The effect of Flue Gas Desulphurization (FGD) gypsum on structural development
of clayey soil from the open cast lignite mining pit in Nochten has been studied
using two soil-physical parameters, shear stress and penetration resistance.
The 100% lime and 100% FGD gypsum saturation in soil was selected as a base
factor. This work also verifies the structural development with respect to the
drainage (swelling and shrinkage) cycles, which are typical in the field and
are compared to the complete drainage in field. Kodikara
et al. (1999) states that shrink/swell or wet/dry cycles can change
the clay structure in soils. This work presents a synthesis on clay structure
development and behavioral changes of soil using FGD gypsum and lime amendments
with drainage cycles.
The Nochten surface clay represents kaolinitic clay according to its mineralogical
composition. It consists of 20.7% quartz and 79.3% clay minerals, of which 90.4%
are two-layered kaolinitic clay minerals. The use of kaolinitic clay is limited
due to its low surface area (Milheiro et al., 2005).
As a result of its low surface area, kaolinitic clay has limited utility as
a carrier for active agricultural ingredients (Bruand et
al., 2001). Therefore, the two-layered kaolinitic clay minerals possess
smaller source ability contrary to that of three-layered clay minerals, as montmorillonite
A special problem regarding the recultivation of Nochten surface clay and sand
substrates is the slow decomposition of clay by swelling and shrinking processes
(Narra, 2008). In order to accelerate the soil-physical
and soil-chemical development of Nochten surface clay for recultivation, the
use of FGD gypsum is suggested due to its superficial penetration into clay
minerals and possible expansion leading to the development of soil structure
Ideal percentage of clay in soil should be ≤20%. Percentage of clay above
20% hinders root growth and when the clay percentage is ≥40%, root growth
will be limited (Miller, 2001) Nochten surface clay
ranged in between 43 and 55% having a pH of 4.2. The objective of this work
is to verify whether FGD gypsum can effectively disrupt the structure of clay
in comparison to lime.
MATERIALS AND METHODS
As a re-cultivation substrate Pleistocene sand and Nochten surface clay from the pre-cut section are mixed in a ratio of 65:35 (sand:clay), similar to the substrate mixture occurring after the consecutive mining operations (excavation, transportation and deposition). The investigations were carried out from year 2006 till year 2008.
To find out the complete amount of FGD gypsum and lime required to obtain 100% salt saturation in the sand-clay mixture, lime and FGD gypsum are added in increments of 0.0453 g lime and 0.0779 g FGD gypsum respectively per every 100 g of sand-clay mixture. The lime and FGD gypsum increment were selected based on the burnt lime present in each of them (Table 1).
The samples are prepared with 100 g sand-clay mixture in small aluminium cups
with increasing concentrations of lime and FGD gypsum. The mixture is mixed
with 20% vol water and is left for 72 h to air-dry. The soil crust is then pulverized
and the electrical conductivity of the soil solution is measured. Electrical
conductivity versus concentrations/increments of lime and FGD gypsum are plotted
and subsequently 50 and 100% saturation values are calculated as shown in Fig.
1. It was expected that there will be an increase in the electrical conductivity
with increasing concentrations of lime and FGD gypsum until the salt concentration
reaches 100% saturation.
values of the lime and FGD gypsum requirement based on their burnt lime
conductivity as a function of incrementally increasing lime and FGD gypsum
) Electrical conductivity of 100 g soil-clay mixture with different lime
(left) and FGD gypsum (right) increments
Thereafter, even with further increasing concentrations, electrical conductivity
will not increase. The value of 50% saturation can be found at the point where
the electrical conductivity is half as high as at 100% saturation. According
to this the amounts required for 50% lime and 50% gypsum saturation were calculated.
Maximum electrical conductivity of the lime solution is 110 μS cm-1,
which corresponds to 0.906 g of lime requirement to obtain 100% lime saturation.
