biomass ash based products for sustainable construction

Journal of Cleaner Production 67 (2014) 117e124

Contents lists available at ScienceDirect

Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro

Sewage sludge/biomass ash based products for sustainable construction Primo
z Pav
si
c a, *, Ana Mladenovi
c b, Alenka Mauko b, Sabina Kramar b, Matej Dolenec c, c Vrta
c e, Peter Bukovec f Ernest Von
cina d, Katarina Pav
si
Building and Civil Engineering Institute ZRMK, Dimi
ceva 12, 1000 Ljubljana, Slovenia Slovenian National Building and Civil Engineering Institute, Dimi
ceva 12, 1000 Ljubljana, Slovenia c University of Ljubljana, Faculty of Natural Sciences and Engineering, A
sker
ceva 12, 1000 Ljubljana, Slovenia d Institute of Public Health Maribor, Prvomajska ulica 1, 2000 Maribor, Slovenia e University of Ljubljana, Veterinary Faculty, Institute for Hygiene and Pathology of Animal Nutrition, Gerbi
ceva 60, 1000 Ljubljana, Slovenia f University of Ljubljana, Faculty of Chemistry and Chemical Technology, A
sker
ceva 5, 1000 Ljubljana, Slovenia a

b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 29 May 2013 Received in revised form 10 December 2013 Accepted 11 December 2013 Available online 21 December 2013

The aim of this research was to determine how best to utilize two environmentally challenging types of waste: sewage sludge, and a particular type of waste ash, biomass ash, which is obtained from biomass combustion processes. The results of the performed research have shown that liquid sewage sludge can, in fact, be successfully stabilized with biomass ash, so that a stable composite material can be obtained, having a compressive strength within the range between 0.5 to 2.5 MPa, with “Controlled Low-Strength Material” properties. During the stabilization process, the microbial activity of sewage sludge is inhibited, due to raised pH levels and temperatures. Analysis of the chemical composition of water leachates from samples of the composite showed that it is inert, and thus does not pose a threat to the environment. The observed decrease, over time, in the concentrations of the pollutants indicated that the latter are immobilized in the hydrated matrix, due to the formation of new hydration products, i.e. mono- and hemi-carboaluminate and Friedel’s salt, and changes associated with pore diameter and distribution. The addition of selected types of recycled aggregates, to the above-described composite material was also investigated, and it was found that a useful material having similar properties could be obtained. This means that such composite materials could be used as low flow material or back fill, road base stabilization material and bedding material for pipes and cables, as well as for daily or intermediate landfill covers. From the point of view of sustainable development, this kind of waste management presents an optimum e zero waste solution, since it results in the cleaner production, while preserving natural resources, reducing CO2 emissions, and lowering the costs of sewage sludge management. Ó 2014 Elsevier Ltd. All rights reserved.

Keywords: Cleaner production Waste minimization Sustainable development Industrial ecology Sewage sludge stabilization Composite construction material

1. Introduction With the improvement of urban waste water collection and treatment in the European Union after the implementation of the Urban Waste Water Treatment Directive 91/271/EEC (EEC, 1991) a significant increase in sewage sludge production was observed (Kelessidis and Stasinakis, 2012; Valderrama et al., 2013). It is estimated that annual sewage sludge production in the EU will, by 2020, exceed 13 million tonnes of dry solids (Léonard, 2011; Milieu Ltd. et al., 2010).

* Corresponding author. Tel.: þ386 31 325 750. E-mail address: [email protected] (P. Pav
si
c). 0959-6526/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.jclepro.2013.12.034

Due to its high organic content, contamination with heavy metals (Eurostat, 2012a; Hong and Li, 2011), and the presence of pathogenic bacteria (Husillos Rodríguez et al., 2012; Kelessidis and Stasinakis, 2012) and organic pollutants (Houdková et al., 2008; Zhu et al., 2011), sewage sludge from urban waste water treatment plants poses a major environmental problem. According to the Eurostat report (Eurostat, 2012a), up-to-date biodegradable sewage sludge management mainly consists of four different types of disposal methods: reuse in agriculture, composting, incineration for volume minimization or energy utilization, and landfilling. Reuse in agriculture is the most widely adopted practice in Spain, Ireland, Lithuania, Hungary, Bulgaria, Cyprus, Luxembourg, France, the Czech Republic and Norway, whereas in Estonia and Slovakia a considerable proportion of the total volume of sewage sludge is treated through composting. Incineration of sewage sludge has

118

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

been reported as a primary pathway for disposal in the Netherlands, Slovenia, Belgium, Germany, Austria and Switzerland. Discharge into controlled landfills is still practised as the primary pathway in Greece, and used exclusively in Malta (Eurostat, 2012a). The reuse of sewage sludge in agriculture is often made more difficult to implement due to the presence of bacteria and heavy metals (Eurostat, 2012a; Qi et al., 2011). In incineration procedures, even though these have undergone significant improvements, several factors concerning the treatment of flue gases and ashes, the emission of dioxins, furans, and heavy metals, and the handling of residues, represent a severe threat to the environment (Fytili and Zabaniotou, 2008), whereas landfilling, as the least favoured option (EP and CEU, 2008), continues to damage the environment. To a minor extent, other methods, such as pyrolysis, temporary storage, long term storage, reuse in green areas or forestry as fertiliser (Romero et al., 2013), landfill cover (Herrmann et al., 2009; Kelessidis and Stasinakis, 2012), as well as use in construction (Barrera-Diaz et al., 2011; Cusidio and Cremades, 2012), have been reported. Because of wide application possibilities and high material demand, the use of sewage sludge in construction is one of the most interesting options. The majority of management practices used for the treatment of sewage sludge, including those in which such sludge is treated for construction purposes, involve dewatering or thermal drying, in order to obtain 20 wt% or more of dry solids (Ländell et al., 2012; Rodríguez et al., 2013). However, these treatments are costly (Furness et al., 2000; Murray et al., 2008) and energy-consuming. This is why the solidification/stabilization of sludge prior to dewatering (with total dry solids typically around 5 wt% (Fytili and Zabaniotou, 2008)) seems to be, with respect to sustainable development, an optimum alternative solution. These technologies are processes in which waste materials are mixed with various binding materials in order to obtain, on the one hand, new and useful composite products (Xu et al., 2008), while, on the other hand reducing the mobility of incorporated pollutants and thus also the potential threat to the environment (Chen et al., 2009). One, in the authors’ opinion very promising material, which could be used in order to develop a synergistic approach to the problem of sewage sludge is ash, whose worldwide annual production has been estimated to be, at the present time, approximately 476 million tonnes (Vassilev et al., 2013). Due to the increase trend of biomass exploitation (Eurostat, 2012b), this quantity may be expected to increase significantly in the near future. Some types of such residues, such as fly ash (Al Bakri et al., 2011; White, 2005), coal bottom ash (Katz and Kovler, 2004), and municipal solid waste incinerator bottom ash (Lam et al., 2010; Li et al., 2012), have already been successfully used as binders in the field of construction, whereas ash obtained from biomass combustion is not widely used.
Based on the results of recent research (Cernec and Zule, 2007; Pav
si
c et al., 2013) it is possible to manage sewage sludge and biomass ash within the scope of a single process, so that a new composite construction material, exhibiting the characteristics of “Controlled Low Strength Materials” (CLSM) (CCAA, 2008a; Trejo et al., 2004), and usable for many purposes, can be obtained. Recycled aggregates from construction and demolition waste, as natural aggregate substitutes, or indeed natural aggregates themselves, can also be incorporated in such new composites. CLSM refers to a cementitious material, whose characteristics are similar to those of stabilized soils, and which permits re-excavation afterwards, if necessary (Gabr and Bowders, 2000; Zhen et al., 2012). Because of reduced labour and equipment costs, faster construction and the possibility of applications in places with restricted access for compaction machinery, these materials continue to gain importance in applications such as bedding materials for pipes and

