Valorization of galvanic sludge in sulfoaluminate cement

Construction and Building Materials 23 (2009) 595–601

Contents lists available at ScienceDirect

Construction and Building Materials journal homepage: www.elsevier.com/locate/conbuildmat

Review

Valorization of galvanic sludge in sulfoaluminate cement C.A. Luz a,*, J.C. Rocha b, M. Cheriaf b, J. Pera c a

Universidade Tecnológica Federal do Paraná, Campus Pato Branco, COEDI, 85501 970 Pato Branco, PR, Brazil Universidade Federal de Santa Catarina, Departamento Engenharia Civil, Campus Trindade, 88040 900 Florianópolis, SC, Brazil Unité de Recherche en Génie Civil, Institut National des Sciences Appliquées de Lyon, Domaine Scientifique de la Doua, Bâtiment J. Tuset, 12 Avenue des Arts, 69 621 Villeurbanne Cedex, France b c

a r t i c l e

i n f o

Article history: Received 21 March 2007 Received in revised form 19 March 2008 Accepted 14 April 2008 Available online 18 June 2008 Keywords: Valorization Galvanic sludge Sulfoaluminate cement

a b s t r a c t Every year increasing amounts of industrial waste are generated worldwide. Depending on their characteristics, wastes can represent an important source of secondary raw materials in order to replace natural resources. In this study, galvanic sludge (LDG) was used as raw material in sulfoaluminate cement. This waste was incorporated to sulfoaluminate cement (CSA), at a weight ratio of 25%, to compose an blended sulfoaluminate cement (BCSA). The compressive strength, drying shrinkage and products of hydration were determined. The efficiency of CSA towards the retention of chromium (the main pollutant present in the sludge) was also investigated. Compressive strength higher than 30 MPa was obtained. The main product of hydration (ettringite) was also identified and the encapsulation of Cr in hydration phases was verified. Ó 2008 Elsevier Ltd. All rights reserved.

Contents 1. 2. 3.

4.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Materials and methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.1. Compressive strength . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2. Mass loss and drying shrinkage . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.3. Products of hydration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4. Analyses of solutions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.5. Leaching tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Portland cement manufacturing represents one of the largest consumers of natural raw materials (limestone) and energy. In the last few years, considerable attention has been given to the development of special cements, capable to reduce CO2 emissions and energy, as well as limestone consumption. Sulfoaluminate cement (CSA) is one type of cement presenting such characteristics. Belite-sulfoaluminate cements were first developed in Russia and Japan [1–7]. These cements were also investigated by Zaharov [8], who proposed to manufacture belite-sulfoaluminate cement * Corresponding author. E-mail address: [email protected] (C.A. Luz). 0950-0618/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.conbuildmat.2008.04.004

595 596 597 597 598 598 600 601 601 601 601

from raw mixture containing 10% of gypsum at temperature 1523–1573 K. Kuznetzova reported on these investigations in 1974 [9]. There is no information on industrial production of this cement. However, this type of clinker/cement is included in GOST 30515-97. Industrial sulfoaluminate cements have essentially been developed in China from natural resources (calcium carbonate, gypsum, and bauxite), since 1975 [10–16]. Their main component (yeelimite) can be synthesized at temperatures 200–300 °C lower than those required by the formation of ordinary Portland cements. Moreover, those cements need a lower amount of limestone in the raw mix and this leads to a reduction of both the thermal input for the calcination process and the emission of CO2. The clinkers resulting after firing are relatively soft and require less grinding energy than Portland clinkers. Consequently these binders can give a

C.A. Luz et al. / Construction and Building Materials 23 (2009) 595–601

substantial contribution to the saving of natural resources, energy and environment. At in industrial scale, sulfoaluminate cement (CSA) is only produced in China. It was developed in the 70s by the CBMA (China Building Material Academy). After 30 years of industrial commercialization, the production was estimated at 1.5 million tons/year in 2005 [16]. Sulfoaluminate cements contain the phases belite (C2S), yeeli and gypsum mite or tetracalcium trialuminate sulfate (C4 A3 S),  2 ) as their main constituents. When CSA cement hydrates, (CSH 3 H32 ) is formed according to the following reactions ettringite (C6 AS [14]:

