- Evaporative cooling options generate more blowdown and therefore increased waste for effluent treatment
- There is a large reduction in waste generated when pretreatment is included ahead of the evaporative cooling tower.
Four Quadrant Analysis
Hatch developed the Four Quadrant Analysis (4QA) approach to compare project options using the economic and sustainability impacts. The 4QA tool plots each option compared to a base case:
► The x-axis is a cost ratio, with a lower cost ratio representing a lower cost option relative to the base case
► The y-axis is a sustainability ratio, which compares the intensity of the option to the base case. A lower sustainability ratio is preferred, which indicates a lower impact on the environment.
The cost ratio (CR) is calculated for each option, relative to the base case (BC), using the following equation:
The CR is the sum of the annual operating cost and the annual capital cost repayments, based on a nominal 5-year repayment (compounded monthly). Any credits received for selling electricity to the market have not been included into the evaluation.
The sustainability ratio (SR) is calculated for each option, relative to the base case (BC), using the following equation:
The weightings can be adapted based on the general importance of each criterion. The weightings used in this evaluation (Table VI) have placed a high importance on water intensity as many plants strive to reduce water consumption. The relative weightings will be site-specific, for example arid locations may consider water impacts more important and footprint less important.
The 4QA shows that for typical site locations:
► Base case with pretreatment has a better sustainability ratio than the base case; however, the cost ratio will increase by 19%. This is mainly due to the high operating cost associated with reagent consumption for cationic resin regeneration. Alternative pretreatment options can be investigated with lower reagent usage and allowing increased cycles of concentration in the cooling tower
► Dry cooling offers considerable improvements to the sustainability ratio, but the cost ratio will increase by 10% for ACC and 4% for ACC and fin-fan.
The cost and sustainability ratios can be affected by several factors, some of which are briefly considered in the following sensitivity analyses.
Fresh water supply costs are location-dependent. The relative water supply costs used for this sensitivity are based on:
► Low water cost (US$0.2 per m3)-typical for locally available water source of good quality (e.g. dam located close to plant) with no additional extraction charges
► Average water cost (US$1.0 per m3)-typical for water sources located a reasonable distance from the plant, requiring minor infrastructure to be built and some minor water treatment on site (e.g. sand filtration)
► High water cost (US$3.0 per m3)-typical for water sources located at a considerable distance or water of poor quality requiring significant treatment (e.g. reverse osmosis). High water cost would also apply for water that is local and of good quality, but with a high extraction charge.
Table VIII summarizes the cost ratios for the water cost sensitivity analyses. The sustainability ratios remain unchanged.
The sensitivity analysis shows that:
► Reduced water costs (US$0.2 per m3) increase the cost ratio of dry cooling options, making them unfavourable compared to the base case
► Increased water costs (US$3 per m3) reduce the cost ratio of dry cooling options to less than the base case (by as much as 23%)
► This supports the general observation that as water extraction costs increase, dry cooling options are preferred
► The base case with pretreatment has the highest overall cost ratio, which corroborates the data in Table VII. At reduced water costs (US$0.2 per m3) the base case with pretreatment compares well with the dry cooling options, but becomes the least favourable overall at increased water costs (US$3 per m3).
The 4QA was updated with the low and high water costs in Figure 7.
Electricity supply costs are also location-dependent. The relative electricity supply costs used for this sensitivity are based on:
► Low electricity cost (US$0.05 per kWh): for locations with abundant low cost electricity, e.g. hydroelectricity
► Average electricity cost (US$0.1 per kWh): for locations with a typical mixed electricity supply, e.g. a mix of coal, renewable, and gas-fired power stations
► High electricity cost (US$0.3 per kWh): for locations where electricity is generated on site, e.g. local diesel or gas-fired generators.
Table IX summarizes the cost ratios for the electricity cost sensitivity analyses.
The sensitivity analysis shows that:
► Lower electricity costs have a minor impact on the 4QA plot
► At increased electricity cost, the advantage of evaporative cooling is clear, due to the increased turbine electricity output
► The base case with pretreatment compares well with the dry cooling options at increased electricity costs; however, it becomes the least favourable at the lower electricity costs.
The 4QA was updated with the low and high electricity costs in Figure 8.