cables, void-filling and backfilling utility trenches, bridge abutments, foundations and retaining walls (Trejo et al., 2004; Zhen et al., 2012). In order to be able to successfully replace the traditional materials for CLSM with such new composite materials, the later must be environmentally inert, must not pose a health hazard, and must exhibit the desired consistency or flowability, with a setting time of 1e5 h, as well as having a compressive strength of 0.5e2.5 MPa after 28 days of curing (Rajendran, 1997). The aim of this work is to present the beneficial utilization of sewage sludge, biomass ash and recycled aggregates in the production of construction composites with CLSM properties, showing their applicability and environmental acceptability, and thus providing a basis for zero waste and cleaner production. From the sustainability point of view, this kind of waste management presents an optimum waste management solution, while preserving natural resources and reducing CO2 emissions. However, efforts still have to be made to address the global warming potential issue (Johansson et al., 2008; Liu et al., 2011). 2. Materials and methods 2.1. Materials Sewage sludge (SS), which is the fluid component of the studied composite materials, was obtained from an aerobic biological waste water treatment plant (dispersed biomass), which belongs to the VIPAP Slovenian paper-mill company, where both industrial and municipal waste waters are treated. Samples of sludge, in total amounting to 100 L, were taken from the aeration basin, before dewatering began. The biomass ash (BA), which was used as a binder in the preparation of the composite materials, originates from the same paper mill company, and represents the combustion residue from a steam boiler, where de-inking fibre paper sludge, waste wood, and bark are used as a fuel. Approximately 100 kg of the BA, which was needed for the research, was collected from a 300 m3 silo. Recycled aggregates (RA) with gradations 0/2 mm and 0/8 mm
produced at the recycled aggregate producer Zuran d.o.o. were sampled in a quantity of 50 kg and tested for size distribution and petrographic composition. Two types of composites were studied. One was a composite prepared from SS and BA, designated as SAC (“sludge-ash-composite”), and the other was the same composite with the difference that RA with nominal gradations of 0/2 mm (RA-0/2) and 0/8 mm (RA-0/8) was added to the basic SAC mixture (samples of this type were designated as SAC-0/2 and SAC-0/8, respectively, referring to the gradation size used). 2.2. Experimental methods 2.2.1. Characterization of materials for the preparation of the composites The materials used in the preparation of the studied composites were characterized. The basic parameters of the SS, such as pH and conductivity, as well as those of the dry residue (according to SIST EN 12880:2001) were first determined. Because of the potential bacteriological contamination of the SS, its microbiological quality was assessed by the aerobic mesophilic bacteria count method, which is based on the standard test method for determining water quality according to ISO 6222:1999. For this purpose 1 g of the sample was diluted, under aseptic conditions, in 100 mL of a sterile Ringer solution and spread on a Standard Plate Count Agar. After incubation for 2 days at 37  C, a colony forming units (CFU) count was performed.

