Journal of Materials Research and Technology 32 (2024) 362–377

Contents lists available at ScienceDirect

Journal of Materials Research and Technology

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

Eco-efficient ultra-high-performance concrete formulation utilizing electric

arc furnace slag and recycled glass powder–advanced analytics and
lifecycle perspectives
M. Iqbal Khan a, ** , Yassir M. Abbas a , Joaquin Abellan-Garcia b, * , Andrea Castro-Cabeza c
a
Department of Civil Engineering, College of Engineering, King Saud University, P.O. Box 800, Riyadh, 11421, Saudi Arabia
b
Department of Civil and Environmental Engineering, Universidad Del Norte, Barranquilla, Colombia
c
Helmut Schmidt University / University of the Federal Armed Forces Hamburg, Hamburg, Germany

A R T I C L E I N F O A B S T R A C T

Handling editor: M Meyers Ultra-high-performance concrete (UHPC) is a promising material for sustainable and resource-efficient con-
struction, but its high Portland Cement content poses environmental challenges. This research proposes an
Keywords: optimized UHPC blend that significantly reduces the Portland Cement dosage by incorporating industrial waste
Compressive strength materials [i.e., Recycled Glass Powder (RGP) and Electric Arc Furnace Slag (EAFS)]. The integration of these
Flowability
waste materials in UHPC formulation addresses critical waste management issues and aligns with sustainable
Multi-criteria optimization
construction practices. Utilizing a tri-variable Design of Experiments methodology in conjunction with the
Response surface methodology (RSM)
Ultrahigh performance concrete (UHPC) Andreasen & Andersen theory, the research aims to reduce Portland Cement in UHPC without compromising its
compressive strength and rheological behavior. The optimized blend incorporates RGP and EAFS, achieving the
threshold compressive strength established by ACI-239 (150 MPa) while maintaining a self-compacting perfor-
mance in the fresh state. This mixture features low cement (621 kg/m3) and silica fume (100 kg/m3) content. Life
cycle assessments (LCA) were conducted to examine the environmental efficiency of the optimized blend. A
comparative LCA demonstrates that the optimized UHPC mixture developed in this study surpasses a control
mixture with higher cement and silica fume contents while maintaining comparable mechanical and rheological
properties. Notably, this improvement is highlighted by a nearly 20% reduction in CO2 equivalent emissions. The
scientific significance of this study lies in its demonstration of a viable pathway to producing UHPC with reduced
environmental impact, contributing to the advancement of sustainable building practices, and offering a practical
solution to waste management issues.

1. Introduction polycarboxylate-ether superplasticizers (HRWRA) and occasionally,

steel fibers [1,5–7]. Given its superior mechanical resilience and
Renowned for its versatility in civil engineering, concrete is foun- enduring nature, UHPC has increasingly become a favored choice in
dational to constructing buildings, bridges, industrial facilities, and modern engineering projects.
dams. Continual advancements have led to enhanced properties of this UHPC typically comprises a substantial amount of cement, finely
material. A groundbreaking development was introduced by the French powdered materials, quartz micro sand, and steel fibers [8]. Neverthe-
firm BOUYGUES, which pioneered ultra-high-performance concrete less, recent investigations have demonstrated that UHPC has the po-
(UHPC). This innovation boasts exceptional compressive strength, tential to incorporate coarse aggregates, including basalt or calcined
attributed to its densely packed structure [1–4]. Further research by bauxite, as a means to reduce production costs and enhance mechanical
Richard and Cheyrezy [1] in the 1990s unveiled that UHPC integrates performance [9–12]. Consequently, the classification of UHPC as con-
more than 800 kg/m3 of Portland cement (PC), significant proportions of crete hinges on the presence of a cementitious matrix and aggregate,
silica fume (SF), quartz powder (QP), and silica sand (SS, sized up to 600 irrespective of the aggregate’s size or type. The key distinction between
μm), along with advanced water-reducing agents such as UHPC and conventional concrete lies in UHPC’s ultra-high strength,

* Corresponding author.
** Corresponding author.
E-mail addresses: [email protected] (M.I. Khan), [email protected] (J. Abellan-Garcia).

https://doi.org/10.1016/j.jmrt.2024.07.171
Received 26 June 2024; Received in revised form 21 July 2024; Accepted 23 July 2024
Available online 25 July 2024
2238-7854/© 2024 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-
nc-nd/4.0/).
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

typically exceeding 120 MPa, and its densified paste matrix, endowing it (GGBFS), which has been extensively studied and used as a binder in the
with increased resistance to various forms of deterioration [12]. In concrete industry, EAFS has seen limited use due to its less favorable
contrast to traditional concrete, which relies on a balance of coarse and chemical composition for hydraulic and pozzolanic properties, among
fine aggregates for strength and workability, the unique formulation of others [51]. Consequently, research on the viability of EAFS as a UHPC
UHPC enables it to attain superior mechanical properties without binder remains scarce. Utilizing this hazardous waste as construction
necessitating the use of traditional coarse aggregates [13]. materials greatly aids in environmental preservation and promotes
While UHPC offers superior properties, its high PC, quartz sand, and sustainable development within the building sector [52,53].
SF content render it costlier than traditional concrete, restricting its For its part, RGP has been extensively employed as supplementary
widespread adoption [14–16]. To counteract this cost barrier, various cementitious material due to its high silicon oxide content which is
supplementary materials and industrial by-products, including lime- bound to provide pozzolanic features [54,55]. However, special care
stone powder (LP), ground granulated blast-furnace slag (GGBFS), rice must be taken with this waste material when utilized as aggregate, as it
husk ash (RHA), fly ash (FA), metakaolin (MK), and recycled waste glass can yield a dangerous alkali-silica reaction due to its elevated concen-
powder (RGP), among others, have been explored as potential PC sub- tration of NaO2 [56]. Nevertheless, several pieces of research have
stitutes. Studies have delved into the impact of blast furnace slag (BFS) demonstrated that, when milled under a size of 300 micros, the pozzo-
on UHPC’s microstructure and mechanics, with certain mixes achieving lanic reaction prevails over the ASR, preventing this dangerous concrete
compressive strengths surpassing 150 MPa at PC concentrations be- pathology [57,58]. The recycled glass powder used in this research has a
tween 680 and 977 kg/m3 [17]. Furthermore, the research highlighted particle size well below this limit, as observed in Fig. 1.
the positive mechanical effects of incorporating 420 kg/m3 GGBFS and To the authors’ knowledge, no previous research has utilized both
120 kg/m3 SF in UHPC. In contrast, others demonstrated the benefits of EAFS and RGP in the production of sustainable and cost-effective UHPC.
adding LP, particularly in enhancing early hydration and particle This investigation aims to demonstrate that the chemical and granulo-
packing density [18–21]. Another intriguing approach involved the in- metric properties of these components allow for the replacement of
clusion of nano-CaCO3, which boosted compressive strength by 17% cement and silica fume (and fully replace QP) without disrupting the
compared to control samples [22–24]. The potential of replacing PC and packing density or the chemical balance between hydraulic and pozzo-
SF with by-products like FA and fluid catalytic cracking residue (FC3R) lanic reactions.
has also been explored. However, achieving the target 150 MPa To that end, this study introduces a factorial design methodology for
compressive strength demanded PC amounts over 740 kg/m3 [25–27]. formulating UHPC composites capable of achieving as high as 150 MPa
Numerous studies have reported the viability of RHA as an SF replace- compressive strength. The emphasis is on minimizing PC usage by
ment in UHPC, with some achieving compressive strengths beyond 150 incorporating RGP and EAFS as replacements for PC and SF. A multi-
MPa [28–30]. Additionally, investigations into substituting SF with fine objective optimization algorithm, coded in R [59], was utilized for the
glass powder (FGP) yielded impressive early compressive strengths, mathematical fine-tuning of the mix. Subsequently, the mathematically
albeit at elevated PC levels [31–33]. Innovative approaches, like derived blend was validated through empirical tests. The study also
employing RGP or integrating waste bottom ash (WBA), showcased aimed to evaluate the environmental consequences of producing con-
mixed outcomes concerning mechanical performance [5,34,35]. These ventional UHPC compared to an optimized UHPC formulation incor-
findings underscore the potential to augment UHPC with more porating EAFS and RGP. This way, the environmental effects of standard
cost-effective and sustainable components without compromising UHPC production against those of UHPC containing EASF and glass
strength. Yet, pinpointing the ideal balance between these alternative powder are compared by conducting their Life Cycle Assessment (LCA).
materials and UHPC’s mechanical attributes remains challenging. The conventional UHPC dosage, i.e., the Control Mix, is characterized by
Moreover, PC, a dominant hydraulic binder in contemporary con- a high concentration of PC and densified SF as cementitious compo-
crete, emerges from an energy-demanding sector (4 GJ/ton of PC) and is nents. Further specifics regarding the Control Mix UHPC are available in
a significant source of CO2 emissions, exacerbating global climate Ref. [60]. It should also be highlighted that EAFS is a byproduct of a
change. Creating a single ton of PC clinker discharges almost an Colombian steel manufacturer. In brief, this study aims to enhance the
equivalent amount of CO2 [36]. The annual global production of 1.5 cost-effectiveness and environmental sustainability of UHPC. Conse-
billion tons, predominantly of PC, accounts for approximately 7% of quently, the investigation focuses on three specific research objectives:
worldwide CO2 emissions [28,37–41]. Therefore, a paramount objective
in sustainable development is the formulation and production of con-
crete that uses reduced clinker, emits less CO2, maintains reliability, and
enhances longevity. UHPC is viewed as a potential solution to mitigate
the environmental impact of building materials because of its excep-
tional mechanical performances that facilitate slimmer constructions.
Yet, as previously indicated, UHPC’s binder content remains notably
high [42–44]. This underscores the significance of substituting a portion
of the PC in UHPC, especially with regionally sourced industrial re-
siduals, drawing substantial attention from researchers [3,27,45–47].
In pursuit of a cost-efficient and greener UHPC, this research aims to
meticulously formulate and fabricate UHPC with minimal PC content,
not exceeding an SF limit of 100 kg/m3. The mixture of these UHPC
blends is driven by the ambition to achieve superior compressive
strength and optimal rheological characteristics, while using the least
possible PC, guided by a tri-variable Design of Experiments (DoE)
approach. The enhanced Andreasen & Andersen particle distribution
model (A&Amod) [48] was employed to guarantee a tightly packed
cementitious matrix. Materials like electric arc furnace slag (EAFS) and
RGP were incorporated as substitutes for traditional PC and SF compo-
nents. EAFS is a by-product generated during the steelmaking process in
an electric arc furnace [49,50]. Historically, disposing of this by-product
incurred high costs. Unlike Ground Granulated Blast Furnace Slag Fig. 1. Grade distributions of the raw and blended materials.

