POP-contaminated sites from HCH production in Sabiñánigo, Spain

Environ Sci Pollut Res
DOI 10.1007/s11356-012-1433-8
11TH FORUM OF THE INTERNATIONAL HCH AND PESTICIDE ASSOCIATION
POP-contaminated sites from HCH production
in Sabiñánigo, Spain
J. Fernández & M. A. Arjol & C. Cacho
Received: 3 May 2012 / Accepted: 13 December 2012
# Springer-Verlag Berlin Heidelberg 2013
Abstract In 2009, hexachlorocyclohexane (HCH) isomers
(α-HCH, β-HCH, and γ-HCH [lindane]) were listed as
persistent organic pollutants (POP) in the Stockholm
Convention. Accordingly, the legacy of HCH/lindane production with the associated large HCH waste deposits has
become recognized as an issue of global concern and is
addressed in the implementation of the Convention. The
current paper gives an overview of the major contaminated
sites from lindane production of the INQUINOSA
Company. This company operated from 1975 to 1988 in
the city of Sabiñánigo, Spain. HCH production resulted in
the production of approximately 115,000 tonnes of waste
isomers which were mainly dumped in two unlined landfills.
These two landfill sites, together with the former production
site, are recognized sources of environmental pollution.
Assessment and remediation activities at these sites are
described. A dense nonaqueous phase liquid (DNAPL)
contaminated inter alia with HCH isomers, benzene, chlorobenzenes, and chlorophenols as the main contaminants and
an associated contaminated groundwater plume have been
discovered at both landfill/dumpsites and at the former
production site. The approximately 4,000 t of DNAPLs
constitute a serious risk for the environment due to the
proximity of the Gállego River. Since 2004, more than
20 tonnes of this DNAPL has been extracted using “pump
and treat” techniques. The Aragon Regional Government
and the Spanish Environmental Ministry are taking action,
focusing on the treatment of DNAPL and on the transfer of
the large quantities of solid POP wastes to a new landfill. A
range of laboratory tests has been performed in order to plan
the aquifer remediation.
Responsible editor: Leif Kronberg
Introduction
This article belongs to the series “Dioxin and POP Contaminated Sites”
(Weber et al. 2008) edited by Roland Weber, Mats Tysklind, and
Caroline Gaus.
Electronic supplementary material The online version of this article
(doi:10.1007/s11356-012-1433-8) contains supplementary material,
which is available to authorized users.
J. Fernández (*) : C. Cacho
Department of Agriculture, Livestock and Environment,
Government of Aragon, San Pedro Nolasco Square,
50071 Zaragoza, Spain
e-mail: [email protected]
C. Cacho
e-mail: [email protected]
M. A. Arjol
SARGA, Cesar Augusto Avenue 14,
50004 Zaragoza, Spain
e-mail: [email protected]
Keywords POP . Lindane . HCH . Organochlorines .
DNAPL . Contaminated site . Stockholm Convention
Technical hexachlorocyclohexane (HCH) and the separated
γ isomer (lindane) were among the most extensively used
organochlorine pesticides. Their production and use have
resulted in global contamination (Breivik et al. 1999; Li
1999; Vijgen et al. 2011). Recently, HCH isomers (αHCH, β-HCH, and γ-HCH [lindane]) were added to the
persistent organic pollutants (POP) list of the Stockholm
Convention (2009). Accordingly, the legacy of HCH and
lindane production has been recognized as an issue of global
contemporary relevance (Vijgen et al. 2011). The largest
challenges are posed by the HCH waste residues generated
from the inefficient production process of lindane: Each
tonne of lindane generated between 8 and 12 tonnes of other
HCH waste isomers (α-HCH, 55–80 %; β-HCH, 5–14 %;
δ-HCH, 2–16 %; and ε-HCH, 3–5 %) (Bodenstein 1972).
Environ Sci Pollut Res
Remaining largely unknown to the public and indeed to the
scientific community until relatively recently, this has generated the globe’s largest POP stockpile—estimated at between four and seven million tonnes of wastes (Vijgen 2006;
Vijgen et al. 2011). These wastes, together with those generated from recycling of HCH residues, were generally
dumped, usually in an uncontrolled manner, in the vicinity
of the production facilities. This practice has resulted in
numerous contaminated sites around the world (Vijgen et
al. 2011; Götz et al. 2012; Jit et al. 2010; Torres et al. 2012;
Weber and Varbelow 2012; Wycisk et al. 2012). These
wastes and the contaminated sites now need to be addressed
as part of the implementation of the Stockholm Convention.
The Stockholm Convention states in Article 6 the measures
to be used to reduce or eliminate releases from stockpiles
and wastes. They must be disposed of in such a way that the
POP content is destroyed or irreversibly transformed so they
do not exhibit the characteristics of POP or must be otherwise disposed of in an environmentally sound manner. The
Convention also mandates that signatory parties shall endeavor to develop appropriate strategies for identifying sites
contaminated by chemicals listed in Annex A, B, or C of the
Convention. If remediation of such sites is undertaken, then
it must be performed in an environmentally sound manner.
Lindane has been produced in Spain at four production
sites. All production has now ceased, with the last plant
operating until 1988. Two production facilities were located
in the Basque country of Northern Spain and these are
thought to have disposed of around 82,000 tonnes of production wastes. One factory produced lindane in Galicia,
generating several thousand tonnes of HCH waste together
with several hundred thousand tonnes of contaminated soil
in the deposition process. Another facility in Aragon generated approximately 115,000 tonnes of dumped HCH wastes
(Vijgen 2006).
In this paper, an overview of contamination from the
production in Sabiñánigo, Aragon is presented. Sabiñánigo
is a small industrial city located in the Aragonian Pyrenees,
in the northeast of Spain (Fig. 1), whose development in the
beginning and middle of the twentieth century was largely
due to the bulk chemical industries located there. During
this period, bad environmental practices associated with the
dumping of the solid wastes generated by the city and its
associated industries resulted in several contaminated sites.