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

The mineral composition of the SS and BA were determined by X-ray powder diffraction (XRD), using a Philips PW 3710 diffractometer and Cu-Ka radiation. The powdered samples were scanned at a rate of 2 per minute, over the range of 8e55 (2q). The results were analysed by X’PertHighScore Plus diffraction software. The SS and BA samples were also analysed by Fourier transform infrared spectroscopy (FTIR), using a Perkin Elmer Spectrum 100 spectrometer. Sixty-four signal-averaged scans of the samples were acquired. Powder pellets were pressed from mixtures of the samples with KBr at a ratio of approximately 1:200. The FTIR spectra were recorded with a spectral resolution of 4 cm1 within the range of 4000e400 cm1. The total element concentrations (As, Ba, Cd, total Cr, Cu, Hg, Mo, Ni, Pb, Sb, Se and Zn), selected according to the Decree on the Landfilling of Waste (Official Gazette of RS, 2011a), were determined in the BA leachate and in the SS by Inductively Coupled Plasma Mass Spectrometry (ICP-MS), based on the procedure described in SIST EN ISO 17294-2:2005. The sample of SS for the determination of total element concentrations was prepared by microwave assisted acid digestion with a nitric (HNO3) and hydrochloric (HCl) acid mixture, based on SIST EN 13656:2004. A quantity of 8.0 mL of aqua regia (HCl:HNO3 3:1) was added to 2.0 mL of the SS sample, followed by microwave digestion according to the program: ramp up to 200  C in 15 min; hold at 200  C for 20 min, and cool down for 20 min. The obtained solution was transferred into a 100 mL volumetric flask, and made up with deionised water. Before the determination of the element concentrations, the solution was filtered through white dot Sartorius filter paper (grade 389, 8e12 mm particle retention). In the case of the BA, selected inorganic parameters (pH, conductivity, selected elements, anions) were determined in the leachate, obtained according to SIST EN 1744-3:2002 (which is equivalent to the standard SIST EN 12457-2:2004 without particle size reduction) at a liquid/solid ratio of 10:1 (L/S ¼ 10 L/kg). Prior to the determination of the selected elements, the leachate was filtered through white dot Sartorius filter paper and acidified to pH 2 by means of HNO3. The filtrates of BA leachate and the SS were analysed by means of Varian 820-MS ICP-MS equipment under the following conditions: carrier gas argon, CRI gas hydrogen, a flow rate of 70 mL/min. The determined isotopes of the elements (m/z) were 53Cr, 60Ni, 63Cu, 66Zn, 95Mo, 111Cd, 121Sb, 137Ba, 202Hg, 206þ7þ8 Pb, 75As (CRI gas) and 78Se (CRI gas), the plasma flow was 18.5 L/min, and 17.0 L/min for As and Se determination, and the plasma power was set to 1.40 kW. The dissolved anions (F, Cl and SO2 4 ) in the filtrates (white dot Sartorius filter paper) were determined by liquid chromatography, according to SIST EN ISO 10304-1:2009. The RA gradations 0/2 mm and 0/8 mm (RA-0/2 and RA-0/8) were characterized by sieving analysis, according to SIST EN 9331:2012. Their petrographic composition according to SIST EN 93311:2009/AC:2010, which provides a classification for the constituents of coarse recycled aggregate, was also determined. 2.2.2. Characterization of the sewage sludge/biomass ash composite (SAC) The fresh SAC mix was prepared by mixing the SS and BA in a 1:1 weight ratio in a laboratory planetary mixer, according to SIST EN 196-1:2005. During the mixing process, changes in the temperature and pH level of the mixture were recorded, as well as the setting time, was also determined according to the procedure given in SIST EN 196-3:2005. In order to assess the potential health hazard of the prepared composite, the microbiological quality of the fresh SAC mix was determined by means of an aerobic mesophilic bacteria count.

119

For the assessment of the mechanical properties of the SAC composite, fifteen 150 mm cube samples were prepared from the fresh mix, according to SIST EN 12390-2:2009, by using a vibrating table. The compressive strength of the composite was determined according to SIST EN 12390-3:2009 after 7, 14, 28, 56, 90 and 433 days of curing the specimens in moist (100% air moisture), and after 7, 14, 28 and 56 days also in wet conditions (submerged in water). For the determination of the compressive strength of the SAC cubes, an Automax 5 automatic concrete compression machine, with a loading rate of 0.6  0.2 MPa/s, was used. Porosity and pore size distribution were determined on samples taken from the hardened SAC cubes after 4, 7, 28, 56 and 87 days of moist curing. These measurements were performed within a pressure range of up to 414 MPa by mercury intrusion porosimetry (MIP), using an AutoPore IV 9500 porosimeter (Micromeritics). Prior to the measurements, samples having a volume of approximately 1 cm3 were cut out from the SAC cubes. They were then stored in an isopropanol solution, and dried at 105  C for 24 h. For calculations of the pore size distribution, the Washburn equation was taken into account, and a contact angle of 130 was assumed. Besides pore size distribution, total porosity, median and average pore diameter, and total pore area, data about the bulk density and apparent density of the SAC composite at different curing times were obtained by using the MIP method. Since this method usually underestimates pore size (Diamond, 2000), it was used just for comparative purposes, and not for the absolute determination of pore size distribution. Gas sorption measurements, using ASAP 2020 Micrometrics equipment, were performed in order to determine the BET (Brunauer, Emmet and Teller) surface area and mesoporosity of the investigated material, for the additional characterization of the SAC cubes with different curing periods. Before the analysis, larger samples were taken out of the isopropanol solution, dried for 24 h at 105  C, crushed and then sieved on 1 and 2 mm sieves in order to obtain a 1/2 mm fraction. Small samples, each having a weight of approximately 2 g, were used for the analysis. These samples were then evacuated at 105  C with an evacuation rate of 0.67 kPa/s until a final vacuum of 2 Pa was achieved. Nitrogen was used as the adsorptive agent for the analysis. The BET surface area, the t-plot micropore area and the pore size distribution were obtained by applying the BJH (Barrett, Joyner and Halenda) method, together with the Halsey equation. Leachates from the composites for the selected inorganic parameters were obtained according to SIST EN 1744-3:2002, at a liquid/solid ratio of 10:1 (L/S ¼ 10 L/kg), from the 150 mm SAC cubes, after 7, 14 and 28 days of curing in moist conditions. The pH and conductivity of the leachates were determined, as well as the total concentrations of selected elements. Prior to this, the leachates were filtered (using a white dot Sartorius filter paper), and acidified to pH 2 with HNO3. An ICP-MS method, under the selected conditions described in subsection 2.2.1, was used to determine the total element concentrations in the acidified filtered leachate. For the identification of organic substances, a diffusioncontrolled 24 h tank leaching test, at L/S ¼ 2.6 L/kg, based on the EA NEN 7375 procedure, was performed on 70 mm hardened SAC cubes, after 3, 10, 17 and 28 days of moist curing. The organic substances in the leachate (L/S ¼ 2.6 L/kg) were identified by Gas Chromatography coupled with Mass Spectrometry (GC/MS), from a dichloromethane extract. A 1 mL splitless injection sample was analysed on Agilent 5973N GC/MS equipment, which was also fitted with an Agilent auto sampler using electron impact (70 eV) in the full spectrum ion monitoring mode. The GC equipment was fitted with a fused silica capillary column that was coated with DB5 (30 m  250 mm i.d.; 0.25 mm film thickness, J&W). The temperature program of the GC oven was as follows: isothermal at 38  C for 2 min, 4  C/min to 320  C, and held isothermal at 320  C for

120

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

15 min. Helium was used as the carrier gas. The organic compounds were identified in the GC/MS data by their characteristic mass fragmentation patterns. In the water leachates (L/S ¼ 2.6 L/kg), the total organic carbon (TOC) concentrations were also determined, according to procedure given in ISO 8245:1999. In order to gather more information about the processes accompanying stabilization, the mineral compositions of the composites after 3, 7, 14, 28, 56, 90 and 433 days (SAC1 e SAC7 respectively) were also determined by XRD, supported by FTIR.