The 50% lime saturation is calculated to be at an electrical conductivity of
55 μS cm-1 and correspond to 0.0453 g of lime requirement. In
case of FGD gypsum, the maximum electrical conductivity is 2315 μS cm-1.
The values of 100 and 50% FGD gypsum saturation for 100 g of sand-clay mixture
are 1.0909 g and 0.3117 g, respectively. Figure 2a and b
shows an overview of electrical conductivity values of sand-clay mixtures with
lime and FGD gypsum increments.
The substrate mixture was prepared with the known concentrations of the lime
and FGD gypsum and with a defined water volume (20%). The mixture is then left
undisturbed for 24 h to release possibly entrapped air and to allow the water
to distribute uniformly within the sample. For the shear test measurements the
sand-clay mixture is thereafter transferred layer by layer using a spoon into
70 cm3 cylinders, with a diameter of 7 cm and a height of 1 cm (Fig.
3a, b). Furthermore, the sand-clay mixture is transferred
into 100 cm3 cylinders for the determination of penetration resistance
and water retention properties having 5.6 cm diameter and 4 cm height (Fig.
3). Two cylinders of same size are joined together with the help of a tape
(cylinder fixing tape). The bottoms of the cylinders are then closed with a
permeable membrane. Below the permeable membrane a perforated plate is used,
which is helpful in transferring samples from the saturation tray to the oven
and vice versa. After the complete preparation of cylinders, the substrate mixtures
are completely saturated in a tray filled with water before starting the drainage
cycles to let any entrapped air out of the sample with which the volume of sand-clay
mixture is ensured to be same in all the repetitions. The samples were prepared
with a defined bulk density of 1.4 g cm-3. After the consecutive
drainage cycles the top cylinder is removed and the soil samples above the lower
cylinder are cut with the help of a knife. Care is taken to prevent any disturbances
in the soil sample while cutting. Soil samples are cut after the respective
drainage cycles to maintain constant sample volume throughout the investigations.
Preparation of laboratory samples for various drying and wetting cycles
using different cylinders for the measurement of penetration resistance
and shear resistance
||Different variants used to evaluate the effect of lime and FGD gypsum
with increasing concentrations at the open mining field nochten
The effect of FGD gypsum and lime on the substrate structural development was
examined over the five different variants (Table 2). In the
case of shear stress measurements, for each variant a measurement under 12 loading
stages (10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150 and 200 kPa) is accomplished.
For the penetration resistance measurements of each variant, 5 replicates, each
with 3 penetration repetitions, are carried out (in total 15 penetrations per
For each variant, samples are treated with four different wetting and drying
cycles (1, 5, 10 and 20 drainage cycles). The drainage cycles were chosen to
intensify the process of the soils aggregation by exposing it to alternating
swelling and shrinkage processes. Shrinkage was performed at temperatures of
33-35°C for 24 h and the following swelling was conducted by saturation
(capillary rise method) during the next 24 h. Water is let into tray drop by
drop to reduce additional forces (surface tension, buoyancsy, dragging, pressure,
etc.) acting on samples when flooded suddenly i.e., rapid wetting produces slaking
by differential swelling, air entrapment and heat of wetting. Soil water suction
has an effect when the soil is being wetted and during the drying phase; whenever
soil water suction is close to zero, effective stress is low and the soil is
unstable (Mullins et al., 1987). Management of
water flow is thus fundamental in stabilization of soils. This special method
is chosen as to reduce sudden disturbances in the sample due to sudden flooding.
After the respective drainage cycles, shear stress and penetration resistance
samples were placed over a suction plate at an matric potential of pF2.5 (316.23
hPa) for 14 days. This value is chosen based on the suggestion given by Klute
(1986), who suggested that the available water retained by soil between
field capacity (pF1.8) and permanent wilting point (pF4.2) is more useful then
field capacity (pF1.8) for measuring relative differences within and amongst
soils. The value pF 2.5 is in the range of plant available water (pF1.8 to 4.2)
and corresponds to typical soil water content during summer.