By far, the largest producer of chemical wastes was the
INQUINOSA Company which was involved in the production of lindane. This company was established in
Sabiñánigo in the 1970s, produced lindane from 1975 to
1988, and continued formulating lindane products until
1992. Waste generation data differ somewhat, depending
upon the information source, but estimates of approximately
6,800 t/year of solid waste and 300 to 1,500 t/year of liquid
waste appear to be reliable. Both waste streams from the γ
isomer enrichment for lindane production. In a first refining
step, a solid α-HCH-rich waste is generated, and in the
second refining step, a liquid/pasty residue with high δHCH concentration (δ-paste) is produced (Jit et al. 2010;
Sievers and Friesel 1989). The HCH waste was initially sent
for some time to another lindane production plant in France
where it was recycled by thermal cracking. At other sites
employing this technique, residues highly contaminated
with polychlorinated dibenzodioxins/polychlorinated dibenzofurans (PCDD/PCDF) are known to have been generated
(Götz et al. 2012; Vijgen et al. 2011). Later, in the absence
of a market for the products of the cracking process (trichlorobenzene/tetrachlorobenzene), the HCH wastes were
packed in drums and dumped at different locations in the
vicinity of the INQUINOSA plant.
Over a period of 7 years, the pollution status of the areas
has been assessed and potential remediation technologies
evaluated. Similar studies have been reported in relation to
sites in Brazil (Torres et al. 2012; La Laina Cunha et al.
2010) and India (Abhilash and Singh 2009).
This article provides an overview on assessment work
conducted on the pollution from lindane production residues
at the main contaminated sites around Sabiñánigo, including
likely and potential impact on the local environment. Details
on the scale and extent of contamination are presented and
the major resulting threats and challenges are described. The
challenges posed by the large volumes of liquid POP waste
generated should be noted in this case. This constitutes a
potentially very high pollution loading with a high migration potential. It poses, therefore, a threat of both groundwater and surface water pollution. The presence of a large
volume of dense nonaqueous phase liquid (DNAPL; containing HCHs, chlorobenzenes, benzene, and other organochlorines) and related surface water and groundwater
contamination at the dumping sites and the old factory have
helped shape the decisions of the Regional Government.
Mitigation of the problems is taking place against a background of limited available resources coupled with the very
large amounts of POP waste which need to be managed.
Materials and methods
The assessment and initial remediation works described in
this article were developed over 7 years with the participation of several companies contracted by the Regional
Government and its technicians. Until 2010, groundwater
analyses were carried out by commercial laboratories. Due
to the contracting of several institutes, analytical methods
used have varied widely. Monitoring of the groundwater
plume has been carried out for a wide range of
(semi)volatile aromatic compounds (including benzene,
chlorobenzenes, phenol, and chlorophenols) and chlorinated
Environ Sci Pollut Res
Fig. 1 Location of the
Sabiñánigo area in Spain
pesticides. The analyses of organochlorine contaminants
(chlorobenzene, HCHs, chlorophenols, and nonchlorinated
compounds) were generally performed using low-resolution
gas chromatography/mass spectrometry (GC/MS) and involved several commercial laboratories (Aycon, Analytico,
Agrolab, and Alcontrol Laboratories). Alcontrol, Analytico,
and Agrolab are companies accredited by Raad voor
Accreditatie, according to ISO/IEC 17025:2005, for almost
all parameters analyzed. GC/MS was used to measure the
volatile and semivolatile compounds.
In Agrolab Laboratories, the aromatic compounds are
determined by the method EN ISO 11423-1, the volatile
compounds by method EN-ISO 10301, and the semivolatile compounds by the method NEN-EN-ISO 6468.
Alcontrol Laboratories uses their own methods,
Analytico Laboratories uses, for all the parameters, the
method W6336 GC-MS. Aycon uses the PE-Q-AG-075
(headspace [HS]-GC/MS) method.
The analytical differences between laboratories were
assessed using blind and duplicate samples and were in an
acceptable range. The complex sampling matrices involved
the heterogeneous character of the contamination source,
and the different analytical methods used by the laboratories
resulted in some discrepancies in results. The variation
between data for the main control parameters (benzene,
monochlorobenzene, and sum of isomers of HCH) ranges
from 20 to 30 %.
This would prevent any clear comparison between campaigns and/or laboratories data and blur the development
trend of the plume. However, with so high concentrations of
pollutants, it is considered that this information is applicable
to delineate the approximate distribution of the contaminants in the aquifer.
On the other hand, discrepancies between analytical campaigns for the same control points are caused by the degree
of dilution (high and low water, pumping-free phase) and
the high variability in DNAPL composition that determines
the load in solution. So, for several campaigns, each control
point remains in the same order of magnitude in the total
pollutant load, but its distribution varies.
In order to reduce analytical errors, facilitate the data
availability, and reduce the analytical cost, the government has established its own GC/MS facility (Agilent
7890A) and has carried out analyses under high standard of quality assurance/quality control procedures,
including interlaboratory comparisons (Aquacheck program). The volatiles compounds were analyzed by HSGC/MS, whereas for semivolatile compounds and phenols, a solid-phase extraction with subsequent organic
solvents dissolution and determination by GC/MS were
used. The uncertainty in intercomparison exercises was
<10 % for all parameters, with the exception of chlorophenols which reaches 70 % due to the absence of a
specific method of extraction.
Environ Sci Pollut Res
Currently, the hydrogeochemical monitoring program in
place generates approximately 160 groundwater samples per
year from the Bailín landfill and 120 groundwater samples
per year from the Sardas landfill. Over time, this has
resulted in more than 1,500 groundwater samples from the
Bailín landfill and more than 500 samples from the Sardas
landfill being taken and analyzed. Before each sampling
campaign, measurements of water and DNAPL level and
vertical profiles of temperature and conductivity are taken in
order to target specific sampling depths. Samples are taken
using bladder pumps and flow cells.
For surface water quality monitoring, a monthly campaign is performed involving sampling of seven stations
on the Gállego River and along the Bailín Ravin.
Sampling is carried out daily or weekly downstream of the
Sardas landfill and 500 m downstream of the Bailín Ravin
waste deposit, depending upon recent local meteorological
conditions. The analytes targeted include benzene, chlorobenzene, and HCH isomers, together with a suite of standard
hydrogeological monitoring parameters.
The surface water quality monitoring program also
includes an annual electrofishing exercise together with
sampling of macrobenthos and sediments at seven stations
in the Gállego River and its tributary.