Table 1 Element concentrations in sewage sludge (SS), biomass ash leachate (BA) and SAC leachate after 7, 14 and 28 days (SAC2 e SAC4) (limit A e Official Gazette of RS, No. 103/2011; limit B e Official Gazette of RS, No. 61/2011). Parameter

SS

Limit A

(mg/L) As

0.137

BA

SAC2

(mg/kgdry 5

<0.02

SAC3

SAC4

Limit B

mass)

<0.001

<0.001

0.5

0.07

20

<0.001 Ba

0.3

50

59

0.91 0.14

2.2.3. Characterization of the SAC with added recycled aggregates (SAC-0/2 and SAC-0/8) In order to extend the potential applicability of the SAC, the effect of the addition of RA gradations of 0/2 mm (SAC-0/2) and 0/ 8 mm (SAC-0/8), was studied. The SAC-0/2 and SAC-0/8 mix designs were based on class S3 (100e150 mm) consistency (SIST EN 2061:2003) of the fresh mix. The consistency of the fresh mixes was determined by the slump test for fresh concrete (SIST EN 123502:2009). For the preparation of the mix SAC-0/2, an SS:BA:RA-0/ 2 ¼ 4.9:4.0:1.0 weight ratio was selected, whereas for the preparation of the mix SAC-0/8, an SS:BA:RA-0/8 ¼ 2.5:1.9:1.0 weight ratio was used. The composites were prepared by mixing in a laboratory planetary mixer (SIST EN 196-1:2005). The compressive strength of the composites was determined, according to SIST EN 12390-3:2009 after 7 and 28 days of moist curing, on 150 mm cube samples of the composites, prepared according to SIST EN 12390-2:2009 by means of a vibrating table. Additionally, the mineral composition of the composites SAC-0/ 2 and SAC-0/8, after 28 days of moist curing, were determined by XRD, supported by FTIR.

Cd

0.004

0.5

<0.005

<0.001

<0.001

0.04

<0.001

0.5

<0.07

2

<0.001 Total Cr

5.3

50

<0.01

<0.001 <0.001

Cu

0.3

10

<0.07

<0.07 <0.07

Hg <0.0005 Mo

0.004 0.01 0.03

0.05

<0.004

<0.0005

1.0

<0.05

<0.01

<0.0005 <0.01

0.5

<0.03

0.4

<0.01

0.5

<0.01 Ni

0.1

50

0.05

<0.03 <0.03

Pb

0.07

10

<0.05

<0.01 <0.01

Sb

0.004

5

<0.006

0.003

0.002

0.06

0.002 Se

0.006

e

<0.01

<0.01

<0.01

0.1

<0.02

4

<0.001

800

0.07

10

<0.01 Zn Cl F

0.5

100

0.1

<0.02 <0.02

50.6

e

25.3

<0.001 <0.001

<1.0

10

4.5

0.91 0.14

SO2 4

44.7

e

<5

<0.001

<0.001

1000

<0.001

3. Results and discussion 3.1. Raw materials SS, BA and RA were used as raw materials in the preparation of the studied composites (SAC, SAC-0/2 and SAC 0/8). The SS from the aerobic biological waste water treatment plant is a fluid with 0.84 wt% of dry matter, a pH of 7.1, and a conductivity of 318 mS/m. It was microbiologically contaminated by 2.2  109 CFU/g of mesophilic bacteria. As can be seen from Table 1, the total concentrations of selected elements in the SS are significantly below the limit values for hazardous waste (Official Gazette of RS, 2011b), which were adopted as the comparative criteria, since SS is a liquid waste, whereas the waste acceptance criteria (Official Gazette of RS, 2011a) are based on leachate quality, thus assuming that the waste is in a solid state. Examination of the mineral composition, determined by XRD (Fig. 1) and FTIR (Fig. 2), showed that the SS contained calcite, quartz, dolomite and kaolinite. The broad peak in the range 15e25 (2q) of the X-ray powder diffraction pattern, together with absorption bands of the FTIR spectrum at around 1740 and 1240 cm1 (assigned to esters in hemicelluloses (Smidt et al., 2011)) confirmed the presence of hemicelluloses. Additionally, aliphatic methylene absorption bands (with a eCH2 vibration) at 2924 and 2852 cm1, and a 1660 cm1 and (assigned to C]O vibrations of amides, CO2 vibrations of carboxylates, and C]C vibrations of aromatics and alkenes), can be deemed (Smidt et al., 2007; Smidt et al., 2011) to originate from the microbial biomass. The BA is, according to the inorganic parameters of the leachate (Table 1), classified as a non-hazardous waste (Official Gazette of RS, 2011a), with Ba and Ni concentrations above the limits for inert waste. The pH of the leachate was 12.5 and its conductivity was 737 mS/m. As determined by XRD (Fig. 1) and supported by FTIR (Fig. 2), the BA consisted of free lime, portlandite, gehlenite, calcite, quartz, talc, hematite and larnite (b-dicalcium silicate).

Fig. 1. X-ray diffraction pattern of sewage sludge (SS), biomass ash (BA), composite (SAC) after 3, 7, 14, 28, 56, 90 and 433 days (SAC1 to SAC7 respectively) and composites with recycled aggregates after 28 days (SAC-0/2 and SAC-0/8).

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

121

Table 2 Compressive strengths of the composites after different time (days-d) of moist or wet curing (ND denotes not determined). Composite

SAC SAC-0/2 SAC-0/8

Fig. 2. FTIR spectra of sewage sludge (SS), biomass ash (BA), composite (SAC) after 3, 7, 14, 28, 56, 90 and 433 days (SAC1 to SAC7 respectively) and composites with recycled aggregates after 28 days (SAC-0/2 and SAC-0/8).