363
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

(1) Formulating UHPC blends with minimal PC content and limited investigation.
SF, incorporating lower-cost and eco-friendlier substitutes (EAFS
and RGP) for achieving the threshold compressive strength and 2.2.1. Experimental methods
rheological characteristics while minimizing the material’s final This research determined the quantities of PC, water, and HRWR
price and carbon footprint. using the DoE methodology. Meanwhile, the A&Amod guided the
(2) A multi-objective optimization algorithm to fine-tune the UHPC incorporation of other components, employing a q value of 0.264 as per
mix derived from the experimental design was used. The vali- Eq. (1). In the equation, D represents the particle size, while Ptar (D) in-
dation through empirical tests will ensure that the formulated dicates the weight proportion of total granular UHPC-making compo-
blend achieves targeted features. nents that are smaller than D. Dqmax and Dqmin correspond to the upper and
(3) Conducting a Life Cycle Assessment (LCA) to evaluate the envi- lower limits of particle sizes, respectively. Here, the distribution
ronmental implications of conventional UHPC production versus modulus is denoted by q. This value determines the ratio between the
an optimized formulation. This evaluation aims to gauge the total aggregate and the cementitious paste, which is crucial for
sustainability and eco-friendliness of the proposed UHPC achieving optimal particle packing and potentially the highest
composition. compressive strength. A lower ratio indicates a higher concentration of
PC and SF, improving packing density and mechanical properties but
2. Materials and methodology also increasing the cost of the resulting UHPC. It is important to note that
the optimal ratio (0.264) was determined in previous research [66].
2.1. Materials ( q )
D − Dqmin
Ptar (D) = ( q ) (1)
This study utilized Colombian local resources to formulate UHPC Dmax − Dqmin
mixtures. The primary binder was ASM Type HE PC, characterized by a Upon finalizing the dosage, a linear mortar mixer was employed for
specific gravity of 3.15 and an average particle diameter (d50) of 8 μm. concrete production. Post-mixing, the fresh attributes of the UHPC
Additionally, SF incorporated adhered to ASTM C-1240 standards, dosages were evaluated. The static slump flow diameter was ascertained
boasting a specific gravity of 2.20 and a d50 of 0.15 μm. The mixture in line with the ASTM C1437 [64], without dropping, utilizing the
incorporated SS with specifications including a specific gravity of 2.65, a dispersion diameter from the mini-slump cone during the flow-table test.
maximum particle size (dmax) of 600 μm, and a d50 of 165 μm. The As presented in Fig. 5, this procedure involved pouring the fresh UHPC
research further employed RGP with properties of a specific gravity of into a mold with a truncated conic shape (until it is completely filled)
2.55 and a d50 of 28 μm, and EAFS with a specific gravity of 3.15 and and then removing the mold to release the fresh-state UHPC. The con-
d50 of 2.1 μm as alternative UHPC-making constituents. The EAFS and crete’s spread flow (Φm ) was then determined by averaging the diameter
RGP were refined from indigenous waste materials through jet milling, measurements taken in four orthogonal directions.
adjusting for variances in fineness by altering the grinding velocities. Subsequent to Φm assessment, the fresh-state UHPC was poured into
Detailed chemical compositions of these constituents are available in 50 mm-side cube formworks. To enhance the concrete’s packing density,
Table 1. Particle size distribution’s visual representation of granular a vibrating table was utilized post-pour. Roughly 24 h after being cast,
constituents can be found in Fig. 1, while Fig. 2 captures the raw ma- the specimens were demolded and then underwent curing in a humidity-
terials employed. To enhance workability, a 1.07-specific-gravity controlled room set at 20 ◦ C, where the relative humidity was main-
HRWRA with a solid content of 40% was integrated into the mixture. tained at 95% ± 5 up to the testing day. Compression tests were con-
Employing scanning electron microscopy (SEM), the alternative ducted on the concrete employing universal testing equipment with a
cementitious materials (i.e., EAFS and RGP), which served as substitutes capacity of 1000 kN, in alignment with ASTM C109 standard [65]. A
for PC and SF, were meticulously examined, as illustrated in Fig. 3. rate of 1000 N/s (i.e., 0.4 MPa/s) was applied during the performance of
Intriguingly, the analyses revealed an absence of porosity in the RGP, the tests in accordance with the mentioned standard. At each specified
indicating a low water absorption of the glass particles. This low ab- testing interval (1-day and 28-days), three specimens were assessed, and
sorption of glass particles, widely reported in the scientific literature the average outcome is presented in this report. Fig. 6 presents some
[61–63] will play a key role in the analysis of RGP effects on the prop- images of the formworks, specimens, and test setup for compressive
erties of UHPC. strength determination.

2.2. Methodology 2.2.2. Response surface methodology

The response surface methodology (RSM) is an ensemble of methods
Fig. 4 and Table 2 delineate the experimental framework, detailing designed for the identification of optimal operational conditions via
the methodological protocols implemented throughout this experimental approaches [67]. This methodology integrates mathe-
matical and statistical tools, primarily rooted in experimental design and
least squares regression analysis. Notably, it is employed in scenarios
Table 1
where multiple input variables significantly impact the outcome [68].
Physicochemical characteristics of the cementitious substances.
Fundamentally, RSM incorporates three elements: a strategic experi-
Cementitious material mental layout, a model of the response surface, and a subsequent opti-
OPC SF EAFS RGP SS mization process [69,70].
SiO2 % 19.42 92.29 21.70 72.89 95.80 Moreover, the Central Composite Design (CCD) stands as a robust
Al2O3 4.00 0.59 6.20 1.67 0.11 experimental framework, favored for its capacity to appraise second-
CaO 64.42 3.89 33.13 9.73 0.38 order impacts and scrutinize a response surface with minimal experi-
MgO 1.52 0.26 10.60 2.08 0.20
mental runs [71]. It is adept at discerning factor interconnections,
SO3 1.93 0.07 1.14 0.01 0.52
Na2O 0.19 0.31 0.15 12.54 0.25
modeling response datasets, and pinpointing the optimal outcome.
K2O 0.39 0.54 0.03 0.76 3.49 Within this methodology, the second-order model is prevalently
TiO2 0.38 0.01 0.42 0.04 0.25 employed due to several rationales: (i) its quadratic structure, being
Mn3O4 0.05 0.01 1.98 0.01 0.01 highly versatile, often serves as a fitting approximation to genuine
Fe2O3 3.61 0.24 18.92 0.81 0.09
response surfaces; (ii) estimating its parameters, denoted as ’ a’ in Eq.
Loss of ignition 2.58 0.60 4.67 1.00 0.31
Specific gravity gr/cm3 3.17 2.20 3.15 2.56 2.67 (3), is straightforward; and (iii) extensive empirical evidence suggests

364
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 2. The raw materials utilized in this research: (a) PC, (b) condensed SF, (c) EAFS, (d) RGP, and (e) SS.

Fig. 3. SEM images of: (a) PC, (b) condensed SF, (c) EAFS, and (d) RGP.

that second-order models are efficacious in addressing authentic In this equation, Xj represents the encoded value of the j th factor, while
response surface challenges.
Zj indicates its actual value. Z0j refers to the j th factor’s real value at the
CCD, often regarded as the quintessential response surface experi-
central point, and Zmaxj and Zminj correspond to its upper and lower
mental framework, predominantly comprises factorial or fractional
bounds within the domain, respectively. To gauge each outcome, a
factorial designs, supplemented with center and axial points. These
second-order approach employed, derived following Eq. (3).
augmentations expand the variable spectrum, enabling the derivation of
∑ ∑ ∑
second-order terms. The specific design deployed in our study consists of Y = a0 + ai xi + aii x2i + aij xi xj (3)
2N factorial points, situated at the cube’s corners, paired with 2 N axial
points distanced at ± α from the center. As illustrated in Fig. 7, this In this model, Y denotes the anticipated response (the dependent vari-
scheme encompasses a cumulative 18 set points, including 4 central able). The term a0 signifies the average response, with ai representing
ones, segmented into 2 blocks (cube and axial). In this research, the linear coefficients, aii as quadratic coefficients, and aij as interaction
value of α was set as ±1.789. To streamline the interpretation of the coefficients. Here, xi and xj are the independent variables (factors). The
findings, statistical evaluation was conducted on the encoded datasets. specified levels for these factors in this research are based on values
This encoding was adhered to Eq. (2). derived from previous work by the authors [72,73] and are depicted in
Zj − Z0j Table 5. The current study selected three primary factors: PC content
Xj = (2) (measured in kg/m3), the water-to-binder ratio (w/b), and the total
Zmaxj − Zminj
HRWRA volume ratio. The A&Amod curve guided the determination of

365
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 4. Flow chart illustrating the research methodology employed in this study.

metric (Φm ).
Table 2
A summary of the conducted tests and their corresponding standards.
2.2.3. Multi-objective optimization
Test Specimen Standard In this study, we utilized an advanced R-coded optimization algo-
Type Age Type Size Number rithm focused on multi-objective targets [59]. This scheme was tailored
(days) (mm) to ascertain the ideal parameters for the independent variables, thereby
Flow 0 – – 4 ASTM C1437a achieving the peak response value. Drawing inspiration from the
[64] dual-response optimization pioneered by Derringer and Suich [74], the
Compression 1, and 28 Cube 50 3b ASTM C109 [65] utilized approach transforms each response metric into a suitable
a
With no drops. approach. This, in turn, optimizes the aggregate suitability for each
b
Per point of the experimental design and age;c Per age. response using a singular objective optimization method. The initial
phase involves crafting a computational relationship between variables
the remaining concrete component ratios. Furthermore, we evaluated and their respective responses. Subsequent to this, every independent
four distinct outcomes: compressive strength under standardized curing variable was systematically and independently adjusted to fine-tune the
conditions at intervals of 1-day (C1), 28 days (C28), and the flowability desired outcomes. This refined solution is designed to meet specific
criteria for each response, maintaining a harmonious equilibrium among

Fig. 5. Stages of the Φm assessment.

