Bioremediation of polluted soil through the combined application of plants,
earthworms and organic matter
Cristina Macci,*Serena Doni, Eleonora Peruzzi, Brunello Ceccanti and Grazia Masciandaro
Received 6th June 2012, Accepted 27th July 2012
Two plant species (Paulownia tomentosa and Cytisus scoparius), earthworms (Eisenia fetida), and
organic matter (horse manure) were used as an ecological approach to bioremediate a soil historically
contaminated by heavy metals and hydrocarbons. The experiment was carried out for six months at a
mesoscale level using pots containing 90 kg of polluted soil. Three different treatments were performed
for each plant: (i) untreated planted soil as a control (C); (ii) planted soil + horse manure (20 : 1 w/w)
(M); (iii) planted soil + horse manure + 15 earthworms (ME). Both the plant species were able to grow
in the polluted soil and to improve the soil’s bio-chemical conditions, especially when organic matter
and earthworms were applied. By comparing the two plant species, few signiﬁcant differences were
observed in the soil characteristics; Cytisus scoparius improved soil nutrient content more than
Paulownia tomentosa, which instead stimulated more soil microbial metabolism. Regarding the
pollutants, Paulownia tomentosa was more efﬁcient in reducing the heavy metal (Pb, Cr, Cd, Zn, Cu,
Ni) content, while earthworms were particularly able to stimulate the processes involved in the
decontamination of organic pollutants (hydrocarbons). This ecological approach, validated at a
mesoscale level, has recently been transferred to a real scale situation to carry out the bioremediation of
polluted soil in San Giuliano Terme Municipality (Pisa, Italy).
Different hydrocarbons and heavy metals are becoming
increasingly prevalent in the soil environment as a result of
anthropogenic and natural activities.
A wide range of physical
and chemical techniques have been developed for the remedia-
tion of contaminated soils. However, these treatments irrevers-
ibly affect soil properties, reduce biodiversity and are often costly
Bioremediation (the use of living organisms to remediate
contaminated soils) seems to be an attractive and promising
approach for addressing both organic and inorganic decontam-
ination and ensuring the conservation of the ecosystem’s
biophysical properties. Moreover, the biological technology is a
cost-effective, non-invasive, and socially acceptable way to
address the removal of environmental contaminants.
The effectiveness of plant and organic matter addition for the
remediation of heavy metal and hydrocarbon polluted soil has
been widely demonstrated in the last few decades.
is a source of organic matter able to add microorganisms and
stimulate soil microorganisms through the presence of organic
and mineral substrates.
In recent years a number of encouraging studies have high-
lighted the potential of earthworms to support soil bioremedia-
The hypothesis is that the synergic action of plants,
microorganisms and earthworms to enhance hydrocarbon
degradation and heavy metal removal is particularly effective.
This novel triple approach has been only recently taken into
account as an effective remediation tool.
In this system (plants–microorganisms–earthworms) plants
act directly by chelating and uptaking metals from the solid
matrix into the roots or shoots, and indirectly, through the
Institute of Ecosystem Study (ISE), CNR, Via Moruzzi, 1, 56124, Pisa,
Italy. E-mail: firstname.lastname@example.org; Fax: +39 050352473; Tel: +39
The applicability of a triple-ecological approach (plants, organic matter and earthworms) has been validated to bioremediate and
functionally recover a soil historically contaminated by heavy metals and hydrocarbons. Moreover, for the ﬁrst time the biore-
mediation potential of Cytisus scoparius has been investigated. On the basis of the results obtained in this study, this ecological
approach has been transferred to a real scale situation.
2710 | J. Environ. Monit., 2012, 14, 2710–2717 This journal is ªThe Royal Society of Chemistry 2012
Dynamic Article LinksC
Cite this: J. Environ. Monit., 2012, 14, 2710
production of root exudates, stimulating the microbial commu-
nity to degrade contaminants through their enzymes.
Microorganisms in the rhizosphere are directly responsible for
the degradation of organic compounds, such as hydrocarbons,
while indirectly, they improve the release of metals from organ-
ically bound matrices.
The selection of vegetal species is particularly important since
they must be able to live and grow in the contaminated soil.
Paulownia tomentosa has received increased attention due to
its marketable value for wood and biofuel production, thanks to
its rapid growth, high biomass production, and elevated stress
tolerance, and it is currently used in Italy for these purposes. In
addition, it has shown strong transpiration rates and elevated
tolerance to high concentrations of metals in both hydroponic
and ﬁeld studies
and it has been already successfully used in
previous bioremediation studies.
Cytisus scoparius has not been investigated yet for its metal
accumulation potential, however, an improvement in soil
nutrient content (C, N and P) due to its ability to ﬁx nitrogen and
by deposition of the leaves has been previously observed.
Finally, earthworms stimulate the microbial activities as they
mix and homogenize soil, increase permeability, aeration, water
retention capacity and root elongation by means of the channels
they dig. In addition, they ingest soil and expel a partially
stabilized product (casting); in this way they ensure the avail-
ability of organic substrates for proliferation of the autochtho-
nous microorganisms in the soil, thus increasing microbiological
and biochemical soil activity.
Moreover, various studies have
demonstrated that earthworm activity increases the mobility and
bioavailability of metals in soil
and they are able to absorb
metals into their tissues without compromising their metabolic
In recent years, Eisenia fetida species, due to its
good adaptability, high mobility and capacity to metabolize
different organic substrates, has been experimented upon in
several bioremediation studies.
