32
© IWA Publishing 2014 Water Science & Technology
|
70.1
|
2014
Evaluation of barrier materials for removing pollutants from groundwater rich in natural organic matter I. Kozyatnyk, P. Haglund, L. Lövgren, M. Tysklind, A. Gustafsson and N. Törneman
ABSTRACT Permeable barriers are used for passive remediation of groundwater and can be constructed from a range of materials. The optimal material depends on the types of contaminants and physicochemical parameters present at the site, as well as the hydraulic conductivity, environmental safety, availability, cost and long-term stability of the material itself. The aim of the presented study was to test a number of materials for their ability to remove heavy metals and organic pollutants from groundwater with a high (140 mg L1) content of natural organic matter (NOM). The following materials were included in the study: sand, peat, fly ash, iron powder, lignin and combinations thereof. Polluted water was fed into glass columns loaded with each sorbent and the contaminant removal efficiency of the material was evaluated through chemical analysis of the percolate. Materials based on fly ash and zero-valent iron were found to be the most effective for heavy metal removal, while fly ash and peat were the most effective for removing aliphatic compounds. Filtration through lignin and peat led to leaching of NOM. Although the leaching decreased over time, it remained high throughout the experiments. The results indicate that remediation of contaminated
I. Kozyatnyk (corresponding author) P. Haglund L. Lövgren M. Tysklind Department of Chemistry, Umeå University, S-901 87 Umeå, Sweden E-mail: [email protected] A. Gustafsson MoRe Research AB, Box 70, 891 22 Örnsköldsvik, Sweden N. Törneman SWECO, Hans Michelsensgatan 2, Box 286, 201 22 Malmö, Sweden
land at disused industrial sites is a complex task that often requires the use of mixed materials or a minimum of two sequential barriers. Key words
| adsorption, groundwater, heavy metals, natural organic matter, organic pollutants, permeable barriers
INTRODUCTION Pollution generated by industrial activities often causes social, health, and environmental problems, such as the contamination of groundwater by organic and inorganic pollutants. The installation of permeable barriers (PBs) downstream of a contamination plume can remediate affected groundwater more cheaply than the conventional method of excavating the contaminated soil. Furthermore, PBs can passively remediate groundwater and prevent groundwater contamination even in extreme environments (Snape et al. ; Amos & Younger ). Thus, they are viable alternatives to the established pump-and-treat technology. The main component of a PB is a reactive or sorbent material. However, no single material can remove all types of pollutants. Therefore, a material with the ability to remove the contaminants present at the site in question has to be selected when installing any PB system. Other doi: 10.2166/wst.2014.192
factors that influence the choice of sorbent include the reactivity, hydraulic conductivity, environmental safety, availability, cost, and long-term stability of the sorbent material itself. Several materials are commonly used in the construction of PBs. Zero-valent iron (ZVI) is most frequently used, followed by granular activated carbon. Other materials used as sorbents include microorganisms, natural zeolites, peat, phosphates, limestone and amorphous ferric oxide (Richardson & Nicklow ). Aliphatic hydrocarbons, which are common pollutants at oil-contaminated sites, can be removed by peat (Guerin et al. , ), zeolites (Northcott et al. ; Vignola et al. ), activated carbons (Hornig et al. ) or biologically active barriers (Kao & Wang ; Vesela et al. ). Polyvalent heavy metals (e.g. Cr, As) can be removed by
33
I. Kozyatnyk et al.
|
Evaluation of barrier materials for removing pollutants from groundwater
exposure to ZVI. Redox reactions taking place at the iron surface affect the solubility of the heavy metals, which are then sequestered through precipitation or adsorption on iron oxides (Lee & Wilkin ; Mak & Lo ). Metals can also be removed by ion-exchange and adsorption on fly ash (González et al. ; Prasad & Mortimer ), zeolites (Vignola et al. ; Prasad et al. ), mining products (Smyth et al. ; Conca & Wright ) or activated carbon (Lakatos et al. ). In the presented study we evaluated the ability of several PB materials to remove heavy metals and organic pollutants (aliphatic and aromatic hydrocarbons) from groundwater at a 10 ha industrial site in northern Sweden. The soil at the site is also rich in natural organic matter (NOM), particularly fulvic substances, and an adjacent river drains untreated groundwater from the site into the Bothnian Sea. The study is part of a broader project to develop economical and environmentally sustainable methods to prevent the migration of hazardous compounds from both active and abandoned industrial sites.
