Journal of Hazardous Materials 283 (2015) 623–632
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Mixture design and treatment methods for recycling contaminated sediment Lei Wang, June S.H. Kwok, Daniel C.W. Tsang ∗ , Chi-Sun Poon Department of Civil and Environmental Engineering, Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• Contaminated sediment can be recycled as fill material for site formation.
• Thermal pretreatment of sediment permits non-load-bearing block application. • CO2 curing enhances strength and reduces carbon footprint. • Inclusion of granular wastes reinforces the solidified sediment matrix. • Sediment blocks are useful resources for construction use.
a r t i c l e
i n f o
Article history: Received 19 May 2014 Received in revised form 2 August 2014 Accepted 22 September 2014 Available online 16 October 2014 Keywords: Carbon dioxide curing Contaminated sediment Stabilization/solidification Thermal pretreatment Waste recycling
a b s t r a c t Conventional marine disposal of contaminated sediment presents significant financial and environmental burden. This study aimed to recycle the contaminated sediment by assessing the roles and integration of binder formulation, sediment pretreatment, curing method, and waste inclusion in stabilization/solidification. The results demonstrated that the 28-d compressive strength of sediment blocks produced with coal fly ash and lime partially replacing cement at a binder-to-sediment ratio of 3:7 could be used as fill materials for construction. The X-ray diffraction analysis revealed that hydration products (calcium hydroxide) were difficult to form at high sediment content. Thermal pretreatment of sediment removed 90% of indigenous organic matter, significantly increased the compressive strength, and enabled reuse as non-load-bearing masonry units. Besides, 2-h CO2 curing accelerated early-stage carbonation inside the porous structure, sequestered 5.6% of CO2 (by weight) in the sediment blocks, and acquired strength comparable to 7-d curing. Thermogravimetric analysis indicated substantial weight loss corresponding to decomposition of poorly and well crystalline calcium carbonate. Moreover, partial replacement of contaminated sediment by various granular waste materials notably augmented the strength of sediment blocks. The metal leachability of sediment blocks was minimal and acceptable for reuse. These results suggest that contaminated sediment should be viewed as useful resources. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Along with accelerated economic growth of China in the past decades, alarmingly high metal concentrations are observed in the estuarine and coastal sediments in the heavily industrialized areas
∗ Corresponding author. Tel.: +852 2766 6072; fax: +852 2334 6389. E-mail address: [email protected] (D.C.W. Tsang). http://dx.doi.org/10.1016/j.jhazmat.2014.09.056 0304-3894/© 2014 Elsevier B.V. All rights reserved.
[1]. In Hong Kong, spatial distribution of sediment contamination and its potential ecological/human risks were shown to be associated with anthropogenic sources and human activities in different areas [2]. In particular, marine sediment was found heavily contaminated at the Kai Tak approach channel, which was a semi-enclosed water body bounded by the former airport runway and the breakwaters of typhoon shelter [3]. The channel had received years of domestic and industrial wastewater discharge in the past but there was limited water circulation, consequently, the marine sediment
624
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
was classified as the most contaminated type [4]. For the development of Cruise Terminal at Kai Tak, about 1.38 million cubic metre of marine sediment required dredging from the existing seabed to provide the necessary water depth within the manoeuvring area of cruise vessels [5]. While sediment remediation were possible via chelant-enhanced washing and electrokinetic extraction [6–8], full-scale application was uncommon due to the enormous costs. Thus, large volumes of dredged contaminated sediments still need to be disposed of at contained marine disposal facilities [9]. Recent studies have attempted to recycle the dredged sediments for construction of road foundation or base layers after cementbased stabilization/solidification [10–12]. However, it was shown that high contents of clay, organic matter, and soluble salts in the marine sediments delayed the setting/hardening processes and weakened the mechanical strength [13,14]. In particular, organic matter in the sediments complexed with calcium and buffered the pH increase, thereby deterring the cement hydration from producing calcium silicate hydrate (C-S-H) and calcium hydroxide (CH) for strength development [15–17]. Chlorides and sulphates severely compromised the strength, durability and corrosion resistance of cementitious products [15,16]. Although clay minerals had negligible chemical interference with cement/aggregate bonding, high content of clay led to capillary porosity in the hydrated assemblage and uneven particle size distribution, which were detrimental to the strength and durability of the cementitious materials [18,19]. These barriers still need to be properly addressed for sediment recycling. While contaminated sediments have been conventionally regarded as hazardous materials for disposal, this study intended to develop a novel integration of mixture designs and treatment methods so that contaminated sediments could be utilized as mechanically and environmentally plausible resources for various applications. Ordinary Portland cement was the most common binder that upon hydration produced C-S-H gel, ettringite hydrate and monosulphate, which served to immobilize contaminants and fill in pores of the matrix to provide mechanical strength [20]. In order to reduce the economic costs and carbon footprint associated with cement production, coal fly ash and lime were also widely used as cement substitutes. The addition of coal fly ash led to pozzolanic reaction and formed secondary C-S-H gel that enhanced strength and reduced water absorption [21,22], whereas lime addition raised pH and supplied dissolved calcium to promote C-S-H and CH formation [17]. In addition, various pretreatment methods were potentially useful for removing organic matter and soluble salts from contaminated sediments, including water washing, alkali extraction, thermal decomposition, and chemical oxidation [16,23–25]. Thus, different binder formulations and pretreatment methods were first evaluated. Moreover, curing methods and granular size distribution were important for strength development in sediment recycling. Accelerated carbonation in a chamber filled with compressed carbon dioxide gas (i.e., CO2 curing) could facilitate setting/hardening rate and reinforce the matrix strength via carbon dioxide dissolution, penetration, and subsequent formation of calcium and magnesium carbonates, such as calcite, aragonite, magnesium calcite, and magnesium carbonates [26,27]. The cement hydration products (CH and C-S-H) could be converted into calcium carbonate, calcite, modified C-S-H gel, or polymerized silica gel in a short period of curing time, thus contributing to microstructural development, reduction in pore size and total pore volume, and increase in apparent density and mechanical strength [28–30]. Besides, water curing was commonly used for ensuring available moisture for prolonged cement hydration inside the matrix such that continued formation of CS-H gel could reduce the porosity and enhance the mechanical strength [31,32]. On the other hand, previous studies showed that coal bottom ash (CBA) [33,34], recycled concrete aggregate (RA)