A higher microbial activity
and an increase in organic compound degradation have been
observed in these studies.
In the present investigation, carried out at a mesoscale level,
two different plant species (a tree: Paulownia tomentosa and a
bush: Cytisus scoparius), Eisenia fetida earthworms, and horse
manure (organic matter) were used to evaluate the effectiveness
of this new triple-ecological approach to bioremediate and
functionally recover a soil historically contaminated by metals
2 Materials and methods
The soil used for the bioremediation process in the experimental
pots was taken from an industrial area (AREA ‘‘EX-ECO-
SIDER’’, San Giuliano Terme Municipality, Pisa, Italy). The
main characteristics of the soil and horse manure are reported in
Paulownia tomentosa and Cytisus scoparius, common species in
the Mediterranean area, were selected for their rapid growth and
capability of developing in the speciﬁc substrate and under the
local climatic conditions. The plants used in the experiment were
approximately 1 year old.
2.2 Experimental layout
The experiment was carried out at a mesoscale level using pots
(about 40 cm i.d. 55 cm h) containing 90 kg of the soil polluted
by metals and hydrocarbons. Each pot was planted with
Paulownia tomentosa or Cytisus scoparius.
Three different treatments with three replicates were
(i) untreated planted soil as a control (C);
(ii) planted soil + horse manure (20 : 1 w/w) (M);
(iii) planted soil + horse manure + 15 earthworms (ME).
The manure was surface broadcast and manually incorporated
to a 25 cm depth. The main characteristics of the manure are
reported in Table 1. The earthworms (Eisenia fetida) were added
together with the manure.
The pots (18 pots: 2 plants 3 treatments 3 replicates) were
kept for 6 months under open ﬁeld conditions exposed to the
same climate as the polluted site.
After six months from the beginning of the experiment, soil
samples and aerial parts of plants were collected. From each pot,
10 soil subsamples were taken by a soil core sampler (1.5 cm
diameter, 30 cm length) and combined. Soil samples were air-
dried, sieved (2 mm), and stored at room temperature until
chemical and biochemical analyses.
The leaves of Paulownia tomentosa and shoots of Cytisus sco-
parius were analysed for metal content after they were washed with
tap water, rinsed with deionised water and dried at 70 C until a
constant weight was obtained. In order to evaluate the plant metal
extraction, plants grown on an uncontaminated agricultural soil
located near to the contaminated site were used as a control.
2.3.1 Chemical analysis. Electrical conductivity (EC) and pH
were measured in soil : water (1 : 10) extracts (w/v).
Total organic C (TOC) and N (TN) were determined by dry
combustion with a RC-412 multiphase carbon and a FP-528
protein/nitrogen determinator, respectively (LECO corporation).
Total-extractable carbon (TEC) was extracted at 60 C for 4 h
under shaking, using Na
(0.1 M, pH 11) as an extractant in
a 1 : 10 solid : liquid ratio. From the total extractable carbon, the
humic acids (HA) were precipitated by adjusting the pH to 2 with
; the extracts were kept at 4 C for 12 h, then the precip-
itate (HA) and the supernatant were separated by centrifugation.
The fulvic acids (FA) were determined in the supernatant, while
the humic acids were obtained from the difference between TEC
Water soluble carbon (WSC) was measured in an aqueous
extract (1 : 10 w/v). WSC, TEC and FA were determined by acid
digestion with K
at 148 C for 2 h. A spec-
trophotometric method was used to quantify the Cr
by the reduction of Cr
measured in an aqueous extract (1 : 10 w/v) with an ammonia-
selective electrode (ORION 95-12), and by a DIONEX ionic
chromatograph (DIONEX 2000i, California, USA), respectively.
Total and available phosphorus were determined by the
after acid digestion with nitric–perchloric
acids and extraction with ammonium acetate–diethylene-
triaminepentaacetic acid (DTPA), respectively.
This journal is ªThe Royal Society of Chemistry 2012 J. Environ. Monit., 2012, 14, 2710–2717 | 2711
The metal concentration analysis was performed by atomic
absorption spectrometry (Perkin-Elmer 3030) after nitric–
perchloric acid digestion for total forms and ammonium acetate–
diethylenetriaminepentaacetic acid (DTPA) for available forms.
Total hydrocarbons were determined by the gravimetric
method (method 1664)
using n-pentane instead of n-hexane, as
modiﬁed by Ceccanti et al.
The soil samples (3 g) were air-dried
and mixed with Na
to remove residual water. Total
hydrocarbons were extracted three times with 9 ml of pentane in
an ultrasound bath for 20 min. The total hydrocarbon content is
estimated by weighing the dry residue after solvent evaporation
under nitrogen ﬂow.
2.3.2 Biochemical assay. To test b-glucosidase activity 0.05 M
4-nitro-phenyl-b-D-glucanopyranoside (PNG) was used as a
substrate, while 0.115 M p-nitrophenyl phosphate (PNPP) was
used as a substrate to measure the phosphatase activity. The
p-nitrophenol (PNP) produced by both hydrolases was extracted
and determined spectrophotometrically at 398 nm.
genase activity was measured using 0.4% 2-p-iodophenyl-3-p-
nitrophenyl-5-tetrazolium chloride (INT) as a substrate;
iodonitrotetrazolium formazan (INTF) produced in the reduc-
tion of INT was measured by means of a spectrophotometer at
2.3.3 Phyto-test. A phyto-test with Lepidium sativum was
carried out following the method reported by Hoekstra et al.
using 2 ml of soil extracts (1 : 10 w/v) and 10 seeds.
