Complete

3rd AMIREG International Conference (2009): Assessing the Footprint of 1
Resource Utilization and Hazardous Waste Management, Athens, Greece
Environmental monitoring of water resources as a project management tool

in the extractive industry
K. Adam
School of Mining and Metallurgical Engineering, National Technical University of Athens, Greece
E. Tziritis
Department of Economic Geology and Geochemistry, Faculty of Geology & Geoenvironment,
National and Kapodistrian University of Athens, Greece
A. Tsakarisianos
ECHMES Ltd., Environmental Consultants, Greece
ABSTRACT spatial distribution of population. The final out-

come is an integrated scheme which incorpo-
Within the framework of the sustainable devel-
rates the fundamental principles of sustainable
opment, environmental monitoring of water re-
development and environmental protection. A
sources provides fundamental data and indica-
metallic mine in Greece is used as a case study.
tors subsequently used for planning and siting of
projects and activities, compilation of strategic
plans, and the assessment of their potential pres-
sures to natural and manmade environment. In-
tegrated water monitoring programmes consist
Project Management tools in the extractive in-
dustry, given the specific features of this indus-
try, relevant to surface and underground exploi-
tation of mineral resources, water abstraction,
land take for the disposal of products, byprod-
ucts and waste etc.. The assessment of both
qualitive and quantitive characteristics of water
resources within the productions site, as well as
upstream and downstream the mining activities
emphasis given for sensitive natural receivers ,
is essential in order to obtain a holistic approach
of the existing conditions, and to predict the fu-
ture trends. Monitoring data are subsequently
used in the design of water resources protection
measures integrated in the project design. The
aim of the present study is to determine the fac-
tors that form the model of an environmental
water monitoring program to be applied in the
extractive industry. These factors include the
qualitative and quantitative characteristics of
ground and surface waters, the total number of
sampling sites, and the frequency of sampling.
The above are directly related with the specific
conditions of the area, such as the hydro-
geological and hydrological setting, vegetation,
land use, sites of special environmental interest,
Europium removal from aqueous solutions using barley rootlets as

biosorbent
V. Anagnostopoulos, M. Anzaoui and B. Symeopoulos

Radiochemistry Lab, Department of Chemistry, University of Patras, Greece
ABSTRACT accumulation of metal ions by materials of bio-

logical origin. The process involves two steps:
The research is focusing on recovery of lantha-
contact of the biomass with the metal contami-
nides either for minimizing the industrial proc-
nated wastewater and separation of the metal -
esses losses or for reasons related to radioactive
loaded biosorbent from the metal - depleted ef-
waste management, since Eu and few more lan-
fluent (Ahmad et al., 2004). Materials such as
thanides simulate the behavior of long lived fis-
barley rootlets, which constitute by-products of
sion products like Am(III) and Cm(III). The po-
malting industrial processes are relatively abun-
tential utilization of barley rootlets, a by-product
dant and low-cost (Low et al., 2000). Europium
of malting industrial process, as a preconcentra-
was chosen as an analogue for trivalent actinide
tion medium of europium is investigated in this
ions (Markai et al., 2003; Kolokassidou and Pa-
study. The effect of solution pH, contact time
shalidis, 2006).
and initial concentration under constant ionic
The objective of the present study is to inves-
strength and temperature were studied. Uptake
was found to be rapid and kinetic data were fit- tigate the potential removal of europium from
ted to pseudo-second order and Morris-Webber aqueous solutions by barley rootlets. It is worth
models. The equilibrium process was described mentioning that this study is the first exploration
by the Langmuir isotherm model yielding concerning the potential use of this material as a
maximum sorption capacity 156 mg Eu/g of dry preconcentration medium of a heavy metal or an
biomass compared to 86 mg/g of maximum ca- organic contaminant.
pacity of activated carbon under the same condi-
tions, proving barley rootlets to be a more effi-
2. EXPERIMENTAL PART
cient sorbent.
Biomass, namely barley rootlets, was supplied
by a local brewery and was dried at 50oC over-
1. INTRODUCTION night. It was sieved and the fraction 0.18<d<1
Contamination of the environment by toxic mm was chosen for further experiments as
metals such as heavy metals and radionuclides higher europium uptake was observed. A stock
is mainly caused either by industrial activities or europium solution was prepared by dissolving
activities related to diverse applications of nu- the appropriate amount of Eu2Ο3 (Alfa Aesar,
clear energy. Although the removal of toxic 99.99% purity) in acidified triply distilled water.
metals has been practiced for decades, the high The resulted solution was standardized by titra-
cost of common physic-chemical techniques tion with a standard EDTA solution (Merck, Ti-
(e.g precipitation, ion exchange, reverse osmo- trisol). Fresh working solutions were prepared
sis, electrochemical treatment etc.) limits their by diluting appropriate portions of the stock so-
routine application and this is where biosorption lution with triply distilled water.
can cover those niches (Volesky, 2003). Batch sorption experiments were carried out
The term biosorption is used to describe the bringing in contact 0.01 g of dry biomass with
10 ml of europium solution of known initial 160

concentration and the suspension was agitated
on a rotary shaker at 25 oC. The suspension was
140
agitated for a length of time varying between 1 120
min and 24 h in order to determine the neces-
q (mg g-1)
100
sary contact time for attaining equilibrium. For 80
the experiments concerning the effect of pH and
60
the sorption kinetics, europium initial concen-
tration was 800 mg/L. For the isotherm experi- 40
ments initial metal concentrations ranged from 20

25 to 800 mg/L. In all experiments the ionic 0
strength (I) was adjusted by NaNO3 solution to 1 2 3 4 5
a total value 0.1 mol/L. At the end of agitation pH
time the suspension was centrifuged and the re- Figure 1: Effect of solution pH on europium uptake by
sidual (final) metal concentration in the super- barley rootlets. Error bars indicate the difference between
natant was determined spectrophotometrically, the two values of q determined by two independent ex-
by Arsenazo III method. The europium uptake periments.
(q/mg.g-1) was calculated from the difference
between the initial (Cin) and the final metal con-
centration (Cfin) using the equation:
q = (Cin − C fin )⋅V / mb (1)
where, mb (g) is the biomass dry weight and V
(l) the volume of the solute.
For each experiment a set of three vials were
used. Two of them, almost duplicate, contained
the metal solution and the biomass, while the
third one contained only the europium solution
without biomass (control vial). All the vials of a
set were treated the same way exactly and si- Figure 2: Speciation of europium under the conditions
studied.
multaneously.
3.2 Sorption kinetics

3. RESULTS AND DISCUSSION
Kinetic data at pH 4.5 at 25 oC are shown in
3.1 Effect of pH Figure 3.
Samples with different pH values ranging from It can be seen that 1 hour is a sufficient time
1.5 to 4.5, were agitated for 24 h, at 25 oC. The to attain equilibrium. Kinetic data were applied
effect of pH on europium uptake is shown in to different equations, namely pseudo-second
Figure 1. As can be seen, uptake increases sig- order kinetic, Equation 2 (Argun et al., 2007)
nificantly with pH. and Morris-Weber, Equation 3 (Karthikeyan
Taking into consideration that the speciation et al., 2005).
of europium remains practically the same in the t / q = 1 / h + (1 / qe ) ⋅ t (2)
pH region between 1 and 5 (Fig. 2), it may be
concluded that the uptake increase is due to the
biomass surface charge. Generally at low pH the q = Ridr t + C (3)
net surface charge of the biomass is positive,
where, qe and q stand for amounts of europium
thus the approach of positively charged euro-
sorbed (mg.g-1) at equilibrium and at time t
pium species is inhibited, whilst at higher pH
(min), k2 is the rate constant of the pseudo sec-
values the net surface charge turns out to be
ond order kinetic equation, h = k2qe2, known as
negative (Göksungur et al., 2005; Say et al.,
initial sorption rate and Ridr is the intra-particle
2001).
160 is deduced that the sorption process occurred in

two stages: by surface sorption and intraparticle
140
diffusion. The larger the intercept, the greater
120 the contribution of the surface sorption in the
rate-controlling step (Kalavathy et al., 2005).
q (mg g-1)
100
3.3 Effect of metal ion concentration
80
Isotherms were drawn at pH value 4.5 and con-
60 tact time 24h. Increase of uptake was observed
40
with the increase of uranium initial concentra-
tion indicating that uranium initial concentration
20 is the driving force for biosorption. For com-
0 50 100 150 200 250
parison reasons, parallel experiments with acti-
t (min) vated carbon (Merck) as sorbent instead of bar-
Figure 3: Effect of contact time on the uptake of europium ley rootlets were carried out under identical
by barley rootlets biomass.
conditions.
The effect of tested metal ion concentration
diffusion rate constant. was analyzed in terms of Langmuir model
A plot of t/q versus t was used to evaluate the (Altin et al., 1998):
pseudo second order model and the values of k2
and h were found 4.3·10-4 g mg-1 min-1 and 10 q = (qmax ⋅ K L ⋅ C fin )/ (1 + K L ⋅ C fin ) (4)
mg g-1 min-1 respectively (correlation coefficient
R2 = 0.991). The results of the data fit to Morris- where qmax (mol g-1) maximum metal uptake, KL
Webber model can be seen in Figure 4. Langmuir constant and Dubinin-Radushkevich
There are four main stages in the process of model (Bering et al., 1972):
adsorption by porous adsorbents: i) solute trans- ln q = ln X m − K ' ε 2 (5)
fer from the bulk solution to the boundary film
that surrounds the adsorbent’s surface, ii) solute where ε is the Polanyi potential
transport from the boundary film to the adsorb-
ent’s surface, iii) solute transfer from the ad- ε = RT ⋅ ln (1 + 1 / C fin ) (6)
sorbent’s surface to active intraparticular sites,
iv) interactions between the solute species and Xm the adsorption capacity of the sorbent (mol
the available adsorption sites on the internal sur- g-1), K´ is a constant related to the adsorption
faces of the adsorbent (Argun et al., 2007). energy (mol2 kJ-2), T temperature (K) and R the
Ridr was found to be 19 mg g-1 min-1/2 (corre- gas constant (kJ K-1 mol-1). The mean adsorp-
lation coefficient R2 = 0.999). From Figure 4, it tion energy E (kJ mol-1) can be obtained from
160
the K´ values using Εquation 7:
E = (− 2 K ')
−1 / 2
140 (7)
120 For E<8kJ mol-1, physisorption dominates the
sorption mechanism, if E is between 8 and 16 kJ
q (mg g-1)
mol-1 ion-exchange is the dominant factor and if

100
80 E>16kJ mol-1 sorption is dominated by particle

diffusion.
60
Results are summarized in Tables 1 and 2.
40 Results suggest that sorption of europium
both onto barley rootlets and activated carbon is
20
2 4 6 8 10 12 14 16 dominated by ion-exhange mechanism.
As may be seen from Figure 5 barley rootlets
t 1/2 (min1/2) are proved to be a much more efficient biosor-
Figure 4: Intraparticle diffusion plot for the sorption of bent for the removal of europium from aqueous
europium onto barley rootlets.
solutions than activated carbon. More specifi-
Table 1: Regression parameters for Langmuir isotherms. ardous materials 141: 77-85.
103 x qmax 10-2 x KL 2 Bering, B.P., M.M. Dubinin and V.V. Serpinsky, 1972.
Biosorbent R On thermodynamics of adsorption in micropores,
(mol g-1) (l mol-1)
Journal of Colloid and Interface Science 38: 185-194.
Barley Göksungur, Υ., S. Üren and U. Güvenc, 2005. Biosorp-
1.0 23 0.999
rootlets tion of cadmium and lead ions from ethanol treated
Activated waste baker’s yeast biomass, Bioresource Technology
0.58 43 0.997
carbon 96: 103-109.
Kalavathy, M.H., T. Karthikeyan, S. Rajgopal and L.R.
Table 2: Regression parameters for Dubinin Radushke- Miranda, 2005. Kinetic and isotherm studies of Cu (II)
vich isotherms. adsorption onto H3PO4-activated rubber wood saw-
103 x K´ E dust, Journal of Colloid and Interface Science 292:
Biosorbent R2
(mol2 kJ-2) (kJ mol-1) 354-362.
Barley Karthikeyan, T., S. Rajgopal and L.R. Miranda, 2005.
5.1 9.9 0.967 Chromium (VI) adsorption from aqueous solutions by
rootlets Hevea Brasiliensis sawdust activated carbon, Journal
Activated of Hazardous Materials 124: 192-199.
0.58 12 0.969
carbon Kolokassidou, K. and I. Pashalidis, 2006. Potentiometric
investigation on the interaction of humic acid with
Cu(II) and Eu(III) ions, Radiochimica Acta 94: 1-4.
cally, according to the qmax calculated from Low, K.S., C.K. Lee and S.C. Liew, 2000. Sorption of
Langmuir model, barley rootlets yield a maxi- cadmium and lead from aqueous solutions by spent
mum sorption capacity under the conditions grain, Process Biochemistry 36: 59-64.
studied of 1.0·10-3 mol g-1 (156 mg g-1) in con- Markai, S., Y. Andrès, G. Montavon and B. Grambow,
2003. Study of the interaction between europium (III)
trast to activated carbon which yields 0.58·10-3 and Bacillus subtilis: fixation sites, bioorption model-
mol g-1 (86 mg g-1). ling and reversibility, Journal of Colloid and Interface
Science 262: 351-361.
Volesky, B., 2003. Sorption and Biosorption, Montreal:
REFERENCES SV Sorbex Inc.
Ahmad, W.A., J. Jaapar and M.A.K.M. Zahari, (ed. Say, R., A. Denizli and M.Y. Arica, 2001. Biosorption of
Ashok Pandey), 2004. Concise Encyclopedia of Bio- cadmium (II), lead (II) and copper (II) from the fila-
resource Technology, USA: The Haworth Press Inc. mentous fungus Phanerochaete chrysosporium, Biore-
Altin, O., H.O. Ozbelge and T. Dogu, 1998. Use of gen- source Technology 76(1): 67-70.
eral purpose adsorption isotherms for heavy metal-
clay mineral interactions, Journal of Colloid and Inter-
face Science 198: 130-140.
Argun, M.E., S. Dursun, C. Ozdemir and M. Karatas,
2007. Heavy metal adsorption by modified oak saw-
dust: thermodynamics and kinetics, Journal of Haz-
140
120
100
q (mg g-1)
80
60
40
20
0
0 100 200 300 400 500 600 700 800
Cfin (mg l-1)

Figure 5: Experimental isotherms for europium sorption
onto barley rootlets (black boxes) and activated carbon
(white boxes).
Preliminary investigation of the in-plane shear behavior of masonry

structures
Ch. Apostolaki, Z. Agioutantis, V. Perdikatsis and Th. Markopoulos

Department of Mineral Resources Engineering, Technical University of Crete, Hania, Greece
ABSTRACT of the world heritage. Unreinforced masonry is

a composite material made of brick units and
The aim of this study was to approach and iden-
mortar joints. Structural walls were mainly de-
tify the mechanical behavior of the interfaces in
signed to resist gravity loads. The overall be-
masonry historical structures. As a result identi-
havior of this set is depended by several factors
cal materials to the original ones such as natural
such as: properties of the component materials
hydraulic lime and clay brick are utilized. The
(bricks, mortars etc.), characteristics of the
natural hydraulic lime was prepared by firing
bonding between the various components, unit
marly limestone available on the Island of Crete
size, joint thickness, joint orientation and
and slaking in open air which results in the con-
workmanship (Sutcliffe et al., 2001; Abdou
version of the CaO content to Ca(OH)2.
et al., 2006; Gabor et al., 2006; Fouchal et al.,
Masonry is a composite material made of
2009; Chaimoon and Attard, 2009).
units such as bricks, stones, etc. and joints
Horizontal loads generate severe forces to the
which can be natural hydraulic based mortar,
walls caused mainly by earthquake and wind
lime mortar etc. The mortar joints of masonry
loads. It is well known that for the successful
act as planes of weakness, which displays direc-
functioning of the masonry in resisting lateral
tional properties. Furthermore masonry material
loads depend strongly on their resistance to
constituting monumental and historical struc-
shear forces (Marzahn et al., 1997; Fouchal
tures is often characterized by low tensile
et al., 2009).
strength.
For the purpose of characterizing a behavior
Regarding natural hydraulic mortars as a
of masonry under shear, different test methods
composite material, current research copes only
to determine the strength parameters c (cohe-
with properties such as durability, permeability,
sion) φ (internal friction angle) have been de-
and compressive strength. This paper presents
veloped by different researchers.
experimental results regarding mechanical
In most of these methods a shear test on one
properties of natural hydraulic lime based mor-
or two surfaces needs to be performed with or
tars without binding to bricks as well as the
without lateral forces. Figure 1 shows a typical
binding characteristic (shear strength) of natural
set up for performing a shear test using two
hydraulic lime based mortar to clay brick. The
bricks (units) and one interface (joint). This ver-
shear strength of the brick - natural hydraulic
sion of shear test set up was introduced by Van
based mortar joints is determined using two s-
Der Pluijm (Chaimoon and Attard, 2009). Fig-
pecially designed devises.
ure 2 shows a typical set up for performing a
Results can be used to determine the charac-
shear test using 3 brick units separated by 2 in-
teristics and potential compatibility with exist-
terfaces (Gabor et al., 2006; Lourenco et al.,
ing brickwork constructions.
2004; Sarangapani et al., 2005; Prakash and
Alagusundaramoorthy, 2008).
1. INTRODUCTION In addition a few researchers (Gabor et al.,
2006) have proposed a shear test on masonry
Masonry buildings constitute a significant part blocks where the block is loaded along the di-
mode a mode b mode c

Figure 4: Failure modes of masonry specimens.
accommodate before any slip occurs. This value

can then be used to solve for normal shear
stresses along the interface.
Figure 1: Couplet test set up for shear-compression load- Typical failure modes for these tests are
ing developed by van der Pluijm. shown on Figure 4 (modes a, b and c) where
masonry may fail along either one of the brick
interfaces or even across its one mass or in a
combination of the two above modes.
The failure behavior of masonry joints under
shear can be represented by the Coulomb fric-
P=0 P=0 tion law, for lower normal compressive stresses
to the joint σ, lower than 2 MPa (Oliveira,
2003), being given by:
τ = c tan φ ⋅ σ (1)
Figure 2: Triplet test set up for shear-compression load- where:

ing. c: represents the cohesion (bond shear
ameter at an angle between the loading axes and stresss)
a masonry plane ranging between 15 and 45 de- tanφ: is the tangent of the friction angle of
grees (Fig. 3). This testing configuration has the the interface between the unit and the
advantage that force needs to be applied along mortar.
only one direction, while the previous methods For higher normal compressive stresses, the
require two sets of force/displacement applica- Coulomb friction law is not valid and the ap-
tion to be in effect. pearance of crushing/shearing of the units along
In all cases results can be obtained in two with cracking occur. In this case a according to
forms: (a) as a force/displacement graph; (b) as Lourenco et al. (2004), a cap model can be
a shear stress/shear normal stress graph for dif- adopted to represent the failure of the combined
ferent confining stress values. Results obtained joint-unit assembly.
from the first set of graphs will help quantify the The first part of this paper deals with the
maximum force (stress) that the interface can characterization of the materials (bricks and
mortar) and the preparation of brick / mortar
specimens for the shear test. The shear behavior
of masonry involving two different assemblies
either composed of one and two interfaces are
tested using two specially designed devices.
2. EXPERIMENTAL METHOD
2.1 Characterization of materials
The binder used for the preparation of examined
mortars in this study was NHL slow air slaked
lime. The chemical and mineralogical analysis
Figure 3: Diagonal compression of the masonry wall. of that binder is given in Apostolaki et al.
GENERAL BUILDING
100
triplets presented in this study were cured for 7
100
90
100 100
89
100 100
98 days by keeping them in laboratory atmosphere
80 74 90
conditions in order to test the below described
% passing the sieve
70 70
70 devices. All tests were carried out in dry speci-
60
50 51 mens. In order to ensure a standardization of
40 40 triplet test specimens along with the verticality,
30
20 15
19
General Building Ch EN 13139 lower margin EN 13139 upper margin
different types of metal moulds have been con-
10 5
4 5
structed (Fig. 6).
0 There are two types of designed devices and
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5
Sieve opening (mm) as a result two types of test specimens. Figure 7
Figure 5: Grain size distribution curves of the aggregate shows the first designed device and a triplet test
fraction. specimen that consists of three bricks and two
(2009). mortar joints. Figure 8 shows the designed con-
Cylindrical mortar specimens were produced figuration for the shear test of prisms. Both de-
with a 1:2:0.7 binder/aggregate/water ratio. The signed devices were tested in order to determine
aggregate was a quartzitic sand. In order to the shear strength of the interface between mor-
identify the grain size distribution of the aggre- tar and brick. Shear bond strength of the brick
gates the use of the mortar must be known. In mortar joints was determined for brick and NHL
this study mortars for masonry and general slow based mortar.
building purposes are examined. Figure 5 shows From Figure 7 it is clear that the horizontal
the grain size distribution of the aggregates used movement was restrained and the middle brick
for the preparation of mortars as well as the was free to move vertically. Vertical shear load
standard constrained margins for this purpose was applied using a constant load rate causing
according to EN 13139. As shown in this figure the bond between brick and mortar joint to fail.
the applied grain size distribution curve lies be- Consequently the shear strength could be esti-
tween the standard limits. mated.
Figures 7 and 8 clearly show the force ap-
2.2 Preparation of specimens for compressive plied to each device. In the first case the shear
strength is equal to the ratio of the applied force
and tensile strength
F to surface A1 divided by 2 due to the existence
For the compressive and tensile tests of mortars of two joint surfaces. It is assumed that the lat-
cylindrical specimens 95 mm in height with a eral force is zero. The horizontal supports are in
diameter of 47 mm and 28 mm height with a di- place to prevent out of plane failure due to
ameter of 47 mm were prepared respectively. bending. In Figure 8 the shear strength is equal
Thoroughly mixed mortar samples were filled to the component force that is parallel to the
into a PVC cylindrical mould in three layers,
each layer is tamped 25 times using a tamping
rod. After 24h of casting the specimens were
removed from the moulds and then placed on
90% RH conditions for curing.
2.3 Materials and preparation of prisms and mould a

triplets
The fired clay bricks used for the experimental
tests were full (solid). The dimensions of clay
brick were 100 mm (length) x 50 mm (width) x
50 mm (height). The bricks used for casting
prisms and triplets were prewetted before lay-
ing. The duration of prewetting of the bricks
was 10 min. A mortar joint thickness of 10 mm
was maintained for all cases. The prisms and mould b
Figure 6: Different types of metal moulds.
Table 1: Compressive and tensile strength of NHL slow curves of shear stress with respect to the relative
binder (NHL based mortar 28 days). displacement were plotted. Results at this stage
Compressive strength Tensile strength of the experiments are considered mainly quali-
Mean of 6 sp. 2.53 MPa Mean of 6 sp. 0.76 MPa tative. These results are presented and discussed
below.
Figure 9 and Figure 10 show shear stress-
displacement and force - displacement curves of
two couplet and one triplet specimen that were
tested after a curing time of 7 days respectively.
The shear strength varies between 0.4 to
0.7 MPa, while the applied force at the moment
of rupture varies between 3000 and 5000 N.
From the above tested couplets it is clear that
Figure 7: Triplet test set up for shear-compression load- there is no plastic zone before rupture. On the
ing.
contrary, these preliminary results for the triplet
joint plane via surface A2. The component force test indicate that triplets may exhibit a different
is equal to F sin(45o). To date these devices behavior. However, this may be due to the ge-
have been tested only for brick and NHL slow ometry and the constraints of the experimental
based mortal with a curing time of 7 days. In setup. Clearly, more data must be collected in
order to determine the shear strength of other order to draw safe conclusions.
binders and other curing times work is still in
progress. 3.3 Failure modes
Results show that all types of failure modes de-
3. RESULTS AND DISCUSSION scribed above (Fig. 4) are present. Likewise, as
3.1 Compressive and tensile strength couplet 1 couplet 2 triplet
The samples were subjected to a monotonously 0.80
increasing load until damage occurred. The 0.70

Shear stress τ(Mpa)
0.60
loading conditions were monitored during the 0.50
test. The mean value of six tested specimens is 0.40
reported in Table 1 for the compressive as well 0.30
as tensile strength of the mortar for a curing 0.20
0.10
time of 28 days. 0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00
Displacement (mm)
3.2 Shear strength
Figure 9: Comparison of numerical results for the shear
In order to establish a shear behavior law, stress - displacement curves between two different de-
vices (couplet and triplet specimens).
couplet 1 couplet 2 triplet

6.00
5.00
4.00
Force (KN)
3.00
2.00
1.00
0.00
0.00 0.50 1.00 1.50 2.00 2.50 3.00
Displacement (mm)
Figure 10: Comparison of numerical results for the force -

Figure 8: Couplet test set up for diagonal compression displacement curves between two different devices (cou-
loading. plet and triplet specimens).
Social Fund (75%) and National Resources

(25%).
REFERENCES
Abdou, L., R.A. Saada, F. Meftah and A. Mebarki, 2006.
Experimental investigations of the joint-mortar behav-
iour. Mechanics research communications 33. pp. 370-
384.
Apostolaki, Ch., Th. Markopoulos, V. Perdikatsis and Z.
Agioutantis, 2009. An investigation of the hydration
and carbonation mechanisms of natural hydraulic
limes. Proceedings, 3rd AMIREG International Con-
ference, September 7-9, Athens, Greece.
Chaimoon, K. and M.M. Attard, 2009. Experimental and
numerical investigation of masonry under three-point
bending (in-plane). Engineering structures 31. pp.
103-112.
Fouchal, F., F. Lebon and I. Titeux, 2009. Contribution to
the modeling of interfaces in masonry construction.
Figure 11: Failure modes. Construction and building materials 23. pp. 2428-
2441.
it is shown in Figure 11, all rupture modes may Gabor, A., E. Ferrier, E. Jacquelin and P. Hamelin, 2006.
develop when testing a single triplet specimen. Analysis and modeling of the in-plane shear behaviour
Work is still in progress in order to determine of hollow brick masonry panels. Construction and
building materials 20. pp. 308-321.
whether the failure mode influences the shear Lourenco, P.B., J.O. Barros and J.T. Oliveira, 2004. Shear
strength of the mortar-brick interface. testing of stack bonded masonry. Construction and
building materials 18. pp. 125-132.
Oliveira, C., 2003. Experimental and numerical analysis
4. CONCLUSIONS of blocky masonry structures under cyclic loading.
PhD thesis, University of Minho.
Comparisons between the binding behavior re- Prakash, S.S. and P. Alagusundaramoorthy, 2008. Load
sults obtained on couplets and triplets of brick/ resistance of masonry wallettes and shear triplets ret-
natural based mortar interfaces showed that the rofitted with GFRP composites. Cement & concrete
magnitude of the shear stress failure point and composites 30, pp.745-761.
the displacements were similar in both cases. Sarangapani, G., B.V. Venkatarama Reddy and K.S. Ja-
gadish, 2005. Brick-mortar bond and masonry com-
The experimental results show that the shear pressive strength. Journal of materials in civil engi-
behavior of the triplet specimen is elastoplastic neering. March/April 2005, pp.229-237.
while in the case of the couplet specimen (di- Sutcliffe, D.J., H.S. Yu and A.W. Page, 2001. Lower
agonal compression) a brittle behavior (sudden bound limit analysis of unreinforced masonry shear
failure) is recorded. walls. Computer and structures 79, pp.1295-1312.
Work is still in progress in order to collect
more data with respect to the mechanical prop-
erties in order to relate them with mechanisms
of hydration and carbonation that take place in
the binders over time as well as with the physi-
cal parameters such as crystallite size described
in Apostolaki et al. (2009).
ACKNOWLEDGMENTS
This research is supported by the General Secre-
tariat of Research and Technology of Greece
through the PENED 2003 program, contract
number 03ED604, co-funded by the European
An investigation of the hydration and carbonation mechanisms of natural

hydraulic limes
Ch. Apostolaki, Th. Markopoulos, V. Perdikatsis and Z. Agioutantis

ABSTRACT lime and named it “Roman cement”. This name

was given due to the fact that Romans and
Natural Hydraulic Limes (NHL) are produced
Greeks had discovered the process of mixing
by firing of natural occurring raw materials,
limestone with “pozzolans” in order to produce
such as marly limestone, below the sintering
a material with hydraulic properties. The use of
temperature. NHL contains mainly portlandite
marly limestone replaced the need of mixing
Ca(OH)2 and larnite (Ca2SiO4). By combining
two materials together as done by Romans be-
NHL with aggregates, a binding material (i.e.
cause marly limestone in itself contains the nec-
mortar) is formed that can be used in restoration
essary mineralogical composition. In 1824 Jo-
and conservation applications on monuments
seph Aspin was credited with the discovery of
and historical buildings. Such materials have
an artificial hydraulic lime known as Portland
been used by Greeks and Romans since antiq-
cement. Although a precursor to the modern
uity.
Portland cement, Aspin’s discovery really bears
During restoration of historic buildings it is
no resemblance to what is known today as Port-
advisable to use compatible materials, both with
land cement. The main reason that Portland ce-
respect to their origin and to their physical and
ment became the major binding material in con-
mechanical properties.
struction from the end of the 19th century on,
This paper presents experimental results re-
was the standardization of its production (Calle-
garding the hydration and carbonation mecha-
baut, 2001).
nisms of NHL. The natural hydraulic lime was
Nowadays binders produced by firing of
prepared by firing marly limestone available on
marly limestone below the sintering tempera-
the Island of Crete and slaked in open air, which
ture, containing mainly portlandite Ca(OH)2 and
results the conversion of the CaO content to
larnite (Ca2SiO4) are known as Natural Hydrau-
Ca(OH)2. Experiments were conducted to de-
lic Limes. Natural Hydraulic Limes are very
termine the mineralogical phase changes in
important for the restoration and conservation of
NHL, during hydration and carbonation proc-
monuments and historical buildings. During
esses, as well as the portlandite crystallite size
these restoration projects the choice of compati-
formation during different slaking rate.
ble materials is the key so as to match, as
closely as possible, those used originally in or-
1. INTRODUCTION der to provide some continuity with the past in
respect to the physical and mechanical proper-
The properties of hydraulic lime were first rec- ties of NHL based mortars. Furthermore the
ognized with the discovery by John Smeath in knowledge of damages caused by the use of
1756, who by examining limes of different ori- Portland cement in the conservation of historical
gin found that those which gave the best results buildings, due to its properties (i.e., hardness,
as mortars contained a considerable proportion rigidity, impermeability, contamination of solu-
of clay matter (Lea, 2004). Forty years later in ble salts etc) increase the necessity of the usage
1796, James Parker patented natural hydraulic of NHL based mortars (Puertas et al., 1994;
Collepardi, 1999; Ciliberto et al., 2008). ine the mechanisms of carbonation and hydra-
This paper is divided into two parts. The first tion processes.
deals with the mineralogy and the process of
production of the end products used in the sec- 2.2 Experimental procedure
ond part as binders. The second part focuses on A quantity of 3 g of both slow slaked and rapid
the examination of carbonation and hydration slaked NL along with 3 g of both slow slaked
mechanisms occurring in these binders. and rapid slaked NHL were exposed in continu-
Current research deals with the carbonation ous humidified airflow of 90% RH. Under the
process of portlandite affected by factors such same conditions a quantity of 3 g of both slow
as relative humidity (RH), temperature (T), con- slaked and rapid slaked NHL along with 3 g of
centration of CO2 and grain size (El-Turki et al., pure larnite were also exposed as pastes by add-
2009; Castellote et al., 2008). This paper exhib- ing 6, 4 and 4 ml of deionized water respec-
its the importance of mineralogical phase tively. The samples were placed in PVC cylin-
changes in NHL, through hydration/ carbona- ders of 2 cm in diameter and 2 cm height to en-
tion processes as well as the significance of sure identical geometry and uni-directional car-
portlandite crystallite size. bonation on the exposed surface. From each of
the above mentioned specimens, four separate
2. EXPERIMENTAL METHOD samples were exposed. The analysis was per-
formed after curing of 0 (reference sample), 10,
2.1 Materials and preparation of binders 20 and 30 days.
The raw materials used were marly limestone
from Crete, a pure fine crystalline calcitic mar- 2.3 Analytical techniques
ble and a pure synthetic larnite, fired at 900oC The mineralogical analysis was carried out by
for 12 hours. The pure synthetic larnite was X-ray powder diffraction analysis (XRD) on a
produced by mixing CaO and SiO2, in 2:1 mol Bruker D8 Advance Diffractometer, using Ni-
ratio. The calcination of limestone or marble filtered Cu Kα radiation (35 kV 35mA) and a
produces lime that is transformed by air or wet Bruker Lynx Eye strip silicon detector. Data
slaking to portlandite (Ca(OH)2) and could be were collected from 4° to 70° 2θ with a step size
characterized as Natural Lime (NL). During fir- of 0.019°and a count time of 31.8 s per step.
ing of marly limestones, larnite (Ca2SiO4) and The crystalline phases were identified using
minor amounts of Ca(Al,Fe)2O4 are formed, as the Diffrac plus Software (Brucker AXS) and
well as lime that changes after slaking into the the PDF database. The presence of amorphous
form of portlandite. phases in hydraulic binders cannot be excluded
Each of the produced NHL and NL (natural a priori, (Gualtieri et al., 2006) therefore; co-
lime) were divided into two parts. The first part rundum powder has been added as an internal
was slaked in open air with RH 55-60% (slow standard. It allows quantification of the amor-
hydration), and the second part was rapidly phous content. The quantitative analysis was
slaked in water. Sampling took place in order to performed by the Rietveld method using the
identify, through mineralogical analysis, the ex- TOPAS software program. This approach leads
act moment of complete transformation of lime to stable refinements and physically meaningful
to portlandite so as to secure the desired minera- information such as crystallite sizes, (Füllmann
logical composition before the carbonation et al., 2001) and profile shapes. They are char-
process starts. The above procedure is crucial, acterised in terms of the physical parameters de-
since the content in free lime should be as low scribing the diffractometer and the diffraction
as possible, and at the same time, reaction of process (Cheary and Coelho, 1992).
water with silicates and aluminates as well as Additionally Infrared (I.R) and thermo-
reaction of portlandite with atmospheric carbon gravimetric analysis were performed.
dioxide must be avoided to preserve the hydrau- Finally, chemical analyses by X-ray fluores-
lic and setting properties (Mertens et al., 2007). cence were conducted with a Brucker S2 Ranger
In addition, this mineralogical composition is of spectrometer.
great importance in this study in order to exam-
Table 1: Chemical and mineralogical analysis of the Table 3: Quantitative analysis of binders in t=0 days.
marly limestone. NHL NL larnite
Bulk composition of marly Insoluble Minerals rapid slow rapid slow C2S
limestone residue At0 Bt0 Ft0 Gt0 Et0
Oxides Minerals Minerals C2S 36.77 55.10 -- -- 100
SiO2 9.87 C3A 1.61 7.43 -- -- --
Al2O3 1.68 Amorf. 4.90 Amorf. 4.90 Periclase -- 1.27 -- -- --
Fe2O3 1.53 Calcite 80.22 -- -- CaO -- 1.89 -- -- --
CaO 49.05 Quartz 4.65 Quartz 4.65 Portland. 25.67 20.79 100 100 --
Chlo- Quartz 0.26 0.26 -- -- --
MgO 1.34 Chlorite 0.85 0.85
rite C4AF 2.04 7.83 -- -- --
K2O 0.52 Illite 7.64 Illite 7.64 C2SH 2.61 2.85 -- -- --
Kaolin- CA -- 0.52 -- -- --
L.O.I. 36.0 Kaolinite 1.73 1.73
ite Grossite 3.96 2.04 -- -- --
Total 99.99 Total 99.99 Total 19.78 Calcite 8.56 -- -- -- --
Arago-
Table 2: Chemical and mineralogical analysis of the NHL 1.74 -- -- -- --
nite
slow slaked binder.
Katoite 16.70 -- -- -- --
NHL slow slaked binder NHL after firing
Total 99.92 99.98 100 100 100
Oxides Minerals Minerals
SiO2 19.70 C2S 55.10 C2S 55.33
Al2O3 10.10 C3A 7.43 C3A 5.24 considerable amount after the almost total lime
Peri- transformation. Both the NL binders (rapid and
Fe2O3 2.68 Periclase 1.27 1.72 slow) in the t=0 days consists only of portland-
clase
CaO 61.30 CaO 1.89 CaO 29.85 ite. Synthetic larnite consists only of a pure
MgO 1.21 Portland. 20.79 Portlan 0.47 phase of larnite without any remnant from the
H2O 4.97 Quartz 0.26 Quartz 0.00 initial mixing material after firing.
C4AF 7.83 C4AF 7.41 Figures 1 to 5 present the mineralogical
C2SH 2.85 C2SH 0.00 changes of binders during the experiment under
CA 0.52 CA 0.00 continuous humidified airflow conditions de-
Gros- scribed above, after 0 (reference sample), 10, 20
Grossite 2.04 0.00 and 30 days. In particular Figures 1 and 2 pre-
site
Total 99.96 Total 99.98 Total 100.02 sent the changes for NHL slow and NHL rapid
binder respectively. The carbonation of port-
landite and the simultaneous increase of carbon-
3. RESULTS AND DISCUSSION ates are clearly observed in both cases. This be-
3.1 Mineralogy havior was expected, but the unexpected though
noticeable feature was the decrease of larnite
XRD phases as well as chemical analysis of the over time, which led to a further increase of car-
marly limestone and the end product of the bonates. The decomposition of larnite was
NHL slow slaked binder from the first part are CALCITE VATERITE ARAGONITE PORTLANDITE C2S
given in Tables 1 and 2 respectively. The NHL C4AF C3A C2SH CA Grossite CA2
slow slaked binder mineralogical analysis given Periclase Quartz CaO
60.000
in Table 2 is used as the initial NHL slow slaked
binder material for the experiment procedure in
50.000
%Concentration (Minerals)
zero curing time. The quantitative analysis of 40.000
the five different binders used in this experiment 30.000
on the first curing stage (t=0 days) is given in 20.000
Table 3. It is clearly shown in Table 3, that the 10.000
dominant silicate in the NHL slow slaked binder 0.000

is the dicalcium silicate (C2S). The aluminates 0 5 10 15 20 25 30 35
are tricalcium aluminate (C3A) and brown- BtN Days
milerate (C4AF). Portlandite is also present in a Figure 1: Mineralogical exchange of NHL slow slaked
binder over time.
CALCITE
C3A
ARAGONITE
C2SH
PORTLANDITE
Grossite CA2
C2S
Katoite
C4AF
Quartz
binder (Fig. 4) forms calcite, while the slow
70.000
slaked NL binder forms vaterite, aragonite and
60.000
calcite (Fig. 3). Furthermore, the total amount of
the carbonate mineral phases is higher com-
50.000
40.000
pared to the rapid slaked NL portlandite. The
same trend and behavior of portlandite is ob-
served also in NHL binders.
30.000
The most interesting observation of this spe-

20.000
10.000
cific experiment is shown in Figure 5; the de-
0.000
0 5 10 15 20 25 30 35 composition of larnite to carbonates. In the pure
AtN Days
larnite sample, the absence of portlandite in the
Figure 2: Mineralogical exchange of NHL rapid slaked reference sample (t=0) indicates that the de-
binder over time. composition of larnite took place according to
the following reactions:
100 Ca2SiO4 + 2H2O → 2Ca(OH)2 + SiO2amorph (1)
CALCITE VATERITE ARAGONITE PORTLANDITE
and
80
60
Ca(OH)2 + CO2 → CaCO3 + H2O (2)
40
This verifies the hypothesis that the reduction
20 of larnite is connected to the increase of port-
landite, which during the carbonation process
leads to the increase of carbonate mineral
0
0 5 10 15 20 25 30
GtN
Days
phases in the NHL binder. The decomposition
Figure 3: Mineralogical exchange of NL slow slaked of larnite offers higher amounts of amorphous
binder over time. SiO2 to the binder. The increase of amorphous
SiO2 improves the pozzolanic activity of the
100
binder during reaction with portlandite.
CALCITE VATERITE ARAGONITE PORTLANDITE In addition, Figure 6 shows TG/DTG curves
of the synthetic larnite after 30 days under con-
80
P
60
tinuous humidified airflow of 90% RH. The
presence of portlandite and calcite again proves
40
the decomposition of larnite according to previ-
20 ously mentioned chemical reactions.
Another very important result of this investi-
gation is connected with the crystallite size of
0
0 5 10 15 20 25 30
Days
FtN portlandite. It has been proved that the slaking

Figure 4: Mineralogical exchange of NL rapid slaked process (rapid or slow) in hydration of CaO of
binder over time. NHL or NL significantly influences the crystal-
known to us from our preliminary experiments. C2S Vaterite Aragonite Portlandite Calcite
This was the main reason for the inclusion of 100
synthetic larnite as well as the pure calcitic 90

80
marble, as reference materials, in this experi- 70
ment. The goal was to determine the mecha- 60
nisms of hydration and carbonation of the NHL 50
40
based binders. In order to verify the above 30
mechanisms the addition of the reference mate- 20
rials in the experiment were vital.

10
Consequently Figures 3 and 4 present the 0 5 10 15 20 25 30 35
carbonation mechanism of NL binders. The car-

Days
bonation of portlandite in a rapid slaked NL Figure 5: Decomposition of synthetic larnite to carbonates

over time.
44.5 0.05 In Figures 9 and 10 the hydrated phases of

44 aluminates over time are shown. In particular
43.5
0
Figure 19 shows the X-ray diffractograms of the
NHL slow slaked binder in 0 and 30 days. Fig-
Derivate Weight
43 TG
Weight (mg)
-0.05
42.5
DTG
ure 10 shows the X-ray diffractograms of the
42
CaCO3 -0.1 NHL rapid slaked binder. In both slaking proc-
esses (slow and rapid) calcium aluminum oxide
`
41.5
hydrate (Ca3Al2O6·xH2O) and katoite

-0.15
41
40.5
Ca(OH)2
-0.2 ((CaO)3Al2O3(H2O)6) are formed. The differ-
0 200 400
o
Temperature ( C)
600 800 1000
ence between them is that in the slow slaking
Et3P process calcium aluminum oxide hydrate is
Figure 6: TG/DTG curves of synthetic larnite after 30 formed in a higher quantity than katoite, which
days under continuous humidified airflow of 90% RH. is present only in small quantities. The opposite
is true in the rapid slaked process where we see
higher quantities of katoite but significantly
20000
slow slaked NL
lower quantities of calcium aluminum oxide hy-
drate.
rapid slaked NL
L in ( C o u n ts )
Portlandite ( 0 0 1) 4. CONCLUSIONS
10000
The conclusions that can be drawn from these

experiments are as follows:
First and foremost is the unexpected decom-
0
position of larnite. Previous studies consistently
13 14 15 16 17 18
2-Theta - Scale
19 20 21 22 23
mention the hydration of larnite. The results
presented in this paper based on analysis of pure
Figure 7: Portlandite’s peak shape of rapid and slow
slaked NL binder.
larnite paste samples support the fact that both
decomposition and carbonation of larnite takes
lite size of portlandite (Figs. 7 & 8). Rapid hy- place instead of hydration.
dration produces well crystallized portlandite Of equal importance is the relationship be-
with a crystallite size > 1800 Å for the NL and tween the crystallite size of portlandite and the
> 700 Å for the NHL, while slow hydration (slow and rapid) slaking processes in the hydra-
forms poor crystallized portlandite with a crys- tion of CaO, as well as with the rate of the car-
tallite size < 250 Å for the NL and < 160 Å for bonation of portlandite. On one hand, it was ob-
the NHL. The full width of half maximum served that the rapid hydration produces well
(FWHM) is inversely proportional to the crys- crystallized portlandite, while slow hydration
tallite size. forms poor crystallized portlandite. On the other
It is also observed that the speed of the car- hand, the carbonation process of the poor crys-
bonation process of portlandite is a function of
crystallite size. In particular NL binder with the
smaller crystallite size of portlandite (Fig. 3) 4000
slow slaked NHL
shows an amount of carbonation; transformation
of portlandite to carbonates, of about 60%. In 3000 rapid slaked NHL
L in ( C o u n ts )
contrary in the NL binder with the larger crys- Portlandite (001)

tallite size of portlandite (Fig. 4) only 25% of 2000
portlandite was transformed to CaCO3 at the cium Aluminum Iron Hydroxide

Larnite
same curing period. The carbonation of the poor 1000
crystalline portlandite is 2.4 times faster than of

well crystalline portlandite. The speed of car- 0
16 17 18 19 20 21
bonation is inversely proportional to the crystal- 2-Theta - Scale
lite size. Figure 8: Portlandite’s peak shape of rapid and slow

slaked NHL binder.
1e4
6000
Sqrt (Counts)
5000
4000
Calcium Aluminum Oxide Hydrate NHL slow slaked 30days
3000
Katoite
2000
1000 Brownmillerite NHL slow slaked 0days
100
0
9 10 20 30 40 50
2-Theta - Scale
Figure 9: X-ray difractogram of the slow slaked NHL in 0 and 30 days.
2e4
1e4
Sqrt (Counts)
6000
5000 NHL rapid slaked 30days
4000 Katoite
3000
2000
Calcium Aluminum Oxide Hydrate NHL rapid slaked 0days
1000 Brownmillerite
100
0
9 10 20 30 40 50
2-Theta - Scale
Figure 10: X-ray diffractogram of the rapid slaked NHL in 0 and 30 days.
talline portlandite is 2.4 times faster than of well Social Fund (75%) and National Resources
crystalline portlandite, which means that the rate (25%).
of carbonation is inversely proportional to the
crystallite size.
Finally, the hydration of aluminates was also REFERENCES
observed. Calcium aluminum oxide as well as Castellote, M., C. Andrade, X. Turrillas, J. Campo and
katoite were the hydrated products of this proc- G.J. Cuello, 2008. Cement and Concrete Research 38.
ess. pp. 1365-1373.
Cheary, R.W. and A.A. Coelho, 1992. A fundamental pa-
Work is still in progress in order to collect rameters approach of X-ray line-profile fitting, J.
data with respect to the mechanical properties of Appl. Crystallogr. 25 pp. 109-121.
the above binders in order to combine them with Ciliberto, E., S. Ioppolo and F. Manuella, 2008. Ettringite
mechanisms of hydration and carbonation that and thaumasite: A chemical route for their removal
take place in the binders over time as well as from cementious artefacts. Journal of Cultural Heri-
tage, Vol. 9, Issue 1, pp.30-37.
with physical parameters such as crystallite size. Collepardi, M., 1999. Thaumasite formation and deterio-
ration in historic buildings, Cement and Concrete
Composites, Vol. 21, Issue 2. pp. 147-154
ACKNOWLEDGMENTS El-Turki, A., M.A. Carter, M.A. Wilson, R.J. Ball and
This research is supported by the General Secre- G.C. Allen, 2009. Construction and building materials
23. pp. 1423-1428.
tariat of Research and Technology of Greece Füllmann, T., H. Pöllmann, G. Walenta, M. Gimenez, C.
through the PENED 2003 program, contract Lauzon, S. Hagopian-Babikian, T. Dalrymple and P.
number 03ED604, co-funded by the European Noon, 2001. Analytical methods, Int. Cem. pp.41-43.
Gualtieri, A.F., A. Viani and C. Montanari, 2006. Quanti-

tative phase analysis of hydraulic limes using the
Rietveld method, Cem. Concr. Res. 36 pp. 401-406.
Lea, F.M. 4th ed., 2004. Lea’s Chemistry of cement and
Concret: Elservier Science & Technology Books.
Mertens, G., P. Madau, D. Durinck, B. Blanpain and J.
Elsen, 2007. Quantitative mineralogical analysis of
hydraulic limes by X-ray diffraction. Cement and
Concrete Research 37. pp.1524-1530.
Puertas, F., M.T. Blanco-Varela, A. Palomo, J.J. Ortega-
Calvo, X. Ariño and C. Saiz-Jimenez, 1994. Decay of
Roman and repair mortars in mosaics from Italica,
Spain. Science of The Total Environment, Vol. 153,
Issues 1-2, pp. 123-131.
New European initiatives on raw materials stimulating mineral resources

sustainability in Greece
N. Arvanitidis
Dr. Economic Geologist, IGME, Regional Division of central Macedonia, Thessaloniki, Greece
ABSTRACT supply is a matter of resources and the knowl-

edge ability to exploit them with respect to sus-
The minerals resources sector is the source of a
tainability. The evidence of recent years on
significant proportion of materials on which the
European metallogenetic belts proves that major
society depends on. It supports regional com-
new ore deposits may still be found. Europe
munities, creates employment, provides facili-
should become world’s key asset in promoting
ties and enhances services, including health,
sustainable use of mineral resources. Both pri-
education and welfare, through its contribution
mary and secondary resources, in terms of re-
to local, regional and national economies. How-
use of by-products and mine wastes should be
ever, the growth of industrial economies, like
exploited. Through in average over 50 years of
for instance the Chinese and the Indian, has led
existence the Greek Institute of Geology and
to a tremendous upward spiral of mineral con-
Mineral Exploration acquired huge amounts of
sumption, in this case accompanied by a shift of
data on Greece’s geology and mineral resources.
emphasis to base metals and industrial minerals
In this respect, there is great potential for future
for steel manufacturing and building. The de-
discoveries of large scale and deep-seated de-
mand is becoming so great that even low and
posits and sustainable minerals supply from
deep-seated mineral concentrations can rea-
Greek source.
sonably considered ore deposits. This means the
European mineral industry will probably have to
be reshaped in terms of exploration and exploi-
tation practises. The main target is to increase
the commodity base and range of European
mineral resources. This has been frequently in-
dicated and documented lately through in a
number of European studies and market analy-
ses (e.g. Analysis of the competitiveness of the
EU Non - Energy Extractive Industry), technol-
ogy platforms (e.g. European Technology Plat-
form for Sustainable Mineral Resources),
strategies (e.g. Towards Thematic Strategy on
the Sustainable Use of Natural Resources), poli-
cies and initiatives (e.g. Raw materials Initia-
tive, GMES, INSPIRE), and RTD projects (e.g.
BioMine, ProMine) related to relevant FP calls.
Europe shows a growing and accelerating con-
sumption of mineral commodities. At the mo-
ment supply to meet these demands is or is not
adequate cannot be stated with any precision for
Application of bentonite modified with L-Histidine for the adsorption of

toxic elements in mine wastewaters
E.N. Bakatula, H. Tutu and E.M. Cukrowksa

Environmental Analytical Chemistry, School of Chemistry, University of the Witwatersrand, South
Africa
I.M. Weiersbye
Restoration & Conservation Biology Research Group, School of Animal, Plant & Environmental
Sciences, University of the Witwatersrand, South Africa
L. Mihaly-Cozmuta and A. Mihaly-Cozmuta

Chemistry-Biology Department of North University of Baia Mare, Romania
ABSTRACT several methods have been used and proposed

world-wide for the removal of pollutants from
This study proposes the use of bentonite modi-
wastewaters leaching from such facilities. Some
fied with L-histidine for the capture of toxic
examples include: ion exchange and adsorption
elements in mine wastewater. The surface area
onto charcoal, activated carbon, zeolites, clays,
of the natural bentonite and the bentonite-
plant materials, membrane filtration. The chal-
histidine were found to be 73.8 and 61.2 m2/g,
lenge of achieving high adsorption efficiencies
respectively. The elemental analysis showed an
still remains as well as the development of
increased amount of 0.258% of carbon and
cheaper materials for adsorption. The sorption
0.066% of nitrogen for the bentonite-histidine.
of dissolved substances by a biomaterial is an
Batch kinetic and batch isotherm studies were
attractive technology that has gained importance
carried out to evaluate the effect of initial con-
due to its advantages over conventional separa-
centration, pH, contact time and temperature on
tion techniques. These advantages are the reus-
the adsorption. The bentonite-histidine showed
ability of biomaterial, low operating cost, im-
higher adsorption capacity (up to 99%) than the
proved selectivity for specific metals of interest,
natural bentonite for the metal ions studied. It
removal of heavy metal from effluent irrespec-
was found that a contact time of 30 minutes was
tive of toxicity, short operation time and no pro-
necessary for the adsorption to reach the equi-
duction of secondary compounds which might
librium. The maximum adsorption value was
be toxic (Spinti et al., 1995; Srinath et al.,
obtained at pH 5 for Ni, Zn, Co, Cu and U on
2002).
the natural bentonite. Except Ni, the metals
L-Histidine is one of the compounds released
studied presented high adsorption capacity on
by the micro-organisms and interacts with the
the functionalized bentonite at pH 2. The ex-
clay mainly through the imidazole group. Stud-
perimental biosorption equilibrium data were in
ies revealed that the adsorption capacity of ben-
good agreement with those calculated by both
tonite could be enhanced (at about ten times
Langmuir and Freundlich models.
higher) by surface modification using organo-
functional silane coupling agents (de Mello,
1. INTRODUCTION 2008; Loredo, 2007). The effects of contact
time, concentration of metal, temperature and
Wastes generated by the mining industry con- pH on the natural bentonite as well as bentonite-
tain high concentrations of metals and metal- histidine were studied.
loids which can be mobile, resulting in leaching
into groundwater and surface water. The man-
agement of such wastes poses a challenge and
2. MATERIALS AND METHODS ficient (Komatsu et al., 1990) were calculated.

2.1 Natural bentonite
The natural bentonite used as adsorbent in the 3. RESULTS AND DISCUSSION
present study, grey in colour, was supplied by The mineralogical composition of the natural
Sigma-Aldrich (Pty.) Ltd. South Africa. The ad- bentonite determined by XRF (in percent) was:
sorbent was dried for 24 hours at 60o C before SiO2 52.26, Al2O3 17.25, Fe2O3 0.53, FeO 4.25,
the experiments. MnO 0.07, MgO 3.67, CaO 2.05, K2O 1.38,
Na2O 0.35, TiO2 0.4, P2O5 0.09 and loss on ig-
2.2 Bentonite-histidine
nition (LOI) 17.72%. BET, CEC and CHNS re-
For the synthesis of the bentonite-histidine ad- sults are given in Table1.
sorbent, 10 g of natural bentonite interacted at Elemental analysis of bentonite-histidine
60oC for 48 h at pH 8 with 100 ml of solution showed an increased amount of carbon
containing L-Histidine hydrochloride which was (0.258%) nitrogen (0.066%).
the same amount as the CEC value of the ben-
tonite. Then, the synthesized sorbent was 3.1 Effect of contact time
washed several times with de-ionized water to The time-dependant behaviour of metals adsorp-
remove the possible unreacted L- Histidine. The tion on natural bentonite as well as on ben-
sorbent was resuspended in de-ionized water, tonite-histidine is shown in Figure 1. It was ob-
stirred for about 1 hr at room temperature and served that the rapid adsorption process oc-
separated by filtration. The suspension was curred within 30 minutes and then followed by a
dried in the oven for 24 hrs at 60oC. slow adsorption process continued for 12 hours.
The adsorption capacity of U and Hg increased
2.3 Reagents and standards
drastically in the modified bentonite. The sorp-
All chemicals used were analytical reagent tion of metals ion on natural bentonite was pro-
grade. Standards used for quantification were duced following the sequence: Fe2+ > Cu2+ >
high purity having trace metals below the detec- Zn2+, Ni2+ > Co2+ > U2+ > Hg2+ whilst in ben-
tion limits. A concentration of 1000 mg L-1 was tonite-histidine, the sequence was: Fe2+ > Hg2+>
prepared for each trace metal (Cu, Zn, Ni, U, Cu2+ > U2+ > Co2+, Zn2+ > Ni2+.
Co, Hg and Fe). Appropriate aliquots were
taken from these standards for subsequent dilu- 3.2 Effect of temperature
tion to the desired concentration level. The adsorption isotherms of Cu, Zn Ni, Co, Fe,
Adsorption of metals (Cu, Zn, Ni, U, Co, Hg U and Hg sorption on the natural and the modi-
and Fe) was studied on natural bentonite as well fied bentonite were obtained at four different
as on bentonite-histidine. An amount of the sor- temperatures while keeping all other parame-
bent was placed into 250 ml bottles and 100 ml ters, i.e. metal concentration, shaking time and
of metal solution of known concentration were pH, constants. The results are given in Table 2.
added. The contents were shaken for a given pe- In general, the adsorption capacity on bentonite-
riod of time and temperature at 170 rpm. The histidine was high (> 99%) compared to that on
solutions were then filtered using Whatman fil- natural bentonite at 18, 22, 26 and 30 oC.
ter paper 0.45 µm; the concentration of metals Thermodynamic parameters, i.e. free energy
was measured by ICP-OES. The pH of the solu- ∆Ho, entropy change ∆Go, activation energy Ea,
tion was adjusted by addition of dilute HNO3 or adsorption capacity Ac and reaction rate for the
NaOH solution. The percentage adsorption sorption of Cu, Zn Ni, Co, Fe, U, Hg on ben-
(Milonjic et al., 1992) and the distribution coef- tonite as well as bentonite-histidine were calcu-
Table 1: Properties of bentonite and bentonite-histidine.

