International Biodeterioration & Biodegradation 86 (2014) 317e326

Contents lists available at ScienceDirect

International Biodeterioration & Biodegradation

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

Accelerated biodegradation of cured cement paste by Thiobacillus

species under simulation condition
Azam Yousefi a, b, Ali Allahverdi a, *, Parisa Hejazi b
a
Cement Research Center, Iran University of Science and Technology, Tehran 1684613114, Iran
b
Biotechnology Research Laboratory, School of Chemical Engineering, Iran University of Science and Technology, Tehran 1684613114, Iran

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

Article history: Biodegradation is one of the most important types of cement deterioration. Complex microbial pop-
Received 24 July 2013 ulations take part in the biodegradation process of cement-based materials. Studies in this field show
Received in revised form that the sulfur-oxidizing bacteria, including Acidithiobacillus thiooxidans, due to sulfuric acid formation,
22 September 2013
play a key role in this process. In this study, with the accelerated leaching process of calcium hydroxide of
Accepted 6 October 2013
Available online 12 November 2013
cement paste, cured under running tap water and exposed to sterile biogenic sulfuric acid for 6 days, the
surface pH of the cement was reduced to a more favorable level for bacterial growth. In this case, the
growth of Thiobacillus proceeded in the presence of cured cement paste specimens. After 90 days of
Keywords:
Biodegradation
exposure to a semi-continuous culture of A. thiooxidans with its pH less than 2 and continuous removal of
Cement paste damaged layers the compressive strength, length and mass of the samples dropped by 96%, 11% and 43%,
Biogenic sulfuric acid in the order given. The mechanism of degradation and the structure of degraded specimens were
Thiobacillus thioparus analyzed by test laboratory techniques such as, XRD, SEM and EDAX analyses.
Acidithiobacillus thiooxidans Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction
Some microorganisms can grow on cement surfaces though
their thin biofilms are invisible. They are not responsible for colored
spots usually observed on the surface of concrete structures, but
they have the ability to destroy them (Escadeillas et al., 2007). Some
types of autotrophic and heterotrophic microorganisms, notably
genus Thiobacillus, as a group of bacteria with serious destruction
ability, were isolated and identified from the degraded concrete
structures (Nica et al., 2000; Okabe et al., 2007). Most research
reports have indicated that Acidithiobacillus thiooxidans plays a key
role in the biodegradation process of cement (Milde et al., 1983;
Sand and Bock, 1984; Diercks et al., 1991; Haile and Nakhla, 2009;
Wei et al., 2010). These bacteria are available in water, air and soil
(Waksman, 1922; Vidyalakshmi et al., 2009) and still the best and
most suitable growth environment is found to be in sewage pipe-
lines (Roberts et al., 2002), so any cement structure located nearby
undergoes biodeterioration. These bacteria have the ability to
oxidize organic and inorganic sulfur compounds in the presence of
oxygen, carbon dioxide and moisture (Waksman, 1922; Wei et al.,
2010). The final product of sulfur-oxidizing bacterial activity is
* Corresponding author. Tel.: þ98 21 77240496; fax: þ98 21 77240495.
E-mail addresses: ali_averdi@yahoo.com, ali.allahverdi@iust.ac.ir (A. Allahverdi).
sulfuric acid (Diercks et al., 1991; Roberts et al., 2002). A white
gypsum layer is found by reaction of sulfuric acid and cement
portlandite [Ca(OH)2]. The reaction of gypsum and aluminate phase
of the cement produces Ettringite (3CaO$Al2O3$3CaSO4$32 H2O),
with expandability and low adhesion properties (Saricimen et al.,
2003; Connell et al., 2010). With sulfuric acid formation by sulfur-
oxidizing bacteria, the strong structure of cement is converted to
a loose structure of gypsum and Ettringite at medium temperature
(>15
C) and Thaumasite (CaSiO3$CaCO3$CaSO4$15H2O) at low
temperature (<15
C) (Cwalina, 2008).
First, Thiobacillus does not exhibit any sign of growth on con-
crete structures at initial pH of around 12e13 (Roberts et al., 2002),
though from the beginning, some algae and fungi grow on cement
surfaces as a suitable substrate (Diercks et al., 1991; Jayakumar and
Saravanane, 2010; Wiktor et al., 2010). At longer time, the
carbonation and leaching processes of cement portlandite occur by
climatic changes such as snow, rain and other running water and
the pH of cement surface are reduced to about 9 (Shook and Bell,
1998). Thiobacillus thioparus is the first acting species of Thio-
bacillus bacteria that has the ability to grow on the cement surface
at pH 10. By production of polythionic acid, elemental sulfur and
sulfuric acid, the pH of cement surface is dropped again and
therefore other Thiobacillus strains start to grow. At pH 4.5, the
various strains of A. thiooxidans find the ability to grow on the
surface of cement (Roberts et al., 2002). By activity of these bacteria