[35,36], crushed glass (CG) [37,38], and municipal waste incineration bottom ash (MIBA) [7,39,40] were relatively inert materials and suitable for replacing natural aggregate. Hence, blending these four types of granular waste materials as coarse aggregate with contaminated sediment might present a simple and feasible approach to improve the granular size distribution and mechanical properties. This study aimed to: (i) identify a suitable binder formulation with varying proportion of cement, fly ash, and lime; (ii) evaluate the effectiveness of different pretreatment methods of contaminated sediment; (iii) assess the potential use of CO2 or water curing for strength enhancement; and (iv) augment the applicability of sediment recycling through integration of different granular waste materials into the mixture. The compressive strength and metal leachability were determined to validate the acceptability for reuse, while microscopic, spectroscopic, and thermogravimetric analyses were conducted to investigate the microstructure and mineralogy of stabilized/solidified sediment. 2. Experimental methods 2.1. Recycling of contaminated sediment Contaminated sediment was dredged from the top 0.5 m of the Kai Tak Approach Channel in Hong Kong. The sediment sample was composed of 27% sand, 58% silt, and 15% clay size fractions on a dry mass basis (determined by wet sieving and hydrometer tests). The as-dredged sediment had a pH of 7.2, 50.8% moisture content, 35.9 g kg−1 salinity (WTW inoLab Cond 720 conductivity meter), and 6.9% organic matter (loss on ignition at 550 ◦ C). Due to historical pollution by wastewater discharge, the total metal concentrations (average) in the sediment were 1600 mg kg−1 of Cu, 410 mg kg−1 of Zn, 130 mg kg−1 of Pb, 230 mg kg−1 of Cr, 65 mg kg−1 of Ni, and 2.4 mg kg−1 of Cd [4]. According to the management guideline in Hong Kong [9], the sediment was classified as highly contaminated (Category H) because multiple metal concentrations exceeded the Upper Chemical Exceedance Levels. The sediment sample was oven-dried, crushed to pass through 2.36-mm sieve, and re-saturated to a moisture content of 60% on a dry weight basis (i.e., same as the optimal water-to-binder ratio). All subsequent experiments were duplicated and the average values were reported with the variation ranges. The binders used in this study were ordinary Portland cement (PC, ASTM Type I, 21% SiO2 , 5.9% Al2 O3 , 64.7% CaO), coal combustion fly ash (FA, ASTM Class-F, 56.8% SiO2 , 28.2% Al2 O3 , 2% CaO) obtained from CLP Power HK Ltd, and lime (CaO, 99% purity). Preliminary tests indicated that, because of the high clay content in the sediment, the water-to-binder ratio was found to be optimal at 0.6 in terms of workability, which was then used in all experiments. Various proportions of binders and aggregates were initially mixed in a standard rotating drum type mixer for 3 min, then transferred into steel cubic moulds (50 × 50 × 50 mm3 ) and compacted on a vibrating table for 30 s to eliminate entrapped air bubbles. The prepared cubic block samples were demoulded after 24 h and wrapped up by waterproof membrane to retain moisture at 20 ± 2 ◦ C for 7or 28-d curing before further analysis. 2.2. Binder formulation, pretreatment, curing method, and aggregate addition To determine the appropriate mixtures for better strength development, 10 different binder formulations (PC, FA, and/or lime, Table 1) were tested for recycling the contaminated sediment at a binder-to-sediment ratio of 3:7, below which preliminary tests revealed that there would be insufficient mechanical strength for possible reuse. To reduce the amount of organic matter
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632 Table 1 Binder formulations for recycling contaminated sediment. Mixture
(1) 20% PC + 10% FA (2) 15% PC + 15% FA (3) 10% PC + 20% FA (4) 30% FA (5) 10% PC + 10% FA + 10% CaO (6) 10% PC + 15% FA + 5% CaO (7) 5% PC + 15% FA + 10% CaO (8) 20% FA + 10% CaO (9)15% FA + 15% CaO (10) 10% FA + 20% CaO a
Binder (%)a
Aggregate (%)a
PC
FA
CaO
SED
20 15 10 0 10 10 5 0 0 0
10 15 20 30 10 15 15 20 15 10
0 0 0 0 10 5 10 10 15 20
70 70 70 70 70 70 70 70 70 70
Mass ratio.
in the sediment, preliminary tests revealed limited success by advanced oxidation processes (only 10–20% removal) such as Fenton’s reaction and Fe(II)-activated persulphate, probably due to low accessibility of the organic matter entrapped in the mineral matrix of the sediment. Therefore, thermal oxidation (at 300 or 400 ◦ C for 1 h in a muffle furnace) and alkali extraction (1-h solubilisation by NaOH at pH 12 and solid-to-solution ratio of 100 g L−1 ) were investigated as pretreatment options for sediment recycling. Then the potential use of CO2 curing was investigated for promoting strength development. Based on preliminary tests, the demoulded sediment block samples (after 24-h initial casting) was pre-dried at 50 ◦ C for 24 h to reduce the moisture content to 4% for ensuring sufficient pathway for gas penetration, because the dissolution of CO2 gas and deposition of CaCO3 on the porous surface could be the limiting factors [41,42]. However, it was noted that elevated temperature may accelerate cement hydration and affect final strength [43], thus the moisture content should be controlled by curing in a drying chamber at room temperature in future studies. The samples were transferred into an air-tight cylindrical chamber that was vacuumed at −0.5 bar and filled with CO2 (99.5% purity) under 0.1 bar pressure for 2-h accelerated carbonation at room temperature, because this substantially reduced the energy consumption for CO2 gas compression [44]. Anhydrous silica gel was put inside the chamber to remove the evaporated water from the samples during the carbonation process. The kinetic change of temperature and relative humidity inside the chamber were recorded. A fraction of the 2-h carbonated samples were evaluated immediately and the rest were subject to 7-d additional water
625
curing (i.e., immersion in a water tank at room temperature). The carbonation front was verified by spraying 1% phenolphthalein (pH indicator) to the freshly split surface of the sediment block samples. In order to provide a continuous granular size distribution of constituents in the blocks for enhancing the mechanical strength, various waste materials were added as coarse aggregates. Recycled aggregates derived from crushed concrete (RA), recycled glass cullet derived from crushed beverage bottles (CG), coal combustion bottom ash (CBA), and municipal waste incinerator bottom ash (MIBA, Guangdong Province, China) were oven-dried and sieved to collect the 2.36–5 mm fraction. The water absorption of CBA, RA, CG and MIBA were 6.8%, 0%, 4.9%, and 9.8%, respectively. The binderto-aggregate ratio was maintained at 3:7, where a total of 70% of sediment (25–55%) and granular waste materials (15–45%) were mixed as shown in Table 2. This proposed approach also offered an advantage of increasing the extent of waste recycling in the treatment of contaminated sediment. 2.3. Compressive strength, X-Ray diffraction, scanning electron microscopy, thermogravimetric analysis, and toxicity characteristic leaching procedure As the primary performance indicator, the uniaxial compressive strength of the samples was measured by using an universal testing machine (Testometric CXM 500-50 KN) at a loading rate of 0.3 mm min−1 according to BS EN 12390-3 [45]. To validate the applicability of sediment recycling, the compressive strength was compared with the criteria for reuse as fill materials for site formation (i.e., ground levelling/stabilization to the design formation level prior to construction works) [46] and non-load-bearing concrete masonry unit [47], respectively. All tests were run in duplicates. The crystalline-phase mineralogy of the crushed samples was evaluated by using a high resolution powdered X-ray diffractometer (XRD, Rigaku SmartLab). The interface morphology of gold-coated samples with highly polished surface was observed by using scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDX, JEOL Model JSM-6490). The crystallization enthalpy between 100 and 800 ◦ C was assessed by performing thermogravimetric analysis with nitrogen stripping gas (Netzch TGA/DSC). The metal leachability from contaminated sediment, waste materials, and recycled sediment products were evaluated in terms of the toxicity characteristic leaching procedure (TCLP)
Fig. 1. Compressive strength of sediment blocks made with different binder formulations.