The particle size distributions of the used RA are typical for their nominal gradations. The RA-0/2 mm consisted of crushed brick (Rb100), whereas the RA-0/8 mm consisted of crushed brick (Rb55) and a mix of concrete and mortar (Rc45). 3.2. The sewage sludge/biomass ash composite (SAC) In the process of the mixing of the SS and the BA, the temperature is elevated up to about 45  C, and the pH value exceeds 12. As determined experimentally (Pav
si
c, 2011), the optimum SS/BA mixing ratio is 1:1 (by weight). This ratio ensures an adequate setting time (start 285 min, end 1140 min), and consequently satisfactory workability of the produced material. The elevated temperature and the pH level decreased the observed mesophilic bacteria count from 2.2  109 CFU/g, as determined in the sewage sludge, to 1.2  105 CFU/g in the composite, which shows that the microbial activity was effectively inhibited, and that the obtained composite product, when used as a construction material, does not pose a health threat (Rupel et al., 2005). The mechanical properties of the composites, which were assessed by compressive strength determination, were timedependent (Table 2). Compressive strength after 28 days ranged from 1.6 MPa to 1.8 MPa, in moist curing or wet curing e submerged, respectively. As can be seen from the development of compressive strength, the hydration process is not yet completed after 28 days. After 56 days the compressive strength had increased

Average compressive strength e moist/wet (MPa) 7d

14 d

28 d

56 d

90 d

433 d

0.4/1.4 0.4/ND 0.3/ND

1.3/1.6 e e

1.6/1.8 1.4/ND 0.6/ND

1.8/3.0 e e

2.0/ND e e

2.2/ND e e

to 1.8 MPa or even as much as 3 MPa (submerged), and even after 15 months (433 days) a slight increase was observed, which shows that the incorporated organic matter does not reduce the hydration, as has been reported for some concrete materials prepared with waste water sludge (Barerra-Diaz et al., 2011). In this way the prepared composite material exhibits a compressive strength characteristic of CLSM (Trejo et al., 2004) with low flow. The workability and compressive strength of the composites also makes it possible to use them as material for intermediate or daily landfill cover (in order to prevent the occurrence of wind-blow litter, and to deter scavenging by birds) (EA, 2010), as well as road base stabilization material, for which a compressive strength of 0.7 to >1.5 MPa has to be achieved (CCAA, 2008b). The predominant mineralogical composition of the composites consists of larnite, portlandite, calcite, gehlenite and quartz (Fig. 1). This composition changed over time, so that the amounts of larnite and portlandite were reduced, whereas the amount of calcite was increased. During the curing period the hydration products were also identified, i.e. calcium hemi-carboaluminate Hc (3CaOAl2O3½Ca(OH)2½CaCO311H2O) and calcium monocarboaluminate Mc (3CaOAl2O3CaCO311H2O), where the Hc/ Mc ratio changes according to the length of the curing period. Towards the end of this period (after 90 days), Friedel’s salt Fs (3CaOAl2O3CaCl210H2O) was formed in composite mixes SAC6 and SAC7 (Fig. 1). The formation of Fs is due to the incorporation of chlorine ions from the solution into the lattice at the time of hydration (Hirao et al., 2005; Rapin et al., 2002). The formation of new products during curing was also confirmed by the FTIR spectra (Fig. 2). This is especially indicated by the appearance of additional bands at around 3675, 3621, 3530 and 3523 cm1, corresponding to OH stretching vibrations in the newly formed hydrates (Xiao et al., 2010); the band at 3640 cm1 is characteristic for portlandite, and was also observed in the BA sample. After stabilisation of the composite, some other modifications of the absorptions bands, clearly associated with the formation of new compounds, can be noted, such as the shift of SieO bands towards lower wavenumbers (from 1015 cm1 to 1002 cm1), the shift of the band from 914 to 960 cm1, and the appearance of a new band at around 805 cm1 (assigned to M-OH vibrations). Absorption at the 1360 cm1 frequency range (assigned to CO3 vibration) in the monocarboaluminate (Kaminskas and Barauskas, 2011) was overlapped by a band characteristic of CaCO3 (1423 cm1), so that only a weak absorption shoulder could be observed. The porosity characteristics of the composite after different curing times were determined by mercury intrusion porosimetry. As can be seen from Table 3, the total pore area shows a general increase over the curing period. The porosity values range from 55.4% to 63.2% and do not appear to be in any correlation with the curing time. Nor does any such correlation appear to exist in the case of the bulk density, which ranged from 0.79 to 0.82 g/mL, and the apparent density, which ranged from 1.78 to 2.15 g/mL. However, the average and median pore diameters reduced significantly with extended curing times, which can also be seen from the pore size distribution curves shown in Fig. 3. These curves show the

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

122

Table 3 Porosity and pore size distribution by mercury intrusion porosimetry and gas sorption measurements of the SAC composites after different curing times. Parameter

Curing time (days) 4

7

28

56

87

Mercury intrusion porosimetry Total pore area (m2/g) 24.56 73.28 70.72 97.65 117.20 Porosity (%) 55.43 63.16 58.30 61.82 58.87 Median pore diameter 0.853 0.519 0.126 0.092 0.056 by volume (mm) 0.044 0.041 0.031 0.025 Average pore diameter (mm) 0.114 Bulk density (g/mL) 0.794 0.784 0.814 0.819 0.802 Apparent density (g/mL) 1.782 2.129 1.952 2.145 1.949 Gas sorption 2 BET surface area (m /g) 33.941 33.096 36.001 35.894 62.604 Total pore volume (cm3/g) 0.172 0.167 0.185 0.181 0.273 2.9600 2.6733 3.2421 3.3411 6.3389 t-Plot Micropore Area (m2/g)

influence of curing time on the representative pore size distribution. At the beginning of the curing process (at 4 and 7 days), the composite shows a clear bimodal distribution, with intrusion peaks at around 0.02 mm and 1.3 mm. Later in the curing period (at 28 days and later), the pore distribution becomes less clearly bi-modal, with a lowering of the pore intrusion peak at around 1.3 mm, and a shift towards 0.7 mm. In the case of the 0.02 mm intrusion peak no clear change or shift can be observed. During the gas sorption measurements it was observed that the surface area of the composite, in terms of the BET surface area, did not change significantly over the first 56 days of curing. It varied between 33.1 and 36.0 m2/g, with no clear trend, whereas after 87 days of curing a large increase up to 62.6 m2/g was observed (Table 3), which could be the result of the formation of Fs. Similarly, after 87 days of curing, the sample exhibited a pore size distribution with a clear maximum of ca. 2.25 nm, whereas the other samples did not show any clear maximum, or else such maxima might have occurred beyond the detection limit (<2.0 nm) (Fig. 4). Goñi et al. (2012) who studied BET surface area and pore size distribution in cement pastes with the addition of paper sludge and fly ash, similarly evaluated pore size distribution with dV/dD pore volume versus average pore diameter peaks. After 90 days of hydration (wet curing) they determined that the proportion of pores with diameter of around 5 nm was greater than in the case of 1, 7 or 28 days of hydration. The volume of pores in the SAC accessible to gas showed a slightly increasing trend over the curing time, with a significant increase after 87 days. The studied composite exhibits adsorptionedesorption isotherms of type IV, which are typical for