90 75

Passing (%)

596

Phosp Clinker LDG

60 45 30 15 0 0.1

1

3 H32 þ 2AH3  þ 2CSH  2 þ 34H ! C6 AS C4 A3 S in absence of calcium hydroxide; 3 H32  þ 8CSH  2 þ 6CH þ 74H ! 3C6 AS C4 A3 S in presence of calcium hydroxide; Sulfoaluminate cements have been used to study the formation of oxyanion-substituted ettringites for the purpose of fixation of heavy metals [17–21]. From a chemical point of view, trivalent ions as Fe3+, Cr3+, and Mn3+ can replace Al3+ ions in the crystal structure of ettringite. Ca2+ ions can be substituted by bivalent ions as Mg2+, can be reZn2+, Mn2+, Fe2+, Co2+ or Ni2+ [20,21]. Moreover SO2 4   2 2 placed by CO2 3 ; NO3 ; SeO4 ; CrO4 , or BðOHÞ4 [22–25]. The purpose of the present study was to compose a new CSA cement to immobilize a galvanic sludge rich in chromium in a sulfoaluminate cement matrix. Some previous studies have shown the ability of such matrix in the fixation of heavy metals [26,27]. 2. Materials and methods Galvanic sludge (LDG) is an industrial waste. It is obtained from the chromium electroplating process and Cr is the main pollutant present in the sludge. This metal occurs in water solution in two states of oxidation: Cr3+ and Cr6+. The last one is more toxic and more soluble and therefore, more difficult to be retained in a cementitious matrix. In this study, the LDG used is from a brazilian industry, in Palhocßa, in the state of Santa Catarina. The galvanic sludge (LDG) was dried at 50 oC and ground (<80 lm), developing a B.E.T specific surface area of 19.5 m2/g, which is pretty high and favorable for the stabilization of wastes by cements [28]. The dried galvanic sludge (LDG) was then used to replace sulfoaluminate cement (CSA) in a weight ratio of 25%, leading to blended sulfoaluminate cement BCSA (25% LDG + 75% CSA). Two compositions of sulfoaluminate cement (BCSA) were utilized, varying the amount of clinker and calcium sulfate. Phosphogypsum was used instead of natural gypsum in the production of sulfoaluminate cement. The two compositions investigated in this study were: – CSA2080: 20% phosphogypsum + 80% CSA clinker – CSA3070: 30% phosphogypsum.70% CSA clinker. Fig. 1 shows the particle size distribution of LDG after drying and grinding. The particle size distributions of phosphogypsum and sulfoaluminate clinker are also shown in this Figure. The range of the distributions is 0.3–120 lm. The dry matter of galvanic sludge (LDG) contained large amounts of CaO (20.93%), SiO2 (16.74%), Al2O3 (5.67%), and a high carbon content (35.40%) which lead to an important loss on ignition (Table 1). The organic carbon content was 1.9% and the total amount of CO2 released at 1000 °C was 33.9%, which is very close to the LOI (35.4%). Trace elements in LDG (Table 1) were identified by Inductively Coupled Plasma Mass Spectrometry (ICP-MS). ICP-

10

100

1000

Diameter (µm) Fig. 1. Particle size distributions of galvanic sludge (LDG), CSA clinker and phosphogypsum (Phosp).