METHODS Water with high levels of dissolved metals and organic pollutants was extracted for the sorption experiments from two groundwater wells at a 10 ha industrial site in northern Sweden. Water from both wells was pumped into and mixed in a 1 m3 tank, which was flushed with an argon–nitrogen gas mixture to prevent oxidation of metal ions. On the first day of each experiment the gas outlet was situated above the water-line; however, for the remainder of the experiment the gas outlet was submerged. The chemical properties of the composite water are summarized in Table 1 and compared to the water quality recommendation of the European council directive 98/83/EC (European Council ), as well as to the Swedish Geological Survey’s regulations on groundwater status classification and environmental quality standards for groundwater, SGU-FS 2008:2 ().
Experimental design Five sorbent materials were selected based on previous studies on NOM adsorption processes (Dries et al. ; Viipsi et al. ; Liu & Lo ) and results of field experiments on the removal of metal and organic contaminants (Richardson & Nicklow ). The materials used in this study are summarized in Table 2.
Water Science & Technology
|
70.1
|
2014
The surface area and pore size distribution of each tested material in the dry state were determined by the conventional Brunauer–Emmett–Teller (BET, Brunauer et al. ) and Barrett–Joyner–Halenda (BJH, Barrett et al. ) methods. Prior to analysis, ∼0.5 g of material was transferred to a 9.5 mm (inner diameter) sample tube and degassed at 120 C for 2 h in a Micromeritics SmartPrep unit. A TriStar 3000 automated nitrogen sorption/desorption instrument (Micromeritics, Norcross, GA, USA) was used for 12-point determinations of the BET specific surface area between 0.05 and 0.28 relative pressure, while the BJH pore size distribution was determined by measurements at 77 points (Wikberg et al. ). Six glass columns (internal diameter 19 cm) were packed with sorbent material. Each column was packed with a 5 cm layer of sand at the bottom, followed by a 40 cm thick layer of a 1:1 mixture of a single sorbent material and sand, and finally a 5 cm layer of sand on top, resulting in an overall bed height of 50 cm. Columns containing mixtures of peat, fly ash and sand (1:1:1) were prepared in a similar manner, with the mixture packed between two layers of sand. The end fittings of the columns contained grooves to distribute the flow evenly and were equipped with a fine metal mesh to keep the sorbent material in place. The polluted water was pumped through the columns materials (from bottom up) and the pollutant reduction was evaluated. The experiments were conducted using three parallel columns and, thus, two sets of experiments were required, see Figure 1. Samples of the percolate were taken out in a sequence representing different liquid to sorbent volume ratios (L/S), i.e. 0, 0.5, 2 and 10. Pump flow through the columns was set at L/S ¼ 1 per 24 hours. Water coming out from the columns was collected in glass bottles which had been flushed with 2 mL of concentrated HCl in an attempt to get a pH of the liquor below 3.5. The water samples were stored in a refrigerator room with a temperature of around þ5 C (±2 C). At the end of each experiment, the remaining column content was carefully extracted and four samples were taken (at 10, 20, 30 and 40 cm bed height), for further analysis. W
W
W
Sample analysis All collected samples were sent to a commercial laboratory for analysis (Eurofins Environment Sweden, Lidköping, Sweden). The substances measured in liquid samples
