Journal of Environmental Management 92 (2011) 2085e2090

Contents lists available at ScienceDirect

Journal of Environmental Management

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

Reuse of de-inking sludge from wastepaper recycling in cement mortar products

Shiqin Yan*, Kwesi Sagoe-Crentsil, Gretta Shapiro
CSIRO Materials Science and Engineering, PO Box 56, Highett, Victoria 3190, Australia

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

Article history: This paper presents results of an investigation into the use of de-inking sludge from a paper recycling
Received 15 September 2010 mill as feedstock material in the manufacture of cement mortar products, including masonry blocks and
Received in revised form mortar renders. Both physical and mechanical properties of mortar specimens containing various
28 February 2011
amounts of de-inking sludge were investigated. It was observed that the addition of de-inking sludge to
Accepted 24 March 2011
cement mortar at a fixed water-to-cement ratio significantly reduced flow properties and increased
Available online 19 April 2011
setting time. Water absorption and volume of permeable voids of cement mortar increased with
increased dosage of de-inking sludge, with a corresponding reduction of bulk density. The 91-day
Keywords:
De-inking sludge
compressive strength of mortar samples with 2.5 wt% and 20 wt% de-inking sludge loadings retained 83%
Cement mortar and 62% respectively of the reference mortar strength. The corresponding drying shrinkage increased by
Compressive strength up to 160% compared to reference samples. However, a de-inking sludge loading of up to 2.5 wt% did not
Drying shrinkage significantly alter measured physical and mechanical properties. The results demonstrate that despite
Environmental management the high moisture absorbance of de-inking sludge due to its organic matter and residual cellulose fibre
content, it serves as a potential supplementary additive and its cellulosic content proving to be an active
set retardant to cementitious masonry products.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction the ink waste residue as biomass resource while examining uses for
metakaolin by-products generated after burning. Cement mortar
Filter cake waste is a by-product derived from effluent treatment products incorporating up to 20% of this highly reactive metakaolin
processes, typically in cardboard processing industries. The material showed improved durability on freeze and thaw cycle exposures
largely comprises of ink waste, generally with 60% moisture content, (Vegas et al., 2009). Ishimoto et al. (2000) reported that the acid
and varying amounts of Si, Al, Cu, Ti and Ca as well as organic matter resistivity of concrete was enhanced when incinerated paper sludge
(Pera and Amrouz, 1998; Behin and Vahed, 2007; Mendez et al., ash was used as an admixture in concrete.
2009). Several thousand tonnes of this by-product material is Another potential option of reusing paper sludge to produce
generated annually in Australia, with the State of Victoria alone lightweight aggregates for non-structural concrete manufacturing
generating in excess of 1300 t of de-inking sludge each year, which is was reported by Liaw et al. (1998). Cement products made with these
largely sent to landfill. Accordingly, effective recycling of paper mill lightweight aggregates have demonstrated that some physical
sludge into building products or other reuse applications remains an properties, such as water absorption, density and abrasion loss, could
attractive environmental option, as also indicated by previous either be similar or superior to commercial lightweight aggregates for
researchers (Pera and Amrouz, 1998; Ishimoto et al., 2000; Liaw non-structural concrete, while other properties, such as compressive
et al., 1998; Naik, 2002; Frías et al., 2008). strength and flexural strength, decreased (Liaw et al., 1998).
The drive to find reuse options for sludge waste has resulted in Limited research has, however, been conducted on direct use of
recent investigations into the potential recycling of paper sludge as de-inking sludge (filter cake) as opposed to biomass processing, i.e.
supplementary additive to cement based construction products. without additional processing in cement products. Naik (2002)
Many studies have been carried out recently on producing a highly investigated concrete properties containing small amounts of de-
pozzolanic material (metakaolin) from calcination of paper sludge inking sludge and found that the concrete exhibited improved
under controlled condition (Pera and Amrouz, 1998; Pera et al., resistance to chloride-ion penetration and freezing and thawing
2001; Villa et al., 2007; Frías et al., 2008). These studies considered action without significant loss of strength. Available data on the
physical and mechanical properties of concrete directly incorpo-
rating de-inking sludge remains however limited and is inadequate
* Corresponding author. Tel.: þ61 3 9252 6294; fax: þ61 3 9252 6244. to guide any potential beneficial pilot-scale reuse trials of this
E-mail address: [email protected] (S. Yan). resource. This study therefore aims to extend the prospects of