366
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 6. (a) Formworks, (b) specimens, and (c) test setup for compressive strength determination.

⎧
⎪
⎪ 0 Yi ≤ L
⎪
⎪ [ ]wti
⎨
Yi − L
d= L < Yi < U (6)
⎪
⎪
⎪ U− L
⎪
⎩
1 Yi ≥ U
⎧
⎨ 0 Yi ≤ L
d= 1 L < Yi < U (7)
⎩
0 Yi ≥ U

where U signifies the upper threshold, while L represents the lower

boundary for a response. Whereas wti is the weight or functional shape
determinant. By manipulating these weights, one can modify each
objective individual desirability shape. The wti values range between 0.1
Fig. 7. The implemented CCD with three design variables, each at two levels. and 10. Weights exceeding 1 accentuate the significance of the objec-
tive, whereas values below 1 suggest diminished importance. A weight
the model’s outcomes [27]. As described in Ref. [74], Eq. (4) presents of precisely 1 infers a linear progression of di . For the study at hand, the
the geometric average of singular suitability. multi-objective optimization was executed in two sequential phases. In
[ ]1/∑ these steps, responses and parameters were earmarked for a distinct
( )1/∑ r n
∏ ri target by allocating a predetermined importance level. Table 3 computes
D = dr11 × dr22 × dr33 × … × drnn i = dri n (4)
i=1
the objectives and parameters of the optimization criteria for each
outcome. The guiding principle was the selection of optimal factor
Where di represents the desirability associated with each specific values to achieve a self-compacting mixture in alignment with EFNARC
response, while ri designates the weight or significance attributed to that criteria [75]. This also ensured minimal PC content while preserving a
response. The variable n quantifies the considered responses’ number. A C28 exceeding 150 MPa. In the context of this study, three critical re-
common solution when all individual responses have the same impor- sponses – C28, Φm , and PC dosage - were deemed of equal consequence.
∑
tance is to consider ri equals to 1/n, in order to force ri = 1[59]. The The refined compositions of both mixtures are presented in Table 4.
parameter di illustrates the individual response desirability metric,
spanning from 0 (representing an unfavorable outcome) to 1 (indicating 2.2.4. LCA methodology
the optimal outcome). Consequently, this individual desirability metric This study employed LCA based on the International Organization
translates the response value into a spectrum between 0 and 1. The for Standards (ISO 15804). The LCA establishes the calculation rules for
concluding value of D resides within this same range. Achieving the the analysis of construction products [76]. The subsequent are the
ideal target values results in D equating to one. It is important to essential stages used to analyze and compare the environmental impacts
mention that, inherent to the features of Eq. (4), the entirety of the of UHPC based on inventory elements: (i) Goal, scope, and System
function collapses to zero if any single response drifts outside its boundaries for LCA; (ii) Compilation of the products and process in-
designated desirability range. The distinct forms of individual desir- ventory of the UHPC concrete; (iii) The assessment of the environmental
ability function for various response scenarios - minimization, maximi- impacts under the inventory; and lastly (iv) Analysis the results and
zation, and range-specific cases - are delineated in Eqs. (5)–(7), putting forwards methods for improvement. To analyze the product
respectively. system, an LCA model is developed using the SimaPro program [77].
⎧
⎪
⎪ 1 Yi ≤ L
⎪ [
⎪ ]wt
⎨
U − Yi i
d= L < Yi < U (5)
⎪
⎪ U− L Table 3
⎪
⎪ Refinement of individual responses aims to achieve a self-compacting UHPC that
⎩
0 Yi ≥ U
boasts a high 28-day compressive strength and minimizes PC content.
Responses and variables Lower (L) Upper (U) Goal wti ri

Φm (mm) 240 260 In range – 1/3

C28(MPa) 150 165 Maximum 1 1/3
PC (kg/m3) 600 650 Minimum 1 1/3

367
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Table 4 2.2.4.3. Life cycle inventory. In the present study, the Life Cycle In-
Optimized composition. ventory (LCI) thoroughly analyzes the inputs and outputs over the
Mix A B C Desirability designated time frame. For the UHPC under examination, the stages
encompass raw material extraction, material production, transportation,
Optimum Coded − 0.149 − 0.149 1.7889 0.675
Real 621.27 0.164 0.0256 and the manufacturing of UHPC, specifically its mixing at a concrete
production facility. The data sources integrated into this investigation
were accurately selected to reflect temporal, geographical, and tech-
nological representativeness. Throughout this segment of the LCA, every
Table 5
Impact categories considered in the Life Cycle Assessment.
piece of data utilized for analysis was articulated to distinctly identify
both primary and secondary data for each process under study, along
Impact category Abbreviation Unit
with their respective references [78]. Primary data pertains specifically
Climate change GWP-Total kg CO2 eq to the production processes directly related to the product or service
Ozone depletion ODP kg CFC11 eq being assessed in the LCA [78,79]. On the other hand, secondary data
Photochemical ozone formation POCP kg NMVOC
eq
provide generalized or average statistics pertinent to the product or
Acidification AP mol H+ eq service in question [80,81].
Eutrophication EP kg N eq Key components such as high-early-strength cement, (HRWRA),
Resource use, fossils ADP – fossil fuels MJ Densified Silica Fume, Glass Powder, Electrical Arc Furnace Slag, and
Resource use, minerals, and metals ADP–minerals and kg Sb eq
Silica Sand were identified in the Ecoinvent database under ’Cement’
metals
Use of net fresh water WDP m3 (Ecoinvent 3). The Electrical Arc Furnace Slag data was sourced from the
Total Energy from non-renewable PENRE MJ Environmental Product Declaration for Slag [82]. The Glass Powder was
sources synthesized using a ball milling machine akin to those employed in
Total Energy from renewable sources PERE MJ clinker production. This powder is derived from larger fragments of
shattered glass. To enhance the pulverization process, specialized
2.2.4.1. Functional unit. In line with the DIN EN 15804 [76] standards, equipment that operates under negative pressure was used. This
which delineate the parameters for defining a product system, the equipment efficiently segregates the fine glass powder from coarser
designated functional unit is defined as one cubic meter of concrete. This fragments during grinding, further milling only the residual larger
serves as the baseline for evaluating its environmental implications. In particles while effectively separating the powder. Subsequently, barring
this context, environmental assessments are methodically computed water, all ingredients were directed to the concrete mixing plant. A truck
based on a per-volume functional unit, corresponding to one cubic meter with a capacity range of 16–32 metric tons, compliant with the Euro3
of concrete. standard, was used to transport all the components. The study simulated
a premixing Plant in SimaPro, boasting a maximum volume of 135,000
m3 and a production rate of 60 m3/h, equipped with a double-shaft
2.2.4.2. System description and boundaries. In the study presented, a
mixer. The considered assumptions in this regard align well with those
comprehensive "cradle-to-gate" approach was employed to delineate the
of previous works in the field [83–85].
system under investigation. The research specifically illustrates the
Fig. 9 depicts the virtual representation of the UHPC production
manufacturing process of UHPC at a ready-mix production facility
process as modeled in SimaPro.
located in Medellín, situated in the northwestern region of Colombia.
Fig. 8 clearly explains the system boundaries and associated trans-
portation distances. Moreover, the scope of the system under scrutiny 2.2.4.4. Life cycle impact assessment. In the present study, we evaluated
not only includes the raw materials and encompasses the entirety of the environmental implications of UHPC mixes that incorporate a high
subsequent processes, culminating in the delivery of the mixed concrete cement content and recycled components, such as RCA, for potential use
at the facility’s exit. in concrete production. This assessment was conducted using an attri-
butional Life Cycle Assessment (LCA) approach, spanning from raw

Fig. 8. System boundaries Illustration.

