Journal of Environmental Management 106 (2012) 75e84

Contents lists available at SciVerse ScienceDirect

Journal of Environmental Management

journal homepage: www.elsevier.com/locate/jenvman

Toxic effects of cadmium (Cd2þ) on anaerobic biomass: Kinetic

and metabolic implications
G. Mockaitis a, J.A.D. Rodrigues b, E. Foresti a, M. Zaiat a, *
a
Hydraulics and Sanitation Department, São Carlos Engineering School, Universidade de São Paulo (SHS/EESC/USP), Av. Trabalhador São-Carlense 400,
CEP 13.566-590, São Carlos, SP, Brazil
b
Mauá Engineering School, Mauá Institute of Technology (EEM/IMT), Praça Mauá 1, CEP 09.580-900, São Caetano do Sul, SP, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: Cadmium ion (Cd2þ) toxicity on anaerobic systems, used for organic matter removal, was assessed by
Received 28 September 2011 studying its effect on kinetic parameters and metabolic changes. This fundamental study was performed
Received in revised form in a continuous fixed bed anaerobic bioreactor that treated synthetic wastewater simulating domestic
7 March 2012
sewage. The biomass was immobilized on a fixed bed made of polyurethane foam. Under influent
Accepted 29 March 2012
Available online 9 May 2012
cadmium concentrations of 0.0, 0.4, 4.4 and 6.2 mg Cd2þ L1 the organic matter removal efficiencies
were 84%, 82%, 72% and 52%, respectively. At influent concentration of 6.2 mg Cd2þ L1 the reactor had
reached its limit for cadmium toxicity. In the removal of dissolved organic matter, the first-order
Keywords:
apparent kinetic coefficients (k1 ) were 0.84, 0.67 and 0.10 h1 for the operations with 0.0, 0.4 and
app
Anaerobic processes
Methanogenesis 4.4 mg Cd2þ L1, respectively. The apparent inhibition coefficient for cadmium (kapp
i
) was 1.69 mg L1.
Toxicity Despite the toxic effects of cadmium on anaerobic organic matter removal at large Cd2þ concentrations,
Cadmium the results demonstrated that the anaerobic process was suitable for cadmium concentrations below
Anaerobic fixed bed bioreactor 29.8 mg Cd2þ L1, considering the bioavailable fraction of adsorbed cadmium in the support when the
cadmium influent concentration was 6.2 mg Cd2þ L1.
Ó 2012 Elsevier Ltd. All rights reserved.

1. Introduction
There are several industrial processes which can generate wastes
containing cadmium. The most common examples are the industries
of electronic components, electroplating, metal mechanic processing
and mining. This metal finds its application in batteries, corrosion
barriers, TV screens and energy cells. Due their toxic characteristics,
wastes from industrial processing of cadmium products should be
displaced properly in order to avoid the environment contamination.
Considering that anaerobic processes are a widespread technology of
wastewater treatment, the evaluation of cadmium toxicity in
anaerobic biomass is an important subject to study.
Cadmium is a known inhibitor in biological processes. This ion has
probably no biological function (Nursita et al., 2009), but its presence
in water and soil, even at low concentrations, is a serious environ-
mental problem and may cause severe health problems like decalci-
fication, arterial hypertension and anemia (Ikeda et al., 1999 and
Shiwen et al., 1990). Other health problems such as intestine
* Corresponding author.
E-mail address: zaiat@sc.usp.br (M. Zaiat).
ulceration, liver necrosis, and prostate and bronchial cancer have also
been reported to relate to cadmium contamination (Nomiyama,
1980).
Cadmium contamination also causes metabolism problems in
microorganisms. In a study conducted to determine the toxic role of
cadmium in Saccharomyces cerevisiae (Heo et al., 2010), the authors
found that cadmium may inhibit the expression of genes related to
copper metabolism which also affect the homeostasis of both
copper and iron. The oxygen and proton flux showed to be affected
by the presence of cadmium in Nitrosomonas europaea and Pseu-
domonas aeruginosa growing in biofilms (Mclamore et al., 2010),
causing pronounced oxidative stress when the concentration of
CdCl2 was 50 mM. It was also observed that this oxidative stress
could inhibit both aerobic respiration and facultative respiration.
Considering anaerobic system a suitable bioprocess for waste-
water treatment, the presence of metals in these processes may be
toxic to anaerobic biomass (Speece, 1996). It should be mentioned
that the majority of works concerning the role of cadmium in
anaerobic processes reports on results of anaerobic sulfidogenic
processes for its removal; thus, fundamental studies about the
effect of heavy metals on kinetics and metabolism of anaerobic
systems in which methanogenesis is the main final metabolic
degradation pathway are incipient.

