Construction and Building Materials 165 (2018) 792–800

Contents lists available at ScienceDirect

Construction and Building Materials

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

A novel type of controlled low strength material derived from alum

sludge and green materials
Lei Wang a, Fuliao Zou a, Xiaoliang Fang a, Daniel C.W. Tsang a,⇑, Chi Sun Poon a,⇑, Zhen Leng a, Kitae Baek b
a
Department of Civil and Environmental Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China
b
Department of Environmental Engineering and Soil Environment Research Center, Chonbuk National University, 567 Baekje-daero, Deokjin-gu, Jeonju 561-756, Republic of Korea

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

 Alum sludge can be recycled as fine

aggregate into eco-friendly CLSM.
 Alum sludge addition increased water
demand and hindered CH formation.
 PFA was a promising supplementary
cementitious material for producing
CLSM.
 Addition of CaCl2 or TEA effectively
shortened overlong stiffening time.

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

Article history: The management and disposal of alum sludge generated from water treatment works is a worldwide
Received 19 October 2017 issue. This study proposed an innovative method to utilize alum sludge as a partial replacement of recy-
Received in revised form 20 December 2017 cled fine aggregate for producing controlled low strength material (CLSM). Various supplementary
Accepted 14 January 2018
cementitious materials (SCMs) and accelerators were incorporated to produce CLSM with high flowability
Available online 20 February 2018
(>200 mm), relatively short stiffening time (<24 h), and moderate compressive strength (0.3–1 MPa).
Cement chemistry and microstructure characteristics were evaluated by using isothermal calorimeter,
Keywords:
quantitative X-ray diffraction, thermogravimetric analysis, and scanning electron microscopy analysis.
Waterworks sludge
Waste recycling
The addition of alum sludge significantly increased water demand for maintaining sufficient flowability,
Cement chemistry resulting in long final stiffening time and low compressive strength. The high content of organic matter in
CLSM alum sludge hampered the Ca(OH)2 formation and pH increase, thus delaying the induction period of
Construction materials cement hydration. The binder composition played an important role in the stiffening time of CLSM.
Flowability Pulverized fly ash (PFA) was found to be the most effective SCM for improving flowability and adjusting
Stiffening time compressive strength. Calcium-based accelerators or triethanolamine (TEA) significantly shortened the
Fly ash stiffening time to 7.4 h, which facilitated the samples to fulfill all the CLSM requirements. This study pre-
sents a new and eco-friendly approach to transform alum sludge into value-added construction materials
in a sustainable way.
Ó 2018 Elsevier Ltd. All rights reserved.

⇑ Corresponding authors.
E-mail addresses: dan.tsang@polyu.edu.hk (D.C.W. Tsang), chi-sun.poon@polyu.
edu.hk (C.S. Poon).