0964-8305/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ibiod.2013.10.008
318 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326
under suitable conditions, pH of the environment is dropped to less
than 1. These bacteria have been found abundant in the 1e5 mm
layer of concrete structures, but in deeper layers, their numbers are
reduced logarithmically due to lower diffusion of oxygen and car-
bon dioxide gases of the air from the surface cement. So, sulfuric
acid is produced on the concrete surface and penetrates into the
lower layers and reacts with cement inner compounds (Nica et al.,
2000; Yamanaka et al., 2002; Wiktor et al., 2010).
There are good review articles (Diercks et al., 1991; Monteny
et al., 2000; Gaylarde et al., 2003; Cwalina, 2008; Connell et al.,
2010) in relation to cement biodegradation. Considerable research
(Magniont et al., 2011) works have been conducted on the
biodegradation of concrete-based structures by synthetic organic
acids (acetic and lactic acids) and comparisons are made with
biogenic acid produced by Escherichia coli and silage effluent
(acetic, propionic, butyric and isobutyric acids) (Bertron et al.,
2005a,b, 2007). Biodegradation mechanism of cement structure is
different in domestic and industrial sewage systems from that of
agricultural silos, etc.; because of different microorganisms pro-
ducing different acids. In agricultural and agro-food effluents, it is
shown that some ionic salts and metals of cement are consumed by
bacteria. Also it is particularly observed that bacteria cause more
intense deterioration of cement compared to a medium without
bacteria or in a synthetic acid solution (Magniont et al., 2011). In
other studies chemical sulfuric acid (Chandra and Berntsson, 1983;
Knight et al., 2002; Saricimen et al., 2003; Hewayde et al., 2007;
Herisson et al., 2013) or sodium and magnesium sulfates
(Monteny et al., 2000) are found to simulate the biodegradation
process of cements, and it is shown that cement degradation with
chemical sulfuric acid and sulfate salts take different course from
biogenic sulfuric acid bacterially produced, even though the so-
dium and magnesium cations play important roles in degradation
of cement. Some research works have focused their studies on
presenting a model for biodegradation (Kaempfer and Berndt,
1999; Vollertsen et al., 2008; De Windt and Devillers, 2010) pro-
cess by in situ tests (Monteny et al., 2000; Okabe et al., 2007; Alum
et al., 2008; Herisson et al., 2013) and by a mixture of several
bacteria (Sand and Bock, 1984; Wei et al., 2010; Herisson et al.,
2013), or just in presence of one bacteria such as A. thiooxidans
(Hormann et al., 1997; Vincke et al., 1999; Knight et al., 2002; Haile
et al., 2008) and finally others have dealt with biodegradation of
concrete structures using algae (Ismail et al., 1993; Bertron et al.,
2007; Escadeillas et al., 2007, 2009; Alum et al., 2008; Jayakumar
and Saravanane, 2010; Wiktor et al., 2010) and fungi (Gaylarde
et al., 2003; De Windt and Devillers, 2010). In these studies, a va-
riety of media are used including sulfur powder and thiosulfate
with calcium, iron, aluminum and magnesium ions with the con-
crete, mortar and cement paste specimens in different shapes and
sizes under variable temperature and humidity conditions. Scat-
tered research data with various microorganisms and methods
have produced complications to reach a unified conclusion for
comparing the observations and data by various experiments on
cement biodegradation. So, in-depth understanding of various
interactive processes in structural biodegradation of cements and
their constant changes are necessary.
In the present work, the growth of two species of Thiobacillus is
studied under normal laboratory conditions in liquid media, free
from ions in common with cements such as calcium, iron and
sulfates, in which bacteria become fully colonized. In retrospect, the
behavior of Thiobacillus has been assessed in the presence of a
cement paste. The reduction in the pH of liquid medium to less than
2 leads to degradation in cement chemical structures. The study
was further continued to evaluate the biodegradation of the cement
in curing and degradation stages by A. thiooxidans in simulated
laboratory conditions.
2. Materials and methods
2.1. Materials
2.1.1. Microorganisms and cultivation media
Two oxidizing-sulfur bacteria T. thioparus PTCC 1668 and
A. thiooxidans PTCC 1717 were purchased from Persian Type Culture
Collection, Iranian Research Organization for Science and Tech-
nology (IROST). The optimum pH for growth of the former was 7,
while it was 4.5 for the latter microorganism at 30
C. These mi-
croorganisms were kept in refrigerator at 4
C in a liquid medium
containing thiosulfate and they were re-cultivated once every 2e3
weeks. Liquid culture media for T. thioparus PTCC 1668 and
A. thiooxidans PTCC 1717 (designated as M1 and M2 in Table 1)
contained mineral salts (Merck Co., Germany) free of calcium, iron,
aluminum, and sulfate ions. The optimum culture media for bac-
terial growth and their corresponding data are not published yet.
The culture media were sterilized at 121
C under 1.2 bar pressure
for 20 min. The optimized volume of the inoculum in their culture
media was 1 v/v(%) towards the end of the logarithmic phase of
bacterial growth.
Symbols for various tests are presented in Table 2.
2.1.2. Cement paste specimen preparation
Portland cement was of Type II with chemical composition as
presented in Table 3 and physical properties of Blaine Fineness of
302 m2/kg and density of 3120 kg/m3.
Cement paste was prepared by tap water of water/cement ratio
of 0.35 and molded in 2  2  2 cm3 cube blocks. The specimens
were stored in an environment with high humidity 95% and room
temperature for 1 day and molded and treated under curing
condition.
2.2. Methods
2.2.1. Simulation experiments
After removing the specimens from the molds, in order to
simulate ion leaching and pH reduction processes, the specimens
were exposed to running tap water for 27 days for hydration of
cement phases and leaching out the portlandite from their surface
at room temperature. Then the specimens were exposed to sterile
biogenic sulfuric acid with pH less than 2 and at room temperature
for 6 days. This acid was obtained from the growth of A. thiooxidans
PTCC 1717 in M2 medium after two days at shaker incubator con-
ditions (30
C, 150 rpm). The culture growth resulted in a pH that
was in agreement with a report given for a 20-year old sewer
condition (Kaempfer and Berndt, 1999). In all the tests, the volume
ratio of biogenic sulfuric acid to cement paste specimens surfaces
was equal to 5.
After simulation process and reduction of pH of the cement
paste to its required level, the biodegradation of specimens was
studied through different pathways of slow (1) and accelerated (2)
Table 1
Components of liquid culture media for T. thioparus PTCC 1668 (M1) and
A. thiooxidans 1717 PTCC (M2).
Mineral salt (g)/1 l distilled water M1 M2
NH4Cl 0.40 2.43
MgCl2$6H2O 0.20 0.41
Na2CO3 0.40 0.00
K2HPO4 2.00 0.00
KH2PO4 2.00 3.00
Na2S2O3$5H2O 5.00 5.00
 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326 319

Table 2 ATO CM2 CATO T= 30 oC

Symbols designated for various culture media and their bacterial growth tests.
100 rpm
Symbol Test V/S = 5
M1 Culture medium of T. thioparus PTCC 1668
M2 Culture medium of A. thiooxidans PTCC 1717
TTP Growth test for T. thioparus in M1
ATO Growth test for A. thiooxidans in M2
CM1 Leaching test of cement paste specimen in M1
CM2 Leaching test of cement paste specimen in M2
CTTP Cement paste specimen exposed to T. thioparus in M1 Fig. 1. Schematic pathway of slow test (ATO: microorganism growth control, CM2:
CATO Cement paste specimen exposed to A. thiooxidans in M2 leaching process control and CATO: main test of bacterial growth on cement paste).