34
Table 1
I. Kozyatnyk et al.
|
|
Evaluation of barrier materials for removing pollutants from groundwater
Water Science & Technology
|
70.1
|
2014
Chemical properties of groundwater before and after treatment with six different PB sorbents. For comparison, water quality standards set out by European Council Directive 98/83/EC (European Council 1998) and Swedish Geological Survey SGU-FS 2008:2 (2008) are included, where available Concentrations in filtrate water (final sampling)
Quality standards
Raw water
Water in tank before treatment
Fly ash
Peat
Sand
Fly ash/ Peat
Iron
Lignin
Aliphatic (C10C12), mg L1
2.5
0.19
© IWA Publishing 2014 Water Science & Technology
|
70.1
|
2014
Evaluation of barrier materials for removing pollutants from groundwater rich in natural organic matter I. Kozyatnyk, P. Haglund, L. Lövgren, M. Tysklind, A. Gustafsson and N. Törneman
ABSTRACT Permeable barriers are used for passive remediation of groundwater and can be constructed from a range of materials. The optimal material depends on the types of contaminants and physicochemical parameters present at the site, as well as the hydraulic conductivity, environmental safety, availability, cost and long-term stability of the material itself. The aim of the presented study was to test a number of materials for their ability to remove heavy metals and organic pollutants from groundwater with a high (140 mg L1) content of natural organic matter (NOM). The following materials were included in the study: sand, peat, fly ash, iron powder, lignin and combinations thereof. Polluted water was fed into glass columns loaded with each sorbent and the contaminant removal efficiency of the material was evaluated through chemical analysis of the percolate. Materials based on fly ash and zero-valent iron were found to be the most effective for heavy metal removal, while fly ash and peat were the most effective for removing aliphatic compounds. Filtration through lignin and peat led to leaching of NOM. Although the leaching decreased over time, it remained high throughout the experiments. The results indicate that remediation of contaminated
I. Kozyatnyk (corresponding author) P. Haglund L. Lövgren M. Tysklind Department of Chemistry, Umeå University, S-901 87 Umeå, Sweden E-mail: [email protected] A. Gustafsson MoRe Research AB, Box 70, 891 22 Örnsköldsvik, Sweden N. Törneman SWECO, Hans Michelsensgatan 2, Box 286, 201 22 Malmö, Sweden
land at disused industrial sites is a complex task that often requires the use of mixed materials or a minimum of two sequential barriers. Key words
| adsorption, groundwater, heavy metals, natural organic matter, organic pollutants, permeable barriers
INTRODUCTION Pollution generated by industrial activities often causes social, health, and environmental problems, such as the contamination of groundwater by organic and inorganic pollutants. The installation of permeable barriers (PBs) downstream of a contamination plume can remediate affected groundwater more cheaply than the conventional method of excavating the contaminated soil. Furthermore, PBs can passively remediate groundwater and prevent groundwater contamination even in extreme environments (Snape et al. ; Amos & Younger ). Thus, they are viable alternatives to the established pump-and-treat technology. The main component of a PB is a reactive or sorbent material. However, no single material can remove all types of pollutants. Therefore, a material with the ability to remove the contaminants present at the site in question has to be selected when installing any PB system. Other doi: 10.2166/wst.2014.192
factors that influence the choice of sorbent include the reactivity, hydraulic conductivity, environmental safety, availability, cost, and long-term stability of the sorbent material itself. Several materials are commonly used in the construction of PBs. Zero-valent iron (ZVI) is most frequently used, followed by granular activated carbon. Other materials used as sorbents include microorganisms, natural zeolites, peat, phosphates, limestone and amorphous ferric oxide (Richardson & Nicklow ). Aliphatic hydrocarbons, which are common pollutants at oil-contaminated sites, can be removed by peat (Guerin et al. , ), zeolites (Northcott et al. ; Vignola et al. ), activated carbons (Hornig et al. ) or biologically active barriers (Kao & Wang ; Vesela et al. ). Polyvalent heavy metals (e.g. Cr, As) can be removed by
33
I. Kozyatnyk et al.
|
Evaluation of barrier materials for removing pollutants from groundwater
exposure to ZVI. Redox reactions taking place at the iron surface affect the solubility of the heavy metals, which are then sequestered through precipitation or adsorption on iron oxides (Lee & Wilkin ; Mak & Lo ). Metals can also be removed by ion-exchange and adsorption on fly ash (González et al. ; Prasad & Mortimer ), zeolites (Vignola et al. ; Prasad et al. ), mining products (Smyth et al. ; Conca & Wright ) or activated carbon (Lakatos et al. ). In the presented study we evaluated the ability of several PB materials to remove heavy metals and organic pollutants (aliphatic and aromatic hydrocarbons) from groundwater at a 10 ha industrial site in northern Sweden. The soil at the site is also rich in natural organic matter (NOM), particularly fulvic substances, and an adjacent river drains untreated groundwater from the site into the Bothnian Sea. The study is part of a broader project to develop economical and environmentally sustainable methods to prevent the migration of hazardous compounds from both active and abandoned industrial sites.