The growthindex, calculated after72 hours, was expressedby the
following formula: GI% ¼P(T/C), where Pis the mean percentage
of seed germination, with respect to the mean value of the control
prepared with distilled water, assumed to be 100%; Tand Care the
length of shoot in the treatment and in the control, respectively.
2.3.4 Statistical analysis. All results reported in the text are
the means of determinations made on three replicates. Differ-
ences between treatments (C, M and ME) and plant species
(Paulownia tomentosa and Cytisus scoparius) were tested by the
analysis of variance (ANOVA). The means were compared
using HSD Tukey’s test. The differences between all treatments
and soil (S) and soil treated with horse manure (SHM) were
tested using Dunnett’s comparison test. A correlation matrix of
the data was also calculated in order to determine the rela-
tionship between the parameters (data not reported). The
signiﬁcant levels reported (p< 0.05) are based on the Student’s
distribution. Principal component analysis (PCA) was applied
to all results obtained during the experiments. The PCA is a
multivariate statistical data analysis technique, which reduces a
set of raw data to a number of principal components that retain
the most variance within the original data, in order to identify
possible patterns or clusters between objects and variables.
statistical evaluations were calculated with Statistica 7.0
3. Results and discussion
3.1 Chemical and biochemical processes involved in the
The plant species were able to grow in the polluted soil, and they
were visibly more developed in the treatments with manure (M)
and manure with earthworms (ME), probably because of the
higher availability of nutrients added with the horse manure.
In the presence of each plant TN, TOC, TEC and its fractions
(HA and FA) gave higher values in M and ME treatments due to
the addition of organic matter (Table 1), which persist after six
months of treatment. As expected, no signiﬁcant differences were
Table 1 Chemical and biochemical parameters in the starting materials: soil (S), horse manure (HM), soil with horse manure (SHM) (mean of three
replicates standard deviation) and in the treatments: soil as control (C); soil + horse manure (M); soil + manure + 15 earthworms (ME) with Paulownia
tomentosa and Cytisus scoparius after six months of the mesoscale experiment
Materials Paulownia tomentosa Cytisus scoparius
S HM SHM C M ME C M ME
pH 7.90 0.12 6.92 0.20 7.89 0.11 9.04 ns 9.06 ns 9.30 ns 9.18 ns 9.33 ns 9.30 ns
EC (dS m
) 0.13 0.03 3.69 0.17 0.17 0.02 0.11aA 0.18bA 0.14aA 0.32aB 0.32aB 0.33aB
TN (%) 0.18 0.01 1.49 0.06 0.197 0.01 0.18aA 0.21bA 0.22bA 0.20aA 0.23bA 0.23bA
TOC (%) 1.45 0.07 45.1 1.5 1.98 0.11 1.34aA 2.07bA 2.20bA 1.68aA 2.15bA 2.24bA
C/N 8.06 30.3 10.1 7.44 9.86 10.0 8.40 9.35 9.74
) 2.30 0.20 118 6.39 4.93 0.31 2.39bA 1.44aA
5.41 cA 2.23aA 2.83abB
) 11.3 0.9 394 22.8 15.1 0.8 0.10aA
13.9 cA 4.15aB
WSC (mg kg
)915 36200 456 133 3 285aA
TEC (mg kg
) 2417 112 59361 988 3570 108 2193aA 3709bA 3767bA 3286aB
FA (mg kg
) 1178 77 33898 665 1609 87 1361aA 1795bA 1752bA 1914aB
HA (mg kg
) 1239 44 25463 2347 1962 119 832aA
1914bA 2015bA 1372aB 2087bA 2036bA
Pav (mg kg
) 0.06 0.00 46.4 1.5 0.33 0.01 0.71bA
TP (mg kg
) 216 13 394 33 241 17 226 ns 235 ns 266 ns 239 ns 260 ns 258 ns
DHase (mg INTF per g per h) 0.67 0.04 64.2 2.1 2.04 0.4 0.88aA
b-Gluc (mg PNP per g per h) 14.3 0.7 1410 65 22.1 1.2 32.8aA
Phosph (mg PNP per g per h) 81.7 4.9 2823 111 106 7 123aA
Different lowercase letters indicate statistically different values (treatment effect) within plants according to HSD Tukey’s test (p< 0.05). Different
uppercase letters indicate statistically different values (plant effect) within treatments according to HSD Tukey’s test (p< 0.05).
Dunnett’s test (p<
0.05) with respect to S for control treatment (C) and with respect to SHM for organic matter treatments (M and ME).
EC, electrical conductivity;
TN, total nitrogen; TOC, total organic carbon; WSC, water-soluble carbon; TEC; total extractable carbon; HA, humic acid; FA, fulvic acid; Pav,
available phosphorus; TP, total phosphorus; DHase, dehydrogenase; b-gluc, b-glucosidase; Phosph, phosphatase.
2712 | J. Environ. Monit., 2012, 14, 2710–2717 This journal is ªThe Royal Society of Chemistry 2012
observed for the total form of N and all the fractions of C
between these treatments (M and ME).