specimens every 10 days to measure the specimens’ compressive

strength, mass and dimensional status.
degradations. The surface microorganisms of cement paste speci-
mens in all the tests were inactivated by immersion in 70e75%
2.2.2. Measurement of degradation intensity of cement paste
ethanol bath for 1e2 days and exposed to UV irradiation of 20 W for
The properties of cement pastes such as compressive strength
1 h and were washed completely with sterilized distilled water.
(MPa), mass (g) and length (mm) were studied after 28-day curing
Pathway (1): By dropping the surface pH to 7, at sterile condi-
and at different exposure times to culture liquid containing
tions, the specimens were exposed to 1% inoculum of T. thioparus
A. thiooxidans. The percentage of changes in properties of the
PTCC 1668 (CTTP test) culture growth followed by A. thiooxidans
specimen in a 90-day period was determined using Eq. (1).
PTCC 1717 (CATO test). Each test was comprised of 2 control sam-
ples; one for bacterial growth (ATO and TTP tests) and the other for  
X0  X
leaching process (CM1 and CM2 tests) according to Table 2. DXð%Þ ¼  100 (1)
Fig. 1 shows a simplified schematic pathway (1) for X0
A. thiooxidans PTCC 1717. One drop of reagent of bromophenol blue
where
(0.01% w/v in distilled water) was used as an internal indicator for
X0: Initial properties of cement paste at the end of curing
culture medium which turned purple at pH > 3.94 and cream at
process.
pH < 3.94 and with a further decrease in pH it appeared yellow.
X: Measured properties of cement paste, after exposure to
Changes in the color of culture medium were indications of the
A. thiooxidans culture, every 10 days up to 90 days.
quality of the bacterial growth. A similar procedure was used for
T. thioparus PTCC 1668 in M1 culture medium without an indicator.
DX (%): Reduction percentage in properties in cement paste after
exposure to A. thiooxidans culture, every 10 days up to 90 days.
Although this pathway took a long time to evaluate the biodegra-
dation of cement in laboratory, but it was indispensable in studying
2.2.2.1. Measuring the compressive strength of the specimens.
the behavior of bacteria on cement paste. The tests were repeated
The surfaces of the cement paste specimens were cleaned after
twice.
reaching the desired age. For accurate measurement of the force the
Pathway (2): This pathway was an accelerated biodegradation
vertical load was applied on one side of the specimen which had
method on cement paste.
been in contact with the wall of the mold. The amount of applied
Due to slow growth of A. thiooxidans and in order to exhibit
force was measured by a hydraulic press machine until a crack
accelerated biodegradation of cement specimens, a semi-
appeared on the specimen. By dividing this force to the specimen
continuous culture of these bacteria with pH < 2 was used at
surface, its compressive strength was obtained. The recorded
30
C. As in Fig. 2, the new specimens were exposed to biogenic
compressive strength was an average of 3 different specimens.
sulfuric acid produced by A. thiooxidans PTCC 1717, with the pH of
about 2, of 5/1 acid volume/cement surface. The change in the color
2.2.2.2. Measuring the mass of the specimens. The mass of each
of the culture medium with respect to increases in pH was attrib-
cement paste specimen was taken by a digital scale with an accu-
uted to the predominant leaching process over bacterial growth,
racy 0.01 g, after having been kept in an oven of 60
C for 1 day for
which was an indication of lower biodegradation rate. So by in-
evaporation of surface water. The specimens were cooled in a
creases in pH of the culture medium and its sudden color changes
desiccator for 1 h and their surfaces were brushed to remove the
from yellow to purple, the microbial culture was immediately
degraded layers. The recorded mass was an average of 3 specimens.
replaced by a fresh one. Also, simulation was made by brushing the
The specimens were placed in the oven after being exposed to
degraded layers of cement paste every 10 days based on actual field
bacterial activity and kept in a 100% power ultrasonic bath for
circumstances in which the sewage flow removes the cement
15 min, at 30
C and 59 kHz frequency to remove the white loose
degraded layers. The whole trial lasted 90 days by removal of 9
layer from their surfaces.