Table 1 Chemical composition of raw materials (%) Elements

Oxides SiO2 Al2O3 Fe2O3 MnO MgO CaO Na2O K2O TiO2 P2O5 SO3 Loss on ignition at 1000 °C CO2 (total) C (organic) Minor (ppm) As Ba Cd Cr Cu Ni Pb Sr Zn

Galvanic sludge

Sulfoaluminate cement Sulfoaluminate clinker

Phosphogypsum

16.74 5.67 1.58
6.62 32.10 8.78
0.22 0.16
33.90 1.90

–

–

1.7 205 1.0 85460 29570 9571 112 549 110

42.9 123 0.4 250 21 39 8.0 2084 54

– 836 0.8 7.0 6.0 6.0 3.0 12090 15

MS is a powerful technique for multi-element analysis. In a single scan in the semi- quantitative mode the analyst is able to acquire estimates on practically every element in the periodic table. In the quantitative mode accuracy and precision is comparable to existing techniques for every calibrated element. Samples for trace metals analysis are acid digested on a hot-plate to solubilize the elements of interest. The digested solutions are pumped into the plasma as a liquid stream at the rate of about 1 mL per minute with a peristaltic pump. At temperatures of 5000–8000 K in the argon plasma, all compounds in the sample stream are dissociated into their most basic components elemental ions. From the plasma, the ion stream enters a vacuum through a pinhole. The ions are focused by a series of voltage modulated lenses into the quadruple mass analyzer which only allows ions in a single mass-to-charge ratio through at a time. Ions are directed into an electron multiplier to increase the signal and be detected. Typically the range from mass 2 to 300 atomic mass units is scanned. In the present study, chromium (Cr) represented the largest part of the pollutants. The chemical composition of the raw materials utilized was obtained by Inductively Coupled Plasma Atomic–Emission

597

C.A. Luz et al. / Construction and Building Materials 23 (2009) 595–601 Table 2 Mineral composition of sulfoaluminate clinker Phase

Percentage

Belite-C2S  Yeelimite-C4 A3 S Mayenite-C12A7 Perovskite-C3FT2 C4AF

17.4 60.9 – 7.9 14.0

Spectrophotometry (ICP-AES) and is presented in Table 1. ICP-AES is an emission spectrophotometric technique, exploiting the fact that excited electrons emit energy at a given wavelength as they return to ground state. The fundamental characteristic of this process is that each element emits energy at specific wavelengths peculiar to its chemical character. ICP-AES analysis requires a sample to be in solution. This is achieved by an acid attack. Phosphogypsum contained 99% pure gypsum. Table 2 shows the mineralogical composition of sulfoaluminate clinker, obtained by X-ray diffraction using the Rigaku Equipment 0 Model Mini Flex, with radiation Cu Ka, k = 1.5406 Å A. The X-ray diffractogram of the dried galvanic sludge (Fig. 2) only revealed the presence of quartz SiO2 (JCPDS 46-1045) and calcite CaCO3 (JCPDS 05-0586) as crystalline phases. The presence of calcite explained the very high loss on ignition (35.40%), observed in the chemical analysis (Table 1): carbon dioxide is released at 1000 °C at a content of 33.9%, which is very close to the LOI (35.4%). The blended sulfoaluminate cement BCSA (25% LDG + 75% CSA) was used in the preparation of standard mortars 1:3:0.5 (cement: sand: water), by weight, as specified in the French standard NF EN 196-1. The drying shrinkage and compressive strength of these mortars were assessed and for two tests, the number of samples was 3 and 6, respectively. The mixtures were prepared at the same consistency and same water consumption (W/C = 0.5) using liquid superplasticizer, and cast in prismatic molds (40 mm  40 mm  160 mm). The use of liquid superplasticizer was very easy in each mortar. After 24 h, the samples were sealed in vinyl bags and maintained in climate chamber (at 22 °C and 50% RH) until the strength tests. This is an experimental plan which is very efficient for mortars curing.