0301-4797/$ e see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jenvman.2011.03.028
2086 S. Yan et al. / Journal of Environmental Management 92 (2011) 2085e2090

reusing filter cake sludge from a paper recycling mill as raw feed- Table 1
stock for manufacturing building products by evaluating the Chemical composition of de-inking sludge.

fundamental effects of sludge additives on the physical and Component Mass percentage
mechanical properties of cement mortar systems. Moisture content 68.0
Loss on ignition 16.9
2. Experimental SiO2 2.0
Al2O3 1.98
CaO 2.69
2.1. Characterisation of de-inking sludge TiO2 3.88
BaO 0.65
The de-inking sludge material used in this work was sourced CuO 0.31
from a major waste treatment plant in Victoria, Australia. Thermal S2O3 0.14
Fe2O3 0.11
and evolved gas analysis was conducted using a Setaram combined Na2O 0.04
TGA/DTA thermal analyser coupled to a Pfeiffer mass spectrometer.
Moisture loss on drying (LOD) and loss on ignition (LOI) were
determined after heating samples at 110  C and 900  C respectively.
Elemental analysis was carried out on the post-LOI ash by induc- procedure of AS 1012.13e1992 (Determination of the Drying
tively coupled plasma mass spectrometry (ICP-MS) after microwave Shrinkage of Concrete for Samples Prepared in the Field or in
decomposition of the material in a mixture of nitric, hydrochloric the Laboratory). Drying shrinkage value is the average results
and hydrofluoric acids. X-ray diffraction (XRD) was also carried out of three mortar specimens.
to identify mineralogical phases in raw dry de-inking sludge.  ASTM C 642e06 (Standard Test Method for Density, Absorp-
tion, and Voids in Hardened Concrete) was adopted for water
2.2. Materials and methods absorption, bulk density and volume of permeable voids tests.
Each test result constitutes the average values of minimum
A general-purpose ordinary Portland cement (Type GP cement) three mortar samples.
complying with Australia Standard AS 3972 (Portland and Blended
Cements) and regular river sand were used in this study. The sand 3. Results and discussion
was dried before use to ensure exact moisture levels in mix
formulations. Mortar samples were prepared in accordance with AS 3.1. De-inking sludge characterisation
2350.12e2006 (Preparation of Standard Mortar Specimens). The
sand-to-cement weight ratio was typically set at 3. To maintain The results of the ICP-MS and X-ray diffraction pattern are
uniformity in sample casting, “as-received” wet sludge filter cake shown in Table 1 and Fig. 1 respectively. Table 1 shows that the as-
was incorporated as an additional material, on a weight percentage received de-inking sludge contains about 68% moisture, 16.9%
(wt%) of sand basis. The water contained in the sludge was included organic matter (cellulose) and 15.1% ash. The mineralogical phases
in calculating the overall water-to-cement ratio (w/c) of a mortar present in the raw dry de-inking sludge are mainly calcite (Ca),
mixture. As the introduction of a relatively high moisture content rutile (R), filler clays (F) and small amount of carbon (C).
de-inking sludge into mortar mix changes flow behaviour, Fig. 2 shows the result of TGA/DTA analysis. The TGA analysis
preliminary tests were carried out to determine a reference w/c for revealed a total weight loss of about 84.9% from ambient to 900  C,
sample preparation. over four distinct temperature ranges. The major weight loss of
Slurries were prepared by initially dispersing the sludge in water water, i.e. 68.0%, occurred in the range from ambient to 200  C. The
in a 10L Hobart mixer, before mixing with cement for about 30 s. second distinct weight loss 7.12% spanned 200e380  C, and
River sand was then added with continued mixing for another 60 s. corresponded to the evolution of Cl2, SO2 and some higher
The flow and the setting time of the mortar slurry were measured in
accordance with ASTM C 1437e07 (Standard Test Method for Flow of
Hydraulic Cement Mortar) and AS 1012.18e1996 (Determination of 1400
setting time of fresh concrete, mortar and grout by penetration R F - Filler clays
resistance), respectively. Ca R - Rutile
1200
The mortar slurry was cast into 50 mm steel cube moulds for Ca - Calcite
compressive strength testing, and 160  40  40 mm moulds for C - Carbon
drying shrinkage measurements. The cast samples were vibrated 1000
Intensity (counts)