Shear stress was determined using a box shear device (Fa. Straßentest. Typ 962). The soil samples are vertically loaded with 12 increasing steps (10, 20, 30, 40, 50, 60, 70, 80, 90, 100, 150 and 200 kPa) and having a shear speed of 0.2 mm sec-1. Mohr-coulomb failure lines or shear lines were drawn with the maximum shear stress values obtained.
Penetration resistance was determined using a converted tri-axial press with
a vertical mobile cross beam. A penetrometer-needle with a tip area of 5 mm2
was used. Each sample was penetrated three times in a triangular shape at an
equal distance from the sample centre. Only the penetration resistance values
between 1 and 3 cm sample depths were recorded, as the upper (0 to 1 cm) and
lower (3 to 4 cm) parts of the sample might show edge effects of the cylinder
(Stock and Downes, 2008). Progression speed of the needle
was 5 mm min-1.
The water retention characteristics are determined with four replicates using the drainage method. The matric suction is equilibrated using suction plates in the range from pF1.8 to 2.8, whereas the value pF3.0 and 4.2 were obtained using a pressure chamber. The value pF7.0 was obtained by completely drying the sample at 105°C.
RESULTS AND DISCUSSIONS
The bulk density values increase with all treatments of lime and FGD gypsum
compared to the zero variant (Table 3). All variants show
an increase in bulk density with increasing number of drainage cycles. Higher
bulk density may be due to the high presence of clay (aggregation) in the sand-clay
mixtures. Zhang and Hartge (1995) observed an increase
in the bulk density due to increase in aggregation with wetting and drying cycles.
Similarly Bruand et al. (2001) observed high
bulk density especially in clayey subsoil resulting from a process involving
cycles of shrinking and swelling.
shear parameters, bulk density values and water retention values
The soil structural stability is influenced by the water content. Water content
decreased with increasing number of drainage cycles, which may be due to the
reduction in the water holding capacity of clay by amelioration with lime and
FGD gypsum. Water content reduced from 50% (after one drainage cycle) to 8%
(after 20 drainage cycles). The huge reduction in water content could be due
to the disruption of clay structure with lime and FGD gypsum salts amendment.
Soil cohesion and strength usually increase with a decrease in water content
due to an increased number of particle contact points and capillary forces (Horn
et al., 1994).
This process of increasing bulk density and decreasing water content in the
sand-clay mixtures after respective drainage cycles can be related to the soil
aggregation and water holding capacity of clay. Ball (2001)
compared the water contents of sand, loam and silty clay loam and observed that
the water holding capacity is in the order sand <loam <silty clay loam
and the increase to be linear. The results show a decrease in water holding
capacity of the substrates with respect to the increasing drainage cycles, indicating
disruption of clay structure (substrate structural development).
The water retention characteristics of various variants showed a reduction of drainage with respect to increasing drainage cycles (Table 3). The effectiveness of drainage is dependent on the effectiveness of the swelling and shrinking pressures. The drainage is not so efficient after one cycle as compared to remaining wetting and drying cycles. Drainage over the suction plate increased from 50% (one cycle) to 80% (20 cycles) in all the variants. With an amendment of 100% lime saturated solution and 100% FGD gypsum saturated solution, the water retention capacity of the sand-clay substrate reduced, indicating success in the disruption of clay structure as the clay structure is broken and can no longer hold water as before. The water retention capacity of the substrate with 50% FGD gypsum saturation treatment is less then that of 50% lime saturation treatment.
The shear line approach gave a proportionate increase in shear stress with
increasing applied loads. Shear stress curves of different variants have been
plotted with respect to various variants (Fig. 4a-e).
After the first drainage cycle, higher cohesion values and lower angle of internal
frictions were recorded. This can be due to the fact that samples did not under
go complete swelling and shrinking stresses after first drainage cycle. For
the samples after five, 10 and 20 drainage cycles, a reduction in cohesion and
increase in angle of internal friction is observed (Table 3).
