Remediation of As contaminated groundwater using constructed wetlands

1
Remediation of As contaminated groundwater using
constructed wetlands
C. Jackson1, R. Gordon2, I. Koch1, K.J. Reimer1,
and A. Mattes3 1. Royal Military College of
Canada, Kingston, ON, Canada 2. Simon Fraser
University, Burnaby, BC, Canada and Beamline
Scientist, PNC/XOR, APS Argonne, IL, USA. 3.
University of Guelph, Guelph, ON, Candada.
Introduction
Methods Bench scale bioreactors
Results Microbial mechanisms

• Bioreactors containing a matrix of celgar only
  did not have a significantly different rate of
  removal from the bioreactors containing a matrix
  of sand and celgar.
• The sand in the bioreactors contains
  oxy-hydroxides which are known sorption agents
  for arsenic. Celgar in the bioreactors acts as a
  substrate for microbial activity.
• The result of this experiment indicates that the
  arsenic sorption to sand is not significant
  relative to the microbial mechanisms of arsenic
  removal. It is likely that bacteria in the
  bioreactors reduce both arsenate and sulfate in
  the feed to form insoluble arsenic-sulfides which
  are filtered out in the reactor matrix.

The purpose of this experiment was to determine
if the Trail bioreactor could operate on a bench
scale and to examine the effects of zinc
addition, which is thought to enhance As removal.
The Trail bioreactor was scaled down to 1 L bench
scale reactors, keeping the 3 day retention time
constant. 14 reactors each contained an
innoculum from the Trail bioreactor in a matrix
of celgar or a mixture of the sand and celgar.
The inlet water for all reactors contained 10 ppm
As but for some reactors, zinc was added to the
inlet water. See figure 6 for the experimental
design. Water samples were analyzed for
dissolved As from each reactor every four days to
examine the effects of sand in the reactor matrix
as well as zinc addition to the inlet water on As
removal.
Smelting has been an important activity in Trail,
British Colombia, Canada over the past century.
Historic mine tailings continue to cause the
groundwater to be contaminated with arsenic (As)
up to 600 times the safe discharge limit. As is
a common metalloid and exists in the earths
crust as stable minerals (e.g.arsenopyrite,
scorodite and orpiment). In Trail, BC, this
arsenic has been mobilized as a result of mining
operations. The arsenic species in the
groundwater is a known toxicant and carcinogen.
Methods Bench scale sand reactors
Figure 14. As results from the bench scale
bioreactors (As inlet concentration 10000 ppb)
A set of 8 bench scale reactors containing a
matrix of construction grade sand was designed to
examine non-biological As removal mechanisms
which may be important to the Trail system.
Water containing 10 ppm As was passed through
each of the reactors with a retention time of
approximately 3 days (100 mL/day).
Results Non-microbial mechanisms
Figure 6 Schematic of the bench scale bioreactor
experiment

• Bench scale sand reactors (containing
  Fe(oxy)hydroxides no celgar) removed 90 of the
  As. The control reactors containing Ottawa sand
  (Si only no adsorption agents) had low As
  removal efficiencies.
• Non-microbial adsorption mechanisms of arsenic
  removal may be important to the overall
  performance of the Trail bioreactor.
• Both bioreactors and sand reactors with Zn added
  to the inlet water removed more As than the
  bioreactors that were treated with As alone. The
  addition of copper in the inlet water for the
  sand reactors also enhances arsenic removal.
  Both zinc and copper are present in the inlet
  water to the Trail bioreactor and are expected to
  play an important role in the performance of the
  system.
• It is likely that some of the As and zinc forms a
  precipitate, Zn3(AsO4)2, which is then filtered
  out in the reactors.