0301-4797/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvman.2012.03.056
76 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84
There are few papers addressing the effect of cadmium on
methanogenic processes, but these works are performance based
and do not describe the behavior of metabolic changes. Altas (2009)
investigated the inhibitory effects of zinc, chromium, nickel, and
cadmium on methanogenic activity in an upflow anaerobic sludge
blanket (UASB) granular sludge with metal concentrations ranging
from 0 to 128 mg L1, and concluded that cadmium had the highest
half maximal inhibitory concentration (IC50) of the investigated
metals, i.e., it was the least toxic to methanogenesis. On the other
hand, Bhattacharya et al. (1995) studied the toxic effects of cadmium
on methanogenic systems (batch and continuous) and assessed
acetate consumption under cadmium concentrations ranging from
0 to 80 mg L1. Instead of total cadmium, free ionic cadmium was
used to more suitably evaluate the response of the system to
toxicity. The limiting concentrations for methanogenesis showed to
be 0.1 mg L1 free cadmium and approximately 7.5 mg L1 total
cadmium. In addition, the inhibitory effects of cadmium tended to
be lower in systems with higher total solids concentration.
Heavy metals removal using anaerobic processes has been
addressed by some authors and is the most studied bioprocess.
According to Neculita et al. (2007) one source of wastewater that
has been treated in this manner is acid mine drainage (AMD);
biological sulfate reduction resulted in the generation of sulfide,
which caused precipitation of the soluble metal content. Interac-
tions of extracellular polymers that are formed by marine sulfate-
reducing bacteria (SRB) and other selected cultures that reduce
metallic ions such as chromium, nickel and molybdenum were
assessed in a previous work (Beech and Cheung, 1995). A study
carried out by Alvarez et al. (2007) proposes an anaerobic biore-
actor fed with fatty acids and sulfate to generate hydrogen sulfide
and precipitate zinc, copper and lead in a separate vessel.
Most of these studies did not show the toxic effects of cadmium on
immobilized biomass bioreactors and biofilms. Biofilms are protected
from cadmium exposure compared to suspended cells because the
mass transfer resistance protects the cells inside the biofilm and
Fig. 1. Schematic of the continuous anaerobic bioreactor with a fixed-structure bed and cross-cut view of the bioreactor showing the immobilization matrix distribution.
longer cellular retention times can be achieved in such systems.
However, cadmium ions may be trapped in the immobilization
matrices because the solids are able to chelate Cd2þ, increasing the
concentration of cadmium in the biofilm. Moreover, there is
a potential increase in total solids in the bed because of precipitation
effects and because biopolymer excretion tends to clog the bioreactor
bed. In this sense, a structured bed configuration appears to mitigate
the problem of clogging because of the free space surrounding the
biomass immobilization matrix (Picanço et al., 2001).
This paper aims to assess the toxic effects of cadmium on
anaerobic biomass, using a continuous fixed bed anaerobic biore-
actor, which was operated at constant temperature and hydraulic
retention time (HRT) and was subjected to increasing cadmium
concentrations.
2. Materials and methods
2.1. Bioreactor configuration
A continuous anaerobic bioreactor with a fixed-structured bed
was constructed of poly-methyl methacrylate (PMMA) with a total
volume of 4.77 L, as shown in Fig. 1. Flexible polyvinyl chloride
(PVC) (TygonÒ) tubing with an internal diameter of 3 mm (an
external diameter of 5 mm) was used for tubing and connections.
The reactor was fed at a flowrate of 0.397 L h1 (6.63 mL min1) by
a GilsonÒ model Minipuls Evolution peristaltic pump, which cor-
responded to a HRT of approximately 12 h. The reactor was placed
in a controlled temperature chamber and operated at 30  C.
2.2. Synthetic medium
The bioreactor was fed with synthetic medium with a similarity
with an organic complex wastewater (prepared accordingly the
works of Ratusznei et al., 2003; Friedl et al., 2009 and Mockaitis
et al., 2010), aiming a mean concentration of 500 mg O2 L1, in
 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84
terms of chemical oxygen demand (COD) (Table 1); the cadmium
source was CdCl2$2H2O.
In the first condition (Condition I), the bioreactor was fed with
synthetic wastewater without cadmium. In subsequent experi-
mental conditions (conditions II through IV), the concentration of
Cd2þ was systematically increased. Thus, each condition corre-
sponds to a different mean cadmium concentration (Condition
II ¼ 0.4 mg Cd2þ L1, Condition III ¼ 4.4 mg Cd2þ L1 and Condition
IV ¼ 6.2 mg Cd2þ L1).
2.3. Immobilization matrix and inoculum
Granulated sludge was obtained from a UASB bioreactor treating
poultry slaughterhouse wastewater with total volatile solids of
35.8 g TVS L1. Biomass quantification, which was performed at the
end of all assays, demonstrated that the bioreactor operated with
58.1 g TVS (12.2 g TVS L1, assuming a total bioreactor volume of
4.77 L).
The immobilization matrix for the biomass was comprised of 17
fixed polyurethane foam strips (with a superficial area of
43.8 m3 g1) with a length of 700 mm, constituting 32 g of biomass
support in the bioreactor. Fig. 1 shows the cross-section of the
reactor, illustrating the position of the bed. The inoculation proce-
dure was performed by immersing the immobilization matrix in
grinded sludge for 2 h at room temperature.
2.4. Physicochemical analysis
Organic matter concentration, amount of organic volatile acids
and total alkalinity were analyzed to determine system efficiency
and stability. Total and free cadmium concentrations were deter-
mined for the reactor affluent and effluent, which allowed per-
forming a mass balance of cadmium retention in the system. The
total and bioavailable cadmium concentrations in the immobiliza-
tion matrix were analyzed to determine the actual concentration of
cadmium affecting the biomass.
The chemical oxygen demand (COD) and solids were analyzed
by Standard Methods for the Examination of Water and
Wastewater (2005), COD-5220 and Solids-2540, respectively.
A titration analysis to determine total alkalinity (TA) and total
volatile acids (TVA) was also performed (Dillalo and Albertson, 1961
modified by Ripley et al., 1986). COD was determined for both non-
filtered (total) and filtered (dissolved) samples (considering colloidal
particles smaller than 1.2 mm), and the pH was determined using
potentiometric measurements (standard glass electrode, Digimed
Inc., Brazil). TA and TVA analysis gives a fast response about process
stability and allows for comparisons with other systems in which
a more accurate fatty acids analysis are not available (e.g.: full scale
processes).
Cadmium analyses were made using atomic absorption spec-
troscopy, as described in the Standard Methods using a VarianÓ
AA240FS spectrophotometer. The light source was a cadmium
lamp with hollow cathode at a current of 4 mA (Varian coded
lamp 5610100800). The amounts of total, free and bioavailable
cadmium were measured. The free cadmium concentration was
obtained by filtering a raw sample through a hydrophilic 0.22 mm
cellulose acetate membrane. The amount of bioavailable
cadmium was determined for the immobilized biomass and was
extracted from the immobilization matrix using an Sr(NO3)2
solution (0.1 mol L1) (Nursita et al., 2009 and Ahnstrom and
Parker, 1999). Both total (non-filtered), free (filtered) and
bioavailable (extracted with Sr(NO3)2 solution) samples were
treated with concentrated aqua regia (HCl/HNO3 2:1 v/v) under
boiling heat.
77
Chromatographic determinations of the fatty acids were
performed by HPLC with a modular ShimadzuÒ chromatograph
using a pump system (LC-10AD), oven (CTO-20A), controller
(SCL-10A) and a photo diode array (PDA) detector adjusted to
read wavelengths ranging from 190 to 370 nm (UV spectra)
with a 1 nm step. A chromatogram showed readings recorded at
a wavelength of 205 nm. A Bio-Rad AminexÒ HPX-87H
3000  7.8 mm column was operated at a constant tempera-
ture of 55  C. The eluent was a 0.005 mol L1 H2SO4 solution at
a flowrate of 0.8 ml min1, and the volume of the injected
sample was 100 mL. The solutions were analyzed for citric, malic,
lactic, formic, succinic, acetic, propionic, butyric, isobutyric,
valeric, isovaleric and caproic fatty acids. The concentrations
were expressed in terms of COD (mg O2 L1) and normalized by
the concentration of dissolved organic matter, assuming that
these acids alone comprised all of the effluent organic matter, to
ensure that the acid concentrations were comparable to the
organic matter concentrations.
2.5. Experimental procedure
The experiment was divided into four conditions, depending on
the cadmium concentration in the synthetic wastewater: Condition
I e no cadmium (control condition), Condition II e 0.4 mg Cd2þ L1,
Condition III e 4.4 mg Cd2þ L1, and Condition IV e 6.2 mg Cd2þ L1.
The reactor was inoculated only at the start of Condition I and was
not re-inoculated during the other experiments. In each of the
conditions, the temperature was maintained at 30  1  C. Table 2
shows the experimental HRT values for each condition associated
with its respective influent flow.
For each condition, the bioreactor was monitored to verify the
stability and efficiency of the process. After achieving stable
operating conditions, spatial profiles of both dissolved and total
organic matter concentrations were used to determine the
intrinsic kinetic parameter (kapp
1 ). The organic acid concentration
profiles along the length of the bioreactor were used to verify the
metabolic pathway. These profiles were obtained by taking
10 mL samples at each sampling point described in Fig. 1, totaling
80 mL from each profile (1.7% of the working volume of the
bioreactor).
2.6. Biomass and immobilization matrix analysis
At the end of the last experimental condition, the biomass
content in the immobilization matrix and the amount of cadmium
adsorbed by the biofilm were assessed through a TVS assay and by
treating the remaining ashes with concentrated aqua regia under
boiling heat; the analysis also considered the bioavailable cadmium
content. Three samples were taken from the immobilization
matrix, corresponding to the top, middle and bottom sections of the
reactor. For each of these samples, the following variables were
determined: total solids (Eq. (1)), TVS (Eq. (2)), total cadmium
content in TVS (Eq. (3)), bioavailable cadmium content in total
volatile solids (Eq. (4)) and critical cadmium concentration in the
bioreactor (Eq. (5)):

