See discussions, stats, and author profiles for this publication at: [Link]

net/publication/382456509

Investigating the Feasibility of Utilizing Electrical Arc Furnace Slag and

Circulating Dry Scrubber Dust as Binder for Cemented Paste Backfill

Conference Paper · July 2024

CITATIONS READS

0 156

5 authors, including:

Noureddine Ouffa Belem Tikou

Université du Québec en Abitibi-Témiscamingue Université du Québec en Abitibi-Témiscamingue
7 PUBLICATIONS 76 CITATIONS 124 PUBLICATIONS 5,478 CITATIONS

SEE PROFILE SEE PROFILE

Romain Trauchessec André Lecomte

University of Lorraine University of Lorraine
27 PUBLICATIONS 697 CITATIONS 80 PUBLICATIONS 1,534 CITATIONS

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Noureddine Ouffa on 22 July 2024.

The user has requested enhancement of the downloaded file.

Investigating the Feasibility of Utilizing Electrical Arc Furnace Slag and Circulating
Dry Scrubber Dust as Binder for Cemented Paste Backfill
Noureddine Ouffa1, Tikou Belem1, Romain Trauchessec2, André Lecomte2, and Mostafa Benzaazoua3
1
Research Institute of Mines and Environment, Université du Québec en Abitibi-Témiscamingue, Rouyn-Noranda, PQ, Canada,
[Link]@[Link]; [Link]@[Link]
2
Université de Lorraine, CNRS, Institut Jean Lamour, Nancy, France, [Link]@[Link]
3
Geology and Sustainable Mining Department, Mohammed VI Polytechnic University, Ben Guerir, Morocco,
[Link]@[Link]

Abstract
In the Abitibi-Témiscamingue mining region in Canada, underground mines utilize cemented paste
backfill (CPB) for secondary ground support. The conventional CPB formulation employs a binder known
as the reference binder (RB), consisting of 20% general use Portland cement (GU) and 80% ground
granulated blast furnace slag (GGBFS). This reference binder (RB = 20 GU/80 GGBFS) demonstrates
excellent mechanical and hydro-geotechnical properties for CPB. However, its high cost, limited
availability of GGBFS, and the environmental impact associated with GU production highlight the critical
need for research into alternative, cost-effective, and environmentally friendly [Link] study explores
the use of electrical arc furnace slag (EAFS) and circulating dry scrubber dust (CDSD) from the
steelmaking industry to partially replace GGBFS and act as a substitute for GU cement. The EAFS
undergoes a processing stage involving screening and grinding, resulting in a product termed ground
EAFS (GEAFS). Various formulations of the ternary blended binder (GGBFS/GEAFS/CDSD) are
evaluated for their unconfined compressive strength (UCS) at 7 and 28 days to assess their suitability as
binders for CPB. Our results demonstrate the feasibility of completely replacing type GU cement and
substituting up to 30% of GGBFS in the RB without compromising the UCS at 28 days (UCS28d). These
highly promising findings suggest the potential to lower the cost and carbon footprint of CPB while
promoting the recycling of metallurgical waste within the mining industry, aligning with a circular
economy approach.

Key words: cemented paste backfills, eco-friendly binders, ground granulated blast furnace slag, electric arc furnace
slag, circulating dry scrubber dusts, binder cost, carbon footprint reduction

Introduction
In the Quebec economy, the mining sector plays a crucial role by supplying essential metals, valuable
minerals, creating employment opportunities, and contributing important government revenue. In 2020,
the mining activities in Quebec made a noteworthy impact on the Gross Domestic Product (GDP),
amounting to approximately 12.9 billion C$, with 10.5 billion C$ specifically attributed to Quebec. In the
Abitibi-Témiscamingue region, these mining activities contributed about 22.8% of total employment and
approximately 35.8% of the regional GDP in the same year, as reported by Ecotec Consultants (2022).
However, a notable challenge arises from the substantial volume of mine tailings generated by the mining
industry. Traditionally, these tailings are managed in storage facilities, incurring economic drawbacks.
The associated costs include those related to development, construction, maintenance, monitoring, and
rehabilitation of these facilities (Bussière and Guittonny, 2020; Carneiro and Fourie, 2019). Furthermore,
this practice raises environmental concerns, encompassing issues related to land usage and water pollution
(Elghali et al., 2023). Additionally, there are geotechnical risks associated with the potential failure of
these tailings’ storage facilities (Bowker and Chambers, 2015).

The introduction of CPB technology represents an innovative alternative for handling mine tailings in the
mining sector. CPB is composed of filtered mine tailings, constituting 70–85% of the total CPB by
weight, plus hydraulic binders (2–10% dry tailings weight), and mixing water from sources such as lakes,
tap or processed water. This technology provides a comprehensive, secure, and sustainable approach to
mine tailings management (Belem and Benzaazoua, 2003; Benzaazoua et al., 2003). In certain instances,
additional components like crushed waste rocks (Hane et al., 2017) and additives (Ouattara et al., 2018)
are integrated to enhance particle size distribution, consistency, and setting time. These additives may
include superplasticizers and set-controlling admixtures, such as accelerators or retarders. The
incorporation of such elements contributes to optimizing the overall performance of the CPB technology
in tailings management. Commonly employed binders in mine tailings management include GU, either
alone or blended with supplementary cementitious materials (SCMs). These SCMs often consist of
GGBFS, a byproduct of cast iron, or various types of fly ashes (FA-F, FA-C) derived from thermal power
processes (Tariq and Yanful, 2013).