Table 3
2.2.2.3. Measuring the length of specimens. The measured length of
Chemical composition of Type 2 Portland cement.
each cement paste specimen, from a specified point with a preci-
Chemical composition Concentration (% w/w) sion caliper 0.01 mm, was an average of 3 specimens, under the
CaO 63.26 condition described in Section 2.2.2.2.
SiO2 22.50
Al2O3 4.15
2.2.3. pH measurement
Fe2O3 3.44
MgO 3.25 The pH of the bacterial suspensions and solutions containing
SO3 1.80 cement paste specimens was determined by a Percisa pH-meter,
K2O 0.65 Switzerland. To measure the surface pH of the cement paste
Na2O 0.20 specimens, each one was placed in 100 ml of distilled water for 1 h
Free-CaO 0.72
LOI 0.61
to achieve a stable pH for the solution. This test was repeated on 3
identical specimens to obtain an average pH value.
320 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326
Fig. 2. Schematic pathway of accelerated biodegradation test of cement paste specimens.
2.2.4. Measuring the sulfate concentration
To measure the concentration of sulfate ions produced by sulfur-
oxidizing bacteria, a method based on barium chloride solution and
turbidimetry measurements (Kolmert et al., 2000) were used. In
this method, barium sulfate precipitate was formed and the culture
turbidity was measured at wavelength of 420 nm by a Metertech
UVeVis spectrophotometer (SP8001 model, Taiwan).
2.2.5. Cell optical density
Microbial culture turbidity measurement, as an indirect method,
was employed to measure bacterial cell concentration. UVeVis
spectrophotometer was used at 600 nm to measure the optical
density of cells (OD600) which was correlated with the rate of
bacterial growth.
2.2.6. Changes in mineral phase composition
To study the changes in mineral phase composition of degraded
layers of the cement paste specimens, a Bruker X-ray diffractometer
(XRD, Cuka, l ¼ 0.154 nm) was used.
2.2.7. Microscopic structural changes
The microstructure of the biodegraded cement paste was inves-
tigated using a TESCAN scanning electron microscopy (model VEEA11,
Czech), and energy dispersive analysis by a TNCA EDAX X-ray (Oxford,
England). The linear elemental analyses of biodegraded specimens
were carried out by EDAX to identify the changes made in their
elemental compositions. The samples were placed in an oven of 60
C
for 3e4 days. After cooling, they were gold coated in a deep vacuum.
3. Results and discussion
3.1. Curing process and pH reduction in cement paste
A 27-day curing process of the specimens resulted in hydration
of different phases of cement and reduction of their pH which were
Fig. 3. Variations in pH and sulfate ion concentration in various suspensions versus time (TTP: microbial control, CM1: leaching process control and CTTP: actual test).
accomplished by running tap water. In this process, the pH of the
specimens at around 12 dropped to about 10 at the rate of
about 0.089 pH/day. To speed up the cement biodegradation, the
surface pH of the specimens was reduced to 7.29 at a rate of
about 0.45 pH/day by their exposure to sterile biogenic sulfuric
acid at pH of approximately 2 for 6 days.
3.2. Thiobacillus growth on cement paste through pathway (1)
At first, the cured cement paste specimens were exposed to
T. thioparus culture for 13 days. By further decrease in pH, they were
exposed to A. thiooxidans growth. In each experiment, two groups
of control samples including biologically controlled tests (TTP and
ATO) and leaching controlled tests (CM1 and CM2) were used.
The sulfate and hydrogen ions were produced due to oxidation
of sodium thiosulfate in presence of oxygen and sufficient moisture
in TTP test (a microbial control) by T. thioparus as in Eq. (2). In this
process, the pH of solution was reduced to 4.5, from its initial value
of 7, after 6 days shown in Fig. 3. In CM1 control test of portlandite
leaching the pH increased from 7 to about 8. In CTTP procedure, as
the actual test, comprised of T. thioparus bacteria and cement paste,
the overall trend in pH variation was similar to microbial control
test, but once the microorganism was in its lag phase (initial 4-day)
there was a slight increase in pH which occurred due to the
leaching process. At bacterial growth phase, entering the loga-
rithmic phase of 4e6 days, the pH was reduced to 5. Beyond 6 days,
at stationary phase of bacterial growth the leaching of calcium
hydroxide was the dominant process and, therefore, the pH of the
medium followed an increasing trend.
Na2 S2 O3 þ 2O2 þ H2 O/Na2 SO4 þ H2 SO4 (2)
The increasing trend in sulfate ion concentration is shown in
Fig. 3. It is evident that sulfur-oxidizing bacteria have consumed
thiosulfate ions in the culture medium as a source of energy, and
converted them into sulfate ions (TTP test). In control test of
 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326
leaching process (CM1), the sulfate ion concentration is negligible
due to low concentration of sulfate ions in initial cement paste, and
so a slight increase is observed in sulfate concentration. In CTTP
test, the production of sulfate is clearly evident due to leaching
process of cement and bacterial growth. In contrast the production
of sulfate ions in this test is higher (w70 mM) than what it had been
expected (40e50 mM), and this could be due to the consumption of
cement’s other components by T. thioparus and further formation of
sulfate ions from the cement paste. Formation of calcium, iron and
aluminum ions of cement, exposed to sulfuric acid produced by
A. thiooxidans, is less than the same test with chemical sulfuric acid
(Knight et al., 2002), due to differences in chemical and biological
sulfuric acid attacking mechanisms on cement paste, and/or it may
be linked to consumption of calcium, iron and aluminum ions by
A. thiooxidans in cement.
With further reduction of pH to w5, A. thiooxidans was able to
grow and colonize in cement. Fig. 4 shows variations in pH of M2
culture medium versus time in solutions with and without
A. thiooxidans and the cement paste specimen. In CM2 control test,
with no bacteria, due to leaching process of portlandite within
cement specimen, the pH of solution increased over time and its
amount increased from 4.52 to w6.5. In ATO test, as the microbial
control, due to oxidation of thiosulfate ions to hydrogen and sulfate
ions, the pH was lowered from 4.52 to 1.85 after 5 days. In CATO
test, there were two opposing actions of bacterial growth (reduc-
tion in pH) and leaching process of portlandite (increase in pH).
Once the bacteria were in their lag phase, a slight increase occurred
in pH due to portlandite leaching process and when at their loga-
rithmic phase (day 2e5) the pH decreased to w2.5. After 5 days,
when the bacteria entered their stationary phase, the leaching
process of portlandite overcame this phase, so the pH of the solu-
tion increased again.
The maximum turbidity of bacterial growth was found in ATO
test with its OD600 ¼ 0.25. Cell turbidity of the suspension of
leaching process control (CM2) was negligible. The cell turbidity of
CATO test (bacterial growth on cement paste specimen) was very
low as well, whereas due to the presence and growth of bacteria,
the OD600 was expected to be at least 0.2. This could be due to
adherence of bacteria cells on the surface of the cement. After 8
days, the specimen surface, exposed to A. thiooxidans growth,
appeared as a soft, white and thin porous layer, but the specimen
that had been placed in solution M2 looked unchanged. In fact, this
layer seemed to have provided a suitable substrate for bacteria to
adhere strongly to cement surface and therefore, resulting in lower
turbidity of solution.
After these biodegradation experiments, the specimens, placed
in M1 and M2 media, were examined for their final compressive
strength, length and mass values. The results of the measurements
Fig. 4. Variations in pH of M2 culture medium versus time in solutions with and
without A. thiooxidans and a cement paste specimen (ATO: microbial control, CM2:
leaching process control and CATO: actual test).
321
indicated that the properties of specimens were kept intact.
Therefore, M1 and M2 media were suitable for the growth of genus
Thiobacillus on cement pastes. Under normal conditions, however,
the cement pastes in neutral states were expected to show higher
compressive strength due to their hydration process, but due to
natural leaching of portlandite from inside the cement into the
culture media within such short testing periods of 8 and 13 days
there were no detectable changes observed in the general proper-
ties of the specimens.
3.3. Accelerated biodegradation of cured cement pastes by
pathway (2)
The results of the previous section showed that genus Thio-
bacillus grew well on a cured cement paste resulting in formation of
biogenic sulfuric acid. This acid showed complex damaging effects
on cement structure. The rate of sulfuric acid production by
pathway (1) was very slow and the assessment procedure of the
changes in cement structure took much longer. Therefore, to speed
up the rate of cement biodegradation and the subsequent assess-
ments the simulated pathway (2) of Section 2.2.1 was adopted.
The results of the accelerated biodegradation of cement pastes
by A. thiooxidans PTCC 1717 culture with pH less than 2 are shown
in Fig. 5. According to Fig. 5a, the degradation depth of the speci-
mens during 90 days is measured 2.5 mm, which is equivalent to
11% reduction in the specimens length. By its extrapolation to one
year, the annual rate of cement paste biodegradation is found to be
10 mm/y. However, the real annual rate of biodegradation of the
concrete is reported as 2e4.7 mm/y (Roberts et al., 2002). Also the
severe degradation of Hamburg concrete sewer pipes is reported
6 cm after 6 years, although the pipeline had been coated by epoxy
resin with 200e300 mm thickness (Sand and Bock, 1984). The
degradation depth of the pipeline contained 50% gypsum and the
surface pH value of the concrete was reported between 1 and 2. In
the same sewage system, degradation took place on concrete while
in our current investigation the degradation has been evaluated on
cement paste and this latter location has displayed a considerable
effect on the degradation rate. As far as we are aware, although no
research is carried out under the exact similar condition as others,
but it seems that our final results and deductions might be com-
parable with the data reported by Hamburg sewage system. In
accelerated methods, the real degradation occurs faster, so the
evaluation processes and their reproducibility are easier, though
the natural degradation rate can be different from an accelerated
degradation rate, therefore, a fixed annual rate cannot be reported
for biodegradation of cement and concrete, because the degrada-
tion rate depends on different parameters such as the quantity of
sulfur compounds, moisture, turbulency and sewage flow rate,
length of pipeline and temperature, etc.
The temperature of 30
C, compared to an ambient condition,
was an optimum at which bacterial growth reached its maximum
and it accelerated the rate of cement biodegradation. It is evident in
Fig. 5b the cultivation of A. thiooxidans PTCC 1717 at a pH below 2
with acid volume to cement surface ratio of 5 to 1, its sudden
change to a pH of w4 and subsequent replacement with the initial
cultivation, and cleaning of loose porous layers in an ultrasonic
bath, all contributed in a relatively high drop in mass of the cement
paste specimen. The mass reduction of 43% after 90 days is clearly
much higher compared to other researchers’ reports listed in
Table 4. It should be noted that these researchers employed con-
crete and mortars which certainly influenced the degradation rate,
due to aggregates which must have shown high resistance to bac-
terial attack. Cleaning in an ultrasonic bath (Hormann et al., 1997)
and drying in an oven of 60
C for 1 day before weighing the
specimens accounted for high mass loss of the present study,
322 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326