for 1 min to achieve adequate compaction, and initially cured at

ambient temperature under general laboratory conditions (23  C 800
and 50% RH) for 24 h. The specimens were then demoulded and
stored in a standard conditioning room (23  C and 100% RH) prior R
to testing. 600
R
The following tests were conducted:
400
 Compressive strength: Samples were tested under saturated F F R
F Ca Ca
surface-dry conditions in accordance with AS 1012.9e1997 F FCaC Ca Ca R
200 Ca
(Determination of the Compressive Strength of Concrete R R Ca FR R
R
Specimens) at a loading rate of 20 MPa/min. Tests were carried
out at 1, 7, 28 and 91 days after casting. Compressive strength 0
value under each condition constitutes the average results of 0 10 20 30 40 50 60 70 80 90
three to four mortar samples.
2-Theta
 Drying shrinkage: Shrinkage measurements were performed at
regular intervals up to 91 days in accordance with a modified Fig. 1. X-ray diffraction pattern of dry de-inking sludge.
 S. Yan et al. / Journal of Environmental Management 92 (2011) 2085e2090 2087

150
0 150 300 450 600 750 900
0 170
120
130
-15

Flow (%)
90 90

Heat flow (mW)

-30
50
TG (%)

DTA plot 60
-45 10

-30 30
-60
-70
-75 0
TGA plot -110
0 5 10 15 20 25
-90 -150
De-inking sludge content (wt%)
Fig. 2. DTA/TGA curves of the filter cake sludge.
Fig. 4. Flow of cement mortar vs. de-inking sludge content at w/c ¼ 0.7.

molecular weight species, such as cyclohexane with alkane func- In this work, adequate flow in both reference and sludge-con-
tional groups, carbonyl and branched alkanes. These species repre- taining samples was achieved at a w/c of 0.7. Fig. 4 shows that the
sented fragmentation ions from the breakdown of chlorinated and flow of mortar mixtures decreased markedly from 140% to 25% with
sulphonated hydrocarbon species. The third weight loss (7.76%) the addition of 20 wt% de-inking sludge.
occurred in the range 380e520  C, and was mainly caused by carbon Fig. 5 shows variation of the mortar setting time with de-inking
in the sample oxidising to CO2, which corresponds to the large sludge content at w/c of 0.7. De-inking sludge significantly
exotherm on the DTA curve in Fig. 2. The last distinct weight loss lengthened both initial and final setting time of cement mortar.
(2.06%) occurred at 520e900  C (corresponding to a small endo- With the addition of 20 wt% de-inking sludge in cement mortar, the
therm on the DTA curve), and is likely due to the calcination of initial and final setting time were respectively delayed by a factor of
residual carbonate species present in the ash. 3.6 and 4.4, compared to the reference mortar mix. This indicates
de-inking sludge can be used as a retarding admixture. It must be
recalled that the organic matter in de-inking sludge is dominantly
3.2. Effects of de-inking sludge on mortar flow and setting time cellulose (Pera and Amrouz, 1998; Vegas et al., 2009), which is well-
known as a retarding concrete admixture (Ramachandran, 1995).
Introducing de-inking sludge into a cement mortar mixture
changes the flow behaviour of the wet slurry, and therefore
3.3. Compressive strength
preliminary tests were carried out to determine a reference w/c.
Fig. 3 shows the variation in mortar flow versus w/c for both
The results from compressive strength testing at 1, 7, 28 and 91
reference (without de-inking sludge) and 10 wt% de-inking sludge
days are shown in Fig. 6, and it is clear that compressive strength
samples. The results demonstrate that de-inking sludge can absorb
decreased with increasing de-inking sludge content up to 20 wt%.
significant quantities of water in the mix, causing a reduction in
The average 91-day compressive strengths of mortar samples
mortar flow. The organic matter constituents and residual cellulose
incorporating 2.5 and 20 wt% de-inking sludge were 27.6 and
fibres in the de-inking sludge are believed to be the primary cause
of this observed high water absorption (Ahmadi and Al-Khaja,
2001; Behin and Vahed, 2007). 2500
Initial setting time