For the identification of hydrated compounds, pastes were cast in cylindrical moulds (U = 20 mm and h = 40 mm). The water to cement ratio was 0.4 and the CSA/LDG ratio was maintained constant at 3 (75% CSA + 25% LDG). At given ages, the hydration was stopped by immersion of the samples into an acetone solution for 2 h, and afterwards the samples were dried by filtration. The fragments obtained were ground to be smaller than 150 lm and analyzed by X-ray diffraction. Some fragments were observed by SEM (scanning electron microscopy associated with EDAX). The X-ray diffraction analyses were carried out at the ages of 3, 7 and 28 days using a SIEMENS D 500 equipment, with Cu Ka = 1.54 Å. The SEM analysis, performed with a PHILIPS Model XL30 apparatus, equipped with a micro-probe EDAX, took place at 90 days. To investigate the effect of chromium (the main pollutant present in LDG) on the hydration of BCSA, three types of salts were used separately: Na2Cr2O7  2H2O (Cr6+), Na2CrO4  4H2O (Cr6+) and CrCl3  2H2O (Cr3+). Each of them was added to pure CSA (without LDG) at a content of 0.024 mol Cr/kg CSA. A suspension with deionized water (water/CSA = 4) was prepared. The mixture was regularly filtered in a membrane Whatman 40, and the distribution  3+ and Cr6+ in the resultant soluanalysis of Al3+, Ca2+, SOþ2 4 Cl , Cr tion was done by ICP-AES. Leaching tests were carried out according to Brazilian standard NBR 10005 on crushed pastes (< 9.5 mm) at 28 days. 3. Results and discussion 3.1. Compressive strength Based on Table 3, it is observed that the compressive strength for mortars prepared with BCSA (CSA + LDG) was always lower than that of the standard mortars. The largest difference was observed at 24 h of age. After 7 days of age, the progress presented the same trend. The dosage with higher amount of clinker (CSA2080) showed a higher early strength, but after 7 days, the ratio clinker/phosphogypsum did not influence anymore the strength of mortars prepared with BCSA. In almost all cases, the ratio between the strength of the mortar containing LDG and that of the standard one is lower than 0.75 which corresponds to the dilution (replacement of 25% CSA by LDG). This probably occurred due to the lower density of mortars

3500 Q =Quartz-SiO2 C = Calcite-CaCO3

C

3000

2500

Q

Cps

2000

1500 C

C C

C

C

1000

C

Q

Q

Q

C

Q

Q

C

C

500

0

0

10

20

30

40

50

2 theta Fig. 2. X-ray diffractogram of dried galvanic sludge (LDG).

60

70

598

C.A. Luz et al. / Construction and Building Materials 23 (2009) 595–601

Table 3 Compressive strength and ratios of the strength and the density between BCSA mortars and CSA control mortars

2080

Days

1 3 7 28 90

3070

1 3 7 28 90

b c

a X

Stdb

mc

Strength

Specific gravity

BCSA CSA_Con BCSA CSA_Con BCSA CSA_Con BCSA CSA_Con BCSA CSA_Con

20.19 37.00 24.50 37.73 26.17 38.23 32.27 48.40 34.90 51.10

0.98 0.20 0.65 0.21 0.87 0.53 1.03 1.67 0.90 0.61

0.05 0.01 0.03 0.01 0.03 0.01 0.03 0.03 0.03 0.01

0.55

0.93

0.65

0.92

0.68

0.94

0.67

0.94

BCSA CSA_Con BCSA CSA_Con BCSA CSA_Con BCSA CSA_Con BCSA CSA_Con

15.72 33.40 24.27 33.83 27.77 35.23 33.97 50.43 35.43 61.22

0.51 0.51 0.67 0.51 0.56 1.35 0.71 2.15 1.80 0.75

0.03 0.02 0.03 0.02 0.02 0.04 0.02 0.04 0.05 0.01

E E E

C

Y

E E E

E G

E

E E Y EE E D C C GD E G

28j

7j

200 3j

0 3

0.68

0.89

0.47

0.92

0.72

0.92

0.79

0.94

0.67

0.94

0.58

0.88

8

13

18

23

28

33

38

43

48

53

58

63

Fig. 4. X-ray diffractogram of BCSA2080 samples at 3, 7 and 28 days. E = Ettringite, G = gypsum, Y = yeelimite, C = calcite.

1000 E

800

E

E E

E

E

G

C Y E E E

E E E G EY C D C E G

E

Y

28j

600

400

prepared with LDG. In Table 3, in the last column, it is possible to verify that the specific gravity of mortars prepared with LDG reaches 88–94% of that of control mortars.