The signiﬁcantly higher WSC content found in the M and
ME treatments with respect to the C treatments was probably
due to the fresh organic material added, but also to the
beginning of the mineralization of the organic carbon added
with horse manure (Table 1). WSC represents the most labile
fraction of the soil organic matter because it is susceptible to
this fraction can be used as carbon and
energy sources by soil microﬂora, and can be positively related
to microbial activity.
A greater positive correlation was, in
fact, found between WSC and DH-ase activity (r¼0.92; p<
0.01), which has been considered to be an indicator of the entire
microbial metabolism as it occurs only within living cells.
higher values of DH-ase observed in M and ME treatments
with respect to the controls (Table 1) were surely due to the
stimulation of autochthonous microbial biomass,
and to the
direct addition of enzymes and microorganisms with organic
Moreover, it is well known that the earth-
worms, through the release of castings and the continuous
mixing of the soil, improve the conditions for the increase of
total microbial biomass number and activity.
higher DH-ase activity was, in fact, found in the presence of
As a consequence of the application of organic matter, a
higher b-glucosidase activity, an enzyme involved in cellulose
was observed in the M and ME treatments (Table
1). However, the higher b-glucosidase activity in each C treat-
ment with respect to the originally polluted soil (S) (Table 1),
suggested that also the plants, through the release of root
exudates, induced the synthesis of this enzyme. As expected,
b-glucosidase activity was positively correlated with DH-ase
activity (r¼0.89, p< 0.01) and WSC (r¼0.96, p< 0.01).
Also the phosphatase activities, enzymes linked to the phos-
phorus cycle, increased in all treatments with respect to the
original soil (S) (Table 1), indicating their release by plants.
particular, the higher activity detected in the ME treatments was
probably due to enzyme production in the worm gut and
excretion through cast deposition.
Several other researchers have found that soil hydrolase
activities are enhanced by the addition of organic materials and
The earthworm activity resulted
particularly evident in the stimulation of N and P cycles, as
shown by the higher available P, ammonium, and nitrate in the
ME treatments (Table 1).
The higher content of nitrate in the ME treatments was
probably due to the capacity of earthworms to favour aerobic
nitriﬁcation processes by their movement, as already found in
Instead, in a study concerning the application
of organic amendment and earthworms in soil, the increase in
available P was attributed to the changes in sorption complexes
induced by the competition for sorbing sites between ortho-
phosphates and carboxyl groups of a glycoprotein, such as the
mucus produced in the gut.
Regarding the use of different plant species, few signiﬁcant
differences were observed in the chemical characteristics of the
soil. With the exception of nitrate and available P, which were
particularly higher in the presence of Cytisus scoparius,an
improvement in the soil nutrient content, due to the ability of
Cytisus scoparius to ﬁx nitrogen and by deposition of the leaves,
has been previously observed.
In contrast, the enzyme
activities, which are very sensitive even to small changes occur-
ring in the soil,
were generally signiﬁcantly higher in the
presence of Paulownia tomentosa than in the presence of Cytisus
scoparius (especially in ME treatment). This result suggested the
formation of a more suitable environment for microorganisms in
carrying out their activity, probably due to the more developed
root systems of Paulownia tomentosa. However, the higher
phosphatase activity found in C and M treatments in the pres-
ence of Cytisus scoparius suggested a greater stimulation of the P
cycle, as conﬁrmed by the amount of higher P available (product
of this enzymatic activity).
The efﬁciency of the remediation process was evaluated
through the assays of organic (TPH) and inorganic pollutants
(total metals) and a phyto-test. The greatest reduction of TPH
(54–75%) was observed in the treatments with earthworms (ME),
even if a signiﬁcant reduction was also observed in M treatments
(Fig. 1). It is known that the humic matter is able to bind
hydrocarbons and enhance their degradation,
adaptation time of microorganisms involved in the metabolism
The further decrease in the polluted organic substrates in the
presence of earthworms suggests a direct degradation of these
compounds by earthworms, and/or an indirect effect through the
stimulation of soil microbial metabolism by the available
substrate from the organic matter and earthworm castings.
higher metabolic activity (DH-ase, b-glucosidase and phospha-
tase activities) was, in fact, detected in the treatments with
The enhancement in bioremediation performance of the
earthworm treatments could be also due to the worm activity in
promoting the release of contaminants entrapped and absorbed
into soil minerals and organic matter fractions. This would make
them more accessible, and thus available for microbial interac-
These results are in agreement with those of Bianchi and
which in a column experiment with the same polluted
Fig. 1 Total hydrocarbon concentration in the soil before (S) and after
treatments: C (control soil); M (manure), ME (manure + earthworms)
with Paulownia tomentosa and Cytisus scoparius. Vertical bars indicate
standard deviation (mean of three replicates). Different lowercase letters
indicate statistically different values (treatment effect) within the plants
according to HSD Tukey’s test (p< 0.05). Different uppercase letters
indicate statistically different values (plant effect) within the treatments
according to HSD Tukey’s test (p< 0.05). *Dunnett’s test (p< 0.05) with
respect to S for control treatment (C) and with respect to SHM for
organic matter treatments (M and ME).
This journal is ªThe Royal Society of Chemistry 2012 J. Environ. Monit., 2012, 14, 2710–2717 | 2713
soil found a greater reduction of hydrocarbons in the treatment
with earthworms and organic matter, with respect to the treat-
ment with only organic matter.