On average, the expenses related to binders constitute approximately 75% of the costs in backfilling
operations, where backfill costs in turn make up 10–20% of the total expenses in mining operations
(Gauthier, 2004; Grice, 1998). In the Abitibi-Témiscamingue region of Québec, Canada, RB is a prevalent
binder (Belem et al., 2010). While the RB consistently meets the mechanical strength and hydro-
geotechnical requirements for ground control (Benzaazoua et al., 2002; Godbout et al., 2007; Sahi, 2016;
Yilmaz et al., 2010; Yilmaz et al., 2011), challenges arise due to its high cost (Curry, 2020a; Curry,
2020b), limited availability of GGBFS (Scrivener et al., 2018), and environmental impacts associated with
GU production (Chen et al., 2010). Consequently, there is a compelling need for research aimed at
identifying alternative, cost-effective, and environmentally friendly binders in the field of mine tailings
management.

Ouffa et al. (2023) have showcased the effectiveness of utilizing circulating dry scrubber dust (CDSD) as
an activator for GGBFS. CDSD, an alkaline waste with a pH > 12, is a finely textured by-product
consisting of calcium sulfites (CaSO3.x(H2O)), calcium hydroxide or portlandite (Ca(OH)2), and gypsum
(CaSO4.2H2O). This by-product originates from the steel production processes employed at the RioTinto
iron and titanium (RTIT) steel works in Sorel-Tracy city, Québec, Canada. CDSD has demonstrated its
viability as a substitute for sodium hydroxide (NaOH) and GU in activating GGBFS. Notably, CDSD can
function as a complete replacement for GU in the reference binder while still maintaining a relatively
comparable unconfined compressive strength at 28 days (UCS28d) when compared to the reference binder.
Specifically, the UCS28d for the blended binder with 20% CDSD and 80% GGBFS reaches 2212 ± 134
kPa. In comparison, the UCS28d for the RB stands at 2016 ± 205 kPa. Notably, the unconfined compressive
strength at 7 days (UCS7d) for the 20% CDSD/80% GGBFS blend measures at 856 ± 26 kPa, a value
considered satisfactory for mining backfill applications, albeit lower than the UCS 7d of the reference
binder, which achieves 1430 ± 55 kPa. In addressing this, Ouffa et al. (2023) suggest the addition of 5%
clinker to the mixture for enhanced UCS7d in mining applications. Moreover, Ouffa et al. (2023) have
recommended considering fly ashes type F and fine glass powders as partial substitutes for GGBFS. This
study aims to propose another alternative approach by introducing the utilization of EAFS, an industrial
by-product readily available in Quebec pending appropriate treatment. EAFS is produced in Quebec by
RioTinto Iron and Titanium in Sorel-Tracy and by ArcelorMittal Long Product Canada in Contrecœur
(Quebec, Canada).

EAFS is a byproduct resulting from the steel production process, specifically generated during the refining
of steel in an electric arc furnace. This furnace employs high-power electric arcs, facilitated by graphite
electrodes, to melt recycled steel scrap and transform it into high-quality steel. Lime is introduced into the
scrap materials to facilitate the melting process, and oxygen is injected into the molten metal to oxidize
undesirable elements. These oxidized elements then combine with the lime to form slag. As the melting
process progresses, liquid steel accumulates at the bottom of the furnace. Once the desired chemical
composition of the steel is achieved, the slag and steel are extracted from the furnace. The molten slag is
then directed to a specialized slag treatment facility using containers or supports (Shi, 2005; Yildirim and
Prezzi, 2011).

The average density of EAFS is 3.35, ranging between 2.8–3.9 (Geiseler, 1996; Teo et al., 2020; Thomas
et al., 2019). Its chemical composition can vary, and is primarily composed of oxides such as CaO (2–
60%), FeO (2–52%), SiO2 (6–34%), Al2O3 (2–14%), and MgO (3–15%). Additionally, EAFS contains
other oxidized impurities like MnO (1–5%), TiO2 (0–1%), SO3 (0.1–2%), and P2O5 (0.5–2%), along with
free lime (f-CaO) and free magnesia (f-MgO) (Thomas et al., 2019; Yildirim and Prezzi, 2011). The
primary mineral phases identified in EAFS include belite, larnite or dicalcium silicate (β and γ-Ca 2SiO4),
alite (Ca3SiO5), merwinite (Ca3Mg(SiO4)2), bredigite (Ca7Mg(SiO4)4), gehlenite (Ca2Al2SiO7), and wustite
(solid solution of FeO) (Brand and Roesler, 2014; Geiseler, 1996; Shi, 2005; Yildirim and Prezzi, 2011).

