Paper Mill Sludge Ash as Supplementary

Cementitious Material
Gabriele Fava1; Maria Letizia Ruello2; and Valeria Corinaldesi3
Downloaded from ascelibrary.org by UFPR - Universidade Federal do Parana on 08/21/18. Copyright ASCE. For personal use only; all rights reserved.

Abstract: Paper mill sludge is often incinerated for heat recovery and volume reduction. In Italy, about 600,000 t of paper mill sludge is
produced yearly. About 6 kg of sludge is produced per ton of paper produced. The paper mill sludge is recovered from a primary mechanical
separation process. When it is combusted as a single fuel source, it produces paper mill sludge ash (PA). In this project, PA was studied to
evaluate its use as a supplementary cementitious material in mortars and concrete manufacturing. On the basis of the data collected, it was
concluded that PA, if used to replace up to 10% of the portland cement, shows a positive effect on the mechanical performance of mortars.
On the other hand, because of its high fineness and consequently high water absorption, it required a higher amount of water. It was
concluded that the use of PA should not be higher than 10% by weight of the cement replaced, unless mortar mixtures are judiciously
proportioned. DOI: 10.1061/(ASCE)MT.1943-5533.0000218. © 2011 American Society of Civil Engineers.
CE Database subject headings: Paper mills; Sludge; Industrial wastes; Compressive strength; Cement; Recycling.
Author keywords: Paper mill sludge ash; Compressive strength model; Water to binder ratio; Modified Abrams relationship.

Introduction
Unsustainable trends in waste generation are causes for concern
because the generation of waste can be a symptom of environmen-
tally inefficient use of resources (OECD 2001). Transforming such
otherwise usable products may mitigate but not solve present and
future environmental challenges (Naik and Kraus 2000). Organic
sludge may play a desirable role as an auxiliary fuel (IPPC
2001; CANMET 2005). Paper sludge is generally considered a
waste stream originating from water-treatment rejects that are sep-
arated from the mill effluent during the treatment process of (recov-
ered) virgin or recycled paper. However, in fact, such materials are
useful (Naik et al. 2004). The objectives of both the European
Union (EU) and Italian national policy are to minimize production
of waste and to promote reuse, recycling, and recovery so as to
reduce its negative environmental impact. The European paper
industry generates about 11 million t of so-called waste
(actually reusable products) annually, 70% of which originates
from recycled paper production. Such reusable products (RP)
are very diverse in composition and consist of rejects, different
types of sludges and in the case of on-site incineration, ashes
(Monte et al. 2009). This huge amount of RPs produces a severe
impact on the environment, and has led European governments to
1 the ash remains to be recycled or disposed. The same process also
Professor, Dept. of Materials and Environment Engineering and Physic
(FIMET), Università Politecnica delle Marche, Via Brecce Bianche, 60131,
Ancona, Italy (corresponding author). E-mail: g.fava@univpm.it
2
Researcher, Dept. of Materials and Environment Engineering and
Physic (FIMET), Università Politecnica delle Marche, Via Brecce Bianche,
60131, Ancona, Italy.
3
Researcher, Dept. of Materials and Environment Engineering and
Physic (FIMET), Università Politecnica delle Marche, Via Brecce Bianche,
60131, Ancona, Italy.
Note. This manuscript was submitted on February 19, 2010; approved
on October 23, 2010; published online on October 29, 2010. Discussion
period open until November 1, 2011; separate discussions must be sub-
mitted for individual papers. This paper is part of the Journal of Materials
in Civil Engineering, Vol. 23, No. 6, June 1, 2011. ©ASCE, ISSN 0899-
1561/2011/6-772–776/$25.00.
encourage and promote any processes that would enable its reduc-
tion, reuse, and/or possible realization of its useful economic-value.
In 2004, the production of paper mill sludge in Italy was around
600,000 t (Asquini et al. 2008). Approximately one-third of the
discarded paper volume processed in recycling plants becomes
sludge and rejects (Göttsching et al. 1996). The disposal of residual
paper sludge is a challenge confronting the paper industry; this
challenge is compounded by the increasing demand toward the re-
cycling of discarded, used papers. Almost paradoxically, increased
recycling has caused a significant increase of the amount of
recycled-paper sludge being generated. Additionally, the interest
for recovering energy from paper sludge is growing, as landfilling
is becoming increasingly expensive and less viable in many coun-
tries. Therefore, paper mill sludge is often incinerated to reduce the
volume for disposal and to recover heat. For these reasons, the use
of paper sludge for energy production has encountered increased
attraction as being a cost-effective and environmentally acceptable
disposal method (Anthony et al. 1993). Paper mill sludge is com-
posed of mineral fillers, inorganic salts, small cellulose fibers,
water, and organic compounds. In Italy, data on total waste gener-
ated are scarcely available because most of the pulp and paper mills
often have processes to internally treat the wastes to reduce the
solid waste generated. This, for instance, applies to bark residues
from debarking, which is incinerated in a boiler. As a result, only
applies to sludge incineration. Some figures on generated waste
found in the environmental reports from different European pulp
and paper mills indicate that approximately 115 to 138 kg of total
waste is produced per ton of paper produced. The main benefit of
the incineration is the reduction of the amount of material to be
landfilled by about 90%; i.e., about 10 to 15 kg of ash is produced
per ton of paper produced (Monte et al. 2009). The volume/weight
reduction process is achieved by (1) dewatering (i.e. by mechanical
means and/or evaporation) at low temperature (< 200°C), and
(2) incineration at high temperature (> 800°C). During incinera-
tion, organic compounds are burned at temperatures of around
350 to 500°C, whereas mineral fillers and inorganic salts are trans-
formed into the corresponding oxides at higher temperatures