https://doi.org/10.1016/j.conbuildmat.2018.01.078
0950-0618/Ó 2018 Elsevier Ltd. All rights reserved.
 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800
1. Introduction
The demand of drinking water treatment is increasing along
with exponential growth of population as clean water is indispens-
able for human life. Alum-based coagulants are the most widely
used chemical agent for coagulation/flocculation process in surface
water treatment works. As a result, alum sludge is massively gen-
erated with water purification [1]. The management and disposal
of alum sludge has been a global issue. Different from sewage
sludge, the low calorific value of alum sludge makes it impossible
to recover energy from incineration treatment. Currently, all alum
sludge is subjected to energy-intensive dewatering process fol-
lowed by non-sustainable landfill disposal due to a lack of robust
recycling approaches. In Hong Kong, disposal of dewatered alum
sludge to landfills is currently 58 tonnes each day and expected
to further increase with urban development [2]. Faced with scarce
land resources and stringent environmental requirements, recent
studies have investigated the utilization of alum sludge in agricul-
tural and horticultural sectors in recent studies [1]. However, there
are some limitations of soil applications such as soil acidification,
aluminium toxicity, and phosphorus deficiency. Therefore, devel-
oping economical and sustainable approaches for alum sludge
recycling has significant social and environmental values.
Alum sludge is composed of high contents of Al2O3 and SiO2
with small particle size, which is similar to conventional clay used
as construction materials [3]. Previous studies suggested that alum
sludge can be incorporated into clay to produce ceramic bricks, but
it deteriorated the properties of ceramic products such as reduced
flexural strength and increased linear firing shrinkage [4]. A low
ratio of sludge replacement should be applied to avoid unaccept-
able compromise in mechanical strength [5], e.g., over 5% of alum
sludge addition deteriorated the mechanical properties and pro-
duced low-quality ceramic bricks [6]. The feasibility of recycling
alum sludge into light-weight composite was also studied [7], in
which every 2% addition of alum sludge increased water-to-
cement ratio by 0.01 due to small particle size of alum sludge. It
was reported to delay the cement hydration and reduce the com-
pressive strength [8]. Although calcination process improved the
microstructure and enhanced the pozzolanic activity of alum
sludge [3], the addition of calcined sludge still increased water
demand and reduced compressive strength of the material [9].
Another concern is that calcination is associated with substantial
energy consumption and carbon footprint. It is imperative to deal
with alum sludge in an environmentally-friendly approach.
In this study, we propose to recycle the alum sludge into Con-
trolled Low Strength Material (CLSM), which is a flowable and
self-compacting fill material widely used in the backfill of founda-
tion pits, subgrades and pipeline beddings, void filling of subsur-
face structures, abandoned pipelines and utility vaults, grooves
and retaining walls [10]. CLSM offers advantages of easy mixing,
easy placing, self-levelling, and fast stiffening. Moreover, it can
be easily excavated at any age with conventional digging equip-
ment when future excavation is required [11]. The application of
CLSM can reduce labour requirement, expedite engineering works,
and lower construction cost. High-quality CLSM should meet strict
requirements including high flowability for self-levelling (> 200
mm), relatively fast stiffening for quick loading support (<24 h),
and controlled low strength for easy excavation (0.3–2.1 MPa)
[12,13]. In general, high aggregate-to-binder ratio was applied for
CLSM production to obtain low strength, and high water-to-
cement ratio was adopted to provide sufficient flowability. Large
amount of supplementary cementitious materials (SCMs) could
be introduced to replace ordinary Portland cement (OPC) to
improve flowability and moderate strength [12]. We propose to
use alum sludge as fine aggregate in CLSM system to improve
793
the flowability and adjust the strength. However, high clay content
would increase water demand and lead to overlong stiffening time
[14]. More importantly, organic matter would complex with Ca2+
and form a coating layer, which significantly retarded cement
hydration [15,16]. These hurdles make it difficult to produce
sludge-derived CLSM.
The binder composition plays a key role in determining the
stiffening time of CLSM. Class-F pulverized fly ash (PFA, waste
material from coal-fired power plants) is the most commonly used
SCM in the production of CLSM. The use of spherical PFA particles
could improve flowability and control moderate strength by the
effect of friction rolling, micro-aggregate filling, and pozzolanic
reaction [17]. But it may consume Ca(OH)2 (CH) and retard the
stiffening time. Other SCMs could also facilitate cement hydration
and shorten stiffening time [18,19], for instance, ground granu-
lated blast furnace slag (GGBS, by-product from iron and steel pro-
duction) contains a high content of calcium and performs as a
hydraulic material [20]. The formation of CH during spontaneous
hydraulic reaction may relieve the delayed effect by organic matter
in alum sludge. Furthermore, incinerated sewage sludge ash (ISSA,
waste material from sludge incineration facility) presents mild
pozzolanic reaction [21], whereas silica fume (SF, by-product of sil-
icon and ferrosilicon alloy production) consists of microspherical
particles that possibly improve flowability and cement hydration
[22].
In addition, the incorporation of accelerators is an effective
method for expediting cement hydration and shortening stiffening
time. Sodium-based admixtures could increase the alkalinity of
pore solution, such that high pH could promote the dissolution of
calcium silicate and calcium aluminate, resulting in short stiffening
time and high early strength [23]. Calcium-based admixtures could
provide oversaturated Ca2+ ion around tricalcium silicate (C3S) and
tricalcium aluminate (C3A) to induce cement hydration [24].
Organic accelerators, especially triethanolamine (TEA), could boost
the reaction between C3A and gypsum to form ettringite (Aft) [25].
Combined use of different accelerators may present synergistic or
superimposed effect on the reduction of stiffening time.
In order to provide scientific insights for a potential engineering
solution to alum sludge utilization, this study aims to: (i) assess the
feasibility of using alum sludge as partial replacement of fine
aggregate in CLSM system; (ii) evaluate the effectiveness of various
SCMs for achieving short stiffening time and suitable compressive
strength; (iii) select suitable accelerators for producing sludge-
derived CLSM; and (iv) investigate the surface chemistry and
microstructure characteristics of sludge-derived CLSM via micro-
scopic and spectroscopic analyses.
2. Materials and methodology
2.1. Materials for CLSM
The alum sludge was obtained from Tai Po Water Treatment
Works of the Hong Kong Water Supplies Department. The alum
sludge cake contained 55.8% water content, and dried alum sludge
contained 21.6% organic matter based on the results of the loss on
ignition (LOI) test according to ASTM D7348 (2013) [26]. The major
inorganic elements of alum sludge were detected by X-ray fluores-
cence (XRF) as shown in Table 1. According to the Toxicity Charac-
teristic Leaching Procedure [27] results, the leachability of heavy
metals and organic contaminants from alum sludge fulfilled the
acceptance criteria for on-site reuse according to the Hong Kong
Universal Treatment Standards [28]. The alum sludge mainly con-
sisted of silt and clay with particle size <0.05 mm. Thus, it could be
directly reused to replace a portion of recycled fine aggregate (RA,
size <2.36 mm) to produce CLSM.
794 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800

Table 1
Chemical compositions of alum sludge and cementitious materials (wt%).