 m105 $mT
TS mg mg support1 ¼ S (1)
mS $mm
 105 

 mS  m550 $mT
1 S
TVS mg mg support ¼ (2)
mS $mm
78 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84

 mCd Table 2
S $mT
TVSTCd mg Cd2þ mg TVS1 ¼ (3) Actual HRT and influent flow in the bioreactor for all conditions: I e without
mS $mm $TVS cadmium contamination (control), II e 0.4 mg Cd2þ L1, III e 4.4 mg Cd2þ L1, and IV
e 6.2 mg Cd2þ L1.

 mCd Bio $m Variable Experimental conditions

mg TVS1 ¼
2þ T
TVSBio
Cd mg Cd
S
(4) I II III IV
mS $mm $TVS
HRT (h) 16  3 (9) 14  1 (14) 12  3 (14) 12  0 (15)
Flow rate (L h1) 0.3  0.1 (9) 0.4  0.0 (14) 0.5  0.2 (14) 0.4  0.0 (15)

 TVSBio
Cd $TVS$mm
CCritical mg Cd2þ L1 ¼ (5)
VBioreactor palindromic sequences on the RNA operon is conserved and related
to a unique phylogenetic pattern (Massol-Deyá et al., 1999).

2.9. Determination of kapp

1 and competitive inhibition
2.7. Mass balances for the determination of cadmium content coefficient (kapp
i
)

To determine the concentration of bioavailable cadmium to A modified first-order model was fitted to the experimental
which the biomass was subjected during the determination of the organic matter concentration results, assuming the existence of
kinetic profiles, cadmium content balances were performed as a concentration of residual organic matter at which the reaction
described in Eq. (6). rate was zero, as shown in Eq. (8). This model is derived from the


Bio
CCd mg Cd2þ L1 ¼ CCd
Bio P Monod microbiological kinetic model by assuming that the
h
 i substrate concentration has much smaller value than kS, so that the
toperation $Q $ T
CCd T
CCd (kS þ C) term in Eq. (9) can be considered to be a single constant k*S .
þ influent
$aBio (6) app
Eq. (9) shows the definition of k1 in terms of the Monod kinetic
VBioreactor
model constants.
The relationship between the bioavailable cadmium concen-
 app
tration and the total cadmium concentration can be expressed by C mg L1 ¼ Cresidual þ ðCinitial  Cresidual Þ$ek1 $HRTS (8)
an empirical factor, aBio, as defined in Eq. (7).