626
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
Table 2 Mixture designs for granular waste material addition. Mixture
Binder (%)a
(1) 15% CBA (2) 30% CBA (3) 45% CBA (4) 15% RA (5) 30% RA (6) 45% RA (7) 15% CG (8) 30% CG (9) 45% CG (10) 15% MIBA (11) 30% MIBA (12) 45% MIBA
30 30 30 30 30 30 30 30 30 30 30 30
a
Aggregate (%)a SED
CBA
RA
CG
MIBA
55 40 25 55 40 25 55 40 25 55 40 25
15 30 45 0 0 0 0 0 0 0 0 0
0 0 0 15 30 45 0 0 0 0 0 0
0 0 0 0 0 0 15 30 45 0 0 0
0 0 0 0 0 0 0 0 0 15 30 45
Mass ratio.
[48], of which the metal concentrations were determined by using an inductively coupled plasma-atomic emission spectrometry (ICPAES, Perkin Elmer Optima 3300DV). The TCLP leachability served as the acceptance criteria for on-site reuse [46]. 3. Results and discussion 3.1. Binder formulations As shown in Fig. 1, the 28-d compressive strength of the sediment blocks stabilized/solidified by 20% PC + 10% FA (1.4 MPa, i.e., 33% PC replacement) and 15% PC + 15% FA (1.2 MPa, i.e., 50% PC replacement) met the acceptance criteria for reuse as fill materials for site formation (1.0 MPa) [46], although PC replacement by FA reduced the mechanical strength. However, a higher volume of FA (67–100% PC replacement) resulted in a significant decrease of strength to 0.18 MPa, because C-S-H gel formation via pozzolanic reaction of FA was insufficient to offset the loss of cement hydration. Hence, FA was partially replaced by lime in the further mix formulations (PC + FA + CaO) to provide additional CH for the pozzolanic reaction. With the presence of 10% PC, both 10% FA + 10% CaO and 15% FA + 5% CaO presented comparable 28-d compressive strength that was above 1.0 MPa and by far exceeded the strength of 10% PC + 20% FA. Nevertheless, when PC was totally replaced (i.e., FA + CaO), the 7-d compressive strength (Fig. 1) was particularly low and signified the slow, kinetically-controlled pozzolanic reaction of FA with CaO. The 28-d strength increased with the CaO content and it was increased from 0.40 MPa (10% CaO) to 0.92 MPa (20% CaO), which, however, was insufficient for reuse. The XRD spectra (Fig. 2) reveal sharp peaks of CH at 18.2◦ and 34.2◦ , calcium carbonate (CC) at 23.2◦ and calcite (CA) at 29.5◦ and 39.6◦ in the 28-d cured sample containing 40% sediment with 30% inert aggregate (for comparison purpose). However, in the samples containing 70% sediment, the peaks of CH could not be identified regardless of the binder formulations (PC + FA, PC + FA + CaO, and FA + CaO). This evidenced that the application of sediment in a large volume significantly hindered the formation of CH (and probably C-S-H though undetectable by XRD) due to calcium complexation by organic matter as well as interferences by heavy metals and salt content in the sediment. Thus, sediment pretreatment and accelerated carbonation were evaluated for facilitating strength development in the subsequent sections.
relatively small difference of strength between 300 ◦ C and 400 ◦ C pretreatment (56% and 90% removal of organic matter as determined by loss on ignition), thus a lower temperature might be sufficient in consideration of energy consumption. Previous investigations revealed that, after thermal oxidation at 400 ◦ C, the residual organic matter in the sediment was recalcitrant, highly condensed soot/graphitic and char/charcoal black carbon [49,50]. Yet, using the PC + FA + CaO and FA + CaO binders, there was more than two-fold difference in strength of the sediment samples pretreated at 300 and 400 ◦ C (Fig. 3). This underlined the importance of binder selection because CaO appeared to be more susceptible to the influence of calcium complexation with residual organic matter. As shown in the SEM images (Fig. 4), samples of sediment pretreated at 400 ◦ C contained fewer impurities (organic matter and other unstable materials), where EDX analysis also revealed a greater amount of C-S-H gel formation. Nevertheless, 400 ◦ C pretreatment was not anticipated to induce major differences in the inorganic matrix, as recent studies showed that a higher temperature (650 ◦ C or above) was needed to activate clay minerals in sediment for subsequent strength development with cement [51,52]. On the other hand, pretreatment by 1-h alkali extraction at pH 12 only removed 17% of organic matter. The resultant compressive strength was, however, even weaker than those of the untreated sediment (Fig. 3). This was attributed to the residual sodium hydroxide in the pretreated sediment. Because alkali-enriched
3.2. Pretreatment for organic matter removal With the 1-h thermal pretreatment of sediment, there were up to 3.5-fold increase in compressive strength compared to the untreated ones (Fig. 3). Using the PC + FA binders, there was
Fig. 2. XRD spectra of 28-day cured sediment blocks.