By comparing the plants, Paulownia tomentosa proved to be
more efﬁcient in removing the organic contaminants, showing
the lowest TPH content in each treatment. In these soil–plant
systems a higher microbial metabolism (higher DH-ase activity)
and b-glucosidase activity, which represents the initial, rate-
limiting step in the decomposition of complex organic
were, in fact, observed. In only six months, the
TPH concentration in the ME treatment reaches values very
close to the limit for industrial use (Table 2).
As far as inorganic pollutants are concerned, the metal content
signiﬁcantly decreased in each treatment, showing higher
reduction in ME treatments (Fig. 2). Different studies have
demonstrated that earthworm activity increases the mobility and
bioavailability of metals in soils.
Probably, as suggested by
Bianchi and Ceccanti
the earthworms, involved at the rhizo-
sphere level as they feed on the microbial biomass of roots, ingest
soil and release metals and nutrients available for plant uptake.
This hypothesis seems to be conﬁrmed by the results of metals
measured in the aerial part of the plants, which were generally
signiﬁcantly higher (p< 0.05) in the ME treatments (Fig. 3). The
higher reduction of metals in the ME treatments could be also
due to the fact that the earthworms are able to absorb a great
quantity of them into their tissues without compromising their
experimental data concerning the
amount of metals absorbed by earthworms have not been
investigated. The lower reduction in metals in the M treatments
with respect to the C treatments (Fig. 2) suggested an immobi-
lization process of metals by organic amendment through cation
exchange, sorption, complexation and precipitation, as other
papers have reported.
As a consequence of metal immobili-
zation, a reduction in the availability of metals for plant uptake
as conﬁrmed by the generally lowest content (even
though not always signiﬁcant) of metals in plant tissues in the M
However, the great reduction in all total and available metals
observed in each treatment (Fig. 4), together with the higher
concentration of the same in plant tissues, conﬁrmed the phy-
toremediation ability of Paulownia tomentosa
and for the ﬁrst
time showed the capability of Cytisus scoparius to extract metals
such as Zi, Cu and Cr (Fig. 2–4). This result was in accordance
with the characteristics of these metals; Zn and Cu, in fact, are
essential micronutrients for plants, while Cd and Pb are known
to be toxic for several vegetal species.
However, it is important to remember that the metal concen-
tration in the aerial part of plants only gives an indication as to
the ability of plants to extract metals, since many plants do not
translocate the metals in the aerial part, but they maintain them
at the rhizosphere level. Anyway, Paulownia tomentosa seemed to
be more efﬁcient in metal extraction with respect to Cytisus
scoparius for all the metals investigated (Fig. 2–4). In the ME
treatment, in the presence of Paulownia tomentosa, a reduction in
soil metal content greater than 50% was observed for each metal
(Zn 69%, Pb 76%, Ni 59%, Cu 80%, Cr 51%, Cd 68%), reaching
for Ni and Cu values lower than the Italian regulation for civil-
reuse of soil (D.lgs. 152/2006, Table 2). Instead, in the presence of
Cytisus scoparius in the ME treatment, only Cu, Cd and Ni
reached a percentage of reduction of about 50%, while for Pb, Cr
and Zn a low reduction of 12%, 20% and 36%, respectively, was
observed. At the end of the experiment, the recovery of the
polluted soil, in terms of the elimination of the contaminant, was
also assessed by carrying out a phyto-test with Lepidium sativum,
selected for its rapid growth and high sensitivity to toxic
The polluted soil (S) showed a germination index
of 61% (Fig. 5); this value can be considered not to be lethal, but
dangerous for plants.
This index increased to 73% with the only
mixture with manure, which supplies nutritive compounds. At
the end of the experiment the GI% reached a value greater than
90% in all the treatments. A higher GI% was observed in the M
and ME treatments (Fig. 5), due to the higher nutritional content
with respect to C soils. Moreover, in these treatments, a higher
Table 2 Total heavy metals and TPH Italian admissible standard limits
for re-use of the cleaned soil after bioremediation (D.lgs 152/06). The
Italian legislation regulates the light (C #12) and heavy (C > 12)
hydrocarbons; in this study TPH were compared with C > 12
) Zn Cu Pb Cd Ni Cr TPH C #12 C > 12
Residential use 150 120 100 2 120 150 — 10 50
Industrial use 1500 600 1000 15 500 800 — 250 750
Fig. 2 (A and B) Total metal concentration in the soil before (S) and
after treatments: C (control soil); M (manure), ME (manure + earth-
worms) with Paulownia tomentosa and Cytisus scoparius. The concen-
tration of Cd was multiplied by 10. Vertical bars indicate standard
deviation (mean of three replicates). Different lowercase letters indicate
statistically different values (treatment effect) within plants according to
HSD Tukey’s test (p< 0.05). Different uppercase letters indicate statis-
tically different values (plant effect) within treatments according to HSD
Tukey’s test (p< 0.05). *Dunnett’s test (p< 0.05) with respect to S.
2714 | J. Environ. Monit., 2012, 14, 2710–2717 This journal is ªThe Royal Society of Chemistry 2012
reduction in TPH was also detected, indicating an inhibition of
organic pollution in Lepidium sativum germination, as already
The stimulation of plant germination and growth indicated
that plants, together with microorganisms and earthworm
activities, were able not only to bioremediate the soil but also to
improve the soil’s agronomic properties.