Fig. 4. Volume of pores accessible to BET for the sewage sludge/biomass ash composites (SAC) after different curing times.

many industrial meso-porous materials (Klobes et al., 2006). The observed changes in porosity and pore distribution indicate that, during curing, the formation of new hydration products takes place, thus lowering the volume of the larger diameter pores and increasing the volume of the micropores. This is also supported by the t-plot micropore area (Table 3), which increases with longer curing times. Since the novel composites are intended for use in the construction industry, the quality of the water leachates, which indicates the environmental acceptability of the product, is also of great importance. From the measured concentrations of inorganic parameters in leachates obtained from the composite after 7, 14 and 28 days of moist curing, which are presented in Table 1, it is clear that the new composite product can be classified as inert (Official Gazette of RS, 2011a) and as such environmentally acceptable. With increased curing times a very slight decrease in the pH value (from 11.8 to 11.4) and the conductivity (from 63.7 to 28.3 mS/m) of the leachates was observed. The same trend was also observed in the case of barium and antimony, whereas the concentrations of other elements are below the limits of detection. These results indicate that the inorganic pollutants which are incorporated in the composite are immobilized by solidification/stabilization processes in the hydrated matrix of the composites (Malviya and Chaudhary, 2006). The organic compounds detected in dichlormethane extracts of the composite leachates are mainly organic acids. Qualitatively, the composition of the extracts did not change with curing time, whereas their actual concentrations reduced over time. Mainly low fatty acids (nC4 e nC12) were detected, followed by phenyl-alcanic acids with basic benzoic acid and methyl benzoic acid derivates. The share of carboxylic acids, biphenyl and naphthalene is lower. Polycyclic aromatic compounds occur in low concentrations. The TOC concentrations (Table 4), which decreases with longer curing times, are in good agreement with the overall leaching of organic compounds from the composite. A decrease in the concentration of organic compounds in the composite leachate with curing time is associated with pore distribution changes, which disable water accessibility to the

Table 4 Results of TOC determination in the SAC leachates at different curing times. Parameter

TOC Fig. 3. Log differential intrusion versus pore size of the sewage sludge/biomass ash composites (SAC) after different curing times.

Curing time (days)

(mg/L) (mg/m2) (mg/kgdry

mass.)

3

10

17

28

22 938 0.102

15 638 0.070

13 553 0.062

13 553 0.062

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

pollutants trapped in the composite matrix (Tiruta-Barna et al., 2006; Wilk, 2007), whereas in the case of the incorporated elements, precipitation, sorption, and chemical incorporation of the hydration products in the matrix can also be assumed (Chen et al., 2009; Malviya and Chaudhary, 2006). 3.3. SAC with recycled aggregates (mixes SAC-0/2 and SAC-0/8) The mixing ratios, the consistency of the fresh mixes, and the compressive strength of these composites, after 7 and 28 days of moist curing, are presented in Table 2. The mineralogical composition of composite SAC-0/2 is similar to the composition of SAC1 to SAC7, but it was found to contain substantially less hemicarboaluminate, whereas the amount of quartz was significantly higher than in any of the previous composites due to the presence of quartz in the brick aggregate. The composite SAC-0/8 has a similar mineralogical composition to that of composite SAC-0/2, with the exception that muscovite and calcium aluminate are additionally present, originated from aggregate (Fig. 1). Similarly, as in the case of the SAC composites, absorption bands in the range 3675 and 3520 cm1, corresponding to OH vibrations in the newly formed hydrates, were observed. The main band attributed to the SieO and AleO vibrations of the mineral phases was observed at around 955 cm1 (Fig. 2). It can be observed that the prepared composites are, in fact, lowstrength materials, with 28 days compressive strengths of 0.6 MPa and 1.4 MPa in case of the 0/8 and 0/2 mm recycled aggregate, respectively (Table 2). The observed difference in compressive strength between SAC-0/2 and SAC-0/8 might be attributed to the lower BA content in the SAC-0/8 (35 wt%) mix in comparison to the SAC-0/2 (40 wt%) mix and the associated liquid to ash (SS/BA) ratios. 4. Conclusions The results of the study showed that the solidification/stabilization, of sewage sludge with a low content of dry solids (before dewatering) with biomass ash can effectively inhibit microbial activity, and thus also the associated degradation, and results in the production of an environmentally acceptable (inert) controlled low-strength, mesoporous, composite construction material, with a compressive strength of around 1.8 MPa. It was observed that mineralogical composition of the investigated SAC composites varied over time. During the curing period, various hydration products were formed. These new products influence the diameters of the pores and their distribution within the composite, reducing the proportion of large pores, while increasing the volume of micropores, thus disabling water accessibility to the incorporated pollutants, and trapping them in the composite matrix. In production of such composites recycled aggregates can also be used, resulting in similar low-strength materials, with a similar mineralogical composition, which differ mainly due to the mineral composition of the added recycled aggregate. All the composites studied within the scope of this research have the characteristics of CLSM, and can be used as a construction material mainly in the area of daily or intermediate landfill covers, and the stabilization of road bases. Due to their CLSM properties they can also be used as low flow fill material, as well as bedding material for pipes and cables, and for the backfilling of utility trenches. With regard to sustainable development, the described waste residues management method represents an optimal e zero waste solution, resulting in the cleaner production, while preserving natural resources, reducing CO2 emissions and lowering the costs of sewage sludge management. A study consisting of an environmental impact comparison of the novel treatment of sewage sludge and biomass ash with conventional treatment technologies is on-going.

123

Acknowledgements This work was partially funded by Ministry of Economic Development and Technology of the Republic of Slovenia, and by the European Regional Development Fund. Contract: RCSG/430123/2011-2011/2015-9895/9896-BE. The authors wish to thank Mr. Peter Sheppard for linguistic help in the proof-reading of the paper.