Since the 1880s, EAFS has been processed and utilized as phosphate fertilizer (Geiseler, 1996), as well as
acting as supplementary cementitious materials, aggregates in concrete, contributing to road construction,
and participating in cement clinker production (Jiang et al., 2018; Motz and Geiseler, 2001). According to
Shi (2005), the cementitious properties of EAFS are associated with the presence of compounds like C 3S
(3CaO.SiO2), C2S (2CaO.SiO2), C4AF ([Link]), and C2F ([Link]). Additionally, a properly
cooled, high-basicity steel slag, as per the same source, can exhibit cementitious properties, albeit
generally weaker due to the low C3S compared to clinker. The limited cementitious properties can be
attributed to the high iron oxide content, the highly crystalline nature of EAFS, and the low amounts of
SiO2 and Al2O (Roslan et al., 2016). As the result, EAFS can only partially substitute Portland cement or
blast furnace slag in systems involving GU/GGBFS, GU/FA, or GU/silica fume (SF) (Amin et al., 2015;
Cristelo et al., 2023; Shi, 2005). The reported maximum substitution rate stands at 30% (Amin et al.,
2015; Muhmood et al., 2009; Ghadimi and Naghipour, 2023; Shi, 2005). Nevertheless, the optimal
percentage seems to be 10%, maintaining strengths comparable to those of Portland cement (Amin et al.,
2015; Hekal et al., 2013; Roslan et al., 2016; Roslan et al., 2020). Cements containing EAFS demonstrate
extended setting times, lower heat release during hydration, enhanced strength development at advanced
curing times (> 28 days), and improved sulfate resistance compared to Portland cement (Cristelo et al.,
2023; Roslan et al., 2016; Shi, 2005). Their strengths correlate with increased alkalinity (Muhmood et al.,
2009; Shi, 2005), prompting the exploration of various treatments to boost the alkalinity of such slag.

Various strategies have been proposed in scientific literature to enhance the reactivity of EAFS, including
rapid pressurized water quenching, chemical composition modification, grinding, CO 2 sequestration, and
alkaline activation. Rapid pressurized water quenching has shown potential in improving EAFS reactivity
(Muhmood et al., 2009), although it does not significantly alter the mineralogical structure compared to
other slag types such as ladle furnace slags (Tossavainen et al., 2007). Kim et al. (2015) suggested a
similar approach, involving chemical composition modification to adjust the CaO/Al 2O3 ratio and reduce
the FeO content to 2–5%, thereby enhancing the reactivity of EAFS. Lu et al. (2019) have proposed a
method of modifying the chemical composition of EAFS by incorporating various industrial wastes and
by-products to increase the CaO/Al2O3 basicity index. This modification has been shown to result in the
production of modified slag with favorable cementitious properties. In addition, alkaline activation has
been identified to enhance EAFS reactivity when combined with other materials like GGBFS or FAF.
Other suggested treatments include carbon sequestration (Mahoutian et al., 2015; Mo et al., 2017) and
grinding (Sun et al., 2022). According to Roslan et al. (2020), the fine grains of EAFS play a role in
bridging the gap between Ca(OH)2, ettringite, and the hard phase.

In the concrete field, EAFS offers interesting characteristics, but these have not been detailed for CPB.
Taking into account economic limitations and the diverse treatments outlined in existing literature, this
study suggests a partial substitution of GGBFS with sieved and ground EAFS in a CDSD/GGBFS system
used as binder for CPB. The proposed ranges drawn from literature review and prior laboratory
experiments are: 50% ≤ GGBFS ≤ 70%, 15% ≤ CDSD ≤ 30%, and 5% ≤ sieved and ground EAFS ≤ 20%.

Materials and Methods

The materials employed in this study encompass Sil-Co-Sil 106 ® procured from US SILICA, GU supplied
by McInnis Cement, GGBFS provided by Lafarge Canada Inc., CDSD obtained from Harsco
Environmental in Sorel-Tracy (Quebec, Canada) and EAFS supplied by ArcelorMittal Long Product
Canada in Contrecœur (Québec, Canada). The EAFS sample under investigation was collected from two
furnaces (50% AMOASC + 50% AMEASC). Notably, CDSD is a byproduct resulting from the air
desulfurization process at the RioTinto Iron and Titanium metallurgical complex in Sorel-Tracy (Québec,
Canada). CDSD and EAFS were acquired through the collaborative effort of the Technology Transfer
Center in Industrial Ecology and Université of Québec at Temiscaming.

Sil-Co-Sil 106®, employed as solid skeleton of cemented paste backfill (as a substitute for actual tailings),
is a high-purity fine sand consisting of 99.8% silicon oxide ([Link]). It possesses a relative
density (Gs) of 2.65 and a BET specific surface area of 880 m2/kg. Peyronnard and Benzaazoua (2011)
compared the particle size distribution of Sil-Co-Sil 106® with that of a tailings sample from LaRonde (in
Abitibi, Quebec, Canada), collected in 2011. This analysis shows that the particle size distribution of Sil-
Co-Sil 106® closely mirrors that of tailings from the LaRonde mine

The processing of EAFS involved screening and grinding (Figure 1). Initially, the EAFS underwent
primary crushing using a Géliko laboratory jaw crusher in a closed circuit with screening, employing a
SWECO vibrating sieve with a diameter ϕ = 2 mm. A segment of the challenging-to-crush slag (ϕ ˃ 2
mm) was separated and termed as REAFS. Following the screening, the crushed portion that passed
through the sieve underwent a secondary crushing process using a Marcy® Gy-Roll Lab Cone Crusher and
a grinding process in a laboratory steel rod mill with a 9 L capacity. The grinding load constituted
approximately 25% of grinder volume and consisted of three types of bars: 1/2 in, 3/2 in and 5/2 in.
 Figure 1. Process of treating and preparing EAFS for utilization as a supplementary cementitious
material.