 CX $mmax
h1
app
k1 ¼ (9)
TVSTCd Yx=s $ðkS þ CÞ
aBio ¼ (7)
TVSBio
Cd To evaluate the toxic potential of cadmium in the process, an
enzyme-based competitive inhibition model was used to assess the
variation of the kapp
1 values under increasing bioavailable cadmium
app
2.8. Microbiological examinations concentrations. The parameter ki is an overall inhibition coeffi-
cient that is directly proportional to the toxic effects of a contami-
Microbiological examinations of the biomass were performed to app
nant in the bioreactor. Eq. (10) shows the model in terms of k1 .
verify changes in the microbiological equilibrium and identify
!1
possible metabolic pathway detours.
 k
app
app 1 app
An amplified rDNA restriction analysis (ARDRA) technique k1 h ¼ k1 j ci¼0 $ 1þ i (10)
(Massol-Deyá et al., 1999 and Griffiths et al., 2000) was applied to
Ci
compare the microbiological diversity along the bioreactor. The
All data fitting was performed by summing the squared errors
rDNA extracted from the samples was amplified by polymerase
and applying the LevenbergeMaquardt optimization algorithm
chain reaction (PCR) using Bacteria- (EUB 21F e 1100R) and Archaea-
using the Linux version of the QtiPlotÓ software version 0.9.7.10.
(1Af e 1100 Ar) specific primers, and the restriction enzymes used in
both the Bacteria- and Archaea-amplified rDNA samples were Hha I
(sequence GCG/C cleavage) and Hae III (sequence GGCC/C cleavage). 3. Results and discussion
ARDRA technique involves cleavage of palindromic regions of rDNA,
increasing the accuracy of comparisons, providing a fingerprinting of 3.1. Operation
bacterial genomes. The principle of ARDRA method is that the
Table 3 summarizes the mean values of the main influent
characteristics for all of the assayed conditions, and Table 4
Table 1
Synthetic wastewater composition.
Table 3
Composition Concentration Approximated Influent characteristics for all experimental conditions: I e without cadmium
(mg L1) COD Contend contamination (control), II e 0.4 mg Cd2þ L1, III e 4.4 mg Cd2þ L1, and IV e 6.2 mg
(mg O2 L1) Cd2þ L1.
Sucrose 35 38 Variable Experimental conditions
Starch 114 103e115
Cellulose 34 34e38 I II III IV
Meat Extract 208 122e138 CT (mg O2 L1) 540  123 (9) 485  95 (10) 517  71 (12) 513  107
Soybean Oil 51 153 (11)
Linear Alkylbenzene Sulfonate (LAS) 15 34 T (mg Cd2þ L1)
CCd e 0.4  0.2 (8) 4.4  0.5 (8) 6.2  0.9 (7)
Sodium Chloride (NaCl) 250 e F (mg Cd2þ L1)
CCd e 0.2  0.1 (8) 3.0  0.4 (8) 4.3  1.0 (7)
Magnesium Chloride (MgCl2$6H2O) 7 e TVA (mg HAc L1) 27  7 (5) 27  4 (3) 22  5 (4) 22  3 (2)
Calcium Chloride (CaCl2$2H2O) 5 e TA (mg CaCO3 L1) 71  4 (4) 108  23 (3) 71  2 (4) 75  3 (2)
Sodium Bicarbonate (NaHCO3) 200 e pH 7.6  0.3 (4) 7.7  0.2 (3) 7.6  0.2 (4) 7.7  0.4 (2)
 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84 79

Table 4 Table 6
Effluent results for all experimental conditions: I e without cadmium contamination Volatile organic acids and ethanol mean concentrations (in terms of COD) in the
(control), II e 0.4 mg Cd2þ L1, III e 4.4 mg Cd2þ L1, and IV e 6.2 mg Cd2þ L1. bioreactor effluent under cadmium-contaminated conditions: II e 0.4 mg Cd2þ L1,
III e 4.4 mg Cd2þ L1, and IV e 6.2 mg Cd2þ L1.
Variable Experimental conditions
Acid (mg O2 L1) Experimental conditions
I II III IV
CT (mg O2 L1) 85  49 (10) 90  40 (14) 115  33 (16) 245  57 (15) II III IV
YT 0.8  0.0 0.8  0.1 0.8  0.1 0.5  0.1 Citric e 2.4  0.1 (6) 2.7  0.1 (11)
CD (mg O2 L1) 47  37 (10) 69  21 (12) 84  23 (16) 202  55 (15) Malic e 1.0  0.1 (5) 1.3  0.6 (8)
YD 0.9  0.1 0.9  0.1 0.8  0.0 0.6  0.1 Formic 1.0  0.1 (14) 0.5  0.1 (13) 0.6  0.1 (11)
T (mg Cd2þ L1)
CCd e 0.2  0.1 (11) 1.0  0.5 (10) 2.6  0.7 (9) Succinic 6  3 (13) 5  1 (13) 7  1 (12)
T
YCd e 0.5  0.3 0.8  0.1 0.6  0.1 Lactic 5  1 (5) 5.2  0.2 (13) 6.3  0.4 (12)
F (mg Cd2þ L1)
CCd e 0.1  0.1 (11) 0.5  0.2 (10) 1.7  0.5 (9) Acetic 5  2 (12) 3.7  0.7 (13) 19  5 (12)
F
YCd e 0.5  0.4 0.8  0.0 0.6  0.1 Propionic 6  1 (11) 5.3  0.5 (13) 17  4 (12)
TVA (mg Hac L1) 24  15 (7) 27  5 (8) 33  9 (6) 64  18 (7) Isobutyric e 7  1 (13) 7  1 (11)
TA (mg CaCO3 L1) 91  37 (7) 115  34 (8) 104  2 (6) 84  10 (7) Butyric 9  3 (14) 8  2 (12) 8  2 (12)
pH 7.2  0.3 (7) 7.4  0.2 (8) 6.9  0.2 (6) 6.9  0.2 (7) Isovaleric 13  4 (10) 16  6 (13) 28  10 (12)
Valeric 16  7 (12) 17  5 (13) 29  7 (12)
Caproic e 13  5 (8) 76  23 (12)
Ethanol 8.0  0.7 (9) e e