7j

200 3j

0 3

8

13

18

23

28

3.3. Products of hydration In pastes BCSA2080 and BCSA3070 (Figs. 4 and 5, respectively), it is possible to observe that ettringite (JCPDS 41-1451) is the main product of hydration, in both cases. Gypsum (JCPDS 03-0053) and yeelimite (JCPDS 33-0256), present in the sulfoaluminate clinker, were not totally consumed. Calcite (JCPDS 05-0586) was also identified due to the presence of galvanic sludge in the paste. Fig. 6 shows the main peak of yeelimite for BCSA2080 and BCSA3070. It is not possible to observe a decrease of the peak intensity versus time (BCSA2080), which shows that most of yeelimite is early consumed. Fig. 6, relative to the BCSA3070 paste, shows that the peak for yeelimite decreases between 3 and 7 days,

48

53

58

63

68

Fig. 5. X-ray diffractogram of BCSA3070 samples at 3, 7 and 28 days. E = Ettringite, G = gypsum, Y = yeelimite, C = calcite.

which indicates that for this dosage (CSA/Phosphogypsum = 70/30), hydration occurs in a slower way. Comparing the BCSA2080 and BCSA3070 pastes, the peaks of ettringite are higher in the first one, which was already expected since it contained a higher amount of clinker, consequently, a higher amount of yeelimite. The formation of ettringite can be better pointed out in Figs. 7 and 8, referring to SEM analyses performed at 90 days of hydration. Fig. 7 refers to BCSA2080 where ettringite is identified by EDAX analyses, indicating the following elements: Ca, S, Al and O. The presence of Cr, one of the compounds of the galvanic sludge, is also observed. The presence of Fe and Si is also pointed out and they are present either in CSA or LDG. The same figure shows the EDAX analyses of the compound with a lighter color, highlighted in the picture, indicating Al, Ca, Si and Cr as the compounds, which represent the main elements present in LDG. -5.00%

-800

-4.00%

Weight loss

-1000

-600 -400

-3.00% -2.00% -1.00%

-200 Ref_CSA2080 CSA2080'

7

43

2 theta

In Fig. 3, it is possible to observe higher values of the mass loss for mortars prepared with LDG sludge (BCSA). Higher drying shrinkage is also observed. At 28 days of age, mortars containing LDG for both dosages (2080 and 3070) presented almost two times the drying shrinkage obtained by the control ones.

0

38

33

3.2. Mass loss and drying shrinkage

0

68

2 theta

Mean. Standard deviation. Coefficient of variation.

Drying shrinkage (µm/m)

Y

600

400

Cps

a

Compressive strength (MPa)

Mortars

E

800

Cps

CSA

Ratios (CSA´/Ref CSA)

E

1000

14

Days

Ref_CSA3070 CSA3070'

21

CSA2080_Ref CSA2080'

0.00% 28

0

7

14

Days

Fig. 3. Drying shrinkage and mass loss of BCSA mortars and control CSA mortars.

CSA3070_Ref CSA3070'

21

28

599

C.A. Luz et al. / Construction and Building Materials 23 (2009) 595–601

800

800

28j

28j 600

600

400

400

7j

23.0

23.5

2 theta

BCSA 2080

3j 24.0

0 22.5

Yeelimite

200

Ettringite

0 22.5

Yeelimite

200

Ettringite

7j

23.0

2 theta

BCSA3070

Fig. 6. X-ray diffractograms. Evolution of the main peak of yeelimite (2h = 23.6°).

Fig. 7. SEM micrograph of BCSA2080 with 90 days: EDAX analyses of points 1 and 2.

Fig. 8. SEM micrograph of BCSA3070 with 90 days: EDAX analyses of point 1.