The simulated cemented paste backfill (SCPB) mixture and uniaxial compression tests were conducted on
cylindrical specimens with a diameter (d) of 30 mm and a height (h) of 60 mm, resulting in an aspect ratio
(h/d) of 2, following the methodology outlined by Ouffa et al. (2023). Notably, the solids mass
concentration (Cw) of SCPB was maintained at 75% and the binder ratio (Bw) was set at 7%, equivalent to
a binder content (Cc) of 6.54%. The definitions are as follows: Cw = Msolid/MSCPB, Bw = Mbinder/MSil-Co-Sil106,
and Cc = Mbinder/(MSil-Co-Sil106+Mbinder) = Bw /(1+Bw ). Additionally, statistical analysis of the results using R
and Minitab and the calculation of environmental footprints and costs followed the methodology detailed
by Ouffa et al. (2023).

We conducted comprehensive characterizations on the investigated cementitious materials encompassing

physical, chemical, and mineralogical aspects. Physical characterization involved relative density and
specific surface area measurements. Relative density determined using the Micrometrics AccuPyc 1330
helium pycnometer, while specific surface area was assessed via the BET method using the Micrometrics
GEMINI surface analyzer, following the approach established by Brunauer, Emmett, and Teller. Chemical
characterization comprised sulfur and carbon content measurement using an ELTRA CS-2000 induction
furnace with a detection limit of 0.009%, free lime analysis according to the ASTM C 114 standard, and
examinations via X-ray fluorescence (XRF) and by inductively coupled plasma mass spectrometry with
strong acid digestion. Major elements were determined by XRF using fusion beads method, conducted on
a Bruker S8 TIGER spectrometer with fused beads prepared using an automatic fusion machine
(Autofluxer®). Minor element analysis was carried out by ICP-MS at SGS Minerals Services Canada in
Lakefield, Ontario. Measurements of sulfur and carbon levels, free lime, and ICP-MS were exclusively
performed on the GEAFS sample. The alkalinity of GEAFS and the investigated cementitious materials
was calculated using Equation 1:

Equation 1

Table 1 provides the outcomes of both physical and chemical characterization conducted by XRF for the
four cementitious materials. In Tables 2 and 3, data elaborates on the ICP-MS analysis and the assessment
of sulfur/carbon and free lime content specific to the GEAFS.

Table 1. XRF analysis and physical properties of the cementitious materials.

Oxide GU CDSD GGBFS GEAFS REAFS
SiO2 19.96 2.08 36.84 16.64 18.92
Al2O3 4.43 0.64 10.04 6.77 6.24
CaO 64.77 52.60 39.98 29.36 31.25
Fe2O3 3.44 0.95 0.69 32.40 27.82
K2O 0.62 0.05 - 0.10 0.10
MgO 1.78 1.00 11.54 12.01 11.60
MnO 0.06 - 0.22 2.14 3.24
P2O5 - - 0.13 0.86 0.21
Cr2O3 - - - 0.68 -
SO3 1.95 32.98 0.56 0.22 0.15
SrO 0.05 0.03 - - 0.02
TiO2 0.32 0.45 1.25 1.26 1.42
V2O5 - - - 0.16 -
ZrO2 - - 0.03 - -
*
LOI
2.63 9.14 -1.22 -2.69 -3.00
(950°C)
Gs 3.08 2.38 2.88 3.81 -
¶
BET SSA
1,462.00 13,441.00 2,227.00 1,984.00 -
(m2/kg)
Mb3 - - 1.10 1.77 1.70
*
LOI = loss on ignition; ¶BET SSA = BET method specific surface area

In this investigation, GU and CDSD are classified as activators, while GGBFS and GEAFS fall under the
category of activated materials. GU is primarily composed of CaO and SiO 2, making up 85% of its
composition, with Al2O3 and Fe2O3 accounting for 8%. In comparison, CDSD is primarily composed of
CaO and SO3, exhibiting a relative density of 2.38 and significant BET specific surface area. As for
GGBFS, it predominantly composed of CaO and SiO2 in nearly equal proportions (CaO/SiO2 ≈ 1.1),
making up 77% of its composition. Additionally, GGBFS contains nearly equivalent proportions of Al 2O3
and MgO (Al2O3/MgO ≈ 0.9), constituting 22%.
Both GEAFS and REAFS showcase nearly identical chemical compositions. GEAFS comprises CaO and
Fe2O3, with a CaO/Fe2O3 ratio of approximately 1 making up 62% of its composition. It also includes
SiO2, MgO, and Al2O3, collectively constituting 35%. The primary difference in the XRF chemical
compositions of GEAFS and REAFS lies in GEAFS having around -5% Fe2O3 relative to REAFS. This
indicates a higher iron concentration in the finer fraction, The BET specific surface area of GEAFS
exceeds that of GU and closely approaches that of GGBFS. Additionally, GEAFS exhibits an alkalinity
index, Mb3 ˃ 1, indicating its potential pozzolanic properties. The free lime (f-CaO) content is 0.12%
lower than the specified limit for clinker, which typically ranges from 0.5–1.5% f-CaO (Alemayehu and
Sahu, 2013).