Surface Area Average Pore volume CEC C H N
Sample
m2/g cm2/g meq/100g % % %
Bentonite 73.8 0.103 64.42 0.555 1.748 0.678
Bentonite-
61.2 0.0087 42.54 0.813 1.490 0.744
Histidime
tion is favored at higher temperature. It was ob-

served that the rate of adsorption for the metal
ions increases with the increasing of the tem-
perature with Fe metal ion being the least af-
fected while mercury was the most affected. The
values of activation energy, enthalpy change for
the reaction and free energy for the adsorption
of heavy metals on bentonite - histidine at each
temperature was greater than those obtained at
the same temperature for the natural bentonite.
This could be due to the complex formation be-
tween the metal ions and the amine groups at-
tached on the bentonite structure. The adsorp-
tion data for 18, 22, 26 and 30o C were found to
be in accordance with the Freundlich and the
Langmuir models.
3.3 Effect of pH
The effect of pH on adsorption of heavy metals
on the bentonite and the modified bentonite was
studied in the pH range 2 to 12. The results are
shown in Figure 2.
The results showed a constant adsorption rate
of Fe on natural bentonite for the all range of
pH. The adsorption rate of U, Ni, Zn, Co and Cu
on natural bentonite was found low at low pH
and the rate increased with the pH. A maximum
adsorption was reached at pH 5 for U, Ni and
Cu and at pH 7 for Zn and Co. The adsorption
Figure 1: Plots of adsorption % versus contact time for
single component system at pH 3.00 for the natural ben- rate of Hg decreased with the pH till pH 5 then
tonite and the bentonite-Histidine. before increasing.
The low adsorption rate observed at low pH
lated. The positive values of ∆Ho showed that in natural bentonite can be attributed to the fact
the sorption of metals on bentonite is an endo- that low pH implies high H+ concentration,
thermic process. Then the enthalpy increases which turns to competes with the metal ions for
with the temperature. The values of ∆Go were the adsorption sites.
negative for the metals except for Hg. These in- Also, hydrolysis of the aluminum oxide sites
dicate the spontaneity of the adsorption process. at very low pH on the natural bentonite results
∆Go becomes more negative with increasing in the loose of the crystallographic structure of
temperature which further shows that the sorp- natural bentonite and consequently its adsorbent
Table 2: Adsorption capacity (mg/g) of metals (single component system) for the natural bentonite and the bentonite-
histidine at different temperatures.
Adsorption capacity (mg/g)
Natural bentonite Bentonite-histidime
Metal 18oC 22oC 26oC 30oC Metal 18oC 22oC 26oC 30oC
Cu 4.717 4.791 4.821 4.887 Cu 4.988 4.989 4.991 4.995
Ni 4.028 4.039 4.155 4.18 Ni 4.986 4.985 4.968 4.964
Zn 4.037 4.377 4.417 4.464 Zn 4.998 4.997 4.994 4.994
Co 3.675 3.842 3.926 3.979 Co 4.997 4.998 4.992 4.992
Fe 4.997 4.985 4.98 4.978 Fe 4.998 4.998 4.998 4.998
Hg 1.623 1.462 1.205 2.018 Hg 4.982 4.993 4.99 4.99
U 3.108 2.616 3.66 3.302 U 4.991 4.992 4.994 4.951
competition for the active site of the bentonite-

histidine, and also, there is destruction of the
bentonite-histidine structure hence resulting to
low porosity and low adsorption rate. At very
high pH values, the precipitation of the metal
hydroxide out ways the adsorption of the metals
on the bentonite - histidine and hence the de-
struction of the bentonite-histidine structure
turns to decrease the rate of adsorption on the
bentonite - histidine.
4. CONCLUSION
The study showed that the adsorption capacity
of bentonite can be enhanced by modification of
the surface using a microbial component as his-
tidine. This hybrid organic-inorganic material
can be an alternative low cost material in the
Figure 2: Plots of the adsorption % versus pH on natural

bentonite of heavy metals in single component solutions.
capacity. The maximum adsorption capacity ob-

tained at the pH range 5 to 7 for all the metals
studied is due to the contribution of adsorption
as well as precipitation process since most of
the metals hydrolyze at this pH range so they
will precipitate in their hydroxide form, as metal
hydroxide. It was observed a maximum and
constant adsorption rate at the pH 7.
The results of adsorption of metals on modi-
fied bentonite (Fig. 3) showed high adsorption
rate at low pH for Co, Zn, Fe, Hg, U. An in-
crease with pH was observed for Ni whilst the
adsorption rate was constant at the all range of
pH for Cu. The formation of complex between
the metal ion and the bentonite-histidine de-
pends on the amine groups present on the ad-
sorbant. The rate of adsorption increases with
the increase of the amine group on the structure.
A decrease of adsorption rate was observed at
very low pH due to the protonation of the amine Figure 3: Plots of the adsorption % versus pH on ben-
group to form ammonium ion. tonite-histidine of heavy metals in single component solu-
At very low pH, there is H+ and metal ions tions.
treatment of mine wastewater and possibly for

the recovery of precious metals for such efflu-
ent. The adsorption rate of bentonite-histidine is
greater than that of natural bentonite at high
concentration of metals due to the structure of
bentonite-histidine.
Chemical modification by amine coupling
agents may be a useful tool for the preparation
of new adsorbents with high adsorption capaci-
ties and selectivity towards metal ions retention
for waste waters from acid mine drainage sys-
tems. Bentonite can also be used as support me-
dia for the culture of other fungi and microalgae
with higher metal-binding capacities.
ACKNOWLEDGEMENTS
The authors would like to thank the National
Research Foundation and the Carnegie Founda-
tion for the financial support.
REFERENCES
Boyd, S.A., 1988. Pentachlorophenol sorption by organo-
clays. Clays and Clay Minerals, 36(2): p. 125-30.
de Mello, A., V. Sampaio, T. Ciminelli and V. van der
Luiz, 2008. Smectite orgnofunctionalized with thiol
groups for adsorption of heavy metal ions Applied
Clay Science 42, 410-414.
Freundlich, H., 1907. Physikalishe Chemie 57, (A) 385-
470.
Komatsu, Y., Y. Fujiki, M. Watanabe and T. Sasaki,
1990. Removal of Pb (II) from aqueous/acidic solu-
tions by using bentonite. Solvent Extraction and Ion
Exchange, 1532-2262, 8, p.173-185.
Langmiur, I. 1918. J. Am. Chem. Soc. 40, 1361.
Milonjic, S.K., M.R. Boskovic and T.S. Ceranic, 1992.
Adsorption of Uranium (VI) and zirconium (VI) from
Acid Solutions on Silica Gel. Separation Science and
Technology, 27, p. 1643-1653.
Spinti, M., H. Zhuang and E.M. Trujillo, 1995. Evaluation
of immobilized biomass beads for removing heavy
metals from wastewaters. Water Environ. Res., 67 (6):
943-952.
Srinath, T., T. Verma, P.W. Ramteke and S.K. Garg,
2002. Chromium (IV) biosorption and bioaccumula-
tion by chromate resistant bacteria. Chemosphere, 48
(4): 427-435.
Study of mechanisms of manganese removal from leachates
G. Bazdanis, D. Zaharaki and K. Komnitsas

ABSTRACT Mn2+ + 2H2O → MnO2 + 4H+ + 2e-,

In the present experimental study the manga-
log K = -41,89 (1)
nese removal mechanisms from leachates were
2+ + -
investigated in batch and column experiments. Mn + 2H2O → MnOOH + 3H + e ,
Batch leaching tests were carried out in shake
flasks using zero-valent iron (ZVI), activated log K = -25,27 (2)
carbon (AC) and natural zeolite as reactive me- The rate of manganese oxidation may in-
dia to obtain kinetic data. Upflow column leach- crease when ions such as aluminum and iron are
ing tests were carried out to simulate the opera- present in solution as well as through microbial
tion of permeable reactive barriers (PRBs) using activity (Vail and Riley, 2000).
coarse and fine limestone as well as goat ma- PRBs containing zero-valent iron, natural
nure as reactive media. Experimental data as zeolite and activated carbon have been used to
well as mineralogical studies show that the re- remove contaminants such as copper (Komnit-
moval mechanisms depend on the reactive ma- sas et al., 2007), hexavalent chromium (Jeen
terial used in each case; the main mechanisms et al., 2008), nickel (Argun, 2008), mercury
are adsorption, ion exchange and/or co- (Chojnacki et al., 2004), selenium (Sasaki et al.,
precipitation. 2008), cadmium (Di Natale et al., 2008), arsenic
(Chutia et al., 2009; Jeon et al., 2009) and other
elements from leachates (Bartzas et al., 2006a,
1. INTRODUCTION b; Gitari et al., 2008).
Manganese is a common contaminant in mine In the present experimental study manganese
waters, industrial wastewaters and effluents. removal from leachates was investigated in
Due to its high mobility and different oxidation batch and column experiments using several re-
state over a wide pH range it is often difficult to active media.
be removed. Although manganese is not consid-
ered as ecotoxic as other metals, it poses several
environmental problems (Bamforth et al., 2006). 2. MATERIALS AND METHODS
Previous studies that were carried out to investi- 2.1 Materials
gate Mn removal from acid mine drainage,
leachates and other effluents used reactive mate- The reactive materials used in batch leaching
rials such as limestone, red mud, fly ash and or- tests were zero valent iron (Gotthart Maier,
ganic matter to increase pH and precipi- Germany), activated carbon (Donau Chemie,
tate/adsorb soluble Mn species (Komnitsas Austria) and natural zeolite (S&B Industrial
et al., 2004, 2006). Minerals S.A., Greece). In the upflow column
Manganese oxidation and subsequent re- leaching tests coarse and fine limestone (Fino-
moval from solutions may be described by the beton Mining S.A, Greece; 2 fractions 1-2 mm
following reactions (Lindsay, 1979): and 2-5.6 mm were tested), as well as goat ma-
nure (containing N, P2O5, K2O, Ca, Mg, Fe, Mn,
Cu, B, Mo and Zn) were used as reactive media.

Chemical analyses and main properties of the
reactive materials are seen in Table 1.
2.2 Methodology
2.2.1 Batch leaching tests
Laboratory batch leaching tests were carried out
in 600 mL conical flasks under magnetic stir-
ring; the reactive media dosage (for activated
carbon, zeolite and ZVI) studied was 0.5, 1.0, 5,
10 and 20 g/L. The initial concentration of Mn Figure 1: Experimental setup for column tests.
in solution was 100 mg/L; this is considered
quite high when compared with actual Mn con- nure was mixed with silica sand in order to
centrations present in leachates and industrial avoid clogging. Two 2.5 cm thick silica sand
effluents. The solution was prepared by dissolv- layers were placed at the base and the top of the
ing the required quantity of manganese sulfate columns in order to ensure optimum flow distri-
monohydrate (MnSO4·H2O) in distilled water. bution and prevent transfer of fine particles. Up-
Adsorption was studied by using Langmuir and flow was applied in order to eliminate channel-
Freundlich equations. ing, maintain optimum contact between solution
and reactive media and increase retention time.
2.2.2 Column experiments The concentration of manganese in the feed
Column experiments were carried out in Plexi- solution was maintained as in the previous se-
glas columns (45 cm length and 5 cm internal ries at 100 mg/L. Feed solution was pumped in
diameter) filled with each reactive media. the columns from 20 L plastic vessels by using
1000 g of coarse or fine limestone and 650 g of PROMINENT GAMMA/4 peristaltic pumps.
goat manure were used in each series; goat ma- The flow rate used was 80 mL/h so that one
pore volume (350 mL) was fed every 4.4 hours.
Table 1: Chemical analysis and main properties of reac- The experimental setup is seen in Figure 1.
tive materials. Aqueous samples were taken from the efflu-
Zero
Acti-
ent of each column and analyzed immediately
valent for pH, oxidation-reduction potential using
(% w/w) vated (% w/w) Zeolite
iron HANNA pH 211 combined pH/Eh meter and
carbon
(Fe0) electric conductivity using a HANNA EC 215
Fe 92.03 Fe2O3 1.23 EC meter. Solutions were filtered using 0.45 μm
C 3.31 79.74 Na2O 1.00 aero disk filters, acidified with concentrated hy-
Si 2.04 SiO2 70.44 drochloric acid to pH<2 and stored at 4 oC until
Mn 0.63 Al2O3 13.11 analysis. Manganese concentration was meas-
Al 0.16 CaO 4.43 ured using a Perkin Elmer Analyst 100 flame
S 0.09 0.36 MgO 0.93 atomic absorption spectrophotometer.
Ni 0.06 TiO2 0.25
Cr 0.05 K2O 1.87
P 0.04 H2O 6.78 3. RESULTS AND DISCUSSION
H2O 0.4 5.45
3.1 Batch leaching tests
Paste pH 5.1-5.3 10.23 7.87
Apparent The evolution of pH vs. time for each reactive
0.85-
density 2.7-2.9 0.45 media used is as follows:
1.1
(g/cm3) Activated Carbon: Depending on the dosage
Grain size
0.2-1.0 1.5-4.0 0.2-1.0
used pH increased rapidly the first 6 hours from
(mm) the initial value of 5.5 to values ranging be-
Specific tween 5.8 and 7.7. Then pH values were stabi-
surface area 0.0482 900.67 15.58 lized and varied between 6.5 and 7.9. Dosages
(m2/g) of 10 and 20 g/L AC maintained solution pH to
8.5
340 μS/cm.
Figures 3-5 show the evolution of Mn con-
8
centration in solution vs. time for each reactive
7.5
media used.
7 Mn is effectively removed from solution
pH
6.5 when the dosage of AC is higher than 10 g/L.

6 After 7 days Mn removal is 82 % and 99 % for
5.5
Activated Carbon AC dosages 10 and 20 g/L, respectively. Lower
removal rates not exceeding 40 % are seen
5
0 50 100 150
when AC dosage is ≤5 g/L.
Time (h)
0.5 1 5 10 20
100
Figure 2: pH vs time (initial Mn concentration 100 mg/L, 80

activated carbon dosage 0.5-20 g/L).
Mn (mg/L)
60
values close to or higher than 7.5 almost from Activated Carbon
the start of the test (Fig. 2); such values favor 40
Mn adsorption. 20
Zeolite: pH increased gradually and at the
end of the test (after 7 days) values varied be- 0
0 50 100 150 200
tween 6.4 and 7.4 for various zeolite dosages Time (h)
used. 0,5 1 5 10 20
Fe0: When zero valent iron was used as reac-
Figure 3: Manganese concentration vs time (initial Mn
tive medium iron corrosion took place and
concentration 100 mg/L, activated carbon dosage 0.5-20
therefore ferrous ions were transferred into solu- g/L).
tion (reactions 3 and 4); these ions may partici-
pate in several reactions that result in the forma- 100
tion of precipitates (Komnitsas et al., 2007). pH
in the system, apart from an initial increase did 80
not follow a certain trend and fluctuated be-
Mn (mg/L)
tween 5.7 and 6.7. 60
Fe0 + 2H2O → Fe2+ + H2 + 2OH- (3) 40
Fe0 + 2H+ → Fe2+ + H2

Zeolite
(4) 20
0 50 100 150 200
Eh values (data not shown) are in accordance
Time (h)
with pH values in all batch systems. Eh of the
0.5 1 5 10 20
solution varies between -75 and 35 mV when
0.5 and 20 g/L AC dosage is used respectively. Figure 4: Manganese concentration vs time (initial Mn
Regarding zeolite, Eh values decreased gradu- concentration 100 mg/L, zeolite dosage 0.5-20 g/L).
ally and at the end of the run varied between -50 100
and 15 mV, slightly lower than in the case of ZVI
AC. In the zero valent iron system Eh values 80
vary between -10 and 56 mV making thus the

Mn (mg/L)
60
progress of the following reaction 5, which ac-
celerates iron corrosion, extremely slow. 40
2Fe3+ + Fe0 → 3Fe2+ (5) 20
0
Conductivity values (data not shown) are in 0 50 100 150 200
line with pH and Eh measurements. It is men- Time (h)
tioned that conductivity in the AC and zeolite 0.5 1 5 10 20
solutions varies between 330 and 430 μS/cm, Figure 5: Manganese concentration vs time (initial Mn
while in the ZVI solution is just above concentration 100 mg/L, ZVI dosage 0.5-20 g/L).
At a dosage of 20 mg/L the maximum re- 1
moval rate recorded over the first 24 h was

3.47 mg/g, while for the next 4 days it dropped 1
1
to 1.42 mg/g. The main removal mechanism in 2 2 3 3
1
3
1
this case, due to favorable solution pH, is ad-

sorption on active sites of activated carbon.
Zeolite is less efficient in removing manga-
5 10 20 30 40 50 60 70
2? scale
nese due to its lower surface area compared to Figure 6: XRD analysis of reactive iron particles (1:
AC and the lower solution pH. Removal per- maghemite, 2: lepidocrocite, 3: Fe-hydroxides).
centages at the end of the test do not exceed
60 % and 40 % when dosages of 20 and 10 g/L probable under the specific solution conditions:
are used respectively. When lower dosages are 4Fe0 + 3O2 +2H2O → 4δ-FeOOH (9)
used manganese removal does not exceed 22 %
after 7 days. Fe2+ + CO 32− /OH- → Ferrous precipitates (10)
The main mechanism for manganese removal
in this case is ion exchange. This is confirmed 4Fe3+ + 3O2+12e− → 2Fe2O3 (11)
by the presence of K, Ca and Mg ions in the so- Fe3+ + 3H2O → Fe(OH)3 + 3H+ (12)
lution, up to 10 mg/L.
When a dosage of 20 g/L is used, ZVI re- It is believed that part of Mn removed from
moves almost 70 % of Mn after 7 days; at dos- solution is co-precipitated with and/or adsorbed
ages ≤5 g/L removal percentage does not exceed on these iron oxides and hydroxides.
15 %. It is underlined that limited Mn removal Figures 7 and 8 show the Langmuir iso-
is anticipated in case solution pH values do not therms in cases AC and zeolite are used as reac-
exceed 6.5-6.7 over the entire leaching period. tive materials. Based on the R2 values it is seen
The concentration of total dissolved iron that adsorption can be best described by the
(data not shown) increased at the beginning of
the tests and reached 135 mg/L after 6 hours 0.25 y = 0.4599x + 0.0583
when a 20 g/L dosage was used. After 24 hours 0.20
R2 = 0.9039
it dropped rapidly and varied until the end of the
tests between 15 and 40 mg/L. ZVI removes 0.15
M/Y
also 67 % of sulfate ions (data not shown) pre- 0.10

sent in solution (initial SO4 concentration
180 mg/L). The reduction of sulfates by ZVI 0.05
can be described by the following reactions; it is 0.00

known that the presence of sulfates enhances 0 0.1 0.2 0.3 0.4
ZVI corrosion (Bartzas et al., 2006b). 1/Ce
Activated carbon Linear (Activated carbon)
4Fe0 + SO42- + 9H+ → 4Fe2+ + HS- + 4H2O (6)
Figure 7: Langmuir isotherm (initial Mn concentration
SO42- + 4H2 → HS- + 3H2O + OH- (7) 100 mg/L, AC dosage 0.5-20 g/L).
4Fe2+ + HS- + 7OH- → FeS(am) + 0.40

0.35
y = 21.558x - 0.1843
R2 = 0.9434
3Fe(OH)2(am) + H2O (8) 0.30
0.25
XRD analysis (Siemens D500 diffractometer
M/Y
0.20
using a Cu tube and a scanning range from 3° to 0.15
70° 2-theta, with a step 0.03° and 4 sec/step 0.10
measuring time) of the iron particles at the end 0.05

0.00
of the tests revealed the presence of maghemite 0.01 0.01 0.02 0.02 0.03 0.03
(γ-Fe2O3), lepidocrocite (γ-FeOOH) and Fe- 1/Ce
hydroxides (δ-FeOOH) (Fig. 6).
ZL Linear (ZL)
These phases may form through the follow-
ing reactions; reactions (9) and (10) are more Figure 8: Langmuir isotherm (initial Mn concentration
100 mg/L, zeolite dosage 0.5-20 g/L).
Langmuir model. moval dropped dramatically in all systems over

a period of 45 days. No system managed to raise
3.2 Column tests
pH to values higher than 7; it is mentioned that
Figure 9 presents the evolution of manganese the behavior of goat manure which as organic
concentration in solution vs. time for each reac- material is not effective in low pHs was antici-
tive medium used, namely coarse/fine limestone pated.
and goat manure.
It is seen from this figure that the efficiency
of coarse and fine limestone increases slowly 4. CONCLUSIONS
with time; after 30 days fine limestone removes Manganese removal from leachates was studied
70 % of manganese while when coarse lime- in batch and column tests. Among the three re-
stone is used the removal percentage exceeds active media tested in batch tests (activated car-
40 % after 47 days. This is due to the slow pro- bon, zeolite and zero valent iron), activated car-
vision of alkalinity in the system and the respec- bon was the most efficient; manganese removal
tive slow increase of pH from 5.5 to 7.2 after 46 reached almost 99 % after 5 days, when the ini-
days. tial Mn concentration was 100 mg/L. At a dos-
On the other hand when goat manure is used, age of 20 g/L the maximum removal rate re-
Mn removal is complete the first 10 days, corded over the first 24 h was 3.47 mg/g, while
whereas the efficiency of the system drops for the next 4 days it dropped to 1.42 mg/g.
gradually and after 47 days reaches 45 %. Goat Main removal mechanisms are adsorption for
manure acts as electron donor in the dissimila- activated carbon, ion exchange for zeolite and
tory reduction of sulphate to sulphide, generat- co-precipitation and/or adsorption on Fe oxy-
ing thus alkalinity (reaction 13) and promoting hydroxides formed in the system containing
metal precipitation as sulphide (reaction 14) zero valent iron.
(Pagnanelli et al., 2009): Column tests showed that goat manure ex-
hibits for a short period (10 days) remarkable
SO 2-4 + 2CH2O + 2H+ → H2S + 2H2CO3 (13) manganese removal efficiency due to adsorption
Me2+ + H2S → MeS↓ + 2H+ (14) of manganese on organic material/precipitation
as sulphide; longer retention time would have
It is believed that the efficiency of goat ma- led to a better efficiency of this system. On the
nure would have been much better over a longer other hand limestone (either coarse or fine) ex-
period if a bigger quantity was used in the col- hibited due to slow solubilisation and therefore
umn or two columns were used in series to in- limited pH buffering effect, low removal rates
crease retention time. over the period tested. When the pH of the feed
As mentioned earlier the initial pH of the solution was decreased to 3, manganese removal
feed solution in column tests was 5.5. When the dropped dramatically in all systems.
pH of the feed solution was decreased to 3 by
the addition of sulfuric acid, manganese re-
ACKNOWLEDGEMENTS
The financial support of the Research Commit-
100
90 tee of the Technical University of Crete (TUC)
80 in the frame of the basic research programme
70
2007-2008 is greatly acknowledged.
Mn (mg/L)
60
50
40
30
20 REFERENCES
10
0 Argun, M.E., 2008. Use of clinoptilolite for the removal
0 10 20 30 40 50 of nickel ions from water: kinetics and thermodynam-
Time (days) ics, Journal of Hazardous Materials 150, 587-595.
Coarce limestone Fine limestone Goat manure
Bamforth, S.M., D.A.C. Manning, I. Singleton, P.L.
Younger and K.L. Johnson, 2006. Manganese removal
Figure 9: Manganese concentration vs time (initial Mn from mine waters-investigating the occurrence and
concentration 100 mg/L, flow rate 80 mL/h). importance of manganese carbonates, Applied Geo-
chemistry 21(8), 1274-1287.

Bartzas, G., K. Komnitsas and I. Paspaliaris, 2006a. Inor-
ganic contaminant fate assessment in zero valent iron
treatment walls, Environmental Forensics 7(3), 207-
217.
Bartzas, G., K. Komnitsas and I. Paspaliaris, 2006b.
Laboratory evaluation of Fe0 barriers to treat acidic
leachates, Minerals Engineering 19 (5), 505-514.
Chojnacki, A., K. Chojnacka, J. Hoffmann H. Gorecki,
2004. The application of natural zeolites for mercury
removal from laboratory tests to industrial scale, Min-
erals Engineering 17, 933-937.
Chutia, P., S. Kato, T. Kojima and S. Satokawa, 2009.
Arsenic adsorption from aqueous solution on synthetic
zeolites, Journal of Hazardous Materials 162, 440–
447.
Di Natale, F., M. Di Natale, R. Greco, A. Lancia, C. Lau-
dante and D. Musmarra, 2008. Groundwater protec-
tion from cadmium contamination by permeable reac-
tive barriers, Journal of Hazardous Materials 160 (2-
3), 428-434
Gitari, M.W., L.F. Petrik, O. Etchebers, D.L. Key, E.
Iwuoha and C. Okujeni, 2008. Passive neutralisation
of acid mine drainage by fly ash and its derivatives: A
column leaching study, Fuel 87, 1637-1650.
Jeen, S.W., D. Blowes and R. Gillham, 2008. Perform-
ance evaluation of granular iron for removing hexava-
lent chromium under different geochemical condi-
tions, Journal of Contaminant Hydrology 95, 76-91.
Jeon, C.S., K. Baek, J.K., Park, Oh Young-Ki and Sang-
Do Lee, 2009. Adsorption characteristics of As (V) on
iron-coated zeolite, Journal of Hazardous Materials
163, 804-808.
Komnitsas, K., G. Bartzas and I. Paspaliaris, 2004. Effi-
ciency of limestone and red mud barriers: laboratory
column studies, Mining Engineering 17(2), 183-194.
Komnitsas, K., G. Bartzas and I. Paspaliaris, 2006. Mod-
eling of reaction front progress in fly ash permeable
reactive barriers, Environmental Forensics 7(3), 219-
231.
Komnitsas, K., G. Bartzas, K. Fytas and I. Paspaliaris,
2007. Long-term efficiency and kinetic evaluation of
ZVI barriers during clean-up of copper containing so-
lutions, Minerals Engineering 20(17), 1200-1209.
Lindsay, W.L., 1979. Chemical equilibria in soils, Wiley,
New York.
Pagnanelli, F., C.C. Viggi, S. Mainelli and L. Toro, 2009.
Assessment of solid reactive mixtures for the devel-
opment of biological permeable reactive barriers,
Journal of Hazardous Materials, doi:10.1016/
j.jhazmat.2009.05.081.
Sasaki, K., D.W. Blowes, C.J. Ptacek amd W.D. Gould,
2008. Immobilization of Se (VI) in mine drainage by
permeable reactive barriers: column performance, Ap-
plied Geochemistry 23, 1012-1022.
Vail, W.J. and R.K. Riley, 2000. The Pyrolusite Process:
A bioremediation process for the abatement of acid
mine drainage, Greenlands 30 (4), 40-46.
Acquisition and assessment of non-intended outputs in the lignite industry
T. Bielig and H.Z. Kuyumcu

Berlin Institute of Technology, Germany
ABSTRACT is introduced and its application on the lignite

industry. From the technological point of view
Land, equipment, material and energy are used
the overall production process in the lignite in-
for the activities of the lignite industry. During
dustry can be divided into mining, power gen-
operation various material and energy flows are
eration in lignite-fired power plants and - future-
released. These material and energy flows are
oriented - in the capture and storage of carbon
nearly always without value for the raw material
dioxide (CO2). The refinement of lignite is not
supply chain and the electric power supply.
considered in the investigations.
They are not production targets. Instead, they
The lignite industry plays an important role
have negative effects on the environment and
in both countries, Greece and Germany. Table 1
the ecology as well as on the economy. There-
gives an overview about the resources, the re-
fore they are referred to as “non-intended”. A
serves and the use of lignite in Greece and Ger-
high degree of transparency of the material and
many. Domestic lignite is the most important
energy flows and especially the knowledge of
indigenous fuel of Greece. Germany and Greece
the quantities and qualities of the non-intended
are the largest lignite producers in the EU. Ex-
outputs are the starting point for technical and
traction of lignite has a very long tradition in
economical measures. The methodology for ac-
both countries (Diercks et al., 2008; Kavouridis,
quisition and assessment of non-intended out-
2008).
puts in the lignite industry bases on a system
Today in Greece and Germany, lignite is ex-
analysis. All relevant material and energy flows
clusively mined in opencast mines and mainly
are assigned to individual processes. Causal re-
used for power generation. Since transport of
lationships, possible interactions, quantities and
lignite is expensive relative to its extraction cost
qualities of the non-intended outputs are exam-
ined as functions of the processes and their pa- Table 1: Lignite and Energy Data, 2007.
rameters. Finally an environmental, technical Information Greece Germany
and economical evaluation follows. Resources [Mt] 6,800 77,200
Reserves [Mt] 3,300 40,800
1. INTRODUCTION Domestic Output [Mt] 65.8 180.4
3,770- 7,800-
At the Chair for Mechanical Process Engineer- Calorific Value [kJ/kg]
9,630 11,500
ing and Solids Processing of the Berlin Institute Ash Content [%] 15.1-19.0 1.5-20.0
of Technology a methodology for the acquisi- Water Content [%] 41.0-57.9 40.0-60.0
tion and assessment of the material and energy Sulphur Content [%] 0.5-1.0 0.15-2.8
flows especially in the mining industry, includ- Primary Energy Con-
11.9 55.0
ing mining and raw material processing, was sumption [Mtce]
developed and already tested in several mines Power Supply [TWh] 31.1 156.0
(Bielig et al., 2005). Power Plant Capacity
5,288 20,385
In the following the developed methodology [MW]
a
Mtce: million tons of coal equivalent
and heating value, lignite can be used economi- Analysis of system structure
cally only within relatively short distances from and Modelling
the opencast mines. Therefore all the lignite-
fired power plants are located in close proximity
to the mines. The lignite opencast mines and Classification of
nearby power plants are technical production the material and energy flows
systems with geographical, geological and tech-
Implementation
nological boundary conditions. As almost all
Alteration
geological and most of the geographical restric- Data acquisition and
tions are not variable, the feasibility and the management
economic success of the mine operation under
given market conditions can only be optimised
by technological parameters. All the technical, Material and energy balancing
economic and organisational measures which Calculation of the non-intended outputs
influence the production of a certain quantity
and a defined product quality from a very well
explored deposit and for electricity production Assessment and deduction
as well as for CO2 capture and storage are to be of appropriate measures
understood as technology. Figure 1: Flowchart of the procedure of acquisition and
Even though especially mines appear diffi- assessment of the non-intended outputs.
cult to access to “benchmarking”, the acquisi- including CO2 capture, transport and storage in
tion and assessment of all the material and en- an adequate deposit is analysed. Therefore, a hi-
ergy flows create the potential for comparisons erarchical system definition specific to the re-
of mining operation as well as for power gen- quirements in the lignite industry is necessary
eration from lignite and for comparisons of that allows to systematically break-down the
similar sites with the option to extend the ex- overall system into its relevant processes and
amination globally to different areas. Improve- associated material and energy flows. For mod-
ment, modification and optimisation can be de- elling the overall system as well as the individ-
rived from this knowledge. By means of the ual processes Umberto is used. Umberto is a
methodology the lignite mining, the power gen- software tool to model, calculate and visualise
eration as well as the CO2 capture and storage material and energy flow systems as material
can be optimised to contribute to a sustainable and energy flow networks.
development in the lignite industry. The material and energy flows or compo-
nents of them are classified as non-intended
outputs according to the definition not to be the
2. METHODOLOGICAL APPROACH
production target. Then operational data (input
For the acquisition and assessment of the mate- and output material and energy flows, cost data,
rial and energy flows not only in the mining in- etc.) has to be gathered for the identified ele-
dustry, but also in the lignite industry, an inte- ments using a set of appropriate questionnaires.
grated approach is necessary. Therefore the ob- For further analysis, the acquired data is then
jective of the work presented in this paper has stored in the database of Umberto.
been the development of a methodology suited Further processing of the data primarily
to the comprehensive technical and economic comprises the data reconciliation of the meas-
analysis of all material and energy flows in the ured and estimated values to fit the set of mate-
lignite industry and its application to selected rial and energy balances. For this purpose proc-
operations. ess models are used. Based on the corrected bal-
Figure 1 shows the overall flowchart for the ances for all material and energy components
procedure of acquisition and assessment of the within the overall system the flows containing
non-intended outputs. The procedure is dis- the non-intended outputs are identified and used
cussed in detail within the next paragraphs. to calculate several performance indicators as a
In the first step the structure of the mine op- decision support in taking appropriate optimisa-
eration and the nearby power generation plant tion measures.
2.1 System definition and system model extraction”, 01.04 “Haulage/transport”, 01.05
“Supporting”, 01.06 “Material supply”, 01.07
In accordance with the system-analytic scope,
“Repair/maintenance”, 01.08 “Personnel trans-
material and energy flows were only considered
port”, 01.09 “Dumping/loading”, 01.10 “Mine
in connection with the active working phase of a
water management”, 01.11 “Mine ventilation”,
lignite mine and a lignite-fired power plant.
01.12 “Storage/disposal”, 01.13 “Immission
Processes and outputs up to the start of regular
control”, 01.14 “Recultivation during mining”,
production and after the close down were not
01.15 “Recultivation after mining” and 01.16
considered. Providing electricity or in the long
“Overall operation of the mine”. The subsys-
term vision the capture and storage of CO2 in an
tems of first order in the partial system 03 “Po-
adequate deposit are other limits of the system
wer generation” are: 03.01 “Processing of fuel”,
for the investigation.
03.02 “Combustion”, 03.03 “Gasification”, 03.
Due to the differences in the balancing pro-
04 “Steam generation”, 03.05 “Energy conver-
cedure, in the system-analytic approach the
sion in turbines”, 03.06 “Condensation”, 03.07
overall system “lignite industry” is divided into
“Flue gas cleaning”, 03.08 “CO2 capture”, 03.09
the partial systems 01 “Mining”, (02 “Process-
“Process water management”, 03.10 “Providing
ing”), 03 “Power generation” and 04 “CO2 se-
electricity”, 03.11 “Loading/unloading”, 03.12
questration”. For all the partial systems not only
“Disposal”, 03.13 “Repair/maintenance” and 03.
subsystems of first and second order, but also
14 “Overall operation of the power plant”. The
the associated elements are defined. By doing so
subsystems of first order in the partial system 04
a process-oriented proceeding is pursued. The
“CO2 sequestration” are: 04.01 “CO2 compres-
main idea is to finally ascertain the material and
sion”, 04.02 “CO2 transport”, 04.03 “CO2 stor-
energy flows in such a way that the reasons,
age”, 04.04 “Repair/maintenance” and 04.05
origins, quantities and qualities of the non-
“Overall operation of the CO2 sequestration”.
intended outputs can be assigned to the individ-
For further subdivision, all the subsystems of
ual processes. The hierarchical structure of the
first order are divided into subsystems of second
system is represented in Figure 2.
order. To the subsystems of second order the
For the better handling of the hierarchical
elements are assigned.
system structure a four-digit coding scheme is
The content of the system is determined by
introduced. The partial systems, the subsystems
the individual elements and their interrelations.
of first and second order as well as the elements
Elements describe processes, which can not fur-
are unambiguously assigned by two-digit num-
ther usefully be subdivided. One or more pieces
bers. So it is possible to match the recorded in-
of equipment can belong to an individual ele-
puts and outputs to the appropriate elements.
ment, for which the material and energy flows
First of all the main, ancillary and auxiliary
are balanced. The flows represent the relations
processes have to be identified and defined as
that link the elements and/or subsystems. An
subsystems of first order. In the partial system
example is shown in Figure 3.
01 “Mining” the following processes are defi-
In accordance with the principle of Petri nets,
ned as subsystems of first order: 01.01 “Deve-
there are three basic types of components to cre-
lopment”, 01.02 “Preparatory”, 01.03 “Mining/
ate material and energy flow networks in Um-
berto. Material and energy transformation proc-
01 Partial system 1
esses are represented in transitions (T). For each
01.01 Subsystem of 1st order
Inputs Element Outputs
01.01.01 Subsystem of 2nd order
Muck Ore
Dust
Energy Motor Device Exhaust gas
01.01.01.01 01.01.01.n 01.01.00.n 01.00.00.n Lubricant Oil loss
Element Element Element Element and material Abrasion
01.03.03.01 „Loading Waste heat
Fresh air with mobile equipment“ Noise
Figure 2: Numbering convention for the hierarchical sys- Figure 3: Inputs and outputs of the element 01.03.03.01
tem definition. “Loading with mobile equipment“.
transition the relation between the input and Material

Material flow
flow
output flows of material and energy has to be Solid

Solid phase
phase
specified by coefficients, functions or a subnet Overburden/waste
Overburden/waste rock
rock Component 11
as process model. Places (P) are the components Run-of-mine
Run-of-mine Component
Component m
in the network that can be stores of materials Back-filling material
material
and energy, input or output places connecting Tailings
Tailings
the material flow network with the environment, Dust

Dust
or places where the output stream of one proc- Solid

Solid material
material
ess becomes the input of another. Arrows link Abrasion

Abrasion
Waste
places and transitions, and thus create the actual
network structure (ifu and ifeu, 2005). liquid
liquid phase
phase
Mine water
water Groundwater
2.2 Material and energy flows
Meteoric
Meteoric water
water
Besides the process-oriented description of the Process
Process water
water
system and its elements, also the relevant mate- Percolating

Percolating water
Liquid
Liquid materials Component 11
rial characteristics are registered in a structured
Component
Component m
way for the evaluation of the material and en- Gas
Gas phase
ergy flows. For that purpose, a flow vector defi- Mine air
air Component 11
nition is introduced. In addition to the mass or Pressure
Pressure air
air Component
Component m
volume flow of the material or energy compo- Mine gas
nents and the linkage of the elements among Exhaust
Exhaust gases
themselves or with the environment they can Figure 4: Structure of the material flow vector.
contain the further material composition (chemi-
cal or mineral), particle-size distribution, state search programmes can be set up. The important
variables (in particular pressure and tempera- objective is to ensure a most comprehensive
ture). data acquisition. Besides general information
The vector is subdivided into the three phases about the mine and the power plant, like mode
solid, liquid and gaseous. To characterise the of operation, location, infrastructure of the re-
solids, liquids and gases, material systems of gion, climate, mining and generation company,
complex composition were defined. Such mate- employees and regulations for safety and envi-
rial systems can be made up of other material ronmental protection, all important data about
systems or of several components. Concentra- processes and machinery, as well as material
tions, contents etc. can be calculated if neces- and energy flows with their specifications, can
sary. Figure 4 shows the structure of the mate- be recorded systematically. Additionally cost
rial flow vector with the most important mate- data are inevitable for the economic evaluation
rial systems in mine operation. of the material and energy flows. For that reason
As energy flows thermal energy has to be a process related acquisition of cost data is pro-
considered primarily. The heat transport is vided.
mainly coupled to air and water (and steam) 2.4 Determination of the non-intended outputs
flows. The most important heat sources in a
mine, which affect the temperature of the mine The non-intended outputs are determined based
water and in particular the temperature of the on the material and energy flows of the system.
mine air underground, are the geothermal heat In the operation often not all the necessary in-
underground, the self compression of the air on formation about the flows is available or the
its way through the mine and the waste heat of measurements are incorrect. For this reason, in-
drives and equipment. In an opencast mine the consistent data have to be corrected or com-
climatic conditions have to be considered. pleted. Umberto offers the possibility to make
up a correct balance and to compensate or calcu-
2.3 Data acquisition late the missing or incorrect values by means of
For the data acquisition, questionnaires, data a gradual algorithm for calculation. If there are
sheets and forms are prepared. With the help of not enough sampling points for a reliable solu-
the questionnaires, data sheets and forms, re- tion of the material balances, process models are
used to complete the system model. rated costs (C) and revenues (R).
There are three different cases to be distin-
2.5 Assessment
guished:
The constraints in mine operation and the han-
- Case 1: Δ2 > 0 The non-intended output
dling of material and energy flows in the lignite
yields a profit.
industry result from technical necessities of the
- Case 2: Δ2 = 0 A profitable use is not directly
production, safety-relevant requirements and
possible. However, it can be quantified, e.g.
regulations for the protection of health and envi-
for preventing follow-up costs.
ronment. For the observance of all regulations
- Case 3: Δ2 < 0 There is no economic use of
and rules, certain outputs have to be declared in
the non-intended output.
detail. Their effects on health and environment
have to be explained. Nevertheless this does not The determination of the costs for avoidance,
necessarily lead to the monitoring and quantifi- treatment and/or disposal of outputs and as-
cation of all non-intended outputs during opera- signment of the costs to the individual processes
tion. Therefore, each procedure for evaluation of (as far as possible to the elements) as well as the
non-intended outputs in the lignite industry has balancing of the costs in accordance with the
to take into account the legal and safety rules as system structure results in the calculation of per-
well as the social-cultural aspects relevant for formance indicators and possibly in the deduc-
the location of the respective mine. tion of appropriate measures according to prior-
Pre-determined parameters of the system can ity.
be assigned to the geology of the site including
the material content of the deposit. Based on the
3. APPLICATION
mine design, the technical and organisational
parameters within mainly economic limits have For the application of the introduced methodol-
to be taken into account for the assessment of ogy to the lignite industry an opencast mine and
non-intended outputs. Processes can only be lignite power plant in Germany is initially cho-
quantitatively described, optimised and com- sen. The raw lignite is extracted by continuous
pared within certain limits. operation of bucket wheel and bucket chain ex-
Figure 5 shows the flowchart of a possible cavators. It is transported and dumped by con-
procedure for an economic evaluation of non- veyor bridges as well as belt conveyors and
intended outputs in the lignite industry, incorpo- spreaders. The raw lignite from the opencast
mines is transported to the nearby power plants
Source of non-intended by conveyor belt and/or train.
Optimisation
outputs: process The system analysis of a lignite opencast
mine with a nearby power plant results in a
Non-intnded output: complex model which can be arranged into mul-
material or energy tiple subsystems and elements. Based on the
available data, the balancing of material and en-
ergy flows for the defined subsystems is per-
Measure: Measure: Measure:
Avoidance Acquisition/disposal Collection/utilisation
formed. Figure 6 shows the material and energy
flow network of the investigated opencast mine.
T1 is specified by a subnet. T2 to T5 are speci-
Costs C2 Costs C1 Revenues R fied by functions, e.g. based on the stoichiome-
try of chemical reactions or physical laws.
There are still no lignite-fired power plants with
Δ 1 = R – C1 CO2 carbon and capture in work. Nevertheless
R > C1 , Δ1 > 0
the subsystems are included in the material and
energy network for rough calculations of the en-
Δ2 = Δ1 – C2
Δ2 > 0, ΣR >ΣC
ergy demand and the effects on the overall sys-
Δ2 < 0, ΣR < ΣC tem.
Figure 5: Algorithm for the evaluation of non-intended The following non-intended outputs are iden-
outputs based on costs and benefits tified as significant: mine water (especially cer-
P6: emissions from P7: mine water P8: emissions from P10: waste heat
opencast mining haulage by train P9: lubricants: air, water,
calcium carbonate
P5: 03.10
P2: lubricants: P3: 01.09.03 "Dumping" T2: 01.04.05.08 "Providing electricity"
fuel and air
"Discontinuous haulage
P14 by train" P4
P13: building
materials
T1: 01 "Mining" T3:03 "Power generation" P15: 01.12.03 "Disposal
in the openc ast mine"
P1: overburden, raw lignite (ash and gypsum)
P11: Electric energy
from the deposit,
groundwater P16
P12: electric energy P20
T5:04.01 "CO2 compression" P19: emissions from

and 04.02 "CO2 transport" P17 power generation
P18: 04.03 "CO2 storage" T4: 03.08 "CO2 capture"

P22: emissions P21: lubricant:
from CO2 c ompression monoethanolamine
Figure 6: Material and energy flow network of a lignite opencast mine in Germany
tain components), scrap metal, rubble, waste

heat, dust, methane, ash, gypsum, CO2 and ex-
haust emissions. REFERENCES
Bielig, T., H.Z. Kuyumcu and W. Hennig, 2005. Devel-
opment of a method for the acquisition and assessment
4. CONCLUSION of non-intended outputs in the mining industry. World
of Mining - Surface and Underground 57 (5): 327 -
The results for the examined mines show that 335.
the methodology for the process based acquisi- Diercks, T., G. Hulik, E.C. Günther, H.-W. Schiffer and
tion and assessment of non-intended outputs in C. Dude (eds.), 2008. Coal Industry across Europe
the mining industry is suitable to the compre- 2008. Brussels: EURACOAL.
hensive ascertainment of non-intended outputs Institut für Umweltinformatik Hamburg GmbH (ifu); In-
stitut für Energie- und Umweltforschung Heidelberg
and their evaluation. The methodology is gener- GmbH (ifeu) (eds.), 2005. Umberto Software für das
ally applicable to different branches of the min- betriebliche Stoffstrommanagement. Hamburg, Hei-
ing and power generation industry. If the neces- delberg.
sary data is available for several systems, indi- Kavouridis, K., 2008. Lignite mining and use in Greece -
vidual processes of the lignite industry or the Energy supply and environment. World of Mining -
Surface and Underground 60 (1): 6 - 15.
complete system from mining to power genera-
tion to CO2 capture and storage can be com-
pared to each other or used as a benchmark.
Appropriate actions can be derived for the
choice of techniques and for operational prac-
tice, e.g. to reduce the costs, control emissions
or save resources (water, land etc.). Further-
more, it is possible to find criteria related to a
specific site in order to optimise the operating
and follow-up costs of mining or to increase the
acceptance of a mine operation or of the lignite
industry in general.
Energy and nutrient recovery from sewage sludge on islands
K. Bognar and F. Behrendt