Sustainability criteria weightings
The sustainability criteria weightings can be adjusted to suit the plant location and requirement for generating electricity, thereby impacting the sustainability ratio. As an example, the weightings can be adjusted as water or power becomes more or less important to the local community. Furthermore, additional sustainability criteria can be included, such as:
► Specific reagent consumption (e.g. high RO membrane costs)
► Downstream impact and stewardship (qualitative)
► Operability and maintainability (qualitative)
► Social, for example using local labour (qualitative)
► Government and externals (qualitative)
► Emissions (qualitative).
The 4QA approach is flexible and can be customized and adapted to meet specific project criteria.
The Hatch 4QA approach compares the economic and sustainability impacts of alternative project options. It can be used for technology and site selection from concept through to feasibility studies and beyond. The 4QA additionally serves as a risk management tool to quantify the impacts of varying sustainability criteria and input costs.
For the water-savings options considered in this paper, it is the acid plant location that largely determines the sustain-ability and cost ratios. Key findings include:
► The cost ratio of evaporative cooling is generally lower, provided there is good-quality and low- to medium-cost water available. The sustainability ratio is generally higher due to the high water consumption, which can make evaporative cooling unfavourable even at sites with low water costs (Maulbetsch, DiFilippo, and Zammit, 2008)
► The cost ratio of the base case with pretreatment is the highest overall due to the high reagent usage. Optimizing the make-up water chemistry by adjusting the pH and adding scale inhibitors might be a more efficient way of increasing the cycles of concentration, but needs to be investigated on a case-by-case basis
► Reverse osmosis (RO) can also offer water savings, but these could be offset by the RO waste generation
► The sustainability ratio of the base case with pretreatment is lower than without pretreatment due to the decreased water usage
► Dry cooling options have a higher capital outlay, but can have lower operating costs in locations where water extraction costs are high. The sustainability ratio is generally lower for dry cooling, due to lower water consumption and similar power consumption to evaporative cooling
► Acid plant electricity generation capacity is lower with dry cooling options and is worsened during higher ambient temperature conditions. For large acid plants generating electricity that is sold to market, this will adversely impact plant revenue.
The authors wish to acknowledge the contributions of Dr Matthew King and Rusi Kapadia, and Hatch for permission to publish this paper.
EnduroSolv. 2012. Cooling Tower Water Saving Strategies, http://endurosolv.com/pdf/cooling_tower_savings_strategies.pdf [Accessed 25 Apr. 2014]. [ Links ]
EPRI. 2002. Comparison of Alternate Cooling Technologies for California Power Plants: Economic, Environmental and Other Trade-offs. Electric Power Research Institute, Aplo Alto, CA, and Californian Energy Commission, Sacramento, CA. [ Links ]
Eskom. 2010. Factsheet, General Communication CO 0005, Revision 7, October 2010. http://wwww.eskom.co.za/AboutElectricity/FactsFigures/Documents/CO000 5CoolingTechniquesRev10.pdf [Accessed 12 June 2013]. [ Links ]
King, M.J., Davenport, W.G., and Moats, M.S. 2013. Sulphuric Acid Manufacture: Analysis, Control and Optimization. 2nd edn. Elsevier, Burlington, MA. ISBN: 978-0-08-098220-5. [ Links ]
Maulbetsch, J.S., DiFilippo, M.N., and Zammit, K.D. 2008. spray cooling for performance enhancement of air-cooled condensers. EPRI Advanced Cooling Workshop, Carolina. [ Links ]
Mortensen, K. 2011. Improved Performance of an Air Cooled Condenser (ACC) Using SPX Wind Guide Technology at Coal-Based Thermoelectric Power Plants. US Department of Energy. http://www.netl.doe.gov/File%20Library/Research/Coal/ewr/water/Proj519.pdf [Accessed 24 April 2014]. [ Links ]
US Department of Energy. 2009. Concentrating Solar Power Commercial Application Study: Reducing Water Consumption of Concentrating Solar Power Electricity Generation. Report to Congress in response to Energy Independence and Security Act of 2007. (Pub. L. No. 110-140), http://www.nrel.gov/csp/pdfs/csp_water_study.pdf [Accessed 25 April 2014]. [ Links ]
Wilson, B. 2011. Detail Engineering and Layout of Piping Systems. 1st edn. On Demand Books, New York. ISBN : 9781926633183. [ Links ]