368
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 9. Virtual process for the preparation of concrete for all the mixes.

material extraction to the production phase. Utilizing the SimaPro systematically listed in Table 7. In the flowability test, the parameter Φm
software (version 9.4.0.1) in tandem with the Ecoinvent database, we
meticulously crafted processes for the constituents required to produce
one cubic meter of concrete and subsequently generated assemblies for Table 7
each mix variant. The environmental footprint for each UHPC mix sce- Composition of UHPC mixes (kg/m3).
nario was determined in accordance with DIN EN 15804 [61]. A PC SF RGP EAFS water HRWRA SS
particular attention was given to the primary indicators (1–7) associated 600 100 375.0 120.0 176.2 21.5 931.3
with resource consumption, as detailed in Table 5. However, 650 100 320.0 94.9 171.3 21.5 981.8
waste-related impact categories were not considered in this analysis. 600 100 375.0 120.0 188.1 21.5 899.6
650 100 320.0 94.9 183.0 21.5 950.9
600 100 375.0 120.0 173.2 25.8 928.7
3. Results and discussion 650 100 320.0 94.9 168.3 25.8 979.1
600 100 375.0 94.9 181.5 25.8 931.5
3.1. The developed model and its verification 650 100 320.0 94.9 188.6 25.8 948.3
625 100 337.2 102.5 175.6 23.6 961.8
625 100 337.2 102.5 175.6 23.6 961.8
This study employed an 18-point central composite design to model 625 100 337.2 102.5 175.6 23.6 961.8
the response across the domain space, factoring in three design variables 625 100 337.2 102.5 175.6 23.6 961.8
at three distinct levels. Table 6 summarizes the CCD’s design points and 580.3 100 375.0 120.0 176.6 23.6 942.8
their experimentally derived values. We utilized R [30], a robust 669.7 100 300.7 100.0 180.1 23.6 960.5
625 100 337.2 102.5 165.2 23.6 989.4
statistical-computational platform, to orchestrate the experimental 625 100 337.2 102.5 186.1 23.6 934.1
design and subsequent study. 625 100 337.2 102.5 178.3 19.8 964.1
The mixture ratios linked to this experimental design have been 625 100 337.2 102.5 172.9 27.5 959.4
refined based on the A&Amod curve. These tailored proportions are

Table 6
The combination of set points and their associated experimental outcomes.
Run Coordinates Actual dosages Responses

A B C PC w/b HRWRA C1 C28 Φm

kg/m3 – %, vol. MPa mm

1 1 1 − 1 600 0.16 0.020 59.5 136.6 259.9

2 − 1 1 − 1 650 0.16 0.020 52.2 137.7 262.5
3 − 1 1 1 600 0.17 0.020 24.7 145.4 270.1
4 1 − 1 − 1 650 0.17 0.020 47.0 151.5 205.6
5 − 1 − 1 1 600 0.16 0.024 47.3 150.9 241.1
6 1 1 1 650 0.16 0.024 12.0 159.9 266.9
7 0 0 0 600 0.17 0.024 51.7 146.1 244.2
8 1 − 1 1 650 0.17 0.024 45.8 153.4 222.7
9 0 0 0 625 0.165 0.022 47.8 148.6 240.4
10 − 1 − 1 − 1 625 0.165 0.022 29.6 140.5 219.8
11 0 − 1.789 0 625 0.165 0.022 46.8 150.3 203.1
12 0 0 0 625 0.165 0.022 44.6 151.6 235.8
13 − 1.789 0 0 580.3 0.165 0.022 47.8 139.3 246.0
14 0 0 0 669.7 0.165 0.022 44.6 151.2 239.9
15 0 0 1.789 625 0.156 0.022 19.0 160.5 251.2
16 0 1.789 0 625 0.174 0.022 40.6 145.5 278.6
17 1.789 0 0 625 0.165 0.0184 50.7 152.1 254.0
18 0 0 − 1.789 625 0.165 0.0256 52.5 139.3 231.6

369
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

(in mm) was determined as the average value of the four orthogonal Table 9
∑4 ANOVA for the established numerical models.
diameters, Φi (i.e., Φm = 1/4 i=1 Φi ). For each response (C1, C28, and
Φm ), a regression approach was formulated derived from Eq. (3). This Model Φm C1 C28
involved two primary phases: (i) constructing an appropriate model and Terms
Coeff P-value Coeff P-value Coeff P-value
(ii) subsequently validating its precision. In this research, the model
Inter. 240.616 <0.001 47.099 <0.001 148.753 <0.001
accuracy was achieved using a backward stepwise methodology. The A − 1.659 0.019 1.082 0.239 3.462 <0.001
comprehensive quadratic polynomial regression was initially estab- B 21.206 <0.001 − 2.253 0.028 − 1.747 0.005
lished to decipher the interlinks among experimental results from CCD C 6.119 <0.001 − 8.224 <0.001 5.645 <0.001
and their respective variables. Following this, variables with the highest A:B 3.344 0.050 − 2.669 0.046 – –
A:C − 4.863 0.002
P-value were incrementally eliminated, retaining only those with a – – – –
B:C − 2.969 0.049 − 11.436 <0.001 2.336 0.005
P-value less than 0.05. Nevertheless, a first-order variable was retained A:B:C – – – – 3.006 <0.001
if it was associated with a significant second-order term, even if it A2 2.945 0.021 – – − 1.179 0.025
exhibited no significance on its own. After model refinement, a residual B2 – – − 1.651 0.071 – –
analysis, encompassing evaluations like residual plots and standard C2 – – − 4.142 <0.001 – –

deviation, was conducted to ascertain the model’s effectiveness. Upon

validating the model’s precision, the response surface diagram was impact on Φm , C1, and C28.
generated [86].
The model’s validity was ascertained through an analysis of variance 3.3. Response of the developed models
(ANOVA) conducted in the R programming language environment [87],
with outcomes presented in Table 8. Both the F-test and P-value metrics 3.3.1. Flowability
assessed the model’s significance. Given that P-values for all regression Fig. 10 exhibits the impact of variables A, B, and C on the Φm . As
approaches were below 0.01%, this underscores their statistical rele- deduced from the ANOVA data in Table 8, there exists a linear associ-
vance. The evaluation of the regression models encompassed several key ation between the Φm response and the water-to-binder proportion, the
parameters, as outlined in Table 8: (i) the r-squared (R2 ), (ii) the latter being paramount. In this context, EFNARC [75] posits that a Φm
adjusted multiple determination coefficient (Adj-R2 ), (iii) root mean value in the range of 250 ± 10 mm is optimal for standard
squared error (RMSE), (iv) F-stat, as well as (v) P-values for lack-of-fit, self-consolidating concrete (SCC). Incorporating more PC into the con-
overall model significance, and the (vi) Shapiro-Wilk test for normality. crete typically augments the mixture’s viscosity, consequently dimin-
Remarkably, the developed models showcased a robust R2 , ac- ishing the spreading flow as PC content (variable A) rises. During initial
counting for 97%, 96%, 96%, and 95% of the variances in responses Φm , cement hydration, rheological parameters—yield stress and plastic vis-
C1, and C28, respectively. Such results underscore the model’s sub- cosity—rise slowly at first and then more rapidly as hydration progresses
stantial fit and its statistical robustness. A strong congruence between [88–90]. This change is, on the one hand, influenced by the formation of
experimental outcomes and model predictions was observed. Notably, calcium-silicate-hydrate (C–S–H), which transforms the suspension into
the proximity of the adj-R2 to the R2 suggests minimal inclusion of a mechanically interconnected network [91,92]. As per Nonat [93], in
extraneous variables. To ascertain the model’s accuracy, we conducted a pure alite systems, the first hydrates formed solidified contact points
lack-of-fit analysis, where an optimal result approaches zero. Based on between particles. At the same time, the initial stiffening of the cement
the ANOVA, all derived P-values reinforced the notion that the model’s paste is attributed to the bridging effect of C–S–H, leading to
lack-of-fit was not statistically significant when in comparison with pure flocculation.
error. Furthermore, the Shapiro-Wilk normality test yielded P-values On the other hand, although the nucleation period of C–S–H is
exceeding 0.05 across the board. generally considered significant for this early hydration stages [93], the
majority of researchers suggest that ettringite may be the most relevant
phase controlling the rheology of the paste at the beginning of cement
3.2. Evaluation of variables utilizing the response surface methodology hydration [94–97]. In this sense, several pieces of research have
demonstrated that early ettringite formation increases yield stress and
In assessing the constructed RSM models, an ANOVA was performed affects workability due to high water consumption and the formation of
to examine the influence of individual factors and their subsequent in- low-density ettringite [94,97]. Therefore, as cement content increases,
terlinks on the outcomes. The t-test was utilized to identify salient pa- more water is needed for hydration, leaving less free water for lubri-
rameters, pinpointing those with a P-value less than the critical value of cation between particles, thus reducing flowability.
0.05; such a value underscores their essential role in enhancing the In addition, the lower the cement content, the higher the RGP
model. Table 9 illustrates the estimated regression approach terms dosage, due to the packing density condition. In this sense, the incor-
alongside their associated P-values. Any parameter that shows a P-value poration of RGP with non-porosity particles (see Fig. 3), and thereby low
surpassing 0.05 was excluded from the model, save for specific noted water absorption [61–63] has as a consequence more free water to help
exceptions. Triangular graphical representations enable the visualiza- with the rheological properties in the fresh state.
tion of response surfaces, delineating the response as a function of three Predictably, both variables B (w/b) and C (HRWRA) amplify the
determinants. Fig. 10 through Fig. 12 display both contour and response flowability metric. While factor B is most impactful in this context,
surface mappings across the specified domain, considering factors such considerations regarding its effect on compressive strength and estab-
as PC content (A), w/b (B), and the volumetric content of high-range- lished upper limits necessitate its moderated application.
water-reducer-admixture (C). These visual depictions shed light on the
interplay of these three independent constituents and their cumulative

Table 8
Evaluation criteria for the established regression models.
Response R2 Adj-R2 RMSE F-stat Lack of fit P-value (%) Shapiro P-value

Φm 0.972 0.954 3.52 60.58 4.50 <0.01 0.966

C1 0.964 0.935 2.30 31.16 3.26 <0.01 0.629
C28 0.955 0.930 1.46 38.67 1.87 <0.01 0.472

370
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 10. Flowability contour plots (down), and 3D-RSM (up).