Berlin Institute of Technology, Germany
ABSTRACT 1. INTRODUCTION
For reasons of soil conservation and consumer Wastewater sludge is an organic resource, and
protection the distribution of sewage sludge on can play a role in resource recovery during the
agricultural land is being considered critical. A wastewater treatment process bringing it closer
decreasing degree of public acceptance and to carbon-neutral operations. This can be
tighter limit values set by the European Union achieved by seeking closed-loop recycling man-
for the utilization of sewage sludge as fertilizer agement avoiding harmful emissions and waste.
contribute to a changing situation of its disposal. At the same time additional value from sludge
As a consequence the volume of thermally can be gained by reuse of energy and recovery
treated sewage sludge has significantly in- of nutrients and metals.
creased in recent years.
Especially on islands waste disposal is a seri-
ous issue. Small-scale thermal-treatment units 2. SEWAGE SLUDGE
for waste, in this case for sewage sludge and or- 2.1 Definition
ganic residues, can contribute to a closed-loop
recycling management. Sewage is wastewater including excrements
For the processing of sewage sludge a huge with a water content of more than 95 % and a
amount of energy is needed, mainly for the dry- share of about 3.5 % dry substances (DS) con-
ing of the sludge. Due to energy recovery ther- taining roughly equal parts of organic residues
mal treatment plants can provide sufficient en- and minerals. Organic residues can consist of
ergy to run the plant in a self-sustained fashion. municipal sewage sludge and bio-waste, agri-
In combination with renewable energy applica- cultural waste like slurry from animal hus-
tions even additional energy can be supplied in bandry or food residues, e.g. from the industry.
some cases. The minerals are mainly ashes.
Furthermore raw materials like nutrients and The amount of sewage sludge produced by a
metals can be recovered out of sewage sludge. community is measured in population equiva-
Mainly phosphorus is being recycled in some lents (PE). PE refers to the amount of biological
processes. oxygen demand (BOD) for biodegradation
In the present paper sewage sludge treatment equaling the average oxygen demand of the
by pyrolysis, gasification, and incineration are wastewater produced by one person. One unit
discussed and some commercialized technolo- equals 54 g BOD per day (BOD, 1997).
gies are presented. In addition some aspects of 2.2 Regulations concerning sewage treatment
phosphorus recovery from sewage sludge are
discussed and its practicability on islands is Currently organic residues are biologically sta-
evaluated. bilized through anaerobic digestion until they
become suitable for application to land as an or-
ganic soil conditioner under existing Provincial
Guidelines. While an agricultural application
still continues to provide the most economical processes even dried. Providing sufficient en-
application solution for sludge, health concerns ergy for this process is one of the biggest chal-
have fuelled recent debates on sludge disposal. lenges during the disinfection process. Dried
The genotoxicity of Cu and Zn in raw and an- sludge has the advantage, that its weight and
aerobically digested slurry and high contents of volume is reduced. Its heating value is increased
heavy metals make sewage sludge a risky fertil- to 11 MJ/kg in the process making it similar to
izer. As a result, thermal treatment becomes in- the value of domestic waste or lignite (Bogner
creasingly the application of choice for sludge et al., 2008).
use including materials recycling and energy re- There are different methods of drying sew-
covery. age. Following procedure principles are distin-
The European Union tries to stipulate the guished:
discharge of waste substances by regulating - Drying by contact (plate dryer, film dryer).
their composition and content. The main incen- - Drying by convection (e.g. belt dryer, drum
tives for the European sludge treatment have dryer).
been the Urban Waste Water Treatment Direc- - Drying by radiation (e.g. Solar Drying); only
tive (91/271/EEC), the Landfill Directive applicable for dewatered sludge with 20%
(Council Directive 1999/31/EC) and the Sewage dry substance (DS) or more.
Sludge Directive (86/278/EEC).
Mainly the Landfill Directive has forced that 2.3 Treatment Technologies with energy recov-
material with an organic content greater than
ery
5 % cannot be disposed on a landfill area and
has initiated a shift away from landfills to more Sludge can be treated by different anaerobic
enhanced sludge treatment methods. Before bio- processes generating methane as a by-product.
solids are used on land, a number of factors With gas cleaning, biogas can be formed for
need to be considered. These include the con- power generation and for the drying of sludge.
tents of the biosolids, soil conditions, water Pyrolysis can also be viewed as an anaerobic
sources around the site, and type of crops to be process but reaches the chemical decomposition
grown (Padmanabhan, 2005). at much higher temperatures of around 500-
An environmental friendly alternative to land 900 °C. Pyrolysis is mostly applied to organic
fills is thermal sewage treatment. In Germany materials. Alternative technologies are the gasi-
mainly co-incineration capacities in coal, do- fication and incineration of the sludge.
mestic waste and cement burning power plants An overview of existing technologies of
are in use. About 55 % of thermal treated sew- small-scale mono incineration plants is shown
age sludge in Germany is combusted in co-
incineration, about 45 % in mono-incineration Table 1: Thermal treatment processes.
plants. Within the mono-incineration especially Gasi-
decentralized systems for some 10,000 PE are pyroly- Incin-
Process fica-
gaining in importance, as they minimize in- sis eration
tion
vestment and transportation costs and generate Pyrobustor x
local benefits (Hanßen, 2003). For islands with HD-Pawa-Therm x
populations quite often below 10,000 the incin- Thermo-catalysis x
eration of sludge is economically not feasible Pyromex x
yet. For such islands biological stabilization G&A/ Unitechnik x x
through anaerobic digestion will probably stay Ohltherm PYRO-3 x x
most frequently applied treatment. For middle Kopf process x
size islands though small-scale mono- PyroFluid x
incineration plants are taken into account within AWINA x
this paper. BioCon x
SWSF-process x
2.2 Drying of the sludge Kalogeo x
In order to combust or gasify sewage sludge it Sludge-2-Energy x
needs to be at least dewatered and for some EcoDry x
in Table 1. Four of these technologies are pre- tion. In the first step sludge is partially dried at a
sented in more detail, the Pyrobustor, the Kopf relatively temperature of about 55 °C using the
process, the Kalogeo, and the Sludge-2-Energy waste heat of the thermal utilization stage and
process. optional solar energy. On average, one ton of
water can be vaporized with an electrical input
Pyrobustor
of approximately 60 kWh. Thermal utilization is
Since the end of 2004, a pyrobustor plant for the achieved by pyrolysis in a fluidized-bed gasi-
thermal utilization of dried sewage sludge is in fier. The ashes produced are inert and can there-
operation in South Tyrol. The energy recovered fore be used as land fill or stored separately in
is used for the upstream sludge drying process. order to use the phosphorus in the ash for the
This drying plant has a process capacity of production of fertilizers for farming purposes.
16,000 t/year of sludge with 20 - 25 % of DS, The final waste product contains inorganic ma-
corresponding to approx. 400,000 PE from 8 terials like ashes and rubble representing about
wastewater treatment plants. Dried sewage 14 % of the original mass. Heat produced can be
sludge as well as other dry residues or solids used into the district heating networks, con-
with a high calorific value such as shredder verted by an organic Rankine cycle (ORC) into
light-weight fractions, meat and bone meal, electricity and fed into the electricity grid, or
residues from biological-mechanical effluent can be used for the sludge drying to run the
treatment or waste from renewable primary plant as self-sustaining as possible. Kalogeo is a
products are first pyrolyzed and then oxidized. small-scale application for about 2000 t of DS
From a storage dry pellets are dosed into the Py- per annum. The Bad Vöslau facility in Austria is
robustor. In the first chamber the material is py- the first project utilizing this technology and op-
rolyzed, in the directly following second cham- erates since 2003 (Stockmayer, 2007).
ber the pyrolysis coke is oxidized to inert ash.
Sludge-2-Energy
The flue gases formed in this secondary com-
bustion process pass in counter current of the Sludge-2-Energy is a project developed by the
raw material and deliver the process heat re- German ATZ research institute and has been
quired for the pyrolysis. Beyond this process- commercialized by the German company Huber
internal use, the energy contained in the dry AG. The process is based on a pebble heater
substance can be used as a substitute for expen- technology producing electrical and thermal en-
sive primary energy (Eisenmann, 2008). ergy at the same time. The patented pebble
heater is combined with a micro gas-turbine al-
Kopf process
lowing the recovery of electrical energy from
This sludge gasification process converts dried the heat of hot flue gas without need for installa-
sludge to gas, which can be used to produce en- tion of a water-vapour cycle with an electric ef-
ergy and a phosphorus-rich granulate, which can ficiency of around 30 %. Through heat ex-
be recycled. In a stationary fluidized bed com- changers the released heat is fed into the dryer.
bustible gas is produced from dried sewage The concept allows a combined treatment of
sludge by gasification at temperatures up to sewage sludge and other organic residues. The
900°C. In a raw-gas quench, the gas is cooled planned demonstration plant is designed for an
annual sludge volume of about 1,000 to 2,200 t
down and subsequently reprocessed by filtering
DS with an electric power output of 100 kW
and drying. In a CHP-type heat-power system,
(Bogner et al., 2008).
the gas is used to produce electricity and heat.
After a retention time of 30 minutes, the sewage 2.4 Costs of treatment technologies
sludge is reduced to an inert granule (Stowa, Data from Bogner et al. (2008) show total costs
2006). of small scale systems for either pyrolysis or
combustion of dried or dewatered or sludges of
Kalogeo approx. 120-190 EUR/t of DS. In case of large
Kalogeo has been developed by Tecon Engi- scale mono-incinerators the cost for dried
neering GmbH in Germany. The sewage sludge sludge is approx. 60 EUR/t while those for de-
is first pre-dried, followed by thermal utiliza- watered raw material go up to 280 EUR/t of DS.
In this case small, decentralized systems do of- Table 2: Recovery options of phosphorus.
fer a cost-advantage over large systems. This Stage of Mass P-con- Bonding
P-pot.*
looks promising for applications on islands. extraction flow rate centr. form
Final ef- 200 max.
< 5 mg/l dissolved
fluent l/(PE*d) 55 %
3. RECOVERY OF NUTRIENTS Sludge liq- 1-10 20-100 max.
dissolved
3.1 Development of Demand uor l/(PE*d) mg/l 50 %
Digested 0.15 chem.
Especially phosphorus (P) and nitrogen (N) are 10 g/kg ~90 %
sludge l/(PE*d) bonded
needed as fertilizer in agriculture worldwide. 0.03 kg/ chem.
There are some options of P-recycling out of Ash 64 g/kg ~90 %
(PE*d) bonded
sewage sludge. The processes are gaining in- *P-recovery potential referring to feed-in
creasing interest since the prices of P increased
dramatically in the last years like shown in Fig- maintaining the pellet bed in a fluidized state. In
ure 1. order to crystallize the target component on the
The recovery of phosphorus can occur at dif- pellet bed, a driving force is created by a reagent
ferent stages (Table 2) of the sewage treatment: dosage and pH-adjustment. By selecting the ap-
propriate process conditions, co-crystallization
- out of the diluted phase, which includes the of impurities is minimized and high-purity crys-
final effluent and the sludge liquor, tals are obtained. The pellets grow and move
- out of the sludge, or towards the reactor bottom. At regular intervals,
- out of the ash. a quantity of the largest fluidized pellets is dis-
charged from the reactor and fresh seed material
3.2 P-recovery out of the diluted phase is added. After atmospheric drying, readily han-
There are some processes that make use of the dled and virtually water-free pellets are obtained
crystallization and precipitation of phosphorous (Giesen, 2009).
compound from the diluted sewage. Since 1997
3.3 P recovery out of digested sludge
a number of applicable solutions have been de-
veloped, some of the latest technologies are: In Germany the recovery of phosphorus out of
digested sludge is a common technology by now
- P-Roc process
and is provided by different producers. Some of
- Ostara-MAP-crystallization
them are:
- PRISA-process
- DHV-Crystalactor - the CAMBI process,
- the Aqua Reci process,
One of the commercialized technologies is
- the KREPRO process, and
the DHV-Crystalactor. The Crystalactor is a
- the Seaborne process.
fluidized-bed type crystallizer, which is a so-
called pellet reactor, partially filled with a suit- The Seaborne process for example is a com-
able seed material like sand or minerals. The mercialized German processes. A typical input
wastewater is pumped in an upward direction consists of 25 t fermentable waste/residual
products resulting in an output stream of 23 t
water, 160 kg fertilizer, 0.6 kg heavy metals as
well as 7600 kWh of heat.
A pilot plant has been set up in Gifhorn,
Germany, producing 1.3 t of a high-quality fer-
tilizer magnesium-ammonium-phosphate
(MAP) per day (Seaborne EPM AG, 2008).
3.4 P-recovery out of ash
A number of processes have been developed for
recovering P out of the ash of sewage sludge
Figure 1: Relative P-price [%] over the years (Dockhorn, over the last years:
2008).
- BioCon process (Denmark, Sweden) and sulfur are added in the third step and ash
- RüPa and PASCH process pellets are pressed.
- Mephrec process A pilot plant based on the SUSAN project
- SEPHOS process has been built and is operational. Approx. 7 t of
- BAM-/ASH DEC-process P-fertilizer is currently produced, a larger plant
is in its planning phase (SUSAN, 2008).
Due to the enormous reduction of the mass
The resistance to corrosion of the pilot plant
flow P-recovery out of the ash of sewage sludge
is uncertain until now. The costs of the process
is a promising recovery method. The ash con-
are estimated 100-130 EUR/t of ash (Gethke,
tains 1.5-10 % P (median 6.2 %) which is not
2006). The expected gain is 18 EUR/t of ash
plant available right away. Heavy metals and
and the plant’s amortisation 5-6 years.
trace elements exceed limit values of the fertil-
P-recovery out of the ash is economically
izer regulations making a direct use of the ash as
a fertilizer impossible. only reasonable for at least 10,000 t ash per year
or more.
Recovery of P is only reasonable after mono-
incineration processes since the P-content after
co-incineration processes would not be high 4. CONCLUSIONS
enough. One example of actually implemented
technologies is the ASH DEC concept sup- Within ongoing research work on sustainable
ported by the EU within the SUSAN project energy supplies for islands in the present paper
(Sustainable and safe re-use of municipal sew- options for the gain of energy combined with
age sludge for nutrient recovery). The principle the retrieval of nutrients from sewage sludge
is shown in Figure 2. have been collected and evaluated.
In step 1 the ash is homogenized and pellet- Mid-size islands with at least 10,000 inhabi-
ized by adding chloride-containing supplements, tants have a high potential to use thermally-
for example CaCl2 and MgCl2 (50-100 g chlo- treated sewage sludge as fertilizer without
rine/kg ash). In step 2 the ash is heated in a health risks while generating sufficient energy
thermoreactor up to 30 minutes at 1000 °C to cover the internal energy demand of the plant.
where 99 % of the heavy metals and supple- In combination with renewable energy sources
ments react and evaporate. The ash is recyclable like solar thermal, geothermal or wind energy
after this stage, but to use it as a fertilizer other some processes even allow a contribution to the
phosphates, nitrogen, potassium, magnesium energy supply of islands in an economically
fashion for some 10,000 person equivalents
(PE).
This represents an increasingly important as-
pect for islands since high costs for imported
fossil fuels can be avoided.
Due to relatively few inhabitants and high
initial investments, P-recovery is economically
not reasonable for small and mid-size islands.
REFERENCES
BOD, 1997. Glossary of Environment Statistics, Studies
in Methods, Series F, No. 67, United Nations, New
York,
http://stats.oecd.org/glossary/detail.asp?ID=2086
Bogner, R., P. Quicker, M. Mocker and M. Faulstich,
2008. Sludge-to-Energy Options with Decentralised
Sewage Reutilisation Needs, Water and Wastewater
International 23(1):34-36.
Dockhorn, Th., 2008. Aspekte des Phosphor-Recyclings,
Figure 2: Schematic presentation of steps involved in the in: Braunschweiger Nährstofftage - Ressourcen scho-
ASH DEC process for recovering P out of ashes (SUSAN, nender Einsatz von Phosphor in der Landwirtschaft,
2008). 10.-11.11.2008.
www.jki.bund.de/cln_044/nn_1269000/DE/veroeff/bs
__naehrstofftage/bs__naehrstofft__node.html__nnn=tr
ue, May 11, 2009.
Eisenmann, 2008. Waste Disposal - Recycling - Thermal
Treatment - Energy Utilization, Flyer,
www.eisenmann.de/include/Download/Abfall-
Entsorgung_en.pdf, May 11, 2008
Gethke, K., 2006. Ergebnisse des UFOPLAN-Projektes
„Rückgewinnung eines schadstofffreien, mineral-
ischen Kombinationsdüngers „Magnesiumammoni-
umphosphat - MAP" aus Abwasser und Klär-
schlamm". Fachgespräch „Rückgewinnung von Phos-
phor aus Abwasser und Klärschlamm", Umweltbunde-
samt Dessau, 24. April 2006.
Giesen, A., 2009. Crystallization Process Enables Envi-
ronmental Friendly Phosphate Removal at Low Costs
www.nhm.ac.uk/research-curation/research/projects/
phosphate-recovery/crystalct.htm, May 11, 2009
Hanßen, H., 2003. Thermische Klärschlammbehandlung
in Deutschland: Stand und Ausbaupotenziale. In: Hans
Huber AG (Hrsg.) Handbuch zur 1. Fachtagung “Klär-
schlamm”, Berching, 9. Sep. 2003, S. 45-67
Padmanabhan, S., 2005. Sludge management shifts to re-
use through innovative thermal solutions, Water &
Wastewater International 20(4).
Seaborne EPM AG, 2008.
www.seaborne-erl.de/englisch/5_News/news3.htm,
May 11, 2009.
Stockmayer, W., 2007. Decentralized sewage manage-
ment in Waste Management World 8(6).
Stowa, 2006. Kopf Gasification Process, Flyer,
www.stowa-selectedtechnologies.nl/Sheets/Sheets/
Kopf.gasification.process.html, May 11, 2009
SUSAN, 2008. http://www.susan.bam.de, May 11, 2009.
Carbon capture and storage in central Appalachia - feasibility, costs and

regulatory impact
S.M. Carpenter
Director, Carbon & International Business, Marshall Miller & Associates, Inc., Bluefield, VA, USA
P. Boileau
Manager, Climate Change, Canadian Standards Association, Ottawa, Ontario, Canada
ABSTRACT 2. CARBON CAPTURE & STORAGE

In an ever increasing “carbon” society; our 2.1 Overview
awareness, management and growth will be
Carbon Capture and Storage (or Sequestration)
measured by our carbon footprint (or output of
“CCS” is the term now commonly applied to
Greenhouse Gas (GHG) emissions). As power
describe the processes whereby carbon dioxide
producers fight the battle between electrical de-
(CO2) is captured during the combustion of
mand vs. carbon reduction, one of the essential
coal, compressed and converted to a supercriti-
tools will be capture and storage (e.g. sequestra-
cal fluid, and stored in deep geologic formations
tion) of carbon dioxide.
via injection wells. Regional geology may allow
Standardization of the regulatory framework
for multiple horizons to be used in a “stacked”
for Carbon Capture & Storage (CCS) will be
array to increase storage capacity in a given re-
key to any success in the “Sequestration” realm.
gion. Increasing the storage capacity in a given
Present progress and an overview of the US De-
area will allow for the reduction of transporta-
partment of Energy’s (DOE) Southeastern Car-
tion issues typically addressed via pipeline in-
bon Sequestration Partnership’s (SECARB)
frastructure (Fig. 1).
Virginia Coal Carbon Sequestration project will
CCS is not new, at least not geologically
be shared. Implementation, economic, risk & li-
speaking. The Intergovernmental Panel on Cli-
ability issues will be discussed.
mate Change has noted that CO2 has been natu-
rally trapped in geologic formations for hun-
1. INTRODUCTION dreds of millions of years, including CO2 that
has been held under the Pisgah Anticline in
Apart from the technical feasibility of Carbon Louisiana and Mississippi for more than 65 mil-
Dioxide Capture (or Sequestration) and Storage lion years. The DOE and International Energy
(CCS), there remain complex regulatory and Agency estimates that the U.S. has over 3,000
public perception issues to overcome before gigatons of CO2 sequestration potential; which
CCS will be accepted and therefore used by in- is equal to the emissions of approximately 1,000
dustries and markets. This paper reviews the coal-fired power plants for 1,000 years.
current standards, codes and regulation frame- Similarly, oil producers in Texas and other
work in the U.S. and Canada as well as the po- oil producing states have been injecting CO2 for
tential NIMBY (e.g. Not In My Back Yard) is- purposes of Enhanced Oil Recovery (EOR) for
sues and newly defined NUMBY issues (e.g. decades. This process has been and is regulated
Not UNDER My Back Yard) raised by CCS as and permitted by the EPA and delegated state
well as the potential risks associated with CCS. agencies under the Safe Drinking Water Act's
Underground Injection Control (UIC) program.
The U.S. regulatory framework is generally
speaking a two-piece puzzle: GHG Reporting
via EPA Mandatory Reporting Rule (MRR) and
Figure 1: Typical Carbon Capture and Storage.
some anticipated form of Carbon Cap & Trade the affected industrial markets.
via Waxman-Markey Bill “American Clean En- Contrary to U.S. policy, the international
ergy & Security Act of 2009”. community in general is a one piece puzzle; re-
The real confusion within the U.S. frame- lying on the Kyoto Protocol and the Interna-
work falls in the Voluntary vs. Regulated de- tional Organization for Standardization (ISO)
bate. Presently, there are no mandatory re- for guidance. Additionally, as most countries
quirements for carbon sequestration. With that are signatory to the Kyoto Protocol, a singular
being said however, most electric regulating approach is at least an option.
bodies in the U.S. are requiring all new applica-
2.2 The Need for Standardization
tions to “address” the carbon question of what
to do with carbon emissions, while not specify- Standardization is critical for success. It allows
ing how - e.g. lack of regulation. all parties equal access. It encourages interac-
The fear or reluctance in the marketplace for tion and is required for Key Nations to be-
industrial investment is the lack of clear direc- come/remain engaged. This is specifically true
tion. “Does my carbon project fall under…” relative to the Kyoto Protocol, where the U.S. is
not presently an option for participation, and
- CAR – CMM Protocol development.
therefore absent from most international trading
- CDM’s within IPCC or UNFCCC.
schemes.
- REC’s – RGGI vs. States (27 presently have
Regardless of Market (GHG, CCS, Clean
some form of Renewable Energy Credit pro-
Energy) standardization must address four key
gram).
factors: Cost, Validation, Standards & Regula-
- Carbon Credits: CCX vs. Over.
tions, and Risk & Liability.
- ANSI vs. CSA -America for ISO 14064-
Whether voluntarily adopted or required by
14066.
law, standards touch us all every day. For ex-
The myriad of choices, and more importantly ample, standards exist for indoor air quality, for
a lack of “enforcement” by virtue of the present sustainable tree harvests and now we are seeing
voluntary status, makes it very difficult to en- the development of standards that define the
courage private, commercial investment from elemental efforts needed to protect the climate –
carbon reductions (also known as carbon off- buyers and sellers both domestically and inter-
sets). nationally, which is why it is understandable
Standards work to everyone's benefit: that standards have begun to enter the arena of
climate change, both to help improve environ-
- by defining many safety requirements, they
mental management but also to support emerg-
help reduce the risk of personal injury;
ing emissions trading markets.
- by setting out performance and design crite-
ria, they bring order and know-how to the
problems, like making plugs fit sockets; 3. SECARB - VIRGINIA COAL PROJECT
- by setting recognized benchmarks, they help
open doors to new markets and facilitate The Southeast Regional Carbon Sequestration
trade around the world; Partnership’s (SECARB) Central Appalachian
- by providing a framework for best manage- Coal Seam Project is one of the seven DOE’s
ment practices, they help advance business Partnerships researching CCS. Design and im-
excellence; plementation of the test are underway. The se-
lection of the well was based on geologic con-
Recent assessments (Blum, 2006) have con- siderations for the site, preliminary reservoir
firmed that standards provide economic benefits modeling, surface access and landowner and
by influencing institutional change and the mineral owner negotiations.
structure of the economy; creating a competitive The regional study area is located within the
advantage in sectors or technologies with high Central Appalachian Basin, a northeast-to-
uncertainty; expanding markets and supply southwest-trending basin encompassing ap-
chains. proximately 10,000 square miles in southwest-
Standards influence institutional change and ern Virginia and southern West Virginia. The
the structure of economies by changing the principal area of investigation for most of the
knowledge content of companies so that they detailed geologic mapping consists of portions
can become more decentralized (i.e. globalized) of five counties located within southwestern
in their operations, supply chains and market Virginia including Buchanan, Dickenson, Rus-
opportunities. Standards allow companies to ac- sell, Tazewell and Wise Counties and four
cess suppliers who can manufacture compo- counties in West Virginia, including Fayette,
nents, vendors who can sell their products and McDowell, Raleigh and Wyoming.
labor and knowledge anywhere in the world. Phase I - Initial Characterization and Pre-
Firms that initially develop and implement liminary Feasibility Study was completed in
standards that become industry accepted have a 2005. Phase II began with an expanded Study
competitive advantage in the marketplace. This Area, Reserve Modeling and Pilot Scale CO2 In-
advantage can be leveraged to fund the evolu- jection. Presently, Phase II is completing Task
tion of that industry standard or to finance de- 10 which focuses on the characterization of the
velopment of new industry standards, thereby region to identify several sites for a potential
using standardization and new product devel- large-volume CO2 injection test to validate the
opment to generate profits. carbon sequestration and enhanced coalbed
Finally, standards can speed product devel- methane recovery potential in the Central Appa-
opment and reduce risk during the first critical lachian Basin with additional characterization of
phases of entering a new marketplace. This is secondary storage reservoirs. A large-volume
primarily the benefit that is driving the devel- (>100,000 ton CO2) test is necessary to create
opment of voluntary carbon standards. New the high level of confidence in this technology
markets, like that associated with carbon trading that is necessary prior to planning for commer-
are characterized by high development and de- cial deployment.
ployment risk and significant financial leverag- The objectives of the project include:
ing. These risks can be reduced by using stan-
dards that reduce the potential uncertainty for - Characterize potential large-volume test sites
both buyers and sellers and provide for in Central Appalachia for coal seam seques-
smoother and less costly transactions. tration and enhanced coalbed methane recov-
All of these benefits facilitate trade between ery.
proximal coal fired electric power plants for 100

years. Figure 2 details the isometric storage in-
tervals based on the present data.
4. ISSUES & CONCLUSIONS

New technology and new regulations almost
always encounter resistance. In the case of CCS,
the resistance is less about change and more
about the lack of clear direction, lack of regula-
tion and lack of international standardization.
This “lack of” effect slows progress and raises
more questions than it answers.
The basic regulatory framework is already
Figure 2: CCS Storage Capacity. under development or exists in different forms.
- Delineate and characterize saline aquifers in The adaptation of these programs to CCS ought
the region that could provide secondary car- to proceed as quickly as possible, including sci-
bon sequestration options. entifically-justified, risk-based performance
- Review and identify depleted or partially- standards and monitoring rules. Even if that oc-
depleted oil and gas fields that could support curs, however, there will still be a number of
large-volume CO2 injection tests. key legal issues to be addressed if CCS is to be
- Identify options for stacked storage reser- widely deployed, most of which are related to
voirs in Central Appalachia. post-injection storage issues.
- Design test site operations, measurement, There are some major issues/questions that
monitoring and verification program and site are presently being discussed; that once re-
closure. solved, should allow the technology and the in-
- Select several test sites in Central Appalachia ternational community to implement a more co-
for a large-volume carbon sequestration coal herent strategy which has the potential to ad-
seam project which could lead to commercial dress the required program elements.
deployment. The U.S. Supreme Court ruling in Massachu-
- Implement public outreach and education. setts v. EPA allowing the regulation of carbon
dioxide as a pollutant under the Clean Air Act
The Phase II Task 10 work is designed to has had several significant impacts. The first is
serve as the initial geologic and engineering that carbon dioxide can be regulated only as it
characterization phase for a future large-volume pertains to air issues (above ground). This does
sequestration project. The aim of the future pro- not co-exist well with the underground portion
ject is to store a large volume of CO2 which can of the sequestration requirements. Secondly, the
stress formations sufficiently to evaluate the potential regulation of carbon dioxide as a waste
structural integrity and long-term storage capac- could have its largest impact on the oil & gas
ity of coal-bearing strata in Appalachia. Rigor- transportation model of “Forced Pooling &
ous measurement, monitoring, and verification Unitization”. This is an evolution of an eminent
tools will be used to ensure the storage and in- domain concept that allows the Federal Energy
tegrity of the CO2 plume in a future large- Regulatory Commission (FERC) to mandate
volume test. The research goals of this project and regulate the storage and transportation of
are to verify the ability of coal seams to store natural gas in geological formations even if the
large volumes of CO2 and to quantify the property owner declines. The concept works be-
ECBM recovery potential. cause natural gas is an asset (a commodity) and
Recent accomplished field testing and com- as such, FERC requires the compensation of
pletion of reserve modelling indicate that the property owners for the use of the geology. Un-
basin has great potential. Based in limited field der Mass. V. EPA, if carbon dioxide is in fact a
data, it is estimated that the limited study area waste, the Forced Pooling and Unitization rules
could accommodate 100% storage of two won’t apply due to the liability of carbon versus
the asset value of natural gas. BIBLIOGRAPHY

This nuance opens the discussion of liability Power, C., 2009. EACC Clean Energy Conference, May
discussions for carbon dioxide storage fields. As 21.
an example, in WV a typical carbon dioxide Irwin, C., 2009. West Virginia Can Lead Transition to a
storage field may be as large as 50 square miles. Low-Carbon Future, January 9, 2009; The State Jour-
Assuming an average of 1,250 parcels/square nal.
mile and one surface right holder and two sub- Carpenter, S., 2009. European-American Chamber of
Commerce Conference on Clean Energy: U.S. and
surface right owners requires 187,500 liability European Perspectives - International Standardization,
and indemnification agreements for one storage May 21.
field. Since this is a non-asset (liability) issue, SECARB, Southern States Energy Board Project as
the landowner has the right to demand a moni- funded by Department of Energy (DOE) National En-
tory value to “assume” the liability for the car- ergy Technology Laboratory (NETL) under contract
DE-FC26-04NT42590.
bon dioxide being stored on their property. It
doesn’t take long for 187,500 land owners re-
quiring any sizable fee ($100,000) applied to the
multitude of storage fields required for the
amount of money required to complete the task
to reach Monopoly© Money status. This same
logic will most likely be applied to the infra-
structure issues of Right of Way (ROW) for the
required pipeline and transportation of the CO2
to the storage field.
The cost to implement this emerging tech-
nology is not cheap. Present Energy Information
Agency (EIA) estimates to implement full scale
(85 %) CCS at the power plant level would have
following impact:
- Increase in current average consumer power
cost of 10.6¢ per kWh.
- New power plant construction costs with in-
tegrated CCS systems would increase 15-
25% increase over current average $2B cost.
- Retrofit existing facilities would have a con-
struction cost of 20-45% of original construc-
tion costs.
- The parasitic load on the U.S. fleet to operate
the CCS technology is 40 GW of parasitic
demand - which is 50-800 MW plants.
With the plethora of NIMBY & NUMBY is-
sues, domestic and international standards,
managed by and regulated in both a voluntary
and mandatory framework, one thing is
clear…there is no single answer to the CCS
problem. Therefore, one should not consider the
possibility of a “silver bullet” solution. The only
option is for “silver buckshot” - all options must
be considered and evaluated to reach an interna-
tional standard which will allow this emerging
technology to be widely used, accepted and suc-
cessful.
Potential of the Greek white and semi-white marble residues under the
prism of updated information
F. Chalkiopoulou
Mineral Processing Dept, IGME
S. Karadassi
Mineralogy-Petrology Dept, IGME
D. Voujoukas
Eastern Macedonia Thrace Branch, IGME
C. Laskaridis
Econonomic Geology Dept, IGME
C. Aggelopoulos
Peloponnesus Branch, IGME
ABSTRACT enhance environmental protection and bring

profits.
The significance of unexploited marble quarry
wastes at European level has been highlighted
many times so far by institutes and researchers, 1. INTRODUCTION
as well as public authorities and quarry owners,
Ornamental Stones’ production remains a dy-
each side seeing the matter diversely. Regarding
namic economic sector in Greece, with an an-
Greece, the total national production of marble
nual production of marble blocks of more than 1
blocks was 13x106 m3 for the period 1980-2006,
corresponding to more than 65x106 m3 of rock, million tonnes (Laskaridis, 2008). On the other
or 170 million tonnes of material, that have been hand, it is a common secret that the percentage
of the wastes stemming from the exploitation of
stockpiled in national level, over the specific pe-
these blocks may be occasionally as high as
riod. Quite a number of projects have been ac-
complished by the Institute of Geology and 95% on the overall quantity of the extracted
rock. Many national and European projects have
Mineral Exploration since 1990 that are, with
dealt with the subject; however, the most sys-
one way or another, related with the exploitation
tematic approach was achieved within a project
of this waste and mainly for the production of
accomplished by the Institute of Geology &
value-added products. The most recent one,
Mineral Exploration with funds from the Third
funded by the Third Support Community
Support Community Framework during the pe-
Framework (CSF 2003-2008), Operational Pro-
riod 2003-2008.
gramme “Competitiveness” was committed
solely to the marble residues of the Eastern,
Central and Western Macedonia areas, as well 2. SCOPE OF THE PROJECT
as the Argolida, Lakonia and Arkadia Prefec-
tures in Peloponnesus. The present article pre- The overall research performed within the above
sents the key findings of the above project, con- mentioned project, was based on the grounds of
tributing thus to the promotion of the attitude the diagram of Figure 1.
that a marble quarry should not be only a marble More specifically, the specific project in-
block producing area, and that an integrating cluded the following main tasks:
mode of operation of such quarries could both - Inventory of the quarries occurring in the ar-
Standards & Spec ifications Legis lative F ramework
Building-up the Profile of

the Material Market Needs
Domestic
Type & History of the Waste
International
Existing Stockpiles
Current Production
Sludge Listing & Grouping
Marbles’ Processing: Trimmings of the Pote ntia l Prod-
ucts/Uses
Qualitative Characteristics
Assays / Mineralogical Features
Mohs Hardness
Powder: Optical Properties
Crushed Products: Physical & Laboratory Testing
Mechanical Properties
– Pilot Testing – In-
Special Properties
dustrial Application
Quantitative Characteristics Feasibility Study

Geological Mapping of Existing Selection of the Installa-
Piles tion’s Location
Current Production Estimations Finalisation of the Flow
Assessment of Quantities vs Sheet of the Production
Types & Qualities Procedure
Decision Selection of Equipment
Location: Key Parameters on End Prod- Identification of End Prod-
Road Network, Distances ucts/Uses ucts—Specifications—Prices
Proximity to Urban / Industrial Marketing Strategy
Areas Funding Opportunities
Proximity to Ports Financial Estimations
Figure 1: Integrated Management of the Marbles’ Extraction Residues: Sequence of Envisaged Actions.
eas under investigation in combination with sented in a way that it could support immedi-
hand sampling, laying emphasis on the vol- ately the initiation of any future activity either
ume of the residues left behind or being pro- research, or investment. Two major outcomes of
duced. the project consist:
- Orientation test work on the samples col- - The compilation (GIS) of qualitative maps
lected in order to classify the materials in identifying the potential of the existing mar-
terms of composition (from calcitic to dolo- ble waste for industrial usage (Fig. 2).
mitic), as well as color evaluation. - The highlighting of new and developing mar-
- Bulk sampling and pilot scale crushing - kets keen to absorb marble waste.
grinding and classification of the bulk sam-
ples, which all underwent tentative mechani- The major results of the project are presented
cal testing regarding their applicability as ag- in the following tables.
gregates for common concrete. Based on the results of the field research and
- Market research that focussed mainly in the the laboratory examination of the samples col-
perspectives of carbonate fillers, the main lected during the project, the rejects of the major
added value product targeted during inte- marble producing areas of the country were
grated management of white marble residues. classified as per their mineralogical composi-
tion, the volume of the stockpiled materials and
the color of the powder produced after fine
3. SIGNIFICANT OUTCOMES & RESULTS grinding (-30 μm).
OF THE PROJECT These results are summarized in Table 1 that
The information collected during the five years follows. According to these, calcitic marbles
of the project was evaluated, screened and pre- with high reflectance values of the ground mate-
rial occur in considerable amounts in the areas
Figure 2: GIS compositive presentation of properties of stockpiled marbles’ extraction residues, in the area of Falakron-
Drama Prefecture-Northern Greece.
of Drama, Kavala and Vermion, whereas the are of diversified nature from pure calcitic to
marbles’ extraction residues of Peloponnesus pure dolomitic and from dark colored to white.
Table 1: Qualitative and quantitative characteristics of the Greek marbles’ residues at major marble producing areas (Ag-
gelopoulos, 2008; Karadassi, 2007; Chalkiopoulou et al., 2008a).
Mineralogical Powder Whiteness Volume of resi-
Area
Composition (Tristimulus value Y) dues,106 m3
Drama - E. Macedonia: Volakas, Granitis,
Piges, Katafito, Tsakalina, Petrousa, Pirgoi, Dolomitic 90-95%
Ohyro
>30
Drama - E. Macedonia: Ksiropotamos,
Varying from cal-
Monastiraki, Livadero, Dendrakia, Makry- <<85-96%
citic to Mg-calcitic
plagi, Taksiarxes, Bathilakkos
Kavala - E. Macedonia Calcitic 93-96% n/a
Vermion - C. Macedonia: Koumaria, Kastania Calcitic 92-97% 2
Magnisia - Thessaly Mg-calcitic 90-94 n/a
Varying from cal-
Argolida - Peloponnesus <<90-92% 3
citic to dolomitic
Lakonia - Peloponnesus: Petrovouni Calcitic 92-94% n/a
Varying from Mg-
Arkadia - Peloponnesus <90-95% n/a
calcitic to calcitic
Furthermore, sixteen (16) bulk samples, each it is deduced that although the Peloponnesus
in a quantity of 4 m3, were collected from dif- materials possess low values regarding the opti-
ferent areas. All the bulk samples were crushed cal properties (Y, L*, a*, b*) of the powdered
and classified into five size fractions corre- forms, they exhibit good mechanical properties
sponding to those of normal aggregate products, rendering them suitable for aggregates applica-
namely +30 mm, -30+12 mm, -12+5 mm, - tions. On the contrary, materials from N. Greece
5+0,5 mm, -0,5 mm. A series of tests were per- show a great potential for fillers’ application.
formed on the products of the pilot scale crush-
ing and grinding of the bulk samples. The mate-
rials were evaluated as per their appropriateness 4. THE GREEK MARKET POTENTIAL FOR
for their application as aggregates and fillers’ MARBLE REJECTS USAGE
production. The physico-mechanical testing of A market research was conducted within the
the bulk samples included examination of Los project for the identification of the potential of
Angeles strength and strength of parent rock as the Greek market for immediate use of marble
well as the Flakiness index. Regarding the fill- rejects. Two direct user-industries were exam-
ers’ application, the materials were examined ined: a) the fillers production industry, and b)
for their calcite content, their abrasiveness and the ready-mixed mortars industry. From this re-
their optical properties. search it was concluded that the capacity of the
The results of the above examinations are in- Greek white carbonate fillers producing enter-
cluded in Tables 2 and 3. Based on these results, prises was doubled over the period 2000-2008
Table 2: Mechanical properties of the crushed products of bulk samples.

Los Angeles (1) Flakiness Index (2) Parent rock
Bulk sample origin Class EN Class EN strength, MPa
Value, % Value, %
12620 12620 (3)
Peloponnesus: Didyma 26.1 LA30 22.3 Fl30 106.8
Peloponnesus: Karnazeika 25.7 LA30 24.8 Fl30 93.2
Peloponnesus: Parnon 52.3 LA52
Vermion: Koumaria, white type 44.6 LA50
Vermion: Lofotopos 35.2 LA40 26 Fl30 70.1
Kavala: Limnia 40.2 LA50
Drama: Monastiraki 42.7 LA50 16.3 Fl20 71.8
(1) ELOT 408 limit: <40%
(2) ELOT 408 limit: <50% (Ordinary limestones ~ 9-13%)
(3) >65 MPa
Table 3: Powdered products of the bulk samples: Fillers’ properties.

Abrasiveness(a) Optical Properties (c)
Bulk sample origin Einlehner, Calcite, %
VWB120 L* a* b* Y
mg
Peloponnesus: Didyma 51.6 169.2 (b) 93.8 0.51 2.93 84.8 70
Peloponnesus: Karnazeika 99.6 326.6 93.4 0.64 4.04 83.9 90.5
Peloponnesus: Parnon 28.6 93.8 97.2 -0.00 0.55 93.0 98
Vermion: Koumaria, white
18.1 59.3 97.9 -0.01 0.49 94.6(d) 98
type
(d)
Vermion: Lofotopos 17.2 56.4 97.9 -0.04 0.38 94.6 97.5
Kavala: Limnia 21.6 70.8 98.5 0.01 0.45 95.1 99
Drama: Monastiraki 19.0 62.3 98.5 -0.06 0.28 95.2 99
(a)
Grain size d90=30μm
(b)
VWB60
(c) Grain size: d90=30μm, 8o Diffuse Reflectance Spectrophotometry CIE 10o Observer, Illuminant D65,
Standard: BaSO4 pressed
(d)
Concerns the size fraction -0,5mm, that is less white than the oversize
and is today at the level of 0,8 Mt annually high calcite content 98-99.5 %, exhibiting
(Chalkiopoulou et al., 2008b). The correspond- high quality optical properties (Y>94 %) and
ing annual white carbonate fillers production low Einlehner abrasiveness (<40 mg) are
was also doubled over the same period, being found in the area of Eastern Macedonia
today at the level of 0.35 Mt annually. A sig- Thrace (i.e. Monastiraki, Limnia). Conse-
nificant percentage, >30 %, of the raw materials quently these rejects are appropriate for the
used to feed the mills refers to white calcitic production of added-value materials, namely
marble rejects, while the interest amongst both fillers for the chemical industry. The me-
marble quarry owners as well as fillers produc- chanical properties of the examined samples
ers is steadily growing for exploitation of mar- from the area under discussion are not satis-
ble rejects. factory and thus the corresponding raw mate-
Another developing market in Greece con- rials are not suitable for production of aggre-
cerns the production of ready-mixed mortars gates.
with an annual figure of 0,7 Mt. High percent- - Similarly, the rejects from the Vermion area
age of these products, more than 70 % com- that were examined, possess a high calcite
prises inert mineral raw materials. Usage of content (97-98 %) and a high whiteness
marble rejects at the level of 25 % is an already (>94 %) value. Some of the examined sam-
established application. ples from this particular area may be used for
The above mentioned applications concern aggregates also.
added-value products, since they possess values - As far as the Argolida - Peloponnesus geo-
in the range of 45-200 €/t. graphical unit is concerned; all the properties
studied showed that the corresponding rejects
are hard materials with high Einlehner abra-
5. REMARKS & CONCLUSIONS siveness values and of generally low optical
Areas of potential investment perspectives were properties. Since the composition of these
identified at national level regarding the mar- rocks is diversified and may vary from pure
bles’ extraction rejects, based on the results of calcitic to pure dolomitic including all inter-
the research accomplished within the CSF III mediate phases, their potential application for
2003-2008 project of the I.G.M.E. “Greek Or- aggregates which is concluded based on the
namental Stones’ by-Products and Wastes’ results of their mechanical testing must be
Management and Utilization Study - Contribu- validated after experimental production and
tion in the Sector’s Viable Growth”. More spe- taking into consideration the compliance of
cifically: the composition with standards and specifica-
tions.
- Materials of considerable amounts with a
- The White Carbonate Filler market has been
globally expanding steadily over the last dec- Roskill, 2005. The Economics of Ground Calcium Car-
ade. The world Ground Calcium Carbonate bonate, 2nd Edition, Roskill Information Services Ltd,
ISBN 0 86214 502 3, pp. 7, 44.
(GCC) capacity (2004) is estimated at
56,5Mtpa, with production of fillers and
coatings for paper accounting for most of the
market expansion (Roskill, 2005). The Euro-
pean prospects of exploiting white marbles
for fillers' production have been stressed re-
peatedly (Chalkiopoulou and Grossou-Valta,
2005). Thus, according to the above analysis
and taking into consideration the market
status, there is a clear field for exploitation of
the Greek marble rejects: a) production of
white carbonate fillers, b) production of
ready-mixed mortars, and c) occasional pro-
duction of aggregates.
REFERENCES
Aggelopoulos, K., 2008. Inventory of Marble Residues in
the Prefectures of Argolida, Lakonia and Arkadia-
Pelopennesus. I.G.M.E., Internal Report, CSF III,
Subproject: Greek Ornamental Stones’ By-products
and Wastes Management and Utilization Study - Con-
tribution in the Sector’s Viable Growth.
Chalkiopoulou, F. and M. Grossou-Valta, 2005. Exploita-
tion Prospects of White Marbles for Fillers’ Produc-
tion in Europe, N.T.U.A., Titular Edition Issued for
the Prof. A.Z. Frangiskos, N.T.U.A. 2005, 249pp.
Chalkiopoulou, F., 2008. The White Carbonate Fillers’
Production in Greece. I.G.M.E., Internal Market Re-
search Report, CSF III, Subproject: Greek Ornamental
Stones’ By-products and Wastes Management and
Utilization Study - Contribution in the Sector’s Viable
Growth.
Chalkiopoulou, F., D. Voujoukas and G. Stavropoulos,
2008a. Inventory of Marble Residues in the Prefec-
tures of Drama and Kavala-Eastern Macedonia.
I.G.M.E., Internal Report, CSF III, Subproject: Greek
Ornamental Stones’ By-products and Wastes Man-
agement and Utilization Study - Contribution in the
Sector’s Viable Growth, p.p. 2-4.
Chalkiopoulou, F., C. Laskaridis and K. Hatzilazaridou,
2008b. Status and Trends of White Carbonate Fillers’
Supply and Demand in Greece - Focus on GCC. 19th
Industrial Minerals Congress, Athens 2008.
Karadassi, S., 2007. Mineralogical-Petrographical Study
of Samples from the Marbles’ Extraction Residues in
the Drama Prefecture. Internal Report, CSF III, Sub-
project: Greek Ornamental Stones’ By-products and
Wastes Management and Utilization Study Contribu-
tion in the Sector’s Viable Growth, p.p. 23, 43, 57, 67,
70, 79, 82.
Laskaridis, C., 2008. Hellenic Marble Through the Ages:
An overview of the Marble Producing Areas and the
Stone Sector of Today. Geosphaera Hellenica-Special
Issue on Industrial Rocks and Minerals, I.G.M.E., p.p.
21-26.
Analysis of perchlorethylene in groundwater before and after its

degradation by helical-sorbent extraction and gas chromatography
I. Ciucanu and V. Agotici

West University of Timisoara, Department of Chemistry, Romania
M. Draga and R. Barna

School of Mines, Albi, France
An integrated method of dynamic extraction of Volatile organic compounds (VOC's) containing
perchlorethylene (PCE) with polydimethylsilox- chlorine substituents such as tetrachloroethene
ane (PDMS) helical-sorbent followed by injec- (PCE) and trichloroethene (TCE) have been
tion into a gas chromatograph for analysis was widely used as industrial solvents for metal
developed for the determination of the real con- cleaning and degreasing processes, in dry clean-
centration of PCE in groundwater before and af- ing factories, heat-exchange fluids, typewriter
ter its degradation by oxidation with KMnO4. correctors, plastic and fumigant manufacturing
The main parameters (agitation, temperature, industries (Siegrist et al., 2001). Chlorinated
salts, pH) that affect the extraction efficiency solvents, known as Dense Non-Aqueous Phase
have been evaluated and optimized. The in- liquids (DNAPLs) are typically denser and less
crease of the extracted amount of PCE due to viscous than water. They have different limits of
the presence of the salts could be partially com- solubility in water. PCE is approximately 40
pensated by the opposite effect of the insoluble times less soluble then trichloromethane (Kue-
MnO2, and of the presence of HCl and the per et al., 2003). Unfortunately, the limit of
global effect of the matrix would be less impor- solubility for PCE was found in the range from
tant for the reproducibility of the PCE extrac- 120 mg/L to 485 mg/L at 25 °C (Horvath et al.,
tion. The quantitative analysis of PCE was car- 1999). All these organochlorinated compounds
ried out with the external standard method using are considered hazardous waste in many indus-
the calibration-curve technique. The limit of de- trialized countries. Intentional or accidental
tection for flame ionization detector was found leaks and spills cause contamination of ground-
in the range 0.05-1.0 mg PCE/L of water. The water and soils. The discharge of even a small
solubility of PCE in water was studied by quantity of organoclorinated compounds can
PDMS helical-sorbent extraction of PCE from lead to long-term contamination of both the un-
standard solution with different amounts of PCE saturated and saturated zones at the site. The
added into the same volume of water and was DNAPLs have the ability to migrate to signifi-
found to decrease with settling time, because the cant depths, in the subsurface, where they
fine droplets created during the agitation of the slowly dissolve into the groundwater causing
mixture were slowly merged in drops on the the aqueous phase plume (Scott and Bruce,
bottle walls and were not transferred into the ex- 2006).
traction vial for GC analysis. The limit of solu- The contamination of the water resource by
bility was evaluated using a graphical method organochlorinated solvents is a matter of great
and was estimated at 224.2 ± 15.7 mg PCE/L of common interest worldwide because they are
water at 25ºC. classified “probably” carcinogenic to humans.
Their persistence in both biotic and abiotic envi-
ronments, their toxicity and potential carcino-
genesis have led to the considerable necessity to
remove these pollutants from contaminated pared by adding to well known amounts of pure
sites. The classical technologies (pump & treat) PCE deionized water up to fill 255 mL glass
capable to treat the aqueous plume are expected bottle. The mixture was magnetically stirred at
to take several decades and are expensive 25 °C for a rapid dissolution. The agitation was
(Reitsma and Dai, 2000). Most popular tech- stopped after 5h and was followed by 24 h of
nologies are: volatilization technologies (vapor static equilibration. An aliquot of solution was
extraction, air sparging or stripping), thermal transferred with a syringe into the extraction
technologies (steam injection, electrical heating, vial.
vitrification), electrokinetics, bioremediation,
2.2 Preparation of Helical Solid Sorbent.
barriers-containment, flushing technologies (al-
cohol or co-solvent, surfactant and oxidant Details of the preparation of the helical solid
flushing (In Situ Chemical Oxidation - ISCO) sorbent have already been described (Ciucanu,
(Fountain, 1998). Practically, during the reme- 2002). The solid sorbent used in these experi-
diation process it is necessary to measure accu- ments was actually a soft solid consisting of
rately the aqueous concentrations of DNAPL PDMS rubber. This soft solid sorbent was
(PCE), for a range from zero to solubility limit. coated on a helical support made by wrapping a
Consequently, the determination of the aqueous wire around a straight wire with the same di-
concentration of DNAPL with a high accuracy ameter at a constant pitch. The helical support
is a very important task in order to confirm the was coated with several layers of PDMS by
degradation of the pollutants and the moments immersion of the helical support in a concen-
when the organic compounds are below suitable trated solution of PDMS in dichloromethane.
level. PDMS cross-linking was accomplished by add-
This paper describes experiments undertaken ing 0.8 % dicumyl peroxide to the coating solu-
to determine the suitability of the extraction tion. After each coating, the solvent was evapo-
with PDMS helical solid sorbent for the moni- rated by heating the helical sorbent at 150 °C for
toring of the PCE level in the groundwater 30 min. The thickness of the final coating poly-
sources during its degradation by in situ chemi- mer depended on the concentration of PDMS
cal oxidation process. The limit of solubility of solutions and the number of coatings. The heli-
the PCE in water has also been evaluated with cal sorbent was weighed after each coating. The
this technique. final length of the helical sorbent was adjusted
to give the same amount of PDMS as on a 100-
μm Supelco PDMS fiber in order to compare the
2. EXPERIMENTAL amounts extracted with these solid sorbents.
2.1 Chemicals and Materials 2.3 Solid-Phase Extraction with PDMS Helical
High-purity PCE and n-hexane were purchased Sorbent
from Fisher Scientific-Labosi (Elancourt, The PDMS helical sorbent sheathed in a needle
France). PCE is toxic and must be handled care- was housed in the manual holder device. All the
fully in fume hood. Kalium permanganate extractions with PDMS fibers were performed
(KMnO4), kalium chloride (KCl), sodium sul- in 2 mL extraction vials filled with 2 mL of
fate (Na2SO4) and manganese dioxide (MnO2) PCE standard solution and involved a few sim-
were from Sigma-Aldrich Chimie (ST Quentin ple steps (Ciucanu et al., 2004). A small mag-
Fallavier, France). Deionized water was ob- netic stirring bar was added to the vial and was
tained with a Milli-Q water purification system tightly capped with Teflon coated silicon septa.
(Millipore, ST Quentin En Yvelines, France) The extraction temperature was adjusted by the
and was held for 10 h in ultrasonic cell at 25 °C immersion of the vial into a regulated tempera-
for degassing. ture controlled water bath and let to stay for
The 100-µm thick non-bonded polydimethyl- 5 min before carrying out the extraction.
siloxane (PDMS) fibers and the manual fiber
holder were from Supelco (Bellefonte, PA, 2.4 Gas Chromatography
USA). The gas chromatographic analysis of the ex-
Standard PCE aqueous solution were pre- tracted analytes was carried out on an Agilent
6890 N gas chromatograph with a flame ioniza- present in sample have a faster access to the
tion detector (FID). By exposing the coating fi- helical sorbent decreasing the extraction time.
ber in the GC injector at 220 °C for 5 min, the When the stirrer speed was higher than 500 rpm,
analytes was desorbed. The injector was operat- no change of the extraction time has noticed.
ing in splitless mode. GC separation of the ana- Increasing the sample temperature generated
lytes was performed on a RTX-624 capillary a decrease of the amount of PCE extracted at the
column (60 m x 0.53 i.d. and 3 µm film thick- same extraction time. For this reason, the high
ness) from Restek (Lisses, France). Helium was temperature for extraction could be a major
used as carrier gas at a constant flow of concern if the sensitivity is not sufficient to
1 mL/min. The GC oven temperature was quantify PCE at ppb levels. In such cases, the
started at 60 °C and after 2 min was increased to extraction at room temperature (25 °C) is indi-
150 °C with 10 °C/min and then to 220°C with cated, even if a higher temperature could im-
30 °C/min. The detector temperature was main- prove the extraction rate.
tained. Before sampling, the new solid sorbent
3.2 Influence of the Sample Matrix
were conditioned for about 2 h in the GC injec-
tor port at 280 °C. Quantitative evaluation was The degradation of PCE from groundwater by
performed by the external standard calibration oxidation with KMnO4 can be described (Hood
method. The calibration curve was done plotting et al., 2000) by the global reaction:
of the GC detector signal versus five different
3C2Cl4+4KMnO4+4H2O→6CO2+4MnO2
standard concentrations.
+4KCl+8HCl [1]
3. RESULTS AND DISCUSSION The aqueous matrix will contain various re-
action products such as KCl, MnO2 and HCl.
3.1 Influence of the Temperature and Agitation During the oxidation process, it has not been
The extraction of PCE with PDMS solid sorbent observed other organic products. For the kinetic
is an equilibrium process and the time necessary study of the oxidation process (Yan and
to reach equilibrium can be estimated studying Schwartz, 1999), an aqueous solution of
the parameters that affect the mass transfer of Na2S2O3 is injected into the oxidation mixture in
the analytes from bulk solution into the PDMS order to stop the PCE degradation by the reduc-
solid sorbent. The extraction time depends for ing of the oxidant in a fast concurrent reaction:
one specific analyte and sorbent on agitation, 3Na2S2O3+8KMnO4+H2O→3K2SO4+
temperature and sample matrix.
The extraction time was measured from the 3Na2SO4+8MnO2+2KOH [2]
insertion of the PDMS sorbent into the sample Two new salts will be generated in the sys-
solution until it was retracted into the protection tem. In the case of a solution of 180 mgPCE/mL
needle. The time required to reach the extraction of water that is under the limit of solubility, the
equilibrium can be determined from the extrac- total amount of inorganic compounds
tion curve when the steady state is reached. In (0.00041 g MnO2/mL; 0.00012 g KCl/mL;
2 mL extraction vial with 200 mg PCE/L pure 0.00011 g HCl/mL; 0.00020 g K2SO4/mL;
water, at 5 h settling time, 500 rpm, and 25 °C, 0.00016 g Na2SO4/mL; 0.00004 g KOH/mL) is
the extraction equilibrium was reached in small. Figure 1A shows the influence of the pH
25 min for Supelco PDMS fiber sorbent and on the PDMS fiber extraction of PCE from wa-
12 min with PDMS helical sorbent. A longer ter. The pH vas modified adding different
extraction time did not increase the amount of amounts of hydrochloric acid or potassium hy-
extracted PCE. droxide. The amount of hydrochloric acid gen-
The agitation of the liquid sample improves erated in the oxidation process might increase
only the mass transport of analytes from the the pH up to 3 and consequently will increase
bulk of the liquid phase into the sorbent layer. the extracted amount of PCE. The influence of
In the above conditions, increasing the speed of the salts generated in the oxidation process on
the magnetic stirrer up to 500 rpm, the aqueous the PDMS helical sorbent extraction is shown
phase is better homogenized so that the analytes Fig. 1B. An increase in the concentration of
presence of the salts could be partially compen-

sated by the opposite effect of the insoluble
MnO2 and of the presence of HCl. Conse-
quently, the global effect of the matrix would be
less important for the reproducibility of the PCE
extraction.
3.3 Quantitative Determination of PCE in Water
The first step in quantification was the estab-
lishment of the limit of linearity of the detector
response without overloading the GC column.
The linearity of the FID was checked out by di-
rect injection of different concentrations of PCE
in n-hexane. Plot the peak area (Y) versus mg
PCE/L hexane (X) was linear (Y=375.9 X) with
a regression coefficient of R2=0.9992 up to
1000 mg PCE/L of n-hexane. The quantitative
analysis of PCE was performed with the exter-
nal-standard method using the calibration-curve
technique. The dissolution protocol consisted in
the complete filling of the bottles with water
followed by magnetic stirring at 500 rpm for
10 h at 25 °C. After this step, each series of vi-
als were maintained at 25 °C without stirring for
Figure 1: Sample matrices effects on PCE extraction effi- 5 h, 30 h, 165 h, and 600 h.
ciency: A) Influence of pH. B) Influence of salts. Condi-
tions: 2 mL extraction vial with 50 mg PCE/L pure water, 3.4 Determination of the PCE Limit of Solubility
5 h settling time, 10 min extraction time, 500 rpm, and in Water
25°C.
The peak area of PCE has a linear variation
Na2SO4 and KCl generated an increase in the function of mg PCE added in 1 L of water up to
amount of PCE extracted by sorbent. K2SO4 had 200 mg PCE/L of water. The increase of the
a similar effect than Na2SO4. An explanation of amount of PCE further gave an asymptotic
these effects from Fig. 1B could be found in the variation for the corresponding peak area. The
increase of the ionic strength of matrix in the plateau of this curve is the result of the satura-
presence of the ionic species from salts, which tion and is an equilibrium process. When the
reduces the solubility of the neutral organic ana- settling time was increased, the peak area values
lyte in polar aqueous matrix and increase its were smaller for the same value of mg PCE/L of
solubility in non-polar PDMS. The presence of water, because the fine droplets created during
MnO2 in the matrix decreased the efficiency of the agitation of the mixture were slowly merged
extraction. This suggests that the insoluble in drops on the bottle walls and were not trans-
MnO2 in this aqueous matrix can cover the sur- ferred into the extraction vial for GC analysis.
face of the PDMS sorbent and reduces the ac- For the settling time higher than 600 h, the peak
cess of the analytes to sorbent. These influences area values are very closed, showing that equi-
of the ionic species presented in the oxidation librium was reached. Consequently, the limit of
aqueous matrix can be minimized by dilution of solubility can be estimated at 600h of the set-
sample. However, one should never forget that tling time. Plotting the amount of PCE extracted
the dilution of sample will automatically reduce as a function of the amount of PCE added in the
the sensitivity of analysis. The sample after oxi- standard solutions for 600 h of settling time, we
dation is a relatively diluted in inorganic com- obtained the graph from Figure 2. The limit of
pounds and they do not have a significant influ- solubility is determined using a graphical
ence on the extracted amount of PCE. Moreover method as the intersection between the tangent
the increase of the extracted amount due to the to plateau of the curve from Figure 2 with the
REFERENCES
Ciucanu, I., 2002. Helical sorbent for fast sorption and
desorption in solid-phase microextrtaction-gas chro-
matographic analysis. Anal. Chem. 74, 5501-5506.
Ciucanu, I., K.C. Swallow and R. Caprita, 2004. Micro-
solid phase extraction with helical-solid-sorbent in the
presence of organic solvent for GC-MS analysis of
per-O-methilated mono- and disaccharides. Anal.
Chim. Acta 519, 93-101.
Figure 2: Graphical method for the determination of the Fountain, J.C., 1998. Technologies for Dense Nonaque-
solubility limit of PCE in deionized water. Conditions: ous Phase Liquid Source Zone Remediation, Technol-
2 mL extraction vial, 600 h of settling time, 12 min ex- ogy Evaluation Report, France.
traction time, 500 rpm and 25°C. Hood, E.D., N.R. Thomson, D. Grossi and G.J. Farquhar,
2000. Experimental determination of the kinetic rate
calibration line. We propose the limit of solubil- low for the oxidation of perchloroethylene by potas-
ity of PCE in water at 25 °C the value of sium permanganate. Chemosphere, 40, 1383-1399.
224.2±15.67 mg PCE/L of water. Horvath, A.L., F.W. Getzen and Z. Maczynska, 1999.
IUPAC-NIST Solubility Data Series 67: Halogenated
Ethanes and Ethenes with Water. J. Phys. Chem. Ref.
Data 28, 395-627.
4. CONCLUSIONS Kueper, B.H., G.P. Wealthall, J.W.N. Smith, S.A. Le-
This study shows that the PDMS helical sorbent harne and D.N. Lerner 2003. An illustrated handbook
extraction followed by GC analysis is a suitable of DNAPL transport and fate in the subsurface, Envi-
ronment Agency: Bistol, England.
and reliable method for quantitative analysis of Reitsma, S. and Q.L. Dai, 2000. Reaction-enhances mass
PCE from groundwater before and after in situ transfer and transport from non-aqueous phase liquid
chemical oxidation of PCE with KMnO4. The source zones. J. Contam. Hydrol. 49, 49-66.
optimum extraction time was 12 min for mag- Scott, G.H. and E.P. Bruce, 2006. In-Situ Chemical Oxi-
netic stirring at 500 rpm at 25 °C. The presence dation. United States Environmental Protection
Agency,Washington DC, USA.
of KCl, Na2SO4 and K2SO4 increased the ex- Siegrist, R.L., M.L. Urynowicz, O.R. West, M.L. Crimi
tracted amount of PCE. HCl and the insoluble and K.S. Lowe 2001. Principles and practices of in
MnO2 have an effect in the reverse direction and situ chemical oxidation using permanganate, Battlle
the global matrix effect on the extraction effi- Press, Columbus, USA.
ciency can be ignored at much diluted solutions. Yan, Y.E. and F.W. Schwartz, 1999. Oxidative degrada-
The flame ionization detector has a very good tion and kinetics of chlorinated ethylenes by potas-
sium permanganate. J. Contam. Hydrol. 37, 343-365.
linearity up to 1000 mg PCE/L of hexane. The
solubility of PCE in water depends on the set-
tling time due to the formation of very fine
droplets of PCE during the agitation of the PCE
and water mixture. The optimum settling time
was 600 h. Consequently, the limit of solubility
was estimated at 224.2±15.7 mg PCE/L of water
at 25 ºC using standard solutions with different
amounts of PCE per liter of water at a settling
time of 600 h on the base of a graphical method.
ACKNOWLEDGMENT
Financial supports for Project 2521/2009 with
CNCSIS- UEFISCSU from Romanian Ministry
of Education and Research were gratefully ac-
knowledged.
Characterisations of silicon carbide coatings deposited on widia plates by

chemical vapour deposition
S. Constantinescu, M. Vlad and L. Orac

University ”Dunarea de Jos” of Galati, Romania
ABSTRACT tional vapor, chemical and special deposition

procedures.
The experiments conducted to obtain thin layer
Compared with the other methods, the vapor
of carbide by the vapor chemical deposition
chemical deposition features the following ad-
method have followed an original path to make
vantages: highly pure thin layers obtained by a
SiC directly in the working room thus avoiding
suitable choice of the initial materials and reac-
the import of these hazardous substances. The
tions; perfectly crystalline layers due to growth
SiCl4 is obtained in the heat treatment chamber
under appropriate equilibrium conditions, rela-
by adding chloride acid vapors passed over the
tively high temperatures possibility to cover the
incandescent ferrosilicium. The thickness of the
samples with thin layers of complex shape, the
deposit layer increases with the time of expo-
process is and can be automated and adapted to
sure to the working temperature.
full scale processing of a large number of sub-
Characterization of coating deposed by CVD
layers (Constantinescu, 1998).
method was done using scanning electron mi-
If the vapor chemical deposition takes place
croscope (SEM), X-ray diffractometer (XRD)
within a tubular continuous reactor, a gas carry-
and microhardness measurements. The thick-
ing the reacting species is passed over the sub-
ness of the SiC layer was determined by the mi-
layer. At the sub-layer surface, the reacting ele-
croscope and the Kalotest device. The operation
ments undergo a number of chemical reactions
of the latter is based on a housing which cuts the
leading to product formation. Part of the prod-
deposited SiC layer. The values of the thin SiC
ucts is deposited on the sub-layer and part of it
layers as measured by the Kalotest device are in
goes back to the gas stream.
good agreement with the values measured by
Before examining the vapor chemical deposi-
microscopic analysis but slightly lower. Sam-
tion reactions it must be determined if the reac-
ples for metallography were prepared by polish-
tion is possible thermodynamically. The reac-
ing, this prevented damage to the dissimilar in-
tion will be possible thermodynamically if the
terface (strate - substrate) during polishing.
calculated concentrations (partial pressures) of
SEM was used to investigate the coating mor-
the reactants, under equilibrium conditions, are
phology and interface structure. X-ray mapping
less than their original concentrations.
were also performed to characterize the ele-
The calculation of the equilibrium concentra-
ments in a semi-quantitative analysis. Dron X-
tions from the equilibrium constant involves a
ray diffractometer with Mo Kα radiation operat-
good choice of the number of gas spaces which
ing was used for phase(s) identification.
can be higher than two and the number of inde-
pendent relations A relation implies the equilib-
1. INTRODUCTION rium expression depending on the free standard
reaction energy and temperature. The other rela-
The vapor chemical deposition is a widely tion consists in that the system pressure is the
spread method of making thin layers. It has been sum of the partial pressures. If some reactants
gaining ground lately as opposed to the conven- possess more than one valence state, the reac-
Table 1: Cutting regime. 2.2. Deposition Method