http//www.blaylock.ca/newPhotos/cominco.jpg
Figure 2 Smelter in Trail, BC in the early
1900s.
Figure 1 Smelter in Trail, BC in 2006
Zinc and copper were added to the inlet water for
some reactors to test the hypothesis that the
addition of cations can enhance As removal by
adsorption. Two control reactors were constructed
using a matrix of Ottawa sand, which is a
silicate material that does not contain iron
oxides or other adsorbent materials. Water
samples were analyzed for dissolved As from each
reactor every four days. The schematic of this
experiment is shown in figure 7.
Figure 8 Bench scale bioreactors.
For the past ten years, Nature Works Remediation
Corporation has been working to remediate the
groundwater using a constructed wetland. The
contaminated groundwater is passed through two
underground anaerobic bioreactors prior to being
treated by a conventional wetland treatment
system. Geochemical and microbial processes
affect the arsenic solubility and toxicity. The
anaerobic bioreactors consist of a mixture of
sand and wood fibre waste from a pulp and paper
mill, commonly referred to as celgar. This matrix
acts as a substrate for both adsorption and
microbial processes that effectively remediate
water contaminated with up to 600 ppm down to the
safe discharge limit of 1 ppm. While the system
works effectively, the exact As removal
mechanisms are not well understood, making it
difficult to install this type of system in other
areas with As contaminated groundwater.
Dissolved As can adsorb to surfaces inside the
reactor containing iron oxy-hydroxides.
Microbial activity in the reactor is likely to
reduce both As and sulfate, creating conditions
suitable for the precipitation of As sulfide
particles that can be filtered out in the reactor
matrix. The characterization of the removed As
product as well as bench scale studies have
provided a better understanding of the As removal
mechanisms in the Trail bioreactor.
Figure 15 As results from the Bench scale sand
reactors
Figure 7 Schematic of the bench scale sand
experiment
Conclusion
Results XAS
Figure 9 Bench scale sand reactors.
The Trail constructed wetland system has been
working effectively for the past ten years to
remove As from ground water but little is known
about the mechanisms of removal. The results
from bench scale sand reactor experiments suggest
that coprecipitation with cations such as Zn and
Cu may be a significant mechanism of removal.
The bench scale bioreactor results and the XAS
analysis of the solid field samples indicate
reduction of arsenate and coordination with
sulphur may be causing a precipitation reaction.
It is likely that several adsorption and
microbial mechanisms are working together to
immobilize dissolved As. More work is required in
the areas of identifying the As removal product
to better understand the dominant removal
mechanism.

• The XANES for the bioreactor samples fit well to
  the As(III) sulphur and orpiment standards.
• Inlet As is As(V), but no As(V) is observed in
  the sample. This implies that As is being
  reduced inside the reactor.

Methods analysis
Methods field sampling
Water samples were diluted in 2 nitric acid and
analyzed using inductively coupled plasma mass
spectrometry (ICP-MS) at the Royal Military
College of Canada
a)
Since the anaerobic bioreactors are very large
and underground, solid samples cannot be taken
directly from the reactors. Packets of celgar and
sand enclosed in mesh bags were placed inside the
sampling wells and left there for 1 year,
exposing them to the conditions of the reactor.
These samples are referred to as teabag samples
and are representative of the solid matrix inside
the reactor.
Solid samples were analyzed for As speciation
using X-ray absorption near edge spectroscopy
(XANES). This was done at the PNC/XOR bending
magnet (BM) line, sector 20 of the Advanced
Photon Source (APS), Argonne National Laboratory
(ANL). Samples were prepared in a glove bag
under an Ar atmosphere and were kept at -50?C
during analysis.
Acknowledgements
Sampling well
b)
Figure 3 (left) Schematic of the Trail
constructed wetland

• The authors gratefully acknowledged the support
  of the Advanced Photon Source (APS) Pacific
  Northwest Consortium/X-ray Operations and
  Research (PNC/XOR), the Department of National
  Defence Canada Academic Research Program, as well
  as the Natural Sciences and Engineering Research
  Council (NSERC) Metals in The Human Environment
  (MITHE) Research Network.
• PNC/XOR facilities at APS, and research at these
  facilities, are supported by the US Department of
  Energy - Basic Energy Sciences, a major
  facilities access grant from NSERC,  the
  University of Washington, Simon Fraser University
  and APS. Use of the APS resources is also
  supported by the U. S. Department of Energy,
  Office of Science, Office of Basic Energy
  Sciences, under Contract DE-AC02-06CH11357)
• A full list of sponsors is available at
  www.mithe-rn.org

• A gold foil was run at the same time as the
  samples for use as a reference for energy
  calibration
• E0 11919.7 eV gold
• E0 11868 eV As

Figure 16 XANES for one of the teabag samples
Teabag sample
Figure 10 Sampling well at the Trail bioreactor
The XANES only give information about the
oxidation state of As and the nearest neighbour.
Although the data fits well to orpiment, a stable
mineral in the earths crust and is unlikely to
dissolve into ground water it is likely that the
As in the reactor is actually an amorphous
As(III)-S compound.
c)
The collected XANES were compared to a library of
standards and fitted using ATHENA software.
Academic Research Program
Figure 13 The cryogenic stage (a) the cryogenic
stage sample holder (b) and sample preparation
inside a glove bag
Figure 4 (right) Construction of the Trail
bioreactors
Figure 12 Schematic of the teabag method of
sample collection
Figure 11 Mesh bag used to construct the teabag
samples