Grant and Blackmore (1991) also observed a reduction
in cohesion and increase in angle of internal friction with increasing drainage
cycles and stated that this reduction in cohesion and increase in angle of internal
friction are due to the disruption of clay structure i.e., new structural development
of clay. The shear resistance curves indicated a formation of new structure
development after 10 drainage cycles with 100% FGD gypsum saturation and 100%
lime saturation variants.
Narra (2008) observed high cohesion constants (>100
kPa) in Lusatia, Germany with clay contents ranging from 20 to 40%. Results
obtained in this study also showed high cohesion constants reaching a value
higher then 100 kPa especially with the 50% FGD gypsum variant (Table
3). However, the cohesion constants with other variants were comparatively
smaller. Zhang and Hartge (1995) found the cohesion values
to be very low (almost zero) in Pleistocene sand, which could be one of the
reasons in obtaining lower cohesion constants compared to Narra
(2008). With smaller cohesion values the erosion stability becomes smaller
(Horn and Rostek, 2000). The cohesion values and the
angles of internal friction increase with increasing number of drainage cycles,
as also observed by Zhang and Hartge (1995).
Shear lines of different variants with respect to different drainage cycles.
Dark black dashed line with arrow marks represents structural shear line
Reduction in angle of internal friction is observed with the amendment of 100%
FGD gypsum saturation and 100% lime saturation compared to the zero variant.
The angle of internal friction for the 100% lime saturated variant is very small
indicating a lubricated effect. The smaller angle of internal friction indicates
higher cohesion presence. The higher the cohesion, the higher is the lubrication
i.e., the substrate mixture behaves more as a gel. In contrast, for the 100%
FGD gypsum saturation variant, the angle of internal friction is higher compared
to the 100% lime variant indicating new structure formation with good crystal
development (Badens et al., 1999). With the good
crystal development, the substrate particles no longer behave in the form of
a gel, instead they behave independently. The black dashed line with arrows
represents the structural shear line of each variant (Fig. 4).
The structural shear line is drawn using the lowest and highest shear values
obtained in all measurements after respective drainage cycles (Lebert
and Horn, 1991). For the zero variant the structural line is developed after
20 drainage cycles, whereas with 100% FGD gypsum saturation and 100% lime saturation
amendments the structural shear line is developed already after 10 drainage
of the shear lines of different variants
A structural shear line after 10 drainage cycles indicates development of structure
earlier compared to the zero variant.
The respective shear lines of the zero variant, 100% FGD gypsum variant and
100% lime variant are drawn together in Fig. 5. Figure
5 shows a clear picture of the cohesion constants and the angle of internal
frictions and their influence on the structural development. The background
marked with grey colour represents the optimum range for the shear lines. When
the shear lines are above the optimum range i.e., with higher angle of internal
friction the substrate would act as hard-set soil especially in the presence
of clay. The shear line obtained with the zero variant showed a tendency towards
hard setting of the clay. When the shear lines have a lower angle of inclination
then the substrate acts as a cohesive soil. The shear line obtained with the
100% lime variant tends to act more as a cohesive soil. The 100% FGD gypsum
variant has an optimum cohesion and angle of inclination indicating a good crystal
development i.e. good structural development of clay. The structural development
is well achieved with the 100% FGD gypsum variant compared to the 100% lime
variant. This is noted based on the shear lines and by their angle of internal
friction and cohesion constants as stated by Wagner et
The penetration resistance is directly related to the soils resistance
against root growth. The higher the penetration resistance, smaller is the root
growth and is subsequently hindered (Clark and Baligar,
2003). The penetration resistance is also related to the packing of particles.
Roots create compressive and shear stresses which can reach up to 2 MPa (Goss,
1991). The radial pressure exerted by growing roots compresses the soil
in their vicinity (Dexter, 1987) and decreases the porosity
in that zone (Guidi et al., 1985).