772 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / JUNE 2011

J. Mater. Civ. Eng., 2011, 23(6): 772-776

 (> 800°C). CaO, Al2 O3 , MgO, and SiO2 are the most abundant
oxides in PA (Liaw et al. 1998). The resulting PA is classified
as a waste rather than the useful product that it is (Naik and Kraus
2000). Therefore, at the present time, it is mainly sent to landfill at
high financial and environmental cost. Recycling PA would have
beneficial effects for paper producers as well as for the environ-
ment. A possible reuse of the paper mill sludge is blending it with
natural raw materials extracted from the ores in the production of
cement, mortar, concrete, or bricks (Naik et al. 2004; Marcis et al.
2005; Ernstbrunner 1997; Liaw et al. 1998), because the main con-
Downloaded from ascelibrary.org by UFPR - Universidade Federal do Parana on 08/21/18. Copyright ASCE. For personal use only; all rights reserved.
stituent elements of paper mill sludge are Al, Mg, Si, and Ca,
whose oxides are largely used in the concrete industries. Therefore,
burning paper sludge represents two sources of cost saving for a
paper mill: reduced disposal costs and reduced fuel consumption
for steam generation/energy production.
The option to utilize paper sludge as a cofiring fuel for energy
production has been considered in many European countries, as
well as in the United States (Naik and Kraus 2000). The organic
fraction in paper sludge is renewable and, therefore, it is not con-
sidered as a contribution to the net carbon dioxide (CO2 ) emissions.
Fig. 1. Results of the thermogravimetric (TG) and differential thermal
analysis (DTA) of PA
Fig. 2. X-ray diffraction of PA
JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / JUNE 2011 / 773
J. Mater. Civ. Eng., 2011, 23(6): 772-776
The present study was conducted in the framework of a research
project aiming at improving the utilization potential of PA and at
evaluating the possibility of its use as a cement replacement in
mortar and concrete manufacturing.
Experimental Investigation
Materials
A commercial portland-limestone (20% maximum limestone
content) blended cement, Type CEM II/A-L 42.5 R, according
to EN-197/1 (CEN 2000a) was used. The Blaine fineness of the
cement was 410 m2 =kg, and its specific gravity was 3:05 kg=dm3 .
The PA used came from the burning of paper mill sludge from a
paper recycling mill, and testing shows the following main proper-
ties: Blaine fineness 635 m2 =kg, and specific gravity 1:72 kg=dm3.
A mixture of PA in water, as an extraction solvent 1∶5 (v/v), fol-
lowing the Italian procedure of IRSA CNR (Quad. Ist. Ric. Acque,
64, 1985), gave a pH of 12.45. To characterize the PA, thermal
analysis and X-ray diffraction were carried out. Thermal analysis
showed that the ash material contained about 6% calcite, CaCO3 .
As can be observed in Fig. 1, a sharp decrease in weight, corre-
sponding to the flex of the differential thermal analysis (DTA)
curve, occurs from about 600°C to 800°C, while a strong heat
absorption is detected. This is attributable to the decomposition
reaction of calcite, which is endothermic. X-ray diffraction analysis
(Fig. 2) also show the presence of calcium oxide (CaO), gehlenite
(Ca2 Al2 SiO7 ), and mayenite (Ca12 Al14 O33 ). In addition, other
components such as talc (3MgO
4SiO2 H2 O), barium sulfate
(BaSO4 ), and zinc oxide (ZnO) probably are also present in small
and variable percentages. PA was observed at increasing magnifi-
cations by using a scanning electron microscope (SEM) (Fig. 3).
The dominant morphology of the PA consisted of minor fragments
of clay/shale residue and agglomerated loose particles. Overall, the
particles ranged from 1 to 200 μm in size and their shapes varied
widely. Such results have been confirmed by others (Naik and
Kraus 2000). On the basis of these observations, it was concluded
that the PA recovering system was particularly selective for non-
spherical morphologies.
Natural sand (range of particle size: 0 to 5 mm) was used in
mortar mixtures. Its main physical properties, determined
according to EN 1097-6 (CEN 2000b) and EN 933-1 (CEN
Downloaded from ascelibrary.org by UFPR - Universidade Federal do Parana on 08/21/18. Copyright ASCE. For personal use only; all rights reserved.