Material v
Type n s t The experiments conducted to obtain thin layer
to ma- [mm/
of plate [rot/min] [mm/rot] [mm] of carbide by the vapor chemical deposition
chine min]
820 180,2 0.096 method have followed an original path to make
P30 35C10 0.5 SiC directly in the working room thus avoiding
900 197 0.096
White 500 122,5 0.096 the import of these hazardous substances. The
K20 0.5 SiCl4 is obtained in the heat treatment chamber
Cast iron 630 154 0.096
by adding chloride acid vapors passed over the
tion should contain the reactant under its most incandescent pure titanium. Lab-scale systems
stable valence state (Constantinescu and Radu, have been designed with the possibility of use at
2003). industry scale for small production. The support
Hard alloys made out of metallic carbides temperature was established at about 1125oC so
manufactured to an industrial scale for cutting that the SiC can provide a suitable deposition of
processing can be divided in two categories, ac- the thin SiC layer. The thickness of the deposit
cording to their use. The second category of in- layer increases with the time of exposure to the
dustrial products comprises the alloys out of working temperature. The SiC coated plates fea-
many carbides used in cutting process of mate- ture higher endurance, capabilities than those
rials with long and continuous chips (all sorts of uncoated for the same cutting speed both for
steel). steel and white cast iron.
2.3. Characterization
2. METHODS
Characterization of coating deposed by CVD
2.1 Materials method was done using scanning electron mi-
SiC coated and uncoated plates types T.P.U.N. croscope (SEM), X-ray diffractometer (XRD)
22.04.08.P30 and S.N.U.N. 15.04.08.K20 have and microhardness measurements. The thick-
been tested. ness of the SiC layer was determined by the mi-
For the purpose of the experiments the maxi- croscope and the Kalotest device (Constan-
mum limit of the cutting speed was taken higher tinescu et al., 2006).
than the usual speeds to obtain a plate durability The operation of the latter is based on a hous-
under the most difficult operation condition ing which cuts the deposited SiC layer. The val-
which should not excessively increase the time ues of the thin SiC layers as measured by the
of the experiments and the material consump- Kalotest device are in good agreement with the
tion (Constantinescu, 2005). values measured by microscopic analysis but
Table 1 gives the ranges of the cutting pa- slightly lower (Constantinescu, 2007a).
rameter values. Samples for metallography were prepared by
Table 2: Compositions and physical properties of the coating and strate material.
Hardness Banding cracking Compression Thermal
Composition % Density
Vickers strength Strength Conductibility
[g/cm3]
WC SiC Co [daN/mm2] [daN/mm2] [daN/mm2] [cal/cmsºC]
94 1 5 14.5-14.7 1750-1850 120-140 555 0.18
87.5 2.5 10 14.0-14.2 1850-1950 140-158 456 0.15
84.5 2.5 13 13.7-13.8 1950-2050 158-198 445 0.14
86 5 9 13.2-13.4 2050-2150 130-140 456 0.14
82 5 13 12.8-13.0 2150-2250 140-158 - -
80 12 8 11.8-12.0 2250-2350 130-150 - 0.077
78 14 8 11.1-11.3 2350-2450 110-120 415 0.08
76 16 6 11.0-11.2 2450-2550 108-122 425 0.09
76 16 8 10.9-11.1 2550-2650 118-128 - 0.068
69 25 6 9.6-9.8 2650-2750 70-90 - 0.049
61 32 7 8.7-9.0 2750-2850 59-80 406 0.038
34 60 6 6.5-6.8 2850-2950 50-59 386 0.029
polishing, this prevented damage to the dissimi- nous solutions of SiC - WC, some richer in SiC
lar interface (strate - substrate) during polishing. (Fig. 1).
SEM was used to investigate the coating mor- The metallographic aspects of cobalt distri-
phology and interface structure. X-ray mapping bution in -alloys of the type, WC - SiC - Co is
was also performed to characterize the elements shown in Figure 2.
in a semi-quantitative analysis. Dron X-ray dif- Micro-hardness is not a constant like Vickers
fractometer with Mo Kα radiation operating was hardness, in spite of the geometrical similarity,
used for phase(s) identification. Microhardness but decreases with higher testing charges de-
was determined by Neophot Micromet micro- pending on the size of the print The micro hard-
hardeness tester at a load of 100 gf (Constan- ness tests show that we have SiC, value HV0,05
tinescu, 2007b). = 29500 MPa is in good agreement with the data
from the literature.
Measurements were made on SiC covered
3. RESULTS AND DISCUSSION thin plates which thickness range between 6, 8
It should be mentioned that the optimum layers and10 μm. Micrographis. Figures 3 - 5 show an
in the cutting process are the SiC layers having adherent layer which is uniform and homoge-
thickness within 4 - 8 μm above these values, nous over the entire depth (Constantinescu
the layers loose tenacity and become fragile. et al., 2008b).
As result of the thermal treatment which The values of the thin SiC layers as measured
means heating up to 1125οC degrees for various by the Kalotest (Fig. 6) device are in good
exposure times, layer thickness within 3 - 12 μm agreement with the values measured by micro-
were achieved (Constantinescu and Orac, 2007). scopic analysis but slightly lower.
The plates were covered with an adherent The steel ball diameter is 12 mm. Since the
coat of SiC of black - silvery color. The thick-
ness of the thin layers increases with the time of
exposure to the working temperature.
The microhardness of WC-SiC-Co alloys is
affected by a large number of elements con-
nected to the raw material, purity and compo-
nent dispersion in the alloy and the solid solu-
tion quality and grain size of components. In the
factory process, these elements are playing an
ultimate role in effective microhardness meas-
urement of the material with a given chemical
composition. Under this structure, the brilliant
angular crystals are of phase WC, the gray
spherical element is the solid solution while the Figure1: Metallographic appearance of alloy with
element in the dark carbon grains is the sinter- 80%WC, 12%SiC, 8%Co, x1500.
ing bondage-the cobalt. The structure becomes
even more complex with the pseudo-alloys of
type WC - SiC - Co depending on the tempera-
ture and reciprocal solubility conditions a num-
ber of solid solutions of SiC - WC can form;
they depend on the WC and SiC solubility and
reciprocal, solubility.
Under structural equilibrium conditions, both
the WC phase and the homogenous solid solu-
tions along with the cobalt can form in the struc-
ture (Constantinescu, 2008a).
Under practical sintering conditions, the
equilibrium is not always achieved and the Figure 2: The metallographic aspect of cobalt distribution
structure frequently reveals solid non- homoge- very good distribution.
Figure 3: SEM micrograph of the cross section of SiC Figure 6: The sphere shell of SiC by the Kalotest.
coating showing strong strate-substrate interface. urement is illustrated in Figure 8.
4. CONCLUSIONS
These coatings have good wear resistance, abra-
sion resistance, corrosion resistance and a strong
strate -substrate interface. This leads to forma-
tion of thick and rough coating. The coating is
fine grained, adherent, dense and fee from
cracks. However, some porosity is observed in
the coating layer.
The widia plates coating with thin SiC layers
entirely suppresses the inconveniences of a rela-
Figure 4: SEM micrograph of the SiC coating showing tively rough topography of the common sinter-
coating morphology.
ized carbides while preserving the adequate ma-
terial mechanical strength. The layer begins
loosing its tenacity if its thickness increases
considerably exceeding the thickness 10μm
mainly due to the lower strength characteristics.
This together with the increase in the inner ten-
sions results in cracks and breakings in the lay-
ers. This has been attributed to poor wetting.
5. REFERENCES
Constantinescu, S., 1998. Studies on Thin Carbide and
Nitride Layers Deposition on Metal Basis, Based on
Figure 5: SEM micrograph of the cross of SiC coating Chemical Reaction at High Temperatures, Academic
showing microhardness variation. Thesis, Galati, p. 58-83.
Constantinescu, S. and T. Radu, 2003. Advanced methods
shell diameter is much less than that of the ball, of obtaining thin coats, Bucureşti, F.M.R. ISBN 973-
8151-25-2
the layer thickness can be calculated by the val- Constantinescu, S., 2005. Analyses and control methods
ues of the steel ball rase, the strate and substrate of metallic materials, Bucureşti, Publisher F.M.R.
thickness (Constantinescu, 2008). ISBN: 973-8151- 38 -4
The best results are obtained for layer depth Constantinescu, S., M. Vlad, L. Radu and A. Ciubotariu,
of 7 μm, with homogeneous and even structure, 2006. CVD of niobiummicrohardness layers type
NbN. Proceeding of the 15th International Metallurgi-
a feature that can be emphasized by means of cal and Materials Conference-METAL 2006, Techni-
diffraction pattern analysis (Fig. 7). cal University Praha, Edited by Ing. Jan Stefan, Tech-
The thickness of SiC film for 4,5 h exposure nical University Ostrava p.62, paper54, ISBN 80-
time at 1125oC for surface profilmeter meas- 86840-18-2, Czech Republic
Figure: 7 XRD pattern of carbide thin layer for 4,5 h exposure time.
Figure 8: Surface profilmeter measurement of the thickness of SiC film for 4,5 h exposure time at 1125oC.
Constantinescu, S., 2007a. Practical experience of using velopment, TQSD 2008, Bucharest, Romania, ISSN:
chemical vapour deposition coatings to resist wear, 1844-9158, AGIR Publisher HOUSE, Editors:
Tehnologii şi produse noi în construcţia de maşini, *University Politehnica of Bucharest, *Academy of
Tehnomus XIV, Published by University „Stefan cel Technical Sciences of Romania, *General Association
Mare”, Suceava, ISBN 1224 - 029X, p.261 Engineers In Romania, p.75-80.
Constantinescu, S., 2007b. Highly Pure Thin Layers Ob-
tained By Vapours Chemical Deposition, Conference
UGALMAT Galati Vol.II Advanced Technology and
Materials, ISSN 1843-5807, p. 436-441.
Constantinescu, S. and L. Orac, 2007. Materials proper-
ties and control methods, Publisher by EUROPLUS
Galati, ISBN 978-973-7845-81-8
Constantinescu, S., T. Radu and L. Orac, 2008a. Wear-
resistance of the WC-Co alloys used for short chip
materials cutting, Milos, Greece Proceedings
GEOENV 2008, Geoenvironment and Geotechnics,
Publisher by Heliotopos Conferences, ISBN 978-960-
6746-01-7
Constantinescu, S., M. Vlad and T. Radu, 2008b. Re-
searches concerning the correlation between properties
and the structural characteristics of the highly loaded
steel plates Proceedings of Conference, Aachen, Ger-
many, p. 162- 168.
Constantinescu, S., 2008. Modern method to perform thin
layers in surface engineering using unconventional
procedure. Proceedings of the 8-th International Con-
ference on Technology and Quality for Sustained De-
New biotechnological aspects for ecological remediation of lands

contaminated with radionuclides in uraniferous mining works
G. Crutu, D. Curelea, G. Groposila and S. Ciuciu

R&D National Institute for Metals and Radioactive Resources-ICPMRR Bucharest, Romania
ABSTRACT iles some plants from local flora and reducer

sulphate bacteria have been used.
The uranium ore exploitation and processing in-
dustry represents one of the most polluting
sources of natural resources - soil and waters- 1. INTRODUCTION
considered to be essential for normal function-
In the exploration and exploitation activity of
ing of the terrestrial and aqua terrestrial ecosys-
complex ores, the impact of ore and waste
tems.
dumps on the environment consists in the con-
The closed mine works are located in the
tamination of the soil under the dumps and of
central part of Apuseni Mountains and Banat
the environment due to the migration phenom-
County, Romania. In 1992 the geological re-
ena of toxic elements such as, Mo, Pub, Zn, Cu
search was closed out. At present there are a lot
etc., or radioactive elements such as U and Ra.
of dumps, consisting (with one exception) of ar-
Among the waste which represent major pollut-
eas with high level of radioactivity up to 5
ing sources and which require special attention
μSv/h, i.e. higher than the standard limit of 0.3
are those resulting from uranium ore mining and
μSv/h.
processing.
The research carried out for decontaminating
The research which has been carried out so
lands and dumps in the uranium industry was
far spotlighted the fact that the use of some
concluded with the application of a new na-
natural materials (clay, peat coal, sawdust) can
tionwide technology of recuperative treatment,
diminish the ground water pollution in the
revegetation and bioremediation for removing
neighbourhood of the mining works.
heavy metals and radionuclides from soil and
The relation between the actions on the envi-
water. Through the research carried out in the
ronment and their impact is a cause-effect rela-
laboratory, vegetation houses and a pilot ex-
tion. For quantifying the impact on the envi-
perimental field, the influence of residual radia-
ronment, waste must be reduced first of all. Not
tions on soil, water and plants on/and around
accidentally, the fourth report of the Club in
waste dumps and deposits in those areas was in-
Rome was named “Waste Era” and highlighted,
vestigated, as well as their rehabilitation.
in the end, that “our society has the tendency to
Considering that the pollutants represent food
waste natural resources, both in the domain of
sources for micro-organisms, the identification
energy, raw materials and food products”. This
of these which can degrade a certain type of pol-
waste is mainly attributed to the technologies
lutant represents a very important mission for
used, even to the most advanced ones (Brown,
the design of the bioremediation system. The
2001).
paper presents a few results obtained regarding
Therefore, it is compulsory to reduce waste,
an ecological remediation activity in the labora-
in order to extend the availability of the natural
tory and in situ in our institute and in our coun-
resources and limit the disturbance of the eco-
try by two methods: bioremediation and phy-
system.
toremediation. For recultivating radioactive ster-
Table 1: Surfaces and volumes of the uranium waste d) Site “planned” use
dumps. e) Correct monitoring capacity.
Mining ac- Dump Dump
Material disposed 2.2 Phytoremediation of soils polluted with ura-
tivity surface volume
Sterile and low ra- nium
S.M. Bihor 56.3 2,300
dioactive sterile Phytoremediation is an attractive alternative to
Sterile and low ra- the current conventional methods for removing
S.M. Banat 21.0 1,820
dioactive sterile
pollutants from soil, being a cheap method
S.M. Sterile and low ra-
52.2 1,850 which may be applied with minimum energy
Suceava dioactive sterile
consumption.
Exploration Sterile and low ra-
11.3 1,750 Soils contaminated with uranium can be
Sector dioactive sterile
Feldioara ameliorated by phytoremediation, using the ca-
Sterile and low ra- pacity of some plants to develop uranium accu-
processing 68.5 3,200
dioactive sterile mulation mechanisms. The uranium forms
plant
which are best accumulated by plants are uranyl
cations (UO2)2+, which are formed in soil at pH
2. MATERIALS AND METHODS 5, in comparison with U hydroxyl and carbonate
An approximate image on the ratio in which the forms, which are formed in soil solution at pH 6
environmental factors in the uranium mining pe- and respectively at pH 8.
rimeter with waste deposits are affected is
2.3 Nutritive and cultivation mediums used in
shown in Table 1 in which the data regarding
the volume of dumps resulting from the uranium the process of ecological remediation
ore exploitation works are presented. The nutritive and cultivation mediums used in
the process of ecological remediation are:
2.1 Bioremediation of soils polluted with ura-
nium - Medium 9 K for T. ferrooxidans
- For emphasizing Acidithiobacillus ferrooxi-
Using bioremediation processes, currently in- dans and Leptospirillum ferrooxidans iron-
cluded in the category of environmental protec- oxidizer bacteria, MACKINTOSH liquid
tion biotechnologies, degradation of polluting
medium is used.
compounds takes place under the action of mi- - For isolating Acidithiobacillus thiooxidans
cro-organisms to produce inoffensive com- sulphur-oxidizer bacteria HUTCHINSON
pounds (CO2, H2O) and bacteria biomass.
culture medium is used.
There are various microbiological processes - For isolating sulphate - reducing bacteria
in natural mediums, depending on the local
POSTGATE medium is used.
conditions and the depth at which they develop
- STARKEY medium - for isolating and study-
in soil, being, from this point of view, aerobic ing sulphate-reducing bacteria.
and anaerobic. Pseudomonas fluorescens pro-
- WIDDEL and PHENING medium with so-
duces both in aerobic and anaerobic conditions,
dium acetate - isolation of sulphate-reducing
uranium precipitation. Concomitantly with ura- bacteria which oxidize organic compounds at
nium, iron is also precipitated. Clostridium sp.
CO2 (Lazar, 2000).
reduces uranium from U (VI) form to U (IV),
insoluble form which does not pass into soil so- 2.4 Dump location and its impact on environ-
lution (Mullen et al., 1983). ment
There are at least 5 critical factors that must
The dump studied in this paper is situated at an
be considered when we assess bioremediation
altitude of 590 m and has a length of 100 m and
use for cleaning the contaminated medium
a height of 20 m in Apuseni (Bihor) Mountains.
(site):
It was investigated for determining gamma debit
a) Magnitude, toxicity and contaminants mobil- dose, through the analysis of samples for U and
ity Ra concentration. On the dump slope, gamma
b) Approach of human and medium receptors debit doses vary between 0.7-2.0 µSv/h, while
c) Contaminants degradation U minimum contents are: U=300 ppm and
3. RESULTS
3.1 Technology for decontaminating lands pol-
luted with radionuclides
The most aggressive pollution form is due to the
mining works, which outrun enormous surfaces
damaged either by open pits or by disposing
residues resulted from underground mining
works. A special risk for the environment is rep-
resented by the sterile resulting from uranium
ore mining and disposed in dumps such as the
dumps in Apuseni Mountain, and so forth
(Rauta et al., 1983). The dumps with rock resi-
dues, which have a high radioactivity, are reha-
bilitated by covering in situ, by relocation at a
Figure 1: Climatic room. centralized deposit which assures the conditions
necessary for protection, or by reintroducing in
Ra=8 Bq/g. The radioactive waste dump is
the mine galleries.
characterized by the non-uniformity of the ma-
Covering is followed by zone revegetaliza-
terial disposed in the dump, which makes the
tion, the cover surface being covered with vege-
recultivation process more difficult. The dump
tation in order to assure erosion control. The
material which should be recultivated/ fixed is
dump form must reshaped in order to fit the sur-
represented by stone (with diameter 20-
rounding landscape and to resist against the im-
200 mm) and gravel category (with diameter 2-
pact of heavy rains.
20 mm), the highest weight (over 90%) being
Soil cover is compulsory to work for at least
consisting in stones.
200 years. The first cover layer, which has the
Water samples collected downstream the
role of seepage barrier, will have a thickness of
waste dumps from Bǎiţa area have the following
0.5 m and will be compacted so that the perme-
contents of radioactive elements: U=0.085 mg/l;
ability coefficient to be equal or less than
Ra=1.7 Bq/L; Rn222=6 Bq/L.
5x10−9 m/s.
In controlled conditions, the plants sampled
The second layer, of seepage and storage,
from the land were kept in a climatic room for
will be covered by vegetation for assuring
growing plants - microclimate model. This room
evapo-transpiration and dump protection against
is presented in Figure 1 and has the following
erosion. The structure of cultures chosen will be
facilities:
according to the local pedo-climatic conditions.
- reproduces with high accuracy the climatic The experimental plants (Lotus corniculatus,
conditions corresponding to the day/night cy- Dactylis glomerata, Lolium perenne, Bromus
cle as concerns humidity, temperature and il- inermis, Festuca pratensis) behaved very well
lumination;
- allows research on plants in all their growing
stages: from seed germination to blossom.
As a result of the exploration geological re-
search and of the activity of uranium ore min-
ing, there is a big volume of sterile rocks which
occupy surfaces that require soil sustainable
shaping and rehabilitation.
Figure 2: Vegetation increase and development.

on the covered dump, assuring a high land cov- the dumped material. After finishing the geocell
ering degree and a good protection against ero- laying operation, the geocells were filled with
sion (Fig. 2). The best thing would be to culti- local soil. The last laid layer was the one with
vate them in mixture, the weight of each of biodegradable geotextile mattresses, pre-seeded
them being of 20%. with seeds of graminee family, Poa and Festuca
A possibility for stabilizing dumps would be sprinkled with sulphate reducing microorgan-
the use of seeded biodegradable geotextiles, isms, in order to reduce the concentration of ra-
namely Geocells, GWL 100 + Mattresses and dioactive ions accumulated in plants (Fig. 3).
Greenfix (200 m2+200 m2), Fibertex F330. The On the last layer which was laid, the left side
use of seeded biodegradable geotextiles for sta- was sprinkled with sulphate-reducing type mi-
bilizing dumps might eliminate the disadvan- croorganisms, in order to deal with the concen-
tages pointed out in the research on recultivation tration of radioactive ions and heavy metals ac-
of dumps using top soil and the acquirement of cumulated in the plants. The surface where the
agricultural products. solution of stabilization and environmental re-
Concerning the rehabilitation and utilization habilitation was experienced on the slope must
of the soils affected by the uranium industry, in be monitored in time, but, even from this stage,
our country, this problem is at present at the in- a conclusion may be drawn, namely: on steep
tention level, without concrete, significant solu- slopes of dumps, it is well to apply the solution
tions and results applicable so far. previously experienced with an addition to it,
First of all, the geotextile material was laid, i.e. supplementary planting of some young birch
with a role of separation, drainage and protec- bushes.
tion against the material dumped with a high For the final protection and rehabilitation so-
granulometry (crushed stone of 5-15 cm). Then, lution of the uranium waste dumps, taking also
the geocells of 10 cm (cell height) were laid into account the research carried out by
connected between them with specific links and INCDMRR in the last 5 years, in addition to the
anchored with iron anchors of about 0.5 m in previous solution, it is also compulsory to use a
a packet of geomaterials as follows: - Trisoplast
with thickness of 7 cm; geotextile for drainage;
geocells with height of 10-15 cm; biodegradable
geotextile and grass.
The results of the measurements carried out
after checking gamma debit dose on dump
slope, before and after covering with geotextile,
soil and grassing, are: 0.8-1.5 µSv/h before cov-
ering and 0.25-0.7 µSv/h after covering.
The monitoring programme will comprise:
collection of soil samples at well established pe-
riods of time (before planting, at 6 months, at 18
months, at 30 months after planting) namely
along at least 3 growing seasons; collection of
plant samples (sprouts and roots) - they will be
sampled at 6, 18, 30 months after planting; the
analysis of the samples collected with a view to
analyzing the concentration of contaminants and
establishing the retaining capacity of the plants;
assessment of the dump covering degree as well
as the development degree of the vegetal carpet
installed (AIEA, 1995).
4. CONCLUSIONS
Figure 3: Degree of vegetation covering with geotextile. As works of control regarding the pollution of
plants of Bromus Inermis Leyss cultivated on

the sterile covered with fertile soil and organi-
cally fertilized are low, not requiring restrictions
in using it as forage.
REFERENCES
Brown, L., 2001. Eco-Economie. Crearea unei economii
pentru planeta noastră. Ed Academiei.
Răuţă C. şi Cârstea St., 1983. Prevenirea şi combaterea
poluării solului. Editura CERES, Bucureşti.
Mullen, M.D., D.C. Wolf, F.G. Ferris, T.J. Beveridge,
C.A. Flemming and G.W. Bailey, 1983. Bacterial
sorption of heavy metals, Appl. Environ. Microbiol.
45 1094-1108.
Lazar, I., 2000. Biotehnologii. Industrii Extractive. Partea
II Minereuri,248 pag. Curs Fac. Biotehnologii, Univ.
de St. Agron. şi Med. Vet. Bucureşti.
AIEA - Safety Series 111 F - 1995. 1. The Principles of
Radioactive Waste Management. Vienna.
Figure 4: Example of gamma debit dose distribution
soil and running waters, there are applied bio-

remediation activities. The research pointed out
that in order for radioactive sterile to be reculti-
vated and to avoid the translocation in the plant
of undesired elements, it is compulsory cover it
with with fertile soil with a thickness of, at least,
20 cm that needs to be organically fertilized and
mineralized as well.
Such steriles were covered by vegetal car-
pets, which not only stop and diminish noxious
radiations but offer also a nice integration into
the landscape in the area of Bihor Mountains.
The measurements obtained in situ for de-
termining gamma debit dose, showed a reduc-
tion of its value from 0.8-1.5 µSv/h to 0.25-
0.7 µSv/h (Fig. 4).
The biosorption tests carried out for U recov-
ery by means of the biomass immobilized in
polyester foam were performed on an uranifer-
rous effluent with a U content of 1g/L. Sul-
phate-reducing bacteria (BSR) role in the epura-
tion process of the water from the pond with
pH=8, with a content of 0.217 mg/l U, 15 mg/L
Mo, 1350 mg/l SO42- was studied by passing the
residual effluent through porous materials on
which BSR (Desulfovibrio and Desulfotomacu-
lum) are immobilized.
The concentrations of heavy metals in the
Speciation of mercury in the environment affected by industrial pollution:

determination and modelling
E.M. Cukrowska, J.L. Makiese, H. Nsengimana and H. Tutu

School of Chemistry, University of the Witwatersrand, Johannesburg, South Africa
D. Amouroux and E. Tessier

Laboratory of Bioinorganic and Environmental Analytical Chemistry, CNRS, University of Pau,
France
ABSTRACT cation in the food chain (Bloom and Watras,

1989).
Mercury concentrations have increased in the
The determination of the total concentration
world as result of industrial activities. It has
of mercury is not sufficient to understand its
been published that mercury emissions are
fate in the environment (Brosset and Lord,
mostly due to coal burning power stations,
1995). The development of a sensitive, reliable,
chlor-alkali facilities, mining and other indus-
simple, and cost effective procedure for speci-
trial processes.
ation analysis of mercury in diffrent environ-
The objectives of the present study were to
mental compartments is currently one of the
develop and optimise analytical procedures for
principal research chalenges in environmental
mercury speciation analysis and their applica-
analytical chemistry (Hoenig, 2001).
tions for analysis of water, sediment profiles,
soils, plants, fish and human hair in the regions
affected by environmental pollution due to re- 2. ANALYTICAL PROCEDURES
processing of old gold tailings dams and chlor-
alkali factory. The study included collection of 2.1 Sampling
ancillary data (pH, redox potential) which are Water sampling: Bottles were rinsed under wa-
critically important for mercury monitoring pro- ter 3 times. Samples were collected the same
gram. way filling them without any headspace and
The development of predictive models based closed still in the water. Samples were acidified
on thermodynamic solution equilibria including at 1% (v/v) with ultra pure HCl acid on the
mercury tranformations, transport, fate and bio- field. Bottles were stored in double plastic bags,
logical uptake are presented. Sediments samples: Fine grained sediments
(clay and silt) without debris were collected,
thus no sieving was necessary. The sediments
1. INTRODUCTION
(auger profiles) were collected using an acid
Mercury is present in the environment in dif- wash plastic cup and transferred into clean PP
frent molecular forms with specific biogeo- vessels. They were transported in a cool box and
chemical transformation and ecotoxicity. Inor- rapidly frozen in a laboratory.
ganic Hg2+ is the main form in water and sedi- Biological samples (fish, plants): The most im-
ment samples. Special attention is given to portant for Hg is to collect muscles from the
transformations which lead to the in situ produc- fish.The dissection of the fish muscles was done
tion of methylmercury (Harris el al., 2007). with clean stainless steel tools (rinsed). They
Concentration levels of organomercury species were stored in clean plastic ware (PP washed
is very low (usualy ng L-1) in environmental with concentrated acid, rinsed with MQ water
aquatic environments but the toxic effect of and finally dried) and rapidly deep frozen (ca -
these compounds can be significant due to their 20°C).
tendency for bioaccumulation and biomagnifi- Sediment, fish and cleaned plant samples
Table 1: Recoveries for certified reference materials. ing have been identified (Chamber of Mines,
2004). Mercury always accompanies gold ores.
Amalgam method for gold extraction was used
from 1850 till 1915. Reprocessing of old tail-
ings dams exposes reduced material to oxidative
environmental conditions and UV radiation and
creates massive leaching of metals (Dabrowski
el al., 2008).
were dry freeze (lyophilized) under vacuum at - Sampling area
50°C. The dry powders were stored in clean
plastic ware at room temperature in the dark at The Greater Johannesburg occupies the area of
40C prior to analysis. the watershed between the Crocodile River
catchment on the north and the Klipriver catch-
2.2 Analytical protocols ment on the south.
An analytical method for simultaneous in situ The southern part is the area of heavy indus-
ethylation, of organotin and organomercury trial activities with predominantly gold mines
compounds in water, sediment and plant sam- and resulting tailings dams.
ples was developed (Cukrowska et al., 2009). The Klipriver catchment ends further south
The determination of mercury and tin com- in the Vaal Dam which provides Drinking Water
pounds was achieved by species-specific isotope for the Greater Johannesburg Metropolitan Area
dilution, derivatisation and gas chromatography with few million population (Fig. 1).
- inductively coupled plasma mass spectrometry
(GC-ICP-MS) detection. The recovery and ac-
curacy of the method were confirmed by the
sediment and plant certified reference material The first set of samples was collected at the end
(IAEA 405 for sediment and CRM 279 for of wet, hot season. The results for water and bi-
plant) (Table 1). ota are presented in Table 2. The accumulation
of mercury in the wetland is evident with high
concentration in biota (algae, water reeds). The
3. SAMPLING wetland conditions with high level of rotten or-
Gold mining resulted in large gold tailing dams ganic substances are also supporting the forma-
south of Johannesburg. More than 270 tailing tion of organomercury.
dams (total area = 180 km2) related to gold min- The sediments profile (Fig. 2) is showing that
Figure 1: Study area.

Table 2: Speciation of mercury in water, algae and plants. rains. Also the main level of organomercury
formation was lower due to lowering of anaero-
bic redox conditions.
5. CONCLUSIONS
The reprocessing of old tailings is creates a
massive leaching of metals to the environment.
Wetlands usually serve as filters for this kind of
pollution but in case of mercury they provide
very good conditions for a formation of ex-
tremely toxic organomercury compounds.
REFERENCES
Bloom, N.S. and C.J. Watras, 1989. Observations of me-
thylmercury in precipitation, Science of the Total En-
vironment 87/88, pp. 199-207.
Brosset, C. and E. Lord, 1995. Methylmercury in ambient
air. Method of determination and some measurement
results, Water Air and Soil Pollution 82, pp. 739-750.
Chamber of Mines, 2004. Facts and Figures. Available
from: www.bullion.org.za.
Concentration are μg L-1, Concentration in μg kg-1 dw for
Cukrowska, E.M., H. Nsengimana, A. Dinsmore, E.
plants. K 1-4 surface water, GW ground water, P 1-4
Tessier and D. Amouroux, 2009. In situ ethylation of
Plants; % RSD in brackets.
organolead, organtin and organomercury species by
bromomagnesium tetraethylborate prior to GC-ICP-
organomercury is produced from accumulated MS analysis J.Sep.Sci. 32 - in print.
inorganic mercury at the level where redox po- Dabrowski, J.M., P.J. Ashton K. Murray, J.J. Leaner, and
tential is around zero volts. R.P. Mason, 2008. Anthropogenic mercury emissions
The second set of samples was collected at in South Africa: Coal combustion in power plants,
the end of cold dry season. In this case the con- Atmospheric Environment 42, pp. 6620-6626.
Harris, R., D.P. Krabbenhoft, R. Mason, M.V. Murray, R.
centrations of mercury were much higher in the Reash, and T. Saltman, 2007. Ecosystem Responses to
wetland not being diluted and washed out by Mercury Contamination, Indicators of Change, SE-
Concentration of MMHg (ng g-1) TAC, New-York, 2007.
Hoenig, M., 2001. Preparation steps in environmental
0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 trace element analysis - facts and traps, Talanta 54, pp.
0-30
1021-1038.
MMHg
ORP
IHg
31-60
Depth (cm)
60-85
ORP
-150 -100 -50 0 50 100
85-100
0 50 100 150 200 250
Concentration of IHg (ng g-1)
Figure 2: Speciation of mercury in sediment profile.
Analysis of water and geothermal-well shallow drilling data via drilling

software allows for rock drillability assessment and drill bit performance
P. Dalamarinis, V.C. Kelessidis, V. Chatzistamou

G. Karydakis
Institute of Geology and Mineral Exploration
M. Chlaboutakis
Manager Geothermy, Public Power Corporation Renewables
ABSTRACT exploitation potential and to produce them.

Drilling, however, can be very expensive.
Drilling maybe the most expensive part of an
Hence, great effort has been devoted in the past
exploration campaign for production of miner-
to develop safe and optimum drilling practices
als, geothermal fluids, water and hydrocarbons.
to reduce the amount of time spent on drilling.
Significant advances have been made on devel-
At the same time, when planning for the next
oping drilling bits and equipment as well as
drilling campaign, it is very beneficial to have
drilling techniques, with hydrocarbon exploita-
the ability to predict drilling rates, given the
tion industry leading the way but no such ad-
field conditions, so that a comprehensive explo-
vances have been made on modeling rock-bit
ration and development plan is developed and
interaction. Some models have appeared for oil
implemented.
industry but with limited success, while no such
Several simulators have been developed in
models have been required for surface drilling,
the past which allow prediction of drilling rates
such as water or shallow geothermal well drill-
with one of them deserving particular attention
ing, because drilling conditions are not so diffi-
as it has proved that it could simulate with fair
cult. Water-well depths, however, have been in-
degree of accuracy hydrocarbon well drilling.
creasing in recent years, and a closer look into
This is the Payzone simulator, developed by
drilling data is needed for choosing better drill
Prof. Cooper of University of California at
bits and implementing optimum drilling condi-
Berkley (Cooper et al., 1995, 1996) which has
tions. An oil-well drilling simulator has thus
proven that it could perform well under given
been tested and used to simulate drilling data
circumstances (Abouzeid and Cooper, 2003;
from two shallow water wells from USA and a
Kelessidis and Dalamarinis, 2009). The particu-
geothermal well in Greece. The data is analyzed
lar simulator has a basic drill bit - rock interac-
with respect to the given lithology, drill bit and
tion model and allows for a few adjustable pa-
bit wear and drilling conditions (weight, torque,
rameters which enable fine tuning of the simula-
fluid flow. The results show that if we have ac-
tor with field data. Besides the capabilities, al-
cess to drilling data from the region, such a tool
though not so advanced, which give a picture of
could be very useful for planning new wells, for
the well drilled and present also the lithology
choosing the right drill bit and right drilling
and drill time log, the main feature is the predic-
conditions, thus reducing costs in an exploration
tion of drilling rate, using Eqn. (1),
campaign.
ROP = ( flow _ factor )(C )(RPM )(tooth _ length )(G ) (1)
1. INTRODUCTION where
Drilling for oil, geothermal or water and mineral
resources is the significant step to uncover sub-
surface resources, to determine their economic
⎡ ⎛ WOB ⎞ curv ⎤ which was drilled in the island of Milos, Greece

⎢− ⎜ ⎟ * ⎥ in the period November 1980 to March 1981
⎢ ⎝ UCS ⎠ ⎥
G = 1 − exp ⎢ ⎥ (2) (Chlaboutakis, 2009). Two sections were simu-
⎢⎛⎜ 12 ⎞⎥
⎟ lated. The first one was in the non-producing
⎢⎜ D 2.5 (0.4 * tooth _ length ) ⎟⎥ formation, between 500 m and 814 m and has
⎣⎝ ⎠⎦
been simulated with 6 intervals. The lithologies
In the equations above, RPM is rotary speed, with their UCS values, taken from Hoek et al.
tooth_length is the average length of the bit (1998), are shown in Table 1. The second inter-
teeth, UCS is the unconfined compressed stress val is in the producing zone, between 1040 m
of the rock, D is the bit diameter, WOB is the and 1180 m, has been simulated with one sec-
applied weight on the bit; the following modifi- tion, with the characteristics shown also in Ta-
able by the user constants are used: C is a con- ble 1. The values of UCS are very important for
stant; (aggressivity) is a formation and bit char- successful simulation. They were modified ac-
acteristic constant ranging between 20 % and cording to the degree of metamorphism of these
100 % and normally is given the value of 35 %; rocks, based on the descriptions from the well
(curv) is a formation - WOB interaction con- report. Thus, formations containing quartz,
stant and it is usually given the value of 1.5; alvite, titanite, chlorite and having high degree
(flow_factor) is a constant, ranging between of metamorphism were assumed to be on the
50 % and 100 % and defines the capability of high end of the UCS range for the particular
the system to adequately clean the bit front by formation. For example, schist has a range of
the cuttings. The above equation is similar in UCS values between 30 to 60 MPa; for the in-
predictions to the equation suggested by Teale terval 555 to 600 m, the rock is described as
(1965) where the specific energy was defined chloritic, calcitic schist with chlorite, hence, it
(Kelessids and Dalamarinis, 2009). has been given the UCS value of 53 MPa.
Using the adjusted with field data Payzone Bit wear was also simulated in the production
simulator, a new campaign, in a similar field, or section to determine its effect on the rate of
development drilling practice scenarios in the penetration. This was accomplished using the
same field could be designed. The particular parameters rock abrasivity and wear factor in
data needed is operational and formation data. the metamorphic rocks so that it is similar to the
The former include, weight on bit (WOB), rota- wear indicated in the well report. Finally, the
tion rate of the drill string (RPM), flow rate and
fluid parameters, drill bit parameters like bit Table 1: Lithology description and UCS values.
type, bit make, types of nozzles, and the bit re- Interval UCS
cord (depth in, depth out and wear condition at Formation
(m) (MPa)
the end of bit cycle); the latter include, well ge- Mica Schists with epidote,
ometry and formation parameters, like lithology 500-525 quartz, albite. Low metamor- 30
and estimations of unconfined rock strength phic grade
(UCS). Calcitic schist with chlorite,
The particular simulator has not been tested, 525-540 epidote, titanite, . No metamor- 35
till today in drilling situations for geothermal phism
wells and for the much shallower water-wells Chloritic, talk, calcitic schist
540-555 40
(Cooper, 2009). It is the intention of this paper with albite. No metamorphism
to demonstrate the use of Payzone for geother- Chloritic, calcitic schist with
mal drilling and water well drilling, to verify 555-600 albite, quartz, quartzite veins. 53
that its capabilities are also well within the Low metamorphism
range of shallow drilling. Mica schist, chloroschist with
chlorite, epidote, calcite, ti-
600-814 56
tanite, graphite. High metamor-
2. METHODOLY AND DATA REDUCTION phic grade.
2.1 Geothermal Well Mica Schist, chloroschist with
chlorite, epidote, calcite, ti-
Data for the simulated geothermal well were 1040-1080 60
tanite, graphite, pyrite, amphi-
taken from the drilling report of well Milos-1, bole. Total metamorphosis.
third adjustable factor ROP_factor of the bit has

been further modified to match real drilling
data.
2.2 Water well
Water well data usually do not have complete
bit records nor do they have good monitoring
logs, as the relatively deeper geothermal wells
and of course hydrocarbon wells do. Some wa-
ter well reports do not even state the bit types
they have been using. An extensive biblio-
graphic search has been performed to gather
data with minimal required information in order
to run the simulator. We have used two water
wells with reports from USA. Nevada-1 well Figure 2: Simulation of 500 to 814 m, real well (blue
(Bechtel, 2005), with simulation of 0 to 760 m, curve #2) and with an increase in drilling parameters by
with 25 lithology intervals consisting of various 50% (red curve #2). Lithology is limestone and metamor-
tuffs and breccia, simulated with UCS values phic rocks.
between 15 and 50.0 MPa and 67.0 MPa respec-
tively; and Nevada-2 well (Mace, 2009), with as: WOB ranging between 3-6 tons, RPM be-
simulation of the section between 329 and tween 50 and 65 RPM and flow rate between
378 m, consisting of one limestone interval with 1800 and 2800 Lpm. When simulating all three
UCS value of 60 MPa. In both wells, foam was intervals with the real data, simulation is very
used as drilling fluid, as is the usual practice in good, with almost one to one correspondence,
many areas for water-well drilling. The simula- indicating the good approach of the simulator.
tor does not simulate foam drilling; hence, the When all significant drilling parameters (WOB,
flow factor has been adjusted to one, thus as- RPM, flow) are increased by a factor of 50 %,
suming full cleaning capabilities of the drilling then a decrease in drilling times is observed of
fluid. The intention was to see whether we can approximately 170 % for each interval.
simulate the drilling parameters in water well Bit wear does not seem to be a factor when
drilling and not the flow parameters. increasing drilling parameter values and in par-
ticular the WOB. We have attempted also an in-
crease in the UCS value by 50 %, which gave an
3. RESULTS - GEOTHERMAL WELL increase in the drilling times, for all three inter-
The results of simulating the 500-814 m interval vals, around 50 %.
section of the Milos-1 geothermal well are The results of drilling simulation for the pro-
shown in Figures 1 and 2. There were three 12 ducing interval of 1040 to 1180 m is shown in
¼-in. milled tooth drill bits used in the three in- Figures 3, 4 and 5. The six drill bits used were
tervals simulated, with drilling parameter values
Figure 3: Simulation of interval 1040-1180 m, of real

Figure 1: Simulation of interval 500-814 m, of real drill- drilling conditions (real data) and of increasing drilling
ing conditions (standard) and of increasing all drilling pa- parameters (all up) as follows: RPM’=RPM+20,
rameters (RPM, WOB, FLOW) by 50% (all up). WOB’=1.5*WOB, FLOW’=FLOW+150 GPM.
each interval, which ranges between 28 and

52 %.
Comparison of bit wear condition, listed in
well report, versus the bit wear condition from
Payzone has indicated that bit wear was also
well simulated. It is also evident from Figure 4
that increasing all parameters to the values indi-
cated, additional bit wear is occurring, seen as
the curved section of curve #2, which of course
is not desirable. Thus, not only good simulation
of rate of penetration is needed, but also good
simulation of the bit wear is necessary in order
to fully model the process. Thus in real life
situation, the driller would proceed to increase
drilling parameters to lower values than the ones
chosen in the example of Figure 4, for the first
Figure 4: Simulation of 1040 to 1180 m, real well (red
curve #1) and with an increase in drilling parameters
three intervals, where significant curvature is
(blue curve #2). Lithology is metamorphic rocks. observed. In the last three intervals the values
used were good because the curve is almost par-
allel to the original (curve #1).
The results of Figure 5 indicate that if UCS
values were merely 50 % higher than the one
anticipated or used, in all six intervals, trying to
simulate the uncertainty of the knowledge of
this value (Kelessidis, 2009), drilling times may
increase significantly, ranging between 26 to
150 %. Furthermore, wear is very significant.
This was not a problem in the real situation, and
thus, one should know with better accuracy the
UCS value for the given formations.
4. RESULTS - WATER WELL

The drill time plot of the Nevada-1 well (Bech-
Figure 5: Simulation of 1040 to 1180 m, real well (blue

curve #2) and with an increase in UCS by 50% (red curve
#1). Lithology is metamorphic rocks.
tricone milled tooth bits, 8 ½-in. Drilling pa-

rameters were, 6 ton WOB, 40 to 60 RPM and
1600 to 2400 Lpm flow rate. From Figure 3 we
see that in the producing interval, real data has
been well simulated. In fact there are six inter-
vals, but in Figure 4, only 3 point is seen, be-
cause of the coincidence of values.
In Figure 4 we see that we used six drill bits.
Curve #2 is the simulation when increasing all
drilling parameters, as shown in the caption.
When introducing the increase of the WOB by
50 %, of RPM by 20 RPM and of flow by Figure 6: Drill time plot of Nevada-1 well (Bechtel,
520 lpm, there is a reduction in drilling time for 2005).
ond well, Nevada-2, with the results of real

drilling data and Payzone simulated data with an
increase in the parameters shown in Figure 8.
One can see, as in the case of the geothermal
well, shown above, but also for the case of hy-
drocarbon wells (Kelessidis and Dalamarinis,
2009), that increasing WOB and RPM has a
positive effect on the reduction of drilling times,
thus proving that the simulator can be used for
future well planning.
Figure 7: Comparison of real drilling time and Payzone
drilling time for Nevada-1 well.
5. CONCLUSIONS
A hydrocarbon drilling simulator has been tried
to test its capabilities for predicting drill times
as well as bit wear for geothermal wells and
much shallower water wells.
It was possible to fine tune the simulator for
the geothermal well, using field data from
Milos-1 well, drilled in 1980s. Tuning of the bit
wear was also possible. Testing the effects of
drilling parameters (WOB, RPM, flow rate) and
of formation parameters (UCS) is possible with
good degree of confidence thus providing an op-
timization tool.
Simulation of two much shallower water
wells was also possible. Good matching of real
data with simulator data has been achieved,
without taking into account the effect of flow
rate, because the wells have been drilled with
foam. Optimization of drilling parameters is
Figure 8: Simulation of water well, between depths of 329 possible for water well drilling, as it has been
to 378 m. Real well drilling (blue curve #1 with, demonstrated with the simulation of Nevada-2
WOB=20klbf, RPM=35, and UCS=60 MPa). Simulated well.
curve (Red curve #2 with optimized drilling procedures,
WOB+50%, RPM=55).
REFERENCES
tel, 2005) is shown in Figure 6. Four milled-
tooth bits were used, two 17 ½-in. in the first Abouzeid, A.A. and G.A. Cooper, 2003. Experimental
two sections up to 346 m and two 12 ¼-in to the verification of a drilling simulator, paper presented at
the 8th International Conference on Mining, Petro-
depth of 762 m. leum and Metallurgical Engineering, Suez Canal Uni-
In Figure 7, the comparison of real time with versity, Egypt, 17 - 19 March.
Payzone time versus depth is shown. The match Bechtel Nevada, 2005. Completion Report for Well ER-
is very good, of course using the adjustable pa- EC-5, US Dept. energy report, DOE/NV/11718-424.
rameters of the simulator. But this proves that, Chlaboutakis, M., 2009. Personal Communication.
Cooper, G.A., A.G. Cooper, G. Bihn, 1995. An interactive
given the field data, the simulator can be tuned drilling simulator for teaching and research, paper
to reproduce them, and thus it can be used in the SPE 30213 presented at the Petroleum Computer Con-
next design phase. ference, Houston, TX, 11-14 June.
This is true even for the shallower water Cooper, G.A., J.F. Mota and A.G. Cooper, 1996. Inte-
wells, hence, one can determine the effects of grated petroleum engineering simulation and decision
altering drilling parameters thus optimizing making teaching program, paper SPE 36660 presented
at the SPE Annual Technical Conf, Denver CO, 6-9
drilling practices. Oct.
Such an attempt has been made with the sec- Cooper, G.A., 2009. Personal Communication.
Hoek, E., P. Marinos and M. Benissi, 1998. Applicability

of the Geological Strength Index (GSI) classification
for very weak and sheared rock masses. The case of
the Athens Schist Formation. Bull. Engr. Geol. Env.
57, 151-160.
Kelessidis, V.C., 2009. Need for better knowledge of in-
situ unconfined compressive strength of rock (UCS) to
improve rock drillability prediction, Paper to be pre-
sented at the 3rd AMIREG Conference, Athens, 7-9
Sept.
Kelessidis, V.C. and P. Dalamarinis, 2009. Monitoring
drilling bit parameters allows optimization of drilling
rates, Paper presented at the International Multidisci-
plinary Scientific Geo-Conference & EXPO, Modern
Management of Mine Producing, Geology and Envi-
ronmental Protection SGEM 2009, Albena, Bulgaria,
June 14-19.
Mace, J., 2009. Using Borehole Geophysics as a Tool to
Determine Potential Productive Zones of Test Wells in
Eastern Nevada, Paper presented at the Nevada Water
Resources Association 2009 Annual Conference,
Reno, Nevada, February 3-5.
Teale, R., 1965. The concept of specific energy in rock
drilling, Int. J. Rock Mechanics and Mining, 2, 57-73.
Challenges and opportunities for beneficial reuse of large volume wastes in

the USA
W. Lee Daniels
Dept. of Crop & Soil Environmental Sciences, Virginia Tech, Blacksburg, Virginia, USA
ABSTRACT maximum soil contaminant levels (e.g. total As

at 0.39 mg/kg) that are considerably below natu-
Large volumes of residual wastes such as coal ral soil and sediment background concentrations
combustion products (CCP’s), dredge sediments (5 to 8 mg/kg). Historically, up to 50% of mu-
and sewage sludge (biosolids) are available for nicipal sewage sludge (biosolids) generated in
beneficial use as soil amendments, manufac- the USA has been land-applied to agricultural
tured soil components or structural fills in the lands. However, similar to CCP’s, mounting
USA, but their utilization is being increasingly public concerns over pathogen regrowth follow-
challenged by a combination of regulatory and ing application, trace organics with hormonal
public acceptance barriers. Over 100 M tonnes activity, and loss of bioaerosols are limiting
of fly ash + flue gas desulfurization (FGD) land application in many states. Interestingly, all
sludge are produced annually and over 40% of of three of these large volume residuals have
the fly ash is recycled annually into cement or been used with great success alone and in vari-
block while over 30% of the FGD is converted ous combinations for mine reclamation and for
into wallboard. Over the past decade, increasing remediation of a wide array of contaminated
amounts of both materials have also been recy- lands. For a combination of reasons, including
cled as structural construction fills and for mine the presence of established monitoring arrays,
reclamation. However, mounting public and potential for large volume applications, and bet-
agency concerns over the potential release of ter public acceptance, these disturbed areas are
toxic elements to the environment is currently emerging as the preferred utilization environ-
reducing recycling efforts and beneficial use. ment for all of these wastes.
These concerns have been amplified by a series
of highly publicized issues including a fly ash
retention dam failure, alleged water quality
damage from structural fills and gaseous H2S
emissions from imported drywall products.
Similarly, over 200 M m3 of dredge sediments
are removed from USA waterways annually and
less than 5% are used beneficially. This is de-
spite that fact that major projects in Illinois and
Virginia have recently clearly demonstrated the
potential of dredge sediments for use as soil
covers at former industrial sites and degraded
mines. One major issue is the almost complete
lack of state regulatory programs and standards
for such uses. Other major constraints include
current federal health based risk screening crite-
ria for soils at remediation sites that specify
Criteria for beneficial utilization of dredge sediments in Virginia, USA
W. Lee Daniels, A. Wick and N. Haus

Virginia Tech, Blacksburg, Virginia, USA
G.R. Whittecar
Old Dominion University, Norfolk, Virginia, USA
C. Carter III
Weanack Land LLLP, Charles City, Virginia, USA
Upland placement of dredge sediments is fa- River and harbor dredging activities generate
vored over bulk disposal options in the USA for hundreds of millions of m3 of dredge materials
many projects where contaminant levels meet in the eastern USA annually, and disposal op-
beneficial utilization standards. Since 2001, we tions are becoming increasing limited and ex-
have placed over 350,000 m3 of fresh water pensive. While certain sediments are heavily
dredge materials (Woodrow Wilson Bridge Pro- contaminated (USEPA, 2005), much of the ma-
ject -WWB) and 250,000 m3 of saline materials terial is quite suitable for placement into upland
from the Earle Naval Weapons Station (N.J.) environments for conversion to topsoiling mate-
into monitored upland placement cells and rials for mining and other disturbed sites (Dar-
documented their conversion to agricultural mody and Marlin, 2002; Darmody et al., 2004;
uses. Detailed groundwater and soil quality Lee, 2001). The basic chemical and physical
monitoring has indicated no adverse effects properties of these dredge sediments vary
from material placement and outstanding agri- widely based on their depositional environment
cultural productivity for the freshwater WWB and watershed characteristics and history. For
materials. Our combined experience to date in- example, exposure and weathering of highly
dicates that an important primary screening pa- sulfidic dredge sediments produces extremely
rameter is acid-base accounting which should acid soil conditions and metal release (Fanning
become a mandatory analytical requirement. and Fanning, 1989), while any materials re-
Our second level of screening is based upon a moved from marine or brackish environments
combination of USEPA risk-based soil screen- will necessarily contain entrained salts and Na
ing levels and related New Jersey standards. that will need to be leached before their conver-
High silt+clay and TOC may also eliminate sion into viable topsoiling materials. Similarly,
many dredge materials for physical and logisti- dredge materials that are too high in fine
cal reasons. This new dredge material screening silt+clay and organic matter may be difficult to
system separates potential upland placement handle, place and till in an upland environment.
candidates into two quality levels with differing However, large volumes of non-sulfidic poten-
management and monitoring requirements. Fi- tially suitable materials are routinely dredged
nally, it is also clear that whenever dredge mate- and disposed of annually, and the federal Inter-
rials from a saline environment are utilized for agency National Dredging Team (USEPA,
upland placement, that salinity and Na will limit 2003) has placed a high priority on moving suit-
plant productivity and soil quality for some pe- able materials to upland beneficial use environ-
riod of time, and that appropriate management ments rather than into disposal impoundments.
practices and/or extended leaching intervals will In this paper we will review results obtained
be necessary. from two large upland dredge placement pro-
jects in Virginia and summarize our new regula-
tory proposal for the screening and classifica-

tion of dredge materials for upland placement
and agricultural use.
2. THE WOODROW WILSON BRIDGE PRO-