Despite the short term experimentation, a complete decon-
tamination of the soil from both organic and inorganic pollut-
ants was obtained from the treatment with organic matter and
earthworms (ME). On the basis of the results obtained from the
treatment with manure (M), an estimation of contaminant
reduction under the Italian regulation for civil-reuse of soil
(D.lgs. 152/2006, Table 2) will be expected in 36 and 48 months
for Paulownia tomentosa and Cytisus scoparius, respectively.
Comparing this bioremediation technology to the commercially
available physical and chemical ones, even though the latter
technologies are more rapid in decontamination they are usually
disruptive to the environment.
3.2 Statistical analysis
To better understand the complexity of chemical and biochem-
ical processes involved in the soil bioremediation, principal
component analysis (PCA) was carried out. PCA is a multivar-
iate statistical data analysis technique that reduces a set of raw
data to a number of principal components that retain the most
variance within the original data in order to identify possible
patterns or clusters between objects and variables.
In order to
minimize the number of variables, each metal was transformed
from mg kg
to meq. kg
and then the contents for total metal
and available metal were summed. Moreover, the ratio C/N and
TEC, being highly correlated with TOC, TN, humic and fulvic
acids, were not considered for the statistical analysis. The PCA of
the dataset indicated 76.9% of the data variance as being con-
tained in the ﬁrst two components (Table 3). The 1
(39.0% of the total variance) included all the nutrients (TOC,
TN, TP, FA, HA), N–NO
, GI and EC. These parameters can
be considered related to the effect of organic matter addition in
the M and ME treatments.
Fig. 3 (A and B) Total metal concentration in the aerial part of
Paulownia tomentosa and Cytisus scoparius at the end of the experiment
in the three treatments: C (control soil); M (manure), ME (manure +
earthworms) and in plants grown in an agronomic soil (A). Vertical bars
indicate standard deviation (mean of three replicates). Different lower-
case letters indicate statistically different values (treatment effect) within
the plants according to HSD Tukey’s test (p< 0.05). Different uppercase
letters indicate statistically different values (plant effect) within the
treatments according to HSD Tukey’s test (p< 0.05). *Dunnett’s test (p<
0.05) with respect to agronomic soil (A).
Fig. 4 Available metal concentration in soil before (S) and after treat-
ments: C (control soil); M (manure), ME (manure + earthworms) with
Paulownia tomentosa and Cytisus scoparius. The concentrations of Ni and
Cd were multiplied by 10. Vertical bars indicate standard deviation (mean
of three replicates). Different lowercase letters indicate statistically
different values (treatment effect) within the plants according to HSD
Tukey’s test (p< 0.05). Different uppercase letters indicate statistically
different values (plant effect) within the treatments according to HSD
Tukey’s test (p< 0.05). *Dunnett’s test (p< 0.05) with respect to S.
Fig. 5 Germination index% determined in soil before (S, soil; SHM, soil
with horse manure) and after treatments: C (control soil); M (manure),
ME (manure + earthworms) with Paulownia tomentosa and Cytisus
scoparius. Vertical bars indicate standard deviation (mean of three
replicates). Different lowercase letters indicate statistically different
values (treatment effect) within the plants according to HSD Tukey’s test
(p< 0.05). Different uppercase letters indicate statistically different values
(plant effect) within the treatments according to HSD Tukey’s test (p<
0.05). *Dunnett’s test (p< 0.05) with respect to S for control treatment
(C) and with respect to SHM for organic matter treatments (M and ME).
This journal is ªThe Royal Society of Chemistry 2012 J. Environ. Monit., 2012, 14, 2710–2717 | 2715
PC loading (36.9% of the total variance) included all
the contaminants (total and available metals, TPH), the enzy-
matic activities (phosphatase, b-glucosidase and dehydrogenase),
WSC, Pav and N–NH
. These parameters can be considered
closely associated with the soil remediation since the enzyme
activities increased when the soil pollution decreased.
In order to correlate the parameters with the treatments, the
scores and loading graphs were combined (Fig. 6). All the
treatments were plotted at the top, while the organic and mineral
contaminants were plotted at the bottom, together with the
originally polluted soil (S and SHM). The addition of organic
matter at the beginning of the experiment greatly modiﬁed the
initial material (S), as shown by the shift on the right (PC1 axis)
of SHM. The right shift (PC1 axis) of the M and ME treatments
with respect to the C soils indicated a modiﬁcation of the
parameters linked to organic matter turnover. Moreover, the ME
treatments were in the upper right quadrant of the plot, far from
the initial soil (S and SHM) and organic and inorganic
contaminants and linked with enzymatic activities and germi-
nation index. The greater efﬁciency of the ME treatments in
improving soil biochemical fertility and in decreasing soil
pollution were conﬁrmed by this change of position on the
The bioremediation treatments carried out at a mesoscale level
were effective in the reclamation of the hydrocarbon and metal
polluted soil. A reduction of hydrocarbons and metals was, in
fact, observed in each treatment. In particular, the treatments
with organic matter and earthworms, which stimulated soil
metabolic processes and showed the high-quality agronomic
properties, achieved the best result as regards to hydrocarbon
degradation. Paulownia tomentosa was more efﬁcient than
Cytisus scoparius in extracting metals, conﬁrming their ﬁtness as
a species for phytoremediation of the polluted soil.
Therefore, on the basis of these results, the ecological
approach has recently been transferred to a real scale situation
(15 000 m
) in order to carry out the bioremediation of the
polluted soil in San Giuliano Terme Municipality (Pisa, Italy).