References Al Bakri, M.M., Mohammed, H., Kamarudin, H., KhairulNiza, I., Zarina, Y., 2011. Review on fly ash-based geopolymer concrete without Portland Cement. J. Eng. Technol. Res. 3, 1e4. Barrera-Diaz, C., Martinez-Barrera, G., Gencel, O., Barneal-Martinez, L.A., Brostow, W., 2011. Processed wastewater sludge for improvement of mechanical properties of concretes. J. Hazard. Mater. 192, 108e115. CCAA (Cement Concrete & Aggregates Australia), 2008a. Controlled Low-Strength Material. Available from: (accessed 25.01.13.). CCAA (Cement Concrete & Aggregates Australia), 2008b. Stabilized and Modified Roadbase Materials. Available from: (accessed 11.10.13.). Chen, Q.Y., Tyrer, M., Hills, C.D., Yang, X.M., Carey, P., 2009. Immobilisation of heavy metal in cement-based solidification/stabilisation: a review. Waste Manag. 29, 390e403. Cusidio, J.A., Cremades, L.V., 2012. Environmental effects of using clay bricks produced with sewage sludge: leachability and toxicity studies. Waste Manag. 32, 1202e1208. Diamond, S., 2000. Mercury porosimetry. An inappropriate method for the measurement of pore size distributions in cement-based materials. Cem. Concr. Res. 30, 1517e1525. EA (Environment Agency UK), 2010. LFE6-Guidance on Using Landfill Cover Materials. The Environment Agency, London, UK. Available from: (accessed 18.10.13.). EEC, 1991. Council directive of 21 May 1991 concerning urban waste water treatment (91/271/EEC). Off. J. Euro. Commun. No. L 135/40. Available from: (accessed 29.04.13.). EP (European Parliament), CEU (Council of the European Union), 2008. Directive 2008/98/EC of the European Parliament and of the Council of 19 November 2008 on waste and repealing certain Directives. Off. J. Euro. Union. No. L 312/3. Available from: (accessed 25.01.13.). Eurostat, 2012a. Water Statistics. Available from: (accessed 25.01.13.). Eurostat, 2012b. Renewable Energy Statistics. Available from: (accessed 25.01.13.). Furness, D.T., Hogget, L.A., Judd, S.J., 2000. Thermochemical treatment of sewage sludge. Water Environ. J. 14, 57e65. Fytili, D., Zabaniotou, A., 2008. Utilization of sewage sludge in EU application of old and new methods e a review. Renew. Sust. Energ. Rev. 12, 116e140. Gabr, M.A., Bowders, J.J., 2000. Controlled low-strength material using fly ash and AMD sludge. J. Hazard. Mater. 76, 251e263. Goñi, S., Frias, M., Vegas, I., García, R., Vigil de la Villa, R., 2012. Quantitative correlations among textural characteristics of CeSeH gel and mechanical properties: case of ternary Portland cements containing activated paper sludge and fly ash. Cem. Concr. Compos. 34, 911e916. Herrmann, I., Svensson, M., Ecke, H., Kumpiene, J., Maurice, C., Andreas, L., Lagerkvist, A., 2009. Hydraulic conductivity of fly ash-sewage sludge mixes for use in landfill cover liners. Water Res. 43, 3541e3547. Hirao, H., Yamada, K., Takahashi, H., Zibara, H., 2005. Chloride binding of cement estimated by binding isotherms of hydrates. J. Adv. Concr. Technol. 3, 77e84. Hong, J., Li, X., 2011. Environmental assessment of sewage sludge as secondary raw material in cement production e a case study in China. Waste Manag. 31, 1364e 1371.
, J., Ucekaj, V., Elsäßer, T., Stehlík, P., 2008. Thermal processing of Houdková, L., Borán sewage sludge e II. Appl. Therm. Eng. 28, 2083e2088. Husillos Rodríguez, N., Granados, R.J., Blanco-Varela, M.T., Cortina, J.L., MartínezRamírez, S., Marsal, M., Guillem, M., Puig, J., Fos, C., Larrotcha, E., Flores, J., 2012. Evaluation of a lime-mediated sewage sludge stabilisation process. Product characterisation and technological validation for its use in the cement industry. Waste Manag. 32, 550e560. Johansson, K., Perzon, M., Fröling, M., Mossakowska, A., Svanström, M., 2008. Sewage sludge handling with phosphorus utilization e life cycle assessment of four alternatives. J. Clean. Prod. 16, 135e151. Kaminskas, R., Barauskas, I., 2011. Influence of pozzolana on C4AF hydration and the effects of chloride and sulfate ions on the hydrates formed. Ceram. Silik. 55, 261e265.