Table 2. ICP-MS analysis of GEAFS

Element Units Value Element Units Value
Total Boron µg/g 66.00 Manganese % 1.40
Molybdenu
Silica % 7.09 µg/g 10.00
m
Aluminum % 3.20 Sodium µg/g 850.00
Arsenic µg/g 6.10 Nickel µg/g 46.00
Barium µg/g 220.00 Lead µg/g 14.00
Beryllium µg/g 1.80 Sulfur % 0.31
Bismuth µg/g 0.14 Antimony µg/g 0.90
Calcium % 20.00 Selenium µg/g < 0.70
Cadmium µg/g 0.36 Strontium µg/g 210.00
Cobalt µg/g 5.70 Tellurium µg/g 0.10
Chromium % 0.20 Titanium µg/g 1.00
Copper µg/g 63.00 Thallium µg/g < 0.02
Iron % 14.00 Uranium µg/g 6.10
Potassium µg/g 280.00 Vanadium µg/g 360.00
Lithium µg/g 5.10 Yttrium µg/g 31.00
Magnesium % 6.20 Zinc µg/g 430.00

Table 3. Sulfur/Carbone and free lime of GEAFS

Ctotal Stotal f-CaO
wt.% wt.% wt.%
DLM* 0.05 0.009
GEAFS 0.07 0.044 0.12
*
DLM = detection limit of the method

The mineralogical analysis was carried out using XRD with a Bruker A.X.S. Advance D8 apparatus
equipped with CuKα1 radiation (λ = 1.54056 Å). Operational parameters were set at 40 kV and 30 mA, with
an increment of 0.54°/min in the 5–70° (2θ) range. Diffractograms were indexed using Diffract Eva 6.0
software along with the PDF 2023 database. Figures 2 and 3 present the indexed diffractograms of the five
cementitious materials under investigation.
 Figure 2. X-ray diffractograms of the GU, CDSD and GGBFS. Abbreviations for minerals are
Ak: akermanite; Ca: calcite; Merw: merwinite; Per: periclase; Geh: gehlinite; Wus: wustite; Mag:
magnetite; Han: hannebachite; P: portlandite; Hiel: hielscherite; A: alite, Be: belite; Gyp: gypse; Br:
brownmerillite.

GGBFS primarily exists in an amorphous state, with minor phases including akermanite (Ca 2MgSi2O7),
calcite, and merwinite Ca3Mg(SiO4)2. CDSD is comprised of calcium sulphites (hanebachite CaSO3.H2O),
calcite (CaCO3), calcium hydroxide or portlandite (Ca(OH)2), hielscherite (Ca3Si(SO4)(SO3)
(OH)6.11H2O), and gypsum (CaSO4.2H2O). GU is composed of alite (3CaO.SiO2), belite (2CaO.SiO2),
brownmillerite (Ca2(Al,Fe)2O5), and gypsum (CaSO4.2H2O).

Figure 3. X-ray diffractograms of the GEAFS and REAFS.

In Figure 3, the diffractograms of the two slags (GEAFS and REAFS) exhibit nearly identical peaks,
indicating the presence of the same phases. The primary distinction lies in the intensity of these peaks,
with GEAFS showing less intense peaks compared to REAFS. This variation could be interpreted as a
higher concentration of the amorphous phase in GEAFS. Both GEAFS and REAFS predominantly contain
wustite (FeO), magnetite (Fe2+Fe3+2O4), bredegite (Ca7Mg(SiO4)4), gehlenite (Ca2Al(AlSi)O7), merwenite
(Ca3Mg(SiO4)2), alite (Ca3SiO5), and larnite (Ca2SiO4). These phases align with those commonly reported
in the literature for EAFS (Teo et al., 2020; Yildirim and Prezzi, 2011).

Results and Discussion

Table 4 displays the results of the UCS7d and UCS28d from the SCPB specimens incorporating diverse
formulations of ternary binders, consisting of GGBFS, GEAFS, and CDSD.

Table 4. Uniaxial compression test results (UCS) for ternary blends GGBFS-GEAFS-CDSD, with solids
mass concentration (Cw) of 75%, binder ratio (Bw) of 7%, binder content (Cc) of 6.54%.
N° Sample Proportions UCS7d (kPa) UCS28d (kPa)
Standard Standard
GGBFS GEAFS CDSD Mean Mean
deviation deviation
RB* 0.80 0.00 0.2 GU 1430 55 2016 205
M0S1 0.80 0.00 0.20 856 26 2212 134
M14S1 0.65 0.20 0.15 337 7 2169 81
M14S2 0.65 0.05 0.30 923 23 2295 100
M14S3 0.50 0.20 0.30 736 49 1816 117
M14S4 0.70 0.15 0.15 365 22 2212 83
M14S5 0.70 0.05 0.25 779 17 2316 107
M14S6 0.68 0.05 0.28 786 166 2240 49
M14S7 0.68 0.18 0.15 400 15 2326 37
M14S8 0.70 0.10 0.20 623 26 2493 104
M14S9 0.58 0.20 0.23 621 16 2059 71
M14S10 0.58 0.13 0.30 933 98 2077 67
M14S11 0.64 0.13 0.23 652 4 2153 101
M14S12 0.65 0.17 0.19 487 13 2159 68
M14S13 0.65 0.09 0.27 746 23 2083 72
M14S14 0.57 0.17 0.27 692 25 1947 38
M14S15 0.67 0.14 0.19 492 11 2300 80
M14S16 0.67 0.09 0.24 723 14 2264 76
*
RB = reference binder = 0.8GGBFS/0.2GU