In order to avoid the recurrence of metals on soil removal of
earthworms and leaves fallen on soil should be considered.
This study was carried out within the framework of a project
ﬁnanced by San Giuliano Terme Municipality and Tuscany
region. We would like to thank Guido Masotti, Agata Manca
and Giada Bartalini for their assistance in the laboratory assays
and Chris Powell for the English revision of the manuscript.
1 C. J. A. Macleod and K. T. Semple, Environ. Pollut., 2006, 139, 455–
2 S. Khan, Q. Cao, A. J. Lin and Y. G. Zhu, Environ. Sci. Pollut. Res.,
2008, 15, 344–353.
3 S. M. N. Chadhain, R. Sean Norman, K. V. Pesce, J. J. Kukor and
G. J. Zylstra, Appl. Environ. Microbiol., 2006, 72, 4078–4087.
4 US-EPA, in An Analysis of Composting as an Environmental
Remediation Technology, US Environmental Protection Agency,
5 US-EPA, in Introduction to Phytoremediation, US Environmental
Protection Agency, (600/R-99/107), 2000.
6 A. Z. Hickman and B. J. Reid, Environ. Int., 2008, 34, 1016–1022.
7 A. Z. Hickman and B. J. Reid, Environ. Int., 2008, 34, 1072–1081.
8 F. Binet, A. Kersante, C. Munier-Lamy, R. C. Le Bayon, M. J. Belgy
and M. J. Shipitalo, Soil Biol. Biochem., 2006, 38, 1255–1263.
9 B. Ceccanti, C. Garcia, G. Masciandaro, C. Macci and S. Doni,
Water, Air, Soil Pollut., 2006, 177, 383–397.
10 S. M. Contreras-Ramos, D. Alvarez-Bernal and L. Dendooven,
Environ. Pollut., 2006, 141, 396–401.
11 V. Bianchi and B. Ceccanti, Toxicol. Environ. Chem., 2010, 92, 477–
Table 3 Principal components (PCs) and component loadings
Variables PC 1 PC 2
pH 0.55 0.48
WSC 0.53 0.72
Fulvic acids 0.83
Humic acids 0.87
Pav 0.33 0.71
Dehydrogenase 0.43 0.80
b-Glucosidase 0.54 0.74
Phosphatase 0.65 0.69
TPH 0.28 0.68
Germination index 0.69
TOT meq. 0.03 0.96
AV meq. 0.20 0.93
Explained variance 6.97 6.85
Total proportionality 0.39 0.38
Variables with component loadings used to interpret the PC.
electrical conductivity; TN, total nitrogen; TOC, total organic carbon;
WSC, water soluble carbon; TP, total phosphorus; Pav, available
phosphorus, TPH, total petroleum hydrocarbons. TOT meq., total
heavy metals; AV meq., available heavy metals.
Fig. 6 Biplot of loadings plot of the analysed variables and scores plot of
soil sampling. EC, electrical conductivity; TN, total nitrogen; TOC, total
organic carbon; WSC, water soluble carbon; HA, humic acid; FA, fulvic
acid; TP, total phosphorus; Pav, available phosphorus; GI%, germina-
tion index, TPH, total petroleum hydrocarbons; DHase, dehydrogenase;
TOT meq., total metal; Av meq., available metal.
2716 | J. Environ. Monit., 2012, 14, 2710–2717 This journal is ªThe Royal Society of Chemistry 2012
12 S. L. Hutchinson, A. P. Schwab and H. K. Banks, in
Phytoremediation: Transformation and Control of Contaminants, ed.
S. C. McCutcheon and J. L. Schnoor, Wiley-Interscience Inc.,
Hoboken N.J., 2003, pp. 355–385.
13 J. A. Rentz, P. Alvarez and J. L. Schnoor, Environ. Microbiol., 2004,
14 S. Khan, A. E. Hesham, G. Qing, L. Shuang and J. He, J. Soils
Sediments, 2009, 9, 482–491.
15 Y. J. Liu, Y. G. Zhu and H. Ding, Environ. Pollut., 2007, 145, 387–
16 S. Doumett, L. Lamperi, L. Checchini, E. Azzarello, S. Mugnai,
S. Mancuso, G. Petruzzelli and M. Del Bubba, Chemosphere, 2008,
17 N. C. Brady and R. R. Weill, in The Nature and Properties of Soils,
Prentice Hall, New Jersey, 1996.
18 G. Fogarty and J. M. Facelli, Plant Ecol., 1999, 144, 27–35.
19 M. Schaefer, S. O. Petersen and J. Filser, Soil Biol. Biochem., 2005, 37,
20 J. Cheng and M. H. Wong, Biol. Fertil. Soils, 2002, 36, 72–78.
21 Y. Ma, N. M. Dickinson and M. H. Wong, Biol. Fertil. Soils, 2002, 36,
22 B. Wen, X. Hu, Y. Liu, W. Wang, M. Feng and X. Shan, Biol. Fertil.
Soils, 2004, 40, 181–187.
23 W. C. Ma, S. Broekhuizen and B. van dassen, Lutra, 1989, 32, 139–151.
24 C. A. Edwards and P. J. Bohlen, Rev. Environ. Contam. Toxicol.,
1992, 125, 23–99.
25 J. C. Yeomans and J. M. Bremner, Commun. Soil Sci. Plant Anal.,
1988, 19, 1467–1476.