Fig. 11. Strength at 1d contour plots (down), and 3D-RSM (up).

3.3.2. First-day compressive strength (C1) 3.3.3. 28-Day compressive strength (C28)
In Fig. 11, both contour and 3-D visualizations depict the impacts of Fig. 12 offers a three-dimensional representation detailing the
the three distinct factors on the 1-day compressive strength. The peak impact of the considered parameters on the compressive strength over
value aligns with the highest PC level (A). Further, the ANOVA analysis 28 days. This visualization underscores an enhanced 28-day compres-
in Table 5 reveals a nonlinear correlation between the C1 response and sive strength in UHPC when both PC and superplasticizer are at an
the considered factors. Fig. 11 illustrates that while an augmentation in elevated level (1), provided the water-to-binder ratio remains subdued
PC (A) drives a boost in early compressive strength, a more pronounced (− 1). Furthermore, Fig. 10 elucidates an intriguing dynamic: as opposed
enhancement is observed with reductions in the w/b (B) and HRWRA to initial strength dynamics, a surge in superplasticizer content bolsters
dosage (C). The detrimental impact of increased water on compressive the rate of 28-day compressive strength. In alignment with previous
strength is a recognized phenomenon. Furthermore, multiple studies findings, a nonlinear correlation persists between this 28-day strength
[74,98] have validated the adverse implications of polycarboxylate on and the triad of investigated parameters.
initial strength progression. Specifically, polycarboxylate-based ether
superplasticizers decelerate the hydration process, particularly for sili-
3.4. Verification of the developed models
cates like the C3S phase, and influence the emergence of Aft (hexacal-
cium aluminate trisulfate hydrate) [99].
Typically, the efficacy of a model is gauged by assessing its residuals.
An ideal model will exhibit a residual plot devoid of discernible

371
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 12. Strength at 28d contour plots (down), and 3D-RSM(up).

structure. In the case of the RSM of the current study (Fig. 13), the re- preserving a compressive strength surpassing 150 MPa.
siduals presented in the Q-Q plot closely align with a linear trajectory, Finally, it is worth highlighting that the optimized dosage has similar
suggesting a standard distribution of errors. This alignment further properties to other sustainable UHPC formulations reported in recent
bolsters the significance of the terms incorporated within the model. In bibliographic references [100–103], but with a reduction in cement and
Fig. 14, we present a plot juxtaposing anticipated outcomes with their SF contents by up to 20 and 50% respectively. It should be noted that the
observed counterparts. This visualization assists in identifying data results obtained are due, on the one hand, to a sensible partial
points or clusters that the model struggles to forecast accurately. The replacement of the classical components, namely cement, and SF, while
figure additionally highlights anomalies in the predicted responses. maintaining a high packing density using the modified Andreasen and
Andersen model. On the other hand, it is also relevant to say that the
3.5. Multi-criteria optimization for ultrahigh-performance concrete strategy ensures a delicate balance between hydraulic and pozzolanic
reactions during concrete curing. In this sense, the partial replacement
The efficacy of dosages determined mathematically was appraised by of cement is supported by the incorporation of electric arc furnace slag
conducting tests under ideal conditions and juxtaposing the anticipated (EAFS), which exhibits a chemically similar composition as detailed in
results with the mixtures detailed in Table 10. Table 10 also illustrates Table 1. In addition, the potential impact on pozzolanic reactions due to
the Absolute Relative Deviation (ARD, %), a metric utilized to gauge the reduction in SF is partially compensated by the introduction of RGP,
prediction accuracy and as a validation instrument. The outcomes which provides a substantial amount of reactive silica (see Table 1),
corroborate that the empirical data closely aligns with forecasts from the counteracting the reduction of SF and maintaining crucial pozzolanic
multi-objective optimization process. Notably, the mixture formulated reactions, vital for the desirable properties of this special concrete.
through the optimization process and subsequently subjected to exper-
imental testing, yielded a significant reduction in PC content by around 3.6. Life cycle assessment
22% and a decrease in SF by more than 50%, relative to the conventional
RPC dosage proposed by Richard and Cheyrezy [1]. By opting for 3.6.1. UHPC mix designs
cost-effective substitutes like RGP and EASF, the mixture’s overall cost The comparison of LCA performance will be established between the
was curtailed by nearly 20%. Remarkably, this was achieved while optimized dosage and a Control dosage without alternative constituents.

Fig. 13. Q-Q plots for the model’s responses: (a) Φm , (b) C1, and (c) C28.

372
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 14. Predicted vs. target for the model’s responses: (a) Φm , (b) C1, and (c) C28.

Table 10 Table 12
Experimentally measured responses vs. model predictions. Results of the Life Cycle Assessment of the UHPC Mixes analyzed.
Mix Φm (mm) ARD C28 (MPa) ARD Impact category Unit Control Optimized
% % Mix
Experimental Model Experimental Model
Climate change kg CO2 eq 703.154 571.332
Optimum 253.00 250.35 1.05 150.21 158.32 1.24
Ozone depletion kg CFC11 eq 0.000 0.000
Photochemical ozone formation kg NMVOC 1.870 1.560
eq
A comprehensive breakdown of these mix components can be found in Acidification mol H+ eq 2.474 2.174
Table 11. This table also details the estimated transportation distances Eutrophication kg N eq 0.706 0.589
for each ingredient, including PC, SF, RGP, EAFS, SS, HRWRA, and Resource use, fossils MJ 3913.888 3723.211
water, based on the respective locations of their producers. The sce- Resource use, minerals, and metals kg Sb eq 0.002 0.002
Use of net fresh water m3 3.574 3.961
narios presented encompass various concrete mixes and the distances Total Energy from non-renewable MJ 4164.634 3968.736
from the suppliers of the individual components to the concrete mixing sources
facility. Total Energy from renewable sources MJ 362.647 330.995

3.6.2. Cradle-to-gate life cycle impacts assessment results

UHPC. One important finding that is being probed is that replacing
Table 12 provides a comparative assessment of the environmental
cement with slag provides a reduction in the Global Warming Potential
and energy indicators for the two UHPC mix variants. Based on the LCI
(GWP). Even though UHPC dosages studied in those pieces of research
findings, the optimized mix, incorporating glass powder (28 μm) and
are different from those evaluated in this investigation. The use of the
electric arc furnace slag throughout its production lifecycle, emerges as
UHPC Control mixture, 0% RGP, and EAFS in the mixes throughout its
the most sustainable option among all contenders. It exhibits superior
manufacturing is regarded as the most ecologically harmful option
environmental performance across nearly all environmental impact and
compared to the optimized mixture. However, it is important to note
energy metrics, except abiotic depletion and ozone layer degradation.
that the Control mixture achieves a C28 value of 167 MPa, while that of
For instance, the cumulative impact scores for GW100a are as follows:
the Optimized Mix is just 150 MPa. Control Mix exhibits the worst
GWP totals 571,332 kg CO2-eq, POCP stands at 1560 kg NMVOCeq, AP
environmental performance in 7 out of the 10 metrics for environmental
registers 2174 mol H+ eq, and EP is 0.588 kg N-eq. These figures
effects and energy consumption. The use of additional cementitious
translate to reductions of 19, 17, 12, and 17%, respectively, when
materials has advantages for the environment, according to research
compared with the potential environmental impacts tied to using the
authored by Faleschini et al. [107] about the decrease in effects, which
control mix, which contains 0% GP and EASF (refer to Table 11). These
may be categorized as direct, which is associated with the actual usage
advantages are graphically illustrated in Fig. 15, where the blue line
of materials, such as leaching and air emissions, and indirect, which is
delineates the percentage disparity between the control and optimized
caused by the energy and fossil resources used in the production process.
mixes. As highlighted earlier, this reveals a decline in four out of the
seven primary indicators and two out of the three resource indicators.
3.6.3. Component contribution analysis for each type of UHPC mix
These advantages align with the results documented by Refs.
Figs. 16 and 17 depict the contributions of various processes to the
[104–106] since the substitution of cement with industrial byproducts
impact categories being evaluated for all UHPC mixtures under
offers a viable alternative for mitigating the environmental effect of
consideration. Predominantly, cement and HRWA production processes
exhibit the most pronounced environmental impacts, a trend consistent
Table 11 across all effect categories and the three UHPC mixtures [108]. As
UHPC mixes alternatives used in this study. depicted in Figs. 14 and 15, the environmental contribution from
Component Control (kg/ Optimized (kg/ Distance Mode cement production can reach up to 93% and 90% for climate change and
m3) m3) (km) photochemical ozone creation metrics, respectively, for the Control Mix.
Cement 852 621 161 Lorry In contrast, HRWA production accounts for a 25% impact in the ODP
Silica Fume 272 100 17 Lorry category for the same Control Mix. Producing glass powder and electric
densified arc furnace slag are additional processes warranting attention, even
28-μm glass powder 0 342 4 Lorry though their peak contributions are limited to 18% for the ADP - min-
Electric Arc Furnace 0 112 414 Lorry
erals and metals score and 9% for the DOP in the Optimized Mix.
slag
Silica Sand 834 956 50 Lorry Conversely, the environmental footprint from the operation of concrete
HRWRA 26,5 27,5 7 Lorry mixing machinery remains relatively insubstantial, with contributions
Water 186 193 0 Lorry of less than 5% (3% EU score for Control and 4% for Optimized Mix).

373
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

Fig. 15. Life cycle impact assessment results. Key: Blue Line difference between Control and Optimized Mix. Orange: Optimized.

Fig. 16. Contribution analysis by component: Control.

Fig. 17. Contribution analysis by component: Optimized.