Na2O MgO Al2O3 SiO2 P2 O5 K2O CaO Fe2O3 LOI

Alum Sludge 0.18 0.86 41.31 40.60 2.76 2.15 1.55 8.56 26.9
OPC 0.17 1.38 4.79 18.99 0.08 0.79 65.72 3.10 2.45
PFA 6.03 1.16 25.01 36.01 0.34 1.07 16.51 8.53 3.57
GGBS 0.46 7.63 15.52 31.07 0.05 0.45 40.17 0.35 2.40
ISSA 6.52 3.52 13.72 27.33 9.23 2.07 10.96 20.70 1.23
SF 0.71 1.13 0.41 85.61 0.18 7.98 1.50 0.43 2.10

LOI: loss on ignition; OPC: ordinary Portland cement; PFA: pulverized fly ash; GGBS: ground granulated blast furnace slag; ISSA: incinerated sewage sludge ash; SF: silica
fume.

The binder was composed of 20% ASTM Type I OPC and 80% sup-
plementary cementitious materials such as PFA, GGBS, ISSA and SF
based on previous report [12]. The chemical compositions and par-
ticle size distribution of these materials are listed in Table 1 and
Fig. S1 (Supplementary Information), respectively. Various addi-
tives were incorporated at the dosage of 5 wt% of binder to accel-
erate the cement hydration and these accelerators were chemical
reagents, which were purchased from Sigma-Aldrich.
2.2. Manufacturing of CLSM
The mixture design for CLSM production from alum sludge was
classified into three groups (Tables 2 and 3). The adequate
proportion of sludge as fine aggregate (0, 12.5, 25, 50 and 100%
replacement of RA by sludge) was determined first. The binder-
to-aggregate ratio was kept at 1: 4 and the OPC-to-PFA ratio was
also 1:4 in the binder system. To fulfil the flowability requirement
[12], the water-to-binder ratio was adjusted to achieve the consis-
tent flowability of 210 mm, with varying densities of the fresh
CLSMs. The effectiveness of varying binder content and different
supplementary cementitious materials was evaluated (Table 2).
The binder content increased from 20% to 40%, and the PFA was
replaced by GGBS/ISSA/SF. Finally, stiffening acceleration by incor-
porating additives (5 wt% of binder) was assessed (thirteen
sodium-based, calcium-based and compound additives, Table 3).
The production of CLSM involved mixing, flowability determi-
nation, casting, and curing phases. In the mixing phase, binder
and aggregate were dry mixed by a planetary stirrer for 1 min, then

Table 2
Mixture formulation (wt%) for CLSM with different sludge replacement and various cementitious materials addition.

Label Binder Aggregate W/B ratio Density

(kg m 3)
(a) Recycling Alum Sludge as Substitute of Aggregate
0% Sludge 4% OPC + 16% PFA 80% RA 0.95 1841
12.5% Sludge 4% OPC + 16% PFA 70% RA + 10% Sludge 1.05 1823
25% Sludge 4% OPC + 16% PFA 60% RA + 20% Sludge 1.30 1695
50% Sludge 4% OPC + 16% PFA 40% RA + 40% Sludge 1.80 1473
100% Sludge 4% OPC + 16% PFA 80% Sludge 2.60 1135
(b) Accelerating Setting and Stiffening by Varying SCMs
Binder  2 8% OPC + 32% PFA 45% RA + 15% Sludge 0.79 1652
GGBS 4% OPC + 16% GGBS 60% RA + 20% Sludge 1.35 1655
ISSA 4% OPC + 16% ISSA 60% RA + 20% Sludge 1.75 1566
SF 4% OPC + 16% SF 60% RA + 20% Sludge 2.10 1481

W/B ratio: water to binder ratio; OPC: ordinary Portland cement; PFA: pulverized fly ash; RA: recycled fine aggregate; SCMs: supplementary cementitious materials; GGBS:
ground granulated blast furnace slag; ISSA: incinerated sewage sludge ash; SF: silica fume.

Table 3
Mixture formulation (wt%) for CLSM with various accelerators incorporation.