26 J. Murphy and J. P. Riley, Anal. Chim. Acta, 1962, 27, 31–36.
27 APHA, Method 5520B, in Standard Methods for the Examination of
Water and Wastewater, 20005, American Public Health Association,
NW, Washington, DC, 1992.
28 C. Garcia, T. Hernandez, F. Costa, B. Ceccanti and A. Gianni,
Bioresour. Technol., 1993, 45, 47–52.
29 G. Masciandaro, B. Ceccanti, V. Ronchi and C. Bauer, Biol. Fertil.
Soils, 2000, 32, 479–483.
30 N. J. Hoekstra, T. Bosker and E. A. Lantiga, Agric. Ecosyst. Environ.,
2002, 93, 189–196.
31 S. Carroll, A. Goonetilleke and L. Dawes, Environ. Geol., 2004, 46,
32 F. Caravaca, A. Pera, G. Masciandaro, B. Ceccanti and A. Roldan,
Chemosphere, 2005, 59, 1625–1631.
33 A. Ghani, M. Dexter and K. W. Perrott, Soil Biol. Biochem., 2003, 35,
34 E. Benitez, H. Sainz and R. Nogales, Bioresour. Technol., 2005, 96,
35 J. A. Pascual, T. Hernandez, C. Garcia and M. Ayuso, Bioresour.
Technol., 1998, 64, 131–138.
36 P. Perucci, Biol. Fertil. Soils, 1992, 14, 54–60.
37 B. L. Turner, D. W. Hopkins, P. M. Haygarth and N. Ostle, Appl. Soil
Ecol., 2002, 20, 157–162.
38 H. Quiquampoix and D. Mousain, in Organic Phosphorus in the
Environment, ed. B. L. Turner, E. Frossard and D. S. Baldwin,
Wallingford, UK7 CAB International, 2005, pp. 89–112.
39 L. S. Ranganathan and S. P. Vinotha, Curr. Sci., 1998, 74, 634–635.
40 S. P. Vinotha, K. Parthasarathi and L. S. Ranganathan, Curr. Sci.,
2000, 79, 1158–1159.
41 I. Marcote, T. Hern
andez, C. Garc
ıa and P. Alfredo, Bioresour.
Technol., 2001, 79, 147–154.
42 J. Tao, B. Grifﬁths, S. Zhang, X. Chen, M. Liu, F. Hu and H. Li,
Appl. Soil Ecol., 2009, 42, 221–226.
43 J. E. Satchell, K. Martin and R. V. Krishnamoorfhy, Soil Biol.
Biochem., 1984, 16, 191–194.
44 G. Masciandaro, B. Ceccanti and C. Garcia, Bioresour. Technol.,
1997, 59, 199–206.
45 T. B. Parkin and E. Berry, Soil Biol. Biochem., 1994, 26, 1233–1238.
46 D. Lopez-Hernandez, P. Lavelle, J. C. Fardeau and M. Nino, Soil
Biol. Biochem., 1993, 25, 789–792.
47 L. J. Smith and R. I. Papendick, in Soil Microbial Ecology, ed. B.
Metting, Marcel Dekker, New York, 1993, pp. 65–94.
48 C. Macci, S. Doni, E. Peruzzi, G. Masciandaro, C. Mennone and
B. Ceccanti, J. Environ. Manage., 2012, 95, S215–S222.
49 A. Verma and M. K. K. Pillai, Curr. Sci., 1991, 61, 840–843.
50 B. Gevao, C. Mordaunt, K. T. Semple, T. G. Piearce and K. C. Jones,
Environ. Sci. Technol., 2001, 35, 501–507.
51 R. L. Sinsabaugh, Biol. Fertil. Soils, 1994, 17, 69–74.
52 B. Moreno, A. Vivas, R. Nogales, C. Macci, G. Masciandaro and
E. Benitez, Environ. Sci. Pollut. Res., 2009, 16, 253–254.
53 P. Mule and P. Melis, Commun. Soil Sci. Plant Anal., 2000, 31, 3193–
54 W. Geebelen, J. Vangronsveld, D. C. Adriano, R. Carleer and
H. Clijsters, Water, Air, Soil Pollut., 2002, 140, 261–277.
55 R. Roman, C. Fortun, M. De Sa and G. Almendros, Arid Land
Research and Management, 2003, 17, 297–311.
56 M. M. Lasat, J. Hazard. Subst. Res., 2000, 2, 1–25.
57 S. C. U. Nwachukwu, P. James and T. R. Gurney, in Joint Millennium
Meeting of the Int Soc Appl Microbiol and Soc Gen Microbiol,
Warwick University, UK, 2000, p. 14.
58 D. Potenz, E. Righetti, A. Bellettieri, F. Girardi, P. Antonacci,
L. A. Calianno and G. Pergolese, in Evoluzione della ﬁtotossicit
un terreno trattato con acque reﬂue di frantoi oleari, 1985, pp. 49–54.
59 M. P. Maila and T. E. Cloete, Int. Biodeterior. Biodegrad., 2002, 50,
60 P. J. Harvey, B. F. Campanella, P. M. L. Castro, H. Harms,
E. Lichtfouse, A. R. Schaffner, S. Smrcek and D. Werck-Reichhart,
Environ. Sci. Pollut. Res., 2002, 9, 29–47.
This journal is ªThe Royal Society of Chemistry 2012 J. Environ. Monit., 2012, 14, 2710–2717 | 2717