Presentaci

1
11th Mediterranean Congress of Chemical
Engineering
(EXPOQUIMIA) Barcelona, Spain October 21-24, 2008
EXPERIMENTAL STUDY AND MODELLING OF LINEAR
ALKYLBENZENE SULPHONATE IN SAND AND SOIL
Boluda-Botella N., Cases V., Gomis V., León V.M.,
and Soriano R. Chemical Engineering Department,
University of Alicante. Apdo. 99, E-03080
Alicante (Spain). Tel.[phone], ext.2647.
Fax[phone]. E-mail nuria.boluda_at_ua.es
INTRODUCTION
REACTION PARAMETERS
The Linear distribution coefficient Kd is
considered to describe adsorption Cads Kd C,
where Cads is the sorbed concentration of a
solute (moles/kg of solid) and C is the
concentration in solution (moles/L of solution)
Over the last two decades, many studies have
been performed to characterize the environmental
behaviour of linear alkylbenzene sulphonate
(LAS), one of the major ingredients of synthetic
detergents. In fact, the fate, effects, behaviour
and sorption of LAS in different soils have
established a good foundation for understanding
its interactions 1-2. However, few reports
analyse how desorption processes occur. In recent
years, high loads of treated wastewater or
sludge, which can contain high concentrations of
LAS, have been applied to agricultural areas, and
therefore migration of these contaminants could
affect groundwater quality.
Adsorption affects the convection term
Convection-dispersion equation
Retardation Factor
Distribution Coefficient
Kd
Obtained graphically with experimental results
Determined in reactive transport experiment
Where ? (velocity), e (porosity), ?gr (density
of grain)
SIMULATION CURVES WITH PHREEQC
PHREEQC (Versión 2) 7 is a computer program for
one-dimensional reactive transport calculations
designed by the U.S. Geological Survey. The
user-friendly interface is useful for simulations
of many practical problems in hydrogeochemistry.
LAS homologues are defined as SOLUTION_MASTER_SPEC
IES. Initial SOLUTION_SPECIES and initial sorbed
species are quantified to start the simulation.
PHREEQC allows for several options in the
simulation of surface reactions. However, in this
case we defined kinetic sorption reaction for
different species (no surface definitions are
needed) Tebes-Steven et al., 8 defined kinetic
sorption for solution species by the rate
equation where Ci is either LAS homologue
(mol/L) and Cads their sorbed concentration
(mol/kg sediment), Km is the transfer coefficient
(hr-1) and Kd is the distribution coefficient
(L/kg). The values of the coefficients are given
in the following table
MATERIALS AND METHODS
To study the physicochemical desorption of LAS,
two laboratory experiments with columns
containing 100 sand (Test I) and 75 sand 25
soil (Test II) have been conducted. The
experimental set-up consisted of a cylindrical
stainless steel column filled with soil and
connected to a HPLC pump 3. Additional details
of specific LAS experiments are reported
elsewhere 4,5.
Tap water composition concentration of major
ions (in table)
Agricultural soil CaCO3 38.3. Organic Carbon
0.78 sand 23.6, silt 38.0 and
clay 38.4
Commercial Sand Sea sand, purified (Merck)
LAS Standard 12.1 C10LAS, 34.1 C11LAS, 30.6
C12LAS and 23.2 C13LAS, donated by PETRESA.
LAS Analysis Samples injected in a HPLC.
Stationary-phaseLichrospher 10?m
100RP-8(25x0.46) Teknokroma Mobile phase
MeOH/H2O (85/15)0.5M NaClO4H2O Flow 0.8
mL/min . UV detector (254 nm)
RESULTS
Performed Column Experiments Thermostated
stainless steel column 22.4cm length, 2.5cm
internal diameter (25ºC). Column connected
to a HPLC pump (Shimadzu LC 9A)
COLUMN TRANSPORT PARAMETERS
The experimental breakthrough curves (with CaCl2
as tracer) were obtained prior to the LAS
desorption experiments. Hydrodynamic column
parameters were obtained using ACUAINTRUSION 6,
designed with Visual Basic 6.0 (Microsoft).
This graphical user interface calculates the best
fit of the experimental data (chloride
concentration (mmol/L) versus experimental time
(h)) with the analytical solution of the
convection-dispersion equation.
LAS DESORPTION EXPERIMENTS
CONCLUSIONS
A continuous 0.5 mL/min in-flow of filtered and
sterilised tap water containing 5 ppm LAS was
injected into both columns for several days until
the concentration at the outlet was close to that
of each homologue (C10, C11, C12, C13) injected.
Formaldehyde was included to avoid growth of
bacteria and hence microbiological
biodegradation. The desorption experiments
started when tap water without LAS was injected,
and the effluent was collected in small
proportions, at first every 20 minutes/sample,
and later every 100 minutes/sample. LAS samples
were analysed by HPLC using a UV detector (254
nm). Experimental results from sand columns
showed that the concentration of different
homologues, in general, decreases sharply within
a relatively short time, whereas the experiment
with sand and soil exhibited more dispersive
spreading.
Two continuous LAS desorption experiments have
been carried out in columns containing 100 sand
and 75 sand 25 soil. Experimental results
from sand columns showed that the concentration
of different homologues, in general, decreases
sharply within a relatively short time, whereas
the experiment with sand and soil exhibited more
dispersive spreading. PHREEQC was applied in both
cases assuming convective-dispersive transport
and kinetic sorption reaction. Distribution
coefficients, determined earlier using
experimental data, are larger in tests employing
soil (greater sorption). Transfer coefficients,
which increase with homologue chain length, were
kept constant during the two tests. Simulated
results are in accord with experimental data.
Calculated sorbed homologue concentrations are
greater in tests employing soil and a longer
desorption time is expected.
REFERENCES
1 Jensen J., 1999. The Science of Total
Environment, 226, 93-111. 2 Verge C., Moreno A,
Bravo J. and Berna J. L., 2001. Chemosphere, 44,
1749-1757. 3 Gomis, V., Boluda, N. and Ruiz,
F., 1997. J. Cont. Hydrol., 29, 81-91. 4 Boluda
N., León V. M., Prats D. and Chorro M.C., 2005.
10th Med. Congress of Chem. Eng.
5 Boluda N., Cases, V., León, V.M., Gomis, V.
and Prats, D., 2007. Hidrol. y aguas subt., 22.
IGME. Spain. 6 Boluda Botella, N., Gomis, V.
and Pedraza, R., 2006. 1st SWIM-SWICA. Cagliary
(Italy). 7 Parkhurst, D.L. and Appelo, C.A.J.,
1999. U.S. Geological Survey. Water Res. Inv.
Report 99-4259, 312 pp. 8 Tebes-Stevens, C.,
Valocchi A.J., VanBriesen J.M., Rittmann B.E.,
1998. J. of Hydrol., v. 209, p. 8-26.