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
627
Fig. 3. Compressive strength of 28-day cured sediment blocks with thermal or alkali pretreatment.
systems would dissolve siliceous components causing microcrack formation and/or result in a more compact C-S-H structure leading to higher porosity [53,54]. Despite convenient and economical deployment, alkali extraction of organic matter was found inadequate for sediment pretreatment. More importantly, sediment blocks produced after 400 ◦ C thermal pretreatment (using 20% PC + 10% FA and 10% PC + 10% FA + 10% CaO) fulfilled the strength requirement for application as nonload-bearing concrete masonry units (4.14 MPa) [47]. Thus, the contaminated sediment could be recycled into partition walls,
planting bricks, landscaping walls, etc. These results showcased that the recycling potential of contaminated sediment in terms of compressive strength could be comparable to or better than controlled low-strength materials produced from waste incineration bottom ash and dewatered sludge in previous studies [55,56]. It should be remarked that although the thermal pretreatment improved mechanical strength, an integration of alternative curing approach and mixture design should be considered for enhancing the strength for wider applications (e.g., load-bearing concrete masonry units).
Fig. 4. SEM images and EDX spectra of 28-day cured sediment blocks with/without thermal pretreatment.
628
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
Fig. 5. Phenolphthalein indicator tests on the freshly split surfaces of sediment blocks after 2-h CO2 curing.
3.3. Accelerated carbonation and water curing Accelerated carbonation by CO2 curing was investigated for facilitating the rate-limited strength development by cement hydration. On the freshly fractured surface of the sediment blocks (Fig. 5), phenolphthalein indicator revealed no colour change (i.e., remained colourless) in all 2-h CO2 cured samples, suggesting that they were completely carbonated to pH below 8.6 by penetration and dissolution of CO2 gas into the core. For comparison, an intense colour change to purple in the uncarbonated samples (top left corner), indicative of high pH of pore solution saturated with calcium hydroxide, whereas only near-surface region was carbonated.
The compressive strength of 2-h CO2 cured samples was almost comparable to those of the uncarbonated 7-d cured samples in the PC + FA systems (Fig. 6), whereas the strength significantly increased by up to 1.2-fold when 10% CaO or above was used in the PC + FA + CaO and FA + CaO systems. This demonstrated the important role of dissolved calcium from lime hydrolysis and freshly precipitated CH and C-S-H, which reacted with dissolved CO2 and transformed into more stable CC and CA inside the porous structure at the early stage. These results echoed previous studies that CO2 curing was effective for strength enhancement through filling up the pore volume and reducing the porosity (molar volumes of CH and CC are 33.0 mL and 36.9 mL, respectively) [28,43,57–59]. On the contrary, the strength was reduced by increasing the FA
Fig. 6. Compressive strength of sediment blocks after 2-h CO2 curing with/without subsequent 7-day water curing.
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
content, signifying that less CH was available for pozzolanic reaction of FA after accelerated carbonation. Thus, calcium-rich binders and fillers (or magnesium-rich) should be selected for capitalizing the potential of CO2 curing. Inside the CO2 curing chamber, there were rapid increases in temperature (above 30 ◦ C) and relative humidity (above 80%) within 30 min, after which the initial conditions were resumed. As a result of exothermic carbonation reactions, 2.8% of water (by weight) was evaporated while 5.6% of CO2 (by weight) was absorbed and stored in the sediment blocks. This study revealed that accelerated carbonation was a feasible alternative of 7-d normal curing for sediment recycling. Its application could present an appealing approach for CO2 capture and sequestration from flue gases of fossil fuel combustion power plants, negating the carbon footprint and greenhouse gas emission [60–62]. In addition, subsequent 7-d water curing was attempted for further strength improvement (Fig. 6), because a recent study demonstrated the essential role of re-hydration cycle in promoting later-stage strength development after carbonation [58]. However, the compressive strength unexpectedly decreased to a notable extent (except the FA + CaO system). This was likely because the high water absorption capacity of clay minerals in the sediment led to protuberance and microcrack formation upon immersion in water. Thus, water curing was found not suitable for sediment recycling. In the TGA curves (Fig. 7), the weight loss at 100–200 ◦ C was attributed to the removal of physically adsorbed water and dehydration of C-S-H gel (loss of molecular water), which was greater in the sediment blocks with 400 ◦ C thermal pretreatment, confirming a larger degree of cement hydration and pozzolanic reaction in these samples (corresponding to the higher strength shown in Fig. 3). Subsequent weight loss at 420–500 ◦ C was due to the CH dehydration, which was similar in all samples. The loss between 500 and 700 ◦ C was associated with decarbonation
629
Fig. 7. TGA curves of sediment blocks with 1-h thermal pretreatment at 400 ◦ C and 2-h CO2 curing at ambient pressure.
of amorphous CC and/or dehydroxilation (removal of hydroxyl groups) of silicates/aluminates, whereas the loss between 700 and 800 ◦ C was attributed to the decomposition of well-crystalline CC [58,63–65]. The substantial weight loss in these temperature windows (Fig. 7) proved the accelerated carbonation of sediment blocks by 2-h CO2 curing (Fig. 6). 3.4. Addition of other wastes as coarse aggregates With 30% replacement of contaminated sediment by four types of granular waste materials, the compressive strength significantly increased (Fig. 8a and b), probably because the granular skeleton was reinforced by addition of the coarse aggregate that provided a continuous particle size distribution. In the PC + FA systems, the difference in the strength of all four waste
Fig. 8. Compressive strength of sediment blocks with the addition of various waste aggregates.
630
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
Fig. 9. TCLP leachability of 28-day cured sediment blocks with different binder formulations.
(0.68 mg L−1 ) in the contaminated sediment exceeded the universal treatment standard for site formation use [46] (Zn: 4.3 mg L−1 ; Cu: 1.65 mg L−1 ; Pb: 0.75 mg L−1 ; Cr: 0.6 mg L−1 ). The 28-d cured sediment blocks using different binder formulations effectively immobilized the leachable metals in the contaminated sediment (Fig. 9). The TCLP concentrations of Zn (
Contents lists available at ScienceDirect
Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat
Mixture design and treatment methods for recycling contaminated sediment Lei Wang, June S.H. Kwok, Daniel C.W. Tsang ∗ , Chi-Sun Poon Department of Civil and Environmental Engineering, Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China
h i g h l i g h t s
g r a p h i c a l
a b s t r a c t
• Contaminated sediment can be recycled as fill material for site formation.