The results reveal UCS28d values that are similar, or even slightly superior, to those of the RB across all
formulations. The UCS28d ranges from 1816 ± 117 kPa (for M14S3 mixture: 0.5 GGBFS/0.2 GEAFS/0.3
CDSD) to 2493 ± 104 kPa (for M14S8 mixture: 0.7 GGBFS/0.1 GEAFS/0.2 CDSD). Notably, these two
formulations enable the complete substitution of GU and a partial replacement of 30 and 10% in the
GGBFS content in the RB, respectively. However, the UCS7d of SCPB specimens based on ternary
mixtures GGBFS/GEAFS/CDSD is lower compared to that of the RB, which exhibits a UCS 7d of 1430 ±
30 kPa. The UCS7d of the ternary mixtures ranges from 337 ± 7 kPa (for M14S1 mixture: 0.65 GGBFS/0.2
GEAFS/0.15 CDSD) to 933 ± 98 kPa (for M14S10 mixture: 0.575 GGBFS/0.125 GEAFS/0.3 CDSD). It
is noteworthy that most, if not all, of the UCS7d values presented are deemed acceptable in the context of
mine backfills.

Within mining sites, the mining operation cycle typically spans approximately one month, representing
the duration between the backfilling of primary open stopes and the excavation of secondary stopes.
Consequently, UCS28d holds significant importance in this context. In the subsequent part of the study, our
aim is to comprehend the impact of each component (ie, GGBSF, GEAFS, and CDSD) on UCS 28d.
The analysis of correlations between the various terms of the full cubic model and UCS 28d is presented in
Table 5. Results indicate a robust positive correlation between GGBFS and UCS 28d. Additionally, it is
noteworthy that positive correlations exclusively involve GGBFS, while all negative correlations involve
GEAFS. Hence, one could infer that the incorporation of GEAFS might have a negative impact on the
UCS28d of GGBFS/CDSD binary mixtures (see M0S1 in Table 4).

Table 5. Correlation analysis of UCS28d for ternary mixtures GGBFS/GEAFS/CDSD.

Positive correlations Negative correlations
Term R Term R
GGBFS 0.90 GEAFS -0.52
GGBFS×CDSD×(GGBFS–CDSD) 0.78 GEAFS×CDSD -0.84
GGBFS× GEAFS×CDSD -0.78

To validate this hypothesis, UCS28d was underwent modeling using a full cubic model, as detailed in the
materials and methods section, and represented by Equation (2) (with a coefficient of determination R 2 =
0.89 and R2Adjusted = 0.79), where:

X1 = GGBFS; X2 = GEAFS, X3 = CDSD Equation 2

Figure 4. Cox trace plot and ternary diagram illustrating UCS28d (kPa) for the GGBFS/GEAFS/CDSD
binder mixtures.
Use of Equation 2 permits the construction of ternary diagram and Cox trace plot (Figure 4). The visual
representation clearly illustrates that the incorporation of GEAFS results in a decrease in 28 day strength,
following an almost linear trend. However, this reduction is limited to around 200 kPa at most, which is
not highly significant when considering the economic advantages of substitution. Additionally, literature
sources emphasize that EAFS exhibits reactions over the long term (beyond 28 days and more) as detailed
in the literature review (see introduction section). Therefore, it becomes crucial to assess these slags over
the long term to ascertain their utility.

Conclusions
The objective of this paper is to showcase the feasibility of utilizing GEAFS as a partial substitute for
GGBFS in a CDSD/GGBFS blended binder. CDSD has previously proven successful as an alternative to
GU in activating GGBFS, presenting a cost-effective and environmentally friendly option in comparison
to the traditional RB. Moreover, both EAFS and CDSD are locally available industrial waste byproducts
in Québec, specifically in Sorel-Tracy and Contrecœur. Based on our research, the following conclusions
can be drawn:

 achieving UCS28d values comparable to those of the RB without using GU and with reduced
costs is a promising outcome from the unconfined compression tests presented in this article
 it is viable to completely substitute GU and decrease the GGBFS content in the RB by up to
30% without significantly impacting UCS28d
 evaluating the long-term reactivity of treated GEAFS in the GGBFS/CDSD system is crucial
to determine its suitability as partial replacement for GGBFS on an industrial scale

Finaly, the incorporation of GEAFS as a substitute for GGBFS has the potential to lower the costs and
decrease the environmental impact of mining with CPB, especially in Québec where GGBFS is imported
while EAFS is locally available.

Acknowledgments
The authors would like to express their gratitude to Mitacs Accelerate (#IT26385) and the Natural Science
and Engineering Research Council of Canada (NSERC) for their generous financial support. Special
appreciation is extended to Pascal Lemoine and Youssef Benarchid from the Centre de Transfert
Technologique en Écologie Industrielle (CTTÉI) for their invaluable assistance in obtaining CDSD and
EAFS from RioTinto Iron and Titanium, ArcelorMittal Long Products Canada, and Harsco
Environmental. A sincere thanks also go to Lafarge Inc. for providing GGBFS and McInnis Cement for
supplying GU Portland cement.