Fig. 5. Changes in testing parameters: (a) length (b) mass and (c) compressive strength of a cement paste specimen during a 90-day test period, and (d) reduction in compressive
strength versus mass.

compared to other studies which just use an ordinary brush. It may
be deduced that there are certainly no fixed rates observed for
annual mass and dimensional reductions of cement as they vary
according to the factors mentioned above. In fact, the annual mass
reduction rate is a strong evidence for biodegradation of cement
structures and it is a justification for the protective measures which
should be taken against degradation of cement-based materials.
Fig. 5c shows the changes in compressive strength of cement
paste specimens versus time. The further reduction of compressive
strength (96%) of cement paste compared to their mass and length
reductions can be due to dimensional deformation of specimens
from their initial cubic shape. This deformation in the shape of
specimens prevents a uniform force to be applied on their surfaces
during compressive strength measurements. Also cement chemical
components involved in chemical reactions in acidic solution are
very intense, forming expansion products in surface areas of the
specimens, causing internal tensions and weakening mechanical
strength. After one month passed from the beginning of the test
period, it occasionally happened that specimens were broken un-
expectedly when attempts were made to brush off the thin porous
layers for final preparation of their mass and length measurements.
Fig. 5d indicates an approximate linear relationship in reduction
of mass and compressive strength of cement specimens during
90-day test period. The slope of the fitted line for compressive
strength and mass reductions is found to be 8.3 MPa/g.
The above tests were completed by visual observations of
specimens. In the first and second 10-day periods, specimens
showed a normal appearance without any change in color and
dimension, though there were reductions noticed in length by
1.03% (Fig. 5a), mass by 2.74% (Fig. 5b) and compressive strength by
26.7% (Fig. 5c). Reduction of length by 1.03% was not detectable by
naked eye after 20 days. The mass reduction of 2.74% in the same
interval was due to leaching processes of cement components,
especially the portlandite into the acidic bacterial solution.
Although the reduced mass and length of the specimens were not
correlated with any change in their appearance to be detected by
naked eye, but these reductions sufficiently agreed with the
reduction of compressive strength by 26.7%. After 30 days from the
start of experiment of accelerated biodegradation process, the
formation of a white, soft and low density layer with low adhesion
was observed on the surfaces of specimens. The layers were easily
separated from the surface of the specimens, as soon as they were
removed from bacterial acidic solution and exposed to ambient air
or placed into the oven of 60
C with severe cracks developed on
the white layers which dropped from the surface instantly. In
addition, when in another test the dried specimen and a control

Table 4
The reduction in mass and thickness of cement-based specimens reported in literature.