JECT WITH FRESHWATER SEDIMENT
The construction of the Woodrow Wilson
Bridge (WWB) replacement spans across the
Potomac River just south of Washington D.C.
excavated approximately 500,000 m3 of fresh-
water tidal dredge sediments between 2000 and
2005. Extensive pre-excavation testing indicated
that the sediments were very low in organic
contaminants and metals, relatively low in or-
ganic matter (< 5% TOC), loam to silt loam in
texture, and moderate in pH. All pesticides, her-
bicides, and other anthropogenic organic com-
pounds were either non-detectable or well be-
low USEPA Region 3 Risk Based Criteria for
residential soils (Daniels et al., 2007)
Weanack LLLP worked cooperatively with
Virginia Tech, Old Dominion University, and Figure 2: Profile of oxidizing dredge soil in April 2002,
Potomac Crossing Consultants to develop an less than 18 months after placement.
upland beneficial use permit structure to use
these materials to construct an agricultural soil cut out to enlarge the capacity of the area and
landscape on former sand and gravel mined the cut spoils were used to construct a 3 to 6 m
lands adjacent to Shirley Plantation in Charles dike around the facility to avoid any loss of
City County, Virginia. The majority of materials sediments to surface waters during operations.
were moved by barge to Weanack in 2000 and 2.1 Agricultural Soil Development
2001 and then off-loaded at its port facility with
a clamshell loader into haulers. The materials The fresh dredge materials were approximately
were then placed into the 20 ha utilization cell 35 to 45% solids when placed in 2000 through
(Fig. 1) and allowed to dewater and consolidate. 2002 and were highly reduced and anaerobic
The upland utilization cell was on a former sand (Fig. 1). Over the course of one year, however,
& gravel mine. The pre-existing mine soils were the materials stabilized and settled to some ex-
tent, and deep polygonal cracking was observed.
The polygons were initially 20 to 50 cm in di-
ameter at the surface, the cracks were 2 to 8 cm
wide, and > 25 cm deep by the summer of 2002.
Observations in the fall of 2002 indicated that
the material was oxidized along cracks to a
depth of > 50 cm, but reduced materials within
polygon/prism centers were observed at 25 cm.
We excavated several soil pits in 2002 (see
Fig. 2) and the surface layer was clearly becom-
ing more uniformly oxidized with time and the
surface became better aggregated. Auger bor-
ings also revealed relatively unconsolidated
“soupy” materials that occurred at 2.0 m+ that
were still strongly reduced and low chroma
Figure 1: Placement of WWB dredge spoil on old sand
and gravel mined lands. (blue-gray color). Total subsidence of the re-
sented in Table 1. Simply put, the corn yields

recorded on the dredge spoil materials in 2002
were outstanding. This region suffered a severe
regional drought in 2002, although a few sum-
mer thunderstorms did provide adequate mois-
ture at critical times for ear filling and grain de-
velopment. Apparently, the corn was able to
root down through the dewatering dredge spoils
(through the deep desiccation cracks discussed
earlier) to tap into the wetter dredge materials at
depth.
While corn grain yield did appear to increase
with compost loading rate up to 112 Mg/ha,
Figure 3: Row crop production trial plot area in the fall of within-plot variability was high, and no signifi-
2001. Multiple crops of wheat and corn were established cant compost rate effects were noted. However,
and grown through 2006. even the 0 compost treatment yields were well
above any 2002 yields on surrounding farms
search plot area is estimated to be 10 to 15% (of that we queried. In contrast to 2002, 2003 was
originally placed 5 to 7 m of dredge) to date, an exceedingly wet year, and we believe that
and further settlement is expected over time contributed to overall yields being lower than
once these deeper layers dewater and consoli- the high yields observed in 2002 (Table 1). It is
date fully. Despite the significant oxidation of also important to note that we added no N-
reduced Fe and Mn species, the pH of the sur- fertilizer to these plots, even though the heavy
face layers has remained > 6.5 to date at all lo- crop the preceding year certainly extracted large
cations and averaged > 7.0. The inbound dredge amounts of plant available N. Nitrogen avail-
sediments had an average calcium carbonate ability is generally the major predictor of corn
equivalence of > 2.5% and were very low in sul- yield when rainfall is not limited, so we are
fides, so this outcome was expected. quite impressed by this overall yield response,
2.2 Crop Production Trials particularly from the 0 compost control plots.
The lower yield at the highest compost loading
By the fall of 2001, the sediments had dewa- rate was most likely related to excess moisture
tered, cured and oxidized sufficiently to allow holding in the very large amount of organic
for cultivation (Fig. 3). Between 2002 and 2005, matter added. Crop yields in 2004 through 2008
a series of row crop production trials was estab- on the WWB materials were consistently higher
lished and run in a replicated design that is de- than local county average yields for prime farm-
scribed fully by Daniels et al. (2007). lands.
The row crop trial plots received varying
rates of yardwaste compost addition and very 2.3 WWB Summary Conclusions
minimal applications of P and K fertilizers be- Based upon these documented chemical, physi-
tween 2002 and 2005. No lime or N additions cal and morphological properties, and the com-
were made. Corn (Zea mays L.) yields are pre- bined crop yield data, we are convinced that
these newly deposited “dredge soils” may be as
Table 1: Corn yield in September 2002 and 2003 at productive as any natural soils in the region.
Shirley Plantation/Weanack.
From a narrow standpoint of bulk soil chemical
Treatment kg/ha properties and fertility, these newly developed
Compost Mg/ha 2002 2003 soils are outstanding and actually superior to
0 13,090 a 7,350 a most native agricultural soils. The pH of the sur-
56 15,820 a 7,910 a face soil ranges from 6.8 to 7.4, despite over
112 16,100 a 7,910 a
five years of oxidation and weathering. Thus, on
224 13,790 a 8,190 a
a net basis, this project replaced low productiv-
336 14,630 a 4,900 b
*Yields in a given year followed by different letters are
ity reclaimed mined lands with high productiv-
significantly different at p < 0.05). ity prime farmland.
3. EARLE SALTWATER SEDIMENTS water quality degradation. One alternative solu-

tion would be to bulk mix agricultural limestone
In 2005 and 2006, approximately 250,000 m3 of
with the inbound sediments (e.g. at 1.5 to 2.5%
saline dredge materials from the Earle Naval
by weight) but this cost would need to be ac-
Weapons Station (N.J.) were placed into a new
counted for and the logistics of this process with
upland placement cell directly to the east of the
wet/soupy sediments are challenging.
WWB materials. These materials were placed
The beneficial use potential of any given
over eroded low productivity farmland and con-
sediment for agricultural uses is also directly
tained within a 2.5 to 4 m high dike and moni-
limited by particle size and organic C content.
tored over time to document their conversion to
In general, when the silt+clay content exceeds
agricultural use. The Earle sediments contained
70%, the materials will be very difficult to man-
several PAH’s (e.g. Benzo(a)pyrene) slightly
age from a tillage perspective, regardless of
above USEPA residential use standards and
their bulk chemical properties. Similarly, when
were higher in TOC and silt+clay than the
the total organic C (TOC) content exceeds 5 to
WWB sediments. Full characterization data on
10% (depending on texture), short-term dewa-
these materials are provided in Daniels and
tering is complicated and physical ripening
Whittecar (2006). As expected, relative to the
processes are greatly slowed.
WWB sediments, the rapid conversion of these
materials to agriculture has been limited by en-
trained salts (Cl and SO4) and Na. Regardless, 5. PROPOSED DREDGE SEDIMENT CLAS-
approximately 60% of the sediment surface was SIFICATION FRAMEWORK
supporting vigorous vegetation by mid-2009
and certain areas were supporting agricultural In association with the work described above,
crops. Current studies are detailing salt leaching we have focused a large collective effort on de-
rates within and under the dewatering sediments veloping a better working knowledge of the ap-
and effects of sand, topsoil and compost proaches used by other USA states and federal
amendments. agencies for screening sediments and solid
wastes for upland use. As a result of these com-
bined efforts, we have developed a proposed set
4. SULFIDES, TEXTURE AND CARBON of sediment screening criteria which are based
Due to the well publicized success of the WWB on a combination of (A) soil screening levels
sediment project, we were asked to review, test used by USEPA (2008) for setting cleanup lev-
and consider many (10+) differing fresh- and els for contaminated sites, (B) upland dredge
saltwater dredge materials for inclusion in our spoil placement criteria developed by the New
cooperative upland placement program. The Jersey Department of Environmental Protection
typical analytical package provided to us in- (1997), and (C) modifications by our group to
cludes total metals, nutrients, herbi- account for agronomic plant growth needs and
cides+pesticides and a wide range of organics certain nonsensical standards as discussed be-
low. This classification system has been pro-
(e.g. PAHs, PCBs and Dioxins). However, we
are convinced that the most essential analysis posed to the Virginia Department of Environ-
mental Quality for adoption. An example of our
that must be conducted on all materials is a sim-
system is presented in Table 2 which is drawn
ple acid-base accounting for acid-forming po-
tential similar to that utilized by the mining in- from a full spreadsheet that specifies a range of
classification levels for over 130 different pa-
dustry. Of the 15+ materials that we have tested
rameters. The base document is an Excel
to date, over 50% have sufficient reactive sul-
fide content vs. net neutralizers that we have re- spreadsheet that allows for entry of sample
characterization data into appropriate review
jected them from consideration for upland use.
fields with instructions for the proposer to
Typically, these acid-forming materials would
require at least 30 Mg/ha of lime additions per bold/highlight all exceedances.
This proposed system is unique in that it pro-
15 cm depth to neutralize their acid loadings to
poses two tiers of acceptable materials with dif-
allow for crop production. Further oxidation be-
neath the limed zone would lead to significant ferent levels of permitting, management and
monitoring requirements. The left-hand column
Table 2: Selected examples of proposed screening limits gested/inhaled by humans at target doses. How-
for beneficial upland use of dredge sediments. These are ever, Zn at these levels in soils could be phyto-
drawn from a full spreadsheet list of > 130 parameters.
toxic at soil pH levels < 7.5, and we therefore
Exclude
have adjusted the Zn screening levels for our
Parameter Clean Fill Limit EPA*
system based on agronomic and soil literature.
mg/kg
In addition to the conventional screening cri-
Metals
teria presented in the large table, we are also re-
As 20 40 1.6
quiring that all proposed sediments be carefully
Pb 300 800 400
screened for acid production potential, soluble
Zn 1500 7500 310,000
salts and other properties such as texture and
Organics
Total PCBs 0.49 25.2 25.2
TOC. Based on our collective experience in this
BAP** 0.21 0.66 0.21
area since 2001, it is obvious that a number of
4,4'-DDT 2.0 7.0 7.0 materials that will be proposed and available for
*USEPA 2008 Region 3 Risk Based Criteria for industrial upland placement are going to fall somewhere
uses. between our “clean fill” and “exclusion” criteria
** Benzo(a)pyrene as explained above. These materials should con-
tinue to be regulated via a site-specific permit
in Table 2 lists screening levels that we propose with soil and water quality monitoring criteria
to be used to separate “clean fill” from material tailored to the nature of the inbound sediments
with moderate levels of contaminants. From a and the characteristics of the utilization site.
regulatory perspective, we propose that these
materials should be released for upland use
without intensive soil and groundwater monitor- 6. CONCLUSIONS
ing requirements once the material properties Upland placement of dredge sediments offers
had been confirmed via inbound sediment sam- great potential to improve land use potentials on
pling and analysis. However, site placement formerly mined and degraded lands with an as-
mapping and minimum fill location require- sociated economic return to landowners. All
ments would need to be documented for these dredge materials need to be rigorously screened
materials. The middle column in Table 2 confor acid-forming potential in addition to conven-
tains what we consider to be “exclusion criteria” tional chemical and physical quality criteria.
where one or more exceedances (of average Freshwater dredge materials are generally better
characterization values) would prevent upland candidates for upland utilization than saline ma-
utilization. The right-hand column in Table 2 terials, but saline materials can be effectively
lists the current USEPA Region 3 risk-based managed over longer timeframes. The new clas-
soil screening levels which are intended for use sification system proposed here could signifi-
in contaminated site cleanup. However, they are cantly improve the review, permitting and over-
often utilized by state agencies and consultants all environmental compliance process for these
as generic screening criteria for other applica- unique materials.
tions.
The various screening levels given in Table 2
were originally developed from human health REFERENCES
risk assessment protocols that may be quite in- Daniels, W.L. and G.R. Whittecar, 2006. Annual Moni-
appropriate for the intended uses discussed here. toring Report - 2005 - Weanack Dredge Spoil Utiliza-
For example, the USEPA industrial site soil tion. Submitted to Va DEQ and PCC 2/15/2006.
screening level for As is 1.6 mg/kg (Table 2) Available at: http://www.cses.vt.edu/ revegeta-
while the normal background As level in native tion/dredgemanu.html.
Virginia soils is approximately 5.0 mg/kg Daniels, W.L., G.R. Whittecar and C.H. Carter, 2007.
Conversion of Potomac River Dredge sediments to
(Smith et al., 2005). Use of this standard for productive agricultural soils. p. 183-199 In: R.I.
screening dredge materials is therefore clearly Barnhisel (ed.), Proc., 2007 Nat. Meeting Amer. Soc.
inappropriate and nonsensical. Similarly, the Mining and Reclamation, Gillette WY, June 2-7,
USEPA screening level for Zn is 310,000 mg/kg 2007. Published by ASMR, 3134 Montavesta Rd.,
since Zn is not particularly toxic when in- Lexington, KY, 40502, 980 p.
Darmody, R.G. and J.C. Marlin, 2002. Sediments and
sediment-derived soils in Illinois: pedological and ag-

ronomic assessment. Env. Mon. and Assess. 77: 209-
227.
Darmody, R.G., J.C. Marlin, J. Talbott, R.A. Green, E.F.
Brewer and C. Stohr, 2004. Dredged Illinois River
sediments: Plant growth and metal uptake. J. Env.
Qual. 33:458-464.
Fanning, D. S., and C. R. Fanning, 1989. Soil Morphol-
ogy, Genesis and Classification. John Wiley and Sons,
New York, Chapter 10 of 395 p.
Lee, C.R., 2001. Manufactured soil field demonstrations
on brownfields and abandoned minelands, DOER
Technical Notes Collection (ERDC TN-DOER-C25),
U.S. Army Engineer Research and Development Cen-
ter, Vicksburg, MS. www.wes.army.mil/el/dots/doer.
New Jersey Department of Environmental Protection, The
Management and Regulation of Dredging Activities
and Dredged Material in New Jersey’s Tidal Waters.
1997. http://www.njstatelib.org/digit/r588/
r5881997.html.
Smith, D.B., W.F. Cannon, L.G. Woodruff, R.G. Garrett,
R. Klassen, J.E. Kilburn, J.D. Horton, H.D. King, M.
B. Goldhaber and J.M. Morrison, 2005. Major- and
Trace-Element Concentrations in Soils from Two
Continental-Scale Transects of the United States and
Canada. USGS Open File Report 2005-1253.
http://pubs.usgs.gov/of/2005/1253/ pdf/OFR1253.pdf.
USEPA, 2003. Dredged Material Management Action
Agenda for the Next Decade, United States Environ-
mental Protection Agency EPA 842-B-04-002.
USEPA, 2005. Contaminated Sediment Remediation
Guidance for Hazardous Waste Sites, EPA-540-R-05-
012 Office of Solid Waste and Emergency Response
OSWER 9355.0-85 Dec. 2005.
USEPA, Region 3 Risk Based Criteria/soil Screening
Levels. 2008. http://www.epa.gov/reg3hwmd/risk/
human/rb-concentration_table/Generic_Tables
/index.htm.
Environmental impacts relative to soil quality caused from the disposal of

olive oil mills’ wastes. Case study: A municipality in Crete, Greece
M. Doula, V. Kavvadias, S. Theocharopoulos and P. Kouloumbis

National Agricultural Research Foundation (NAGREF) - Soil Science Institute of Athens, Greece
D. Oikonomou and D. Arapoglou

NAGREF - Institute of Technology of Agricultural Products
ABSTRACT wastes, European Commission co-funded in the

framework of LIFE+ funding scheme a four-
Olive oil mills wastes (OOMW), a by-product
year project entitled “Strategies to improve and
of the olive mill industry, are produced in large
protect soil quality from the disposal of olive oil
quantities in Mediterranean countries. OOMW
mills’ wastes in the Mediterranean region” (Ac-
contain high organic load, substantial amounts
ronym: PROSODOL; start date: 1st January,
of plant nutrients but also several compounds
2009). Five Institutions from the three major
with recognized toxicity towards living organ-
olive oil productive countries worldwide (Spain,
isms. The effect of the disposal of untreated
Italy, Greece) participate in the project, which
OOMW on soil chemical properties was inves-
aims to:
tigated by collecting and analyzing soil samples
from areas near evaporation lagoons in the is- - Develop and disseminate innovative, envi-
land of Crete, south Greece. ronment friendly, low cost technologies for
the protection of soil and water pollution
from olive oil mills’ wastes.
1. INTRODUCTION - Establish an info-library/knowledge-base
In Greece there are about 108 olive trees and system to assess environmental impacts from
2800 olive oil mills while the average oil pro- wastes in the Mediterranean region.
duction in cultivation period of 2006/2007 was - Facilitate the implementation of Soil The-
about 250.000 tn. Although the disposal of un- matic Strategy in areas close to olive oil
treated mills wastes in the environment is not mills.
permitted, it is estimated that up to 1.5 million - Design, implement and support a monitoring
tonnes of wastes are disposed each year. The system for the assessment of the soil and wa-
usual treatment and disposal method in Greece ter quality affected directly or indirectly from
is the evaporation in lagoons/ponds after neu- mills’ activities in relation to factors, pres-
tralization with lime. There are also many cases sures and responses.
of sea, river and underground disposal. In prac- - Identify potential safest uses in the agricul-
tice the evaporation lagoons/ponds are rarely of tural sector of olive oil mills’ wastes and
proper size and construction and wastewaters possible contribution to agricultural produc-
often overflow affecting the neighbouring sys- tion.
tems (soil, surface and groundwater) but also One of the main project’s technical part is the
professional activities of the residents (agricul- development of a soil and water monitoring sys-
ture, livestock farming). The base of the lagoons tem which foresees soil and water sampling
is in most cases, permeable and thus, the prob- every two months. First results regarding soil
ability for groundwater and deep soil horizons quality and its potential degradation due to
contamination is high. wastes disposal are presented in this work.
Due to the serious environmental problems
caused by the uncontrolled disposal of these
2. SOIL SAMPLING AND ANALYSIS

2.1 Implementation area
The project area is located in the municipality of
Nikiforos Fokas, prefecture of Rethymnon,
Crete. The municipality’s jurisdiction extends
over fourteen former community wards, com-
prises a total of 21 villages, has a total area of
95 km2 and a population of approximately 6600.
Geological formations of the area under study
are mainly identified as limestones, dolomites
and marbles. Soils are slightly alkaline to alka-
line (pH between 7,3 and 8,0), rich in carbon-
ates (40 %-60 %) and clay or silty clay in tex-
ture.
Soil samples were collected from five dis-
posal areas: three active and one inactive for 5
years.
All of them contain lagoons which were built
by excavating the superfacial materials of soil,
and (for some of them) by heaping excavating
materials around the lagoon to form low retain-
ing walls. Protective impermeable membranes
or other protective media were not used. The
disposed wastes derived from three-phase mills
with continuous centrifuge extraction systems. Photo 1: The two sampling disposal areas in Municipality
of Nikiforos Fokas, Rethymnon, Crete.
In this work data from two active disposal
areas, NFR and NFC (Photo 1, Table 1), are
presented. Both lagoons are used for waste dis- and exchangeable K, Ca and Mg by BaCl2 ex-
posal for more than 10 years. traction (ISO 11260:1994); and available Mn
Sampling took place one month after com- and Fe with DTPA extraction (ISO
pletion of wastes disposal, between 6th and 9th 14870:2001).
May, 2009.
2.2 Soil parameters 3. RESULTS EVALUATION
Soil samples were collected from different plac- Regarding OM content, it was observed that
es and depths of 0-25 cm, 25-50 cm, 50-75 cm, there was no substantial difference among sam-
75-100 cm, 100-125 cm and 125-150 cm. The pling locations in NFR site (Fig. 1). Since mills’
preparation of samples for analysis was per- owners, as they say, do not dispose wastes di-
formed according to ISO 11464:2006 method. rectly on land (except one of them-NFC area),
Laboratory determinations were performed these results are expectable and are similar to
according to the methods usually used for soil values for uncultivated areas. Further investiga-
characterization (Miller and Keeny, 1982). Par- tion of factors potentially affecting OM content
ticle size distribution analysis was carried out by should be carried out.
Bouyioukos method; the pH and the Electrical For NFC place the surface disposal resulted
Conductivity (EC) were measured in paste ex- in substantial increase in OM content at all soil
tract; organic matter (OM) was determined by depths for most sampling locations (Fig. 2) due
dichromate oxidation; carbonates by using Ber- to waste sedimentation on the soil surface and
nard calcimeter; total N by the Kjeldahl method infiltration to other horizons.
(ISO 11261:1995); available phosphorous with In general, soil samples collected were car-
sodium hydrogen carbonate extraction (ISO bonate rich and have basic pH. These two pa-
11263:1994); Cation Exchange Capacity (CEC) rameters buffer wastes’ acidity (pH ̴ 4.6). This,
Table 1: Sampling sites of NFR and NFC disposal areas.

Sampling 8 % Organic Matter
Characteristics Direct disposal on soil 0-25cm
Site
7 25-50cm
NFR 1 Internal lagoon’s walls
50-75cm
NFR 2 2m from the lagoon, (h) 6
75-100cm
NFR 3 4m from the lagoon, (u)
5 100-125cm
NFR 4 10m from the lagoon, (u) 125-150cm
NFR 5 12m from the lagoon (u) 4
NFR 6 Unaffected-Control
3
NFC 1 External lagoon’s wall (u)
NFC 2 3m from the lagoon, (u) 2
NFC 3 3m from the lagoon, (u) 1
25m from the lagoon, surface dis-
NFC 4 0
posal, (u)
NFC 5
posal, (u)
NFC 6 Figure 2: OM content for NFC disposal area. The content
posal, (u) of the sample taken from the internal wall of the lagoon
75m from the lagoon, surface dis- was 36.1%.
NFC 7
posal, (u)
NFC 8 External lagoon’s wall (e) wastes’ content in potassium, chloride, sulfate,
NFC 9 Unaffected-Control and nitrates produced through wastes minerali-
(h):higher from wastes level; (u):under the wastes level; zation and transformation. EC increase is corre-
(e): at the waste level lated with seed germination toxicity and thus,
toxic effects can be expected in soils where
in turn, results in carbonates decrease in relation mills’ wastes have been disposed more than 2
to that of the control sample (Fig. 3). months previously.
With regard to EC, the values for affected For NFR area there was considerable en-
soils are in general higher than those for the richment of exchangeable K (Fig. 5). Particu-
control (Fig. 4). larly for NFR2 location (2 m from the lagoon)
According to Paredes et al., (1987) and to Si- the concentration of exchangeable K was very
erra et al., (2001), the increase in soil EC ap- high which, in turn, may result in phytotoxicity.
pears to be irreversible and is correlated with In general the increase in the content of K and
25 60
Organic Matter % CaCO3 0-25cm
25-50cm
control
%
0-25cm 50-75cm
20 50
75-100cm
25-50cm
50-75cm
75-100cm 40
15
30
10 m
10
2m
20
control
2m 4m 12 m
5 4m 10 m 12 m
10
0 0
NFR.1 NFR.2 NFR.3 NFR.4 NFR.5 NFR.6 NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6
Figure 1: OM content for NFR disposal area. Figure 3: Carbonate content for NFR disposal area.
4 mS/cm EC 0-25cm mg/g 2m

25-50cm 5
Total N
50-75cm 4,5
3,5 0-25cm
75-100cm
4 25-50cm
3 50-75cm control
3,5 75-100cm
2,5
3 10 m
2 2,5 4m
12 m
1,5 2
2m 4m
control 1,5
1 10 m
12 m 1
0,5
0,5
0 0
NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6 NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6
Figure 4: Electrical Conductivity for NFR disposal area. Figure 6: Total N for NFR disposal site.
other nutrients (N, P) as well as, OM could have mg/kg

300 Available P
beneficial effect on soil fertility therefore olive
0-25cm
oil mills wastes could help to reduce, or avoid 25-50cm
the application of chemical fertilizers. However, 250
50-75cm
the significant increase in cations content (e.g. 2m
75-100cm
K) causes increase in soil electrical conductivity 200
and contributes to the increase of soil salinity
and thus to desertification. 150
Total nitrogen (Fig. 6) and available phos-
phorous (Fig. 7) content were also higher in af-
100
fected soils, with concentrations decreasing with
depth and distance from the lagoon. For nitro-
control
gen the increase in content appeared mainly on 50
4m 10 m 12 m
deeper horizons since the surface samples con-
0
16 cmol/kg NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6
Exchangeable K
14 Figure 7: Available phosphorous for NFR disposal site.
0-25cm
12 25-50cm tent did not differ considerably compared to
50-75cm control. With regard to phosphorous there was a
10 75-100cm notable decrease at all soil depths with the dis-
8 tance from the lagoon while, for NFR2 place,
2m like K, the P content was very high.
6 Exchangeable Mg was also affected from the
disposal. As Figure 8 presents, all sampling sites
4
4m 10 m have higher content of Mg than the control.
control
2 12 m However, no relation was observed between
Mg content and distance from the lagoon.
0 Exchangeable Ca content was not affected by
NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6 the disposal since, all samples were similar in
content to the sample taken from the internal la-
Figure 5: Exchangeable K for NFR disposal site. goon’s wall and slightly higher than the control
mg/kg
140 Available Mn 0-25cm
25-50cm
50-75cm
120 75-100cm
4m
100
10 m
80
12 m
60 2m
40 control
20
0
NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6
Figure 8: Exchangeable Mg for NFR disposal site. Figure 10: Mn content for NFR disposal area.
mg/kg
350 Available Fe
300 0-25cm
25-50cm
50-75cm
250 75-100cm
200
2m
150
100 4m
10 m 12 m control
50
0
NFR 1 NFR 2 NFR 3 NFR 4 NFR 5 NFR 6
Figure 9: Exchangeable Ca for NFR disposal site. Figure 11: Fe content of NFR disposal area.
(Fig. 9). phenols entering the soil with wastes.

In general, the increase in exchangeable ca- In this process, polyphenols are oxidized to
tion content after the disposal was followed by polyquinones, which then polymerized with
enhancement of CEC (data not shown). amino acids to form humic materials. This oxi-
Regarding available Mn and Fe, there was dative polymerization reaction is catalyzed by
markedly decrease in their concentrations with polyphenol oxidase enzymes such as tyrosinase,
the distance from the lagoon, except the 2 m however, Fe and Mn oxides promote also the
sampling location for Mn, which will be further reaction (McBride, 1987). Thus, Mn(IV) oxides
investigated during the future monitoring of the are reduced to extractable forms of Mn2+ while
area (Figs. 10, 11). Fe(III) is reduced to Fe(II) and then oxidized
According to Piotrowska et al., (2006), Mn again by the reaction's products (i.e. quinones)
and Fe catalyze the oxidative transformation of and liberated to the soil solution phase as Fe3+
(Kung and McBride, 1988).
3. CONCLUSIONS
Disposal of untreated OOMW at evaporation
lagoons without using protective materials (e.g.
impermeable membranes) has significant effect
on soil chemical properties. Soil samples col-
lected one month after the completion of wastes
disposal are characterized by enhanced content
in nitrogen; OM; exchangeable K, Mg; CEC;
available Mn and Fe as well as increased EC
and decreased CaCO3. Changes in soil proper-
ties were depended on depth and distance from
the disposal lagoon.
REFERENCES
Kung, K.-H., M.B. McBride, 1988. Electron transfer
processes between hydroquinone and iron oxides.
Clays and Clay Minerals 36: 303-309.
McBride, M.B., 1987. Adsorption and oxidation of phe-
nolic compounds by iron and manganese oxides. Soil
Sci. Soc. Amer. J. 51: 1466-1472.
Miller, R.H. and D.R. Keeny, 1982. Methods of Soil
Analysis, Part 2. Madison: American Society of
Agronomy.
Paredes, M.J., E. Moreno, A. Ramos-Cormenzana and J.
Martinez, 1987. Characteristics of soil after pollution
with waste waters from oil extraction plants. Chemos-
phere 16: 1557-1564.
Piotrowska, A., G. Iamarino, M.A. Rao and L. Gianfreda,
2006. Short-term effects of olive mill waste water
(OMW) on chemical and biochemical properties of a
semiarid Mediterranean soil. Soil Biology & Bio-
chemistry 38: 600-610.
Sierra, J., E. Marti, G. Montserrat, R. Cruanas and M.A.
Garau, 2001. Characterization and evolution of a soil
affected by olive oil mill wastewater disposal. Science
of the Total Environment 279: 207-214.
Bioremediation and toxicity mechanism of some herbicide dinitrophenyl

derivatives
G. Drochioiu
Al. I. Cuza University of Iasi-Romania
M. Murariu
Petru Poni Institute of Macromolecular Chemistry, Iasi-Romania
S. Jurcoane
University of Agronomical Sciences and Veterinary Medicine of Bucharest, Romania
ABSTRACT Dinitrophenyl derivatives such as Dinoseb (2-

sec-butyl-4,6-dinitrophenol), Karathane (2,6-
Biodegradation and toxicity mechanisms of
dinitro-4-octylphenyl crotonate) or DNOC (4,6-
some dinitrophenols (DNP) and dinitrophenyl
dinitrocresol) have been widely used pesticides
ethers (DNE), such as 2,4-, 2,6-, 3,5-dinitro-
that persist in some contaminated soils, and
phenol, dinitro-o-cresol (DNOC), dinitroanisole
have been found in ground-waters, causing
(DNA), dinitro-phenetole (DNF), 2,4-dinitro-1-
health and environmental hazards (Stevens
(octadecyloxy)-benzene (DNO), Karathane (Krt,
et al., 1991). Dinitrophenols act as uncoupling
2,6-dinitro-4 or 6-octylphenyl crotonate), and 3-
agents of oxidative phosphorylation at low con-
(2,4-dinitro-phenoxy)propan-1,2-diol (DNG)
centrations, and inhibit the electron transport
have been studied on yeast and wheat. Dinitro-
chains (Matsumoto et al., 2008; Takahashi et al.,
phenols and dinitrophenyl ethers inhibited
2009). Their excess toxic potency could be ex-
wheat germination, whereas 2,4-dinitrobenzoic
plained by their molecular structure involving in
acid proved to have a stimulatory action. Un-
vivo reaction with bio-macromolecules (Wang
coupling dinitrophenols as well as dinitrophenyl
et al., 2002; Fedoseeva et al., 2008). However,
ethers quench tryptophan fluorescence. The un-
several microbes degrade or use dinitrophenols
coupling activity of these compounds was corre-
as a nitrogen source (Zablotowicz et al., 1999;
lated with an intense absorption in near infrared
Stevens et al., 1991). The degradation pathway
region over 6000 cm-1, at quantum energies
for dinoseb under anaerobic reducing conditions
which are similar to that needed for releasing
involves reduction of the nitro groups to amino
the ATP molecule from ATPase. Fluorescence
groups followed by replacement with hydroxyl
quenching, FT-IR spectra, circular dichroism
groups (Kaake et al., 1995).
data, and solubility characteristics of dinitro-
The fundamental mechanisms of plant toler-
phenyl compounds shed a new light on their
ance and resistance to xenobiotic chemicals are
mechanism of toxicity and that of biological ac-
related to the phytotoxicity of major classes of
tivity. Chemical degradation and that by micro-
contaminants (Popa et al., 2007). The relation to
organisms are discussed in order to increase the
phytotoxicity provides a practical understanding
theoretical, medical, agricultural, forensic, and
of how tolerance and resistance may affect phy-
environmental interest in these compounds.
toremediation applications. Despite impressive
potential and progress, many technical questions
1. INTRODUCTION remain to be solved before in situ bioremedia-
tion becomes a reliable and accepted tool for
In spite of their toxicity to the environment, hazardous-waste management. Therefore, the
many dinitrophenols and dinitrophenyl derivati- bioremediation by several microorganisms
ves are still in use as herbicides, insecticides, might be discussed as a function of dinitro-
metabolic inhibitors or even therapeutic agents. phenyl group toxicity. The uncoupling effect of
some nitroderivatives could be related to a 2.3 Synthesis of dinitrophenyl ethers

higher phytotoxicity with no relationship with
The dinitrophenol ethers were synthesized using
the microbial bioremediation.
dinitrochlorobenzene, anhydrous carbonate and
This work aims at investigating the dinitro-
the corresponding alcohol (Molnar et al., 2006).
phenol derivatives in order to understand better
their phytotoxicity related to their chemical 2.4 Circular Dichroism
structure and spectral properties. Fluorescence
CD spectra of proteins in the presence of dini-
quenching, circular dichroism studies and infra-
trophenyl derivatives were recorded as shown
red absorption of the investigated compounds
previously (Murariu et al., 2007), in 0.5-mm
suggest an energy-based mechanism of toxicity,
quartz cells at pH 7.4 and 200 μM peptide con-
confirming our previous findings (Drochioiu
centration.
et al., 2008).
2.5 Germination tests
2. EXPERIMENTAL Four replications of 100 seeds were counted for
each replicate, whereas lots of 50 seeds were
2.1 Materials used for seedling measurements. Solutions used
Most of the reagents, biological material, and for treatment of the seeds were also incubated
instruments were as previously shown (Molnar together with the seeds. The first count was
et al., 2006; Popa et al., 2007). Dinitrophenols, made after 3 days (energy of germination, EG)
4-nitrophenol, 2,4-dinitrobenzoic acid, dinitro- and a second measurement was performed after
salycilic acid, dinitrochlorobenzene, and picric 7 days (germination rate, GR). Quantity of ger-
acid were purchased from Merck, whereas Ka- minated, abnormal, dead and fresh seeds was
rathane and yeast from the specialized compa- recorded.
nies. One week after the initial treatment, all
Seeds of spring wheat (Triticum aestivum), plants were harvested. The stems cut from the
Henika variety (32.7 g/1000 dried seeds), were seeds were measured (H, expressed as cm) and
used in the germination tests. weighed (W, expressed as grams). The roots
were also cut and weighed.
2.2 Equipment
1 2.6 Statistics
H-RMN (Bruker Advance DRx400, 400 MHz)
spectra and mass spectra (Vestec 601) were per- The results were processed using the Tukey test
formed. The infrared spectra were taken on a (Snedecor, 1994).
Jasco FT/IR660 Plus Fourier spectrometer in the
range from 400 to 8000 cm-1. UV-vis spectra
were produced in the range from 190 to 700 nm
with a Libra S35PC spectrophotometer (Bio- 3.1 Chemical syntheses
chrom Ltd., Cambridge, England). CD spectra
The structure of all newly synthesized and puri-
were recorded on a Jasco J-715 spectropo-
fied dinitrophenyl derivatives was proved by
larimeter (Labor und Datentechnik GmbH,
melting points, elemental analysis, NMR and
Germany) equipped with a thermal circulator
FTIR techniques.
accessory and spectrometer software (Dr. Jörg
Nerkamp). Steady state fluorescence measure- 3.2 Effect of DNP on wheat germination
ments were conducted with a Kontron SFM25
All the dinitrophenol derivatives inhibited wheat
spectrofluorimeter. The experiments and the
germination, whereas 2,4-dinitrobenzoic acid
germination determinations were performed in
proved to have a stimulatory action. For exam-
Petri dishes, on filter paper at room temperature.
ple, although DNO has no free OH group as di-
The new compound purification was carried out
nitrophenols do, a similar toxicity pattern was
by TLC on silica gel (Kieselgel 60F254, Merck)
observed (Fig. 1). In addition, DNG with alco-
and on silica gel column.
holic -OH groups but phenol ones inhibited the
wheat germination even at 3·10-4 M, when re-
duced the total height of the plantlets (H) from
nols.
3.4 FT-IR study
FT-IR spectra (Fig. 3) of the compounds with-
out uncoupling activity show a less intense ab-
Figure 1: Effect of 2,4-dinitro-1-(octadecyloxy)benzene sorption of near infrared radiation at wave num-
(DNO) on wheat germination and growth. 1) 10-3 M; 2)
8x10-3 M; 3) 6x10-3 M; 4) 5x10-3 M; 5) 3x10-3 M; 6) 2x10-
bers higher than 6000 cm-1, comparatively to
3
M; 7) 10-3 M; 8) Control with H2O. dinitrophenols and dinitrophenyl ethers, such as
2,6-dinitrophenol and 2,4-dinitroanisol. Dinitro-
625.7 cm to 584.3 cm. The mean height of the benzoic acid showed neither uncoupling activity
lots treated with DNG and DNA was 2.5 cm and nor the characteristic absorption over 6000 cm-1,
4.6 cm respectively, as compared to control, in infrared region.
which was 6.4 cm. Besides, the masses of the The energy of the electromagnetic vibration
resulted plantlets were diminished by both dini- was calculated around 6000 cm–1; this was then
trophenyl ethers and stimulated by dinitroben- multiplied by Avogadro's number and resulted
zoic acid. The fundamental mechanisms of plant about 17 kcal/mole.
tolerance and resistance to DNP seem to be re- Generally it is considered that ATP contains
lated to the phytotoxicity of investigated dini- an amount of energy of 7.3 kcal/mole. However
trophenyl derivatives. NADH, which generates 3 moles of ATP, con-
tains 52.6 kcal/mole. It is almost impossible to
3.3 Effect of DNP on yeast
explain why only about 42 % of NADH energy
The pH values of a yeast suspension increased is used to yield ATP. The rest of this huge
much in the presence of dinitrophenols, depend- amount of energy could increase much the body
ing on the type of compound (Fig. 2). Because temperature. Nevertheless, such thermic effect
amino acids and other molecules were present in 100
2,4-Dinitrobenzoic acid
the suspension, we considered that DNP dam-
aged the biostructure of yeast cell, as Ma- 80
covschi’s theory claims (Drochioiu, 2006).
Transmitance (%)
Both DNO and DNA or DNG increased glu-

60
cose consumption by yeast suspensions under
aerobic conditions (data not shown). In addition,
the supernatant was found to be yellow-red col- 40
ored within the 3-day experiments, especially in
the case of DNOC and less that of DNA. Dini- 20
trophenyl ethers proved to have similar behav-
iour to those of the corresponding dinitrophe- 1000 2000 3000 4000 5000 6000 7000 8000
100
6
4-Nitrobenzoic acid
80
Transmitance (%)
5.8
60
5.6
pH value
40
5.4 Yeast
2,5-DNP
2,4-DNP 20
5.2
DNOC
1000 2000 3000 4000 5000 6000 7000 8000
5 Wave number (cm -1)
0 40 80 120
Figure 3: FTIR spectra of 2,4- dinitrobenzoic acid and 4-
Time of contact (min)
nitrobenzoic acid, which show that the less intense ab-
Figure 2: The effect of some dinitrophenols on the pH of sorbance around 6000 cm-1 is correlated with the lack of
yeast suspensions. uncoupling effect.
100 35
4-Nitrobenzaldehyde
Transmitance (%)
Tryptophan fluorescence
80 28
y = 31,161e-0,0169x
60 21 r = 0,999
40 14
7
20
1000 2000 3000 4000 5000 6000 7000 8000 0

100
0 50 100 150 200 250
DNOC
2,5-DNP concentration (μΜ)
80 Figure 5: Quenching of tryptophan fluorescence by 2,5-
Transmitance (%)
dinitrophenol.
60 Because the hydrogen atom within the aldehyde

group is tightly tied, uncoupling activity based
40 on proton translocation cannot be considered.
3.5 Fluorescence study
20
Tryptophan, an essential amino acid, is highly
1000 2000 3000 4000 5000 6000 7000 8000
fluorescent. Uncoupling dinitrophenols as well
as dinitrophenyl ethers proved to quench its
Wave number (cm-1)
Figure 4: FTIR spectra of the uncoupling agents DNOC
fluorescence. The intense fluorescence quench-
and 4-nitrobenzaldehyde, which show an intense absorp- ing of DNOC, more powerful than that of 2,5-
tion at 6000 cm-1. dinitrophenol, was associated with a stronger
uncoupling effect. 2,4-Dinitrobenzoic acid and
was not noticed in the living cells. Besides, Paul 4-nitrobenzoic acid, without such uncoupling
Boyer (2002) showed that the metabolic energy activities did not manifest a similar quenching,
(12-13 kcal/mole) is used to release ATP mole- although it contains an ionisable group (Fig. 5).
cules from ATP synthase and not for their de
novo synthesis. 3.6 Circular Dichroism
FADH2 and NADH generated in Krebs cycle A beneficial action of low concentrations of di-
will produce 2 moles of ATP (2x17.97 = nitrophenols, including neuroprotection against
35.94 kcal/mole) and 3 moles of ATP (3x17.5 = different types of insult, blockade of amyloid
52.52 kcal/mole), respectively. Hence, if a di- aggregation, stimulation of neurite outgrowth
rect transfer of a quantum of energy might pro- and neuronal differentiation, and even extension
duce a molecule of ATP according to quantum of lifespan in certain organisms was reported
mechanics rules, an electromagnetic radiation (Felice and Ferreira, 2006). Therefore, we in-
with about 13-17 kcal/mole, corresponding to a vestigated the effect of dinitrophenols and their
wave number range from 4539 to 6112 cm-1 derivatives on the secondary structure of pro-
would be necessary. This electromagnetic quan- teins. First, we used albumin (Fig. 1) and found
tum could emerge from an electron transition small conformational changes upon 2,6-DNP,
from an energetically higher level to another 2,4-DNP, DNOC and DNG binding.
lower one. Thus, albumin has a high proportion (29.4 %)
Indeed, FT-IR spectra of DNOC, which of α-helical conformers and 31.5% β-turn ones.
showed an intense absorption area around 2,6-DNP in 1:5 molar ratio changed its confor-
6000 cm-1, correlated well with the uncoupling mation to 27.5 % α-helix and 30.5 % β-turn
effect (Fig. 4). All the uncoupling agents inves- forms. DNOC (1:5 molar ratio) induced also a
tigated here had the same large band of absorp- small decrease in α-helix proportion (28.4 %),
tion, as shown for 4-nitrobenzaldehyde (Fig. 4). while DNG reduced α-helical forms to 28.0 %.
3.7 Bioremediation and biodegradation lar pattern of biological activity inhibiting seed
germination or increasing the fermentative
The bioremediation by several microorganisms
processes in yeast. The uncoupling activity of
should be discussed as a function of dinitro-
these compounds was correlated with an intense
phenyl group toxicity. The uncoupling effect of
absorption in near infrared region around 6000
some nitroderivatives could be related with a
cm–1, at energies which are similar to that for
higher phytotoxicity with no relationship with
releasing the ATP molecule from ATPase. Fluo-
the microbial bioremediation. The degradation
rescence quenching of tryptophan, circular di-
pathway for dinoseb under anaerobic reducing
chroism data, and solubility characteristics of
conditions involves reduction of the nitro
uncouplers shed a new light on their mechanism
groups to amino groups followed by replace-
of biological activity. This one seems to be re-
ment with hydroxyl groups (Kaake et al., 1995).
lated to energy transfer during the process of
Sphingomonas strain UG30 oxidatively metabo-
ATP formation and not with proton transloca-
lizes certain 2,4-dinitrophenol (2,4-DNP), and
tion through the biological membranes. The mi-
4,6-dinitrocresol with liberation of nitrite
crobial biodegradation or that by yeast can be
(Zablotowicz et al., 1999) PCP-4-
applied in the case of uncoupling dinitrophe-
monooxygenase (pcpB gene expressed in Es-
nols.
cherichia coli) is likely central to degradation of
the herbicide dinoseb (Ramos et al., 2005). A
2,4-dinitrophenol-degrading bacterial strain, ACKNOWLEDGEMENTS
DNP505(T), was isolated from an industrial
The funding by MEdCT-ANCS Bucharest,
wastewater (Yoon et al., 2000).
CNMP project 2746/2007 (Contract 31-017
3.8 Toxicity study /2007) is gratefully acknowledged.
Both toxicity mechanism and the biodegradation
one might be related to the dinitrophenol inter- REFERENCES
ference with energy transfer in ATP formation,
Argese, E., C. Bettiol, D. Marchetto, S. De Vettori, A.
but with proton translocation through the bio- Zambon, P. Miana and P.F. Ghetti, 2005. Study on the
logical membranes or their chemical reactions. toxicity of phenolic and phenoxy herbicides using the
For example, the uncoupling dinitrophenyl de- submitochondrial particle assay, Toxicol in vitro, 19:
rivatives have also a large absorption range in pp. 1035-1043.
infrared region around 6000 cm-1 and show a Boyer, P.D., 2002. A research journey with ATP syn-
thase, J Biol Chem, 277: pp. 39045-39061.
specific fluorescence quenching. Drochioiu, G., 2006. Eugen Macovschi’s concept of bio-
Dinitrophenols prevent aggregation and structure and its current development. In Life and
cause disaggregation of Aβ fibrils and that they mind. In search of the physical basis. (S. Savva ed.)
strongly prevent the neurotoxicity of Aβ to rat Trafford Publ., Canada, USA, Ireland & UK, pp. 43-
hippocampal neurons in culture (Felice et al., 60.
2004). However, it is not clear how these com- Fedoseeva, I.V., K.Z. Gamburg, N.N. Varakina, T.M. Ru-
saleva, E.L. Tauson, I.V. Stupnikova, G.B. Borovskii,
pounds might affect the secondary structure of A.V. Stepanov, E.A. Davydenko, E.G. Rikhvanov and
proteins of ATP synthase. V.K. Voinikov, 2008. The effect of sodium azide and
There are different mechanisms of action for 2,4-dinitrophenol on the development of thermotoler-
the various classes of pesticide compounds (Ar- ance and induction of Hsp101 in cultured Arabidopsis
gese et al., 2005). Although the findings ob- thaliana cells. Russian J Plant Physiol, 55: 225-231.
Felice, F.G.D. and S.T. Ferreira, 2006. Novel neuropro-
tained for dinitrophenolic herbicides are consis- tective, neuritogenic and anti-amyloidogenic proper-
tent with a protonophoric uncoupling mecha- ties of 2,4-dinitrophenol: The gentle face of Janus,
nism, our results revealed a more complicated IUBMB Life, 58: pp. 185 - 191.
one, based on a possible energetic transfer. Felice, F.G.D., J. Houzel, J. Garcia-Abreu, P.R.F.
Louzada, R.C. Afonso, M.N.L. Meirelles, R. Lent,
V.M. Neto and S.T. Ferriera, 2004. Inhibition of Alz-
4. CONCLUSIONS heimer’s disease β-amyloid aggregation, neurotoxic-
ity, and in vivo deposition by nitrophenols: implica-
The newly synthesized dinitrophenyl ethers and tions for Alzheimer’s therapy. FASEB J, 15: pp. 1297-
the well known dinitrophenols displayed a simi- 1299.
Kaake, R.H., D.L. Crawford and R.L. Crawford, 1995.