Results and discussion
The 7-year assessment of pollution from the former production of lindane and associated waste dumpsites of the
INQUINOSA plant revealed that, during its 17 years of operation, the company had virtually no environmental policy in
operation and particular waste disposal. It is estimated that
approximately 115,000 tonnes of hazardous organochlorine
wastes (largely HCH waste isomers) were disposed in unsecured landfills (dumpsites). The investigation further revealed
that four major contaminated sites exist (Fig. 2): the former
factory area of INQUINOSA which is now derelict, the Sardas
and Bailin dumpsites, and the Sabiñánigo dam. In particular,
the two dumpsites were found to be the major legacies of
lindane production with aggregated waste quantities estimated
at between 30,000 and 80,000 tonnes of HCH for the respective dumpsites.1 These estimates place the sites among the
largest HCH waste deposits worldwide (Vijgen 2006; Vijgen
et al. 2011; Jit et al. 2010).
Former workers of the plant were interviewed as part of
the site investigations. From these interviews, it is thought
that part of the generated waste stream was disposed of at
1
It is based on the annual production of waste according to different
information sources. During some periods, HCH waste was disposed to
different landfills which results in difficulties in detailed determination
of waste volumes for each landfill. From recent documentation materials discovered in the factory, the upper estimate seems most probable.
more distant sites, but the locations remain to be established.
Additionally, therefore, several smaller dumpsites probably
exist. These have not been addressed by the current investigation which focused on only four major contaminated
sites outlined previously.
For three of these sites, the Regional Government of
Aragon and the Government of Spain are executing the
investigation/remediation works. Only the dam is solely
supervised by a national authority and the work underway
is not described in this study. In the following sections,
contamination at these sites and their current situation are
described. The assessment activities conducted in recent
years and current plans for further remediation are outlined.
INQUINOSA’s former production site
The production facility of INQUINOSA was closed in 1992
and is today abandoned and derelict. When it was closed,
the company removed manufacturing equipment, but left
raw materials and waste on the site (see Fig. 3), which have
remained on-site ever since.
Today, after access to the site was granted by judicial
means, soil and groundwater characterization and an inventory of the remaining wastes on the site are being undertaken. The conducted assessments have revealed significant
contamination of soil and groundwater (Table 1). Further,
the presence of a DNAPL was discovered approximately
30 m from the Sabiñánigo reservoir (Dirección General de
Calidad Ambiental, Gobierno de Aragón 2011a). The
DNAPL (Table 1) stems from the liquid waste derived from
process residues for the enrichment of the γ-isomer and
from the mixture of faulty production batches.
The main causes of the pollution on-site were the management of solid waste and the infiltration of liquid waste
(resulting in DNAPL). The HCH contamination in soil is
associated with solid waste dispersion, especially important
in the early years of production. In this period, solid HCH
waste was managed without control measures. Some of the
area in the factory was without concrete flooring. The high
contamination levels in the waters are clear indicators of the
presence of DNAPLs in the aquifer at the site.
About 100 tonnes of HCH, 6 tonnes of o,o-dimethyl phosphorodithioic acid (used in the manufacture of phosmet insecticide), and other chemicals still remain on the abandoned
production site. Currently, further characterization of the chemicals and wastes left on-site is being undertaken and further
management of this waste is planned. Other work planned in
the short term includes a monitoring program needed to characterize the distribution of the DNAPL in the aquifer at the site
and formulate plans for its extraction and treatment.
Any further demolition of the facility and remediation of
the soil are constrained by the legal situation which currently applies to the property and ultimately by the available
Environ Sci Pollut Res
Fig. 2 Geographic location of
the major contaminated sites
budget. The task of remediation requires very thorough
planning, taking into account the challenges encountered
at similar factories where such work has been carried out
(Weber and Varbelow 2012)
The Sardas landfill
The Sardas landfill is located along a geological ridge structure of folded rock layers. The northern flank or limb of this
Fig. 3 Abandoned raw
materials and wastes at the
former INQUINOSA
production site
anticline is declining with about 30° into the subsurface,
while the southern limb is orientated almost vertically.
Eocene marls form the geological basement of the Sardas
landfill (Fig. 4).
It is estimated that between 30,000 and 80,000 tonnes2 of
solid HCH waste isomers and 2,000 t in liquid form
(DNAPL) were dumped in this site. The total waste volume
of the site is approximately 350,000 m3. The site lacks a
liner system in its basin and does not have leachate
c
b
a
μg/kg
μg/kg
ε-HCH
Total HCH
2,700,000
74,730,000
3,200
640
57,000,000
5,600,000
9,700,000
2,200,000
138,885
3,200,165
452
161
2,303,197
245,523
406,837
105,722
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
1
22
730
712
1,158
2,269
640
5,509
7,600
1,400
15,000
37,633
139
16,428
2,535
701
93
6
15
34
–
3
77
2,000
1,400
2,800
120,000
310
42,000
8,200
2,007
245
16
43
123
–
Mean
13
31
94
<1
<1
21
20
9
1.800
13
370
<1
<1
<1
<0.1
<1
<1
–
Minimum
Chromatographically not resolvable constituents
Nivel Genérico de Referencia (NGR or generic reference level [GRL]) maximum concentration acceptable for land use in Spanish law
Units in micrograms per kilogram dry matter
–
–
1,000
100,000
1,000
1,000
1,000
–
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
–
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
2,546
341
65,167
10,561
38,162
12,504
955
308
648
–
Pentachlorophenol
Monochlorophenols
α-HCH
β-HCH
γ-HCH
δ-HCH
60,000
2,200
1,600,000
247,300
801,600
279,000
20,000
1,900
4,200
9.6
10,000
100,000
35,000
40,000
90,000
–
–
10,000
100,000
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
μg/kg
Benzene
Phenol
Monochlorobenzene
Dichlorobenzenes
Trichlorobenzenes
Tetrachlorobenzenes
Pentachlorobenzene
Dichlorophenols
Trichlorophenols
Tetrachlorophenols
Maximum
Unit
Mean
NGRb industrial
(soil limited
value)
Unita
Maximum
Water (16 samples)
Soil (28 samples)
Parameter
16.01
5.95
15.29
14.78
1.74
24.06
1.41
Unknownc
Density (g/cm3)
0.25
4.31
0.91
Tetrachlorocyclohexene
Pentachlorocyclohexenes
Hexachlorohexadiene
Heptachlorocyclohexane
α-HCH
β-HCH
γ-HCH
δ-HCH
3
4
7.92
6.47
3.66
0.24
Weight percent
Benzene
Monochlorobenzene
Dichlorobenzenes
Trichlorobenzenes
Tetrachlorobenzenes
Pentachlorobenzene
Parameter
DNAPL (1 sample)
Table 1 Concentration of HCH isomers, chlorobenzenes, benzene, chlorophenol, and other chlorinated and related nonchlorinated compounds in soil and groundwaters at INQUINOSA production
site
Environ Sci Pollut Res
Environ Sci Pollut Res
Fig. 4 Conceptual model of the Sardas landfill
treatment or other protection measures in place. The codisposal of waste from chloralkali production has resulted
in a high pH of 13 (Fig. 5). The wastes also contain, in
addition to HCH, large amounts of hydrocarbons and metals. Leachates and dispersion of waste have contaminated
soils, groundwater, and surface water.