• Thermal pretreatment of sediment permits non-load-bearing block application. • CO2 curing enhances strength and reduces carbon footprint. • Inclusion of granular wastes reinforces the solidified sediment matrix. • Sediment blocks are useful resources for construction use.
a r t i c l e
i n f o
Article history: Received 19 May 2014 Received in revised form 2 August 2014 Accepted 22 September 2014 Available online 16 October 2014 Keywords: Carbon dioxide curing Contaminated sediment Stabilization/solidification Thermal pretreatment Waste recycling
a b s t r a c t Conventional marine disposal of contaminated sediment presents significant financial and environmental burden. This study aimed to recycle the contaminated sediment by assessing the roles and integration of binder formulation, sediment pretreatment, curing method, and waste inclusion in stabilization/solidification. The results demonstrated that the 28-d compressive strength of sediment blocks produced with coal fly ash and lime partially replacing cement at a binder-to-sediment ratio of 3:7 could be used as fill materials for construction. The X-ray diffraction analysis revealed that hydration products (calcium hydroxide) were difficult to form at high sediment content. Thermal pretreatment of sediment removed 90% of indigenous organic matter, significantly increased the compressive strength, and enabled reuse as non-load-bearing masonry units. Besides, 2-h CO2 curing accelerated early-stage carbonation inside the porous structure, sequestered 5.6% of CO2 (by weight) in the sediment blocks, and acquired strength comparable to 7-d curing. Thermogravimetric analysis indicated substantial weight loss corresponding to decomposition of poorly and well crystalline calcium carbonate. Moreover, partial replacement of contaminated sediment by various granular waste materials notably augmented the strength of sediment blocks. The metal leachability of sediment blocks was minimal and acceptable for reuse. These results suggest that contaminated sediment should be viewed as useful resources. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Along with accelerated economic growth of China in the past decades, alarmingly high metal concentrations are observed in the estuarine and coastal sediments in the heavily industrialized areas
∗ Corresponding author. Tel.: +852 2766 6072; fax: +852 2334 6389. E-mail address: [email protected] (D.C.W. Tsang). http://dx.doi.org/10.1016/j.jhazmat.2014.09.056 0304-3894/© 2014 Elsevier B.V. All rights reserved.
[1]. In Hong Kong, spatial distribution of sediment contamination and its potential ecological/human risks were shown to be associated with anthropogenic sources and human activities in different areas [2]. In particular, marine sediment was found heavily contaminated at the Kai Tak approach channel, which was a semi-enclosed water body bounded by the former airport runway and the breakwaters of typhoon shelter [3]. The channel had received years of domestic and industrial wastewater discharge in the past but there was limited water circulation, consequently, the marine sediment
624
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
was classified as the most contaminated type [4]. For the development of Cruise Terminal at Kai Tak, about 1.38 million cubic metre of marine sediment required dredging from the existing seabed to provide the necessary water depth within the manoeuvring area of cruise vessels [5]. While sediment remediation were possible via chelant-enhanced washing and electrokinetic extraction [6–8], full-scale application was uncommon due to the enormous costs. Thus, large volumes of dredged contaminated sediments still need to be disposed of at contained marine disposal facilities [9]. Recent studies have attempted to recycle the dredged sediments for construction of road foundation or base layers after cementbased stabilization/solidification [10–12]. However, it was shown that high contents of clay, organic matter, and soluble salts in the marine sediments delayed the setting/hardening processes and weakened the mechanical strength [13,14]. In particular, organic matter in the sediments complexed with calcium and buffered the pH increase, thereby deterring the cement hydration from producing calcium silicate hydrate (C-S-H) and calcium hydroxide (CH) for strength development [15–17]. Chlorides and sulphates severely compromised the strength, durability and corrosion resistance of cementitious products [15,16]. Although clay minerals had negligible chemical interference with cement/aggregate bonding, high content of clay led to capillary porosity in the hydrated assemblage and uneven particle size distribution, which were detrimental to the strength and durability of the cementitious materials [18,19]. These barriers still need to be properly addressed for sediment recycling. While contaminated sediments have been conventionally regarded as hazardous materials for disposal, this study intended to develop a novel integration of mixture designs and treatment methods so that contaminated sediments could be utilized as mechanically and environmentally plausible resources for various applications. Ordinary Portland cement was the most common binder that upon hydration produced C-S-H gel, ettringite hydrate and monosulphate, which served to immobilize contaminants and fill in pores of the matrix to provide mechanical strength [20]. In order to reduce the economic costs and carbon footprint associated with cement production, coal fly ash and lime were also widely used as cement substitutes. The addition of coal fly ash led to pozzolanic reaction and formed secondary C-S-H gel that enhanced strength and reduced water absorption [21,22], whereas lime addition raised pH and supplied dissolved calcium to promote C-S-H and CH formation [17]. In addition, various pretreatment methods were potentially useful for removing organic matter and soluble salts from contaminated sediments, including water washing, alkali extraction, thermal decomposition, and chemical oxidation [16,23–25]. Thus, different binder formulations and pretreatment methods were first evaluated. Moreover, curing methods and granular size distribution were important for strength development in sediment recycling. Accelerated carbonation in a chamber filled with compressed carbon dioxide gas (i.e., CO2 curing) could facilitate setting/hardening rate and reinforce the matrix strength via carbon dioxide dissolution, penetration, and subsequent formation of calcium and magnesium carbonates, such as calcite, aragonite, magnesium calcite, and magnesium carbonates [26,27]. The cement hydration products (CH and C-S-H) could be converted into calcium carbonate, calcite, modified C-S-H gel, or polymerized silica gel in a short period of curing time, thus contributing to microstructural development, reduction in pore size and total pore volume, and increase in apparent density and mechanical strength [28–30]. Besides, water curing was commonly used for ensuring available moisture for prolonged cement hydration inside the matrix such that continued formation of CS-H gel could reduce the porosity and enhance the mechanical strength [31,32]. On the other hand, previous studies showed that coal bottom ash (CBA) [33,34], recycled concrete aggregate (RA)