23.5

3j 24.0

C.A. Luz et al. / Construction and Building Materials 23 (2009) 595–601

The EDAX analyses made over the highlighted region reveals Al, Ca and S as the main compounds, which are the compounds of ettringite. Besides these elements, Cr and Si are also present in LDG (Fig. 8). 3.4. Analyses of solutions

SO4-2 Concentration (ppm)

2+ 3+ 3+ and Cr6+, in Fig. 9 shows the behavior of SO2þ 4 , Ca , Al , Cr CSA solutions (without LDG), containing Na2Cr2O7  2H2O, Na2CrO4  4H2O and CrCl3  2H2O and analyzed for 15 h. Fig. 9 shows the release of sulfate ions. Comparing both graphs, it is noticed that there is a greater release of sulfate ions for the CSA3070 composition, as expected since it presents a higher

2500 2000 1500 1000 500 0 00:00

02:24

04:48

07:12

09:36

12:00

14:24

16:48

amount of phosphogypsum. In CSA 2080, the concentration decreases which shows that the consumption of sulfate ions for the formation of ettringite occurs in a faster way. It is noticed that by 15 h, sulfate ions are totally consumed, even for solutions where Cr is present. For the 3070 composition, the sulfate concentration is in the range 1500–2000 ppm, even for the control one, showing that the hydration reaction for CSA3070 is done in a slower way. In both situations, the Cr6+ (Na2Cr4O7  2H2O and Na2CrO4  4H2O) seems to reduce the consumption of sulfate ions for the formation of ettringite (for CSA 2080 up to 8 h) while the CrCl3  2H2O seems to accelerate it. Fig. 9 shows that the consumption of Ca ions occurs in a slower way for the cement composed of 30% phosphogypsum and 70% SO4-2 Concentration (ppm)

600

2500 2000 1500 1000 500 0 00:00

02:24

04:48

07:12

Time (h:min)

800 600 400 200

02:24

04:48

07:12

09:36

12:00

14:24

16:48

Ca +2 Concentration (ppm)

Ca+2 Concentration (ppm)

1000

0 00:00

Ref Na2CrO4.4H2O

2

Na2Cr2O7.2H2O CrCl3.2H2O

Ref Na2CrO4.4H2O

1

600 400 200

02:24

04:48

07:12

09:36

800 600 400 200 0 00:00

02:24

04:48

07:12

12:00

14:24

16:48

Cr+6 Concentration (ppm)

Na2Cr2O7.2H2O CrCl3.2H2O

200 150 100 50

02:24

04:48

07:12

09:36

12:00

14:24

16:48

Ref Na2CrO4.4H2O

16:48

Na2Cr2O7.2H2O CrCl3.2H2O

600 400 200 0 00:00

02:24

04:48

07:12

09:36

12:00

14:24

16:48

Na2Cr2O7.2H2O CrCl3. 2H2O

Ref Na2CrO4.4H2O

250 200 150 100 50 0 00:00

02:24

04:48

07:12

09:36

12:00

14:24

Time (h:min)

Time (h:min)

1

14:24

800

2

250

0 00:00

12:00

Time (h:min)

Cr+6 Concentration (ppm)

Ref Na2CrO4.4H2O

09:36

Ref Na2CrO4.4H2O

Time (h:min)

1

16:48

1000

2

Na2Cr2O7.2H2O CrCl3. 2H2O

800

0 00:00

14:24

Time (h:min)

Al+3 Concentration (ppm)

Al+3 Concentration (ppm)

Ref Na2CrO4.4H2O

12:00

Na2C r2 O7.2H2 O CrCl3.2H2O

Time (h:min)

1

09:36

Time (h:min)

Na2Cr2O7.2H2O CrCl3.2H2O

2

Ref Na2CrO4.4H2O

Fig. 9. Ionic concentrations for CSA with Na2Cr2O4  4H2O, Na2Cr2O7  2H2O and CrCl3  2H2O.