compares the obtained effluent results. The observed difference

between the total (CCdT ) and free (C F ) cadmium concentrations in
Cd
the influent indicates that the medium may form colloids that can
trap the metal in the form of a chelate. The occurrence of this
phenomenon is also indicated by the cadmium concentrations in
the effluent which presented CCd F values lower than CCd T , as
depicted in Table 4, showing that part of the metal may be caught
in a dissolved micelle and avoid treatment. The differences
between the cadmium concentrations in the influent and the
effluent show that cadmium accumulation occurred in the reactor
throughout the experiments. Because it was impossible to sample
the biomass during the assays, the cadmium concentration to
which the biomass was subjected and the cadmium mass balance
were evaluated, as described by Eq. (6), at the end of each
experimental condition by considering the total cadmium
concentrations (CCdT ). Due the system immobilization matrix and
inoculum were the same in all operations, the initial cadmium
concentration in the bioreactor was the same as the final
concentration from the previous operation (*CCd T ). For example,
Condition II (in which the bioreactor was fed with 0.4 mg
Cd2þ L1) began with no cadmium contamination. The mass
balance results are presented in Table 5, which shows the total and
bioavailable cadmium concentrations to which the biomass was
subjected and compares these concentrations to the influent
cadmium concentration.
The increasing toxic effects of cadmium can be established by
the decrease in the organic matter removal efficiency (Table 4: YT
for total and YD for dissolved forms) with the increase in the
influent cadmium concentration. The effluent by-product compo-
sition was also affected, as shown by the increased values of TVA
and by the accumulation of some specific fatty acids, as shown in
Table 6. In Condition III, citric, malic, isobutyric and caproic acids
were produced, but these acids remained in low and stable
concentrations, indicating that there was a change in the formation
of metabolites from acidogenic and acetogenic processes compared
differences in concentration from Condition II to Condition III. In
Condition IV, high values of caproic, valeric, isovaleric and pro-
pionic acids were observed, indicating that the bioreactor was
unstable under this condition. It is probable that the acidogenic and
acetogenic microbiological communities were affected and that the
effect on these microorganisms that was observed in Conditions II
and III was amplified in this condition. It is possible that in
Condition IV, the activity of acetoclastic methanogens was also
inhibited, as indicated by the high concentration of acetic acid.
Under the experimental conditions reported here, the maximal
toxic effect (when the process became unviable) occurred in
Condition IV, when the bioreactor was unstable. Fig. 2 shows the
unstable behavior of the organic matter concentration (in terms of
COD) for non-filtered and filtered samples and the respective effi-
ciencies. The bioreactor operating under Conditions II and III was
affected by the toxic effect of cadmium, showing a decrease in the
organic matter removal efficiency values and an increase in the acid
concentrations. However, the bioreactor remained stable under
these conditions, which allowed for the determination of spatial
profiles that were used to obtain the kinetic parameters.
app
3.2. Evaluation of k1 and kapp
i
All of the spatial profiles were collected at the end of the each
experimental condition under a steady-state regime. Fig. 3 depicts
O

to Condition II. All of the other acids showed no considerable

Table 5
Cadmium mass balance at end of each condition: II e 0.4 mg Cd2þ L1, III e 4.4 mg
Cd2þ L1, and IV e 6.2 mg Cd2þ L1.

Experimental Total operating time (d) Cadmium concentration (mg L1)

conditions
Influent Bioreactor Bulk

Total Bioavailable
II 50 0.4 18.5 1.37
III 26 4.4 221 16.4
Fig. 2. Organic matter concentration (--YD and A-YT) and organic matter efficiencies
IV 26 6.2 401 29.7
(:-CD and ;-CT) in the bioreactor operating under Condition IV.
80 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84

A B

T
C
T

Fig. 3. Organic matter concentration profile (,-CD and --CT) for the determination of the first-order kinetic coefficient (kapp
1 ) for A e Condition I, B e Condition II and
C e Condition III.

the experimental data normalized by the mean value of the initial 1.69  0.23 mg L1 (R2 ¼ 0.99) for the dissolved organic matter
concentration. The first-order kinetic decay was fitted to the kinetic concentration profiles (CD) and 3.32  0.99 mg L1 (R2 ¼ 0.97) for
model of organic matter concentration across the bioreactor for the total organic matter concentration profiles (CT). Attempts to fit
total (CT) and dissolved (CD) samples for Conditions I, II and III. The uncompetitive and non-competitive inhibition models were per-
app
kinetic parameters of the best fit, which are k1 , Cinitial, Cresidual and formed, but they showed no convergence and are therefore not
2
the correlation coefficient R , are presented in Table 7. considered in this discussion.
The apparent kinetic coefficient decreased exponentially with As expected, increasing cadmium concentrations directly
increasing cadmium concentration (Fig. 4). In Condition IV, in inhibited organic matter removal, as quantified by the kinetic
which the kinetic profile showed a bioavailable cadmium concen- constant of the competitive inhibition model. In this case, the
tration of 29.7 mg Cd2þ L1, the bioreactor became unstable, and inhibition model had no direct mechanistic meaning. However,
the kapp
1 could not be determined. Because long-term exposure of
the biomass to this critical cadmium concentration would halt
operation of the reactor, it is reasonable to state that the kapp 1 in
Condition IV was zero. Therefore, the kapp i
for cadmium was

Table 7
app
First-order apparent kinetic coefficient (k1 ), initial and residual organic matter
)

concentrations (Cinitial and Cresidual, respectively) obtained from the fitting of dis-
solved (CD) and total (CT) organic matter concentration profiles.