[35,36], crushed glass (CG) [37,38], and municipal waste incineration bottom ash (MIBA) [7,39,40] were relatively inert materials and suitable for replacing natural aggregate. Hence, blending these four types of granular waste materials as coarse aggregate with contaminated sediment might present a simple and feasible approach to improve the granular size distribution and mechanical properties. This study aimed to: (i) identify a suitable binder formulation with varying proportion of cement, fly ash, and lime; (ii) evaluate the effectiveness of different pretreatment methods of contaminated sediment; (iii) assess the potential use of CO2 or water curing for strength enhancement; and (iv) augment the applicability of sediment recycling through integration of different granular waste materials into the mixture. The compressive strength and metal leachability were determined to validate the acceptability for reuse, while microscopic, spectroscopic, and thermogravimetric analyses were conducted to investigate the microstructure and mineralogy of stabilized/solidified sediment. 2. Experimental methods 2.1. Recycling of contaminated sediment Contaminated sediment was dredged from the top 0.5 m of the Kai Tak Approach Channel in Hong Kong. The sediment sample was composed of 27% sand, 58% silt, and 15% clay size fractions on a dry mass basis (determined by wet sieving and hydrometer tests). The as-dredged sediment had a pH of 7.2, 50.8% moisture content, 35.9 g kg−1 salinity (WTW inoLab Cond 720 conductivity meter), and 6.9% organic matter (loss on ignition at 550 ◦ C). Due to historical pollution by wastewater discharge, the total metal concentrations (average) in the sediment were 1600 mg kg−1 of Cu, 410 mg kg−1 of Zn, 130 mg kg−1 of Pb, 230 mg kg−1 of Cr, 65 mg kg−1 of Ni, and 2.4 mg kg−1 of Cd [4]. According to the management guideline in Hong Kong [9], the sediment was classified as highly contaminated (Category H) because multiple metal concentrations exceeded the Upper Chemical Exceedance Levels. The sediment sample was oven-dried, crushed to pass through 2.36-mm sieve, and re-saturated to a moisture content of 60% on a dry weight basis (i.e., same as the optimal water-to-binder ratio). All subsequent experiments were duplicated and the average values were reported with the variation ranges. The binders used in this study were ordinary Portland cement (PC, ASTM Type I, 21% SiO2 , 5.9% Al2 O3 , 64.7% CaO), coal combustion fly ash (FA, ASTM Class-F, 56.8% SiO2 , 28.2% Al2 O3 , 2% CaO) obtained from CLP Power HK Ltd, and lime (CaO, 99% purity). Preliminary tests indicated that, because of the high clay content in the sediment, the water-to-binder ratio was found to be optimal at 0.6 in terms of workability, which was then used in all experiments. Various proportions of binders and aggregates were initially mixed in a standard rotating drum type mixer for 3 min, then transferred into steel cubic moulds (50 × 50 × 50 mm3 ) and compacted on a vibrating table for 30 s to eliminate entrapped air bubbles. The prepared cubic block samples were demoulded after 24 h and wrapped up by waterproof membrane to retain moisture at 20 ± 2 ◦ C for 7or 28-d curing before further analysis. 2.2. Binder formulation, pretreatment, curing method, and aggregate addition To determine the appropriate mixtures for better strength development, 10 different binder formulations (PC, FA, and/or lime, Table 1) were tested for recycling the contaminated sediment at a binder-to-sediment ratio of 3:7, below which preliminary tests revealed that there would be insufficient mechanical strength for possible reuse. To reduce the amount of organic matter
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632 Table 1 Binder formulations for recycling contaminated sediment. Mixture
(1) 20% PC + 10% FA (2) 15% PC + 15% FA (3) 10% PC + 20% FA (4) 30% FA (5) 10% PC + 10% FA + 10% CaO (6) 10% PC + 15% FA + 5% CaO (7) 5% PC + 15% FA + 10% CaO (8) 20% FA + 10% CaO (9)15% FA + 15% CaO (10) 10% FA + 20% CaO a
Binder (%)a
Aggregate (%)a
PC
FA
CaO
SED
20 15 10 0 10 10 5 0 0 0
10 15 20 30 10 15 15 20 15 10
0 0 0 0 10 5 10 10 15 20
70 70 70 70 70 70 70 70 70 70
Mass ratio.
in the sediment, preliminary tests revealed limited success by advanced oxidation processes (only 10–20% removal) such as Fenton’s reaction and Fe(II)-activated persulphate, probably due to low accessibility of the organic matter entrapped in the mineral matrix of the sediment. Therefore, thermal oxidation (at 300 or 400 ◦ C for 1 h in a muffle furnace) and alkali extraction (1-h solubilisation by NaOH at pH 12 and solid-to-solution ratio of 100 g L−1 ) were investigated as pretreatment options for sediment recycling. Then the potential use of CO2 curing was investigated for promoting strength development. Based on preliminary tests, the demoulded sediment block samples (after 24-h initial casting) was pre-dried at 50 ◦ C for 24 h to reduce the moisture content to 4% for ensuring sufficient pathway for gas penetration, because the dissolution of CO2 gas and deposition of CaCO3 on the porous surface could be the limiting factors [41,42]. However, it was noted that elevated temperature may accelerate cement hydration and affect final strength [43], thus the moisture content should be controlled by curing in a drying chamber at room temperature in future studies. The samples were transferred into an air-tight cylindrical chamber that was vacuumed at −0.5 bar and filled with CO2 (99.5% purity) under 0.1 bar pressure for 2-h accelerated carbonation at room temperature, because this substantially reduced the energy consumption for CO2 gas compression [44]. Anhydrous silica gel was put inside the chamber to remove the evaporated water from the samples during the carbonation process. The kinetic change of temperature and relative humidity inside the chamber were recorded. A fraction of the 2-h carbonated samples were evaluated immediately and the rest were subject to 7-d additional water
625
curing (i.e., immersion in a water tank at room temperature). The carbonation front was verified by spraying 1% phenolphthalein (pH indicator) to the freshly split surface of the sediment block samples. In order to provide a continuous granular size distribution of constituents in the blocks for enhancing the mechanical strength, various waste materials were added as coarse aggregates. Recycled aggregates derived from crushed concrete (RA), recycled glass cullet derived from crushed beverage bottles (CG), coal combustion bottom ash (CBA), and municipal waste incinerator bottom ash (MIBA, Guangdong Province, China) were oven-dried and sieved to collect the 2.36–5 mm fraction. The water absorption of CBA, RA, CG and MIBA were 6.8%, 0%, 4.9%, and 9.8%, respectively. The binderto-aggregate ratio was maintained at 3:7, where a total of 70% of sediment (25–55%) and granular waste materials (15–45%) were mixed as shown in Table 2. This proposed approach also offered an advantage of increasing the extent of waste recycling in the treatment of contaminated sediment. 2.3. Compressive strength, X-Ray diffraction, scanning electron microscopy, thermogravimetric analysis, and toxicity characteristic leaching procedure As the primary performance indicator, the uniaxial compressive strength of the samples was measured by using an universal testing machine (Testometric CXM 500-50 KN) at a loading rate of 0.3 mm min−1 according to BS EN 12390-3 [45]. To validate the applicability of sediment recycling, the compressive strength was compared with the criteria for reuse as fill materials for site formation (i.e., ground levelling/stabilization to the design formation level prior to construction works) [46] and non-load-bearing concrete masonry unit [47], respectively. All tests were run in duplicates. The crystalline-phase mineralogy of the crushed samples was evaluated by using a high resolution powdered X-ray diffractometer (XRD, Rigaku SmartLab). The interface morphology of gold-coated samples with highly polished surface was observed by using scanning electron microscopy with energy dispersive X-ray spectroscopy (SEM-EDX, JEOL Model JSM-6490). The crystallization enthalpy between 100 and 800 ◦ C was assessed by performing thermogravimetric analysis with nitrogen stripping gas (Netzch TGA/DSC). The metal leachability from contaminated sediment, waste materials, and recycled sediment products were evaluated in terms of the toxicity characteristic leaching procedure (TCLP)
Fig. 1. Compressive strength of sediment blocks made with different binder formulations.