In 2009, a DNAPL was detected from the surface
(Table 2; Fig. 5). Immediately, collection works and studies
of the hydrogeological behavior of the site were initiated.
This initial work was focused on identifying viable confinement and treatment options for later implementation.
A general high heterogeneity of the subsurface conditions
at the landfill site was revealed by the application of various
investigation techniques. This includes geological drilling
and borehole data, geophysical seismic and electrical tomography, hydraulic characterization by heat–pulse flowmeter measurements as well as pumping tests, and extensive
hydrochemical analyses.
Based on these investigations, the landfill site was characterized as follows: (1) the landfill is not secured by a
bottom liner system, (2) the base of the landfill is saturated
Fig. 5 Surface alkaline
leachates and DNAPL
and in contact with the groundwater, and (3) an extensive
DNAPL pool in the downstream area of the landfill was
identified (Fig. 4) (Dirección General de Calidad Ambiental,
Gobierno de Aragón 2010).
The conceptual model of the site suggests that groundwater flow mainly occurs in the Quaternary top layers as
well as in the upper 15 m of the altered marl horizon
(Dirección General de Calidad Ambiental, Gobierno de
Aragón 2011b). A general leachate accumulation is present
in the anthropogenic fillings and alluvial silt sediments in
the downstream area of the landfill. The presence of the
Sabiñánigo dam maintains the water table to a level very
constant, between 5 and 10 m below the surface. The
DNAPL was detected at very variable depths, even on the
surface, where the topography short marl outcrops, up to
40 m deep near the dam.
The DNAPL movement is density-driven and follows
mainly preferential flow pathways through fractured media.
The DNAPL contamination already affected the alluvial
deposits downstream that are hydraulically linked to the
Gállego River bed. Fortunately, the contamination of surface
Environ Sci Pollut Res
Table 2 DNAPL composition at the Sardas dumpsite (date: 16 February 2009
Parameter
Unit
Density
Benzene
Monochlorobenzene
1,2-Dichlorobenzene
1,3-Dichlorobenzene
1,4-Dichlorobenzene
1,2,3-Trichlorobenzene
1,2,4-Trichlorobenzene
1,3,5-Trichlorobenzene
1,2,3,5-Tetrachlorobenzenene
1,2,4,5-Tetrachlorobenzene
1,2,3,4-Tetrachlorobenzene
Pentachlorobenzene
kg/L
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
1.42
0.426
0.994
5.254
2.13
6.674
0.426
10.224
4.118
8.662
n.d.
3.976
n.d.
Phenol
Chlorophenol
2,4-Dichlorophenol
2,6-Dichlorophenol
Trichlorophenols
Tetrachlorophenols
Tetrachlorocyclohexenes
Pentachlorocyclohexenes
Hexachlorohexadiene
Hexachlorocyclohexane
Heptachlorocyclohexane
Heptachlorocycloheptane
α-HCH
β-HCH
γ-HCH
δ-HCH
ε-HCH
Other HCH isomers
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
n.d.
n.d.
n.d.
n.d.
0.142
n.d.
1.42
21.726
26.838
n.d.
n.d.
0.71
6.532
0.142
12.354
9.656
3.692
7.242
water is retained by the dam that keeps the superficial water
level to an upper bound to the alluvial; the contaminant
discharge from the gravels is at least somehow retarded
and avoids a direct and rapid contamination of the surface
water.
The monitoring network of the landfill has 51 piezometers; 7 of these are connected to a pumping station for
DNAPL extraction. Currently, the DNAPL is controlled by
these seven piezometers. The pumped residues, after decanting, are sent to an incinerator. The aqueous phase is stored in
two ponds for in situ treatment in a wastewater treatment
plant (physicochemical and activated carbon treatment).
In order to plan and decide on future decontamination
strategies, a range of preliminary laboratory tests have been
carried out (with persulfate, peroxide, and nanoparticles of
zerovalent iron). To date, the results of these tests have not
been conclusive and these studies will probably be
extended.
To assess the treatability of soil and wastes located at the
bottom of the landfill, a pilot thermal desorption test with
43 tons of these materials (with concentrations of 143 mg/kg
of total HCH) has been performed. The tests were conducted
in a thermal desorption plant designed by EMGRISA (a
public company involved in industrial waste management).
This achieved removal efficiencies of 99 % of organochlorines with emissions below the established regulatory limits.
The quantities of the waste volumes at this site
(350,000 m3 of mixed mainly hazardous wastes) and the
extremely complex and unpredictable handling caused by
the variety of wastes involved (organochlorine wastes,
chloralkali wastes with pH 13, wastes with mercury contamination, wastes impacted by other heavy metals, fluorides, oils and hydrocarbons, doughy or saturated wastes,
etc.) has currently ruled out an excavation of the landfill and
the transfer of contents to a new secure landfill site.
Furthermore, the incineration of waste or other alternative
approaches for waste removal are currently not being further
evaluated. Such comprehensive management approaches
involve high and unpredictable costs which in the current
Spanish economic situation are simply not available.