Variable/ Experimental conditions

Parameter
I II III

CD CT CD CT CD CT
Bioavailable 0 (Control 1.37 16.4
cadmium condition)
concentration
(mg L1)
Cinitial 374  15 374  15 384  23 426  34 341  11 385  20
(mg O2 L1)
90  6 129  8 87  11 103  17 0a 0a
Cresidual
)
(mg O2 L1)
kapp (h1) 1.2  0.2 0.8  0.2 0.7  0.1 0.7  0.2 0.08  0.10  app
1 Fig. 4. First-order kinetic coefficient (k1 ) decay considering both 6-CD and :-CT
0.01 0.02
2 with increasing in situ bioavailable cadmium concentration (Ci) to determine the
R 0.93 0.88 0.91 0.84 0.93 0.84
competitive inhibition coefficient (kapp
i
). The model was fitted for both CD and CT
a
These values had to be forced to make the model converge. (dashed and bold lines, respectively).
 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84 81

A B

C D

E F

G H

I J

Fig. 5. Volatile organic acid profiles of the all stable conditions (--Condition I, :-Condition II and A-Condition III). A e formic, B e succinic, C e lactic, D e acetic, E  propionic,
F e isobutyric, G e butyric, H e isovaleric, I e valeric and J e caproic acid.

there are many studies reporting that cadmium is a competitive
inhibitor of some enzymes (Norris and Kelly, 1977; Stohs and
Bagchi, 1995; Congiu et al., 2000), and considering that some
enzymes are inhibited by the presence of cadmium during intra-
cellular metabolism, the inhibition mechanism may be reproduced
at the cellular level. Consequently, the kinetic profile of the
metabolites may reflect this type of inhibition in the enzymes
present in the microorganisms of this bioreactor. Nevertheless, the
kapp
i
value is suitable to compare and measure the toxic effect of
cadmium on the anaerobic system.
3.3. Volatile organic acid profiles
Classic anaerobic digestion comprehends four main stages:
hydrolysis (which transform complex organic matter into smaller
molecules), acidogenesis (which convert the products of hydrolysis
step in organic fatty acids), acetogenesis (acetate production) and
methanogenesis (consumption of acetate and CO2/H2 with methane
production). Acidogenesis regulation is a very important step to
maintain the overall process stability (Mosey, 1983). Thus, the eval-
uation of toxic effects over the dynamic of volatile organic acids in
82 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84

Table 8 polymer in Condition IV. Based on these results, it is possible to

Analysis of the immobilization matrix and the extracellular polymer after Condition infer that the bottom section had the highest cadmium concen-
IV.
tration. In this section, the high TVS value indicates that extracel-
Analysis Position in bioreactor Extracellular lular polymer was formed, a phenomenon that can be attributed to
polymer the high cadmium concentration at the bottom of the bioreactor.
Top Middle Bottom
Mean relative position (m) 0.83 0.50 0.17 e Therefore, aBio and CCritical were determined only for the bottom
TS (mg TS mg support1) 4.1 4.5 5.3 0.02 section of the bioreactor because this section was affected most
TVS (mg TVS mg support1) 3.7 4.0 4.7 0.01 greatly by the toxic effects of cadmium. The overall aBio value was
TVSTCd (mg Cd mg TVS1) 24 29 31 54
0.0741, and the CCritical was 29.82 mg Cd2þ L1.
TVSBio 1 2.0a 2.1a 2.2 e
Cd (mg Cd mg TVS ) The extracellular polymer produced during the operation of
a
Values calculated using the aBio value determined at the bottom of bioreactor. bioreactor was a viscous material with a strong yellowish color,
which is typical of cadmium compounds. The total amount of this
material was approximately 200 mL, corresponding to 4.2% of the
bioreactor volume. As shown in Fig. 6, this material showed some
the reactor gives an overview of how cadmium affects organic
morphologies representative of diverse microorganisms, and
matter metabolism.
epifluorescent microscopy showed evidence of the presence of
Fig. 5 illustrates the production and consumption of the volatile
Archaea in this polymer. Therefore, solids quantification (Table 8)
acids in the bioreactor under steady-state conditions. In Condition I,
and microbiologic examinations were performed on this polymer.
all of the analyzed organic acids were quickly produced and subse-
The method for determining bioavailable cadmium, described by
quently consumed across the reactor, readily reaching low residual
Ahnstrom and Parker (1999) and adapted by Nursita et al. (2009),
concentrations. In Condition II, in spite of some toxic effects, the
is not applicable to liquid samples such as this polymeric
same behavior was observed, but the initial concentrations of formic,
material.
lactic, acetic, propionic, isobutyric, butyric, isovaleric and valeric acid
The small deviation (0.5%) between the experimental value of
were higher than in Condition I and had higher residual concen-
CCritical and the value obtained from the mass balance in Eq. (6) and
trations, indicating that the acidogenic and acetogenic communities
demonstrated in Table 5 corroborates the accuracy of the mass
were inhibited. Caproic acid was not detected in this condition, and
balance, which was performed to estimate the actual bioavailable
the higher initial value for the formation of acids indicates that the
cadmium concentrations to which the biomass was subjected.
hydrolysis step was also affected, indicating a possible metabolic
The ARDRA technique was applied to samples taken from the
pathway detour towards hydrolysis and the formation of longer
top and bottom sections of the bioreactor and from the extra-
chain acidogenic products. In Condition III, the initial concentrations
cellular polymer. These samples, when compared to the initial
were lower than those in Condition II, and an inhibitory effect on
inoculum, showed that both the Bacteria and Archaea microbio-
hydrolytic bacteria was observed. The caproic acid profile showed
logical communities underwent some modifications, indicating
a formation curve followed by a constant high concentration,
that some communities present in the initial inoculum were not
corroborating that the hydrolysis step was affected by cadmium
found after the experiment, and also that some microorganisms
toxicity. A peak in the formation of acetic, propionic, and isobutyric
that did not appear in the initial inoculum were found in the
acids with an HRT ranging from 0.5 to 2.0 h also indicates the
bioreactor after the experiment. These results may be related to
possible inhibition of acidogenic and acetogenic bacteria and ace-
the greater presence of longer-chain volatile acids in the reactor
toclastic methanogens.
effluent, which demonstrates that the toxicity of cadmium may
primarily affect the acidogenic and acetoclastic methanogenic
3.4. Analysis of biomass and the immobilization matrix at the communities.
critical cadmium concentration No changes were observed between the bottom and top sections
of the reactor, indicating a lack of biomass stratification. However,
At the conclusion of Condition IV, when the bioreactor showed some differences were detected for the Bacteria domain between
a lack of stability, a cadmium content analysis was performed on the extracellular polymer and the immobilization matrix. There-
the immobilization matrix to determine the critical cadmium fore, it can be inferred that the microorganisms present in both
concentration. Table 8 shows the values of TS, TVS, TVSTCd and media were from different communities. If the extracellular poly-
TVSBio mer were produced because of the toxic effect of cadmium, the
Cd found in the immobilization matrix and the extracellular