METHODS Water with high levels of dissolved metals and organic pollutants was extracted for the sorption experiments from two groundwater wells at a 10 ha industrial site in northern Sweden. Water from both wells was pumped into and mixed in a 1 m3 tank, which was flushed with an argon–nitrogen gas mixture to prevent oxidation of metal ions. On the first day of each experiment the gas outlet was situated above the water-line; however, for the remainder of the experiment the gas outlet was submerged. The chemical properties of the composite water are summarized in Table 1 and compared to the water quality recommendation of the European council directive 98/83/EC (European Council ), as well as to the Swedish Geological Survey’s regulations on groundwater status classification and environmental quality standards for groundwater, SGU-FS 2008:2 ().
Experimental design Five sorbent materials were selected based on previous studies on NOM adsorption processes (Dries et al. ; Viipsi et al. ; Liu & Lo ) and results of field experiments on the removal of metal and organic contaminants (Richardson & Nicklow ). The materials used in this study are summarized in Table 2.
Water Science & Technology
|
70.1
|
2014
The surface area and pore size distribution of each tested material in the dry state were determined by the conventional Brunauer–Emmett–Teller (BET, Brunauer et al. ) and Barrett–Joyner–Halenda (BJH, Barrett et al. ) methods. Prior to analysis, ∼0.5 g of material was transferred to a 9.5 mm (inner diameter) sample tube and degassed at 120 C for 2 h in a Micromeritics SmartPrep unit. A TriStar 3000 automated nitrogen sorption/desorption instrument (Micromeritics, Norcross, GA, USA) was used for 12-point determinations of the BET specific surface area between 0.05 and 0.28 relative pressure, while the BJH pore size distribution was determined by measurements at 77 points (Wikberg et al. ). Six glass columns (internal diameter 19 cm) were packed with sorbent material. Each column was packed with a 5 cm layer of sand at the bottom, followed by a 40 cm thick layer of a 1:1 mixture of a single sorbent material and sand, and finally a 5 cm layer of sand on top, resulting in an overall bed height of 50 cm. Columns containing mixtures of peat, fly ash and sand (1:1:1) were prepared in a similar manner, with the mixture packed between two layers of sand. The end fittings of the columns contained grooves to distribute the flow evenly and were equipped with a fine metal mesh to keep the sorbent material in place. The polluted water was pumped through the columns materials (from bottom up) and the pollutant reduction was evaluated. The experiments were conducted using three parallel columns and, thus, two sets of experiments were required, see Figure 1. Samples of the percolate were taken out in a sequence representing different liquid to sorbent volume ratios (L/S), i.e. 0, 0.5, 2 and 10. Pump flow through the columns was set at L/S ¼ 1 per 24 hours. Water coming out from the columns was collected in glass bottles which had been flushed with 2 mL of concentrated HCl in an attempt to get a pH of the liquor below 3.5. The water samples were stored in a refrigerator room with a temperature of around þ5 C (±2 C). At the end of each experiment, the remaining column content was carefully extracted and four samples were taken (at 10, 20, 30 and 40 cm bed height), for further analysis. W
W
W
Sample analysis All collected samples were sent to a commercial laboratory for analysis (Eurofins Environment Sweden, Lidköping, Sweden). The substances measured in liquid samples
34
Table 1
I. Kozyatnyk et al.
|
|
Evaluation of barrier materials for removing pollutants from groundwater
Water Science & Technology
|
70.1
|
2014
Chemical properties of groundwater before and after treatment with six different PB sorbents. For comparison, water quality standards set out by European Council Directive 98/83/EC (European Council 1998) and Swedish Geological Survey SGU-FS 2008:2 (2008) are included, where available Concentrations in filtrate water (final sampling)
Quality standards
Raw water
Water in tank before treatment
Fly ash
Peat
Sand
Fly ash/ Peat
Iron
Lignin
Aliphatic (C10C12), mg L1
2.5
0.19