Based on the data and insights gathered, several observations can be the specific case study. Notably, integrating supplementary cementi-
made. However, it’s essential to interpret these as general trends, tious materials into UHPC formulations offers a promising avenue for
recognizing that the technical, historical, and geographical nuances of significantly mitigating environmental impacts throughout their life

374
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

cycle. This observation is underpinned by the understanding that References

cement has been identified as the primary contributor to environmental
burdens [79]. The predominant reason behind cement’s substantial [1] Richard P, Cheyrezy M. Composition of reactive powder concretes. Cement Concr
Res 1995;25:1501–11. https://doi.org/10.1016/0008-8846(95)00144-2.
environmental footprint lies in the clinker production process, which [2] Song J, Liu S. Properties of reactive powder concrete and its application in
requires substantial quantities of fossil fuels and minerals. highway bridge. Adv Mater Sci Eng 2016;2016:1–7. https://doi.org/10.1155/
2016/5460241.
[3] Abbas S, Nehdi ML, Saleem MA. Ultra-high performance concrete: mechanical
4. Implications, limitations, and conclusions performance, durability, sustainability and implementation challenges. Int J
Concr Struct Mater 2016;10:271–95. https://doi.org/10.1007/s40069-016-0157-
4.
The study outlines a method for producing eco-friendly UHPC with
[4] Jammes F, Cespedes X, Resplendino J. Design of offshore wind turbines. In:
reduced PC content. By incorporating EAFS and RGP, the optimized RILEM-Fib-AFGC int. Symp. Ultra-high perform. Fibre-reinforced concr. UHPFRC;
blend achieves high compressive strength while lowering carbon emis- 2013. p. 443–52.
sions and nonrenewable energy use. This research provides a sustainable [5] Tagnit-Hamou A, Soliman N, Omran A. Green ultra-high-performance glass
concrete. In: First international interactive symposium on UHPC. Iowa State
pathway for the construction industry to achieve high-performance University Digital Press; 2016. https://doi.org/10.21838/uhpc.2016.35.
infrastructure without compromising ecological responsibility. The [6] de Larrard F, Sedran T. Mixture-proportioning of high-performance concrete.
findings are essential for policymakers and industry leaders to balance Cement Concr Res 2002;32:1699–704. https://doi.org/10.1016/S0008-8846(02)
00861-X.
structural demands with environmental stewardship. However, the [7] Sharma R, Jang JG, Bansal PP. A comprehensive review on effects of mineral
study’s scope may not reflect future technological advancements or admixtures and fibers on engineering properties of ultra-high-performance
regional variations. The controlled environment of the research might concrete. J Build Eng 2022;45:103314. https://doi.org/10.1016/j.
jobe.2021.103314.
not foresee challenges in interactions, scale-up, and external factors like [8] Abellan-Garcia J, Fernández J, Khan MI, Abbas YM, Carrillo J. Uniaxial tensile
regulatory changes and market dynamics that could influence the ductility behavior of ultrahigh-performance concrete based on the mixture design
broader application of the findings. In light of the current study, the – partial dependence approach. Cem Concr Compos 2023;140:105060. https://
doi.org/10.1016/j.cemconcomp.2023.105060.
following conclusions were drawn:
[9] Liu J, Han F, Cui G, Zhang Q, Lv J, Zhang L, Yang Z. Combined effect of coarse
aggregate and fiber on tensile behavior of ultra-high performance concrete.
(1) Acknowledging the indispensable role of superplasticizers in Construct Build Mater 2016;121:310–8. https://doi.org/10.1016/j.
conbuildmat.2016.05.039.
UHPC production, the research underscores the adverse effects of
[10] Li PP, Yu QL, Brouwers HJH. Effect of coarse basalt aggregates on the properties
polycarboxylate on initial strength progression. However, it also of Ultra-high Performance Concrete (UHPC). Construct Build Mater 2018;170:
validates the positive influence of polycarboxylate super- 649–59. https://doi.org/10.1016/j.conbuildmat.2018.03.109.
plasticizers on the compressive strength at 28 days of UHPC. [11] Liu Y, Wei Y. Abrasion resistance of ultra-high performance concrete with coarse
aggregate. Mater Struct 2021;54:157. https://doi.org/10.1617/s11527-021-
(2) The study’s procedural approach, integrating the design of ex- 01750-6.
periments, modified packing theory, and multi-objective opti- [12] Yujing L, Wenhua Z, Fan W, Peipei W, Weizhao Z, Fenghao Y. Static mechanical
mization, proves its worth in achieving a UHPC dosage with properties and mechanism of C200 ultra-high performance concrete (UHPC)
containing coarse aggregates. Sci Eng Compos Mater 2020;27:186–95. https://
outstanding compressive strength (>150 MPa), self-compacting doi.org/10.1515/secm-2020-0018.
properties, and the incorporation of eco-friendly by-products. [13] Du J, Meng W, Khayat KH, Bao Y, Guo P, Lyu Z, Abu-obeidah A, Nassif H,
This optimized mixture features low cement (621 kg/m3) and Wang H. New development of ultra-high-performance concrete (UHPC). Compos
B Eng 2021;224:109220. https://doi.org/10.1016/j.compositesb.2021.109220.
silica fume (100 kg/m3) contents. [14] Meng W, Samaranayake A, Khayat KH. Factorial design and optimization of ultra-
(3) The inclusion of materials such as EAFS and RGP not only reduces high-performance concrete with lightweight sand. ACI Mater J 2018;115. https://
reliance on conventional cementitious components but also doi.org/10.14359/51700995.
[15] Abdal S, Mansour W, Agwa I, Nasr M, Abadel A, Onuralp Özkılıç Y, Akeed MH.
highlights the transformative potential of material substitution
Application of ultra-high-performance concrete in bridge engineering: current
for ecological sustainability. This is evidenced by the significant status, limitations, challenges, and future prospects. Buildings 2023;13:185.
reduction in carbon emissions and nonrenewable energy https://doi.org/10.3390/buildings13010185.
[16] Singh N, Colangelo F, Farina I. Sustainable non-conventional concrete 3D
consumption.
printing—a review. Sustainability 2023;15:10121. https://doi.org/10.3390/
(4) A comparative analysis reveals that the optimized UHPC dosage su151310121.
developed in this investigation outperforms a control mixture [17] Abdulkareem OM, Ben Fraj A, Bouasker M, Khelidj A. Effect of chemical and
with higher cement and silica fume contents. This is particularly thermal activation on the microstructural and mechanical properties of more
sustainable UHPC. Construct Build Mater 2018;169:567–77. https://doi.org/
evident in a nearly 20% reduction in kg CO2 eq. throughout the 10.1016/j.conbuildmat.2018.02.214.
LCA. [18] Ghanem Hassan, Obeid Yehia. The effect of steel fibers on the rheological and
mechanical properties of self compacting concrete. Eur Sci J 2015;11:85–98.
[19] Li W, Huang Z, Zu T, Shi C, Duan WH, Shah SP. Influence of nanolimestone on the
Future research could refine superplasticizer use in UHPC, explore hydration, mechanical strength, and autogenous shrinkage of ultrahigh-
novel eco-friendly materials, enhance procedural approaches, and assess performance concrete. J Mater Civ Eng 2016;28. https://doi.org/10.1061/(ASCE)
scalability and long-term performance. This builds on the study’s blue- MT.1943-5533.0001327.
[20] Khan K, Ahmad W, Amin MN, Rafiq MI, Abu Arab AM, Alabdullah IA,
print for sustainable UHPC, emphasizing optimization, material substi- Alabduljabbar H, Mohamed A. Evaluating the effectiveness of waste glass powder
tution, and real-world applicability, contributing to environmentally for the compressive strength improvement of cement mortar using experimental
responsible high-performance construction practices. and machine learning methods. Heliyon 2023;9:e16288. https://doi.org/
10.1016/j.heliyon.2023.e16288.
[21] Nguyen N-H, Abellán-García J, Lee S, Garcia-Castano E, Vo TP. Efficient
estimating compressive strength of ultra-high performance concrete using
Declaration of competing interest XGBoost model. J Build Eng 2022;52:104302. https://doi.org/10.1016/j.
jobe.2022.104302.
The authors declare that they have no known competing financial [22] Huang Z, Cao F. Effects of nano-materials on the performance of UHPC. 材料导报
研究篇 2012;26:136–41.
interests or personal relationships that could have appeared to influence [23] Wu Z, Khayat KH, Shi C, Tutikian BF, Chen Q. Mechanisms underlying the
the work reported in this paper. strength enhancement of UHPC modified with nano-SiO2 and nano-CaCO3. Cem
Concr Compos 2021;119:103992. https://doi.org/10.1016/j.
cemconcomp.2021.103992.
Acknowledgments [24] Wu Z, Shi C, Khayat KH. Multi-scale investigation of microstructure, fiber pullout
behavior, and mechanical properties of ultra-high performance concrete with
nano-CaCO3 particles. Cem Concr Compos 2018;86:255–65. https://doi.org/
The authors extend their appreciation to Researcher Supporting
10.1016/j.cemconcomp.2017.11.014.
Project number (RSPD2024R692), King Saud University, Riyadh,
Kingdom of Saudi Arabia.