Fig. 3. SEM observations of PA at different magnifications: (a) 500 μm; (b) 100 μm; (c) 50 μm; (d) 10 μm

Table 1. Physical Properties of Sand and PA Fraction Table 2. Mixture Proportions of Mortars Prepared with Different
Percentages of PA
Fraction Water Specific gravity Passing 75 μm
absorption (%) in SSDa condition sieve (%) Ingredient dosage (g)
(kg=m3 ) Mixture Water/binder Water Cement PA
Natural sand 4.1 2,530 0.7
0.4–0% PA 0.400 180 450.0 0.0
PA 25 1,720 80
0.4–5% PA 0.405 180 427.5 22.5
a
SSD = saturated surface-dried. 0.4–10% PA 0.411 180 405 45.0
0.4–15% PA 0.418 180 382.5 67.5
0.4–20% PA 0.425 180 360.0 90.0
1997), are shown in Table 1. A typical water-reducing admixture, 0.5–0% PA 0.500 225 450.0 0.0
based on polycarboxylate polymers (30% aqueous solution), was 0.5–5% PA 0.511 225 427.5 22.5
used for preparing all mortar mixtures. 0.5–10% PA 0.522 225 405.0 45.0
0.5–15% PA 0.535 225 382.5 67.5
Specimens 0.5–20% PA 0.550 225 360.0 90.0
Mortar specimens used for mechanical properties measurements 0.6–0% PA 0.600 270 450.0 0.0
were prepared with different water-to-cementitious-materials ratios
0.6–5% PA 0.616 270 427.5 22.5
(W=Cm). In the control specimen, the W=Cm was 0.4, 0.5, and 0.6.
0.6–10% PA 0.633 270 405.0 45.0
PA was added to the mortars mixtures at different percentages: 0, 5,
0.6–15% PA 0.653 270 382.5 67.5
10, 15, and 20% by weight for replacement of the cement. In all
0.6–20% PA 0.675 270 360.0 90,0
these specimens, the W=Cm was kept nearly constant and approx-
imately equal to that of the control mixtures, within experimental
error. The mixture proportions are summarized in Table 2. Each
mixture was prepared in a mixer with 2 L capacity. The dry ma- mortar workability was similar for all the mortar mixtures and equal
terials were mixed for 2 min. Then, the water containing the plas- to 180  10 mm slump/spread (measured according to EN 12350-
ticizer was added (1.0% by weight of the cement plus PA), and the 2CEN 1999). To determine the mechanical properties of the mortar
mixing was continued for an additional 2 min. The specimens mixture, 15 mortar prismatic specimens (40 × 40 × 160 mm)
for strength measurements were cast in 40 × 40 × 160-mm steel from each mixture were prepared. All the specimens with the
molds. All specimens were vibrated for 20 s on a vibrating table W=Cm ratio of nominal values of 0.4 and 0.6 were wet-cured at
and then covered with a plastic sheet to minimize water evapora- the temperature of about 20°C, while the others, with a W=Cm ratio
tion. The dosage of superplasticizer was maintained equal to 1.0% of 0.5, were cured at a higher temperature of about 27°C, which is
by weight of the cement plus PA to reduce the water dosage. The typical of summer temperatures. Curing temperature was included

774 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / JUNE 2011

J. Mater. Civ. Eng., 2011, 23(6): 772-776

 as an additional numerical input, along with the age of the speci-
men, W=Cm ratio, and mortar mixture composition to obtain a
more qualitative understanding of the effects of PA on mortar.