Label Binder Aggregate Additivea W/B ratio Density

(kg m 3)
(a) Accelerating Stiffening by Sodium Additives
NaOH 4% OPC + 16% PFA 70% RA + 10% Sludge NaOH 1.30 1660
NaCO3 4% OPC + 16% PFA 70% RA + 10% Sludge NaCO3 1.30 1660
Na2SiO3 4% OPC + 16% PFA 70% RA + 10% Sludge Na2SiO3 1.30 1660
NaAlO2 4% OPC + 16% PFA 70% RA + 10% Sludge NaAlO2 1.20 1860
NaHMP 4% OPC + 16% PFA 70% RA + 10% Sludge NaHMP 1.30 1660
(b) Accelerating Stiffening by Calcium Additives
CaO 4% OPC + 16% PFA 70% RA + 10% Sludge CaO 1.30 1660
CaCl2 4% OPC + 16% PFA 70% RA + 10% Sludge CaCl2 1.20 1860
Ca(NO3)2 4% OPC + 16% PFA 70% RA + 10% Sludge Ca(NO3)2 1.20 1660
Ca(OAc)2 4% OPC + 16% PFA 70% RA + 10% Sludge Ca(OAc)2 1.20 1660
(c) Accelerating Stiffening by Compound Additives
TEA 4% OPC + 16% PFA 70% RA + 10% Sludge TEA 1.30 1819
TEA + CaCl2 4% OPC + 16% PFA 70% RA + 10% Sludge TEA + CaCl2 1.10 1700
TEA + Ca(NO3)2 4% OPC + 16% PFA 70% RA + 10% Sludge TEA + Ca(NO3)2 1.10 1740
TEA + Ca(OAc)2 4% OPC + 16% PFA 70% RA + 10% Sludge TEA + Ca(OAc)2 1.10 1720

W/B ratio: water to binder ratio; OPC: ordinary Portland cement; PFA: pulverized fly ash; RA: recycled fine aggregate; NaHMP: sodium hexametaphosphate; TEA: tri-
ethanolamine; Ca(OAc)2: calcium acetate.
a
Additive dosages: 5 wt% of binder. For compound additives, the mass ratio of two chemicals is 1:1.
 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800
moderate amount of water was quickly added and mixed for
another 2 min to ensure a homogeneous mixture. In the flowability
determination stage, fresh mixture was introduced into a flow
cylinder to test the flowability according to the standard test
[29]. Water was gradually added until the flowability satisfied
the requirement of 210 mm ± 10 mm [12]. In the casting phase,
fresh mixture was poured into 50  50  50 mm steel moulds.
All samples were kept in moulds for 2 days for initial strength
development. After demoulding, samples were wrapped in a
waterproof membrane to retain moisture at 23 ± 2 °C for total
7-d or 28-d air curing for further analysis. All experiments were
conducted in triplicate and the average values were presented with
the variations.
2.3. Physical property and spectroscopic/microscopic analyses
To validate the applicability of the alum sludge derived CLSM,
the flowability was determined by measuring the spread range of
CLSM from a fixed-volume flow cylinder [29]. The stiffening time
was examined by testing the resistance of the materials to the pen-
etration rod [30]. The resistances to penetration at the initial and
final stiffening time were 0.5 N mm 2 and 3.5 N mm 2, respec-
tively. The mechanical strength of CLSM products after 7-d or
28-d curing was characterized in terms of unconfined uniaxial com-
pressive strength test [31] by a multi-function mechanical testing
machine (Testometric CXM 500-50 KN) at a loading rate of
0.5 mm min 1. The water absorption of samples after 24-h water
immersion [32] was measured to assess the dimensional stability.
Hydration heat of fresh mixture and mineral composition of
products were analysed to elucidate the macro performance of
CLSM. The hydration heat for the pastes was measured with
isothermal calorimeter (Calmetrix I-CAL 4000). The operating
Fig. 1. Stiffening time and compressive strength of CLSM at varying sludge
replacement.
795
temperature of the instrument was maintained at 25 °C. The pastes
contained all the raw materials except recycled fine aggregate. The
morphology of CLSM products were observed by scanning electron
microscopy with energy dispersive X-ray spectroscopy (SEM-EDX,
TESCAN VEGA3 XM). Gold was coated on the powdered samples to
ensure good electrical conductivity and SEM was operated at 20 kV
voltage. Thermogravimetric analysis (TGA) of raw sludge and sam-
ple powders was performed from 100 to 1000 °C at a heating rate
of 10 °C min 1 with Ar stripping gas (TGA, Rigaku Thermo Plus).
The crystalline-phase mineralogy of samples was evaluated by
Quantitative X-ray diffraction (QXRD) using a high-resolution
powdered X-ray diffractometer (XRD, Rigaku SmartLab). The CaF2
powder (20 wt%) as internal standard was incorporated into the
samples to determine the content of amorphous phase. The mixed
powders were scanned within a 2h range of 15° to 50° with 3°
min 1. Rietveld refinement quantitative phase was analysed using
the whole powder pattern fitting (WPPF) method in Rigaku’s XRD
software (PDXL).
3. Results and discussion
3.1. Effects of alum sludge replacement
Fig. 1a illustrates that the stiffening time of CLSM dramatically
extended with increasing alum sludge replacement ratio. The
Fig. 2. Rate of hydration heat and cumulative of hydration heat of CLSM at varying
sludge replacement.
796 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800