For penetration resistance values almost 85% of amelioration is achieved with
100% FGD gypsum whereas with 100% lime only 76% is achieved. This indicates
a high percentage of amelioration achievement with FGD gypsum usage as an amendment.
With amendments of 50% lime and 50% FGD gypsum, amelioration success achieved
is very less (only about 31 and 40%, respectively). Narra
(2008) showed that there is an increase in penetration resistance for substrates
of different glaciation ages from lower Lusatia, to be 10 times higher than
the rooting limit of 4.2 MPa as described by Materetschera
et al. (1991). The penetration resistance values obtained are seven-fold
higher for the zero variant, whereas with 100% FGD gypsum saturation treatment
the values are only slightly higher than 4.2 MPa (Fig. 6)
and were close to the value described by Materetschera et
resistance as a function of the number of accomplished drainage cycles
at a suction of pF 2.5
Materetschera et al. (1994) suggested that a
higher proportion of small aggregates in cultivated than in uncultivated soils
are resulting from the breakdown of bigger aggregates by the penetration of
roots. For re-cultivation locations with certain clay contents, the values of
shear stress and penetration resistance increase with increasing drainage. This
may be the reason for a remarkable increase in shear stress and penetration
resistance with increasing number of drainage cycles. An increase in penetration
resistance may also be due to the fact that water must be displaced as the needle
is moving into the soil, especially when clay is present (Perumpral,
1987). The amelioration achieved with respect to shear stress parameters
can be related to structural formation whereas the amelioration achieved with
respect to the penetration resistance can be related to roots penetration (Arshad
et al., 1993).
There is an increase in bulk density values with increasing drainage cycles, which can be related to the function of clay with the reduction in water content. There is a reduction in drainage with an increase in wetting and drying cycles, which is observed with the help of water retention curves. In case of zero variant, the water retention curve obtained is almost a straight line (i.e., even though there is presence of water, it cannot be drained). This could be due to the water holding capacity of clay. With the amendment of 100% lime and 100% FGD gypsum drainage is observed even after 20 wetting and drying cycles. Drainage observed is more with 100% FGD gypsum compared to that of 100% lime indicating amelioration success with the use of FGD gypsum to be more efficient.
Amelioration with different amendments is well achieved in the case of 100% FGD gypsum in comparison with 100% lime. The values of cohesion and angles of internal frictions of the shear lines are smaller. Cohesion constants are almost same in case of zero variant, 100% FGD gypsum and 100% lime variants but the angle of internal friction is higher in case of 100% FGD gypsum variant in comparison to 100% lime variant. With increase in drainage cycles cohesion and angle of internal friction increased along with the shear stress. With cohesion values being less FGD gypsum or lime addition can not prevent erosion. This increase in shear stress could be due to the increased aggregation with drainage cycles.
Clark and Baligar (2003) states an increase in penetration
resistance reduces the root growth. Observations show very high penetration
resistance with zero variant. With the amendment of 100% lime and 100% FGD gypsum,
the penetration resistance reduced. 100% FGD gypsum is comparatively a better
amendment as compared with different variants. With 100% FGD gypsum the penetration
resistance values are around 5 MPa, even though these values are a bit higher
than the limiting penetration value as mentioned by Materetschera
et al. (1991). The penetration resistance values reduced 6 fold lesser
with the amendment of FGD gypsum as compared to no amendment.
Results of this study show that the substitution of FGD gypsum for lime is
more efficient under soil structural amelioration and is also economic. Sub-soiling
enhanced penetration and ameliorative action of surface applied calcium (Ca)
salts. Similarly Bocskai (1974) found that sub-soiling
in combination with a surface application of lime and a subsurface introduction
of FGD gypsum improved crop yield more than sub-soiling alone. Arshad
et al. (1993) observed that with amendment of FGD gypsum, amelioration
has increased significantly on wheat yield. Sustainable alternative FGD gypsum
with the help of parameters considered concludes a very good structure development
of clay and also on the development of roots penetration in the course of re-cultivation.