Fig. 6. Microphotography of the morphology of the extracellular polymer found in the bioreactor at the end of Condition IV. The photographs were obtained using A e epifluorescent
microscopy and B e phase-contrast microscopy.
 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84
communities found in this polymer are likely to be more resistant
to the toxic effects of cadmium.
4. Conclusions
The toxic effects of cadmium in a continuous anaerobic biore-
actor with a fixed-structured bed was observed for all conditions, as
the decay kapp 1 value appeared with increasing bioavailable
cadmium concentration. The bioreactor showed stable and efficient
operation for the removal of organic matter when operated with
cadmium concentrations below the critical concentration
(CCritical ¼ 29.82 mg Cd2þ L1).
Cadmium accumulation in the bioreactor was observed during
the assays by analyzing the differences between the influent and
effluent cadmium concentrations. The adsorption of cadmium on
the immobilization matrix in the bioreactor increased the in situ
cadmium concentration; nevertheless, this adsorption made the
cadmium partially unavailable for the microbiological communities
in the bioreactor, which reduced the toxic effects of the increased
concentration of cadmium.
The kinetic profiles, organic volatile acids behavior and the
ARDRA technique made it possible to infer changes to the metabolic
pathways. A shift in the microbiological communities towards the
production of fatty acids with longer chains was observed, which
affected the microorganisms responsible for acetic acid consump-
tion, and primarily changed the caproic acid profiles.
This study showed that this configuration is an attractive and
suitable method for the treatment of wastewater containing heavy
metals; however, it is important to monitor the concentration of
metal in the immobilization matrix and the formation of extracel-
lular polymers.
Acknowledgments
This study was supported by the Fundação de Amparo à Pes-
quisa do Estado de São Paulo, Brasil (FAPESP), process numbers 05/
51.702-9 and 07/07.574-1 to G. Mockaitis. The authors gratefully
acknowledge Dr. Eloisa Pozzi for her help with the molecular
biology techniques and discussions.
References
Ahnstrom, Z.S., Parker, D.R., 1999. Development and assessment of a sequential
extraction procedure for the fractionation of soil cadmium. Soil Science Society
of America Journal 63, 1650e1658.
Altas, L., 2009. Inhibitory effect of heavy metals on methane-producing anaerobic
granular sludge. Journal of Hazardous Materials 162, 1551e1556.
Alvarez, M.T., Crespo, C., Mattiasson, B., 2007. Precipitation of Zn(II), Cu(II) and Pb(II)
at bench-scale using biogenic sulfide from the utilization of volatile fatty acids.
Chemosphere 66, 1677e1683.
Beech, I.B., Cheung, C.W.S., 1995. Interactions of exopolymers produced by sulphate-
reducing bacteria with metal ions. International Biodeterioration and Biodeg-
radation 35, 59e72.
Bhattacharya, S.K., Madura, R.L., Uberoi, V., Hahighi-Podeh, M.R., 1995. Toxic effects
of cadmium on methanogenic systems. Water Research 29, 2339e2345.
Congiu, L., Chicca, M., Pilastro, A., Turchetto, M., Tallandini, L., 2000. Effects of
chronic dietary cadmium on hepatic glutathione levels and glutathione
peroxidase activity in Starlings (Sturnus vulgaris). Archives of Environmental
Contamination and Toxicology 38, 357e361.
Dillalo, R., Albertson, O.E., 1961. Volatile acids by direct titration. Journal Water
Pollution Control Federation 33, 356e365.
Friedl, G.F., Mockaitis, G., Rodrigues, J.A.D., Ratusznei, S.M., Zaiat, M., Foresti, E.,
2009. AnSBBR applied to organic matter and sulfate removal: interaction effect
between feed strategy and COD/Sulfate ratio. Applied Biochemistry and
Bioengineering 159, 95e109.
Griffiths, R.I., Whiteley, A.S., O’Donnell, A.G., 2000. Rapid method for coextration of
DNA from natural environments for analysis of ribosomal DNA and rRNA-based
microbial community composition. Applied and Environmental Microbiology
66, 5488e5491.
Heo, D.H., Baek, I.J., Kang, H.J., Kim, J.H., Chang, M., Jeong, M.Y., Kim, T.H., Choi, I.D.,
Yun, C.W., 2010. Cadmium regulates copper homoeostasis by inhibiting the
83
activity of Mac1, a transcriptional activator of the copper regulon, in Saccha-
romyces cerevisiae. Biochemical Journal 432, 257e265.
Ikeda, M., Zhang, Z.W., Higashikawa, K., Watanabe, T., Shimbo, S., Moon, C.S.,
Nakatsuka, H., Matsuda-Inoguchi, N., 1999. Background exposure of general
women populations in Japan to cadmium in environment and possible health
effects. Toxicology Letters 108, 161e166.
Massol-Deyá, A., Odelson, D.A., Hickey, R.F., Tiedje, J.M., 1999. Bacterial community
fingerprinting of amplified 16S and 16e23S ribosomal DNA gene sequences and
restriction endonuclease analysis (ARDRA), pp. 3.3.2:1e8. In: Akkermans, A.D.L.,
van Elsas, J.D., de Bruijn, F.J. (Eds.), Molecular Microbial Ecology Manual. Kluwer