Specimen type Thickness Mass reduction (%) Experiment Bacteria Ref.

reduction time (day)

Concrete 0.75e0.8 mm 9e11 51 T. neapolitanusT. Thiooxidans T. ntermedius/T. novellus (Vincke et al., 1999)
3.8 mm 13.5 360 T. thiooxidans (Ismail et al., 1993)
e 5.8 270 Thiobacillus neapolitarius, T. intermedius and T. thiooxidans (Sand et al., 1984)
50% 35 365 Sewer atmosphere (Okabe et al., 2007)
Mortar e 18e31 150 T. thiooxidans (Hormann et al., 1997)
Cement paste e 5.7 90 Thiomonas perometablis T. thiooxidans A. thiooxidans and Wei et al., 2010)
T. ferrooxidans T. intermedia and T. peromotabolis Probe
11% 43 90 A.thiooxidans Current research
 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326 323

specimen (not having been in biogenic sulfuric acid solution) were

put into the water, plenty of bubbles were rapidly developed from
the degraded specimen up to about 1 h. These bubbles from the
degraded specimens were attributed to the leaching process of the
internal chemical components of the cement and formation of
porous and tiny ducts within it. This phenomenon and separation
of layers persisted till the end of the accelerated biodegradation
period. In images of Fig. 6, there are changes in the appearance of
the specimens representing the degraded paste after 60 days
exposure to cultivation of bacterial growth at pH below 2,
compared to a pure cement paste. The changes in the appearance of
the specimens during the sixth to ninth testing periods are clearly
visible to the naked eye.
By visual assessment of the cross-section of degraded speci-
mens, 3 distinct regions were noticed as the followings: first, a
completely degraded white thin layer on the surface, lacking any
cementitious property; the second region, consisting of a moder-
ately thin layer with cement property and distinct bright color
relative to other intact parts of the specimen and the third region
was the remaining part that seemed completely safe. The sampling
and XRD analysis were performed on each region to study the
phase constituents of degraded sections (surface white layer) and
the deteriorating layer (thin layer under the white region). The
results of this analysis are shown in Fig. 7. By comparing the XRD
data with standard patterns (33-306, 05-0586, 04-0733, 33-0311
and 41-1451) it appeared that the white loose layer formed on the
cement paste was mainly consisted of gypsum (33-0311).
The comparison of XRD patterns of the deteriorating layer with
standard patterns (04-0733 and 05-0586), it was revealed that the
dominant phases were indicated as calcite and portlandite and the
secondary phases consisted of calcium silicate hydrate and gypsum.
The specific Ettringite phase peaks were not observed at 2q ¼ 9.17

and 15.82
. The reason could be due to unstable Ettringite phase at

pH under 10.6 (Allahverdi and Skvara, 2000b) which because of its
Fig. 7. X-ray diffraction patterns of the surface layer, the deteriorating layer and the
close contact with biogenic sulfuric acid solution at pH below 2, it
intact part of cement paste specimen (G, gypsum; P, portlandit; C, calcite; CSH, calcium
decomposed to gypsum and aluminum hydroxide. Also Thaumasite silicate hydrate).
phase could not be produced at 30
C test temperature which was
higher than Thaumasite formation temperature (15
C).
The XRD pattern of the intact part of the samples showed that According to Fig. 7, the XRD patterns of the “deteriorating layer”
portlandite and calcium silicate hydrate were dominant phases are comparably different from the intact part of specimen. In this
compared to the secondary phase of calcite. Fig. 7 shows the layer, there is the reduction in the amount of portlandite phase and
fundamental changes in mineral phase constituents of biodegraded the high presence of calcite phase. These changes may be due to the
cement paste specimen with biogenic sulfuric acid. fact that in this region the atmospheric carbon dioxide has easily
The presence of calcite phase in the deep regions of cement paste reacted with portlandite which has led to increased calcite phase.
after 60 days of exposure to biogenic sulfuric acid was probably due The presence of a small percentage of gypsum in this layer may also
to either carbonation by atmospheric carbon dioxide during sample be attributed to the diffusion of a biogenic sulfate and its reaction
preparation or carbonation during exposure to biogenic sulfuric acid. with portlandite. It is noted that at cleaning stage some specimens
In fact, one can conclude that pathways must have been created for were broken before weighing. This could be due to internal ten-
diffusion of atmospheric carbon dioxide which reacted with por- sions created by gypsum crystals sediment in the surface regions of
tlandite present in deep parts of cement paste. the specimen leading to micro-cracks formation.

Fig. 6. The changes in the appearance of specimen by exposure to cultivation of A. thiooxidans with pH below 2 after 60 days (control (left) and biodegraded (right).
324 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326

In the white surface layer which was completely degraded, no
sign of portlandite, calcite and calcium silicate hydrate phases were
observed. At low pH, the physical and chemical balance of the
cement matrix was disrupted and its hydrated compounds started
to decompose. The first compound that began to dissolve and leach
out included the portlandite followed by decalcification of calcium
silicate, calcium aluminates and calcium aluminum ferrite hydrates
producing the amorphous hydrogels. Following the sulfuric acid
attack onto the cement paste, the final products consisted of cal-
cium sulfate and the hydrogels of silica and aluminum and ferric
oxide (Allahverdi et al., 2000a,b, 2005; Lajili et al., 2008).
In the degradation process due to biogenic sulfuric acid attack,
the diffusion of the acid ions into the specimen surface may facil-
itate reactions with the leaching calcium and the growing gypsum
crystals. As the degradation process proceeds, the gypsum layer
thickens further and it finally imposes a protective effect and
controls the rate of degradation process by diffusion phenomenon.
However, due to elimination of the gypsum layer in every 10-day
interval, a fresh surface of the cement paste came into contact with
the solution of biogenic sulfuric acid, so in total, the degradation
rates were controlled by the diffusion phenomenon combined with
the surface degradation processes.
Fig. 8 shows SEM images of cement paste specimen after 60 days
of exposure to biogenic sulfuric acid with different magnifications.
While preparing the samples for SEM and EDAX analyses, the
major part of the surface degraded layer that contained gypsum
was separated from the surface while drying the samples for 4 days
at 60
C. Despite of the gypsum presence which was confirmed by