Biodegradation of the nitroaromatic herbicide dinoseb
(2-sec-butyl-4,6-dinitrophenol) under reducing condi-
tions, Biodegradation, 6: pp. 329-337.
Matsumoto, M., T. Furuhashi, C. Poncipe and M. Ema,
2008. Combined repeated dose and reproduc-
tive/developmental toxicity screening test of nitrophe-
nolic herbicide dinoseb, 2-sec-butyl-4,6-dinitrophenol,
in rats. Environ Toxicol, 23: pp.169-183.
Molnar, R., D. Creanga, R. Ruscanu, M. Murariu, C. Cio-
banu, M.I. Neica, A. Pui and G. Drochioiu, 2006. Di-
nitrophenol pesticides: biological activity, toxicity,
and a novel mechanism of action, Lucr Stiint Univ
Agron Iasi, Romania, 49: pp. 352-357.
Murariu, M., E.S. Dragan and G. Drochioiu, 2007. Syn-
thesis and mass spectrometric characterization of a
metal-affinity decapeptide: copper-induced conforma-
tional changes. Biomacromolecules, 8: pp. 3836-3841.
Popa, K., M. Murariu, R. Molnar, G. Schlosser, A. Cecal
and G. Drochioiu, 2007. Effect of radioactive and non-
radioactive mercury on wheat germination and the
anti-toxic role of glutathione. Isot Environ Health S,
43: pp. 105-116.
Ramos, J.L., M.M. González-Pérez, A. Caballero and P.
van Dillewijn, 2005. Bioremediation of polynitrated
aromatic compounds: plants and microbes put up a
fight, Curr Opinion Biotechnol, 16: pp. 275-281.
Snedecor, G.W., 1994. Statistical methods applied to ex-
periments in agriculture and biology, The Iowa Stat
Univ. Press, U.S.A., pp. 255-274.
Stevens, T.O., R.L. Crawford and D.L. Crawford, 1991.
Selection and isolation of bacteria capable of degrad-
ing dinoseb (2-sec-buty-4,6-dinitrophenol), Biodegra-
dation, 2: pp. 1-13.
Takahashi, M., M. Sunaga, M. Hirata-Koizumi, A. Hi-
rose, E. Kamata and M. Ema, 2009. Reproductive and
developmental toxicity screening study of 2,4-
dinitrophenol in rats. Environ Toxicol, 24: pp. 74-–81.
Wang, X., J. Yu, Y. Wang and L. Wang, 2002. Mecha-
nism-based quantitative structure–activity relation-
ships for the inhibition of substituted phenols on ger-
mination rate of Cucumis sativus, Chemosphere, 46:
pp. 241-250.
Yoon, J.H., Y.G. Cho, S.S. Kang, S.B. Kim, S.T. Lee and
Y.H. Park, 2000. Rhodococcus koreensis sp. nov., a
2,4-dinitrophenol-degrading bacterium”, Int J Syst
Evol Micr, 50: pp. 1193-1201.
Zablotowicz, R.M., K.T. Leung, T. Alber, M.B. Cassidy,
J.T. Trevors, H. Lee, L. Veldhuis and J.C. Hall, 1999.
Degradation of 2,4-dinitrophenol and selected ni-
troaromatic compounds by Sphingomonas sp. UG30”,
Can J Microbiol, 45: pp. 840-848.
Odorless and cohesive zeo-sewage sludge produced by Hellenic natural

zeolite treatment
A. Filippidis, G. Papastergios, N. Apostolidis, I. Paragios and S. Filippidis

Department of Mineralogy-Petrology-Economic Geology, Aristotle University of Thessaloniki, Hel-
las
C. Sikalidis
Department of Chemical Engineering, Aristotle University of Thessaloniki, Hellas
ABSTRACT The large natural zeolite deposits and the low

cost of mining, gave access to large-scale utili-
Aerobic treatment of municipal sewage (pHinitial
zation (e.g. Pond and Mumpton, 1984; Tsitsish-
8.2) by the Hellenic Natural Zeolite (HENAZE)
vili et al., 1992; Carr, 1994; Collela and Mump-
gave as precipitate, odorless and cohesive zeo-
ton, 2000; Bish and Ming, 2001; Harben, 2002;
sewage sludge, suitable for safe deposition, but
Filippidis et al., 2005; Filippidis et al., 2008e-g).
also for the amendment of agricultural soils.
In the Trigono Municipality (Evros Prefec-
Simultaneously, the aerobic treatment resulted
ture, N.E. Greece) and in the Ntrista stream of
to the production of clear water of pH 7.3, free
Petrota village has been located a HEU-type
of odors and improved by 89% for the sus-
zeolite deposit, the Hellenic Natural Zeolite
pended particles, 90% for the color, 93% for the
(HENAZE) of GEO-VET N. Alexandridis & Co
P2O5 content, 94% for the chemical oxygen de-
O.E. concession (e.g. Filippidis and Kantiranis,
mand (COD) and 98% for the NH4 content.
2002, 2005, 2007; Filippidis, 2005, 2007; Filip-
These final values of the pH and of the previous
pidis et al., 2007). The purification of municipal
mentioned quality parameters, measured in the
sewage, as well as the production of odorless
overflowed clear water, are fulfilling the re-
zeo-sewage sludge, using HENAZE and stirring
quirements for disposition as downstream, irri-
time 7-60 min, has been investigated (e.g. Filip-
gation, swimming and fish waters. The
pidis, 2008; Filippidis et al., 2008a-d).
HENAZE comes from Ntrista stream of Petrota
The present study investigates the production
village, N.E. Greece. Contains 89 wt.% HEU-
of odorless zeo-sewage sludge, using the same
type zeolite, exhibit an ion exchange capacity of
material and conditions, but in the reduced stir-
226 meq/100g and regulates to neutral the pH of
ring time of 5 min.
waters. The addition of HENAZE in the agricul-
tural soils, increases the crops yield by 29-57%
and improves the quality by 4-46% of agricul- 2. MATERIALS AND METHODS
tural products, reduces the use of fertilizers by
The Hellenic Natural Zeolite (HENAZE) sam-
55-100%, reduces the usage of irrigation water
ple was selected from a vertical profile of the
by 33-67%, prevents the seepage of dangerous
Ntrista stream within the GEO-VET’s conces-
species into the water environment (e.g., NO3-
sion. Petrographic investigation of HENAZE
by 55-57%), protecting thus the quality of sur-
was performed on thin and polished thin sec-
face and underground waters.
tions. The morphology and chemistry of the
HEU-type zeolite were studied by Scanning
1. INTRODUCTION Electron Microscopy-Energy Dispersive Spec-
troscopy (SEM-EDS) with Link-AN 10000 EDS
High quality HEU-type natural zeolites, dis-
system. To minimize volatilization of alkalis,
plays unique physical and chemical features and
the electron beam spot size was enlarged and
have a great variety of environmental, indus-
the counting time decreased. The mineralogical
trial, aquacultural and agricultural applications.
composition was determined by X-Ray Powder
Diffraction(XRPD). Semi-quantitative estimates Chalkidiki Prefecture, exhibiting an amphoteric

were performed on the basis of the intensity character (Filippidis and Kantiranis, 2002, 2007;
(counts) of specific XRPD reflections, the den- Filippidis, 2005; Filippidis et al., 2006).
sity, the mass absorption coefficient (CuKa) and The Kilkis City (Northern Greece) sewage
external mixture standards of the identified wastewaters of pH 8.2, were treated at room
mineral phases. The chemical composition of temperature with HENAZE of grain-size <1.5
the HENAZE was determined by Atomic Ab- mm (Fig. 2), in aerobic butch-type experiment.
sorption Spectrometry. The ammonia ion is The experiments were carried out using mag-
most commonly used for CEC testing of materi- netic stirrer, mixing 300 ml municipal sewage
als and the result there from is referred to as with 7.5 g of HENAZE; after 5 min stirring,
ammonia ion exchange capacity. The ammonia polyaluminium chloride and cationic polyelec-
ion exchange capacity (sorption ability) of the trolyte was added. The overflowed clear water
HENAZE was determined according to the and the precipitated zeo-sewage sludge were
Ammonium Acetate Saturation (AMAS) separated by filtering and consequently the zeo-
method. The pH variations and the removal sewage sludge was dried overnight at room
ability of metals and anions were performed in temperature (RT). The starting municipal sew-
butch-type experiments at RT (e.g., Filippidis & age and the overflowed clear water, were ana-
Kantiranis 2002, 2007, Filippidis 2005, Filippi- lyzed for (method): pH (electrometric), color
dis et al. 2006). (photometric), suspended particles (filtering and
The typical platy crystals of HEU-type zeo- centrifugation), COD (method of K2CrO6), P2O5
lite have a grain-size of 5-25 μm (Fig. 1). The and NH4 (molecular absorption spectropho-
chemical formula of the zeolite is Ca1.5K1.4 tometry).
Mg0.6Na0.5Al6.2Si29.8O72.20H2O. HENAZE, on
average contains 89 wt.% HEU-type zeolite, 3
wt.% mica + clays, 6 wt.% feldspars and 2 wt.% 3. RESULTS
quartz. Chemically the HENAZE consists The treatment of urban wastewater gave as pre-
mainly of 68.62 wt.% SiO2, 11.80 wt.% Al2O3, cipitate odorless and cohesive zeo-sewage
2.92 wt.% K2O, 2.14 wt.% CaO, 1.13 wt.% sludge, dried overnight at room temperature
Na2O and 0.75 wt.% MgO. HENAZE shows a (Fig. 3).
remarkable ammonia ion exchange capacity of The treatment of urban wastewater (pH 8.2)
226 meq/100g and removes from their aqueous with the HENAZE gave also overflowed clear
solutions 74 % of Pb and 79% of Ag and 55- water (Fig. 4) of pH 7.3, free of odors and im-
57% NO3-, as well as an ability to neutralize the proved concerning the quality parameters by 89-
pH of basic lake water (pH 9.5) from the lake 98 % (Table 1).
Koronia (Prefecture of Thessaloniki) and of
acidic stream mine water (pH 5.5) from NE
Figure 1: SEM microphotograph of typical platy crystals Figure 2: The grain-size < 1.5 mm of HENAZE, used for
of HEU-type zeolite of the HENAZE. the aerobic butch-type experiment.
Table 1: Chemistry of starting municipal sewage (SMS)

and of overflowed clear water (CW)
Parameters (det. limit) SMS CW ±%
pH (0.1) 8.2 7.3 -11
Susp. particles, mg/L(5) 210 23 -89
Color, mg/L, Pt.scale(5) 1180 119 -90
P2O5, mg/L (0.02) 9.24 0.66 -93
COD, mg/L O2 (15) 410 26 -94
NH4, mg/L (0.02) 30.52 0.51 -98
nor, exhibiting thus an amphoteric character

(Charistos et al., 1997).
The Hellenic Natural Zeolite (HENAZE) is
of very high quality (> 85 wt.% HEU-type zeo-
Figure 3: Odorless and cohesive zeo-sewage sludge, dried lite), removes inorganic, organic, organometal-
overnight at RT. lic, gas species, metals, cations and anions from
their aqueous solutions. Also, it shows an ability
to neutralize the pH of acidic and basic waters.
HENAZE removes from their aqueous solutions
74% of Pb, 79% of Ag and 55-57% of NO3-
(Filippidis, 2005; Filippidis et al., 2006). The
increase of the pH in the acidic pH-range, could
mainly be attributed to the binding of the pro-
tons to the Lewis basic sites of the zeolite. The
decrease of the pH in the basic pH-range could
be the result of the removal of protons from sur-
face Brønsted acidic sites or even of the de-
tachment of protons from water molecules sur-
rounding the exchangeable cations, caused by
Figure 4: Center: Odorless and cohesive zeo-sewage
OH– attack on the zeolite (e.g., Godelitsas et al.,
sludge, Right: Starting municipal sewage (SMS) and Left: 1999, 2001, 2003). The mineralogical composi-
Overflowed clear water (CW) after the HENAZE treat- tion and the unique physico-chemical proper-
ment. ties, make HENAZE suitable material for nu-
merous environmental, industrial, agricultural
and aquacultural applications, such as: Animal
4. DISCUSSION AND CONCLUSIONS nutrition, soil amendment for agriculture, condi-
The natural zeolites show a remarkable ability tioning of acid and basic soils, greenhouse and
to remove inorganic, organic, organometallic, flowers substrates, durability and health im-
gas species, metals and radionuclides from their provement of lawn, purification of industrial
aqueous solutions. The sorption of the different and urban wastewaters, treatment of sewage
species from their solutions by the micro- meso- sludge, odor control, fishery and fish breeding,
and macroporous of natural zeolite can be at- gas purification and drying systems, oxygen en-
tributed to absorption (mainly ion exchange), richment of aqua ecosystems, improvement of
adsorption and surface precipitation processes drinking water, constructed wetlands and
(e.g., Tsitsishvili et al., 1992; Misaelides et al., wastewater treatment units (e.g. Collela and
1993, 1995; Godelitsas et al., 1999, 2001, 2003; Mumpton 2000; Harben, 2002; Filippidis, 2007;
Collela and Mumpton 2000; Bish and Ming, Filippidis et al., 2007, 2008a-g).
2001). The sorption of gas phases results to oxy- The HENAZE treatment of urban wastewater
gen enrichment of the air and to the remarkable gave as precipitate, odorless and cohesive zeo-
decrease of the malodour. Also, they show an sewage sludge, suitable for safe deposition but
ability to neutralize the pH of acidic and basic also for the reclamation of agricultural soils.
waters, acting either as a proton acceptor or do- The same stands for the odorless and cohesive
zeo-sewage sludge produced by mixing the characteristics of five natural zeolite samples for N.
sewage sludge and the HENAZE. Alexandridis & Co O.E., Thessaloniki: Internal report
(in Greek).
The HENAZE treatment of municipal sew- Filippidis, A., 2007. Zeolites of Trigono Municipality of
age (pH initial 8.2) gave also, overflowed clear Evros Prefecture in industrial, agricultural, cattle-
water of pH 7.3, free of odors and improved by raising and environmental technology, Proceedings,
89% for the suspended particles, 90% for the National Congress on Development Perspectives of
color, 93% for the P2O5 content, 94% for the Northern Evros, Petrota, August 4, pp. 89-107 (in
chemical oxygen demand (COD) and 98% for Greek).
Filippidis, A., 2008. Treatment and recycling of munici-
the NH4 content. The improvement in previous pal and industrial waste waters using Hellenic Natural
experiments (e.g., Filippidis et al., 2008a-d) Zeolite, CD-Proceedings, AQUA 2008, 3rd Interna-
reached higher values (up to 99% for suspended tional Conference on Water Science and Technology,
particles, 97% for color, 99% for P2O5, 97% for Athens, October 16-19, 5p.
COD and 99% for NH4), due to the differences Filippidis, A. and N. Kantiranis, 2002. Morphology, min-
eralogy, chemistry, mineralchemistry and ion ex-
in the composition of the starting municipal change capacity of five natural zeolite samples for N.
sewage and the longer stirring time (7-60 min). Alexandridis & Co O.E., Thessaloniki: Internal report
These final values of the pH and of the previous (in Greek).
mentioned quality parameters, measured in the Filippidis, A. and N. Kantiranis, 2005. Industrial, agricul-
overflowed clear water, are fulfilling the re- tural and environmental uses of the natural zeolites of
quirements for disposition as downstream, irri- Thrace. Bulletin Geological Society of Greece, 37: 90-
101 (in Greek with English summary).
gation, swimming and fish waters. Filippidis, A. and N. Kantiranis, 2007. Experimental neu-
The presence of HENAZE in the agricultural tralization of lake and stream waters from N. Greece
soils, increases the production by 29-57% and using domestic HEU-type rich natural zeolitic mate-
improves the quality by 4-46% of agricultural rial. Desalination, 213: 47-55.
products, reduces the use of fertilizers by 55- Filippidis, A., N. Kantiranis, A. Drakoulis and D. Vogiat-
zis, 2005. Quality, pollution, treatment and manage-
100%, reduces the usage of irrigation water by ment of drinking, waste, underground and surface wa-
33-67%, prevents the seepage of dangerous spe- ters, using analcime-rich zeolitic tuff from Samos is-
cies into the water environment (e.g., NO3- by land, Hellas, Proceedings, 7th Hellenic Hydrogeologi-
55-57%), protecting thus the quality of surface cal Conference, Athens, October 8-10, vol. 2, pp. 219-
and underground waters (e.g. Filippidis, 2005, 224.
2007; Filippidis et al., 2006, 2007, 2008c). Filippidis, A., N. Kantiranis, A. Drakoulis and D. Vogiat-
zis, 2006. Improvement and protection of the lake
Koronia using natural zeolite. Proceedings, 2nd Con-
gress of Aristotle University Environment Council,
ACKNOWLEDGEMENTS Thessaloniki, June 1-4, pp. 273-279 (in Greek with
We express our gratitude to the GEO-VET N. English summary).
Alexandridis & Co O.E., for the supply and Filippidis, A., A. Siomos, N. Barbayiannis and S. Filippi-
dis, 2007. Agriculrural and environmental applications
treatment of HENAZE, as well as for their eco- using Hellenic Natural Zeolite of Petrota (Evros), CD-
nomical support. Proceedings, Jean Monnet Congress, Veria, November
30-December 1, 10p (in Greek with English sum-
mary).
REFERENCES Filippidis, A., N. Apostolidis, S. Filippidis and I. Par-
Bish, D.L. and D.W. Ming, 2001. Natural Zeolites: Oc- agios, 2008a. Purification of industrial and urban
currence, Properties, Applications. Washington DC: wastewaters, production of odorless and cohesive zeo-
Mineralogical Society of America. sewage sludge using Hellenic Natural Zeolite, Pro-
Carr, D.D., 1994. Industrial Minerals and Rocks. Ann Ar- ceedings, 2nd International Conference on Small and
bor, Michigan: Braun-Brumfield Inc. Decentralized Water and Wastewater Treatment
Charistos, D., A. Godelitsas, C. Tsipis, M. Sofoniou, J. Plants, Skiathos, May 2-4, pp. 403-408.
Dwyer, G. Manos, A. Filippidis and C. Triantafyllidis, Filippidis, A., N. Apostolidis, I. Paragios and S. Filippi-
1997. Interaction of natrolite and thomsonite inter- dis, 2008b. Purification of dye-work and urban
growths with aqueous solutions of different initial pH wastewaters, production of odorless and cohesive zeo-
values at 25o C in the presence of KCl: Reaction sewage sludge, using Hellenic Natural Zeolite, CD-
mechanisms. Applied Geochemistry, 12: 693-703. Proceedings, 1st International Conference on Hazard-
Collela, C. and F.A. Mumpton, 2000. Natural Zeolites for ous Waste Management, Chania, October 1-3, 8p.
the Third Millenium. Napoli: De Frede Editore. Filippidis, A., N. Apostolidis, I. Paragios and S. Filippi-
Filippidis, A., 2005. Mineralogy and physico-chemical dis, 2008c. Safe management of sewage sludge, pro-
duced by treatment of municipal sewage with Hellenic

Natural Zeolite, CD-Proceedings, AQUA 2008, 3rd In-
ternational Conference on Water Science and Tech-
nology, Athens, October 16-19, 5p.
Filippidis, A., N. Apostolidis, I. Paragios and S. Filippi-
dis, 2008d. Zeolites clean up. Industrial Minerals,
April: 68-71.
Filippidis, A., N. Kantiranis, S. Filippidis, I. Vordogian-
nis, N. Apostolidis and I. Paragios, 2008e. Purification
of Textile-work Waste Waters with Natural Zeolite.
Patent Number: 1006140, Industrial Property Organi-
sation, Athens (in Greek).
nis, N. Apostolidis and I. Paragios, 2008f. Purification
of Sewage with Natural Zeolite. Patent Number:
1006145, Industrial Property Organisation, Athens (in
Greek).
nis, N. Apostolidis and I. Paragios, 2008g. Purification
of Tanning-work Waste Waters with Natural Zeolite.
Patent Number: 1006146, Industrial Property Organi-
sation, Athens (in Greek).
Godelitsas, A., D. Charistos, J. Dwyer, C. Tsipis, A. Filip-
pidis, A. Hatzidimitriou and E. Pavlidou, 1999. Cop-
per (II)-loaded HEU-type zeolite crystals: characteri-
zation and evidence of surface complexation with
N,N-diethyldithiocarbamate anions. Microporous and
Mesoporous Materials, 33: 77-87.
Godelitsas, A., D. Charistos, A. Tsipis, C. Tsipis, A.
Filippidis, C. Triantafyllidis, G. Manos and D. Siap-
kas, 2001. Characterisation of zeolitic materials with a
HEU-type structure modified by transition metal ele-
ments: Definition of acid sites in Nickel-loaded crys-
tals in the light of experimental and quantum-chemical
results. Chemistry European Journal, 7: 3705-3721.
Godelitsas, A., D. Charistos, C. Tsipis, P. Misaelides, A.
Filippidis and M. Schindler, 2003. Heterostructures
patterned on aluminosilicate microporous substrates:
Crystallisation of cobalt (III) tris(N,N-diethyl-
dithiocarbamato) on the surface of HEU-type zeolite.
Microporous and Mesoporous Materials, 61: 69-77.
Harben, P.W., 2002. The Industrial Minerals HandyBook.
Blackwood, UK: Pensord.
Misaelides, P., A. Godelitsas, D. Haristos, F. Noli, A.
Filippidis and C. Sikalidis, 1993. Determination of
heavy metal uptake by the sodium form of heulandite
using radiochemical techniques. Geologica Carpathica
- Series Clays, 44(2): 115-119.
Misaelides, P., A. Godelitsas, A. Filippidis, D. Charistos
and I. Anousis, 1995. Thorium and uranium uptake by
natural zeolitic materials. The Science of the Total
Environment, 173/174: 237-246.
Pond, W.G. and F.A. Mumpton, 1984. Zeo-Agriculture,
Use of Natural Zeolites in Agriculture and Aquacul-
ture. Brockport, NY: I.C.N.Z.
Tsitsishvili, G.V., T.G. Andronikashvili, G.N. Kirov, and
L.D. Filizova, 1992. Natural Zeolites. Chichester,
West Sussex: Ellis Horwood Ltd.
Permeable reactive barriers using red mud to treat acid mine effluents
K. Fytas
Dept of Mining, Metallurgical and Materials Engineering, Laval University, Quebec City, PQ,
Canada
M.-E. Desjardins
Genivar, Quebec City, PQ, Canada
ABSTRACT tralization and metal binding capacity (McCon-

chie et al., 2002). Mining and metallurgical ac-
Acid Mine Drainage (AMD) is one of the most
tivities produce very high volumes of waste
serious environmental problems facing the Ca-
rock and finely crushed mill tailings. Production
nadian Mineral Industry. Acid Mine Drainage
of acid mine drainage can occur by chemical
results from oxidation of sulphide minerals (e.g.
and microbiological oxidation of the sulphide
pyrite or pyrrhotite) contained in mine waste or
phase in piles of waste rock or fine tailings that
mine tailings. It is characterized by acid efflu-
are in contact with air and water. The resulting
ents rich in heavy metals that are released into
water is usually high in acidity and dissolved
the environment. A new acid remediation tech-
metals and when the process is started, it’s ex-
nology is presented in this paper by which alka-
tremely difficult to stop it. Overall, acid mine
line metallurgical residues (red mud) from the
drainage is a serious environmental problem
aluminum extraction industry are used to con-
with estimated costs of treatment across the
struct permeable reactive barriers (PRB) to treat
planet in the tenths of billions of dollars (Natu-
acid mine effluents. This paper describes col-
ral Resources Canada, 2000).
umn tests performed to simulate a PRB con-
On the other hand however, the metallurgical
structed using BauxsolTM., a chemically and
industry of aluminium produces every year 70
physically treated bauxite refinery residue, as a
million tonnes of highly alkaline residues (red
reactive material. The results from these ex-
mud) that also represent harmful effects to the
periments show that these permeable reactive
environment. Sustainable development pro-
barriers (PRB) are efficient in neutralizing pH
motes the use of one waste to treat another
and removing metals from acidic mine effluents.
waste. Virotec International developed Baux-
solTM. by treating red mud (a residue from alu-
1. INTRODUCTION mina production) with seawater, or other mag-
nesium- and calcium-rich brines, or magnesium
The utilisation of permeable reactive barriers and calcium chloride salts, or some combination
(PRBs) in treating several groundwater inor- of these materials, to transform it into a near-
ganic and organic contaminants has a long track neutral material that can be used to treat AMD
record of over 15 years. Contaminants can be (McConchie et al., 2003). Several studies have
contained (immobilized) and/or transformed demonstrated that red mud can fix metals and
(into non-toxic forms) within a PRB by adsorp- increase the pH of effluents (Lopez et al., 1996;
tion, absorption, precipitation, redox, or biologi- Doye and Duchesne, 2002), but also showed
cal processes (Naftz et al., 2002). Most of the that this residue is not permeable enough to be
pilot and full scale PRBs have been using zero used directly for a PRB; there can also be prob-
valent iron (Fe0) (Bone et al., 2005). Recently, lems with an initial caustic discharge at start-up.
seawater neutralised red mud and BauxsolTM., This paper describes the behavior of PRBs
have been promoted for the treatment of acid that use BauxsolTM as a reactive material. To
rock drainage because of their high acid neu- simulate PRBs, several mixtures were leached
in Plexiglas columns with acid effluent from the

Doyon Mine in Val d’Or, Canada. The experi-
ments described in this paper are a continuation
of previous work (Lapointe et al., 2005) when
tests were carried out in columns filled with dif-
ferent mixes of BauxsolTM and river sand. Dur-
ing those tests, there were clogging problems
that reduced the lixiviation capacity of the col-
umns. Those problems were caused by the high
iron concentration of the mine effluent used
(around 2500-3000 mg/L) when the rapid in-
crease in the pH caused the precipitation of
large volumes of iron and aluminum oxyhy-
droxides that blocked the columns (Lapointe
et al., 2006).
Figure 1: Schematic diagram of an experimental column.
2. EXPERIMENTAL METHODS AND MA-
TERIALS USED wood chips (0 to 1 cm). The role of these addi-
tional materials was to ensure a homogeneous
Seven experimental columns were set up in or- matrix and avoid blockage or development of
der to simulate groundwater seepage flow, study preferential flow paths of simulated contami-
the reaction kinetics and ultimately predict the nated groundwater flow. Furthermore, they in-
decontamination efficiency and effective life- creased the void volume available for the pre-
time of permeable reactive barriers composed of cipitation of iron and aluminum oxyhydroxides,
red mud. The reactive material studied in these which previously caused clogging of the col-
experiments is a pelletized form of alkaline umns (Lapointe et al., 2005). A 10 cm layer of
residues from the aluminum extractive industry. quartz 2095 (95 % passing a 20 micron sieve)
The pellets were provided by Virotec and are was placed above and below the reactive me-
called BauxsolTM. pellets. The composition of dium in the column to ensure good flow distri-
these residues is: 32,3 % iron oxides & oxyhy- bution. Ten sampling ports were installed every
droxides, 18,2 % hydrated alumina, 17,5 % so- 10 cm along each column. The ports were ad-
dalite, 7,1 % quartz, 6,6 % cancrinite, 5 % TiO2, justed to the column wall, and a glass syringe
4,6 % Ca(Al) hydroxides & hydroxycarbonates, was inserted to allow sampling. Table 1 shows
3,9 % Mg(Al) hydroxides & hydroxycarbonates, the different mixes used in the experimental
2,3 % calcium carbonates (and others 2,5 %). columns. Figure 2 shows the particle size distri-
They correspond to a metal binding capacity of bution of the ground BauxsolTM. pellets.
over 1,500 meq of metals/kg. The pellet form is Mine waste effluents from the Doyon Mine
preferred because of its high porosity, as princi- in Val d’Or were used to simulate contaminated
pal reactive material of the PRB. The CaCO3 groundwater flow. The pH of these effluents is
equivalent (alkalinity) of the pellets was esti- 2,2 and contains high concentrations of heavy
mated to be 0.3 kg CaCO3 eq/g (Lapointe et al., metals (Table 2). We chose these effluents in
2006). order to introduce the “worse case scenario” for
A series of 7 experimental columns in a row the clean up of contaminated plumes in mining
was used. Each 100cm long and 15 cm internal sites. The Doyon mine spends a few $ million
diameter PlexiglasTM column (Fig. 1) was every year for the neutralization of these efflu-
packed with ground BauxsolTM. pellets mixed ents and this will continue for at least another 30
with various additional materials at different to 40 years (i.e. well beyond the end of mine
proportions. The additional materials used were life).
sand, ceramic hollow cylinders (1 cm long and All seven columns were leached (using a
0,5 cm of diameter), synthetic sponges variable multi-channel speed pump) from bot-
(1 cm x 1 cm), quartz, charcoal (of an average tom to top twice a week at a flow rate of
size of 1 cm), volcanic rock (0 to 2 cm) and 10 mL/min for three hours. Leaching tests were
Table 1: Composition of columns (G. Bauxsol: Ground BauxsolTM.).

Void Hydraulic Treated Quantity of
Porosity Column life
No Mix ratio conductiv- volume Bauxsol/L
(%) (days)/comments
(e) ity k(m/s) (L/day) treated
1 G. Bauxsol/sand 17.96 0.22 - 0.25 47.01 127/clogging
G. Bauxsol/ceramic -3 98/Second best, break-
2 52.55 1.11 1.1x10 0.38 54.19
cylinders through
-6
3 G. Bauxsol/Sand 21.95 0.28 1x10 0.21 61.32 103/clogging
4b Bauxsol/Quartz 45.61 0.84 1x10-4 0.28 57.68 91/breakthrough
5 G. Bauxsol/sponges 92.64 12.60 7.5x10-4 0.49 41.59 116/Best, breakthrough
6 G. Bauxsol/charcoal 46.67 0.88 - 0.47 55.49 111/breakthrough
G. Bauxsol/volcanic
7 52.72 1.12 8.8x10-5 0.33 93.24 66/breakthrough
rock
discontinuous in order to better represent likely volumes required for taking these measurements
field conditions. The pH and Fe, Zn, Ni, Cu, and with typical probes. Filtered solutions of the ef-
Pb concentrations in the outgoing leachate were fluent and the sampling ports using a 0.45μm
determined weekly, using a PerkinElmer 3110 filter were analyzed for all heavy metals using
atomic absorption spectro-photometer. The atomic absorption spectrophotometer and sul-
mineralogy and composition of the reactive ma- phates (colorimetry). At the end of each ex-
terial in the columns were characterized before periment, solid samples (consisting of reactive
and after leaching by X-ray diffraction with a materials and precipitates) will be examined by
Siemens D5000 diffractometer using Cu radia- XRD and microscopy techniques in order to
tion (40 kV/30 ma), SEM using a JEOL 840- study the nature and the properties of the pre-
with an EDS Tracor-Northern detector, and neu- cipitates formed and the geochemical reactions
tron activation using a Slowpoke reactor to irra- taken place.
diate the sample and two GEM-20180P co-axial
detectors, with a resolution between 1.8 keV
and 1332 keV, as well as higher flow rates in 3. TEST RESULTS
order to account for seasonal variations and pre- In the beginning of the test, BauxsolTM. demon-
dict the effective life time of the barriers. The strated a strong alkalinity for 15-20 days, until
samples were taken very slowly to avoid distur- the material reached equilibrium. Following this
bance of the flow. Samples were analyzed for equilibrium, the pH of the outgoing leachate
pH with a pH/Conductivity meter (OAKTION). from all columns stayed around 7 until break-
The pH profiles for each column were generated through (or clogging as was the case for col-
less frequently because of the higher sample umns 1 and 3).
Table 3 shows the average quality of outgo-
ing leachate from each column as opposed to
the average ingoing acid effluent and the metal
effluent regulations in Quebec. Overall, with the
Table 2: Characteristics of the acidic effluent used during

experiments.
pH 2,2
Acidity 16500 meq/CaCO3
Conductivity 265 mSiemens
Cu 14.5 mg/L
Pb 0.49 mg/L
Zn 7 mg/L
Ni 3.2 mg/L
Figure 2: Particle size distribution of the ground Baux- Fe 4222 mg/L
solTM. Pellets. Al 2089 mg/L
Table 3: Column performance rankings.

Parameter/Column pH Fe (ppm) Cu (ppm) Zn (ppm) Ni (ppm) Pb (ppm)
Acid effluent 2.2 4222 14.5
1 9 1.5 0.25 0.02 0.01 0.05
2 7 2 0.1 0.25 0.35 0.45
3 7.7 0.5 0.1 0.03 0.025 0.6
4b 8 2.4 0.08 0.3 0.4 0.5
5 7.7 0.5 0.1 0.04 0.2 0.5
6 6.5 1.5 0.08 0.1 0.45 0.3
7 7.5 1 0.05 0.05 0.3 0.35
Directive 019 6.5 to 9.5 3 0.3 0.5 0.5 0.2
exception of Pb, the treated water quality during our acid effluent has very high in iron (4222
the column tests met all metal effluent regula- ppm) and aluminum concentrations (2089 ppm).
tions in the province of Quebec according to Di- In order for the PRB material to be efficient it
rective 019 (Gouvernement du Québec, 2005). must present a lot of interstitial space and fine
pores in order to act like a filter and to accom-
modate the high volumes of iron- and aluminum
4. OVERALL COLUMN EFFICIENCIES oxyhydroxides that will inevitably precipitate. It
Three indexes of overall column efficiencies in appears likely that the available void ratio may
treating the acid effluent were defined: the be a major limiting factor in relation to the per-
treated volume per day, the treatment capacity formance of PRBs that may be used to treat
(i.e. the quantity of reactive material in g of AMD water with a high iron and aluminum con-
BauxsolTM. pellets per volume {L} treated) and tent.
the total column life (days) before clogging or
breakthrough. As we can notice in Table 1, col-
5. METAL REMOVAL EFFICIENCY AND
umn 1 (ground BauxsolTM. pellets/sand) was the
best in terms of column life. It lasted 127 days CAPACITY
before clogging. The second best was column 5 Table 5 presents the average metal removal effi-
(ground BauxsolTM. pellets/synthetic sponges) ciency (%) for each column and for each metal.
that lasted 116 days before breakthrough. Col- It was determined by calculating the total metal
umn 6 (ground BauxsolTM. pellets/Charcoal) concentration reduction from the inflow to the
was the third best and lasted 111 before break- outflow. We can notice that with the exception
through. However, if we consider the treatment of Pb, the columns remove almost completely
capacity (quantity of reactive material {g of the metals examined (i.e. Cu, Fe, Zn and Ni).
BauxsolTM. pellets} per L treated) the best per- Lead is the first metal to reach its loading limit
former was column 5 (ground BauxsolTM. pel- whereas Fe continued to be removed long after
lets/synthetic sponges) since it corresponded to the columns were saturated with respect to the
the lowest quantity of reactive material per L
treated (41,59 g of BauxsolTM. pellets/L). The Table 4: Column performance rankings.
second best was column 1 with 47 g of Baux- Column Rankings (1 is the highest)
solTM. pellets/L and the third best was column 2 Treated
Treatment ca-
Column
with 54,19 g of BauxsolTM. pellets/L. Column
volume
pacity (quan-
life
On the other hand, considering the total vol- tity of Baux-
(L/day) (days)
ume treated per day, column 5 was the best fol- sol/L treated)
lowed by columns 6 and 2. Table 4 summarizes 1 6 2 1
the rankings of all columns for each index and 2 3 3 4
can conclude that the best column considering 3 7 6 6
all three indexes was column 5 (ground Baux- 4b 5 5 5
solTM. pellets/synthetic sponges). This was ex- 5 1 1 2
pected since column 5 corresponds to the high- 6 2 4 3
est porosity (92,6%) and void ratio (12,6) and 7 4 7 7
Table 5: Column metal removal efficiencies (%).

COLUMN 1 2 3 4B 5 6 7 4B
Parameters Mean Mean Mean Mean Mean Mean Mean Mean
Fe (%) 99.9 100 100 99.9 99.9 97 99.9 99.9
Cu (%) 99.4 93.4 99.5 99 95.2 99 95 99
Zn (%) 99.7 60 99.4 98.2 94 98 90 98.2
Ni (%) 97.7 84 96 89.5 88 94 90 89.5
Pb (%) 90.2 52 42 80 48 66 68 80
other metals. The sequence of metal loading along the columns. The last row of the table in-
limits displayed by the columns follows the fol- dicates the pH at which precipitation mainly
lowing order (from less efficient to more effi- takes place for each metal (Eary, 1999).
cient): Pb < Ni < Zn < Cu < Fe. These results We can notice from these data that for most
are similar to those obtained by Lapointe et al., columns and heavy metals complete metal re-
2006. Furthermore, a comparable study of a moval takes place at a pH that is 1 to 2 pH units
PRB constructed by Munro et al., (2004) using (orders of magnitude in relation to hydrogen ion
very fine-grained BauxsolTm. (>90% of the ma- concentration) lower than their precipitation pH.
terial <10 µm) showed a similar sequence of It can be noticed for example in Table 6 that
metal loading limits following the increasing with the exception of column 4b, zinc, copper
pKa of hydration values and suggested that and lead are removed at a pH which is 1 to 2 or-
metal removal processes (for Cu, Zn, Ni and Pb) ders of magnitude lower than their precipitation
are controlled by the equilibrium formation of pH (in relation to hydrogen ion concentration).
the hydrous metal species (MOH+). However, On the other hand Ni is removed at a pH which
they had also reported that the continuing re- is almost lower by 3 orders of magnitude. This
moval of Fe is not consistent with the pKa of is a clear indication that the principal mecha-
hydration values. This inconsistency suggests nism by which BauxsolTM. particles remove
that metal removal processes other than forma- heavy metals from an acid effluent is not pre-
tion of MOH+ and binding to the surfaces cipitation resulting from the increased pH as
BauxsolTM. particles may also be occurring; acid is neutralized but by absorption of metal
such processes which may include direct ab- cations onto the BauxsolTM. particles. It is also
sorption, and/or co-precipitation, isomorphous reasonable to assume that metal uptake involves
substitution and solid state diffusion. chemisorption and not simple surface adsorption
because the ion exchange capacity of the Baux-
solTM. material is less than 40 meq/kg (McCon-
6. MECHANISMS OF METAL REMOVAL chie et al., 2003) compared to a metal uptake
There are two main mechanisms that control capacity in excess of 1,500 meq.kg. Other data
heavy metal removal processes in a liquid/solid (McConchie et al., 2003) suggest that processes
system such as in a permeable reactive barrier: including coprecipitation, isomorphous substitu-
precipitation and adsorption. Precipitation is tion and solid state diffusion may make an im-
mainly controlled by the solution pH and redox portant, even if minor, contribution to metal
potential and it takes place at a specific pH zone
defined by the solubility curves of each sub- Table 6: Average pH of metal removal in each column.
stance. One of the objectives of the columns ex- Average pH of metal removal in
periments was to investigate the relative impor- Column each column
tance of precipitation and adsorption onto the Cu Pb Zn Ni Fe
mechanisms of heavy metal removal of a Baux- 2 7 7 7 5 2.5
solTM. based PRB. Table 6 shows a summary of 3 6.8 8.3 6.8 6.8 8.5
the average pH zone where each of the heavy 4b 8.7 8.7 8.7 8.7 8.5
metals (Cu, Fe, Ni, Pb and Zn) was mainly re- 5 7.1 8 7.1 7.1 8.5
moved from the leaching solution. The average 6 6.5 6.5 7.1 7.1 7.5
7 7.3 7.3 7.3 7.3 7.3
pH of each metal removal is based on the analy-
Precipitation
ses of samples taken through the sampling ports 8-10 9-10 8-9 10 7-8
pH
binding by BauxsolTM. particles; the importance Cambior for his continuous interest and encour-
of these processes appears to increase as metal agement.
saturated BauxsolTM. particles age.
REFERENCES
7. CONCLUSIONS Bone, R.D., J.W. Harris, G.A. Smith, R.E. Boshoff, R.T.
An experimental method was developed in a Kalin and P. Morgan, 2005. The development and use
of permeable reactive barrier technologies and poten-
previous study to simulate permeable reactive tial future applications in the UK, Permeable Reactive
barriers using BauxsolTM. as a reactive material. Barriers, edited by Genevieve A. Boshoff & Brian
Here, the use of five different mixes were Bone, International Association of Hydrological Sci-
evaluated to treat acid rock drainage was evalu- ences, IAHS Press.
ated. The objective was to increase pH, remove Doye, I. and J. Duchesne, 2002. Neutralisation of acid
metals and avoid the clogging of the columns mine drainage with alkaline industrial residues: labo-
ratory investigation using batch-leaching tests. Ap-
previously caused by excessive iron and alumin- plied Geochemistry 18: 1197-1213.
ium oxyhydroxide precipitation. The results of Gouvernement du Québec, 2005. Directives 019, Min-
the tests show that PRBs using BauxsolTM. have istère du Développement durable, de l’Environnement
the potential to treat acidic effluents and remove et des Parcs, Direction des politiques de l’eau, Service
metals over a long period of time; most metals des eaux industriels.
Lapointe, F., K. Fytas and D. McConchie, 2005. Using
are trapped in the cores of pellets and are likely permeable reactive barriers for the treatment of acid
to be very difficult to remove by conventional rock drainage. International Journal of Surface Min-
leaching reagents (e.g. using the TCLP test). ing, Reclamation and Environment 19(1): 57-65.
However, due to slow reaction kinetics, the ratio Lapointe, F., K. Fytas and D. McConchie, 2006. Effi-
of reactive mixture to acidic effluent would ciency of BauxsolTm. in permeable reactive barriers
have to be increased to attain the full potential to treat acid rock drainage, Mine Water and the Envi-
ronment, 25, pp 37-44.
of the treatment. Future work should consider Lopez, E., B. Soto, M. Arias, A. Núñez, D. Rubinos and
both long-term tests and large scale in situ ap- M.T. Barra, 1996. Adsorbent properties of red mud an
plications of BauxsolTM. in a PRB. Estimation of its use for wastewater treatment. Water Research
the required PRB thickness and void volume in 32(4): 1314-1322.
the PRB should be also carried out in order to McConchie, D., M. Clark and F. Davies-McConchie,
2002. New strategies for the management of Bauxite
meet the environmental standards for mine ef- refinery residues (red mud), Proceedings of the 6th In-
fluents. In the future, it may be desirable to de- ternational Alumina Quality Workshop, 8-13 Septem-
velop design criteria for both BauxsolTM. - based ber, Brisbane, pp. 327-332.
PRBs, and for PRBs constructed using other re- McConchie, D., M. Clark, G. Maddocks, F. Davies-
active media, that specify the void volume in the McConchie, S. Pope and W. Caldicott, 2003. The use
PRB required to accommodate the Fe and Al of BauxsolTm. technology in mine site management
and remediation. Proceedings of the CIM Mining In-
oxyhydroxide precipitates that could form from dustry Conference, Montreal, Canada.
the effluent being treated. The same applies for Munro, L.D., M.W. Clark and D. McConchie, 2004. A
the flow path length in the PRB required to pro- BauxsolTm.-based permeable reactive barrier for the
vide sufficient residence time for complete reac- treatment of acid rock drainage. Mine Water and the
tion and the cross-sectional area of the PRB Environment 23: 183-194.
Naftz, D.L., S.J. Morrison, J.A. Davis and C.C. Fuller,
perpendicular to the flow direction required to 2002. Handbook of groundwater remediation using
handle the applicable flow rates. permeable reactive barriers, Academic Press, pp 4-8.
Natural Resources Canada, 2000. MEND Manual, Report
5.4.2. CD-ROM.
ACKNOWLEDGEMENTS
We thank Virotec International Ltd., the Societé
IAM Gold Inc. and the National Science and
Engineering Research Council of Canada
(NSERC) for their financial support and for
their permission to publish the research results.
In particular, we thank Dr. Serge Vezina, former
V.P. Industrial Engineering and Environment of
Recovery of waste lubricant oil contained in used automotive filters
G. Gaidajis and K. Angelakoglou

Department of Production Engineering and Management, School of Engineering, Democritus Uni-
versity of Thrace, Xanthi, Greece
ABSTRACT According to data from the National Statis-

tics Office in Greece (2009), the number of pas-
About 7.2 million automotive filters utilized in
senger cars that were used regularly in 2007 was
passenger cars are annually consumed in
approximately 4.800.000. Considering that a
Greece, with only a small percentage of them
passenger car travels annually an average
being recycled and the rest being disposed in the
15.000 kilometers (km), and the majority of the
environment. In other words, based on our ex-
automobile industries recommend a change in
periments it is estimated that a maximum of
the used oil filters every 10.000 kilometers, it is
1260 tons of used oil and 1713 tons of steel are
estimated that 7.200.000 lubricant filters are
disposed every year in Greece, as a result of the
used in Greece every year, taking into account
non-effective management of used automotive
only the filters used by passenger cars.
filters. In this paper, the factors affecting the re-
It is obvious that due to the great develop-
covery of the lubricant oil contained in used
ment of the transportation sector, the overuse of
automotive filters are evaluated and quantified
passenger cars and the characteristics of the
through experimental measurements. Further-
automobile industries, trying to find better tech-
more, a Life Cycle Assessment (LCA) of a used
niques for reusing or recycling even the smaller
automotive oil filter containing lubricant oil is
parts of a vehicle, could have a significant im-
performed, in order to evaluate and quantify the
pact on the environment.
environmental benefits arising from the applica-
tion of a used filter recycling scheme.
2. EXPERIMENTAL
1. INTRODUCTION 2.1 Filter Description and Choice
Transportation is considered to be one of the A typical automotive oil filter has a cylindrical
most developing sectors in Europe. It is esti- shape with a closed dome on the upper side. In-
mated that between 1990 and 2004, vehicles in ternally, there is a paper-made cylindrical filter
Europe increased by nearly 40%, whereas since enhanced with cellulose, through which the oil
then the annual increase continually grows. In- flows, and the undesirable components are fil-
dicatively, there were 216 million passenger tered. At the bottom of the filter there is a valve
cars in total in the EU25 in 2004, according to that averts the non-filtered oil from flowing in
the European Statistics Office (Eurostat, 2006). the engine when it is turned off. The particular
The impacts of such an increase are mainly at- valve is also the inhibitory factor that prevents
mospheric pollution, a rise in the consumption the oil from leaking during the removal of the
of raw materials and fuels, noise pollution, traf- filter from the vehicle. The shell, the base and
fic accidents and the need for treatment of waste the central chamber of the filter are made of
products such as tires, oils, etc. All those aggra- steel. There are generally two kinds of automo-
vate the quality of urban environment and entail tive oil filters, the light duty filters used in pas-
significant financial cost. senger cars and motorcycles and the heavy duty
filters used in trucks, buses etc. The light duty

(LD) filters comprise 75% of the oil market
while the heavy duty (HD) filters comprise the
remaining 25% (Peaslee and Roberts, 1997a).
The majority of LD oil filters are relatively
small, with a diameter between 6 and 12 cm and
a height between 7 and 15 cm.
The disposal of the used filters without fur-
ther treatment to remove the lubricant which
remains confined, not only wastes precious re-
sources of steel, petroleum and landfill volume,
but is also hazardous for the environment, as the
lubricant contained in the filter will eventually
pollute both the soil and the water streams
nearby. The most developed techniques used for
the removal of the used oil are simple infiltra-
tion, punctuation of the upper dome and infiltra-
tion, dismantling (using this technique the steel Figure 1: The experimental arrangement used in this
is also reclaimed) and shredding (using this study.
technique the steel is also reclaimed).
Due to the large variety of filters, this study flow gives remarkable quantity of lubricant.
focused on specific filters. For their selection,
their market share in Greece was used. Accord- 2.3 Sampling
ing to data from the Association of Motor Vehi- Several blank filters of each filter type were
cle Importers-Representatives (AMVIR, 2009) weighted prior their use in order to estimate
in Greece, the top three best-selling car indus- their net weight, without the presence of used
tries for 2007 were TOYOTA (Filter Type 1) oils, metal scrap and other substances within the
with a share of 9,9%, OPEL (8,3%, Filter Type filter body. Then, every used filter was weighted
2) and VOLKSWAGEN (7,9%, Filter Type 3), again solely, so as to figure out the weight of
resulting to a combined 26,1% of total sales in embodied lubricant (Table 1).
Greece. The measurements were conducted in a con-
2.2 Sampling arrangement trolled environment room where the temperature
was maintained constant at 25°C. For low tem-
The aim of the present experimental study was perature measurements the filters were placed
to examine the maximum capability of infiltra- outdoors on periods of cold temperatures (0°C).
tion of the embodied lubricant oil contained in Used filters were stored in shelves; during their
used filters. The most important factors affect- transportation filters were kept at an upside
ing the infiltration of the filter is the angle, the down position.
punctuation of the dome, as well the tempera- The concept that was followed during sam-
ture and the viscosity of the embodied oil (Rob- pling aimed to investigate only one factor per
erts and Peaslee 1998). The filters were placed time. Then, by using the best factor according to
in different orientation in order to examine the the results, we proceeded to the examination of
above mentioned factors (Fig. 1). the next factor and so on. As a result, a best case
Furthermore, the results were examined ac- scenario was identified. The flow chart of the
cording to the duration in which the observed experiments is shown in Figure 2 along with the
Table 1: Net and gross weight of the filters.

Net weight of fil- Gross weight of Net weight of Standard deviation of
Filter Type
ter (kg) filter (kg) oil (kg) net weight of oil
Type 1 (n=51) 0.218 0.351 0.133 0.016
Type 2 (n=51) 0.260 0.420 0.160 0.018
Type 3 (n=51) 0.364 0.598 0.234 0.019
Pierced Non-Pierced 3. RESULTS AND DISCUSSION

Filter Filter
A description of the percentage of infiltration
for every stage examined is given on Figure 3.
Pierced
Best percentage
of infiltration Results have shown that the maximum infil-
Filter tration (Table 2) is achieved under higher tem-
peratures (25οC), on filters that have been punc-
tuated and their infiltration was performed at 0ο
0° angle 30° angle 60° angle 180° angle angle (dome being on the top, Fig. 1a). More-
over, at a time period greater than 600 seconds
(10 minutes) the infiltration rate was practically
close to zero (Zokas, 2008).
Best percentage According to these results, an easy, simple
0° angle of infiltration
and quick protocol can be followed at the work-
shops where the filters of the cars are changed,
which will include:
– filter dome puncturing,
Temperature Temperature – infiltration for 10 minutes,
0°C 25 °C
– at 00 angle,
– at the highest temperature available
Temperature (room).
25 °C Best percentage By following these simple steps, 50-55% of
of infiltration used oil that is trapped in the filter, can be re-
Figure 2: Flow chart of the experiments performed. covered. Taking into account that the amount of
experimental conditions that produced the most this oil was found to be an average 175 g then
favorable infiltration results (grey boxes). 630-693 tons of lubricant oil from filters can be
During the experiments, 153 filters were used incorporated to the quantity of lubricants for re-
and 3813 measurements were performed. The cycling. Furthermore, if a recycling method
average duration of the experiment for every fil- such as dismantling or shredding is used, then a
ter was 30 minutes. considerable amount of steel can be reclaimed.
In our experiments the average weight of our
filters before their use was 280 g in accordance
Figure 3: Percentage of infiltration for all stages examined.

Table 2: Description of the experiments performed.

Percentage of Infiltra-
Number of filters Angle (degrees) Temperature (°C) Punctuation
tion
Stage 1 3 0 25 Yes 50-51%
3 0 25 No 9-12%
Stage 2 60 0 25 Yes 50-55%
45 30 25 Yes 45-50%
30 60 25 Yes 28-34%
15 180 25 Yes 15-17%
Stage 3 3 0 0 Yes 30-35%
with the average light duty oil filter proposed by ess of the filter, and only its material content is
Oil filter Manufacture Council (283 g) (Smailer important, since for the recycled and non recy-
et al., 2002). Considering that a new oil filter is cled filters the impacts due to processes and en-
made of 85%wt steel (Peaslee and Roberts, ergy use will not differ; furthermore, no reuse is
1997b), the saving for 100% recycling of used possible. In other words only extraction and
automotive oil filters would be 1713 tons of processing of raw materials and waste scenarios
steel. However this is a very optimistic state- will be taken into account, setting the bounda-
ment as 100% recycling of steel from used oil ries of the LCA. During the use phase it was
filters is most of the times practically impossible considered that the filter is not consuming any
due to scrap quality and steel disposed with the energy or material thus it does not affect the as-
filter media. sessment.
In addition, because no comprehensive data
from the filter’s manufacturers were available
4. USING LCA TO ANALYZE THE IM- (such as exact type of steel, processes and ma-
PACTS OF FILTER RECYCLING chines used for the assembly, energy consumed,
During the experimental part it was concluded etc.), some assumptions and simplifications had
that there is a potential of saving a maximum of to be made. Only data provided by SimaPro da-
1260 tons of used lubricant oil and 1713 tons of tabases were used.
steel coming from the disposal of untreated used In order to examine the goal of this study,
automotive oil filters. Yet, in order to fully un- two different scenarios were created. For the
derstand the positive environmental impacts of pessimistic scenario, the filters are disposed to a
saving these amounts, further analysis is landfill untreated, including all the quantity of
needed. Environmental impacts caused by addi- oil remaining within their body. No further
tional activities in order to recycle, separate, de- processes are entered in the life cycle for this
liver the filters to recycling facilities, etc. need scenario. For the second (optimistic) scenario, it
to be taken into account. Therefore the Life Cy- was considered that a lorry (7.5-16 tons capac-
cle Assessment (LCA) method was chosen for ity) collects the filters and transports them to re-
our analysis. cycling facilities where 100% of the steel and
In general, LCA is a methodology developed oil are recycled, whereas the cellulose (filter)
to examine the environmental impacts of a prod- and the rubber (gasket) are disposed in a land-
uct or a system considering all steps, from fill. The functional unit was one (1) average
manufacture to end of life. Although LCA is not used light duty oil filter.
fully developed, some standard steps have been The impact assessment method used in this
created for those who wish to implement it (ISO study was Eco-indicator 99 (EI99). A non-
14040 series). The current LCA was performed metric unit, called Points (Pt), is used in Eco-
using the SimaPro 7.1 software. indicator to evaluate the impact of a product or
The goal of the current LCA is to evaluate process. More points mean worst efficiency for
and quantify the environmental benefits arising the environment.
from the application of a used filter recycling Results have shown that recycling is prefer-
scheme. For the specific LCA analysis there is able when it comes to the issues of resources
no need to define the complete production proc- depletion and human health deterioration. How-
Figure 4: Single scores for the two scenarios examined (note that less Pt is better for the environment).
ever, there is no significant difference on eco- affecting the infiltration of used filters were ex-
system quality. More specifically, use and de- amined, proposing a simple framework that can
pletion of fossil fuels and minerals have the save up to 693 tons of lubricant oil every year.
most negative impact and are those factors Furthermore a Life Cycle Assessment (LCA) of
greatly affected by a filter’s life cycle. Of less a used automotive oil filter containing lubricant
importance, but generally high, is the produc- oil was performed, in order to evaluate and
tion of respirable inorganics and the effect to quantify the environmental benefits arising from
climate change. In all categories except ecotox- the application of a used filter recycling scheme.
icity recycling is better than untreated disposal. Results have shown that recycling seems prefer-
A single score for every scenario is given in able from an environmental point of view.
Figure 4. As a final statement we can claim that However the LCA performed was more like a
recycling oil filters makes sense from an envi- screening life cycle rather than a complete one.
ronmental point of view. However the differ- From that aspect more data are needed in order
ence between recycling and disposal was not to fully evaluate our goals.
found to be very significant. This is probably
because, first of all, many simplifications have
been made and also the results from recycling REFERENCES
the lubricant oil are unspecified due to lack of Association of Motor Vehicle Importers-Representatives
data from the software. AMVIR, 2009. Greece, http://www.seaa.gr/common/
statistics.asp?language=english and
http://www.seaa.gr/data/pdf/2007-12-b.pdf (access 19-
5. CONCLUSIONS 01-2009).
Eurostat News Release, 125/2006-19 September 2006,
The competition of car industries, the develop- issued by Eurostat Press Office.
ment of technology, the opening of new mar- Greek National Statistics Office, http://www.statistics.gr/
kets, and the advantages of using cars, result in gr_tables/S602_SME_3_TB_AN_07_2_Y_BI.pdf
(access 19-01-2009)
a constant increase of the number of cars world- Peaslee, K.D. and D.E. Roberts, 1997a. The Future of
wide. Consequently, their impact on the envi- Used Oil Filter Recycling in Missouri: An Evaluation
ronment is growing. In Greece, the underling of Potential Processes, Product Quality, Recycling
practices for the treatment of waste products Locations, and Economics. Technical Bulletin.
such as the used automotive filters need further Peaslee, K.D. and D.E. Roberts, 1997b. Characterization
of used automotive oil filters for recycling. Resources,
improvement. In this paper, some basic factors Conservation and Recycling, 19, pp. 81-91.
Smailer, R.M., G.L. Dressel and J.H. Hill, 2002. A Feasi-

bility Study for Recycling Used Automotive Oil Fil-
ters in a Blast Furnace, AISI Contract TRP-9, Project
No.0041, American Iron and Steel Institute, January
2002.
Roberts, D.E. and K.D. Peaslee, 1998. Analysis of the
factors influencing the removal of oil from used auto-
motive oil filters during recycling operations. Re-
sources, Conservation and Recycling, 22, pp. 97-113.
Zokas, Z., 2008. Method Development for the recovery of
waste lubricative oil contained in used automotive fil-
ters. Graduation Thesis, Department of Production and
Management Engineering, School of Engineering,
Democritus University of Thrace.
Crystalline silica dust determination in limestone aggregate quarries in

Greece
M. Galetakis, G. Alevizos and I. Aspirtakis

ABSTRACT mostly of quartz, where as the quartz content of

crushed stone will vary from region to region
The aim of this study was to determine the per-
(USBM, 1992; Watts and Parker, 1995).
sonal exposure to respirable dust and free crys-
Crystalline silica dust has been recognized as
talline silica in an aggregate quarry. Personal
an occupational health hazard to the respiratory
and stationary samples were obtained and ana-
system since chronic overexposure to respirable
lyzed by X-ray diffraction (XRD). Dust surveys
silica dust (particle diameter < 10 microns)
were also conducted by on-line dust analyzers to
leads to the progressive lung disease known as
investigate the main dust sources. Results indi-
silicosis (IARC, 1997; American Thoracic Soci-
cated that the exposure levels for respirable
ety, 1997).
crystalline silica are below the permissible lim-
In surface mining this silica dust is created
its. However the estimated limits for the sever-
from several sources and activities such as drill-
ity showed that the monitoring of the dust must
ing or crushing of rock material. In the quarry-
be continued and possible actions to reduce the
ing sector and particularly in crushed limestone
occupational risk must be examined.
aggregates quarries, crystalline silica is found in
small quantities as an inherent constituent of the
1. INTRODUCTION limestone deposit.
Occupational Safety and Health Administra-
The generic term silica refers to the chemical tion (OSHA) has an established Permissible Ex-
compound silicon dioxide (SiO2), which occurs posure Limit (PEL), which is the maximum
naturally in crystalline and amorphous states. amount of crystalline silica to which employees
Crystalline silica exists in seven different forms may be exposed during an 8-hour work shift.
four of which are extremely rare. The three ma- Dusts containing more than 1% of crystalline
jor forms, quartz, cristobalite, and tridymite, are silica are considered fibrogenic and PEL is es-
stable at different temperatures. In nature, quartz timated as below:
in its alpha, or low form, is a very common
mineral. It is a major component of rocks, and ⎧ 10
mineral ores and is the second most common ⎪⎪ q + 2 if q > 1%
mineral in the earth’s crust, next to feldspar. PEL ( mg / m ) = ⎨
3 (1)
Because of its abundance in the earth, silica, in ⎪
both its crystalline and non-crystalline states, is ⎩⎪ 5 if q ≤ 1%
present in nearly all mining operations. It is in Where q is % wt of crystalline silica in the
the host rock, in the ore being mined, as well as respirable dust and PEL is expressed as a con-
in the overburden. Most ores are mined from centration of respirable dust in milligrams per
deposits containing crystalline silica. The min- cubic meter. For inert dusts, containing less than
eralogy of the deposit and, to some extent, the 1% of crystalline silica, PEL is 5 mg/m3. In the
processing of the ore determines the quartz con- Greek Regulations for Mining and Quarrying
tent of the final product. Sand and gravel consist (GRMQ, 1985) the above limits have been
adopted for respirable dusts containing free diffractometry, indicated that it is consisting
crystalline silica. Monitoring the respirable mainly of calcite (85-99%), dolomite (0.5-8%)
quarry dust exposure for compliance with regu- and other minor minerals such as quartz (0.2-
lations is mandatory to protect the health of 1.0%), clay minerals (<0.5%) and iron oxides
working personnel. (<0.5%).
In this study total and respirable full-shift The aggregates are produced via traditional
personal and fixed-position dust samples were quarrying methods. The limestone is obtained
collected from several workplaces of a lime- by blasting a quarry face, then crushing and
stone aggregate quarry, in Crete, Greece. Sam- screening material to produce a wide range of
pling and analysis of crystalline silica was de- aggregates in various sizes and to specified
termined according to MDHS101 (HSE, 2005). quality limits.
According to this method a sample of respirable As a consequence, workers are exposed to
dust is collected on a membrane filter using a the risk of inhaling dusts during all of the quar-
respirable dust sampler. The filter is then placed rying phases (drilling, loading, conveying and
directly into the sample beam of either an infra- dumping) as well as during crushing, screening
red spectrophotometer or an X-ray diffractome- and handling of produced aggregates and the as-
ter. The mass of crystalline silica on the filter is sociated by-products.
determined from the infrared or X-ray diffrac-
2.2 Sampling and analysis
tion response, calibrated against filters loaded
with known amounts of standard quartz or cris- To determine the concentration of total and res-
tobalite or tridymite (Virji et al., 2002). Since pirable dust from all the emission sources of the
the volume of air sampled is known, the concen- quarry several samples were collected. The dust
tration of airborne crystalline silica is readily survey was conducted at the dump/crusher facil-
calculated. The choice of analytical technique, ity, on trucks and front-end wheel loader drivers
i.e. infrared spectroscopy or X-ray diffractome- to determine which activity during the produc-
try, depends largely on other materials present tion cycle generated the greatest potential for
on the filter which may interfere in the analysis exposing the operator to silica dust. Respirable
(Page, 2006). In this study the X-ray diffracto- airborne dust was sampled using a suitable sam-
metry was used. pler, filter and an air pump.
The Higgins-Dewell conductive cyclone
sampler with 25 mm diameter filters was used
2. MATERIALS AND METHODS
2.1 Geological characteristics of quarry and
phases of extraction
For the purposes of this project, airborne dust
samples (total and respirable) were collected a
quarry located in the area of the Island of Crete,
at Southern Greece. The quarry produces
800000 t of aggregates annually and supplies
the nearby concrete and asphalt-mixture manu-
facturers. Aggregates are extracted from car-
bonate sedimentary formations consisting main-
ly of limestone in the upper part, and from
dolomitic limestone and dolomites in the lower
part. Maximum stratigraphic thickness of these
formations is at places measured up to 300 m.
Aggregates are extracted mainly from lime-
stone layers, which are medium bedded, and are
visually distinguished from its light grey to dark
grey color. The mineralogical composition of Figure 1: Personal sampler for the respirable airborne
the extracted limestone, determined by X-ray dust.
(Fig. 1). The selected type filter was the GF/A-

1.6 μm-25 mm type. This type of filter has a
capture efficiency greater than 90% and is suit-
able for collection of samples containing free
crystalline silica. It was also suitable for X-ray
diffraction analysis since it has have no interfere
with the diffraction peaks of the dust constitu-
ents.
The SKC pump (Model -224-PCXR8) was
used. This sampling pump has an adjustable
flow rate, and incorporates a flowmeter with a
flow fault indicator (Fig. 1). It is capable of
maintaining the selected flow rate to within
±5% of the nominal value throughout the sam- Figure 2: On-line monitor for respirable airborne dust.
pling period.
time the concentration of the respirable dust
During sampling the cyclone was attached to
must be known. For this purpose a particle ana-
the worker’s lapel and as close to the mouth and
lyzer (TSI DustTrak 8520) which measures and
nose in order to determine respirable dust in the
records the concentration of respirable airborne
breathing zone. Samples of airborne respirable
dust (includes all dust types and particulate) us-
dust at fixed points were also taken.
ing a light scattering technique (Fig. 2). Light-
For fixed-position samples the sampler was
scattering instruments offer only a relative
mounted at approximately head height, away
measure of concentrations but provide a con-
from obstructions, fresh air inlets or strong
tinuous record of dust levels so that concentra-
winds. The sampling procedures are otherwise
tions can be evaluated over any time interval
the same as for personal sampling. Fixed-
during the sampling period. Taking into account
position samples may be useful in identifying
the DustTrak measurements (from 0.7 to
the main sources of crystalline silica exposure.
3.4 mg/m3 with an average value of 2.8 mg/m3),
Comparison of airborne concentration meas-
the airflow rate of the pump the measuring time,
urements from personal and fixed point samples
for both personal and fixed samples, was se-
may give some indication of the extent to which
exposure arises from local or general conditions lected to 420 min. The volumetric flow was set
to 2.2 l/min, to an accuracy of ±5%
(Bahrami et al., 2008).
(MDHS14/316, 2000).
Personal and fixed-position samples were ob-
tained during February 2008. The sampling At the end of the sampling period the volu-
metric flow rate and the sampling time were re-
aimed to monitor truck drivers, loader drivers
corded. The filters were removed from the sam-
and crusher operators through the loading-
transportation-dumping-crushing cycle, to pro- pler and were transported to the laboratory in a
container capable of preventing damage. The
file their dust exposure during these different
filters were conditioned in a desiccator for 24 h
operations. Fixed samples were taken from
crushers, which are considered the main source at 25°C and weighed by a balance with an accu-
racy of 0.01 mg. The same procedure was ap-
of the airborne dust. Samples were also obtained
plied for the determination of weigh of the
from the offices for monitoring the exposure of
the administrative personnel to the respirable empty filters. The weight of the filter before and
after sampling was used to determine the con-
dust.
centration of airborne dust (gravimetric method)
The determination of sampling time is crucial
in sampling airborne dust. A long sampling time according to the following formula:
ensures a heavier deposit of dust on the filter, Wafter − Wbefore
thus reducing measurement inaccuracies. How- D = 1000 ⋅ (2)
VF ⋅ ST
ever overloaded filters are problematic in trans-
portation and particular care must be taken to Where, D is the concentration of airborne
prevent material being dislodged from the filter. dust in mg/m3, Wbefore and Wafter are the weights
For the determination of the optimal sampling of the filter before and after sampling in mg, VF
Table 1: Results of the gravimetric estimation of the respirable dust.