Current efforts, therefore, are focusing on increasing the
capacity of DNAPL extraction, increasing the number of
pumping points, and evaluating suitable surfactants for the
types of waste and specific site conditions that might allow
in situ remediation to some extent.
Within the framework of these initiatives, perimeter isolation measures (up to 15 m depth) for the landfill are
included. In addition, the management of polluted soil and
waste located outside the landfill and their transfer to the
new security cell built at the Bailin site (Fig. 8) has been
included. Activities necessary for the collection of the leachate draining from the bottom of the landfill need to be
further addressed, including an optimization of the overall
approach.
The thermal cracking plant of HCH—the process with
high PCDD/PCDF release in other facilities (Vijgen et al.
2011; Götz et al. 2012)—in INQUINOSA Company never
came into operation. A first analytical tests carried out to
investigate the presence of PCDD/PCDF in the DNAPL of
the Bailin landfill were below detection limits (Supporting
information), consistent with the absence of thermal cracking in INQUINOSA and with the use of this landfill exclusively for HCH residues. The conditions, however, under
which the Sardas landfill was used suggest a strong possibility of PCDD/PCDF being present: Apart from the heterogeneous chemical waste dumped there, it was common
practice also to dump residues from the incineration of
municipal solid waste (MSW) and residues from chlorine
Environ Sci Pollut Res
production—both processes where residues can contain
high levels of PCDD/PCDF (UNEP 2005; Weber et al.
2008). In addition, the presence of highly alkaline wastes
is known to facilitate the degradation of HCH isomers
which have resulted in PCDD/PCDF contamination at another HCH deposit (Braga et al. 2003; Weber 2007; Torres
et al. 2012). Further investigation of this potential pollution
will be carried out as part of the future work program.
The Bailín dumpsite
Between 1984 and 1992, the Bailin creek (see Figs. 2 and 6),
approximately 3 km south of Sabiñánigo, was used for the
disposal and dumping of industrial solid waste and MSW and
contains the other large HCH reservoir from the production of
HCH at INQUINOSA. The total volume of the landfill is
estimated at 180,000 m3, containing 30,000 to 80,000 tonnes3
of HCH solid waste and 2,000 t in liquid form (DNAPL)
consisting mainly of organochlorine chemicals (Table 3).
Hydrogeological setting
The dumpsite is located on top of an interbedded vertical
sequence of sandstone and fine-grained limolite layers. This
landfill/dump does not have any bottom liner system as a
Fig. 6 Landfill sites and geological frame at the Bailin creek
securing measure. In 1996, a surface liner, using highdensity polyethylene (HDPE) material, was installed.
The contamination plume is a dense mixture of benzene,
chlorobenzenes, chlorophenols, alcohols, and HCH isomers
(Table 3), comprising densities of 1.5 cps to 950 g/kg. This
DNAPL contamination moves downstream through fractured media towards the Gállego River about 800 m away.
Here, the contamination plume moves mainly along four
vertically orientated sandstone layers that are hydraulically
linked by horizontal fractures to adjacent limestone formations (see Fig. 6) (Dirección General de Calidad Ambiental,
Gobierno de Aragón 2009). The DNAPL movement highly
depends on the underlying rock units and related fracture
system which is characterized by corresponding fracture
widths and lengths. The plume has reached depths between
20 and 40 m and was observed in a maximum distance of
150 m downstream the landfill site.
The plume in the groundwater (Table 4; Fig. 7) has reached
the Gállego River and peaks of 1 μg/L of HCH have been
measured 0.5 km downstream in the river, with an average
concentration of 0.57 μg/L. This drops below detection limits
(0.1 μg/L) 2.5 km down the river due to flow increase from a
hydropower plant channel and consequent dilution.
Occasionally, however, a level around the detection limit of
0.1 μg/L has been recorded at this point downstream. The
Environ Sci Pollut Res
Table 3 DNAPL composition at the Bailin dumpsite (for four measurements; details in Supporting information Table 1)
Parameter
Unit
Maximum
Minimum
Viscosity at 25°
Density
Water
Benzene
Monochlorobenzene
1,2-Dichlorobenzene
1,3-Dichlorobenzene
1,4-Dichlorobenzene
1,2,3-Trichlorobenzene
1,2,4-Trichlorobenzene
1,3,5-Trichlorobenzene
1,2,3,5-Tetrachlorobenzenene
1,2,3,4-Tetrachlorobenzene
cps
kg/L
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
12.1
1.543
90
11.16
140
18.89
4.93
20.5
6.50
74.92
0.31
13.49
12.72
14.2
1.3662
2.2
9.4
75.53
7.4
2.2
17.03
2.96
29.57
<0.025
5.1
5.3
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
α-HCH
β-HCH
γ-HCH
δ-HCH
ε-HCH
Other HCH isomers
Phenol
Chlorophenol
2,4-Dichlorophenol
2,6-Dichlorophenol
2,4,5-Trichlorophenol
2,4,6-Trichlorophenol
Tetrachlorophenols
Tetrachlorocyclohexenes
Pentachlorocyclohexenes
Hexachlorohexadiene
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
g/kg
11.1
1.7
60.92
1.91
148.86
129.05
35.90
39.97
1.72
<0.025
0.045
<0.025
1.18
3.426
0.254
18.64
130.69
65.1
n.a.
<0.025
48.3
<0.025
134.00
102.2
21.43
n.a.
n.a.
n.a.
n.a.
n.a.
0.9
0.31
n.a.
1.97
74.3
<0.025
Hexachlorocyclohexane
Heptachlorocyclohexane
Alcohols, volatile fatty acids,
and other HCH metabolites
g/kg
g/kg
g/kg
167.00
96.38
148.85
n.a.
1.39
136.1
n.a. not analysed
average HCH limit in surface water specified by the Council
Directive for priority substances under the Water Framework
Directive (Directive 2008/105/CE) is 0.02 μg/L, with 0.04 μg/
L for temporary maximum peaks. To understand the consequences resulting from the need to comply with this legal limit
requires the establishment of reliable water monitoring with a
detection limit below 0.02 μg/L.