626
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
Table 2 Mixture designs for granular waste material addition. Mixture
Binder (%)a
(1) 15% CBA (2) 30% CBA (3) 45% CBA (4) 15% RA (5) 30% RA (6) 45% RA (7) 15% CG (8) 30% CG (9) 45% CG (10) 15% MIBA (11) 30% MIBA (12) 45% MIBA
30 30 30 30 30 30 30 30 30 30 30 30
a
Aggregate (%)a SED
CBA
RA
CG
MIBA
55 40 25 55 40 25 55 40 25 55 40 25
15 30 45 0 0 0 0 0 0 0 0 0
0 0 0 15 30 45 0 0 0 0 0 0
0 0 0 0 0 0 15 30 45 0 0 0
0 0 0 0 0 0 0 0 0 15 30 45
Mass ratio.
[48], of which the metal concentrations were determined by using an inductively coupled plasma-atomic emission spectrometry (ICPAES, Perkin Elmer Optima 3300DV). The TCLP leachability served as the acceptance criteria for on-site reuse [46]. 3. Results and discussion 3.1. Binder formulations As shown in Fig. 1, the 28-d compressive strength of the sediment blocks stabilized/solidified by 20% PC + 10% FA (1.4 MPa, i.e., 33% PC replacement) and 15% PC + 15% FA (1.2 MPa, i.e., 50% PC replacement) met the acceptance criteria for reuse as fill materials for site formation (1.0 MPa) [46], although PC replacement by FA reduced the mechanical strength. However, a higher volume of FA (67–100% PC replacement) resulted in a significant decrease of strength to 0.18 MPa, because C-S-H gel formation via pozzolanic reaction of FA was insufficient to offset the loss of cement hydration. Hence, FA was partially replaced by lime in the further mix formulations (PC + FA + CaO) to provide additional CH for the pozzolanic reaction. With the presence of 10% PC, both 10% FA + 10% CaO and 15% FA + 5% CaO presented comparable 28-d compressive strength that was above 1.0 MPa and by far exceeded the strength of 10% PC + 20% FA. Nevertheless, when PC was totally replaced (i.e., FA + CaO), the 7-d compressive strength (Fig. 1) was particularly low and signified the slow, kinetically-controlled pozzolanic reaction of FA with CaO. The 28-d strength increased with the CaO content and it was increased from 0.40 MPa (10% CaO) to 0.92 MPa (20% CaO), which, however, was insufficient for reuse. The XRD spectra (Fig. 2) reveal sharp peaks of CH at 18.2◦ and 34.2◦ , calcium carbonate (CC) at 23.2◦ and calcite (CA) at 29.5◦ and 39.6◦ in the 28-d cured sample containing 40% sediment with 30% inert aggregate (for comparison purpose). However, in the samples containing 70% sediment, the peaks of CH could not be identified regardless of the binder formulations (PC + FA, PC + FA + CaO, and FA + CaO). This evidenced that the application of sediment in a large volume significantly hindered the formation of CH (and probably C-S-H though undetectable by XRD) due to calcium complexation by organic matter as well as interferences by heavy metals and salt content in the sediment. Thus, sediment pretreatment and accelerated carbonation were evaluated for facilitating strength development in the subsequent sections.
relatively small difference of strength between 300 ◦ C and 400 ◦ C pretreatment (56% and 90% removal of organic matter as determined by loss on ignition), thus a lower temperature might be sufficient in consideration of energy consumption. Previous investigations revealed that, after thermal oxidation at 400 ◦ C, the residual organic matter in the sediment was recalcitrant, highly condensed soot/graphitic and char/charcoal black carbon [49,50]. Yet, using the PC + FA + CaO and FA + CaO binders, there was more than two-fold difference in strength of the sediment samples pretreated at 300 and 400 ◦ C (Fig. 3). This underlined the importance of binder selection because CaO appeared to be more susceptible to the influence of calcium complexation with residual organic matter. As shown in the SEM images (Fig. 4), samples of sediment pretreated at 400 ◦ C contained fewer impurities (organic matter and other unstable materials), where EDX analysis also revealed a greater amount of C-S-H gel formation. Nevertheless, 400 ◦ C pretreatment was not anticipated to induce major differences in the inorganic matrix, as recent studies showed that a higher temperature (650 ◦ C or above) was needed to activate clay minerals in sediment for subsequent strength development with cement [51,52]. On the other hand, pretreatment by 1-h alkali extraction at pH 12 only removed 17% of organic matter. The resultant compressive strength was, however, even weaker than those of the untreated sediment (Fig. 3). This was attributed to the residual sodium hydroxide in the pretreated sediment. Because alkali-enriched
3.2. Pretreatment for organic matter removal With the 1-h thermal pretreatment of sediment, there were up to 3.5-fold increase in compressive strength compared to the untreated ones (Fig. 3). Using the PC + FA binders, there was
Fig. 2. XRD spectra of 28-day cured sediment blocks.