Results and Discussions

The compressive strengths of the mortars were measured at the

ages of 1, 7, 28, and 60 days after casting. The data fitted well
the Carino-Knudsen relationship based on the Maturity Method
(Carino 1984; Naik 1985), giving the opportunity to evaluate
Downloaded from ascelibrary.org by UFPR - Universidade Federal do Parana on 08/21/18. Copyright ASCE. For personal use only; all rights reserved.

the maximum compressive strength (f max ) and the related param-

eters through the expression
kðt  t 0 Þ
f c ¼ f max
1 þ kðt  t 0 Þ

where f c = compressive strength predicted at a given time t; t0 =

time needed before the strength gain begins (t 0 ¼ 0 was used);
f max = final strength when t tends to infinity; and k = days, with
1 as a constant. The results obtained and synthesized are shown in
Tables 3 and 4 and also in Fig. 4.
The data for the good performances were obtained with PA
replacing 5% of cement with all the W=Cms studied. For a proper
evaluation of the PA effect, it was also necessary to correct the
W=Cm for all mortars containing PA, taking into account the water Fig. 4. Strength developments of mortars with nominal water-to-
absorptivity of PA. The nominal W=Cm, when corrected, varied in cement ratio of 0.4, 0.5, and 0.6 with 5% and 15% PA substituted
the range from 0.40 to 0.67, for mortar mixtures nominally at for portland cement
W=Cm between 0.40 and 0.60. Cement replacement from 10%
up to 20% with PA seems to reduce the mechanical performance
of mortars. The maximum compressive strength (f max ) reduction
Table 3. Compressive Strength Values (MPa) Measured on Cement
Mortars Prepared with Different Percentages of PA ranged from 12 to 26%, with an increase of PA from 0 to 20%.
With higher ash content, the strength development seems to slow
Curing time (days) down, particularly for the higher W=Cm (0.60).
Mixture 1 7 28 60 Further insight was obtained to clarify the reasons for the
behavior detected for the mortars prepared with 5% PA replacing
0.4–0% PA 13.4 37.8 44.1 51.2
cement. Therefore, the Abrams (1918) equation, as modified by
0.4–5% PA 7.3 36.0 44.9 51.2 Popovics (2008) and Popovics and Ujhelyi (2008), was used to ac-
0.4–10% PA 2.6 30.4 37.4 50.0 count for a possible chemical or physical effect for PA.
0.4–15% PA 9.4 27.0 42.3 46.5
0.4–20% PA 11.8 35.7 42.4 45.0 k1 k1
f max ¼ w=bcem=100
or f max ¼ w=bash=100
0.5–0% PA 19.5 33.2 37.1 39.4 k2 k2
0.5–5% PA 15.2 33.3 36.5 39.9 The data (Fig. 5) fitted quite well, giving the following
0.5–10% PA 14.5 30.5 32.5 35.7 relationship:
0.5–15% PA 15.5 27.8 31.2 32.6
0.5–20% PA 12.2 23.0 29.1 29.9 f max ¼ 1=ð9:19 · 103: þ 6:60 · 102: w=CmÞ
0.6–0% PA 7.7 19.8 26.1 30.7
Pearson’s correlation coefficient R2 ¼ 0:92
0.6–5% PA 4.2 20.2 24.8 30.5
0.6–10% PA 3.2 18.1 24.9 28.3
0.6–15% PA 4.5 14.9 21.1 26.6 A better description can be obtained by taking into account a
0.6–20% PA 5.1 15.3 20.6 22.7 second-order interaction between the cement and the PA, as shown
in Fig. 6 and the following equation:

Table 4. Carino-Knudsen Parameters for Mortars Prepared with Different Percentages of PA

f max k f max k f max k
% PA w=b ¼ 0, 4 w=b ¼ 0, 4 w=b ¼ 0, 5 w=b ¼ 0, 5 w=b ¼ 0, 6 w=b ¼ 0, 6
0 51.49 0.36 38.89 0.95 29.65 0.30
5 53.92 0.24 40.08 0.63 31.45 0.22
10 51.81 0.22 36.11 0.69 30.50 0.19
15 50.98 0.17 32.74 0.88 28.19 0.15
20 47.01 0.39 30.29 0.59 23.84 0.26

JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / JUNE 2011 / 775

J. Mater. Civ. Eng., 2011, 23(6): 772-776


Downloaded from ascelibrary.org by UFPR - Universidade Federal do Parana on 08/21/18. Copyright ASCE. For personal use only; all rights reserved.
Fig. 5. Modified Abrams relationship
Fig. 6. Mortar compressive strength model
f max ¼ 96:4  101:9 × W=Cm þ 1:54 × 104 × cem × PA  8:43
×K
R ¼ 0:98
2 Liaw, C. T., Chang, H. L., Hsu, W. C., and Huang, C. R. (1998). “A novel
Conclusions
Based on the experimental results of this study, the following
observations can be drawn:
• At 28 days, the mortars containing 5% PA exhibited a compres-
sive strength higher than or as high as that of mortar made from
cement only; and
• The results presented encouraged the researchers to undertake
further study on the use of PA in concrete, which could lead to
a reduction in the cost of mortars and concrete, as well as a
method for minimizing cost for disposal of PA.
On the basis of the data collected within this experimental
project, it can be concluded that the PA, particularly if it replaces
up to 10% of cement, shows a positive effect on the development of
the mechanical performance of mortars. Judicious use of PA and
proper mixture proportioning can give a contribution to the cement
paste hardening, leading to an increase in the cement activity index.
On the other hand, because of its high fineness and, consequently,
high water absorption, the dosage of PA should not be too high (the
upper limit currently appears to be 10% of the weight of cement).
776 / JOURNAL OF MATERIALS IN CIVIL ENGINEERING © ASCE / JUNE 2011
J. Mater. Civ. Eng., 2011, 23(6): 772-776
Further research is continuing in the benefits of PA in cement-
based construction materials and especially for the use of PA as
a cement replacement material by greater than 10% of the weight
of cement.
References
Abrams, D. A. (1918). “Design of concrete mixtures.” Structural Materials
Research Laboratory, Bull. 1, Lewis Institute, Chicago.
Anthony, E., Preto, F., Herb, B. E., and Lewnard, J. J. (1993). “The tech-
nical, environmental, and economic feasibility of recovering energy
from paper mill residual fibers.” Fluidized Bed Combustion, ASME, 1,
239–247.
Asquini, L., Furlani, E., Bruckner, S., and Maschio, S. (2008). “Production
and characterization of sintered ceramics from paper mill sludge and
glass cullet.” Chemosphere, 71, 83–89.
CANMET. (2005). “Pulp and paper sludge to energy—Preliminary assess-
ment of technologies.” Energy Technology Centre, Ottawa.
Carino, N. J. (1984). “The maturity method: Theory and application.” Cem.
Concr, Aggregates, 6(2), 61–67.
European Committee for Standardization (CEN). (1997). “Tests for
geometrical properties of aggregates—Determination of particle size
distribution—Sieving method.” EN 933-1, Brussels.
European Committee for Standardization (CEN). (1999). “Testing fresh
concrete—Slump test.” , EN 12350-2, Brussels.
European Committee for Standardization (CEN). (2000a). “Cement—Part
1: Composition, specifications, and conformity criteria for common
cements.” EN 197-1, Brussels.
European Committee for Standardization (CEN). (2000b). “Tests for
mechanical and physical properties of aggregates—Determination of
particle density and water absorption.” EN 1097-6, Brussels.
Ernstbrunner, L. (1997). “Rejects from paper manufacture utilized in the
cement works.” Das Papier, 51(6), 284–286.
Göttsching, L., Hamm, U., Platzer, E., and Putz, H. J. (1996). “Analysis of
waste paper recycling and disposal options in Germany.” IIED Rep.
Integrated Pollution Prevention and Control (IPPC). (2001). “Reference
document on best available techniques in the pulp and paper industry.”
European Commission Joint Research Centre, Seville, Spain.
method to reuse paper sludge and cogeneration ashes from paper mill.”
J. Hazard. Mater., 58, 93–103.
Marcis, C., Minichelli, D., Bruckner, S., Bachiorrini, A., and Maschio, S.
(2005). “Production of monolithic ceramics from incinerated municipal
sewage sludge, paper mill sludge and steelworks slag.” Ind. Ceram.,
25(2), 89–95.
Monte, M. C., Fuente, E., Blanco, A., and Negro, C. (2009). “Waste
management from pulp and paper production in the European Union.”
Waste Manage., 29, 293–308.
Naik, T. R. (1985). “Maturity functions for concrete cured during winter
conditions.” ASTM, STP 852.
Naik, T. R., Friberg, T. S., and Chun, Y-C. (2004). “Use of pulp and paper
mill residual solids in production of cellucrete.” Cem. Concr. Res.,
34(7), 1229–1234.
Naik, T. R., and Kraus, R. N. (2000). “James fillcoat materials: A potential
source for valuable construction materials.” UWM center for by-
products utilization, Rep. No. CBU-2000-28, Rep. 402.
Organization for Economic Cooperation and Development (OECD).
(2001). “Environmental outlook for the chemical industry.” Paris.
Popovics, S. (1998). Strength and related properties of concrete, Wiley,
New York.
Popovics, S., and Ujhelyi, J. (2008). “Contribution to concrete strength
versus water-cement ratio relationship.” J. Mater. Civ. Eng., 20(7),
459–463.