For the Bailin site, a yearly release to the Gállego River
of 20 kg HCH was estimated from the surface water concentrations. For the previously mentioned Sardas landfill,
the yearly release was estimated to approximately 120 kg/
year. The estimates are currently mainly based on surface
water measurements and do not consider other diffuse sources. The releases are currently not included in the European
Pollution Release Transfer Register, where for POP, levels
above 1 kg release are reported.
Already the water pollution levels of HCHs of more than
an order of magnitude higher than the regulatory limit reveal
that action is required for the reduction of releases and
inflow to surface water (in addition to the inherent need
for protection of the environment).
The treatment strategy at this site is also driven by the
presence and consequent threat of the migrating DNAPL.
Since the discovery of the DNAPL in 2006, the priority has
been to stop its flow and control the contamination levels in
the aqueous phase. As a result of this initial work, a flow
control protocol has been established. Efforts are now focused
on studying and understanding how the aquifers behave. First,
a conceptual model (Fig. 6) and then a 3D transient mathematical model assimilating a low permeability porous multilayer fractured media (based on Modflow code) was
developed, based on a recent study commissioned by the
Regional Government to URS Spain. The model includes an
area of approximately 90 ha, delimited by the new landfill on
the east, the Gállego River on the west, the mountains in the
north, and the Bailin stream in the south. The study focuses on
details in the sandstone layer which allow the highest percolation of leachate. Several pumping tests, tracers, head pulse
gauge, and temperature and conductivity logs, have allowed
the permeability of sections and even particular fractures to be
established. Subsequently, media permeability assigned by
discreet depths in the model has been used to simulate the
fractures at different levels of permeability. The model, which
fits well with experimental data, was used, in turn, to simulate
the hydrogeological equilibrium and to design first pilot tests
of potential surfactant treatments.
The control of pumping from the aquifer is managed
using a network of 150 piezometers and a pumping and
wastewater treatment system. The DNAPL is extracted from
the aquifer by a programmed pumping system that effectively prevents the flow of the plume. Twenty thousand
liters of DNAPL and an average of 15,000 m3/year of
leachate have been extracted since 2006.
The DNAPL pumped from the piezometers, the sludges, and
the loaded activated carbon from the leachate treatment plant is
transferred by an authorized waste management company to a
hazardous waste incineration company located in France. The
leachates, once they have been treated, are stored in a pond and
the disposal parameters established by the authorities are
checked before the leachates are discharged to the environment.
Once the movement of the contamination plume was
brought under control, two additional tasks were addressed:
the removal of the pollution source and the study and test of
potential aquifer remediation techniques.
Environ Sci Pollut Res
by, in the same valley, was selected. Due to local concerns
and security considerations, other sites farther away, but
with better hydrogeological conditions, were rejected.
The alternative to landfilling, the destruction of the solid
wastes, was also rejected for several reasons:
In anticipation of an excavation of the landfill and in order
to decide what would be the most appropriate remediation
techniques in this case, a series of laboratory tests on destruction and remediation methods (see Supporting information 3)
have been carried out. The detailed description of these tests
and the results will be the subject of future publications.
–
The excavation of the landfill
–
In 2007, a project plan for excavation of the wastes and
transfer of the Bailin dumping site was decided. A site close
The limited numbers of destruction techniques and the
duration of several years to complete the treatment
including the necessary transfer and storage facilities.
The need to remove the source of pollution in the short term
as a first step in order to proceed with the aquifer remediation.
Table 4 Composition of the leachates from Bailin site along one of the sandstone layers
Piezometer numbers
Date: 22–28 May 2012
UNID
139
81
26
26
129
143
104
142
Distance to landfill
Distance to Gállego River
Sampling depth
Chloromethane
1,1-Dichloroetane
m
m
m
μg/L
μg/L
41
837
25
<2.5
<2.5
153
725
35
<2.5
<2.5
220
658
20
<2.5
<2.5
220
658
50
<2.5
<2.5
315
563
50
<2.5
<2.5
453
425
40
<2.5
<2.5
630
248
30
<0.1
<0.1
830
48
45
0.7
<2.5
Trichloroethylene
Tetrachloroethylene
Benzene
Toluene
Ethylbenzene
m-Xylene and p-xylene
o-Xylene
Monochlorobenzene
1,3-Dichlorobenzene
1,4-Dichlorobenzene
1,2-Dichlorobenzene
1,3,5-Trichlorobenzene
1,2,4-Trichlorobenzene
1,2,3-Trichlorobenzene
1,2,3,5-Tetrachlorobenzenene
1,2,4,5-Tetrachlorobenzene
Pentachlorobenzene
Phenol
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
9.2
0.9
33,860.6
24.4
2.1
0.8
2.6
28,188.4
101.0
1,057.4
1,216.8
4.0
731.2
69.1
29.3
58.3
7.8
252.3
8.9
<2.5
24,657.3
20.0
2.6
0.6
2.7
24,595.3
96.4
999.1
912.5
3.9
549.8
50.3
19.3
52.6
7.5
247.0
<2.5
<2.5
7,016.6
21.6
1.1
0.2
2.0
14,885.3
35.4
440.3
371.1
0.9
145.9
15.5
2.7
4.7
0.5
51.5
<2.5
<2.5
26,634.0
26.9
2.8
1.9
0.3
32,730.0
75.3
1,016.0
862.2
3.0
343.8
33.7
6.3
15.9
1.3
207.8
21.7
1.2
54,210.5
79.0
9.4
7.9
11.9
69,378.8
188.0
2,585.7
2,068.1
10.1
1,095.6
101.0
40.7
40.8
2.1
223.2
15.7
3.5
37,708.4
17.6
2.4
2.2
<2.5
29,617.2
142.1
2,301.0
1,632.6
8.8
1,102.4
79.2
32.3
32.7
2.1
256.2
<0.1
<0.1
5.3
1.2
<0.1
<0.1
<0.1
12.5
<0.1
3.3
2.4
<0.1
1.8
0.2
0.8
0.3
0.2
1.8
15.5
4.1
35,946.3
23.5
3.3
2.7
0.7
32,088.9
172.6
3,895.0
1,876.1
13.9
1,316.0
83.8
39.8
20.6
1.3
76.1
2- and 4-chlorophenol
2,4-Diclorofenol
3-Chlorophenol
2,6-Dichlorophenol
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
μg/L
271.0
327.0
2,410.0
65.2
9,166.3
271.1
1,740.1
219.1
7,051.5
259.1
289.0
3,595.7
62.8
12,680.9
778.4
1,618.1
735.6
6,387.0
36.7
123.9
567.8
7.7
760.3
95.7
169.1
30.8
284.9
242.5
205.1
4,704.2
47.1
7,040.2
227.5
560.3
276.1
614.2
252.4
241.9
4,535.4
48.1
8,154.0
163.9
1,059.3
238.5
3,283.6
148.8
123.6
1,361.5
24.4
3,715.4
192.6
706.0
188.9
2,350.5
<0.1
<0.1
<0.1
<0.1
<0.1
<0.2
15.4
0.7
97.3
71.2
<0.1
929.8
11.3
1,745.3
87.6
315.8
61.3
460.5
μg/L
μg/L
μg/L
16,701.8
967.4
26,679.9
17,886.4
1,030.7
27,657.8
1,890.7
155.1
2,530.6
9,892.5
766.6
12,109.8
13,085.2
714.3
18,380.8
8,588.9
517.2
12,351.6
75.5
7.3
196.2
4,939.9
368.0
6,145.6
Tetrachlorophenols
α-HCH
β-HCH
γ-HCH
δ-HCH
ε-HCH
Total HCH
Environ Sci Pollut Res
Fig. 7 Distribution of benzene and HCH plume in May 2011 (cross-section in Fig. 6)
–
The high quantities involved require proper pretreatment and best available techniques destruction facilities,
with associated high cost.