Respirable Volume of Respirable dust
Sampling Sampling rate
Site/employee dust weight air sampled concentration
time (min) (l/min)
(mg) (m3) (mg/m3)
Crusher 420 2.2 2.91 0.924 3.15
Dumper driver 420 2.2 3.2 0.924 3.46
Loader driver 420 2.2 2.74 0.924 2.97
Office 420 2.2 2.38 0.924 2.2
is the flow rate of pump in l/min and ST is the mated.

sampling interval in minutes. 3. RESULTS-DISCUSSION
The analysis was done by X-ray diffraction
The results of the gravimetric determination of
(XRD) using a Brucker AXS (Model D8 Ad-
respirable dust are given in Table 1. The X-ray
vanced) diffractometer. Initially samples of total
diffraction analysis of the total airborne dust is
airborne dust as well as respirable dust depos-
shown in Figure 3. Dust contains mainly calcite
ited on the filters were analyzed. These samples
with, small amounts of dolomite and lesser
were analyzed, without any further treatment, to
amounts of quartz. The analysis of the insoluble
identify the major minerals of the collected dust.
residual of the respirable dust indicated that is
To improve the accuracy of the estimation of
consisting of quartz, muscovite and albite
crystalline silica content in the respirable dust,
(Fig. 4). The free crystalline silica in all ana-
the filters were treated with a 2N HCl solution
lyzed samples is exclusively in the form of
to dissolve calcite and dolomite. The acid in-
quartz (low temperature quartz).
soluble residual was then analyzed by X-ray dif-
The concentration of respirable quartz (expo-
fraction. The mineralogical constituents of the
sure), the permissible exposure levels (PELs)
dust were determined and the concentration of
and the severity were calculated and shown in
crystalline silica (quartz, cristobalite and tridy-
Table 2. Severity is derived by dividing the ex-
mite) in the respirable dust was calculated. Sub-
posure by the PEL.
sequently the concentration of respirable crys-
talline silica in mg/m3 and the respective per-
missible exposure limits (Eq. 1), were esti-
15000
14000
Calcite
13000
12000
11000
10000
9000
8000
7000
6000
Sqrt (Counts)
5000
4000
Calcite
Calcite
Calcite
Calcite
Calcite
3000
Calcite
2000
Calcite
Quartz
Calcite
1000
Dolomite
100
10
16 20 30 40 50
2-Theta - Scale
Figure 3: X-ray diffraction diagram showing the main minerals of the total airborne dust.
2400
2300
2200
2100
2000 Qz = Quartz
1900
1800
Ab = Albite
1700
1600 Mu = Muscovite
1500
Lin (Counts)
1400
1300
Qz
1200
1100
1000
Ab
900
800
Qz
Ab
700
Mu
Ab
600
Mu
Mu
Ab
500
Qz
Mu
Qz
400
Qz
300
200
100
9 10 20 30 40 50 60 70
2-Theta - Scale
Figure 4: X-ray diffraction diagram showing the main minerals of the insoluble residue of the respirable dust.
measured exposure is over the PEL. However, if

Exposure ( mg / m 3 ) LCL is lower than 1 this may indicate that expo-
Severity = (3)
PEL ( mg / m 3 ) sure is below the PEL. In this case the monitor-
ing of the exposure should be continued and it is
Considering that the total sampling and ana-
recommended to take appropriate actions to re-
lytical error for quartz determination is ~0.3%
duce exposure. When severity and LCL is
wt, the confidence limits (Upper Control Limit
greater than 1 the exposure is over the PEL and
or UCL and Lower Control Limit or LCL) were
appropriate actions must be taken to reduce the
estimated. UCL and LCL calculations were car-
exposure.
ried out by using the web-based e-tool for silica
Results, summarized in Table 2, indicated
crystalline, developed by OSHA (2009). Ac-
that the concentration of quartz in respirable
cording to OSHA, a severity less than or equal
dust is below to 1%. Therefore the dust is char-
to 1 means that the measured exposure is under
acterized as inert and the permissible exposure
the PEL. However if UCL is greater than 0.5, it
level is 5 mg/m3. Personal and site sampling in-
is recommended to examine the workplace for
dicated that respirable dust concentration was
possible actions that can be taken to further
below the permissible exposure level. The expo-
lower exposure. When UCL is greater than 1,
sure levels of truck drivers are the highest, fol-
additional measurement are required to confirm
lowed by crusher operators and loader drivers.
that the exposure is below the PEL, and it is
The office employees were the least exposed.
recommended to take appropriate actions to re-
Data from the dust survey, conducted by the
duce exposure.
DustTrak on-line analyzer, confirmed that a ma-
A severity greater than 1 indicates that the
Table 2: Results of the gravimetric estimation of the respirable dust.
Exposure mg/m3 (res- Permissible Ex-
Quartz
Site/employee pirable dust concentra- posure Level Severity UCL LCL
%
tion in air) mg/m3
Crusher 3.15 0.85 5 0.63 0.93 0.33
Dumper driver 3.46 0.65 5 0.69 0.99 0.39
Loader driver 2.97 0.71 5 0.59 0.89 0.29
Office 2.20 nd* 5
not determined
jor source of dust exposure for the truck drivers 404-408.

was the unloading of the truck at the dump. Page, S., 2006. Crystalline Silica Analysis: A Comparison
of Calibration Materials and Recent Coal Mine Dust
Respirable dust from the loading operation had Size Distributions. ASTM Symposium on Silica:
a lower contribution of dust exposure. The esti- Sampling and Analysis, 22-23 April, Salt Lake City,
mated severity was less than 1 but the UCL was UT.
greater than 0.5 and slightly lower than 1.0. Watts, W.S. and D.R. Parker, 1995. Quartz Exposure
These values indicated that the monitoring of Trends in Metal and Nonmetal Mining. Appl. Occup.
the dust must be continued and possible action Environ. Hyg. 10(12), 1009-1018.
Occupational Health and Safety Agency, (OSHA). 2009.
for the further reduction of the exposure must be www.osha.gov/SLTC/etools/silica/index.html.
examined. Greek Regulation for Mining and Quarrying (GRMQ),
1985. www.ypan.gr/fysikoi_poroi/emne_vasiki_no-
mothesia.htm
4. CONCLUSIONS
In this study the crystalline silica content of to-
tal and respirable dust from a limestone aggre-
gates quarry was determined. The crystalline sil-
ica was exclusive found in the form of quartz,
while its concentration in respirable dust was
measured between 0.65 and 0.85%. The expo-
sure levels of trucks’ drivers were the highest
while the major source of dust generation was
the unloading of the truck. Although the expo-
sure levels were under the PELS, the UCL value
for severity indicated that the monitoring of the
dust must be continued and possible action for
the further reduction of the exposure must be
examined.
REFERENCES
American Thoracic Society, 1997. Adverse Effects of
Crystalline Silica Exposure. Am J Resp Crit Care Med
155, 761-765.
International Agency for Research on Cancer (IARC),
1997. Monograph on the Evaluation of Carcinogenic
Risks to Humans: Silica, Some Silicates, Coal Dust
and Para-Aramid Fibrils. Volume 68, IARC, Lyon,
France.
MDHS14/316, 2000. Methods for the Determination of
Hazardous Substances: General methods for sampling
and gravimetric analysis of respirable and inhalable
dust. HSE Books, Sudbury.
MDHS101, 2005. Methods for the Determination of Haz-
ardous Substances: Crystalline silica in respirable air-
borne dusts-Direct-on-filter analyses by infrared spec-
troscopy and X-ray diffraction. HSE Books, Sudbury.
United States Bureau of Mines (USBM), 1992. Crystal-
line silica primer. USBM Special Publications, USA.
Virji, A. D. Bello, S. Woskie, M. Liu and A. Kalil, 2002.
Analysis of Quartz by FT-IR in Air Samples of Con-
struction Dust. Applied Occupational and Environ-
mental Hygiene, 17(3), 165-175.
Bahrami, A., F. Golbabai, H. Mahjub, F. Qorbani, M.
Aliabadi and M. Barqi, 2008. Determination of Expo-
sure to Respirable Quartz in the Stone Crushing Units
at Azendarian-West of Iran. Industrial Health, 46,
Prediction of annual CO2 emissions of lignite mined from the South Field
Mine, Greece
M. Galetakis and D. Vamvuka

C. Roumpos and F. Pavloudakis

Power Public Corporation, Greece
ABSTRACT 14 Mt per year (Kavouridis, 2008).

For the exploitation of the lignite deposits the
In this paper a methodology for the long-term
continuous surface method is mainly used. The
prediction of the CO2 emission factor, based on
applied method employs high capacity bucket
2D models of selected lignite quality parameters
wheel excavators, conveyors belts and stackers.
and on strategic mine planning, is presented.
Also conventional mining equipment including
The calculation of the CO2 emission factor was
trucks, front-end loaders, electric rope and hy-
based on the calorific value of the lignite and
draulic shovels and dozers, is utilized for spe-
the efficiency and oxidation factor of the lignite-
cific mining operations. These large scale min-
fired power unit. The developed methodology
ing operations have been selected in order to
was applied for the long-term prediction of CO2
achieve the required production and cost targets.
emission factor in South Field Mine in the Ptol-
The future exploitation of lignite deposits in
emais area (Greece).
Greece depends on the ability of the coal indus-
try to adapt the environmental requirements of
1. INTRODUCTION Kyoto Protocol, regarding the control of green-
house gas emissions. For lignite to remain as a
Lignite is the main domestic energy resource of major component of the Greek energy mix, new
Greece since the mineable lignite deposits technological solutions and policies, which will
amount to 4200 Mt. Lignite exploitation in enable lignite to contribute to the short-term so-
Greece has drastically increased from 900 thou- lutions for climate change, are needed. These
sand tons per year to 65-70 Mt per year, during solutions are focused in increasing power plant
last 50 years. This production rates Greece as efficiency by improving the combustion tech-
the second lignite producer in the EU and the nology and/or by upgrading and controlling the
fifth largest in the world (Kavouridis, 2008). lignite quality in order to reduce the CO2 emis-
The main lignite producer in Greece is the sion factor (t CO2/MWh).
Power Public Corporation SA (PPC). Approxi- In this study a methodology for the predic-
mately 96 % of the mined lignite used to supply tion of CO2 emission factor, based on the qual-
the lignite-fired power plants operated by PPC. ity of the mined lignite and the power plant op-
The contribution of lignite to total electricity erational parameters, is presented. The devel-
production in Greece is approximately 58%. At oped methodology was used for the prediction
present PPC operates five open-cast lignite of CO2 emission of lignite mined from South
mines. The four mines (Main Field, Kardia Field mine and fed to Agios Dimitrios Power
Field, South Field and Amynteon - Florina plant for electricity generation.
Fields) constitute the Lignite Center of Western The layout of the paper is as follows. In the
Macedonia (Northern Greece) with an annual second section of this paper the South Field
production of about 55 Mt, while the fifth mine mine is briefly described and the problem of
in Peloponese region, under the name of Lignite quality variation of the excavated lignite is dis-
Centre of Megalopolis, produces approximately cussed. The methodology for the prediction of
CO2 emissions and its application in South Field of the lignite layers are mainly related to deposi-
mine is given in the third part. Finally, the dis- tional parameters (sediment supply, paleocli-
cussion of the results and the conclusions are mate and tectonics). Run-of-mine quality vari-
given in fourth part. ability is further affected by dilution of lignite
with thin waste layers, due to unavoidable co-
excavation during selective mining operations
2. SOUTH FIELD MINE - LIGNITE QUAL- (Galetakis and Kavouridis, 1998; 1999).
ITY AND CO2 EMISSIONS Feeding the power plants with lignite which
The South Field Lignite Mine is the largest of does not meet the specifications results in de-
the four operational surface mines at the Lignite crease of efficiency, loss of energy, high gase-
Center of Western Macedonia, covering an area ous emissions (mainly CO2 and NOx), and con-
of 24 km2 (Fig. 1). The original mineable re- sequently increase of the cost of power genera-
serves in the area were estimated at 825 Mt of tion. Operation of units at maximum efficiency
lignite, while the remaining at 450 Mt. Mining mitigates the gaseous emissions and improves
operations commenced in August 1979. Cur- the environmental performance (Galetakis and
rently the mine operates on ten benches, using Kavouridis, 1998; 1999). The introduction of
mainly the continuous mining method, which CO2 emission allowances has increased the cost
employs bucket wheel excavators, conveyors of fossil-fuel power plants. Lignite produced
and stackers. Also conventional mining equip- electricity is affected much more than gas pro-
ment including big trucks, front end loaders, duced electricity, because of the higher (ap-
shovels and dozers are utilized for mining the proximately double) CO2 emission per unit of
hard rock formations which are encountered in output. The estimation of CO2 emissions is
the overburden strata. The annual lignite pro- therefore crucial for the evaluation of the addi-
duction at the South Field mine (2008) is tional cost that affects significantly the competi-
21.3 Mt lignite, while overburden total annual tiveness of lignite. The development of a meth-
excavations are 97.3 Mm3 (bank material). The odology for the estimation of annually CO2
majority of the lignite mined at the South Field emissions which can be incorporated into long-
Mine is fed to the nearby power plant of Agios term planning and production scheduling of the
Dimitrios, with a nominal output of 1585 MW. lignite mining, is therefore of paramount impor-
Lignite mined from South Field lignite de- tance.
posit exhibits significant variation in quality.
This variation is related to the structure of the
3. METHODOLOGY FOR THE PREDICTION
deposit (multiple-seam) and the applied mining
OF ANNUAL CO2 EMISSIONS
methods. The variation in thickness and quality
Mine design and long-term sequencing is an
important part of a mining operation. It provides
the technical plan to be followed from mine de-
velopment to mine closure, having a profound
SECTOR 6
ΚLΙΤΟS
effect on the economic value of the mine (Dimi-
trakopoulos et al., 2007). The uncertain nature
of mineral evaluation data requires the frequent
revision of long-term mine planning to incorpo-
SECTOR
11
rate new data, such as the CO2 emission allow-
ances. The methodology for the prediction of
SECTOR 7
SECTOR
10
annual CO2 emissions is based on a 2D model
SECTOR
8
SECTOR
9 of selected quality attributes of the mineable
lignite, the operating characteristics of the
power plants, the mining sequence and the long-
term production scheduling. The main steps of
the proposed methodology are:
Figure 1: Map of the South Field Mine indicating the min-
ing sequence. - Division of deposit into sectors according to
the strategic mine planning. (Galetakis and Vamvuka, 2009).

- Selection of boreholes within each sector and
3153
evaluation of the mineable lignite quality. Qs = ⋅ CL ⋅ OF (1)
- Estimation of the CO2 emission factor for LHV ⋅ n
each borehole. where:
- Development of a 2D model of the CO2 LHV: low heating value of mineable lignite
emission factor and of the heating value of (kcal/kg),
lignite. CL: total carbon content of lignite in as re-
- Prediction of annual CO2 specific emission ceived basis, %,
by taking into account the mining sequence, OF: oxidation factor, indicating the per-
the planned annual production rate and the centage of carbon converted to CO2, %,
specific emission factor. n: power plant efficiency, %.
3.1 Division of deposit into sectors-evaluation While the lower heating value, LHV, of
of mineable lignite quality mineable lignite was available from the bore-
hole data, the total carbon content of lignite was
The South Field lignite deposit was explored by determined in a limited number of borehole as-
480 drillholes, covering an area of 25 km2, with says. It is known however that CL is strongly re-
an average spacing of 250 m. It has been di- lated to A, W and LHV and consequently CL can
vided into 11 sectors as shown in Figure 1. be calculated as linear functions of these quality
The quality of mineable lignite layers was parameters (Galetakis and Vamvuka, 2009). The
determined by using certain evaluation criteria estimated linear equation by using the linear re-
concerning their geometric and quality charac- gression method is:
teristics. Thin successive lignite and waste lay-
ers are grouped using specific criteria regarding CL= 23.7+0.006·LHV -0.14·W -0.13·A (2)
thickness and ash content, to define the mine- The power plant efficiency, n, as well as the
able lignite (Table 1). Dilution and mining loss oxidation factor, OF, was estimated from the ex-
occurring during mining were also taken into isting operational data. It was observed from the
consideration (Galetakis and Hristopoulos, operational data of Agios Dimitrios power plant,
2004). during 1999-2007, that n and OF does not vary
The quality parameters of the mineable lig- significantly and thus their mean values were
nite, in each borehole, were estimated as the used in Eq. 1 ( n =37.2 % and OF =98%).
weighted average of the individual mineable
layers. These quality parameters were the ash 3.3 Development of a 2D model for Qs
(A), the lower heating value (LHV) and the Geostatistical kriging methods were used for the
moisture (W). The model was based on the ex- development of the 2D model of Qs. The devel-
isting drillholes data and consists of 3799 oped model consists of 3790 blocks 100x100 m2
blocks, 100x100 m2 each. each and it was based on data from 473 drill-
3.2 Estimation of the specific emission factor holes (drill holes spacing is approximately
250m). Statistical analysis indicated that Qs val-
The specific emission factor Qs (tons of pro- ues follow the normal probability distribution,
duced CO2 per generated MWh), was estimated as shown in the normal probability plot illus-
for each block of the above model, as follows trated in Figure 2.
The calculated experimental variograms γ(h)
Table 1: Evaluation parameters.
of Qs, in four different directions (0o, 45o, 90o
Minimum thickness of lignite layer for se- 0.50
and 135o), did not indicate any substantial dif-
lective excavation (m)
Minimum thickness of intermediate waste 0.30
ferences. In Figure 3 the experimental omni-
layer for selective excavation (m) directional variogram of Qs, is plotted. Several
Maximum allowable ash (in dry basis) of 55 variogram models were fitted to this experimen-
the mineable lignite % tal variogram, including combinations of nugget
Dilution factor (w/w) % 7 terms with spherical and exponential compo-
Mining loss (w/w) % 5 nents, as well as the non-stationary linear
99.9 Qs (t/MWh)
99 21000
Cumulative frequency %
1.35
95 20000 1.3
80 19000
1.25
50 1.2
18000
20 1.15
17000
5 1.1
1 16000
1.05
0.1 15000 1
1 1.1 1.2 1.3 1.4 1.5 -15000 -14000 -13000 -12000 -11000 -10000 -9000 -8000 -7000
Qs (t/MWh) Figure 4: Contour map for predicted Qs. values on

100x100 grid.
Figure 2: Normal probability plot of Qs.
LHV kcal/kg
0.0022
21000
1700
0.002
20000 1600
0.0018
0.0016 19000
1500
0.0014 1400
18000
0.0012
γ(h)
1300
17000
0.001 1200
0.0008 Exponential model 16000

1100
γ(h)=c(1-exp(-|h|/a) c=0.00145 a=150
0.0006
15000 1000
Linear model
0.0004
γ(h)=bh b=0.0000002 -15000 -14000 -13000 -12000 -11000 -10000 -9000 -8000 -7000
0.0002 Figure 5: Contour map for predicted LHV values on

0
100x100 grid.
0 500 1000 1500 2000 2500 3000 3500 4000
Lag distance h (m) ity is closely related to tectonics and deposi-
Figure 3: Experimental omni-directional variogram of Qs tional parameters during the formation of the
with the fitted theoretical model. deposit (Galetakis and Hristopoulos, 2004).
model. Finally the experimental variogram was 3.4 Prediction of annual CO2 emissions
modelled by a combined exponential and linear
The prediction of annual CO2 emissions in year
model as shown in Figure 3.
t, Q(t), was obtained by comparing the esti-
The universal kriging method was used for
mated annual Qs(t) and the planned production
the estimation of Qs at each block of the devel-
rate of lignite L(t), to the corresponding values
oped 2D model. A 2D model for the LHV was
in 2005. During 2005, 22.09 Mt of lignite were
also constructed in a similar way. The contour
mined from South Field mine and fed to the
maps, shown in Figures 4 and 5, illustrate the
Agios Dimitrios power plant. The generated
spatial variation of Qs and LHV respectively, in
electricity was 11634375 MWh, the emitted
the South Field mine. As clearly shown in Fig-
CO2 was 14.860 Mt, while the resulting specific
ure 5, the middle part of the lignite deposit, con-
emission factor was 1.277 t CO2/MWh. Since
sisting of sectors 4,7,8,9 and 10, indicates high
annual CO2 emissions in 2005, Q(2005), is
calorific values (LHV). Thus the resulting Qs
known, Q(t) is given by:
values (Eq. 1) are consequently low in these
sectors (Fig. 4). On the contrary, in the sectors Q s (t) L(t)
3, 5, 6 and 11, located in the borders of deposit, Q(t) = ⋅ ⋅ Q ( 2005 ) (3)
Q s ( 2005 ) L ( 2005 )
the calorific value of the lignite is significant
lower and the corresponding Qs values higher. Qs(t) was calculated by averaging the values of
The observed spatial variability of lignite qual- Qs of the lignite blocks excavated in year t. The
Table 2: Long-term production scheduling and predicted annual CO2 emissions.

CO2 emission Qs (t ) L( t ) Annual CO2
Sectors under Lignite produc-
Year, t factor emissions in
exploitation tion in kt, L Qs (2005) L(2005)
(t/MWh), Qs(t) kt, Q
2009 22287 1.250 0.979 1.009 14676
Sector 4, 5 2010 19926 1.250 0.979 0.902 13121
2011 19899 1.250 0.979 0.901 13104
Sector 4, 5 2012 17460 1.263 0.989 0.790 11620
and 6 2013 17499 1.278 1.006 0.792 11841
2014 18083 1.301 1.019 0.819 12394
Sector 6
2015 13483 1.301 1.019 0.610 9241
2016 13363 1.306 1.018 0.605 9152
2017 11767 1.297 1.018 0.533 8057
Sector 6, 7
2018 12665 1.295 1.013 0.573 8632
2019 11978 1.249 0.978 0.542 7879
2020 13290 1.237 0.969 0.602 8660
Sector 7
2021 15997 1.237 0.969 0.724 10424
2022 18356 1.222 0.957 0.831 11817
Sector 8 2023 17400 1.222 0.957 0.788 11201
2024 15991 1.222 0.957 0.724 10294
2025 17079 1.210 0.948 0.773 10886
2026 18198 1.210 0.948 0.824 11600
Sector 9
2027 18375 1.210 0.948 0.832 11713
2028 21075 1.210 0.948 0.954 13434
2029 15083 1.205 0.944 0.683 9574
2030 12962 1.205 0.944 0.587 8228
Sector 10
2031 13024 1.205 0.944 0.590 8268
2032 13024 1.205 0.944 0.590 8268
2033 11815 1.288 1.009 0.535 8017
2034 8914 1.288 1.009 0.404 6048
Sector 11
2035 8914 1.288 1.009 0.404 6048
2036 7515 1.288 1.009 0.340 5099
determination of the annually excavated blocks ods with increasing trend. More specifically
was based on existing long-term exploitation from 2019 to 2028 the annual emissions raises
planning of the South Field mine, shown in Fig- from 7.8 Mt to 13.4 Mt, which is the highest ob-
ure 1 and Table 2. tained value. This increase is related to the
The predicted annual CO2 emissions, as well planned production rate during this period.
as the variation of Qs(t)/Qs(2005) and The specific CO2 emissions factor, Qs, de-
L(t)/L(2005), are given in Table 2. The time- pends primarily on the quality of coal consumed
sequence plot of these values is illustrated in (the carbon content and the calorific value). Qs
Figure 6. indicates a relative low fluctuation from 2009 to
2036. From 2019 to 2028, when the planned
production rate is high, the estimated Qs values
4. DISCUSSION OF RESULTS - CONCLU- are low resulting in lower CO2 emissions. For
SIONS instance, while in 2028 the production rate is
The predicted annual CO2 emissions between 4.5% lower than the corresponding of 2005, the
2009 and 2036, as well as the specific emission annual CO2 emissions are reduced by 9.6%.
factor, indicated that: It is evident from the above, that the devel-
Although the annual CO2 emissions show opment of a reliable methodology for the pre-
clearly a decreasing trend, resulting mainly from diction of CO2 emissions during the long-term
the decline in lignite production, there are peri- mine planning is of great importance. It can be
1.200 CO2 emissions in 2005 16000
1.100 14000
1.000
12000
Annual CO 2 emissions, kt
L(t)/L(2005) or Q s (t)/Q s (2005)
0.900
10000
0.800
8000
0.700
6000
0.600
4000
0.500
Productivity ratio
Specific emission factor ratio 2000
0.400
Annual CO2 emissions, kt
0.300 0
2009 2011 2013 2015 2017 2019 2021 2023 2025 2027 2029 2031 2033 2035 2037
Year
Figure 6: Graph indicating the evolution of the annual CO2 emission from South Field mine, as well as, the variation of
specific emission factor and production rate, compared to corresponding values of 2005.
used to estimate the additional cost due to CO2

allowances. By examining different mining se-
quences the optimal exploitation strategy can be
determined.
REFERENCES
Dimitrakopoulos, R., L. Martinez and S. Ramazan, 2007.
A maximum upside/minimum downside approach to
the traditional optimization of open pit mine design.
Mining Science, 43, 73-82.
Galetakis, M. and D. Vamvuka, 2009. Lignite quality un-
certainty estimation for the assessment of CO2 emis-
sions. Energy & Fuels, 23, 2103-2110.
Galetakis, M. and D.T. Hristopoulos, 2004. Prediction of
Long-Term Quality Fluctuations in the South Field
Lignite Mine of West Macedonia, Proceedings
AMIREG 2004, 133-138 (eds. Z. Agioutantis and K.
Komnitsas), Heliotopos.
Galetakis, M. and K. Kavouridis, 1999. Homogenization
of quality of lignite mined from the Ptolemaes -
Amynteon lignite basin, Scientific Journal of the TCG
(V) 1-2, 41-52.
Galetakis, M. and K. Kavouridis, 1998. Quality control of
lignite produced by continuous surface mining, using
statistical methods, Scientific Journal of the TCG (V)
1-2, 49-61.
Kavouridis, K., 2008. Lignite industry in Greece within a
world context: Mining, energy supply and environ-
ment. Energy Policy 36, 1257-1272.
Characterization of quarry fines for recycling uses in construction
M. Galetakis and K. Leventakis

ABSTRACT from limestone deposits that are widespread all

over the country, via quarrying operations
Quarrying and ornamental stone processing, be-
(Galetakis and Raka, 2004a). More than 300
ing a massive contributor to any construction
small to medium size quarries produce about
project regardless of the technology level in-
100 million tones of crushed limestone aggre-
volved, has been proven to attribute to a great
gates annually (Menegaki, 2008).
extent to the accumulation and harmful disper-
During the extraction, hauling and crushing
sion in water, air and soil of solid waste such as
processes for the production of aggregates, large
fine stone dust. In this study, the dominant
quantities of limestone fines, hereon referred to
mechanisms of dust production during extrac-
as quarry dust, are generated. Particularly at the
tion, crushing and processing have been ad-
stage of crushing and screening, fine airborne
dressed. Consecutively, the technical, legal, and
material such as silt and crusher dust can be lib-
economical issues arising from the lack of
erated and emitted into the surrounding atmos-
proper management of such a fine stone by-
phere. Silt is a material between 2 and 60 μm
product have been addressed. A characterization
reduced to this size by natural processes and it is
methodology, based on selected quality charac-
found in aggregates won from natural deposits.
teristics of quarry dust, has been developed and
Crusher dust is a fine material formed during
limestone dust, collected from a concrete and
the process of rock comminution. In concrete
asphalt mixture producer in Crete, has been
fine material is present in aggregates in the form
characterized as a case study.
of surface coating which interferes with the
bond between aggregate and cement paste.
1. INTRODUCTION Greek National Concrete Regulation
(Karavasili, 2003) limits the content of fine ma-
Recent years have seen an unprecedented terial in coarse crushed aggregates for structural
growth of interest on the environmental per- concrete, to 1% and in crushed stone sand to
formance of all industry sectors, as the global 15%. Limestone dust compared with coarser
economy is forced to evaluate its environmental aggregates used in structural concrete, exhibits
impact, under the pressure of evident recent higher specific surface and higher water absorp-
changes in climate, natural habitat, fauna and tion resulting to higher water demand. High wa-
flora. Construction, being a major component of ter-to-cement ratio, is in turn responsible for
human activities, regardless of the state of evo- higher shrinkage and creep values for the hard-
lution of state economy, has been accused from ened concrete. Quarry dust in Greece, usually
early years on for its heavy impact on natural removed from the aggregates by bag filters, suc-
environment. tion, or air cyclones, is estimated to 10% of the
Aggregate materials, used extensively for all total aggregate production (Galetakis and Raka,
construction applications, are essential for the 2004b). This quarry dust amounts to ~10 Mt an-
development of any modern economy. In nually (Menegaki, 2008) and poses severe dis-
Greece, the majority of aggregates are extracted posal and environmental problems. Since devel-
opment needs to take place in a way, which pro- Centre for Aggregate Research (ICAR) sympo-
tects the environment, in order to achieve sus- siums (1994 to 2008), devising ways of charac-
tainability, waste or by-product material re- terizing, assessing and testing aggregate proper-
quires management for proper disposal or fur- ties. The concept was to predict the properties of
ther utilization (Hill et al., 2001). concrete or concrete-based building elements
The Mineral Industry Research Organization made from it. Much of this research concerns
(MIRO), of UK, has issued a number of reports the fine fraction (<75µm) of aggregates. Efforts
(Manning, 2004; Petavratzi and Wilson, 2008) have been made by ICAR to the direction of de-
evaluating the following uses: quarry restora- veloping such a classification procedure, for the
tion, backfill/infill of voids, general fill (i.e. em- use of aggregate fines in concrete (ICAR 101-
bankments), road pavement construction (i.e. 2F).
sub-base, capping), remediation works, artificial In order to achieve sustainability, out-dated
soils, compost, filler applications (for example, or superseded specifications must be changed,
paper, paint, plastics, rubber), portland cement when appropriate, to allow the use of so-called
kiln meal, innovative products (for example, secondary material. Inappropriate specifications
green roofs, eco-slates, cob building ), con- and/or standards are a significant problem in the
trolled low strength materials - self compacting USA and there are numerous ICAR symposium
concrete, manufactured concrete, heavy ceram- papers related to this issue. For example, stan-
ics, manufactured aggregates, hydraulically dards set for the use of natural sands in con-
bound mixtures, asphalt. As far as construction crete, are still active in many states where
is concerned a lot of uses for quarry dust have crushed sand is used. In natural sands, the fines
been investigated such as: construction of vari- present tend to be clay minerals which reduce
ous types of masonry bricks and artificial concrete strength by inhibiting the bond be-
stones, mineral admixture in Self Compacted tween aggregate and cementitious mortar (Du-
Concrete (SCC), and high performance micro- katz, 1995). The restriction on the percentage of
concrete (Fowler, 1995). fine material present in concrete has originally
Quarrying enterprises in Greece belong in the risen from the later observation. When crushed
construction sector of the economy. Conse- sand is used, fines are typically non clay. To
quently construction uses for the fine quarry by- clarify this, chemical analysis can be used to de-
product would be much easier to encompass termine the clay mineral content (for example
into their production chain. methylene blue, titration test).
In this study a characterization methodology, Dumitru et al. (2001) in a study presented at
based on selected quality characteristics of the 2001 ICAR symposium, propose alternative
quarry dust, is presented. Its future utilization or complimentary methods for assessing fine
would point out the most suitable use for quarry aggregates. They concluded that mineralogical
dust of different qualities. In the second section analysis (for example X-ray diffraction analy-
of this paper the quality requirements as well as sis), should be used to identify and quantify
the characterization methods of quarry dust for secondary minerals harmful in some applica-
uses in construction industry is discussed. The tions (in particular clay minerals).
quality characteristics of the limestone dust col- Prior to 1997, quarries certified by the US
lected from a quarry in Crete is also presented. Department of Transportation were required to
In the third section results are discussed and fi- maintain grading target values of 0-7% passing
nally conclusions are given in the fourth part. mesh No 100 (150 μm) sieve and 0-2% passing
mesh No200 (75 μm) sieve. Users were forced
to add fines in order to improve workability and
2. CHARACTERIZATION OF QUARRY finishability. This specification for very low
FINES percentages of -75 mm fines, common in as-
2.1 Characterization and quality requirements phalt and concrete mix aggregates, in conjunc-
of quarry fines for construction uses tion with increased demand for crushed concrete
sand, have increased the rate of waste fines pro-
Over the years, there has been an increase in reduction (Smith and Slaughter, 1996).
search efforts, presented at the International It is evident that size distribution, as well as
chemical and mineralogical composition is the the operation of the asphalt-mix plant. Fine
most important characteristics of bulk limestone aggregate particles are removed by the ex-
quarry dust. Since though, they exhibit wide haust fumes of the fuel oil-fired kiln and are
variations, even within a single day’s produc- then collected by air cyclones and bug filters.
tion, they should be monitored in order to indi- Consequently the collected dust contains a
cate the best suitable use for recycling this by- considerable amount of unburned particles of
product. Evaluation of the resulting composi- the heavy-oil. The resulting color of the col-
tional and physical data can be used as input in lected dust is directly related to the amount
the development of specifications and test pro- of the contained unburned fuel. The material
cedures used to evaluate and approve fines for is then wetted (for the elimination of air dust
the production of recycled products. Further- emissions) and disposed in tanks or piles lo-
more, this data should aid in identifying the cated inside the plant yard.
most appropriate economic use for fines, based Samples of white, dry material were col-
on spatial constraints associated with lithology lected by using a slotted sampling probe suitable
variations (McClellan et al., 2002). for fine bulk materials. This sampler assures the
collection of representative samples on all pile
3. CHARACTERIZATION AND EVALUA- locations.
TION OF AGGREGATE FINES COLLECTED Slurry or semi-dry dust samples (disposed in
FROM A CONCRETE / ASPHALT MIXING the setting tanks) were collected uniformly from
PLANT the tanks’ surfaces due to the small and roughly
constant depth of the tanks, using a shovel. Col-
3.1 Quarry lithology and sampling lected samples of both dust varieties were ho-
For the purpose of this project, limestone dust mogenized and reduced in quantity by using a
samples where collected from a concrete ready- riffle sample divider to obtain suitable sub-
mix and asphalt mix production plant, situated samples required for the laboratory analysis.
in the Heraklion prefecture on the Island of 3.2 Laboratory analysis and results evaluation
Crete, Southern Greece. The company uses ex-
clusively aggregates provided from the near-by Laboratory tests carried out on selected samples
quarry, producing over 800.000 t of aggregates of the quarry dust, included grain size and mor-
annually, where rock samples were also col- phology analysis, determination of mineralogi-
lected. cal and chemical composition, specific surface
Aggregates are extracted from geological for- and bulk density.
mations consisting mainly of limestones me- Grain analysis of limestone dust was carried
dium bedded, light to dark gray. Two visually out by using a laser particle-size analyzer (Mal-
distinguishable limestone dust varieties gener- vern Instruments, Mastersize-S). Results, shown
ated as by-products, lead to different strategies in Figure 1, indicated that the white limestone
of sampling. This is due to the different mecha- dust collected from concrete aggregate is
nisms of limestone dust accumulation during the coarser to that collected from the asphalt plant
various production stages, attributing to differ- (brown and dark grey dust).
ent physical and chemical characteristics of the Grain morphology was studied by means of
final bulk fine by-product. These different varie- optical and scanning electron microscopy
ties are: (SEM). Size fractions of +75 μm, -75+56 μm
and -56 μm were taken by wet sieving from se-
- Dry White limestone dust (sample designa- lected samples of the quarry dust. The +75 μm
tion WD). This material is collected when and -75+56 μm fractions were optically studied
dedusting of coarse concrete aggregates is by using a stereomicroscope while the -56 μm
performed by mean of air separation. Dust is fraction by SEM. Typical microphotographs of
removed from coarse aggregates by high the white and the brown dust, taken from the
pressure air vents and then collected by cy- stereomicroscope, are shown in Figure 2. In
clones and bag filters. Figure 3 SEM images of the -56 μm fraction of
- Brown to dark grey limestone dust (samples brown dust are illustrated.
1-30). This type of dust is generated during
100 White dust Brown dust Dark grey dust
90
80
+75 μm
Cumulative % (passed)
70 White dry dust

1500 μm
60
50 Brown and dark grey dust
-75 +56 μm
40
30 500 μ m
20
Figure 2: Stereomicroscopic images indicating grains
10 morphology, for the different types and gradations of the
0 quarry dust.
0 40 80 120 160 200
Particle size (μm) is shown in Table 1. Dust consists mostly of
calcite, while small amounts of dolomite and
Figure 1: Particle size analysis of selected samples of
limestone dust. traces of quartz where also detected.
In a limited number of brown to dark grey
In Figure 2, it can be observed that the coarse dust samples, small amounts of NaCl were also
granulometric fraction (+75 μm) of both types detected. The presence of NaCl in the wetted
of dust is non-homogeneous. Relatively large dust is associated with the occasionally sub-
and flaky particles, that were likely drifted by saline water (especially during summer months)
the air vent, are also present. On the contrary, in used in the Heraklion industrial region. This ob-
fractions -75+56 μm, and -56 μm (Fig. 3) grains servation might lead to concern when examin-
appear homogeneous. Dust consists almost ex- ing the potential uses for quarry dust, as NaCl
clusively of angular calcite particles. has been identified as deleterious in many con-
Mineralogical analysis of dust, as well as bed struction applications.
rock samples (designated BD), was carried out The acid insoluble fraction of each sample of
by using an X-Ray diffractometer (Siemens by-product fines was determined after digestion
D500). Results indicated that quarry dust con- in 2N HCl and then analyzed by x-ray diffrac-
sists mainly of calcite (~98 wt %) with traces of tion (XRD) to determine its mineralogical com-
dolomite and less frequently of quartz (Fig. 4). position. The resulting data (Table 2) illustrate
The chemical composition of selected samples that the white limestone dust possesses much
(WD1, WD2, 28, 11, 13, 14, 19 and 24) of less acid insoluble content than the brown to
quarry dust as well as of bed rock samples dark grey limestone dust. XRD analysis of the
(BRA and BRB) collected from the quarry, was acid insoluble residues shows them to compose
determined by X-ray fluorescence spectrometry mainly of quartz, clay minerals and iron oxides.
(Siemens SRS 303 sequential spectrometer) and Most common is, quartz, which was detected in
Table 1: Chemical composition of quarry dust samples.
CaO Fe2O3 SiO2 Al2O3 MgO K2O Na2O CO2 Calcite Dolomite
Sample code
% % % % % % % % % %
WD1 51.46 0.06 0.52 0.15 3.36 <0.1 <0.1 44.55 83.60 14.85
WD2 30.40 0.07 0.63 0.14 21.90 <0.1 <0.1 47.70 <0.5 99.13
8 55.11 0.21 0.53 0.12 0.25 <0.1 <0.1 43.90 98.41 <0.5
11 55.17 0.27 0.55 0.11 0.23 <0.1 <0.1 43.85 98.51 <0.5
13 53.46 0.21 0.48 0.16 1.68 <0.1 <0.1 43.86 91.17 7.8
14 53.76 0.25 0.63 0.21 1.48 <0.1 <0.1 43.75 92.07 6.7
19 53.06 0.165 0.49 0.19 1.48 <0.1 0.6 43.26 90.88 7.02
24 53.46 0.179 0.55 0.17 1.68 <0.1 <0.1 43.85 91.17 7.80
BRA 55.44 0.03 0.26 0.10 0.24 <0.1 <0.1 43.89 99.00 <0.5
BRΒ 55.34 0.03 0.36 0.10 0.23 <0.1 <0.1 43.90 98.83 <0.5
Blaine values are high and exhibit moderate to

high variability (from 2988 to 5651 cm2/g).
Blaine value, also referred to as fineness, is re-
lated to the gradation of the dust. Other meas-
ured parameters were apparent density, pH and
organic matter content. pH was measured in
10 μm dust suspensions in distilled water. The calcu-
lated values were typical for limestone filler.
Figure 3: Scanning electron microscopy images showing The organic matter was determined by thermo-
grain size and morphology of the -56 μm granulometric
fraction of the dark grey dust.
gravimetric analysis (TGA-6 Perkins Elmer sys-
tem). The mean values of the above parameters
all studied samples. Clay minerals’ presence is are listed in Table 2.
also common, as a result of clay concentration
enrichment in the fine fraction. The quantifica-
4. CONCLUSIONS SUGGESTIONS
tion of clay minerals’ content is essential when
examining the potential applications for quarry The accumulation of fine limestone dust in
dust, as many of them possess shrink/swell quarries and concrete/asphalt-mix plants is a
characteristics, which have been recognized as chronic problem that threats the industry with
harmful in concrete structures. great legal and economic implications, in the
The specific surface of limestone dust was present environmentally sensitive global econ-
measured by the Blaine method (air permeabil- omy. The huge quantities of this by-product that
ity) and the calculated mean values for both have been stockpiled and are also currently pro-
types of dust are given in Table 2. Measured duced, call for a solution that could consume
Cc
7000
♦ Cc=Calcite
• Do=Dolomite
6000
5000
Lin (Counts)
4000
3000
2000
1000
Do
3 10 20 30 40 50 60 70
2-Theta - Scale
Figure 4: Typical X-ray diffractometer diagram of the quarry dust, indicating that quarry dust is consisted mainly of cal-
cite and of a small amount of dolomite.
Table 2: Average values of measured parameters for both types of dust

Number of Specific surface Apparent Acid insoluble Organic matter
Type of dust pH
samples cm2/g (Blaine) density g/cm3 residue % %
White 2 2988 1.26 0.75 9.34 0.02
Brown/dark
30 5645 1.12 2.27 9.04 1.27
grey
great quantities and produce even marginal ACKNOWLEDGMENTS

profits for the companies involved.
The authors wish to gratefully acknowledge the
The characterization and evaluation of quarry
Research Committee of the Technical Univer-
dust samples collected from an asphalt/concrete
sity of Crete (ELKE) for the financial support of
plant in the area of Heraklion indicated that:
this research project.
- Limestone dust produced is not of uniform
quality, but exhibits large fluctuation, espe-
cially in gradation and composition. This REFERENCES
fluctuation relates to lithology of the ex- Dukatz, E.L., 1995. Effective use of aggregate fines. Pro-
tracted material as well as to dust-collection ceedings of 3rd ICAR annual symposium,
processes. www.icar.utexas.edu.
Dumitru, I., V. Paraschiv, C. Glass and G. Mandarakas,
- Dust was classified into two categories: 2001. Alternative or Complementary Methods for As-
white dust consisting of fines collected dur- sessment of Fine Aggregates. Proceedings of 9th
ing the dedusting of coarse concrete-grade ICAR annual symposium, www.icar.utexas.edu.
aggregates, and brown to grey dust, consist- Fowler, J., 1995. Construction uses of quarry fines. Pro-
ing of fines collected form the asphalt-mix ceedings of 3rd ICAR annual symposium,
plant. White dust is coarser than brown to www.icar.utexas. edu.
Galetakis, M. and S. Raka, 2004a. Utilization of lime-
grey dust. In both types the +75 μm fraction stone dust for artificial stone production: an experi-
is inhomogeneous and is characterized by the mental approach. Mineral Engineering, 17, 355-357.
presence of relatively large flaky particles. Galetakis, M. and S. Raka, 2004b. Assessment of lime-
- The most common contaminant of the col- stone quarry dust by using factorial design of experi-
lected brown to grey dust was unburned re- ments. Proceedings, 1st Int. Symposium on Advances
in Mineral Resources Management and Environmental
sidual matter from heavy oil used in the as- Geotechnology (AMIREG), Hania, Greece, pp. 685-
phalt kiln. In a limited number of dust sam- 689.
ples, small amount of NaCl were also identi- Hill, A.R., A.R. Dawson and M. Mundy, 2001. Utilization
fied. Its presence is related to the occasion- of aggregate materials in road construction and bulk
ally sub-saline water table used for spraying fill. Resources Conservation & Recycling, 32, 305-
the dust. 320.
ICAR 101-2F: Framework for Development of a Classifi-
- Since uniform quality of the quarry dust and cation Procedure for Use of Aggregate Fines in Con-
absence of harmful substances are prerequi- crete, www.icar.utexas.edu.
sites for any construction application, it is Karavasili, M., 2003. Greek Regulations of Concrete
mandatory that dust is collected and handled Technology. P-Systems, Athens, 103p. (in Greek).
in a more appropriate manner. Such suitable Manning, D., 2004. Exploitation and Use of Quarry Fines.
MIRO-Final Report, 60p.
dust collecting and handling practices should McClellan, G.H., J.L. Eades, K.B. Fountain, P. Kirk and
include: C. Rothfuss, 2002. Research and techno-economic
- Dust collection by air suction (sandbags, air evaluation: uses of limestone byproducts, Department
cyclones), blending/homogenization and of Geological Sciences, College of Liberal Arts and
storage of dry dust in silos, similar to those Sciences, University of Florida, Florida Department of
used by the cement industry. The implemen- Transportation State Contract No. BA589, WPI
0510798.
tation of this process will significantly im- Menegaki, M., 2008. Aggregate Quarries and Environ-
prove the uniformity of dust quality and will mental Protection. Latomeio, 8, pp. 28-32 (in Greek).
eliminate the use of sub-saline water for its Petavratzi, E. and S. Wilson, 2008. Report of Research
treatment. Program “Sustainable provision of aggregates, sus-
- Reduction of the unburned residual in the tainable utilization of quarry by-products”, Defra Ag-
gregates Levy Sustainability Fund, 79p.
dust collected from the asphalt kiln, by Smith, J.O. and P. Slaughter, 1996. Uses for by-product
coarse aggregate fraction (-100 + 50 mm) fines. Proceedings of 4th ICAR annual symposium,
dedusting, before its feeding to the asphalt www.icar.utexas.edu.
kiln.
Study of reformation and re-establishment of vegetation during alternative

management of quarries in the municipality of eastern Mani
J.N. Georgi
Hellenic Open University, Greece
A. Kampouris
Technological Educational Institute of Kavala, Department of Landscape Architecture, Kavala,
Greece
ABSTRACT
The present paper studies a) the restoration of 2. METHODS AND MATERIALS
the landscape in a quarry site and b) the estab-
lishment of vegetation and its incorporation into The case study area is a private area located in
the natural landscape without affecting the aes- Loukadika of Messinia, Greece, which belongs
thetics of the area. The aim of this study was the to the municipality of Eastern Mani. It is an iso-
re-establishment of the landscape which is over- lated partly rocky area of 16.186,88 m² (4 acres)
loaded as a result of quarry activities and con- where no other activities are carried out except
struction of streets for heavy equipment. The breeding of livestock.
The area is characterized by:
study focuses on the organisation and reforma-
tion of countryside for the accommodation of 1. sloping terrain
several events. 2. particular forestry value (Quercus pubescens,
Olea europaea.)
1. INTRODUCTION 2.1 Research and Study Methodology
The installation and operation of several indus- In order to shape the environment in such an
trial units in the natural landscape increases rap- area proper investigation and successful design
idly and causes changes in the landscape and the are required. Proper design improves substan-
microclimate due to the operation of equipment tially the quality of life. The science that con-
that release pollutants which contribute to mi- tributes to development, enhancement, conser-
croclimate changes. The high cost of installation vation and restoration completing thus composi-
combined with the high cost of land purchase tion and configuration of the landscape is Land-
lead the owners to establish the units in areas scape Architecture. The landscape architect as a
outside of cities or close to the natural landscape leader in science is the person who through his
where the costs are lower (Sakellakis, 2006; successful design offers qualitative use of the
Eleftheriadis, 2008) landscape in harmony with human needs and ac-
In other countries restoration of the land- tivities (Arbogast et al., 2000; Rokos, 2001;
scape in similar areas is accomplished from the Dimitriou and Georgi, 2008).
start of the construction phase having as result The research methodology is based on the
the immediate absorption of the impacts and the principles of landscape architecture and aims to
harmonization of the unit with the landscape. achieve the following:
The aim of this study was the re-establishment
of the landscape which is overloaded as a result 1. Search similar cases in Greece and abroad
of quarry activities and construction of streets 2. Analyze the area of interest
for heavy equipment. The study focuses on the 3. Design the proposal
organisation and reformation of countryside for
the accommodation of several events.
2.2 Similar Cases - Examples Restoration in Butchart Gardens in British Co-

Our search on similar cases in Greece and lumbia in Canada
abroad was not successful and gave no results. Mining areas and quarries have a long tradition
For this reason the study group investigated of after use as botanical gardens. One such case
similar areas, in terms of conditions and charac- is the mining area in the Butchart Gardens in
teristics. British Columbia in Canada.
The examples given below are mainly from In 1904 began the restoration of the aban-
quarries which have similar restoration needs as doned limestone quarry with a total area of 500
the study area examined in this work. acres as presented in Figs 2a and b.
Darlton Quarry The botanical garden opened in 1940 and re-
ceives annually over one million visitors. This is
In this case the restoration is classified in the one of many similar examples. A more recent
“Optical Hide (Rehabilitation) Option”, Fig.1. example is the approval given by the Memorial
There are cases where the extraction of raw University of Newfoundland in Canada to estab-
materials requires the development and opera- lish a botanical garden in an abandoned quarry.
tion of large fronts for long periods. In these The role of the Landscape Architect is very
cases the boundaries of the mining/quarry sites important for the rehabilitation of such areas.
are particularly noteworthy and the establish- The involvement of a landscape architect from
ment of vegetation should be combined with the the early stages of the work is very important as
modification of soil characteristics (landform the study and understanding of the actual land-
transformation).
The vegetation that is going to be used must
be compatible with the character of the area and
the existing vegetation in the quarry. The time
period for the establishment of plantations is
very important so that the visual coverage is
achieved at the right time.
In Darlton quarry several uneven slopes in
terms of height were created. The slopes were
developed in such a way as to act as a link with
the boundaries of the quarry. The elements of
the area outside the quarry were extended to the
inner part of the restoration area to enable better
interaction. The type, colour and layout of the
area play a very important role especially when
the site is visible from long distances.
Figure 2a and b: The Butchart Gardens in British Colum-

Figure 1: Darlton Quarry. bia in Canada.
be used so that they will not increase the tem-

perature in the area by absorbing solar radiation.
Another issue that has to be considered by the
Landscape Architect is the technology that ex-
ists concerning the study area. The cultural and
natural characteristics should be combined and
mixed with the scientific knowledge to achieve
a successful result.
Another issue that designers should consider
is the safety measures required for proper plan-
ning. The safety standards during construction
and proper use should be inspected by a special-
ist in order to achieve an effective use of the re-
stored area.
3. ANALYSIS OF THE STUDY AREA