375
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

[25] Camacho E, López JA, Serna P. Definition of three levels of performance for [50] Lin X, Peng Z, Yan J, Li Z, Hwang J-Y, Zhang Y, Li G, Jiang T. Pyrometallurgical
UHPFRC-VHPFRC with available materials. In: 3rd international symposium on recycling of electric arc furnace dust. J Clean Prod 2017;149:1079–100. https://
UHPC and nanotechnology for construction materials; 2013. p. 249–56. doi.org/10.1016/j.jclepro.2017.02.128.
[26] Camacho Torregosa E. Dosage optimization and bolted connections for UHPFRC [51] Coppola L, Buoso A, Coffetti D, Kara P, Lorenzi S. Electric arc furnace granulated
ties. of Valencia: Polytechnic University; 2013. PhD Thesis. slag for sustainable concrete. Construct Build Mater 2016;123:115–9. https://doi.
[27] Abellan-Garcia J, Martinez DM, Khan MI, Abbas YM, Pellicer-Martínez F. org/10.1016/j.conbuildmat.2016.06.142.
Environmentally friendly use of rice husk ash and recycled glass waste to produce [52] Lin X, Peng Z, Yan J, Li Z, Hwang J-Y, Zhang Y, Li G, Jiang T. Pyrometallurgical
ultra-high-performance concrete. J Mater Res Technol 2023;25:1869–81. https:// recycling of electric arc furnace dust. J Clean Prod 2017;149:1079–100. https://
doi.org/10.1016/j.jmrt.2023.06.041. doi.org/10.1016/j.jclepro.2017.02.128.
[28] Van Tuan N, Ye G, van Breugel K, Fraaij ALA, Bui DD. The study of using rice husk [53] Coppola L, Buoso A, Coffetti D, Kara P, Lorenzi S. Electric arc furnace granulated
ash to produce ultra high performance concrete. Construct Build Mater 2011;25: slag for sustainable concrete. Construct Build Mater 2016;123:115–9. https://doi.
2030–5. https://doi.org/10.1016/j.conbuildmat.2010.11.046. org/10.1016/j.conbuildmat.2016.06.142.
[29] Van Tuan YGN, Van Breugel K. Mitigation of early age shrinkage of ultra high [54] Zheng K. Recycled glass concrete. In: Eco-efficient concrete. Elsevier; 2013.
performance concrete by using rice husk ash. In: 3rd international symposium on p. 241–70. https://doi.org/10.1533/9780857098993.2.241.
UHPC and nanotechnology for construction materials; 2012. p. 341–8. [55] Jani Y, Hogland W. Waste glass in the production of cement and concrete – a
[30] Abellán-García J. Effect of rice husk ash as partial replacement of ordinary review. J Environ Chem Eng 2014;2:1767–75. https://doi.org/10.1016/j.
Portland cement in ultra-high-performance glass concrete. European Journal of jece.2014.03.016.
Environmental and Civil Engineering 2023:1–23. https://doi.org/10.1080/ [56] Guo P, Meng W, Nassif H, Gou H, Bao Y. New perspectives on recycling waste
19648189.2023.2219722. glass in manufacturing concrete for sustainable civil infrastructure. Construct
[31] Soliman NA, Tagnit-Hamou A. Partial substitution of silica fume with fine glass Build Mater 2020;257:119579. https://doi.org/10.1016/j.
powder in UHPC: filling the micro gap. Construct Build Mater 2017;139:374–83. conbuildmat.2020.119579.
https://doi.org/10.1016/j.conbuildmat.2017.02.084. [57] Shao Y, Lefort T, Moras S, Rodriguez D. Studies on concrete containing ground
[32] Amin M, Agwa IS, Mashaan N, Mahmood S, Abd-Elrahman MH. Investigation of waste glass. Cement Concr Res 2000;30:91–100. https://doi.org/10.1016/S0008-
the physical mechanical properties and durability of sustainable ultra-high 8846(99)00213-6.
performance concrete with recycled waste glass. Sustainability 2023;15:3085. [58] Redondo-Pérez NM, Redondo-Mosquera JD, Abellán-García J. A comprehensive
https://doi.org/10.3390/su15043085. overview of recycled glass as mineral admixture for circular UHPC solutions.
[33] Abellan-Garcia J, Iqbal Khan M, Abbas YM, Martínez-Lirón V, Carvajal-Muñoz JS. Sustainability 2024;16:5077. https://doi.org/10.3390/su16125077.
The drying shrinkage response of recycled-waste-glass-powder-and calcium- [59] Roth T. Working with the qualityTools package. http://Www.r-Qualitytools.Org;
carbonate-based ultrahigh-performance concrete. Construct Build Mater 2023; 2016.
379:131163. https://doi.org/10.1016/j.conbuildmat.2023.131163. [60] Abellán García J. Optimización de dosificaciones de hormigones de ultra altas
[34] Soliman NA, Tagnit-Hamou A. Using glass sand as an alternative for quartz sand prestaciones reforzados con fibras (UHPFRC) y aplicaciones como material de
in UHPC. Construct Build Mater 2017;145:243–52. https://doi.org/10.1016/j. refuerzo sísmico = Dosage optimization and seismic retrofitting applications of
conbuildmat.2017.03.187. Ultra-High-Performance Fiber Reinforced Concrete (UHPFRC). Universidad
[35] Yu R, Tang P, Spiesz P, Brouwers HJH. A study of multiple effects of nano-silica Politécnica de Madrid; 2021. https://doi.org/10.20868/UPM.thesis.66122.
and hybrid fibres on the properties of Ultra-High Performance Fibre Reinforced [61] Guo P, Bao Y, Meng W. Review of using glass in high-performance fiber-
Concrete (UHPFRC) incorporating waste bottom ash (WBA). Construct Build reinforced cementitious composites. Cem Concr Compos 2021;120:104032.
Mater 2014;60:98–110. https://doi.org/10.1016/j.conbuildmat.2014.02.059. https://doi.org/10.1016/j.cemconcomp.2021.104032.
[36] Alturki A. The global carbon footprint and how new carbon mineralization [62] Ismail ZZ, Al-Hashmi EA. Recycling of waste glass as a partial replacement for
technologies can Be used to reduce CO2 emissions. ChemEngineering 2022;6:44. fine aggregate in concrete. Waste Manag 2009;29:655–9. https://doi.org/
https://doi.org/10.3390/chemengineering6030044. 10.1016/j.wasman.2008.08.012.
[37] Hou Z, Luo J, Xie Y, Wu L, Huang L, Xiong Y. Carbon circular utilization and [63] Redondo-Pérez NM, Redondo-Mosquera JD, Abellán-García J. A comprehensive
partially geological sequestration: potentialities, challenges, and trends. Energies overview of recycled glass as mineral admixture for circular UHPC solutions.
2022;16:324. https://doi.org/10.3390/en16010324. Sustainability 2024;16:5077. https://doi.org/10.3390/su16125077.
[38] Anika OC, Nnabuife SG, Bello A, Okoroafor ER, Kuang B, Villa R. Prospects of low [64] American Society for Testing and Materials. Standard test method for flow of
and zero-carbon renewable fuels in 1.5-degree net zero emission actualisation by hydraulic cement mortar. Conshohocken, PA: ASTM C1437); 2016.
2050: a critical review. Carbon Capture Science & Technology 2022;5:100072. [65] American Society for Testing and Materials. Standard test method for
https://doi.org/10.1016/j.ccst.2022.100072. compressive strength of hydraulic cement mortars (using 2-in . or [50-mm] cube
[39] Huseien GF, Shah KW, Sam ARM. Sustainability of nanomaterials based self- specimens). West Conshohocken, PA: ASTM; 2010. C109/C109M.
healing concrete: an all-inclusive insight. J Build Eng 2019;23:155–71. https:// [66] Abellan J, Torres N, Núñez A, Fernández J. Influencia del exponente de Fuller, la
doi.org/10.1016/j.jobe.2019.01.032. relación agua conglomerante y el contenido en policarboxilato en concretos de
[40] Zhou D, Huang F, Wang Q, Liu X. The role of structure change in driving CO2 muy altas prestaciones. In: IV Congreso Internacional de Ingeniería Civil, La
emissions from China’s waterway transport sector. Resour Conserv Recycl 2021; Habana (Cuba); 2018.
171:105627. https://doi.org/10.1016/j.resconrec.2021.105627. [67] Lenth RV. Response-surface methods in R. J Stat Software 2009;32. https://doi.
[41] Peng J, Xie R, Lai M. Energy-related CO2 emissions in the China’s iron and steel org/10.18637/jss.v032.i07.
industry: a global supply chain analysis. Resour Conserv Recycl 2018;129: [68] Raviselvan RJ, Ramanathan K, Perumal P, Thansekhar MR. Response surface
392–401. https://doi.org/10.1016/j.resconrec.2016.09.019. methodology for optimum hardness of TiN on steel substrate. International
[42] Yu R, Spiesz P, Brouwers HJH. Mix design and properties assessment of ultra-high Journal of Materials and Metallurgical Engineering 2015;9:1448–54.
performance fibre reinforced concrete (UHPFRC). Cement Concr Res 2014;56: [69] Ghafari E, Costa H, Júlio E. RSM-based model to predict the performance of self-
29–39. https://doi.org/10.1016/j.cemconres.2013.11.002. compacting UHPC reinforced with hybrid steel micro-fibers. Construct Build
[43] Mu X, Zhang S, Ni W, Xu D, Li J, Du H, Wei X, Li Y. Performance optimization and Mater 2014;66:375–83. https://doi.org/10.1016/j.conbuildmat.2014.05.064.
hydration mechanism of a clinker-free ultra-high performance concrete with solid [70] Abellán J, Fernández J, Torres N, Núñez A. Statistical optimization of ultra-high-
waste based binder and steel slag aggregate. J Build Eng 2023;63:105479. performance glass concrete. ACI Mater J 2020;117. https://doi.org/10.14359/
https://doi.org/10.1016/j.jobe.2022.105479. 51720292.
[44] Xu D, Tang J, Hu X, Yu C, Han F, Sun S, Deng W, Liu J. The influence of curing [71] Branchu S, Forbes RT, York P, Nyqvist H. A central composite design to
regimes on hydration, microstructure and compressive strength of ultra-high investigate the thermal stabilization of lysozyme. Pharm Res (N Y) 1999;16:
performance concrete: a review. J Build Eng 2023;76:107401. https://doi.org/ 702–8. https://doi.org/10.1023/A:1018876625126.
10.1016/j.jobe.2023.107401. [72] Abellan-Garcia J, Redondo-Mosquera J, Khan MI, Abbas YM, Castro-Cabeza A.
[45] Abellán García J, Fernández Gómez J, Torres Castellanos N. Properties prediction Development of a novel 124 MPa strength green reactive powder concrete
of environmentally friendly ultra-high-performance concrete using artificial employing waste glass and locally available cement. Arch Civ Mech Eng 2023;23:
neural networks. European Journal of Environmental and Civil Engineering 2022; 161. https://doi.org/10.1007/s43452-023-00695-7.
26:2319–43. https://doi.org/10.1080/19648189.2020.1762749. [73] Abellan-Garcia J, Iqbal Khan M, Abbas YM, Castro-Cabeza A, Carrillo J. Multi-
[46] Abellan-Garcia J, Fernández J, Khan MI, Abbas YM, Carrillo J. Uniaxial tensile criterion optimization of low-cost, self-compacted and eco-friendly micro-
ductility behavior of ultrahigh-performance concrete based on the mixture design calcium-carbonate- and waste-glass-flour-based ultrahigh-performance concrete.
– partial dependence approach. Cem Concr Compos 2023;140:105060. https:// Construct Build Mater 2023;371:130793. https://doi.org/10.1016/j.
doi.org/10.1016/j.cemconcomp.2023.105060. conbuildmat.2023.130793.
[47] Abellan-Garcia J, Iqbal Khan M, Abbas YM, Castro-Cabeza A, Carrillo J. Multi- [74] Derringer G, Suich R. Simultaneous optimization of several response variables.
criterion optimization of low-cost, self-compacted and eco-friendly micro- J Qual Technol 1980;12:214–9. https://doi.org/10.1080/
calcium-carbonate- and waste-glass-flour-based ultrahigh-performance concrete. 00224065.1980.11980968.
Construct Build Mater 2023;371:130793. https://doi.org/10.1016/j. [75] The European Project Group. The European guidelines for self-compacting
conbuildmat.2023.130793. concrete. 2005.
[48] Funk JE, Dinger D. Predictive process control of crowded particulate suspensions: [76] Deutsche NORM. Core rules for the product category of construction products EN
applied to ceramic manufacturing. New York: Springer Science; 1994. 15804:2012+A2:2019 + AC:2021. www.beuth.de; 2021.
[49] Coppola L, Buoso A, Coffetti D, Kara P, Lorenzi S. Electric arc furnace granulated [77] Pré Sustainability BV. SimaPro. 2022.
slag for sustainable concrete. Construct Build Mater 2016;123:115–9. https://doi. [78] Deutsche NORM. Core rules for the product category of construction products EN
org/10.1016/j.conbuildmat.2016.06.142. 15804:2012+A2:2019 + AC:2021. 2021.