Final setting time

2000
180
Setting time (min)

Reference
150 10% de-inking sludge
1500
Flow (%)

120
1000
90

60
500
30

0 0
0.4 0.5 0.6 0.7 0.8 0.9 0 5 10 15 20 25
Water-cement ratio De-inking sludge content (%)
Fig. 3. Flow of hydraulic cement mortar versus w/c. Fig. 5. Effect of de-inking sludge content on mortar setting time at w/c ¼ 0.7.
2088 S. Yan et al. / Journal of Environmental Management 92 (2011) 2085e2090

35 0 5 10 15 20
-0.080
Compressive strength (MPa)

30

% change in length
25 -0.120

20
-0.160
15 Reference
2.5% sludge
5% sludge -0.200
10
10% sludge
15% sludge
5 -0.240
20% sludge

0
-0.280
0 10 20 30 40 50 60 70 80 90 100
Mortar age (days) De-inking sludge content (wt%)
Fig. 6. Compressive strength development with time (w/c ¼ 0.7). Fig. 8. Drying shrinkage at 91 days of mortar samples with different de-inking sludge
content.

20.5 MPa respectively, corresponding to 83 and 62% of the refer- sludge was 22.3 MPa, i.e. 82% of the reference mortar strength
ence strength (33.2 MPa). (27.3 MPa), which is in line with Naik’s results.
The initial strength development rate up to 7 days appeared
relatively unaffected by the de-inking sludge loading, and a close
3.4. Drying shrinkage
similarity in the rate of strength development between 7 and 91
days was also observed. Thus, while the organic matter loading in
Drying shrinkage is an important durability parameter, because
the sludge appeared to alter the fresh mortar properties, in
of its implication on potential crack formation in the finished
particular the flow characteristics, there were no observed effects
product. Concrete drying shrinkage is mainly induced by the
on the long-term kinetics of cement hydration and properties of the
cement paste, and can lead to debonding at the cement paste/
hardened paste.
aggregate interfaces and/or to the development of intergranular
Fig. 6 further shows that only small variations in compressive
cracks (de Sa et al., 2008).
strength were achieved for de-inking sludge loadings between 10
The results of drying shrinkage measurements in the current
and 20 wt%.
study are shown in Fig. 7. The drying shrinkage of samples
Naik (2002) investigated the effects of adding de-inking residual
increased significantly with increasing de-inking sludge content. At
solids with 50% moisture content to concrete, and found that the
2.5 and 20 wt% de-inking sludge content, the drying shrinkage of
28-day compressive strength of concrete containing up to 1.2 wt%
samples corresponded to 1169 and 2547 microstrains at 91 days,
de-inking residual solids (by weight of concrete) was between 45
relatively corresponding to w21% and w160% increase respectively
and 75% of the corresponding reference strength. The total dry
compared to the reference mortar measured at 964 microstrains.
sludge content of 1.2 wt% (by weight of total concrete mix) in Naik’s
A linear relationship was observed when the measured drying
mix equates to 20% higher than the equivalent dry sludge of 5.0 wt%
shrinkage at 91 days was plotted against de-inking sludge loading
de-inking sludge loading used in the current work (1.0 wt% dry
(Fig. 8), suggesting that the addition of de-inking sludge material in
sludge by weight of mortar mix). As shown in Fig. 6, the 28-day
the samples initially caused high water absorption, as discussed in
compressive strength of mortar samples with 5.0 wt% de-inking
Section 3.2, which was released with time and subsequently lost,

