This page provides information on aluminum and iron in water treatment residuals. While providing extensive information on water treatment residuals, there is a section focused specifically on stormwater applications for aluminum and iron in water treatment residuals.
Water treatment residuals are the by-products of water treatment for drinking water. Drinking water treatment residuals are primarily sediment, metal (aluminum, iron or calcium) oxide/hydroxides, activated carbon, and lime removed from raw water during the water purification process (Agyin-Birikorang et al., 2009). Aluminum sulphate (commonly known as alum), ferric chloride and lime are added as flocculants in the water treatment process. This process results in the generation of vast quantities (generally between 10 and 30 mL of WTRs for every litre of water clarified) of a sludge-like waste (or by-product) known as water treatment residuals (WTRs), which require an outlet for their disposal or end use (Dassanayake et al. 2015). The most common method of disposing these WTRs is by sending them to landfills. However, WTRs have many physical and chemical properties that lend them to potential positive reuse routes.
WTR is a by-product generated from the addition of alum or ferric salts used in the coagulation–flocculation process during drinking water treatment, and contains minimal amount of toxic substance (Ippolito et al., 2011) Water treatment residuals are widely available and generally low cost. The use of WTR in bioretention media will prevent this material of ending up in a landfill. WTR was shown to efficiently remove more than 90% of phosphorus from wastewater (Lee et al., 2015) Different sources of WTR will have different adsorption levels due to the amount of Iron and Aluminum coagulants using. The adsorption of phosphorus in stormwater is correlated to the amorphous hydrous metal oxide content in the WTR due to their strong affinity for oxyanions like phosphate.
Phosphorus saturation index (PSI) of two WTRs were relatively low (Fe-WTR: 0.01 and Al-WTR: 0.02). When PSI was less than 1, there would be excess Fe and Al for binding of P or the presence of low P availability, which suggested that WTR has high P adsorption capacity (Elliott et al., 2002).
97% removal with WTR, 7-21% without WTR used in Babatunde et al. (2009)'s study had excellent P removal capacity (23.1 mg P/g)
This section provides information on the physical and chemical properties of spent lime.
Geotechnical behavior of WTR material is a function of the physical and chemical composition of its solid contents and the type, amount, and chemical nature of the pore fluid. As the structure and the solids, content of the WTP residual change, interactions between the solid and the liquid phases such as cementation takes place. This affects geotechnical properties such as compaction, shear strength and permeability. Change in the water content alters the floc structure and the particle sizes of the solids but will also change the ion concentration and complex formation within the residuals.
The following tables provide a summary of physical characteristics of WTRs.
The following table summarizes chemical data for aluminum and iron water treatment residuals (WTRs). Soil reference values (SRVs) and soil leaching values (SLVs) are included and discussed in the next section.
Chemical properties of water treatment residuals. Values are medians for all data collected from the literature.
Link to this table
Chemical | Type of water treatment residual | Tier 2 Soil Reference Value (mg/kg) | Soil Leaching Value (mg/kg) | |
---|---|---|---|---|
Aluminum and iron | Calcium | |||
Aluminum | 51000 | 2211 | 100000 | |
Ammonia | nd | |||
Antimony | nd | 1 | 100 | 10.8 |
Arsenic | 35 | 4.3 | 20 | 11.6 |
Barium | 205 | 30.3 | 18000 | 3370 |
Beryllium | 1.55 | nd | 230 | 5.44 |
Boron | nd | 47000 | 124 | |
Cadmium | 0.8 | 0.2 | 200 | 17.6 |
Calcium | 16700 | 494367 | ||
Carbon | 172500 | 114000 | ||
Chromium III | 52.5 | 1.6 | 650 | 200000 |
Copper | 48 | 2 | 9000 | 1400 |
Iron | 51000 | 956.5 | 75000 | |
Kjeldahl nitrogen | nd | |||
Lead | 49.5 | 2.8 | 700 | 5400 |
Magnesium | 2300 | 8530 | ||
Manganese | 1950 | 73 | 8100 | 260 |
Mercury | nd | nd | 1.4 | 6.58 |
Molybdenum | 2 | 0.2 | 32.2 | |
Nickel | 32.5 | 1.6 | 2500 | 352 |
Nitrogen | 8350 | 330 | ||
Phosphorus | 2995 | 74 | ||
Potassium | 5550 | 846 | ||
Selenium | nd | 1 | 1300 | 5.28 |
Silver | nd | 1 | 1300 | 15.7 |
Sodium | 855 | 335 | ||
Strontium | 45 | 274 | 100000 | 5620 |
Vanadium | 68 | 3 | 250 | 8 |
Zinc | 183 | 5.4 | 75000 | 6010 |
nd=not detected. Reporting limits were below SRVs and SLVs.
SLV assumes 3 foot separation from groundwater and a media depth of 3 feet.
Numbers in bold exceed one or more of the risk criteria.