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
627
Fig. 3. Compressive strength of 28-day cured sediment blocks with thermal or alkali pretreatment.
systems would dissolve siliceous components causing microcrack formation and/or result in a more compact C-S-H structure leading to higher porosity [53,54]. Despite convenient and economical deployment, alkali extraction of organic matter was found inadequate for sediment pretreatment. More importantly, sediment blocks produced after 400 ◦ C thermal pretreatment (using 20% PC + 10% FA and 10% PC + 10% FA + 10% CaO) fulfilled the strength requirement for application as nonload-bearing concrete masonry units (4.14 MPa) [47]. Thus, the contaminated sediment could be recycled into partition walls,
planting bricks, landscaping walls, etc. These results showcased that the recycling potential of contaminated sediment in terms of compressive strength could be comparable to or better than controlled low-strength materials produced from waste incineration bottom ash and dewatered sludge in previous studies [55,56]. It should be remarked that although the thermal pretreatment improved mechanical strength, an integration of alternative curing approach and mixture design should be considered for enhancing the strength for wider applications (e.g., load-bearing concrete masonry units).
Fig. 4. SEM images and EDX spectra of 28-day cured sediment blocks with/without thermal pretreatment.
628
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
Fig. 5. Phenolphthalein indicator tests on the freshly split surfaces of sediment blocks after 2-h CO2 curing.
3.3. Accelerated carbonation and water curing Accelerated carbonation by CO2 curing was investigated for facilitating the rate-limited strength development by cement hydration. On the freshly fractured surface of the sediment blocks (Fig. 5), phenolphthalein indicator revealed no colour change (i.e., remained colourless) in all 2-h CO2 cured samples, suggesting that they were completely carbonated to pH below 8.6 by penetration and dissolution of CO2 gas into the core. For comparison, an intense colour change to purple in the uncarbonated samples (top left corner), indicative of high pH of pore solution saturated with calcium hydroxide, whereas only near-surface region was carbonated.
The compressive strength of 2-h CO2 cured samples was almost comparable to those of the uncarbonated 7-d cured samples in the PC + FA systems (Fig. 6), whereas the strength significantly increased by up to 1.2-fold when 10% CaO or above was used in the PC + FA + CaO and FA + CaO systems. This demonstrated the important role of dissolved calcium from lime hydrolysis and freshly precipitated CH and C-S-H, which reacted with dissolved CO2 and transformed into more stable CC and CA inside the porous structure at the early stage. These results echoed previous studies that CO2 curing was effective for strength enhancement through filling up the pore volume and reducing the porosity (molar volumes of CH and CC are 33.0 mL and 36.9 mL, respectively) [28,43,57–59]. On the contrary, the strength was reduced by increasing the FA
Fig. 6. Compressive strength of sediment blocks after 2-h CO2 curing with/without subsequent 7-day water curing.
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
content, signifying that less CH was available for pozzolanic reaction of FA after accelerated carbonation. Thus, calcium-rich binders and fillers (or magnesium-rich) should be selected for capitalizing the potential of CO2 curing. Inside the CO2 curing chamber, there were rapid increases in temperature (above 30 ◦ C) and relative humidity (above 80%) within 30 min, after which the initial conditions were resumed. As a result of exothermic carbonation reactions, 2.8% of water (by weight) was evaporated while 5.6% of CO2 (by weight) was absorbed and stored in the sediment blocks. This study revealed that accelerated carbonation was a feasible alternative of 7-d normal curing for sediment recycling. Its application could present an appealing approach for CO2 capture and sequestration from flue gases of fossil fuel combustion power plants, negating the carbon footprint and greenhouse gas emission [60–62]. In addition, subsequent 7-d water curing was attempted for further strength improvement (Fig. 6), because a recent study demonstrated the essential role of re-hydration cycle in promoting later-stage strength development after carbonation [58]. However, the compressive strength unexpectedly decreased to a notable extent (except the FA + CaO system). This was likely because the high water absorption capacity of clay minerals in the sediment led to protuberance and microcrack formation upon immersion in water. Thus, water curing was found not suitable for sediment recycling. In the TGA curves (Fig. 7), the weight loss at 100–200 ◦ C was attributed to the removal of physically adsorbed water and dehydration of C-S-H gel (loss of molecular water), which was greater in the sediment blocks with 400 ◦ C thermal pretreatment, confirming a larger degree of cement hydration and pozzolanic reaction in these samples (corresponding to the higher strength shown in Fig. 3). Subsequent weight loss at 420–500 ◦ C was due to the CH dehydration, which was similar in all samples. The loss between 500 and 700 ◦ C was associated with decarbonation
629
Fig. 7. TGA curves of sediment blocks with 1-h thermal pretreatment at 400 ◦ C and 2-h CO2 curing at ambient pressure.
of amorphous CC and/or dehydroxilation (removal of hydroxyl groups) of silicates/aluminates, whereas the loss between 700 and 800 ◦ C was attributed to the decomposition of well-crystalline CC [58,63–65]. The substantial weight loss in these temperature windows (Fig. 7) proved the accelerated carbonation of sediment blocks by 2-h CO2 curing (Fig. 6). 3.4. Addition of other wastes as coarse aggregates With 30% replacement of contaminated sediment by four types of granular waste materials, the compressive strength significantly increased (Fig. 8a and b), probably because the granular skeleton was reinforced by addition of the coarse aggregate that provided a continuous particle size distribution. In the PC + FA systems, the difference in the strength of all four waste
Fig. 8. Compressive strength of sediment blocks with the addition of various waste aggregates.
630
L. Wang et al. / Journal of Hazardous Materials 283 (2015) 623–632
Fig. 9. TCLP leachability of 28-day cured sediment blocks with different binder formulations.
(0.68 mg L−1 ) in the contaminated sediment exceeded the universal treatment standard for site formation use [46] (Zn: 4.3 mg L−1 ; Cu: 1.65 mg L−1 ; Pb: 0.75 mg L−1 ; Cr: 0.6 mg L−1 ). The 28-d cured sediment blocks using different binder formulations effectively immobilized the leachable metals in the contaminated sediment (Fig. 9). The TCLP concentrations of Zn (