124

P. Pav
si
c et al. / Journal of Cleaner Production 67 (2014) 117e124

Katz, A., Kovler, K., 2004. Utilization of industrial by-products for the production of controlled low strength materials (CLSM). Waste Manag. 24, 501e512. Kelessidis, A., Stasinakis, A.S., 2012. Comparative study of the methods used for treatment and final disposal of sewage sludge in European countries. Waste Manag. 32, 1186e1195. Klobes, P., Meyer, K., Munro, R.G., 2006. Porosity and Specific Surface Area Measurements for Solid Materials. In: NIST Recommended Practice Guide, Special Publication. NIST National Institute for Standards and Technology, Washington. Lam, H.K., Barford, J.P., McKay, G., 2010. Utilization of incineration waste ash residues as Portland cement clinker. Chem. Eng. Trans. 21, 757e762. Ländell, M., Håkansson, K., Mácsik, J., 2012. Evaluation of properties of fly ash stabilized sewage sludge (FSS) as liner on landfills. In: WASCON 2012 Conference Proceedings. ISCOWA and SGI, Gothenburg, Sweden. Léonard, A., 2011. Management of wastewater sludge’s: a hot topic at European level. J. Residuals Sci. Tech. 8, 38. Li, X.G., Lv, Y., Ma, B.G., Chen, Q.B., Yin, X.B., Jian, S.W., 2012. Utilization of municipal solid waste incineration bottom ash in blended cement. J. Clean. Prod. 32, 96e100. Liu, Q., Jiang, P., Zhao, J., Zhang, B., Bian, H., Qian, G., 2011. Life cycle assessment of an industrial symbiosis based on energy recovery from dried sludge and used oil. J. Clean. Prod. 19, 1700e1708. Malviya, R., Chaudhary, R., 2006. Leaching behavior and immobilization of heavy metals in solidified/stabilized products. J. Hazar. Mater. 137, 207e217. Milleu Ltd., WRc and Risk, Policy Analysts Ltd. (RPA), 2010. Environmental, Economic and Social Impacts of the Use of Sewage Sludge on Land. Final Report, Part III: Project Interim Reports. DG ENV.G.4/ETU/2008/0076r. 10.2.2010. Available from: (accessed 24.01.13.). Murray, A., Horvath, A., Nelson, K.L., 2008. Hybrid life-cycle environmental and cost inventory of sewage sludge treatment and end-use scenarios: a case study from China. Environ. Sci. Technol. 42, 3163e3169. Official Gazette of RS, 2011a. Decree on the Landfill of Waste. Official Gazette of the Republic of Slovenia No. 61/2011. Available from: (accessed 25.01.13.). Official Gazette of RS, 2011b. Decree on Waste. Official Gazette of the Republic of Slovenia No. 103/2011. Available from: (accessed 25.01.13.). Pav
si
c, P., 2011. Research Report on Biodegradable Sludge Stabilized by Combustion Residues from Boiler K5 at the Company VIPAP (Report No. 105/11/A5-PP) (in Slovenian). PMA d.o.o., Ljubljana. Pav
si
c, P., O
stir, D., Mladenovi
c, A., Kramar, S., Dolenec, M., Bukovec, P., 2013. Sewage sludge stabilization with biomass ash. Mater. Tehnol. 47, 561e574. Qi, Y., Szendrak, D., Yuen, R.T.W., Hoadley, A.F.A., Mudd, G., 2011. Application of sludge dewatered products to soil and its effects on the leaching behaviour of heavy metals. Chem. Eng. J. 166, 586e595. Rajendran, N., 1997. Controlled Low Strength Materials (CLSM). Reported By ACI Committee 229. Available from: (accessed 18.10.13.). Rapin, J.P., Renaudin, G., Elkaim, E., Francois, M., 2002. Structural transition of Friedel’s salt 3CaOAl2O3CaC210H2O studied by synchrotron powder diffraction. Cem. Concr. Res. 32, 513e519. Rodríguez, N.H., Martínez-Ramírez, S., Blanco-Varela, M.T., Donattelo, S., Guillem, M., Puig, J., Fos, C., Larrotcha, E., Flores, J., 2013. The effect of using thermally dried sewage sludge as an alternative fuel on Portland cement clinker production. J. Clean. Prod. 52, 94e102.

Romero, C., Ramos, P., Costa, C., Márquez, M.C., 2013. Raw and digested municipal waste compost leachate as potential fertilizer: comparison with a commercial fertilizer. J. Clean. Prod. 59, 73e78. Rupel, T., Pavlica, T., Planina, P., Vidrih, J., Glas, S., Car, T., Kobola, L., Lu
sicky, M., Bo
snjak, D., Sabotin, D., 2005. Guidelines for Microbiological Food Safety for the End User (in Slovenian). Institute of Public Health of the Republic of Slovenia. Available from: (accessed 25.01.13.). Smidt, E., Meissl, K., Tintner, J., 2007. Investigation of 15-year-old municipal solid waste deposit profiles by means of FTIR spectroscopy and thermal analysis. J. Environ. Monit. 9, 1378e1393. Smidt, E., Böhm, K., Schwanninger, M., 2011. The application of FT-IR spectroscopy in waste management. In: Nikolic, Goran (Ed.), Fourier Transforms e New Analytical Approaches and FTIR Strategies. InTech, ISBN 978-953-307-232-6. Available from: http://www.intechopen.com/books/fouriertransforms-newanalytical-approaches-and-ftir-strategies/the-application-of-ft-irspectroscopy-in-wastemanagement (accessed 25.01.13.). Tiruta-Barna, L., Fantozzi-Merle, C., De Brauer, C., Barna, R., 2006. Leaching behaviour of low level organic pollutants contained in cement-based materials: experimental methodology and modelling approach. J. Hazard. Mater. 138, 331e342. Trejo, D., Folliard, K.J., Du, L., 2004. Sustainable development using controlled lowstrength material. In: Proceedings of the International Workshop on Sustainable Development and Concrete Technology, pp. 231e250. Bejing, China. Valderrama, C., Grandos, R., Cortina, J.L., Gasol, C.M., Guillem, M., Josa, A., 2013. Comparative LCA of sewage sludge valorisation as both fuel and raw material substitute in clinker production. J. Clean. Prod. 51, 205e213. Vassilev, S.V., Baxter, D., Andersen, L.K., Vassileva, C.G., 2013. An overview of the composition and application of biomass ash. Part 1. Phase-mineral and chemical composition and classification. Fuel 105, 40e76. White, D.J., 2005. Fly Ash Soil Stabilization for Non-uniform Subgrade Soils. In: Engineering Properties and Construction, Guidelines (IHRB Project TR-461, FHWA Project 4), vol. I. Iowa Highway Research Board, Federal Highway Administration. Wilk, C.M., 2007. Principles and use of solidification/stabilization treatment for organic hazardous constituents in soil, sediment, and waste. In: Proceedings WM’07 Conference, Tuscon Ariz., USA, 2007. Xiao, J., Guo, C., Jin, Y., Wang, Y., 2010. Effect of CaCO3 on hydration characteristics of C3A. J. Cent. South Univ. Technol. 17, 918e923. Xu, G.R., Zou, J.L., Li, G.B., 2008. Stabilization of heavy metals in ceramcite made with sewage sludge. J. Hazard. Mater. 152, 56e61. Zhen, G., Zhou, H., Zhao, T., Zhao, Y., 2012. Performance appraisal of controlled lowstrength material using sewage sludge and refuse incineration, bottom ash. Chinese J. Chem. Eng. 20, 80e88. Zhu, R., Wu, M., Yang, J., 2011. Mobilities and leachabilities of heavy metals in sludge with humus soil. J. Environ. Sci. 23, 247e254.
Cernec, F., Zule, J., 2007. Stability properties of biosludge-wood ash composites. In: International Conference Waste Management, Environmental Geotechnology and Global Sustainable Development (ICWMEGGSD’07-GzO’07). Available from: http://www.srdit.si/gzo07/papers/97FCernec_FinalPaperGzO07.pdf (accessed 25.01.13.).