Fig. 8. SEM images of cement paste specimen: (a) with 70 magnification and (b) 150, (c) its deteriorating layer with 1000 magnification and (d) 2000, (e) the boundary of its
deteriorating layer and (f) its intact part after 60 days of exposure to biogenic sulfuric acid.
 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326
XRD in the degraded surface layer, these crystals could not be easily
traced by SEM analysis on the remaining surface layer of the
specimens. The layer boundary line of the deteriorating and the
intact deep layers of the cement paste is clearly observed in Fig. 8b.
The present images in Fig. 8c and d show the deteriorating layer
with 1000 and 2000 magnifications. With regard to these images,
the vertical border between the deteriorating layer of at least
w100 mm thickness and the intact area is clearly visible. Fig. 8d
shows the deteriorating layer with higher magnification (2000)
having a structure covered with small and shattered pieces of
particles. Due to consumption of portlandite in the deteriorating
layer, there is no sign of its crystals present, though some scattered
microscopic cracks are evident for the progressive degradation of
the cement paste.
The images in Fig. 8e and f show the deteriorating layer and the
intact areas of cement paste clearly. The image on Fig. 8e shows
portions of the extended vertical cracks in the boundary of the
deteriorating layer and the intact areas. The image on Fig. 8f shows
the intact part of cement paste after 60 days of exposure to biogenic
sulfuric acid. The amorphous matrix of calcium silicate hydrates is
clearly visible inside the circle.
Fig. 9 indicates the EDAX linear elemental analysis of the cross-
section of cement paste specimen after 60 days of exposure to
biogenic sulfuric acid. The high sulfur content close to the surface is
a strong evidence for the presence of gypsum, as a residual
Fig. 9. EDAX Elemental analysis from the cross-section of the cement paste specimen
after 60 days exposure to biogenic sulfuric acid.
325
degraded layer, compared to other areas. So in the outermost layer,
the amount of silicon is low due to the formation and precipitation
of gypsum and because of its insolubility, silica gel, contrary to
portlandite, it cannot be leached out into the solution. But, ac-
cording to the results of element analysis of the deteriorating layer,
as the next layer, the amount of silicon is high while the amounts of
iron and aluminum are low. The increase in the amount of silicon is
due to the leaching portlandite and the reductions in the amounts
of leaching iron and aluminum can be due to the acidic pH of this
region which transfers iron and aluminum ions or other com-
pounds into the acid solution. In the third region, the furthest away
layer from biogenic sulfuric acid attacks, the amounts of calcium,
iron, aluminum and silicon are comparatively constant. In fact,
diffusion of the constituents of cement paste into the solution is
dependent on the pH of different regions. Therefore, calcium can
leach out from all parts of the specimen, because at pH below 12.6,
the portlandite begins to dissolve, while aluminum and iron of the
cement can enter into solution from the deteriorating layer and its
silica content is insoluble in all acidic conditions of the present
media. It may be theoretically stated that the pH increases from w2
of the surface layer to above 12 of deep intact region, although it is
not possible to measure the pH of different parts of specimen.
4. Conclusions
Biodegradation of cement is a slow process and its study needs
to be accelerated by simulation. In this research, by simulation of
sewage system conditions, i.e. using running tap water and semi-
continuous culture of A. thiooxidans PTCC 1717 bacteria, the
biodegradation of the cement paste has been investigated. The
results show that:
 T. thioparus PTCC 1668 and A. thiooxidans PTCC 1717 bacteria
have well grown in the presence of cured cement paste
specimens.
 Cement biodegradation is due to sulfuric acid production and
consumption of cement components by bacteria.
 Using a semi-continuous culture of A. Thiooxidans induces
intensive biodegradation of the cured cement paste specimens.
 By bacterial attack under simulated conditions during 90 days,
the compressive strength, length and mass of the cured cement
paste specimens are reduced by 96%, 11% and 43% in the order
given.
References

Allahverdi, A., Skvara, F., 2000a. Acidic corrosion of hydrated cement based mate-
rials, part 1. Mechanism of the phenomenon. Ceram. Silik. 44, 114e120.

Allahverdi, A., Skvara, F., 2005. Sulfuric acid attack on hardened paste of geo-
polymer cements, part 1. Mechanism of corrosion at relatively high concen-
trations. Ceram. Silik. 49, 225e229.