12.5% sludge incorporation lengthened final stiffening time from

19.3 h to 29.8 h, which exceeded the minimal requirement of 24
h [13]. This phenomenon could mainly result from the high water
absorption of alum sludge. To maintain self-levelling property (fl
owability > 200 mm), alum sludge addition significantly increased
water demand due to high content of clay (Table 2). As a result,
the stiffening time of CLSM was significantly delayed [33,34]. High
content of organic matter in sludge (21.6% of LOI) also interfered
cement hydration, because it would hamper the CH formation
and pH increase [35,36]. Consequently, the compressive strength
significantly decreased along with increasing sludge replacement
ratio (Fig. 1b). Only 12.5% sludge replacement led to the 75% reduc-
tion in 28-d compressive strength, although all the samples ful-
filled the strength requirement of 0.3 to 2.1 MPa [12]. Excessive
sludge addition resulted in high water absorption of CLSM prod-
ucts and even dissolution after 1-d water immersion (Fig. S2). Thus,
12.5% and 25% sludge incorporated CLSM was investigated in the
subsequent experiments.
As shown in the curve of hydration heat rate (Fig. 2a), there were
five periods in the pure binder samples (0% sludge), namely initial
reaction period, induction period, acceleration period, deceleration
period, and slow reaction period as described in the literature [33].
After the induction period, the first humps caused by the silicate Fig. 4. XRD spectra of 3-d air cured CLSM at varying sludge replacement.
hydration (C3S) appeared at approximately 17 h [37], which aligned
with the final stiffening time (Fig. 1a). The second dumps occurring dissolution, it significantly delayed the induction period. Two
at approximately 28 h presumably corresponded to the renewed dumps distinctively decreased and shifted to long curing time when
reaction of aluminate phases [38]. Although addition of alum sludge there was 12.5% sludge replacement (Fig. 2). This revealed that alum
enhanced heat release in the initial reaction period via sludge sludge addition significantly hampered cement hydration, account-
ing for the observation of long stiffening time and low early strength.
The TGA curves showed (Fig. 3) that the weight of raw sludge
continuously decreased with elevating temperature until 500 °C,
which was associated with the decomposition of organic matter
and water loss from alum products. In contrast, control sample