The activities started in 2010 with the construction of
infrastructure to facilitate the excavation (Fig. 8). Due to the
inadequate geological characteristics of the selected new site,
the new cell has been constructed with additional isolating
measures going beyond the Spanish legal requirements. The
landfill basin has a set of 5-m-deep drainage trenches, excavated in the rock, to depress the water table. The collected
leachates will be taken to a tank and the quality monitored
Fig. 8 Infrastructures
established to excavate and
transfer the Bailin dumping site.
a The transfer station for the
wastes. b Chutes and dust
suppression system. c
Pretreatment of the DNAPL. d
New landfill for the HCH solid
wastes
prior to its release to the environment. The new cell capacity is
250,000 m3, enough to store the current Bailin landfill, the
residues remaining in INQUINOSA and the soils affected
outside the Sardas landfill.
As basement on top of the rock, a compacted clay layer, a
geo-drain, and a gravel layer connected to a drainage system
that divides the base and takes the flows to a control tank was
built to isolate the landfill from the environment and to avoid
future leakage. Two sets of geomembrane liners of HDPE
(1.5 mm), bentonite (5,500 g/m2), and HDPE (1.5 mm) impermeabilize the base. The sets are separated by a geo-drain in
order to control possible leakages. A final gravel layer with
Environ Sci Pollut Res
drainage tubes collects the leachate. This is then taken to a
tank connected to the wastewater treatment plant.
The final cap at the site will consist of a gas collection
system, a set of geomembrane liners of HDPE (1.5 mm),
bentonite (5,500 g/m2), and HDPE (1.5 mm), a geotextile, a
geo-drain, and soil. The transfer station is designed for the
pretreatment of the solid wastes, depending on the size, the
moisture content, and the HCH concentration (Fig. 8). It has
four chutes, a packing system, a solidification–stabilization
unit for high moisture content wastes, a centrifuge unit, a
separator unit, and a loading bay. A dust suppression system
is installed to minimize the dust release while operational.
The excavation is planned for May–October 2013 or
2014, coinciding with the low precipitation period. The
operating area will be covered with an HDPE sheet with
an installed dust suppression system. A grid of 30×30 m
drilling test, sampling 2 m deep (the excavation is planned
in 2-m layers), will serve to characterize the solid wastes and
to establish the pretreatment regime at the transfer station.
To minimize the risks associated with the transfer of the
solid waste caused by weather, a meteorological system is
available. The operation will be suspended for those days
with rain and with medium to high wind conditions.
The transfer to the new constructed landfill is not a
sustainable long-term solution (Weber et al. 2011), but it
will allow actions to remediate the contaminated aquifer and
provide a much more secure storage infrastructure until such
times as the development of adequate remediation techniques can be devised. These will need to be both costeffective and capable of carrying out the task over a reasonable time frame.
Conclusions
The manufacture of lindane by the INQUINOSA Company
in Sabiñánigo, Spain has resulted in contaminated megasites
in the former factory area and the surrounding area. HCH
production resulted in approximately 160,000 tonnes of
waste isomers which were mainly dumped at two unlined
landfills/dumps which pose today the highest threat to the
environment and a significant challenge for remediation.
With this amount of disposed HCH residues, the site is
one of the largest HCH deposit. With the location close to
surface waters, the case is additionally of specific urgency.
DNAPLs with HCH isomers, benzene, chlorobenzenes,
and chlorophenols as main contaminants and an associated
groundwater plume have been discovered at both landfill/
dumpsites and the former production site. The approximately 4,000 t of DNAPLs constitute a serious threat to the
environment due to the potential impact to the Gállego
River. The presence of DNAPL and the large volumes of
solid waste (at about 500,000 m3 of solid wastes polluted by
a range of contaminants from HCH/other pesticide production, chloralkali, and others) have triggered the remediation
strategy of the Government of Aragon. As a priority,
DNAPL management is currently being addressed. For the
solid POP wastes, currently, only transfer from the unsecured Bailin site to a new secured landfill is planned. Due to
the large quantities involved, the resources, time, and appropriate technology for the destruction of wastes is not
currently available.
For remediation of the aquifer, a range of laboratory tests
have been performed, but the technologies assessed
(Supporting information 3) need further testing before they
could be applied. The case also reveals that remediation of
contaminated sites in difficult economic situation is a particular challenge. The costs of securing measures are already
high: Since 2005, the average investment for the Bailin
landfill is 750,000€/year for the control of the aquifer and
750,000€/year for the management of the leachates and the
analytical tests. The resources required to excavate the
Bailin dumpsite and to transfer the waste to the secured site
are estimated at approximately 19,000,000€ over a period of
3 years. For the landfill of Sardas, the average investment
has been 500,000€ yearly since 2009. For the industrial
plant of INQUINOSA, cost has been kept below 100,000
€/year in the last 3 years. These costs cover only the
expenses of managing and controlling the pollution. They
are still considered “low” in comparison to the cost of the
final destruction of the POP wastes, the cost of aquifers
remediation, and the cost of the investigation and treatment
of soils affected around the main contaminated sites. The
final elimination of these wastes, as mandated by the
Stockholm Convention, will probably have to wait until an
affordable treatment is available or until other priorities in
the frame of sustainable development are set.