0 10 20 30 40 50 60 70 80 90 100
12.5

0.000
12
Water absortion (%)

-0.050
% change in length

11.5
-0.100

-0.150 11

-0.200
10.5 Absorption after immersion
Absorption after immersion & boiling
-0.250
Reference 2.5 wt% 5 wt%
10 wt% 15 wt% 20 wt%
10
-0.300 0 5 10 15 20 25
Age (days) De-inking sludge content (%)
Fig. 7. Variation of drying shrinkage with time (w/c ¼ 0.7). Fig. 9. Water absorption of mortar samples after immersion and boiling.
 S. Yan et al. / Journal of Environmental Management 92 (2011) 2085e2090 2089

22.5
Volume of permeable voids (%)

22

21.5

21

20.5

20
0 5 10 15 20
De-inking sludge content (%)
Fig. 10. Volume of permeable voids versus de-inking sludge content.

resulting in increased drying shrinkage. Hence, higher de-inking

sludge loading would result in greater free water absorption in the
sludge, and a higher drying shrinkage in the final product.

3.5. Water absorption, volume of permeable voids and bulk density

The water absorption, volume of permeable voids and bulk

density of the mortar cubes were measured and the results shown
in Figs. 9e11. From Figs. 9e11, the water absorption and permeable
voids increased with increasing de-inking sludge content, with
a corresponding reduction of bulk density.
The increase of permeable voids could be related to surfactants
such as fatty acids and sulphate salts that are used in froth flotation in
the de-inking process (Behin and Vahed, 2007). Dissolved lignins are
also a type of surfactant (Xu et al., 2009). These surfactants are well-
known as air-entraining admixtures in concrete (Ramachandran,
1995). Air entrainment improves the workability and consistency of
plastic concrete and reduces its bleeding and segregation, while Fig. 12. SEM photomicrographs of the fracture-face morphology of: (a) reference
improving the durability of concrete against freezing exposure sample, and (b) sample containing 15 wt% de-inking sludge.
(Ramachandran, 1995).
The correlation between water absorption (Fig. 9) and volume of measurements (Fig. 11) was somewhat different. A very large initial
permeable voids (Fig. 10) was evident, in that these water-related reduction in bulk density from 2200 to 2050 kg/m3 was observed
parameters both increased somewhat linearly with increasing for the 2.5 wt% de-inking sludge sample, with only marginal change
sludge loading. However, the trend observed in bulk density in measured bulk density thereafter. This observation may be due
to mortar aeration caused by organic matter in the sludge material,
which levelled off once a threshold aeration level was achieved. The
marginal reduction in bulk density when the sludge loading was
2200 over 5 wt% further supported the compressive strength results in
Fig. 6, which showed marginal strength reduction at sludge
contents of 10e20 wt%.
2150 For comparison purposes, Fig. 12 shows SEM photomicrographs
of the fracture-face morphology of a reference sample and a sample
containing 15 wt% de-inking sludge. The more porous structure
2100 evident in the sample containing the sludge suggests that the
addition of de-inking sludge induced mortar aeration.

2050 4. Conclusions

1. The addition of de-inking sludge in cement mortar significantly

2000 reduces flow properties at a fixed water-to-cement ratio. The
0 5 10 15 20 25 results have demonstrated that de-inking sludge re-absorbs
De-inking sludge content (%) significant amounts of water in the mix, largely due to the
organic matter and residual cellulose fibre content, leading to
Fig. 11. Variation of bulk density of mortar with de-inking sludge content. reduction in measured mortar flow properties.
2090 S. Yan et al. / Journal of Environmental Management 92 (2011) 2085e2090