Fig. 5. Stiffening time and compressive strength of CLSM with different binder
Fig. 3. TG and DTG of 3-d air cured CLSM at varying sludge replacement. compositions.
 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800
(0% sludge) presented a distinctive peak at 410 °C corresponding to
the decomposition of Ca(OH)2, which is an important indicator of
cement hydration (Table S1) [39]. However, alum sludge replace-
ment led to the almost disappearance of CH peak even at low
replacement ratio (12.5%). This verified that organic matter from
alum sludge interfered the CH formation and impeded further
cement hydration [40]. The XRD spectra (Fig. 4) demonstrated
the remarkable peak of CH (18.4°) nearly disappeared in sludge
incorporated samples compared to control sample. The QXRD
results quantified that the CH content declined from 6.7% to 1.1%
after 12.5% sludge replacement, which corroborated the TGA
results. A gentle peak of unreacted C3S (31.1°) and an obvious peak
of unreacted dicalcium silicate (C2S) (32.2°) were identified in
sludge incorporated samples, which corresponded to long
stiffening time and low mechanical strength. Hence, to promote
cement hydration and stiffening, supplementary cementitious
materials were needed to produce sludge-derived CLSM in the next
section.
Fig. 6. Morphology of 3-d air cured CLSM with different binder formation.
797
3.2. Significance of binder formulation
As shown in Fig. 5a, at 25% sludge replacement, the doubled
binder content slightly shortened the final stiffening time by
18.6%, and the stiffening time was still unable to fulfil the require-
ment [13]. However, the compressive strength was significantly
increased to nearly 1 MPa (Fig. 5b), which would increase future
excavation cost as mechanical excavators are required. Unexpect-
edly, the addition of supplementary cementitious materials
resulted in longer stiffening time compared to PFA samples
(Fig. 5a). The hydration heat curves (Fig. S3) verified that the alter-
native SCMs did not facilitate cement hydration, which could be
attributed to the high water demand of these SCMs for maintaining
sufficient flowability (Table 2) [34].
As illustrated in the SEM image (Fig. 6a), irregular silt and clay
particles (<10 mm in size) were present in the raw alum sludge,
whereas the PFA particles showed smooth and spherical structure
(Fig. 6b), which had a lower water demand to achieve good
798 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800
flowability. Higher binder content facilitated the agglomeration of
solid materials (Fig. 6c), which may facilitate the formation of
dense structure with short stiffening time (Fig. 5a). Compared to
PFA samples, GGBS samples contained abundant sheets of parti-
cles, which required more water to maintain sufficient flowability.
The high water-to-solid ratio hampered stiffening, although the
spontaneous hydraulic reaction of GGBS may alleviate the delayed
effect. The addition of irregular and porous ISSA significantly
increased the water demand, corresponding to longer stiffening
time. For SF, despite its spherical structure, the particle size was
<0.5 mm (Fig. 6f), thus its large specific surface area resulted in high
water absorption. To achieve the flowability of 210 mm, the water-
to-binder of SF samples increased by 62% compared to PFA sample,
which was associated with stiffening time extension. This demon-
strated that the water demand of binders played an important role
in the stiffening time [34]. Considering relatively short stiffening
time, PFA was regarded as applicable SCMs for CSLM products.
However, the stiffening time of sludge-derived CLSM still failed
to meet the requirement. Thus, in following experiments, the alum
sludge replacement was reduced to 12.5% and various accelerators
were incorporated to shorten the stiffening time.
3.3. Performance enhancement by accelerators
Fig. 7 illustrates the mixed effects of incorporating additives
(5 wt% of binder) into the 12.5% sludge-derived CLSM. Sodium-
based admixtures could facilitate the dissolution of calcium silicate
and calcium aluminate [23], and Na2SiO3 and NaAlO2 shortened
the initial stiffening time but they could not effectively
reduce the final stiffening time (Fig. 7a). This was probably
ascribed to the pH buffering capacity of alum sludge and
sodium-calcium aluminate hydration, which is adverse for early
strength development [41].
Fig. 7. Stiffening time and compressive strength of CLSM with various accelerators
incorporation.
By contrast, Ca-based admixtures effectively shortened initial
and final stiffening time, especially CaCl2. The final stiffening time
of Ca-incorporated CSLM was shorter than the standard require-
ment of 24 h in Hong Kong [13]. The oversaturated Ca2+ ion from
additives could surround C3S and C3A particles as active sites and
induce the generation and growth of calcium silicate hydrate
(CASAH) gel [24]. Moreover, chloride ions from CaCl2 would react
with C3A and facilitate the precipitation of insoluble calcium
chloroaluminate hydrates. Such exothermic chemical reactions
could accelerate cement hydration [36,42].
The addition of organic accelerator TEA was also found favour-
able. After 5 wt% TEA addition, the initial and final stiffening time
reduced by 82% and 75%, respectively. This may be attributed to
the acceleration effect of TEA on the reaction between C3A and
gypsum to form Aft [25]. However, the dosage of TEA should be
delicately balanced because TEA addition would retard the hydra-
tion of C3S and reduce early strength [43]. Combined addition of
TEA and Ca-based additives, there was no obvious synergistic effect
on the reduction of stiffening time. All the studied accelerators