References
Abhilash PC, Singh N (2009) Seasonal variation of HCH isomers in
open soil and plant–rhizospheric soil system of a contaminated
environment. Environ Sci Pollut Res Int 16:727–740
Bodenstein G (1972) Disposal of wastes from Lindane manufacture.
In: Uhlmann E, Verlag K (eds) Lindane monograph of an insecticide. Schillinger, Freiburg im Breisgau, pp 23–77
Braga A, Krauss T, Reidos Santos C, Mesquita de Souza P (2003)
PCDD/F-contamination in a hexachlorocyclohexane waste site in
Rio de Janeiro, Brazil. Chemosphere 46:1329–1333
Breivik K, Pacyna JM, Munch J (1999) Use of α-, β-, γhexachlorocyclohexane in Europe, 1970–1996. Sci Total
Environ 239:151–163
Dirección General de Calidad Ambiental, Gobierno de Aragón (2009)
Monitoring and control service of HCH Bailin landfill (tm
Sabiñánigo, Huesca) in 2009. Monitoring hydrogeological
Environ Sci Pollut Res
network of aquifers affected and decontamination work.
Unpublished (in Spanish) (available upon request)
Dirección General de Calidad Ambiental, Gobierno de Aragón (2010)
Research survey and detail geotechnical and hydrogeological study
for the mitigation of pollution in the affected area of the Sardas
landfill, Sabiñánigo (Huesca) in 2009. Alternative approach and
additional work for a definitive solution. Unpublished (in Spanish)
(available upon request)
Dirección General de Calidad Ambiental, Gobierno de Aragón (2011a)
Soil characterization work in the INQUINOSA former industrial
plant, Sabiñánigo (Huesca). Unpublished (in Spanish) (available
upon request)
Dirección General de Calidad Ambiental, Gobierno de Aragón (2011b)
Study of technical alternatives to solve the environmental problem
of Sardas landfill T.M. of Sabiñánigo (Huesca). Unpublished (in
Spanish) (available upon request)
Götz R, Sokollek V, Weber R (2012) The dioxin/POPs legacy of
pesticide production in Hamburg: part 2—waste deposits and
remediation of Georgswerder landfill. Env Sci Pollut Res.
doi:10.1007/s11356-012-0986-x
Jit S, Dadhwal M, Kumari H, Jindal S, Kaur J, Lata P, Niharika N, Lal
D, Garg N, Gupta SK, Sharma P, Bala K, Singh A, Vijgen J,
Weber R, Lal R (2010) Evaluation of hexachlorocyclohexane
contamination from the last lindane production plant operating
in India. Env Sci Pollut Res 18(4):586–597
La Laina Cunha AC, Ruiz MS, Echevenguá Teixeira C (2010)
Environmental assessment of remediation technologies: an analytical framework for a hexachlorocyclohexane (HCH) contaminated site in Brazil. In: Proceedings of the 2nd International
Conference on “Hazardous and Industrial Waste Management”,
Chania, Technical University of Crete, Greece, 5–8 October
Li Y-F (1999) Global technical hexachlorocyclohexane usage and its
contamination consequences in the environment: from 1948 to
1997. Sci Total Environ 232:121–158
Sievers S, Friesel P (1989) Soil contamination patterns of chlorinated
organic compounds: looking for the source. Chemosphere
19:691–698
Stockholm Convention (2009) Report of the conference of the parties
of the Stockholm Convention on persistent organic pollutants on
the work of its fourth meeting. UNEP/POPS/COP.4/38, 8 May
Torres JPM, Fróes-Asmus CIR, Weber R, Vijgen JMH (2012) Status of
HCH contamination from former pesticide production and formulation in Brazil—a task for Stockholm Convention implementation. Environ Sci Pollut Res. doi:10.1007/s11356-012-1089-4
UNEP (2005) Standardized toolkit for identification and quantification
of dioxin and furan releases, 2nd edn. UNEP Chemicals, Geneva,
February
Vijgen J (2006) The legacy of lindane HCH isomer production. Main
Report, IHPA, Holte, January
Vijgen J, Abhilash PC, Li Y-F, Lal R, Forter M, Torres J, Singh N,
Yunus M, Tian C, Schäffer A, Weber R (2011) HCH as new
Stockholm Convention POPs—a global perspective on the management of lindane and its waste isomers. Env Sci Pollut Res
18:152–162
Weber R (2007) Relevance of PCDD/PCDF formation for the evaluation of POPs destruction technologies—review on current status
and assessment gaps. Chemosphere 67:109–117
Weber R, Gaus C, Tysklind M, Johnston P, Forter M, Hollert H,
Heinisch H, Holoubek I, Lloyd-Smith M, Masunaga S,
Moccarelli P, Santillo D, Seike N, Symons R, Torres JPM, Verta
M, Varbelow G, Vijgen J, Watson A, Costner P, Woelz J, Wycisk
P, Zennegg M (2008) Dioxin- and POP-contaminated sites—
contemporary and future relevance and challenges. Env Sci
Pollut Res 15:363–393
Weber R, Watson A, Forter M, Oliaei F (2011) Persistent organic
pollutants and landfills—a review of past experiences and future
challenges. Waste Manag Res 29(1):107–121
Weber R, Varbelow G (2012) Dioxin/POPs legacy of pesticide production in Hamburg: part 1—securing of the production area. Env Sci
Pollut Res. doi:10.1007/s11356-012-1011-0
Wycisk P, Stollberg R, Neumann C, Gossel W, Weiss H, Weber R (2012)
Integrated methodology for assessing the HCH groundwater pollution at the multi-source contaminated mega-site Bitterfeld/Wolfen.
Env Sci Pollut Res. doi:10.1007/s11356-012-0963-4