2. De-inking sludge addition to cement mortar increases mortar References

setting time, indicating its potential use as a retarding chemical
admixture. Ahmadi, B., Al-Khaja, W., 2001. Utilization of paper waste sludge in the building
construction industry. Resources, Conservation and Recycling 32, 105e113.
3. The addition of 2.5 wt% de-inking sludge results in a reduction Behin, J., Vahed, Sh, 2007. Effect of alkyl chain in alcohol deinking of recycled fibres
of the mortar bulk density from 2200 to 2050 kg/m3. Further by flotation process. Colloids and Surfaces A: Physicochemical and Engineering
addition in excess of 2.5 wt% produces marginal reduction in Aspects 297, 131e141.
de Sa, C., Benboudjema, F., Thiery, M., Sicard, J., 2008. Analysis of microcracking
bulk density. induced by differential drying shrinkage. Cement and Concrete Composites 30,
4. The average compressive strengths of mortar samples incor- 947e956.
porating 2.5 and 20 wt% de-inking sludge were 27.6 and Frías, M., Rodriguez, O., Vegas, I., Vigil, R., 2008. Properties of calcined clay waste
and its influence on blended cement behaviour. Journal of American Ceramic
20.5 MPa respectively at 91 days, corresponding to 83 and 62% Society 91 (4), 1226e1230.
of the reference strength (33.2 MPa), well within range of Ishimoto, H., Origuchi, T., Yasuda, M., 2000. Use of papermaking sludge as new
predicted masonry product usage. material. Journal of Materials in Civil Engineering 12 (4), 310e313.
Liaw, C., Chang, H., Hsu, W., Huang, C., 1998. A novel method to reuse paper sludge
5. The measured drying shrinkages of mortar mixes incorporating
and co-generation ashes from paper mill. Journal of Hazardous Materials 58
2.5 and 20 wt% de-inking sludge correspond to 1169 and 2547 (1e3), 93e102.
microstrains respectively at 91 days, which represent w21% Mendez, A., Fidalgo, J.M., Guerrero, F., Gasco, G., 2009. Characterization and
and w160% increases respectively compared to the reference pyrolysis behaviour of different paper mill waste materials. Journal of Analytical
and Applied Pyrolysis 86, 66e73.
mortar (964 microstrains). Naik, T., 2002. Greener concrete using recycled materials. Concrete International 7,
6. Water absorption and volume of permeable voids increase with 49e53.
increased de-inking sludge content, and is associated with an Pera, J., Ambroise, J., Chabannet, M., 2001. Transformation of wastes into comple-
mentary cementing materials. In: Seventh CANMET/ACI International Conference
apparent drop in bulk density. on Fly Ash, Silica Fume, Slag and Natural Pozzolans in Concrete. SP199, pp. 459e75.
7. The addition of up to 2.5 wt% de-inking sludge in cement Pera, J., Amrouz, A., 1998. Development of highly reactive metakaolin from paper
mortar does not significantly affect its physical and mechanical sludge. Advanced Cement Based Materials 7, 49e56.
Ramachandran, V.S., 1995. Concrete Admixtures Handbook: Properties, Science, and
properties, strongly favouring potential of incorporating de- Technology, second ed. Noyes Publications, New Jersey.
inking sludge waste as a supplementary addition to cementi- Vegas, I., Urreta, J., Frías, M., García, R., 2009. Freezeethaw resistance of blended
tious masonry products. cements containing calcined paper sludge. Construction and Building Materials
23, 2862e2868.
Villa, R., Frías, M., Rojas, M., Vegas, I., García, R., 2007. Mineralogical and morpho-
Acknowledgements logical changes of calcined paper sludge at different temperatures and retention
in furnace. Applied Clay Science 36, 279e286.
Xu, Q., Fu, Y., Gao, Y., Qin, M., 2009. Performance and efficiency of old newspaper
This work was conducted with the support of the EPA Victoria deinking by combining cellulase/hemicellulase with laccase-violuric acid
(Australia) through the Australian Industry Group. system. Waste Management 29, 1486e1490.