Academic Publishers, Dordrecht, The Netherlands.
Mclamore, E.S., Zhang, W., Porterfield, D.M., Banks, M.K., 2010. Membrane-aerated
biofilm proton and oxygen flux during chemical toxin exposure. Environmental
Science and Technology 44, 7050e7057.
Mockaitis, G., Friedl, G.F., Rodrigues, J.A.D., Ratusznei, S.M., Zaiat, M., Foresti, E.,
2010. Influence of feed time and sulfate load on the organic and sulfate removal
in an ASBR. Bioresource Technology 101, 6642e6650.
Mosey, F.E., 1983. Mathematical modeling of the anaerobic digestion process:
regulatory mechanisms for the formation of short-chain volatile acids from
glucose. Water Science and Technology 15, 209e232.
Neculita, C.M., Zagury, G.J., Bussière, B., 2007. Passive treatment of acid mine
drainage in bioreactors using sulfate-reducing bacteria e Critical review and
research needs. Journal of Environmental Quality 36, 1e16.
Nomiyama, K., 1980. Recent progress and perspectives in cadmium health effects
studies. The Science of the Total Environment 14, 199e232.
Norris, P.R., Kelly, D.P., 1977. Accumulation of cadmium and cobalt by Saccharomyces
cerevisiae. Journal of General Microbiology 99, 317e324.
Nursita, A.I., Singh, B., Lees, E., 2009. Cadmium bioaccumulation in Proisotoma
minuta in relation to bioavailability in soil. Ecotoxicology and Environmental
Safety 72, 1767e1773.
Picanço, A.P., Vallero, M.V., Gianotti, E.P., Zaiat, M., Blundi, C.E., 2001. Influence of
porosity and composition of supports on the methanogenic biofilm character-
istics developed in a fixed bed anaerobic reactor. Water Science and Technology
44, 197e204.
Ratusznei, S.M., Rodrigues, J.A.D., Camargo, E.F.M., Ribeiro, R., Zaiat, M., 2003.
Effect of feeding strategy on a stirred anaerobic sequencing fed-batch
reactor containing immobilized biomass. Bioresource Technology 90,
199e205.
Ripley, L.E., Boyle, W.C., Converse, J.C., 1986. Improved alkalimetric monitoring for
anaerobic digestor of high-strength wastes. Journal Water Pollution Control
Federation 58, 406e411.
Shiwen, C., Lin, Y., Zhineng, H., Xianzu, Z., Zhaolu, Y., Huidong, X., Yuanrrong, L.,
Rongdi, J., Wenhua, Z., Fangyuan, Z., 1990. Cadmium exposure and health effects
among residents in an irrigation area with ore dressing wastewater. The Science
of the Total Environment 90, 67e73.
Speece, R.E., 1996. Anaerobic Biotechnology for Industrial Wastewaters. Vanderbilt
University, Archae Press, Nashville, Tennessee.
Standard Methods for the Examination of Water and Wastewater, 2005. APHA,
AWWA, WPCF, 21th ed. American Public Health Association, Washington.
Stohs, S.J., Bagchi, D., 1995. Oxidative mechanisms in the toxicity of metal ions. Free
Radical Biology and Medicine 18, 321e336.
Nomenclature
Symbols
C: organic matter concentration (mg O2 L1)
CD: dissolved organic matter concentration (mg O2 L1)
CT: total organic matter concentration (mg O2 L1)
CCritical: critical cadmium concentration (mg Cd2þ L1)
Ci: inhibitor concentration (mg Cd2þ L1)
Cinitial: initial organic matter concentration at a HRT of 0 h (mg O2 L1)
Cresidual: residual organic matter concentration (mg O2 L1)
Bio : estimated bioavailable cadmium concentration (mg Cd2þ L1)
CCd
Bio P : estimated bioavailable cadmium concentration from previous condition
CCd
(mg Cd2þ L1)
F : free cadmium concentration (mg Cd2þ L1)
CCd
T : total cadmium concentration (mg Cd2þ L1)
CCd
2þ 1
ðCCd
T Þ
effluent : effluent total cadmium concentration (mg Cd L )
T Þ 2þ 1
ðCCd influent : influent total cadmium concentration (mg Cd L )
Cx: biomass concentration (mg TVS L1)
HRT: hydraulic retention time (h)
HRTS: hydraulic retention time in sampling point (h)
1
kapp
1 : apparent first-order kinetic coefficient (h )
ki : enzyme-based competitive inhibition coefficient (mg L1)
app
kS : half-saturation constant (mg L1)
mm: immobilization matrix mass (mg support)
mS: sample mass (mg)
m105 
S : sample mass after 24 h at 105 C drying (mg)

m550
S : sample mass after 2 h at 550 C calcination (mg)
2þ
mCd
S : total cadmium mass content in sample (mg Cd )
mCd Bio : bioavailable cadmium content in sample (mg Cd2þ)
S
84 G. Mockaitis et al. / Journal of Environmental Management 106 (2012) 75e84

mT: immobilization matrix and biomass total mass (mg) VBioreactor: bioreactor volume (L)
Q: bioreactor inlet/outlet mean flow (L h1) F : free cadmium removal efficiency
YCd
toperation: total time of condition (h) T : total cadmium removal efficiency
YCd
TA: total alkalinity (mg CaCO3 L1)
TS: total solids concentration (mg TS mg support1) YD: dissolved organic matter removal efficiency
TVA: total volatile acids (mg Hac L1) YT: total organic matter removal efficiency
TVS: total volatile solids concentration (mg TVS mg support1) Yx/S: substrate to biomass yielding factor
TVSTCd : total cadmium concentration in biomass (mg Cd2þ mg TVS1) aBio: bioavailable cadmium mass fraction on biomass
TVSBio
Cd : bioavailable cadmium concentration in biomass (mg Cd
2þ
mg TVS1) mmax: maximum specific growth rate (h1)