imposed marginal effects on the final compressive strength
(Fig. 7b), which was in line with previous findings [36,41,44].
Considering the moderate compressive strength of TEA and
CaCl2 added samples (approximately 0.6 MPa), the long-term
strength of CLSM products could fulfil the required strength range
(0.3–1 MPa) [12]. Hence, individual use of CaCl2 or TEA was
regarded as an effective option.
Fig. 8. Rate of hydration heat and cumulative of hydration heat of CLSM with
various accelerators incorporation.
 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800
Fig. 9. TG and DTG of 3-d air cured CLSM with various accelerators incorporation.
The incorporation of accelerators, CaCl2 in particular, promoted
the initial reaction period during first 2 h (Fig. 8a), and it also
enhanced the hydration heat release during cement hydration
(Fig. 8b). This evidenced that these admixtures accelerated cement
hydration. Among them, NaOH samples presented the highest heat
release in initial reaction period because of NaOH dissolution, but
it was surpassed by other samples afterwards, which corroborated
the minimal effect of NaOH for the reduction of stiffening time
(Fig. 7a). From TG curves (Fig. 9a), the mass loss from hydrates
decomposition in the accelerators incorporated samples were
higher than that of the control samples. However, it should be
noted that the remarkable decomposition peaks of sodium hydrox-
ide at 230 °C and TEA at 420 °C could be observed in NaOH and TEA
samples, respectively (Table S1). This implied that large amounts of
unreacted accelerators remained or adsorbed in alum sludge.
Therefore, the optimal dosage of promising accelerators (CaCl2
and TEA) should be further investigated to produce eco-friendly
CLSM in future studies.
4. Conclusions
This study demonstrated an innovative method that success-
fully utilized alum sludge for producing eco-friendly CLSM. The
addition of silt/clay-rich alum sludge in CLSM dramatically
increased water demand, resulting in long stiffening time and
low compressive strength. High content of organic matter in alum
sludge significantly delayed the induction period of cement hydra-
tion. In the binder system, PFA used as SCM could effectively
799
reduce water demand and provide moderate compressive strength.
The addition of Ca-based admixtures notably shortened stiffening
time, whereas organic TEA was the most effective accelerator and
reduced the initial and final stiffening time by 82% and 75%, respec-
tively. The results suggested that PFA-based binder system with
TEA addition was a feasible mixture for transforming alum sludge
into CLSM. This study showcased a novel engineering solution
based on hydration chemistry for transforming alum sludge into
value-added construction materials. In future studies, the optimal
dosage of accelerators in field trial and the durability of CLSM
under varying environmental conditions should be further
evaluated.
Acknowledgement
The authors appreciate the financial support from the Hong
Kong Water Supplies Department (project account: P15-0569) for
this study.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.conbuildmat.2018.
01.078.
References
[1] K.B. Dassanayake, G.Y. Jayasinghe, A. Surapaneni, C. Hetherington, A review on
alum sludge reuse with special reference to agricultural applications and
future challenges, Waste Manage. 38 (2015) 321–335.
[2] HK EPD, Monitoring of Solid Waste in Hong Kong: Waste Statistics for 2015.
Environmental Protection Department, Hong Kong, 2017.
[3] M.A. Tantawy, Characterization and pozzolanic properties of calcined alum
sludge, Mater. Res. Bull. 61 (2015) 415–421.
[4] S.R. Teixeira, G.T.A. Santos, A.E. Souza, P. Alessio, S.A. Souza, N.R. Souza, The
effect of incorporation of a Brazilian water treatment plant sludge on the
properties of ceramic materials, Appl. Clay Sci. 53 (2011) 561–565.
[5] Y.P. Ling, R.H. Tham, S.M. Lim, M. Fahim, C.H. Ooi, P. Krishnan, A. Matsumoto, F.
Y. Yeoh, Evaluation and reutilization of water sludge from fresh water
processing plant as a green clay substituent, Appl. Clay Sci. 143 (2017) 300–
306.
[6] C. Martínez-García, D. Eliche-Quesada, L. Pérez-Villarejo, F.J. Iglesias-Godino, F.
A. Corpas-Iglesias, Sludge valorization from wastewater treatment plant to its
application on the ceramic industry, J. Environ. Manage. 95 (2012).
[7] A. Sales, F.R. De Souza, F.D.C.R. Almeida, Mechanical properties of concrete
produced with a composite of water treatment sludge and sawdust, Constr.
Build. Mater. 25 (2011) 2793–2798.
[8] M. Frías, R. Vigil De La Villa, I. De Soto, R. García, T.A. Baloa, Influence of
activated drinking-water treatment waste on binary cement-based composite
behavior: Characterization and properties, Compos. Part B Eng. 60 (2014) 14–
20.
[9] H. El-Didamony, K.A. Khalil, M. Heikal, Physico-chemical and surface
characteristics of some granulated slag–fired drinking water sludge
composite cement pastes, HBRC J. 10 (2014) 73–81.
[10] M. Bassani, F. Bertola, M. Bianchi, F. Canonico, M. Marian, Environmental
assessment and geomechanical properties of controlled low-strength
materials with recycled and alternative components for cements and
aggregates, Cem. Concr. Compos. 80 (2017) 143–156.
[11] ACI 116R, 2000. Cement and concrete terminology. American Concrete
Institute, 2000.
[12] ACI 229R, 1999. Controlled low-strength materials. American Concrete
Institute, 1999.
[13] HK HD, 2016. Guidance notes 014C, Permanent trench reinstatement.
Highways Department, Hong Kong, November 2016.
[14] L. Wang, D.C.W. Tsang, C.S. Poon, Green remediation and recycling of
contaminated sediment by waste-incorporated stabilization/solidification,