Na2Cr2O7.2H2O CrCl3.2H2O

16:48

C.A. Luz et al. / Construction and Building Materials 23 (2009) 595–601

clinker. For this composition, the presence of synthetic salts seems to influence the consumption of ions, making it slower. It is also possible to observe that the curve relative to the control is situated under the other ones (containing synthetic salts) which indicates that Cr, either as trivalent or hexavalent, slightly inhibits the consumption of calcium ion for ettringite formation. The concentration of aluminum ions is initially the same for both dosages but the consumption of these ions occurs in a faster way for the composition 3070, independently from the presence of salts. For this formulation, the presence of Na2Cr4O7  2H2O (Cr6+), after 8 h, seems to inhibit Al consumption. For both compositions, Cr3+ (salt CrCl3, 2H2O) was not identified, which evidences an easy fixation by CSA. For Cr6+ (salts Na2Cr4O7  2H2O and Na2CrO4  4H2O) the concentration reduced versus time, showing a retention in CSA. In the 2080 dosage, after 14 h, the concentration of Cr was lower than the limit of detection, showing that this dosage is more efficient for the fixation of hexavalent chromium. 3.5. Leaching tests Leaching tests were carried out according to Brazilian standard NBR 10005 on crushed pastes (<9.5 mm) at 28 days. Pastes CSA 3070´ were leached in de-ionized water at water to solids ratio of 16 for 24 h. Leachate was analyzed by ICP-MS to determine the chromium content. The value obtained was 1.1 ppm which is largely smaller than the limit of Brazilian standard NBR 10004: 5 ppm. 4. Conclusion Mortars prepared with blended sulfoaluminate cement BCSA (CSA + LDG) shows very interesting results. About the mechanical performances, the strengths found, lower than the control values, are acceptable since there was a lower amount of cement in the mortar. The hypothesis of a more porous system for the matrix with sludge is based on the lower values of specific gravities and higher values of mass loss and drying shrinkage. Yet, it was seen that at the beginning, the BCSA2080 showed higher results than the BCSA3070 but, by 28 days, the values were the same. Since the last one presents a lower amount of clinker, in other words, a lower cost too, it would be more interesting for the stabilization of the galvanic sludge. About the tests carried out on the solutions with synthetic salts, the retention of Cr3+ and Cr6+ for BCSA2080 was excellent. For the composition with less clinker (BCSA3070), it was observed a complete fixation of Cr3+ but only a partial fixation of Cr6+, indicating, in this case, that the composition with more clinker is more efficient. An increase in the release of sulfate ions in both compositions was pointed out, indicating a replacement of these ions by Cr in the structure of ettringite (replacement of Cr-sulfate). The presence of Cr, identified by the micro probe EDAX also indicates that such pollutant was incorporated in ettringite. Leaching tests carried out on crushed samples shows a good retention of chromium. Acknowledgement The authors wish to thank the Capes and Cofecub for their support.

