Untersuchungsergebnisse zur Mobilität und Remobilisierung von Kupfer und Antimon in wasserwirtschaftlich relevanten, porösen Lockergesteinen durch Säulenversuche und mit reaktiver Transportmodellierung

The aim of this study is to obtain a better understanding of the
hydrological and geochemical contexts of the heavy metal transport
in watersaturated porous aquifer sediments, where both copper (Cu)
and antimony (Sb) are the main focus. The background of
investigation is based on questions within the scope of the
redevelopement of past industrial waste deposits areas and
groundwater protection. The functional and experimental part of
this study was initially comprised of planning, conception and
set-up of a column arrangement which served as a model of the
sediment-/groundwater system as well as the development of
appropriate preparation and sampling techniques. The experimental
set-up run was suitable for simulation of groundwater flow with
velocities between centimetre and metre per day. In order to
differentiate results concerning the migration behaviour of the
heavy metals copper and antimony, laboratory experiments were
conduced in three water aquifer systems Quaternary Gravel, Tertiary
Sand and Dogger-Sand of southern Germany. Between the systems
carbonate, clay, or iron contents vary. Quartz-Sand was served as
the reference material. Copper was injected as the cationic form of
copper nitrate (Cu(NO3)2) and antimony as anionic potassium
antimonat (K[Sb(OH)6]). The investigations included short-term
(Dirac-impulse) and long-term (Pulse-injection) experiments. A
calcareous water served as an eluent for sediments with high buffer
capacity and a rainwater-mimic modelwater as an eluent for the
sediment with lower buffer capacity. Determination of distribution
coefficients (batch-experiments) and thermodynamic modelling of
solubility in different experimental waters led to maximum values
in the calcareous water for copper as well as antimony, where the
solubility of copper was 30 % lower than for antimony. Due to the
additional complexes found in the DOC-water of the Dachauer Moos
copper has a significantly higher mobility. The effect is even more
significant by applying EDTA-water. Mass balance of the column
experiments show that copper is fundamentally less mobile (recovery
0-18 %) in contrast to antimony (recovery 85-99 %). Differences in
the sorption behaviour were reflected in the retardation values,
which subsequently distinguished almost three orders for both
elements. Cumulatively, these results show that the sorption
capacities for copper can not be achieved even after a total input
of 201 mg Cu (Quaternary Gravel) by injection pulses over two
years. However, sorption capacities can be attained with antimony
after only 10.4 days (Tertiary Sand), after 11.4 days (Dogger-Sand)
and after 1.5 years (Quaternary Gravel), respectively.
Qualitatively, the results of copper can be described with a fast
irreversible sorption kinetic, and antimony with a slow reversible
one. Three mathematical models were applied to simulate the
experimental data set. The observed Sb-breakthrough curves modelled
quite well with the one-dimensional linear reaction model of
CAMERON & KLUTE (1977) and the determination of corresponding
dynamic reaction parameters. However, a satisfying fit could not be
found for copper with this model in a buffered system for the two
following reasons: 1.) The precipitation processes and 2.) the
complex surface-active processes which were assumed next to the
predominant sorption reactions. Therefore non-linear model
approaches were employed for the fit. Initially, the observed
Sb-curves were fitted using two-site Langmuir-isotherm first order
kinetics, allowing for quantification of sorption capacities,
-affinities and –rates. Applying three-site Langmuir-isotherm first
order kinetics also includes precipitation processes. With all
models, a satisfying fit of the desorption processes in the
declining curve part emerged. However the sorption processes could
still not be satisfactorily described. Experiments investigating
the influence of both complexing agents, DOC and EDTA, on the
migration behaviour of copper showed more or less a stoichiometric
complexation of the heavy metal copper. With the stronger
complexing agent EDTA at least twice as high remobilisation as with
DOC could be achieved. The strongest EDTA-mediated remobilisation
could be observed in the unbuffered sediments (49-50%),
significantly lower values were attained in the buffered sediments
with 17-26 %. Of the injected copper mass in unbuffered sediment 82
% could be recovered from Dogger-Sand, and 83 % from Quartz-Sand. A
maximum of 55 % could be recovered from the buffered sediments. In
order to support mobility data of copper and antimony sequential
extractions of column beds were performed after the column
experiments. As an essential result, it emerged that copper is
accumulated increasingly in areas near the surface, while antimony
is distributed in the whole profile. Therefore, copper in contrast
to antimony is stable bound at neutral pH and is not transferred.
As a consequence, a different behaviour of availability of both
elements can be observed. The results of the column experiments
indicate that copper is preferentially occluded at organic matter
binding fractions as well as in the immobile aqua regia fraction.
Antimony is mainly bound in the mobile and easy exchangeable
fraction and in both the iron and manganese fraction. As this
result allows immediate conclusions on the availability and
transfer potential of heavy metals, it also permits estimations of
the ecotoxicological efficacy of copper and antimony and the
assessment of the endangering risk potential of the contaminated
sediments.