Allahverdi, A., Skvara, F., 2000b. Acidic corrosion of hydrated cement based mate-
rials, part 2 e kinetics of the phenomenon and mathematical models. Ceram.
Silik. 44, 114e120.
Alum, A., Rashid, A., Mobasher, B., Abbaszadegan, M., 2008. Cement-based biocide
coatings for controlling algal growth in water distribution canals. Cem. Concr.
Compos. 30, 839e847.
Bertron, A., Duchesne, J., Escadeillas, G., 2007. Degradation of cement pastes by
organic acids. Mater. Struct. 40, 341e354.
Bertron, A., Duchesne, J., Escadeillas, G., 2005a. Accelerated tests of hardened
cement pastes alteration by organic acids: analysis of the pH effect. Cem. Concr.
Res. 35, 155e166.
Bertron, A., Duchesne, J., Escadeillas, G., 2005b. Attack of cement pastes exposed to
organic acids in manure. Cem. Concr. Compos. 27, 898e909.
Chandra, S., Berntsson, L., 1983. Technical notes: influence of polymer microparti-
cles on acid resistance of structural lightweight aggregate concrete. Int. J. Cem.
Compos. Lightweight Concr. 5, 127e131.
Connell, M., McNally, C., Richardson, M., 2010. Biochemical attack on concrete in
wastewater applications: a state of the art review. Cem. Concr. Compos. 32,
479e485.
326 A. Yousefi et al. / International Biodeterioration & Biodegradation 86 (2014) 317e326
Cwalina, B., 2008. Biodeterioration of concrete. Archit. Civil Eng. Environ. 4, 133e
140.
De Windt, L., Devillers, P., 2010. Modeling the degradation of Portland cement
pastes by biogenic organic acids. Cem. Concr. Res. 40, 1165e1174.
Diercks, M., Sand, W., Bock, E., 1991. Microbial corrosion of concrete. Cell. Mol. Life
Sci. 47, 514e516.
Escadeillas, G., Bertron, A., Ringot, E., Blanc, P.J., Dubosc, A., 2007. Accelerated testing
of biological stain growth on external concrete walls. Part 1: development of
the growth tests. Mater. Struct. 40, 1061e1071.
Escadeillas, G., Bertron, A., Ringot, E., Blanc, P.J., Dubosc, A., 2009. Accelerated
testing of biological stain growth on external concrete walls. Part 2: quantifi-
cation of growths. Mater. Struct. 42, 937e945.
Gaylarde, C., Silva, M.R., Warscheid, T., 2003. Microbial impact on building mate-
rials: an overview. Mater. Struct. 36, 342e352.
Haile, T., Nakhla, G., 2009. Inhibition of microbial concrete corrosion by
Acidithiobacillus thiooxidans with functionalised zeolite-A coating. Biofouling
25, 1e12.
Haile, T., Nakhla, G., Allouche, E., 2008. Evaluation of the resistance of mortars
coated with silver bearing zeolite to bacterial-induced corrosion. Corros. Sci. 50,
713e720.
Herisson, J., van Hullebusch, E.D., Moletta-Denat, M., Taquet, P., Chaussadent, T.,
2013. Toward an accelerated biodeterioration test to understand the behavior of
Portland and calcium aluminate cementitious materials in sewer networks. Int.
Biodeterior. Biodegrad. 84, 236e243.
Hewayde, E., Nehdi, M.L., Allouche, E., Nakhla, G., 2007. Using concrete admixtures
for sulphuric acid resistance. Constr. Mater. 160, 25e35.
Hormann, K., Hofmann, F., Schmidt, M., 1997. Stability of concrete against biogenic
sulfuric acid corrosion, a new method for determination. In: Proceedings of the
10th International Congress on the Chemistry of Cement, Gothenberg. 4vi038.
Sweden.
Ismail, N., Nonaka, T., Noda, S., Mori, T., 1993. Effect of carbonation on microbial
corrosion of concretes. J. Constr. Manag. Eng. 474, 133e138.
Jayakumar, S., Saravanane, R., 2010. Biodeterioration of coastal concrete structures
by marine green algae. Int. J. Civil Eng. 8, 352e361.
Kaempfer, W., Berndt, M., 1999. Estimation of service life of concrete pipes in sewer
networks. Durab. Build. Mater. Compon. 8, 36e45.
Knight, J., Cheesemana, C., Rogersb, R., 2002. Microbial influenced degradation of
solidified waste binder. Waste Manag. 22, 187e193.
Kolmert, Å., Wikström, P., Hallberg, K.B., 2000. A fast and simple turbidimetric
method for the determination of sulfate in sulfate-reducing bacterial cultures.
J. Microbiol. Methods 41, 179e184.
Lajili, H., Devillers, P., Grambin-Lapeyre, C., Bournazel, J., 2008. Alteration of a
cement matrix subjected to biolixiviation test. Mater. Struct. 41, 1633e1645.
Magniont, C., Coutand, M., Bertron, A., Cameleyre, X., Lafforgue, C., Beaufort, S.,
Escadeillas, G., 2011. A new test method to assess the bacterial deterioration of
cementitious materials. Cem. Concr. Res. 41, 429e438.
Milde, K., Sand, W., Wolff, W., Bock, E., 1983. Thiobacilli of the corroded concrete
walls of the Hambourg sewer system. J. Gen. Microbiol. 129, 1327e1333.
Monteny, J., Vincke, E., Beeldens, A., De Belie, N., Taerwe, L., Van Gemert, D.,
Verstraete, W., 2000. Chemical, microbiological, and in situ test methods for
biogenic sulphuric acid corrosion of concrete. Cem. Concr. Res. 30, 623e634.
Nica, D., Davis, J.L., Kirby, L., Zuo, G., Roberts, D.J., 2000. Isolation and character-
ization of microorganisms involved in the biodeterioration of concrete in
sewers. Int. Biodeterior. Biodegrad. 46, 61e68.
Okabe, S., Odagiri, M., Ito, T., Satoh, H., 2007. Succession of sulfur-oxidizing bacteria
in the microbial community on corroding concrete in sewer systems. Appl.
Environ. Microbiol. 73, 971e980.
Roberts, D.J., Nica, D., Zuo, G., Davis, J.L., 2002. Quantifying microbially induced
deterioration of concrete: initial studies. Int. Biodeterior. Biodegrad. 49, 227e234.
Sand, W., Bock, E., 1984. Concrete corrosion in the Hamburg sewer system. Environ.
Technol. Lett. 5, 517e528.
Saricimen, H., Shameem, M., Barry, M.S., Ibrahim, M., Abbasi, T.A., 2003. Durability
of proprietary cementitious materials for use in wastewater transport systems.
Cem. Concr. Compos. 25, 421e427.
Shook, W., Bell, L., 1998. Corrosion control in concrete pipe and manholes. In: Annul
Conference and Exposition Water Environmental Federation, pp. 843e852.
Orlando, Florida.
Vidyalakshmi, R., Paranthaman, R., Bhakyaraj, R., 2009. Sulphur oxidizing bacteria
and pulse nutrition e a review. World J. Agric. Sci. 5, 270e278.
Vincke, E., Verstichel, S., Monteny, J., Verstraete, W., 1999. A new test procedure for
biogenic sulfuric acid corrosion of concrete. Biodegradation 10, 421e428.
Vollertsen, J., Nielsen, A.H., Jensen, H.S., Wium-Andersen, T., Hvitved-Jacobsen, T.,
2008. Corrosion of concrete sewersdthe kinetics of hydrogen sulfide oxidation.
Sci. Total Environ. 394, 162e170.
Waksman, S.A., 1922. Microorganisms concerned in the oxidation of sulfur in the
Soil: III. Media used for the isolation of sulfur bacteria from the soil. Soil Sci. 13,
329e336.
Wei, S., Sanchez, M., Trejo, D., Gillis, C., 2010. Microbial mediated deterioration of
reinforced concrete structures. Int. Biodeterior. Biodegrad. 64, 748e754.
Wiktor, V., Grosseau, P., Guyonnet, R., Garcia-Diaz, E., Lors, C., 2010. Accelerated
weathering of cementitious matrix for the development of an accelerated
laboratory test of biodeterioration. Mater. Struct. 44, 623e640.
Yamanaka, T., Aso, I., Togashi, S., Tanigawa, M., Shoji, K., Watanabe, T., Watanabe, N.,
Maki, K., Suzuki, H., 2002. Corrosion by bacteria of concrete in sewerage sys-
tems and inhibitory effects of formates on their growth. Water Res. 36, 2636e
2642.