References: Turner et al. 2019; Elliott et al 2002; Lang, 2009; Shrestha et al, 2019; Elliott et al., 1990; 2009; EPA, 2011; Barr Engineering, 2014; Wang et al., 2014
Concentrations of arsenic, manganese, and vanadium in WTRs exceed Tier 2 Soil Leaching Values (SLVs). The Tier 2 Soil Reference Value (SRV) was exceeded for arsenic. Mixed with a typical bioretention media consisting primarily of a sand-compost mixture, it is unlikely SLVs or SRVs would be exceeded.
A concern with the application of WTRs is the potential hazard of leached Al/Fe (Basta et al., 2000). This may damage the aquatic environment when leachate enters surface water bodies (Codling, 2008). It becomes more critical at low soil pH values (pH < 5.2), whereby Al becomes more soluble (Foy, 1996). The average pH of WTR is 6.5, while typical stormwater runoff is at neutral pH (6.5–7.5, Lee et al., 2015). In this case, the potential of Al leaching would not be prominent according to the distribution of alum species under neutral pH range. Similar research supported that organic-complexed Al forms were the dominant form detected in the supernatant of WTR samples under the neutral pH condition, rather than Al3+ (Agyin-Birikorang and O'Connor, 2009). Another study conducted on grass cropping grown with WTR soil showed no signs of Al toxicity in the plants studied (Oladeji et al., 2009). However, there are still limited studies on the impact of leaching phenomenon especially on WTRs with metal-modified surface.
SPSC (mgP/kg)=(0.15 - PSR)* (Alox + Feox)*31
PSR=Phosphorus sorption ratio = (Pox)/(Alox+Feox)
SPSC values can indicate the risk arising from P loading as well as the inherent P sorption capacity of the soil. The SPSC values range from negative values (for highly P-impacted soils with no remaining P retention capacity) to positive values (for less P-impacted soils, excess P retention capacity)
Studies have shown that the phosphorus removal performance (adsorption) is strongly dependent on stormwater runoff pH (Lee et al., 2015). pH dependence P adsorption may be due to different solubility of Al and Fe (AlOX and FeOX), and changes to the surface charge of Al-hydroxide at different solution pHs (Kim et al., 2003). • As pH was adjusted from 3 to 6, a sharp decrease in Al solubility and a slight increase of P adsorption suggested that the increased surface charge of Al-hydroxides contributed to higher P adsorption capacity. • pH below 6, the formation of Al–P complexes on the surface of Al(OH)3 had a lower solubility than that of Al(OH)3. • pH above 7, the sharp decrease in P adsorption was apparently related to both increased Al solubility and reduced surface charge of Al hydroxides. • Under neutral pH condition, namely in the range of 6.5–7.5 for stormwater (Lee et al., 2015), minimal pH effect on P adsorption capacity was observed and minor fluctuations in pH would not affect P adsorption performance significantly (O'Neill and Davis, 2012).
Relationship with flow concentration: Inconsistent findings regarding inflow concentration of P on the performance of WTR-supplemented bioretention system. Relationship with flow rate: Inconsistent findings regarding inflow rate on the performance of WTR-supplemented bioretention system.
• Median removal efficiencies from 86% to 99%for BSMs containing at least 10% WTR by volume https://edis.ifas.ufl.edu/pdffiles/SS/SS51300.pdf
• Hydraulic conductivity (Ksat) rates for the 60sd/30comp/10wtr treatment were generally lower than the other treatments due to the addition of fine–textured, water treatment residuals (WTR). WTR may also have a high water holding capacity
Juan et al. (2017) observed no negative effects of al- and Fe-WTRs on algal survival and growth. Juan et al. (2016) observed no negative impacts to the bacteria Aliivibrio fischeri. Wang et al. (2014) observed no increase in metal lability under anaerobic conditions, but recommended further study since some metals showed increased lability following anaerobic incubation.
• One of the most cost-effective and sustainable additives is WTR, which is locally sourced and is the only additive tested that is a low-energy byproduct of another industrial process. However, based on the results from the WSU Mesocosm Study, this additive actually decreased system performance. This result was inconsistent with at least one other study that indicate WTR can improve total phosphorus removal (Lucas and Greenway 2011).
Presently, only dewatered WTRs are utilized in stormwater and other land applied applications. These WTRs are considered to be aged. Birikorang and O'Connor (2009) studied the aluminum reactivity of freshly-generated Al-WTRs and observed that reactivity decreased with time. They estimated that reactivity would continue to decrease over a 6 month period for field-applied WTRs. They recommended that application of freshly-generated Al-WTRs with similar physicochemical characteristics as the one utilized for their study should be avoided.
High phosphorus adsorption capability of WTR would lead to an increase of P concentrations in soil and would gradually reach the media saturation after long-term application. Bayley et al. (2008) studied the long-term effects of WTR. It was demonstrated that the WTR maintained its capacity to immobilize the significant inorganic P and remained stable after 13 years application. Similarly, the study reported that WTR reduced total labile P in runoff by 60% compared with control after 7.5 years (Agyin-Birikorang and O'Connor, 2009).