The study area (Fig. 3) of approximately
16186.88 m² is located at a distance of approxi-
mately 1 km from the settlement of Loukadika,
Figure 3: Aerial Photograph of the wider area of Mani and in Messinia, Greece; since it is isolated and in-
Loukadika. clined it is ideal for the installation of such a
unit. The provincial road which connects the
scape becomes apparent from the first stage settlement with the area crosses its upper part.
where information on the design and planning The vegetation in the surrounding area is of par-
of site restoration should be obtained. The in- ticular value and forestry for the proposed con-
formation obtained needs to be summarized in figuration (Fig. 4); the aim is to link the unit
the following categories: with the larger area of the region which is visi-
- Ecology and Development ble from the road without affecting the Mediter-
- Research and Technology ranean landscape which is typical in the region.
- Culture The interventions made by the study group
- Nature aimed at preserving the naturalness of the land-
- Science and scape while seeking to make the space as much
- Art safe as possible and functional for users. Also,
through vegetation, designed to enable continu-
In the natural environment beyond the eco- ity and connection to the wider landscape, to es-
logical consciousness and awareness, the mate- tablish an air filter for microparticles released
rials as well as the ways used for constructions during processing of plant material and thus im-
are also considered important. Materials with prove the microclimate of the restored area.
light colours friendly to the environment have to Generally, the regeneration of this area is neces-
sary in order to connect the unit with the natural
landscape and transmit through this the envi-
ronmentally friendly nature that distinguishes
this facility.
The surrounding study area is characterized
by naturalness and Mediterranean character.
Due to the rocky surface, no other activi-
ties,except breeding livestock, are carried out
around the site. The characteristics of the site
make the study of the configuration particular
since the aim should be achieved through a
Figure 4: The existing situation of the area. physical view connecting the region with the
unit. to drought in arid and rocky soils, low tempera-

The proposed actions should be considered tures and strong winds. These species include:
so as to contribute substantially to the dual use Ostrya vulgaris, Quercus pubescens, Aria Quer-
of region; emphasis should be placed on the fact cus ilex, Punica granatum, Olive Olea europaea,
that the restored area should be user friendly Juniperus oxycedrus, Ceratonia siliqua.
and provide space that will accommodate vari- Finally, for aesthetic reasons and in order to
ous activities. avoid monotony various resistant to drought
During the design and in order to implement shrubs and especially lavender, levantine,
a complete study for the rehabilitation of the thyme, oregano and other brushwood were in-
area issues that were taken into consideration stalled (Castro et al., 2002; 2004).
included the present and anticipated use of the The dominant idea has similar characteristics
site as well as processing of the parameters and with the rest of the region. The simple lines and
constraints. nature form complex designs. The nature of
physical space is maintained and is not con-
3.1 The Proposal
verted into a human construction in the natural
The proposed design aims through appropriate landscape. There were used colours of wood,
manipulation of vegetation and the landscape to natural flooring materials and equipment and
unify the various operations carried out on the generally plain and simple intervention will fol-
perimeter and the area of study and to achieve low the natural character of the area.
the physical integration of this region within the The configurations including roads for vehi-
rest of the mountainous - Mediterranean land- cles, pedestrian zones, green spaces, fences and
scape (Gómez-Aparicio et al., 2004). constructions will be considered so as to con-
For the better planning and exploitation of tribute substantially to the dual use of the area
the site, due to the presence of steep slopes, the and to highlight the role of space as environ-
formation of levels throughout the area was mentally friendly and able to host various activi-
proposed, which according to the purpose that ties (Fig. 5).
will serve they will have similar dimensions. During the design of the portal, marking of
The basic idea behind the proposed formula- the entrance of the facility was considered an
tion is to maintain the initial naturalness of the aspect of great importance; therefore the con-
landscape. This can be achieved by maintaining struction of two stone columns with a spacing of
and enhancing the existing vegetation and by 5.5 m was proposed to enable vehicles to cross.
creating a large hedge of plants and trees to hide Outside the entrance a synthesis of stone and
the area and maintain the vulnerability of the tree was considered.
aesthetic landscape which is achieved largely by
3.2 Additional Constructions
the creation of steps at which trees, mainly
broadleaf (deciduous and evergreen), will be Next to the office building the construction of a
planted. This approach apart from the visual 15 m² chalet which will function as a kiosk pro-
improvement will contribute to changing the viding information on the operation of the unit
microclimate of the area, will absorb as filter as well as serving as environmental education
large percentage of pollutants produced by the pavilion is proposed.
activities carried out and will act as natural At the lowest point of the unit it is proposed
acoustic screen which will greatly reduce any that the recovered steep slopes of soil and ter-
noise generated by the operations. rain will form the basis for the construction of
In order to achieve this it is important to take an outdoor amphitheatre, using iron and wood,
into account the needs of plants required for which is expected to host several events and
their growth. For this reason, since the soil is constitute a place for gathering of visitors dur-
rocky it is proposed that 1 m³/m² of soil is ing summer and when the weather permits. The
placed at all levels. proposed shape is suggested to be triangular,
The selection of species proposed for the es- with the scene at the lowest point. It consists of
tablishment of vegetation was made on the basis 9 rows of seats in order to accommodate ap-
of climatic and soil conditions in the region. proximately 90 people (Fig. 6).
More specifically selected species are resistant
The amphitheatre is proposed to be con- ing: Aesthetical organization that reflects the
structed as an elevated small-scale construction, character of the area, Use of vegetation as a
based on a metal skeleton of which only the ver- main element, Flexibility of the space with the
tical lines will be placed in the ground. The con- possibility of carrying out social open air activi-
struction is proposed to have this shape for the ties (exhibitions, performances, rallies), Visual
following reasons: and functional complexity of each area to pro-
vide the necessary wealth. Finally, all provi-
- It doesn’t require special configuration of the
sions in the light of the Environmental Planning
excavation of soil only for the installation of
and Principles of Landscape Architecture to
piles that will form the central skeleton of the
protect and enhance the cultural and natural
structure.
landscape of the region will be considered.
- No cement or other materials will be used
since it is a lightweight construction which in
future can be easily removed if required. REFERENCES
- It is placed in an inclined area and at low
Arbogast, B.F., D.H. Knepper Jr. and W.H. Langer, 2000.
level so that rain water can easily drain under US Geological Survey (USGS) - The Human Factor In
the construction without causing any prob- Mining Reclamation.
lems. Castro, J., J. Zamora, A. Hodar, J.M. Gómez and L. Go-
mez-Aparicio, 2002. Use of Scrubs as nurse Plants: A
New Technique For Reforestation In The Mediterra-
nean Mountains., Restoration Ecology 10(2): 297-305.
4. CONCLUSIONS Castro, J., J. Zamora, A. Hodar, J.M. Gómez and L. Go-
The design, in terms of functionality of the site mez-Aparicio, 2004, Benefits of Using Shrubs as
Nurse Plants For Reforestation In Mediterranean
is summarized below. A traffic clear network of Mountains: A 4 Year Study. Restoration Ecology
footpaths will connect the various parts. Alter- 12(3): 352-358.
native forms of operations so that different areas Gómez-Aparicio L., Zamora, R., Gómez, J.M., Hódar,
can be used both ways, through individual and J.A., Castro, J., Baraza, E., 2004, Applying Plant Fa-
organized activities, will be developed includ- cilitation To Forest Restoration: A Meta-Analysis Of
The Use of Shrubs As Nurse Plants, Ecological Appli-
cations 14(4): 1128-1138.
Sakellakis, M.V., 2006. Longitudinal Development of
Natural Landscape in Terms of Penteli. Renovation of
open spaces and Territorial Integration of Antiquities.
Graduation Project, Hellenic Open University.
Eleftheriadis, N., 2008. Landscape Management,
Photo/Graphs Studio.
Rokos, D., 2001. Integrated development of mountainous
areas. Theory and Praxis, 3rd Inter-science Inter-
university Conference organized by National Techni-
Figure 5: 3D Model of the proposed Configuration in the cal University of Athens, Metsovon Conference Cen-
Unit. tre, Metsovon, Greece, 7-10 June 2001.
Dimitriou, D. and J.N. Georgi, 2008. The Contribution of
Urban Green Spaces to the Improvement of The city
Environment. Analysis of Experimental Data for the
city of Chania.
Figure 6: Photograph of the model with the proposed con-

figuration.
Copper cementation from the waste effluents of primary copper industry
I. Giannopoulou and D. Panias

National Technical University of Athens, Greece
ABSTRACT Ag, Se, Te, etc.) from the anodic slime pro-
duced in the stage of copper electrorefining.
In this paper, the cementation of copper from
The most important waste effluents in the
the waste effluents generated in the primary
whole pyrometallurgical production process of
copper pyrometallurgical production industry is
copper sulphide concentrates originated from
studied. These wastewaters are strongly acidic
the refining of copper, the regeneration of elec-
aqueous solutions with high concentration of
trolyte in copper electrorefining and the produc-
copper, significant concentration of nickel and
tion of precious metals. All these waste efflu-
minor contents of other metals, such as Pb, Zn,
ents are aqueous acidic solutions, due to the
Fe, As, Sb, Bi, etc. As cementing agent, zero-
high sulphuric acid concentration, with high
valent iron in three different types (particles,
content of copper. They also contain significant
sponge and fillings) was used. The effect of the
quantities of Ni and minor quantities of other
specific surface area and excess of iron used, as
metals, such as Pb, Zn, Fe, As, Sb, Bi, Sn, Cr,
well as the solution pH and agitation rate, on the
etc. According to the European legislation
cementation of copper was investigated, in order
(European Commission, 2000), the treatment of
to determine the optimum operating conditions
these waste effluents is necessary before their
of the process. According to the experimental
discharge into the environment. Additionally,
results, cementation is an easily applied, cost-
the high copper concentration in these effluents
effective and efficient method for the recovery
raises an economic interest for its recovery, in
of copper from the waste effluents generated in
order to be recycled or re-used.
the pyrometallurgical production of copper. A
The most used methods for metal recovery
compromise of its main operating parameters is
from industrial waste effluents with high metal-
necessary for the nearly total copper recovery.
lic content are the chemical precipitation, elec-
The recovered copper can be recycled into the
trolysis and cementation. Although chemical
smelting stage of the copper pyrometallurgical
precipitation is considered as the most efficient
plant.
and economic among these methods for the
treatment of waste effluents (Christiansen and
1. INTRODUCTION Delwiche, 1982; Ying et al., 1988; Tunay and
Kabdasli, 1994; Baltpurvins et al., 1996; Cha-
The pyrometallurgical processing of copper sul- rerntanyarak, 1999; Giannopoulou and Panias,
phide concentrates for the copper production in- 2008), it has important drawbacks. It requires
cludes mainly the stages of roasting, smelting, extremely long setting times and results in a
converting, refining and electrorefining. Gener- bulky sludge that contains usually all the metals
ally, an integrated industry of primary copper of the treated waste effluents in the form of hy-
production includes also a plant for the produc- droxides or sulfides. In addition, this sludge
tion of sulphuric acid from the off-gases of the may contain hazardous and/or toxic metals and
roasting stage that contain SO2, as well as a thus, it is not easily handled. The waste cake
plant for the production of precious metals (Au, produced after the sludge dewatering and filter
pressing is not acceptable for the metallurgical cious metals production, a strongly acidic waste
furnaces, due to the high water content (ap- effluent with a copper concentration of about 6
proximate 70-80%). Generally, its utilization is g/L was generated. Considering that the average
limited, mainly in the manufacture of building volumetric outflow of this effluent was equal to
ceramics as colorizing admixture. Therefore, 16 m3/day almost 35 tons of copper are annually
this sludge has to be dumped on appropriate lost, due to its improper treatment.
waste landfills, increasing this way the final The present work aims at determining the op-
treatment cost of the waste effluents (Ying timum operating parameters of cementation
et al., 1988). process, so as to be effective, reliable and viable
Electrolysis gains increasing attention during for the recovery of copper from the studied
the last years, as an efficient method for the waste effluent. Particularly, the effect of the ex-
treatment of waste effluents containing metals. cess and type of iron used as cementing agent,
However, it is effective when applied on the as well as of the solution pH and agitation rate,
treatment of waste effluents containing a limited on the efficiency of cementation process was
number of metals with high concentration (For- studied.
nari and Abbruzzese, 1999; Juttner et al., 2000;
Janssen and Koene, 2002). Electrolysis requires
specific operating adjustments and controls, 2. EXPERIMENTAL
which increase its maintenance and operating 2.1 Materials
cost. In addition, it necessitates specific reactors
and equipment that render often its application In order to simulate the waste effluent studied in
uneconomical. Generally, the operating cost of this work, a synthetic solution with the chemical
electrolysis is enormous, in comparison to the composition given in Table 1 was used in all the
other treatment methods, as it operates by con- cementation experiments. This solution was
suming electric energy. prepared using chemical reagents of analytical
Cementation is considered as a reasonable grade and distilled water and had pH= 0.4.
and simple method for the recovery of copper As cementing agent, zero-valent iron (ZVI)
from industrial wastewaters (Stefanowicz et al., in three different types (Table 2), namely
1997; Donmez et al., 1999; Dib and Makhloufi, sponge (Alfa Aesar), granules (Alfa Aesar) and
2004; Mubarak et al., 2004; El-Batouti, 2005). It fillings from the cast iron grit (Gotthert Maier
is based on a heterogeneous electrochemical re- GmbH) was used.
action involving the spontaneous reduction of a 2.2 Experimental procedure
metal from ionic solutions and its precipitation
on another, more electropositive metal, which is All the experiments were carried out in batch
oxidized (“sacrificial metal”). Cementation has mode and at room temperature. For each ex-
several advantages, including recovery of met- periment, 100 mL of the synthetic solution was
als in an essentially pure metallic form, simple added to a 250 mL conical flask containing the
control requirements, low energy consumption, appropriate amount of iron, so as to achieve the
high efficiency and low cost. Moreover, the re- pre-defined iron excess. Flasks were put on a
covered metallic sediments are suitable for met- laboratory-scale shaking plate, which operated
allurgical processing (Stefanowicz et al., 1997). Table 1: Chemical composition of the synthetic solution
The main disadvantage of the process is related simulating the studied waste effluents.
to the consumption of the sacrificial metal, es- Element Concentration (g/L)
pecially at low pH values (Dib and Makhloufi, Cu 6,520
2004). Ni 0,688
This paper deals with the application of the Pb 0,003
cementation process for the treatment of the Fe 0,150
waste effluents generated in the primary copper Zn 0,026
pyrometallurgical plant at Bor, Serbia, in order Sb 0,013
to recover the contained copper. In this plant, by Bi 0,040
combining the wastewaters streams originated As 0,458
from the processes of copper refining and pre- SO42- 81,571
at a pre-defined stirring rate (from 0 to 250 rpm) granules fillings sponge

for an appropriate time (from 0.5 to 6 hours). 100
Thereinafter, the solution was filtered and 80
Cu cemented, %
analyzed in order to determine the concentration
of the contained metals according to the Atomic 60
Absorption Spectrophotometry method (Perkin
40
Elmer 2100). The resulted solid phase was dried
and weighed. All the solid phases taken after the 20
cementation experiments were analyzed by X-
Ray Diffractometry (Siemens, D 5000, CuKα 0
radiation, λ=1.5418 Å, 40 kV and 30 mA), in 0 1 2 3 4 5 6
order to define the mineralogical phases that Time, h
were cemented. Figure 1: Cemented copper versus time and type of ZVI
used. Experimental conditions: pH=0.43, stirring
rate=100 rpm, Fe/Cu mass ratio=0.879.
3. RESULTS AND DISCUSSIONS
ing agent, the highest the rate of metal cementa-
3.1 Effect of iron type and excess tion on it.
Copper is less electropositive in respect to iron, According to Table 2, the specific surface
and thus, the following chemical reactions were area of the granules type of ZVI was extremely
expected to occur during cementation treatment lower than that of fillings and sponge types,
of the studied acidic waste effluent (Dib and which had quite similar specific surface areas.
Makhloufi, 2004): Thus, the high specific surface area of fillings
and sponge ZVI types enhanced the copper ce-
Cu2+ + Fe(s) → Fe2+ + Cu(s) (1) mentation rate and the process was completed
very fast, almost in the first hour. At this time,
2H+ + Fe(s) → Fe2+ + H2(g) (2) about 72% and 80% of copper was cemented
Since iron was also contained in the studied using the fillings and sponge types of ZVI, re-
effluent, the following side reactions were also spectively, while only 30% of copper was ce-
expected to occur. mented using the granules ZVI type.
Except for copper, bismuth, antimony, lead
2Fe3+ + Fe(s) → 3Fe2+ (3) and arsenic were also recovered during the
treatment of the waste effluent (Fig. 2). In oppo-
2Fe3+ + Cu(s) → 2Fe2+ + Cu2+ (4) site, neither nickel nor zinc cementation took
In addition, the reduction of other metals, place.
which were contained in the studied effluent and The excess of iron improved the efficiency of
are less electropositive in respect to iron (such copper cementation process (Fig. 3), for all the
as Bi, Sb, As, Pb), was expected to take place. ZVI types used.
According to Equation (1), the stoichiometric The solid phases taken at the end of each ce-
iron demand for the reduction of copper is equal mentation experiment were red-brown, due to
to 0.879 g/g. The fraction of cemented copper as the copper cemented. According to the X-Ray
a function of time for all the types of ZVI used Diffraction analysis performed in these solid
in this work is presented in Figure 1. As shown phases, copper was cemented as Cu and CuO,
in Figure 1, for the sponge and fillings types of when the granules type of ZVI was used, while
ZVI used, the copper reduction reaction (Eq. 1) it was also cemented as Cu2O, when fillings and
reached the equilibrium in the first hour of the Table 2: Characteristics of the used zero-valent iron
process. In opposite, when the granules type of Particle Specific sur-
ZVI was used, the reduction of copper was per- Fe form 0
% Fe size face area
formed with substantially lower rate (Fig. 1). (mm) (m2/g)
Given that the reaction of metal cementation oc- Granules 99.8 1.0-2.0 0.0014
curs on the surface of the cementing agent, the Fillings 92.03 0.2-1.2 0.0482
highest the specific surface area of the cement- Sponge 99.8 0.14-0.29 0.0502
80 Pb Sb As Bi
1
1. Cu 4. Cu2O
4 2. CuO 5. AsCu3
2
Metals recovery, %
60 35 3. Fe
4 1
4 4
Intension, cps
sponge
40
fillings
20
granules
0
20 30 40 50 60 70
0 1 2 3 4 5 6
2θ, degrees
Time, h
Figure 4: XRD diagrams of the solid phases taken after
Figure 2: Metals recovery versus time for the fillings ZVI the experiments, where 20% excess of iron was used, for
type. Experimental conditions: pH=0.43, stirring rate=100 all the ZVI types.
rpm, Fe/Cu mass ratio=0.879.
granules fillings sponge increase of the stirring rate from 0 to 250 rpm
resulted in increasing the cemented copper frac-
100
tion, almost linearly. At the higher applied stir-
Cu cemented, %
ring rate of 250 rpm, about 99% and 86% of the

80 copper contained in the solution was cemented
using ZVI sponge and fillings types, respec-
60 tively. By comparing these copper cemented
fractions with the respective achieved after 6
40 hours cementation under the stirring rate of 100
rpm (Fig. 1), it is obvious that the intense stir-
0 10 20
ring rate improved the efficiency of copper ce-
Iron excess, % mentation.
Figure 3: Cemented copper versus iron excess for all the Generally, stirring of the cementing agent or
ZVI types used. Experimental conditions: pH=0.43, circulation of the solution shortens the metals
time=2 h, stirring rate=100 rpm. cementation time and favors the efficiency of
the process (Stefanowicz et al., 1997). The in-
sponge types of ZVI were used (Fig. 4).
tense and continuous motion of ZVI provided
In addition, the mineralogical phase AsCu3
the maximum contact of copper ions with the
was detected, in the case of using the sponge
iron surface. Thus, high efficiency of copper
ZVI type (Fig. 4). Excluding copper, arsenic
cementation was achieved in a shorter time.
was contained in the waste effluent with the
Moreover, the iron surface was continuously re-
highest concentration among the other cemented
newed through the mechanical friction created
metals and its recovery exceeded 95%, when the
(Stefanowicz et al., 1997), improving the effi-
sponge type of ZVI was used.
100
3.2 Effect of stirring rate
Cu cemented, %
80
The effect of the solution stirring rate on the re-
covery of copper was studied by applying dif- 60
ferent stirring rates on the laboratory shaking sponge
plate used for the cementation experiments. The 40
duration of these experiments was 30 min, while fillings
20% iron excess of sponge and fillings ZVI 20
types was used. The stirring rates of 0, 100 and 0 50 100 150 200 250
250 rpm were applied. Stirring rate, rpm
The fraction of cemented copper versus stir-
Figure 5: Cemented copper versus stirring rate for the
ring rate, for both sponge and fillings ZVI types, sponge and fillings ZVI types used. Experimental condi-
is presented in Figure 5. According to Figure 5, tions: pH=0.43, time=30min, iron excess=20%.
70 cover the contained copper, by using iron as

cementing agent. The process is easily applied
Metals recovery, %
60 Cu
As
and may be cost-effective, taking into account
50 that iron scrap or by-products can be used as
Bi cementing agents. A compromise of the main
40 Sb process operating parameters can achieve nearly
30 Pb total copper recovery from the waste effluents.
The specific surface area and the excess of
20 iron were proved as crucial parameters affecting
0,4 0,8 1,2 1,6 the rate of copper cementation. By using iron
pH with high specific surface area, copper cementa-
Figure 6: Recovery of metals versus pH for the fillings tion was completed after 60 min, under inter-
ZVI type. Experimental conditions: time=30min, iron ex- mediate conditions of solution stirring rate and
cess=20%, stirring rate=100rpm. pH.
Increase of the solution stirring rate shortens
ciency of the process.
the time of copper cementation and increases its
3.3 Effect of pH rate. Besides, the process is favored at solution
pH values between 0.8 and 1.2.
The effect of the solution pH on the efficiency
Except for copper, lead, arsenic, bismuth,
of copper cementation was investigated for the
and antimony contained in the waste effluent are
pH vales of 0.43, 0.79, 1.18 and 1.57. In this
also cemented on iron. Copper was cemented as
experimental series, cementation time, excess of
pure metal, as well as cupric and cuprous ox-
iron and stirring rate were maintained at 30 min,
ides, while it also co-deposits with arsenic. The
20% and 100 rpm, respectively. The fillings
cemented copper, along with the other cemented
type of ZVI was only used in this experimental
metals, can be recycled into the smelting stage
series, taking into account that it is a by-product
of the primary copper pyrometallurgical plant.
generated in the process of cast iron gritting and
thus, is cheaper than iron sponge.
According to Figure 6, cementation process REFERENCES
was favored in the solution pH range from 0.8 Baltpurvins, K.A., R.C. Burns and G.A. Lawrance, 1996.
to 1.2, for all the metals recovered. Generally, Heavy metals in wastewater: modeling of hydroxide
the pH values between 1 and 1.5 are considered precipitation of copper (II) from wastewater using
as optimum for efficient copper cementation. lime as the precipitant, Waste Management 16 (8):
At pH < 1, a compact copper deposit or a fine 717-725.
oxide film is formed on iron surface Charerntanyarak, L., 1999. Heavy metals removal by
chemical coagulation and precipitation, Water Science
(Karavasteva, 2005) inhibiting the dissolution of and Technology 39 (10): 135-138.
iron, as well as the cementation of copper (Ste- Christiansen, E.R. and J.T. Delwiche, 1982. Removal of
fanowicz et al., 1997; Karavasteva, 2005). At heavy metals from electroplating rinsewaters by pre-
pH values higher than 1.5, hydrolysis of iron cipitation, flocculation and ultrafiltration, Water Re-
and antimony contained in the solution may be sources 16: 729-737.
Dib, A. and L. Makhloufi, 2004. Cementation treatment
occurred (Giannopoulou and Panias, 2008) re- of copper in wastewater: mass transfer in a fixed bed
sulting in the deposition of hydroxide precipi- of iron spheres, Chemical Engineering and Processing
tates on the iron reacting surface (Stefanowicz 43: 1265-1273.
et al., 1997; Karavasteva, 2005). Under these Donmez, B., F. Sevim and H. Sarac, 1999. A kinetic
conditions, iron dissolution and copper cemen- study of the cementation of copper from sulphate solu-
tation are inhibited. tions onto a rotating aluminium disc, Hydrometallurgy
53: 145-154.
European Commission, 2000. Commission Decision of 3
May 2000 replacing Decision 94/3/EC establishing a
4. CONCLUSIONS list of wastes pursuant to Article 1(a) of Council Di-
Cementation treatment of the strongly acidic rective 75/442/EEC on waste and Council Decision
waste effluents generated in the primary copper 94/904/EC establishing a list of hazardous waste pur-
suant to Article 1(4) of Council Directive 91/689/EEC
pyrometallurgical industry can effectively re- on hazardous waste. Official Journal of European
Communities (6/9/2000) L226: 3-24.

El-Batouti, M., 2005. Removal of copper metal by cemen-
tation using a rotating iron cylinder, Journal of Colloid
and Interface Science 238: 123-129.
Fornari, P. and C. Abbruzzese, 1999. Copper and nickel
selective recovery by electrowinning from electronic
and galvanic industrial solutions, Hydrometallurgy 52:
209-222.
Giannopoulou, I. and D. Panias, 2008. Differential pre-
cipitation of copper and nickel from acidic polymetal-
lic aqueous solutions, Hydrometallurgy 90 (2-4): 137-
146.
Janssen, L.J.J. and L. Koene, 2002. The role of electro-
chemistry and electrochemical technology in envi-
ronmental protection, Chemical Engineering Journal
85: 137-146.
Juttner, K., U. Galla and H. Schmieder, 2000. Electro-
chemical approaches to environmental problems in the
process industry, Electrochemica Acta 45: 2575-2594.
Karavasteva, M., 2005. Kinetics and deposit morphology
of copper cementation onto zinc, iron and aluminium,
Hydrometallurgy 76: 149-152.
Mubarak, A.A., A.H. El-Shazly and A.H. Konsowa, 2004.
Recovery of copper from industrial waste solution by
cementation on reciprocating horizontal perforated
zinc disc, Desalination 167: 127-133.
Stefanowicz, T., M. Osinska and S. Napieralska-Zagorda,
1997. Copper recovery by cementation method, Hy-
drometallurgy 47: 69-90.
Tunay, O. and N.I. Kabdasli, 1994. Hydroxide precipita-
tion of complex metals, Water Resources 28 (10):
2117-2124.
Ying, W.C., R.R. Bonk and M.E. Tucker, 1988. Precipita-
tion treatment of spent electroless nickel plating, Jour-
nal of Hazardous Materials 18: 69-89.
The pH effect on lactose uptake rate by kefir cells, during whey

fermentation
A. Golfinopoulos, M. Soupioni, M. Kanellaki and A.A. Koutinas

Food Biotechnology Group, Section of Analytical Environmental and Applied Chemistry, Depart-
ment of Chemistry, University of Patras, Greece
ABSTRACT kefir was thought to be ideal for exploitation of

whey lactose (Kourkoutas et al., 2002; Athana-
Whey is the liquid by-product of cheese produc-
siadis et al., 2004). Kefir is a mixed culture con-
tion and a grave environmental pollutant in
sisting of various yeasts (Kluyveromyces, Can-
Greece. Also, whey is of great nutritious value
dida, Saccharomyces, Pichia et al.), lactic acid
as it contains proteins, fats and mainly carbohy-
bacteria (Lactobacillus, Streptococcus, Leu-
drates (lactose ~ 5 %). This study aims on de-
conostoc et al.) and acetic bacteria embedded
termining pH effect on lactose uptake rate by
into a matrix consistent of polysaccharides
the mixed co-culture of kefir during whey fer-
(Garrote et al., 1997; Pintado et al., 1996). The
mentation. To study the effect of pH value, a se-
microbiological composition of kefir ensures
ries of whey fermentations were carried out at
simultaneous alcoholic and lactic acid whey
4.0, 5.0, 5.5, 6.0 and 6.5 pH values using
fermentations resulting useful products forma-
pressed wet weight kefir cells of 2.4 % w/v at
tion like ethanol, lactic and propionic acid. In
30 °C. Lactose uptake rate was recorded by us-
order to convert this waste to value added prod-
ing 14C-labelled lactose. Fermentation kinetic
ucts, first of all the kinetic parameters of whey
and ethanol productivity were monitored by
lactose fermentation need to be examined thor-
specific lactose measurements using HPLC
oughly. Because the whey fermentation rate is
method. Results illustrated that lactose uptake
affected by the lactose uptake rate, in the pre-
rate and ethanol productivity were higher at 5.5
sent work the effect of pH value on lactose up-
pH value.
take rate by kefir cells was studied. In this in-
vestigation 14C-labelled lactose was used to
1. INTRODUCTION study kefir fermentation ability through lactose
uptake rate as has before done (Soupioni et al.,
Last thirty years, Greece has strong developed 1998). Also, residual lactose and ethanol pro-
milk industry and therefore about 700,000 tones ductivity were monitored using HPLC method.
of whey are produced annually. Whey is the liq-
uid by-product of the cheese processing units,
which represents an important environmental 2. MATERIALS AND METHODS
problem due to its bulk capacity and its high or-
2.1 Microorganism and cell growth
ganic load of 40-70 g/L BOD5 and 60-80 g/L
COD (Athanasiadis et al., 2004). At the same Kefir, a commercial product in the form of
time, whey is of great nutritious value as it con- grains was employed in the present study. Kefir
sists of proteins (0.8 - 1.6 %), fats (0.3 - 1.3 %) grains were preserved into 1 L of fresh pasteur-
and mainly lactose (4.8 - 5.3 %). Therefore ised full cream milk, which was renewed every
there should be further whey processing aiming week, at approximately 4 °C. Cell growth of ke-
in reducing its polluting load and possible pro- fir biomass took place in a sterilized synthetic
duction of useful fermented products as ethanol. medium consisted of 2 %, yeast extract 0.4 %
Last decade due to its ability to ferment lactose, (Fluka), (NH4)2SO4 0.1 %, KH2PO4 0.1 % and
MgSO4.7H2O 0.5 % (Merck 99,5 %), at 30 °C.

Pressed wet-weight cells were prepared centri- 5
pH 4
fuged and used directly in fermentations. pH 5
4 pH 5.5
2.2 Effect of pH value pH 6
Lactose (g/100mL)
pH 6.5
A series of whey fermentations were carried out 3
at 4.0, 5.0, 5.5, 6.0 and 6.5 pH values at 30 °C.

Every time 250 mL of sterilized whey were 2
transferred into a 500 mL Erlemeyer flask with

6 g of wet-weight kefir biomass. At the begin- 1
ning of each run a small quantity of labelled lac-

tose [D-glucose-1-14C], (ARC 0466 lactose, 0
0.1 mCi/mL) was added. All treatments were
carried out in triplicate for statistical reasons. 0 20 40 60 80 100 120
Time (h)
2.3 14C labelled lactose determination Figure 1: Lactose fermentation kinetic observed for vari-
The labelled lactose was fermented in the same ous initial pH values during whey fermentation.
way as the non active one. At various time in- tion kinetic is illustrated. It is obvious that 5.5
tervals samples of 2 mL were filtered using cel- pH value was the best for whey fermentation by
lulose membrane filters (0.45 μm) and 14C was kefir cells. At 6 and 6.5 pH values a reduction of
determined by liquid scintillation. The amount the fermentation rate was observed. Therefore,
of the labelled lactose consumed by a specific the high effect of pH value in the case of kefir
amount of kefir biomass during fermentation yeast natural co-culture fermentation can be at-
was determined and expressed as cpm of lactose tributed either to lactose hydrolysis was affected
per gram of kefir biomass per hour. by pH through of the pH effect on lactasse en-
2.4 Liquid scintillation measurements zyme activity or through of the low resistance at
lower pH value of various lactic acid bacteria
All cellulose filters with cells containing 14C, are contained in kefir microflora.
were put one by one in appropriate vials and 5 In Figure 2, it is shown that 5.5 pH value was
mL of liquid scintillation cocktail Opti Fluor the one where the higher lactose uptake rate was
(Perkin Elmer) were added. Measurements were indicated. Lactose uptake rate was increased as
performed on a PACARD-3255 liquid scintilla- pH increases up to 5.5. At higher pH values lac-
tion counter and then interfaced to an APPLE-2 tose uptake rate reduction was observed. This
personal computer for data evaluation. conclusion was identified by fermentation rate
2.5 Determination of Ethanol and Residual changes, which was increased up to 5.5 pH
Sugar
50
Residual sugar and ethanol concentrations in the
fermented liquid were determined on a Shima-
Lactose uptake rate [cpm/(gbiomass.h)]
pH 4
40 pH 5
dzu LC-9A HPLC system consisting of a Shim- pH 5.5
pack SCR-101N column, an LC-9A pump, an 30
pH 6
pH 6.5
RID-6A refractive index detector, a CTO-10A
column oven, and a DGU-2A degassing unit.
20
Supra pure water was used as the mobile phase
(0.8 mL/min), and 1-butanol (0.1 % v/v) was
used as an internal standard. Column tempera- 10
ture was 60 oC. Sample dilution was 1 % v/v,

and injection volume was 40 mL. 0
0 20 40 60 80 100 120
Time (h)
3. RESULTS AND DISCUSSION Figure 2: Lactose uptake rate vs time during whey fer-
In Figure 1, pH effect on the lactose fermenta- mentation at various pH values.
laki, I.M. Banat and R. Marchant, 2002. Continuous

0,6 whey fermentation using kefir immobilized on delig-
pH 4
nified cellulosic material, J. Agric. Food Chem., 50,
0,5
pH 5 2543-2547.
pH 5.5 Pintado, M.E., J.A. Lopes Da Silva, P.B. Fernades, F.X.
EtOH [(g/(L.h)]
0,4 pH
pH
6
6.5
Malcata and T.A. Hogg, 1996. Microbiological and
rheological studies on Portugese kefir grains, Interna-
0,3 tional Journal Food Science and Technology, 31, 15–
26.
0,2 Soupioni, M., E. Polichroniadou, M. Τokatlidou, M.
Kanellaki and A.A. Koutinas, 1998. Glucose uptake
0,1 rate by Saccharomyces cerevisiae in the presence of
promoters of alcoholic fermentation using 14C- la-
0 20 40 60 80 100 120
belled glucose, Biotechnology Letters, 20(5), 495-497.
Time (h)
Figure 3: Ethanol productivity during whey fermentation

at various pH values.
value and then it went decreasing at higher pH

values (Fig. 1).
In Figure 3 pH effect on ethanol productivity
during whey fermentation by kefir cells is illus-
trated. At 5.5 pH value the highest ethanol pro-
ductivity was observed. This result was identi-
fied by fermentation rate as well as lactose up-
take rate changes, which were increased up to
5.5 pH value and then they reduced at higher pH
values (Figs. 1 and 2).
4. CONCLUSIONS
It was clear that kefir culture was suitable for
whey fermentation in order to reduce whey pol-
lutant load and produce useful products as etha-
nol. 5.5 pH value was the best for whey fermen-
tation by kefir culture. This is probably due to
the fact that kefir is consisted of a number of
microorganisms for which 5.5 pH value is quite
conventional condition.
Further scientific research is needed for
studying the effect of other conditions such as
temperature, initial kefir concentration and use
of biocatalyst on increasing fermentation rate
and alcohol productivity.
REFERENCES
Athanasiadis I., A. Paraskevopoulou, G. Blekas G., and
V. Kiosseoglou, 2004. Development of a Novel Whey
Beverage by Fermentation with Kefir Granules. Effect
of Various Treatments, Biotechnol.Prog., 20, 1091-
1095.
Garrote, G.L., A.G. Abraham and G.L. De Antoni, 1997.
Preservation of kefir grains, a comparative study, Le-
bensmittel-Wissenschaft un Technologie, 30, 77–84.
Kourkoutas, Y., C. Psarianos, A.A. Koutinas, M. Kanel-
Assessing biosorption on anaerobic dyes removal
I.C. Gonçalves, S. Abrantes and M.I. Ferra

University of Beira Interior, Covilhã, Portugal
H.M. Pinheiro
Technical University of Lisbon, Portugal
ABSTRACT region and the main characteristic peak re-

mained in HPLC chromatograms. This indicated
An improvement of color removal was achieved
that adsorption was prevalent during the initial
(96-100%) in an anaerobic bioreactor in which a
stages of dye removal in anaerobic bioreactors,
natural adsorbent, spent brewery grains (SBG)
bearing in mind that, for less soluble dyes, pre-
was incorporated. SBG could work as adsorbent
cipitation could also contribute to decoloriza-
as well as an electron shuttle (lignin frac-
tion.
tion/humic acids) in the dye removal mecha-
nism. Anaerobic biomass also can undergo ad-
sorption and reductive mechanisms. However, 1. INTRODUCTION
little is known about which dye removal mecha-
nism, degradation/adsorption, predominates and Azo dyes represent 60-70% of the more than
to what extent. The aim of the present work was 10000 dyes currently manufactured (Carliell
to evaluate the adsorption susceptibility of dyes et al., 1996). Most dyes cause negative aesthetic
on SBG and anaerobic biomass adapted and non effects. In addition, they can be toxic to aquatic
adapted to SBG (active and autoclaved). Batch life and carcinogenic and mutagenic to humans
tests were performed for 60 minutes at 37±1ºC (Brown and Hamburguer, 1987). Azo dyes may
with SBG, and both active and autoclaved an- be decolorized by several mechanisms including
aerobic mesophilic biomass. The adsorb- adsorption and reductive cleavage of the azo
ent/solution ratio was of 1:4 (v/v) and the bio- bond, to which the color is associated, via an-
mass concentration was of 6-8 gSS/L. In pre- aerobic degradation (Brás et al., 2001).
liminary tests performed with 24 azo dyes at a Adsorption is probably the first step of the
initial concentration of 100 mgl-1 and active an- mechanisms associated to anaerobic azo dye
removal, and depends on the sorbate-sorbent in-
aerobic biomass, a color removal rate between 0
teraction and the system conditions. It is known
and ~99% was achieved. Depending on the dye,
that materials of biological organic compounds,
adsorption and precipitation seem to be associ-
such as vegetal and bacterial biomass, display a
ated to color removal. More soluble dyes pre-
high adsorption capacity and have been used as
sented lower values for decolorization. A dye
biosorbents in the removal of several pollutants
concentration of 30 and 50 mg/L was then
(Akzu, 2001; Silva et al., 2004).
tested, with 5 dyes, and different types of bio-
Brewers’ spent grain (BSG) is a lignocellu-
mass (bacterial and vegetal). AO8 was then se-
losic material with a high content of protein and
lected for further studies. It is a soluble dye and
fibre (ca 20 and 70% dry basis, respectively),
displayed a significant adsorption (the percent-
non-cellulosic polysaccharides, lipids and some
age adsorbed was of 71% for SBG, 36% for ac-
polyphenolic components. Such rich composi-
tive biomass and 62% for autoclaved biomass).
tion makes it viable to be used as a conditioning
Ultraviolet-visible spectra of dye solutions did
agent in anaerobic reactors along with favoring
not show marked alterations (in systems with
the biodiversity in anaerobic biomass and bio-
and without BSG). The absorbance decreased in
gas production (Mussato et al., 2006).
the whole range of the spectrum in UV-visible
In tandem with adsorption the following 60; (NH4)6MoO7.4H2O 180; NiCl2.6H2O 100;
mechanism that presumably occurs in anaerobic Na2SeO3.5H2O 200 and HCl (37%) (1). Sodium
reactors, is reduction of azo dyes via anaerobic bicarbonate was used (6-9 g l-1) to neutralize the
bacteria as an non-specific process, in which re- medium fed to the anaerobic vessels. For the ad-
ducing equivalents from an external electron sorption tests with BSG a phosphate buffer solu-
donor (biologically or chemically generated) are tion was used (1.28 g l-1 NaH2PO4.2H2O and
transferred to the dye (Santos et al., 2006). This 0.42 g l-1 NaH2PO4). All solutions were pre-
results in the formation of aromatic amines that pared with tap water.
are hardly metabolized anaerobically, with the
2.4 Chemicals
exception of a few examples bearing hydroxyl
and carboxyl groups, which can be fully de- All dyes, including selected azo dyes Acid Or-
graded (Razo-Flores et al., 1996). So a higher ange 8 (AO8), Acid Red 151 (AR151), Acid
biodiversity of anaerobic biomass could also Red 114 (AR114), Acid Blue 120 (AB120) and
improve the elimination of the persistent aro- Acid Blue 113 (AB113) were purchased from
matic amines. Sigma-Aldrich (Germany). Their chemical
The present study was undertaken to assess structures are indicated in Table 1. Chemicals
the contribution of biosorption on the color re-
Table 1: Dyes selected for adsorption studies with differ-
moval process that occurs in the presence of an- ent types of biomass (bacterial and vegetal).
aerobic biomass (adapted and non adapted to Commercial
BSG), in active and autoclaved conditions. This Dye structure
name
will help to clarify which mechanism (adsorp-
tion/reduction) is prevailing, or if both play a vi- Acid Orange 8
tal role. (AO 8)
C.I. *: 15575
2. MATERIALS AND METHODS Acid Red 151
2.1 Raw material (AR 151)
C.I. *: 26900
The BSG used in the experiments had the fol-
lowing composition (% dry weight, w/w): 24%
protein, 60% fibre, 10% ashes and moisture and
6% lipids. It was gently supplied by a brewery Acid Red 114
from Lisbon region. (AR 114)
2.2 Anaerobic biomass C.I. *: 23635
Anaerobic biomass was collected from two

mesophilic lab-scale continuous reactors, fed
with a synthetic wastewater containing a basal
medium and glucose as soluble carbon source.
In one of the reactors BSG was incorporated as Acid Blue 113
biosorbent/conditioning agent (BSG adapted (AB 113)
biomass). C.I. *: 26360
2.3Basal medium
The basal medium used in the anaerobic reac-
tors and adsorption tests contained (mg l-1)
Ca(OH)2 90; NH4Cl 170; NaH2PO4.2H2O 48; Acid Blue 120
MgCl2.6H2O 11; KCl 25; and 0,5 ml l-1 of a (AB 120)
trace elements solution with (mg l-1) H3BO3 C.I. *: 26400
100; FeCl2.4H2O 4000; EDTA (Ethylene Dia-
mine Tetracetic Acid, sodium salt) 2000;
ZnCl2.4H2O 100; MnCl2.4H2O 50; CuCl2.2H2O *C.I. – Colour Index
were dissolved in the basal medium in concen-

trations of 30 mg l-1, 40 mg l-1 and 50 mg l-1.
These concentrations are usually found in textile
effluents. Methanol and acetonitrile were pur-
chased from Sigma and Merck respectively.
All chemicals were of high purity and com-
mercially available.
2.5 Analyses
Color was measured spectrophotometrically
with Helios Alpha spectrophotometer (Unicam,
UK.) at the maximum visible absorbance wave-
length of the dye (487 nm for AO8, 513 nm for
AR151, 516 nm for AR114, 559 nm for AB120
and 536 nm for AB113). Absorbance at these
wavelengths was correlated with dye concentra-
tion and used to quantify decolorization.
Suspended solids (SS), volatile suspended
solids (VSS) and pH in liquid samples were de-
termined according to Standard procedures
(APHA, 1995).
2.6 Batch adsorption tests
Figure 1: Scheme of the adsorption tests performed with
A total of 24 azo dyes were studied with active several azo dyes.
mesophilic anaerobic biomass in batch condi-
tions. Preliminary decolorization tests were car- BSG with buffer solution was used as blank
ried out in the same conditions as indicated for (Fig. 1).
biosorption studies, with an initial dye concen- UV-visible spectra and HPLC chroma-
tration of 100 mg/L. From those, 5 dyes were tograms were run to follow the adsorption of
selected to proceed with decolorization tests. those compounds. The amount of dye adsorbed
These 5 dyes (Table 1) and two types of an- was calculated as the difference between the
aerobic mesophilic biomass, with and without amount of test substance initially present in so-
BSG (both active and autoclaved), and dried lution and the amount remaining at the end of
BSG were suspended in 50 ml of solution con- the experiment.
taining a known concentration of dye. Dye solu-
tions (30, 40 and 50 mg l-1) were prepared with 3. RESULTS AND DISCUSSION
basal medium for tests performed with the two
types of bacterial biomass and with the buffer As pH is an important parameter to control on
solution for tests with dried BSG. The experi- adsorption and biological systems, all experi-
ments were conducted in amber glass flasks ments were carried out at neutral pH (7.5±1). In
(100 ml capacity) for 60 minutes, equipped with preliminary tests performed with active meso-
magnetic stirring, at 37±1ºC. The ad- philic anaerobic biomass, significant differences
sorvent/solution ratio was of 1:4 (v/v) and the were observed on color removal rate, for the 24
biomass (bacterial and vegetal) concentration dyes studied (Table 2). The number of aromatic
was of 6-8 gSS l-1. Samples were taken before rings, azo bonds and sulphonic and others sub-
mixing the biosorbent and dye bearing solutions stituent groups in the dye molecule determine
and at pre-determined time intervals for the re- their hydrophilic/hydrophobic character and the
sidual dye concentration in solution. Before type of interactions to be established with other
analysis samples were centrifuged at 14000 rpm groups present in substrates to be linked (as tex-
and the supernatant fraction, diluted when nec- tile substrates). So the behaviour of the different
essary, was analyzed for the remaining dye. A dyes when in presence of bacterial biomass will
flask containing biomass and basal medium or be affected in a similar way. Despite of all dyes
Table 2: Color removal in preliminary batch tests per- of absorbance in the visible region, suggesting
formed with mesophilic anaerobic biomass, at the end of biosorption/precipitation as mechanisms respon-
60 min.
sible for decolorization, for the three biosor-
Dye Colour removal (%)
bents tested. In the UV region the absorbance is
Acid Orange 6 0
mainly due to the blank test (Fig. 2)
Acid Orange 7 24.1
Color removal ranged from 25.0% to 92.4%
Acid Orange 8 34.6
at the end of 60 minutes contact time (Table 3).
Acid Orange 10 18.0
Acid Orange 12 35.2
AO8 reached the highest value for BSG and the
Acid Red 8 10.6
lowest for active biomass. It is the smallest dye
Acid Red 14 13.9 tested and has the smallest number of aromatic
Acid Red 18 18.8 rings. As obtained from preliminary tests,
Acid Red 73 21.6 AR151 and AR113 have a very efficient color
Acid Yellow 9 25.8 removal rate, for the three types of biomass
Acid Yellow 17 6.4 tested (from 89.7 to 98.3%).
Acid Yelow 36 27.3 AB120 andAB113 display similar biosorp-
Acid Blue 29 33.8 tion of the three type of biosorbents, although
Acid Blue 113 96.6 with lower values than those obtained in the
Acid Blue 120 >99 preliminary experiments, which can be associ-
Acid Blue 161 46.9 ated to the lower concentrations essayed.
Acid Black 1 36.9 Also it can be inferred that adsorption occurs
Acid Violet 7 15.9 essentially in the first 30 minutes (Table 3a, b).
Brilliant Black BN 49.1 Dye AO8 was selected for further studies
Brilliant Yellow 8.4 since it displays a markable absorption peak
Reactive Orange 16 0.0 (high absortivity) in the visible range, even for
Acid Red 88 12.1 low concentrations, which renders it easier to be
Acid Red 114 94.6 measured by UV-visible spectrophotometry
Acid Red 151 >99
Acid Orange 8
contain azo bonds, as shown in Table 1, the
color removal rate has a broad range of values 4,000
AO8
(ranging from no removal to almost a complete
Absorbance
3,000 t30min
removal). Spectra of UV-visible run in the be- 2,000 t60min
ginning and at the end of the experiments indi- 1,000 BSG blank
cate that the main mechanism involved on color 0,000
removal, in the first 60 minutes, was adsorption, 200,0 400,0 600,0 800,0
since the absorbance decreased in the whole
w avelength(nm )
range of the spectra. Nevertheless, some dyes
seem to display low solubility (as for instance
a)
AR151, AB113 and AB120). So in tandem with
adsorption a precipitation mechanism, rendering Acid Blue 120
the increase of dye removal rate from the solu-

4,000
tion, should not be discharged.
Absorbance
3,000 AB120
From these, 5 dyes, 4 with a high color re- 2,000 t30min
moval rate (AR114 AR151, AB113 and AB120) 1,000 t60min
and 1 with ca 35% color removal (AO8) were 0,000
studied for lower initial concentrations (30- 200,0 400,0 600,0 800,0
50 mg l-1), in the presence of 3 types of biosor- w avelength (nm )
bents. Examples of some ultraviolet-visible
spectra of dye solutions collected at the begin- b)
ning and at 30 and 60 minutes contact time, for
Figure 2: Examples of spectra obtained for dyes Acid Or-
different types of biomass, are depicted in Fig- ange 8 (a) and Acid Blue 120 (b) in tests performed with
ures 2 and 3. Results indicate a sharp decrease BSG.
Table 3: Dye removal (%) in BSG, active and autoclaved 4. CONCLUSIONS

biomass tests at 30 (a) and 60min (b) contact time.
a) In order to clarify some aspects involved on azo
Biosorbent Active Autoclaved dyes decolorization in anaerobic bio-reactors, it
BSG
Dyes Biomass Biomass should be pointed out:
AO8 67.4 33.1 52.6
- Results from experiments performed with 24
AR151 92.0 95.3 97.0
azo dyes and active biomass show that, in the
AR114 90.0 91.2 92.9
first 60 min, only dyes Acid Orange 6 and
AB113 56.0 74.9 59.5
AB120 17.2 38.4 58.9
Reactive Orange 16 were not removed. For
b) the other 22 dyes decolorization ranged from
Biosorbent Active Autoclaved 6.4 to >99% (Table 2).
BSG - For the 5 dyes studied with active and auto-
Dyes Biomass Biomass
AO8 71.0 36.0 62.1 claved biomass and BSG, removal rate
AR151 92.0 96.0 98.3 ranged from 36 to 98.3%, being the highest
AR114 94.9 89.7 92.4 values achieved for dyes with the lowest
AB113 77.7 70.0 66.3 solubility. The most soluble dye (AO8) dis-
AB120 50.1 41.8 58.9 played lower color removal rates. In this case
dye precipitation should not be relevant.
(Fig. 3). - Depending on the dye tests also suggest that
From the results obtained with anaerobic adsorption and/or precipitation could be the
biomass incubated with BSG, adsorbed and non main mechanisms responsible for color re-
adsorbed dye was measured and the recovered moval in anaerobic bio-reactors, in the first
dye was calculated from extraction tests and 30-60 min.
analyzed by HPLC. Adsorption seems to be the - AO8 was also tested with active biomass in-
predominant mechanism of color removal. cubated in the presence of BSG leading to an
improvement of dye adsorption (higher de-
AO8
colorization).
5
Absorbance
4 ACKNOWLEDGEMENTS
3 AO8
2 t30min The authors are grateful to the project
1
t60min
PDCT/AMB/59388/2004 from Fundação para a
0
t7days
Ciência e Tecnologia (FCT), Portugal
200 400 600 800
w avelength(nm )
REFERENCES
a) Akzu, Z., 2001. Biosorption of reactive dyes by dried ac-
tivated sludge: equilibrium and kinetic modelling.
AB120
Biochemical Engineering Journal, 7: 79-84.
Brás, R., M.I.A. Ferra, H.M. Pinheiro and I.C. Gonçalves,
6,000 2001. Batch tests for assessing decolourisation of azo
A b so rb an ce
dyes by methanogenic and mixed cultures. Journal of

4,000 AB120 Biotechnology, 89, 155-162.
2,000 t30min Brown, D. and B. Hamburguer, 1987. The degradation of
dyestuffs: part III - investigations of their ultimate de-
0,000 t60min gradability. Chemosphere, 16, 1539-1553.
200 400 600 800 t7days Carliell, C.M., S.J. Barclay and C.A. Buckley, 1996.
Treatment of exhausted reactive dyebath effluent us-
w avelength (nm) ing anaerobic digestion: laboratory and full-scale tri-
als. Water SA, 22, 3, 275-233.
b) Mussato, S.I., G. Dragone and I.C. Roberto, 2006.
Figure 3: Examples of spectra obtained for dyes Acid Or- Brewer´s spent grain: generation, characteristics and
ange 8 (a) and Acid Blue 120 (b) in tests performed with potential applications. Journal of Cereal Science, 43:
active biomass. 1-14.
Santos, A.B., M.P. de Madrid, F.A.M. de Bok, A.J.M.

Stams, J.B. van Lier and F.J. Cervantes, 2006. The
contribution of fermentative bacteria and methano-
genic archaea to azo dye reduction by a thermophilic
anaerobic consortium. Enzyme and Microbial Tech-
nology, 39, 38-46.
Silva, J.P., S. Sousa, J. Rodrigues, H. Antunes, J.J. Porter,
I. Gonçalves and S. Ferreira-Dias, 2004. Adsorption of
acid orange 7 in aqueous solutions by spent brewery
grains, Separation and Purification Technology, 40:
309-315.

Complete

Uploaded by

Copyright:

Available Formats

Complete

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Complete

Uploaded by

Copyright:

Available Formats

3rd AMIREG International Conference (2009): Assessing the Footprint of 1

Resource Utilization and Hazardous Waste Management, Athens, Greece

Environmental monitoring of water resources as a project management tool

ABSTRACT spatial distribution of population. The final out-

Europium removal from aqueous solutions using barley rootlets as

V. Anagnostopoulos, M. Anzaoui and B. Symeopoulos

ABSTRACT accumulation of metal ions by materials of bio-

10 ml of europium solution of known initial 160

agitated for a length of time varying between 1 120

min and 24 h in order to determine the neces-

ments initial metal concentrations ranged from 20

3.2 Sorption kinetics

160 is deduced that the sorption process occurred in

mol-1 ion-exchange is the dominant factor and if

80 E>16kJ mol-1 sorption is dominated by particle

Cfin (mg l-1)

Preliminary investigation of the in-plane shear behavior of masonry

Ch. Apostolaki, Z. Agioutantis, V. Perdikatsis and Th. Markopoulos

ABSTRACT of the world heritage. Unreinforced masonry is

mode a mode b mode c

accommodate before any slip occurs. This value

Figure 2: Triplet test set up for shear-compression load- where:

2.3 Materials and preparation of prisms and mould a

3.1 Compressive and tensile strength couplet 1 couplet 2 triplet

The samples were subjected to a monotonously 0.80

increasing load until damage occurred. The 0.70

test. The mean value of six tested specimens is 0.40

reported in Table 1 for the compressive as well 0.30

as tensile strength of the mortar for a curing 0.20

couplet 1 couplet 2 triplet

Figure 10: Comparison of numerical results for the force -

Social Fund (75%) and National Resources

An investigation of the hydration and carbonation mechanisms of natural

Ch. Apostolaki, Th. Markopoulos, V. Perdikatsis and Z. Agioutantis

ABSTRACT lime and named it “Roman cement”. This name

slow slaked binder mineralogical analysis given Periclase Quartz CaO

zero curing time. The quantitative analysis of 40.000

the five different binders used in this experiment 30.000

on the first curing stage (t=0 days) is given in 20.000

Table 3. It is clearly shown in Table 3, that the 10.000

dominant silicate in the NHL slow slaked binder 0.000

are tricalcium aluminate (C3A) and brown- BtN Days

The most interesting observation of this spe-

FtN portlandite. It has been proved that the slaking

This was the main reason for the inclusion of 100

synthetic larnite as well as the pure calcitic 90

ment. The goal was to determine the mecha- 60

nisms of hydration and carbonation of the NHL 50

mechanisms the addition of the reference mate- 20

rials in the experiment were vital.

Consequently Figures 3 and 4 present the 0 5 10 15 20 25 30 35

carbonation mechanism of NL binders. The car-

bonation of portlandite in a rapid slaked NL Figure 5: Decomposition of synthetic larnite to carbonates

44.5 0.05 In Figures 9 and 10 the hydrated phases of

hydrate (Ca3Al2O6·xH2O) and katoite

The conclusions that can be drawn from these

contrary in the NL binder with the larger crys- Portlandite (001)

portlandite was transformed to CaCO3 at the cium Aluminum Iron Hydroxide

same curing period. The carbonation of the poor 1000