601

References [1] Budnikov PP, Kravtschenko IV. Properties of belite-sulfoaluminate cement. Fifth international congress on the chemistry of cement, vol. 4. The Cement Association of Japan; 1968. p. 319–28. [2] Okushima M, Kondo R. Development of expansive cement with calcium sulphoaluminous cement clinker. Fifth international congress on the chemistry of cement, vol. 4. The Cement Association of Japan; 1968. p. 419–38.  C2 S. In: 7th International congress on [3] Ikeda K. Cements along the join C4 A3 Sthe chemistry of cement, vol. 3; 1980. p. 31–6. [4] Ikeda K, Fukuda K, Shima H. Calcium sulphoaluminate cements prepared from low-alumina waterworks slime, In: Justnes HJ, Amarkai AB, Congrex, editors. 10th International congress on the chemistry of cement, vol. 1; June 2–6 1997. p.1i025, 8. [5] Kuznetsova TV. Aluminate and sulfoaluminate cements. Stroijzdat, Moscow; 1986. p. 206. [6] Kravchenko IV, Kouznetsova TV, Chlusov VB. Alite – sulphoaluminate clinker and cements on its base. Cement 1978;8:7–9. [7] Kuznetsova TV. Chemistry and technology of expansive cements. Cement 1980;2:10–11. [8] Zaharov LA. Belite–sulfoaluminate cement. Communication, Erevan, 1966. [9] Kuznetsova TV. Self-stressing of expansive cements. In: Sixth international congress on the chemistry of cement, vol. 3; 1974. p. 219–28. [10] Beretka J, Shermamn N, Marroccoli M, Pompo A, Valenti GL. Effect of composition on the properties of rapid hardening sulfoaluminate cement. In: Justnes HJ, Amarkai AB, Congrex, editors. 10th International congress on the chemistry of cement, vol. 2; June 2–6 1997. p. 2ii029, 8. [11] Kasselouri V, Tsakirids P, Malami Ch, Georgali B, Alexandrio C. A study of the hydratation products of a non-expansive sulfo-aluminate cement. Cement Concrete Res 1995;25(8):1726–36. [12] Zivica V. Properties of blended sulfoaluminate belite cement. Construct Build Mater 2000;14:433–7. [13] Chatterjee AK. Special cements. In: Bensted J, Barnes P, editors. Structure and performance of cements. Spon Press; 2002. p. 226–31. [14] Odler I. Cements containing calcium sulfoaluminate. In: Bentur A, Mindess S, editors. Special inorganic cements. E&FN Spon; 2000. p. 69–87. [15] Su M, Kurdowski W, Sorrentino F. Development in non-Portland cements. In: Ninth international congress on the chemistry of cement, New Delhi, India, vol. 1; 1992. p. 317–54. [16] Glasser FP, Zhang L. High-performance cement matrices based on calcium sulfoaluminate–belite composition. Cement Concrete Res 2001;31:1881–6. [17] Luz CA, Rocha JC, Cheriaf M, Pera J. Use of sulfoaluminate cement and bottom ash in the solidification/stabilization of galvanic sludge. J Hazard Mater 2006;136(3):837–45. [18] Luz CA, Pera J, Cheriaf M, Rocha JC. Behaviour of calcium sulfoaluminate cement in presence of high concentrations of chromium salts. Cement Concrete Res 2007;37(4):624–9. [19] Peysson S, Pera J, Chabannet M. Immobilization of heavy metals by calcium sulfoaluminate cement. Cement Concrete Res 2005;35(12):2261–70. [20] Klemm WA. Ettringite and oxyanyon-substituted ettringites – their characterization and application in the fixation of heavy metal: a synthesis of the literature, El Boletín RD116W de Investigación y Desarrollo, Portland Cement Association, Skokie, Illinois, EUA; 1998. [21] Kumarathasan P, McCarthy GJ, Hasset DJ, Pflughoeft-Hassett DF. Oxyanion substituted ettringites: synthesis and characterization and their potential role in immobilization of As, B, Cr, Se and V. Fly ash and coal conversion byproducts: characterization, utilization and disposal, vol. 178. Pittsburgh: Materials Research Society; 1990. p. 83–104. [22] Pölmann H, Kuzel HJ, Wenda R. Solid solution of ettringite: Part I. Incorporation of OH and CO23 in 3CaO  Al2O3  3CaSO4  32H2O. Cement Concrete Res 1990;20(6):303–22. [23] Pölmann H, Kuzel HJ, Wenda R. Solid solution of ettringite: Part II. Incorporation of B(OH)4 and CrO2 4 in 3CaO  Al2O3  3CaSO4  32H2O. Cement Concrete Res 1993;23(2):422–30. [24] Bensted J, Varma SP. Ettringite and its derivatives. II. Chromium substitution. Silicates Indust 1973;23:315–8. [25] Gougar MLD, Scheetz BE, Roy DM. Ettringite and C–S–H– Portland cement phases for waste ion immobilization: a review. Waste Manage 1996;16:295–303. [26] Pera J, Ambroise J. New applications of calcium sulfoaluminate cement. Cement Concrete Res 2004;34:671–6. [27] Pera J, Ambroise J, Chabannet M. Valorization of automotives shredder residue in building materials. Cement Concrete Res 2004;34:557–62. [28] Shi CJ, Lin CF. Arsenic contaminated site at abandoned copper smelter plant: waste characterization and solidification/stabilization treatment. Chemosphere 2003;53:691–703.