Chemosphere. 122 (2015) 257–264.
[15] L. Wang, J.S.H. Kwok, D.C.W. Tsang, C.S. Poon, Mixture design and treatment
methods for recycling contaminated sediment, J. Hazard. Mater. 283 (2015)
623–632.
[16] E. Jeon, S. Ryu, S. Park, L. Wang, D.C.W. Tsang, K. Baek, Enhanced adsorption of
arsenic onto alum sludge modified by calcination, J. Clean. Prod 176 (2017)
54–62, https://doi.org/10.1016/j.jclepro.2017.12.153.
[17] D.C.W. Tsang, A.C.K. Yip, W.E. Olds, P.A. Weber, Arsenic and copper
stabilisation in a contaminated soil by coal fly ash and green waste compost,
Environ. Sci. Pollut. Res 21 (2014) 10194–10204.
[18] C.B. Cheah, K.Y. Chung, M. Ramli, G.K. Lim, The engineering properties and
microstructure development of cement mortar containing high volume of
800 L. Wang et al. / Construction and Building Materials 165 (2018) 792–800
inter-grinded GGBS and PFA cured at ambient temperature, Constr. Build.
Mater. 122 (2016) 683–693.
[19] J. Li, L. Wang, D.C.W. Tsang, Jingzi Beiyuan, C.S. Poon,, Dynamic leaching
behavior of geogenic as in soils after cement-based stabilization/solidification,
Environ. Sci. Pollut. Res. 24 (2017) 27822–27832.
[20] Y. Arail, B.A. Powell, D.I. Kaplan, Sulfur speciation in untreated and alkali
treated ground-granulated blast furnace slag, Sci. Total Environ. 589 (2017)
117–121.
[21] J.S. Li, J. Beiyuan, D.C.W. Tsang, L. Wang, C.S. Poon, X.D. Li, S. Fendorf, Arsenic-
containing soil from geogenic source in Hong Kong: leaching characteristics
and stabilization/solidification, Chemosphere 182 (2017) 31–39.
[22] A.R. Goodarzi, H.R. Akbari, M. Salimi, Enhanced stabilization of highly
expansive clays by mixing cement and silica fume, Appl. Clay Sci. 132–133
(2016) 675–684.
[23] J. Han, K. Wang, J. Shi, Y. Wang, Influence of sodium aluminate on
cement hydration and concrete properties, Constr. Build. Mater. 64 (2014)
342–349.
[24] T.B. Min, I.S. Cho, W.J. Park, H.K. Choi, H.S. Lee, Experimental study on the
development of compressive strength of early concrete age using calcium-
based hardening accelerator and high early strength cement, Constr. Build.
Mater. 64 (2014) 208–214.
[25] Y.L. Yaphary, Z. Yu, R.H.W. Lam, D. Lau, Effect of triethanolamine on
cement hydration toward initial setting time, Constr. Build. Mater. 141
(2017) 94–103.
[26] ASTM D7348, Standard Test Methods for Loss on Ignition (LOI) of Solid
Combustion Residues. ASTM D7348, 2013.
[27] US EPA, Method 1311, Toxicity Characteristic Leaching Procedure, U.S. Envi-
ronmental Protection Agency, Washington, DC, USA, 1992.
[28] HK EPD, Practice Guide for Investigation and Remediation of Contaminated
Land. Environmental Protection Department, Hong Kong, 2011.
[29] ASTM D6103, Standard Test Method for Flow Consistency of Controlled Low
Strength Material (CLSM). ASTM Standard D6103, 2004.
[30] BS EN 13294, Products and systems for the protection and repair of concrete
structures -Test methods -Determination of stiffening time. British Standards
Institution, London, 2002.
[31] BS EN 12390-3, Testing hardened concrete Part 3: Compressive strength of test
specimens, 2009
[32] BS 1881-122, Testing concrete Part 122: Method for determination of water
absorption. British Standards Institution, London, 2011.
[33] J. Hu, Z. Ge, K. Wang, Influence of cement fineness and water-to-cement ratio
on mortar early-age heat of hydration and set times, Constr. Build. Mater. 50
(2014) 657–663.
[34] J. Zhang, T. Fan, H. Ma, Z. Li, Monitoring setting and hardening of concrete by
active acoustic method: Effects of water-to-cement ratio and pozzolanic
materials, Constr. Build. Mater. 88 (2015) 118–125.
[35] L. Wang, S.S. Chen, D.C.W. Tsang, C.S. Poon, K. Shih, Recycling contaminated
wood into eco-friendly particleboard using green cement and carbon dioxide
curing, J. Clean. Prod. 137 (2016) 861–870.
[36] L. Wang, S.S. Chen, D.C.W. Tsang, C.S. Poon, K. Shih, Value-added recycling of
construction waste wood into noise and thermal insulating cement-bonded
particleboards, Constr. Build. Mater. 125 (2016) 316–325.
[37] K. Hoang, H. Justnes, M. Geiker, Early age strength increase of fly ash blended
cement by a ternary hardening accelerating admixture, Cem. Concr. Res. 81
(2016) 59–69.
[38] T. Hemalatha, M. Mapa, N. George, S. Sasmal, Physico-chemical and
mechanical characterization of high volume fly ash incorporated and
engineered cement system towards developing greener cement, J. Clean.
Prod. 125 (2016) 268–281.
[39] L. Wang, S.S. Chen, D.C.W. Tsang, C.S. Poon, J.G. Dai, CO2 curing and fibre
reinforcement for green recycling of contaminated wood into high-
performance cement-bonded particleboards, J. CO2 Util. 18 (2017) 107–116.
[40] L. Wang, T.L.K. Yeung, A.Y.T. Lau, D.C.W. Tsang, C.S. Poon, Recycling
contaminated sediment into eco-friendly paving blocks by a combination of
binary cement and carbon dioxide curing, J. Clean. Prod. 164 (2017) 1279–
1288.
[41] R.P. Salvador, S.H.P. Cavalaro, I. Segura, A.D. Figueiredo, J. Pérez, Early age
hydration of cement pastes with alkaline and alkali-free accelerators for
sprayed concrete, Constr. Build. Mater. 111 (2016) 386–398.
[42] L. Wang, I.K.M. Yu, D.C.W. Tsang, S. Li, C.S. Poon, Mixture Design and Reaction
Sequence for Recycling Construction Wood Waste into Rapid-Shaping
Magnesia-Phosphate Cement Particleboard, Ind. Eng. Chem. Res. 56 (2017)
6645–6654.
[43] J. Han, K. Wang, J. Shi, Y. Wang, Mechanism of triethanolamine on Portland
cement hydration process and microstructure characteristics, Constr. Build.
Mater. 93 (2015) 457–462.
[44] N. Shanahan, A. Sedaghat, A. Zayed, Effect of cement mineralogy on the
effectiveness of chloride-based accelerator, Cem. Concr. Compos. 73 (2016)
226–234.