References
Alemayehu, F., & Sahu, O. (2013). Minimization of variation in clinker quality. Advances in Materials, 2(2), 23-28.
Amin, M. S., El-Gamal, S. M. A., Abo-El-Enein, S. A., El-Hosiny, F. I., & Ramadan, M. (2015). Physico-chemical
characteristics of blended cement pastes containing electric arc furnace slag with and without silica fume.
HBRC Journal, 11(3), 321-327. doi:[Link]
Belem T., Benzaazoua, M. (2003) Utilisation du remblai en pâte comme support de terrain. Partie I : de sa
fabrication à sa mise en place sous terre. Après-mines 2003, "Impacts et gestion des risques: besoins et
acquis de la recherché." 5–7 February 2003, Nancy. GISOS, CD-ROM, 12p.
Belem, T., Peyronnard, O., Benzaazoua, M. (2010). A model of formulation of blended binders for use in cemented
mine backfills. Proceedings of 1st International Seminar on Reduction of Risk in the Management of
Tailings and Mine Waste – Mine Waste’10, Perth, WA, Australia, 29 sept – 1st oct. 2010, pp. 433-448.
Benzaazoua, M., Belem, T., & Bussière, B. (2002). Chemical factors that influence the performance of mine
sulphidic paste backfill. Cement and Concrete Research, 32(7), 1133-1144.
doi:[Link]
Benzaazoua M., Belem T., Ouellet S., Fall M. (2003) Utilisation du remblai en pâte comme support de terrain. Partie
II : comportement à court, à moyen et à long terme. Après-mines 2003, "Impacts et gestion des risques :
besoins et acquis de la recherché." 5–7 February 2003, Nancy. GISOS, CD-ROM, 12p.
Bouzoubaâ, N., & Fournier, B. (2005). Current situation with the production and use of supplementary cementitious
materials (SCMs) in concrete construction in Canada. Canadian Journal of Civil Engineering, 32(1), 129-
143.
Bowker, L. N., & Chambers, D. M. (2015). The risk, public liability, & economics of tailings storage facility
failures. Earthwork Act, 24, 1-56.
Brand, A. S., & Roesler, J. R. (2014). Concrete with steel furnace slag and fractionated reclaimed asphalt pavement.
Retrieved from: Bussière, B., & Guittonny, M. (2020). Hard rock mine reclamation: from prediction to
management of acid mine drainage: CRC press.
Carneiro, A., & Fourie, A. B. (2019). An integrated approach to cost comparisons of different tailings management
options. Paper presented at the Paste 2019: Proceedings of the 22nd International Conference on Paste,
Thickened and Filtered Tailings, Cape Town. [Link]
Chen, C., Habert, G., Bouzidi, Y., & Jullien, A. (2010). Environmental impact of cement production: detail of the
different processes and cement plant variability evaluation. Journal of cleaner production, 18(5), 478-485.
doi:[Link]
Cristelo, N., Coelho, J., Rivera, J., Garcia-Lodeiro, I., Miranda, T., & Fernández-Jiménez, A. (2023). Application of
electric arc furnace slag as an alternative precursor to blast furnace slag in alkaline cements. Journal of
Sustainable Cement-Based Materials, 12(9), 1081-1093. doi:10.1080/21650373.2022.2161660
Curry, K. C. (2020a). Cement Statistics and Information. Retrieved from [Link]
statistics-and-information
Curry, K. C. (2020b). Iron and Steel Slag Statistics and Information. Retrieved from
[Link]
science_support_page_related_con=0#qt-science_support_page_related_con
EcoTec Consultants. (2022). Retombées économiques de l’industrie minière au Québec en 2020. Retrieved from
[Link]
Elghali, A., Benzaazoua, M., Taha, Y., Amar, H., Ait-khouia, Y., Bouzahzah, H., & Hakkou, R. (2023). Prediction of
acid mine drainage: Where we are. Earth-Science Reviews, 241, 104421.
doi:[Link]
Gauthier, P. (2004). Valorisation des liants et des rejets industriels dans les remblais miniers. Rapport DESS,
Université du Québec en Abitbi-Témiscamingue, Rouyn-Noranda, Canada.
Geiseler, J. (1996). Use of steelworks slag in Europe. Waste management, 16(1), 59-63.
doi:[Link]
Godbout, J., Bussière, B., Aubertin, M., Belem, T. (2007). Evolution of cemented paste backfill saturated hydraulic
conductivity at early curing time. Proceedings of 60th Canadian Geotechnical Conference and the 8th Joint
CGS/IAH‐CNC Groundwater Conference, Ottawa, 21‐24 October 2007, p. 2230‐2236.
Grice, T. (1998). Underground mining with backfill. 2nd Annual Summit œ Mine Tailings Disposal Systems,
Brisbane, Nov, 24-25.
Hane, I., Belem, T., Benzaazoua, M., & Maqsoud, A. (2017). Laboratory characterization of cemented tailings paste
containing crushed waste rocks for improved compressive strength development. Geotechnical and
Geological Engineering, 35, 645-662.
Hekal, E. E., Abo-El-Enein, S. A., El-Korashy, S. A., Megahed, G. M., & El-Sayed, T. M. (2013). Hydration
characteristics of Portland cement – Electric arc furnace slag blends. HBRC Journal, 9(2), 118-124.
doi:10.1016/[Link].2013.05.006
Jiang, Y., Ling, T.-C., Shi, C., & Pan, S.-Y. (2018). Characteristics of steel slags and their use in cement and
concrete—A review. Resources, Conservation and Recycling, 136, 187-197.
Kim, H.-S., Kim, K.-S., Jung, S. S., Hwang, J. I., Choi, J.-S., & Sohn, I. (2015). Valorization of electric arc furnace
primary steelmaking slags for cement applications. Waste management, 41, 85-93.
doi:[Link]
Lu, X., Dai, W., Liu, X., Cang, D., & Zhou, L. (2019). Effect of basicity on cementitious activity of modified electric
arc furnace steel slag. Metallurgical Research & Technology, 116(2), 217.
Mahoutian, M., Shao, Y., Mucci, A., & Fournier, B. (2015). Carbonation and hydration behavior of EAF and BOF
steel slag binders. Materials and structures, 48(9), 3075-3085. doi:10.1617/s11527-014-0380-x
Mo, L., Zhang, F., Deng, M., Jin, F., Al-Tabbaa, A., & Wang, A. (2017). Accelerated carbonation and performance
of concrete made with steel slag as binding materials and aggregates. Cement and Concrete Composites, 83,
138-145. doi:[Link]
Motz, H., & Geiseler, J. (2001). Products of steel slags an opportunity to save natural resources. Waste management,
21(3), 285-293. doi:[Link]
Muhmood, L., Vitta, S., & Venkateswaran, D. (2009). Cementitious and pozzolanic behavior of electric arc furnace
steel slags. Cement and Concrete Research, 39(2), 102-109.
doi:[Link]
Ghadimi, M. j., and Naghipour, A. (2023). Investigating the Physical and Chemical Effects of Adding Electric Arc
Furnace Slag in Different Percentages to Portland Cement. doi:[Link]
Ouattara, D., Belem, T., Mbonimpa, M., & Yahia, A. (2018). Effect of superplasticizers on the consistency and
unconfined compressive strength of cemented paste backfills. Construction and Building Materials, 181, 59-
72.
Ouffa, N., Benzaazoua, M., Belem, T., Trauchessec, R., and Lecomte, A. (2023). An alternative to NaOH in the
alkali-activation of ground granulated blast furnace slag in the formulation of cemented paste backfills.
Paper presented at the Paste 2023: 25th International Conference on Paste, Thickened and Filtered Tailings,
Banff. [Link]
Ouffa, N., Trauchessec, R., Benzaazoua, M., Lecomte, A., and Belem, T. (2022). A methodological approach applied
to elaborate alkali-activated binders for mine paste backfills. Cement and Concrete Composites, 127,
104381. doi:[Link]
Roslan, N. H., Ismail, M., Abdul-Majid, Z., Ghoreishiamiri, S., and Muhammad, B. (2016). Performance of steel
slag and steel sludge in concrete. Construction and Building Materials, 104, 16-24.
doi:[Link]
Roslan, N. H., Ismail, M., Khalid, N. H. A., and Muhammad, B. (2020). Properties of concrete containing electric
arc furnace steel slag and steel sludge. Journal of Building Engineering, 28, 101060.
doi:[Link]
Sahi, A. (2016). Validation expérimentale d'un modèle de sélection optimale des liants dans la fabrication des
remblais miniers cimentés. Mémoire de Maîtrise, Génie minéral, École Polytechnique de Montréal-UQAT,
196 p.
Scrivener, K., Martirena, F., Bishnoi, S., and Maity, S. (2018). Calcined clay limestone cements (LC3). Cement and
Concrete Research, 114, 49-56.
Shi, C. (2005). Steel Slag — Its Production, Processing, Characteristics, and Cementitious Properties. Cheminform,
36. doi:10.1002/chin.200522249
Sun, X., Liu, J., Zhao, Y., Zhao, J., Li, Z., Sun, Y., Qiu, J., Zheng, P. (2022). Mechanical activation of steel slag to
prepare supplementary cementitious materials: A comparative research based on the particle size
 distribution, hydration, toxicity assessment and carbon dioxide emission. Journal of Building Engineering,
60, 105200. doi:[Link]
Tariq, A., & Yanful, E. K. (2013). A review of binders used in cemented paste tailings for underground and surface
disposal practices. Journal of environmental management, 131, 138-149.
Teo, P. T., Zakaria, S. K., Salleh, S. Z., Taib, M. A. A., Mohd Sharif, N., Abu Seman, A., Mohamed, J. J., Yusoff,
M., Yusoff, A. H., Mohamad, M., Masri, M. N., [Link], S. (2020). Assessment of Electric Arc Furnace
(EAF) Steel Slag Waste’s Recycling Options into Value Added Green Products: A Review. Metals, 10(10),
1347. Retrieved from [Link]
Thomas, C., Rosales, J., Polanco, J. A., and Agrela, F. (2019). 7 - Steel slags. In J. de Brito and F. Agrela (Eds.),
New Trends in Eco-efficient and Recycled Concrete (pp. 169-190): Woodhead Publishing.
Tossavainen, M., Engstrom, F., Yang, Q., Menad, N., Lidstrom Larsson, M., & Bjorkman, B. (2007). Characteristics
of steel slag under different cooling conditions. Waste management, 27(10), 1335-1344.
doi:[Link]
Yildirim, I. Z., and Prezzi, M. (2011). Chemical, mineralogical, and morphological properties of steel slag. Advances
in Civil Engineering, 2011.
Yilmaz, E., Belem, T., Benzaazoua, M., and Bussière, B. (2010). Assessment of the modified CUAPS apparatus to
estimate in situ properties of cemented paste backfill. Geotechnical Testing Journal, 33(5), 351-362.
Yilmaz, E., Belem, T., Bussière, B., and Benzaazoua, M. (2011). Relationships between microstructural properties
and compressive strength of consolidated and unconsolidated cemented paste backfills. Cement and
Concrete Composites, 33(6), 702-715.

View publication stats