376
M.I. Khan et al. Journal of Materials Research and Technology 32 (2024) 362–377

[79] Cassiani JD, Wünsch J, Keßler S. Sustainable concrete formulations for [95] Pott U, Ehm C, Jakob C, Stephan D. Investigation of the early cement hydration
sustainable reinforced structures exposed to chloride induced corrosion. MATEC with a new penetration test. Rheometry and In-Situ XRD 2020:246–55. https://
Web of Conferences 2022;361:06001. https://doi.org/10.1051/matecconf/ doi.org/10.1007/978-3-030-22566-7_29.
202236106001. [96] Struble LJ, Lei W-G. Rheological changes associated with setting of cement paste.
[80] European Commission. Joint research centre. Institute for environment and Adv Cement Base Mater 1995;2:224–30. https://doi.org/10.1016/1065-7355(95)
sustainability, international reference life cycle data system (ILCD) handbook 90041-1.
general guide for life cycle assessment : detailed guidance. Publications Office; [97] Jakob C, Jansen D, Ukrainczyk N, Koenders E, Pott U, Stephan D, Neubauer J.
2010. Relating ettringite formation and rheological changes during the initial cement
[81] Araujo DLV, Santos J, Martinez-Arguelles G. Environmental performance hydration: a comparative study applying XRD analysis, rheological measurements
evaluation of warm mix asphalt with recycled concrete aggregate for road and modeling. Materials 2019;12:2957. https://doi.org/10.3390/ma12182957.
pavements. Int J Pavement Eng 2022:1–14. https://doi.org/10.1080/ [98] Kubens S. Interaction of cement and admixtures and its influence on rheological
10298436.2022.2064999. properties, Göttingen, vol. 49; 2010.
[82] ASTM International. An industry average environmental product declaration for [99] Puertas F, Santos H, Palacios M, Martnez-Ramrez S. Polycarboxylate
slag cement. 2021. superplasticiser admixtures: effect on hydration, microstructure and rheological
[83] Abellan-Garcia J, Martinez DM, Khan MI, Abbas YM, Pellicer-Martínez F. behaviour in cement pastes. Adv Cement Res 2005;17:77–89. https://doi.org/
Environmentally friendly use of rice husk ash and recycled glass waste to produce 10.1680/adcr.17.2.77.65044.
ultra-high-performance concrete. J Mater Res Technol 2023;25:1869–81. https:// [100] Soliman NA, Tagnit-Hamou A. Using glass sand as an alternative for quartz sand
doi.org/10.1016/j.jmrt.2023.06.041. in UHPC. Construct Build Mater 2017;145:243–52. https://doi.org/10.1016/j.
[84] Pradhan S, Tiwari BR, Kumar S, Barai SV. Comparative LCA of recycled and conbuildmat.2017.03.187.
natural aggregate concrete using Particle Packing Method and conventional [101] Huang H, Gao X, Wang H, Ye H. Influence of rice husk ash on strength and
method of design mix. J Clean Prod 2019;228:679–91. https://doi.org/10.1016/ permeability of ultra-high performance concrete. Construct Build Mater 2017;
j.jclepro.2019.04.328. 149:621–8. https://doi.org/10.1016/j.conbuildmat.2017.05.155.
[85] Turk J, Cotič Z, Mladenovič A, Šajna A. Environmental evaluation of green [102] Rangaraju PR, Li Z. Development of UHPC using ternary blends of ultra-fine class
concretes versus conventional concrete by means of LCA. Waste Manag 2015;45: F fly ash, meta-kaolin and Portland cement. In: First international interactive
194–205. https://doi.org/10.1016/j.wasman.2015.06.035. symposium on UHPC. Iowa State University Digital Press; 2016. https://doi.org/
[86] Montgomery DC. Design and analysis of experiments. John wiley & sons; 2017. 10.21838/uhpc.2016.64.
[87] R Core Team, R. A language and environment for statistical computing. 2018. [103] Ibrahim MA, Farhat M, Issa MA, Hasse JA. Effect of material constituents on
Vienna, Austria. mechanical and fracture mechanics properties of ultra-high-performance
[88] Struble LJ, Lei W-G. Rheological changes associated with setting of cement paste. concrete. ACI Mater J 2017;114. https://doi.org/10.14359/51689717.
Adv Cement Base Mater 1995;2:224–30. https://doi.org/10.1016/1065-7355(95) [104] Zhong R, Wille K, Viegas R. Material efficiency in the design of UHPC paste from a
90041-1. life cycle point of view. Construct Build Mater 2018;160:505–13. https://doi.org/
[89] Uchikawa H, Ogawa K, Uchida S. Influence of character of clinker on the early 10.1016/j.conbuildmat.2017.11.049.
hydration process and rheological property of cement paste. Cement Concr Res [105] Shamass R, Rispoli O, Limbachiya V, Kovacs R. Mechanical and GWP assessment
1985;15:561–72. https://doi.org/10.1016/0008-8846(85)90053-5. of concrete using blast furnace slag, silica fume and recycled aggregate. Case Stud
[90] Pott U, Ehm C, Jakob C, Stephan D. Investigation of the early cement hydration Constr Mater 2023;18:e02164. https://doi.org/10.1016/j.cscm.2023.e02164.
with a new penetration test. Rheometry and In-Situ XRD 2020:246–55. https:// [106] Martin C, Manu E, Hou P, Adu-Amankwah S. Circular economy, data analytics,
doi.org/10.1007/978-3-030-22566-7_29. and low carbon concreting: a case for managing recycled powder from end-of-life
[91] Pott U, Ehm C, Jakob C, Stephan D. Investigation of the early cement hydration concrete. Resour Conserv Recycl 2023;198:107197. https://doi.org/10.1016/j.
with a new penetration test. Rheometry and In-Situ XRD 2020:246–55. https:// resconrec.2023.107197.
doi.org/10.1007/978-3-030-22566-7_29. [107] Faleschini F, De Marzi P, Pellegrino C. Recycled concrete containing EAF slag:
[92] Uchikawa H, Ogawa K, Uchida S. Influence of character of clinker on the early environmental assessment through LCA. European Journal of Environmental and
hydration process and rheological property of cement paste. Cement Concr Res Civil Engineering 2014;18:1009–24. https://doi.org/10.1080/
1985;15:561–72. https://doi.org/10.1016/0008-8846(85)90053-5. 19648189.2014.922505.
[93] Nonat A. Interactions between chemical evolution (hydration) and physical [108] Cassiani JD, Wünsch J, Keßler S. Sustainable concrete formulations for
evolution (setting) in the case of tricalcium silicate. Mater Struct 1994;27: sustainable reinforced structures exposed to chloride induced corrosion. MATEC
187–95. https://doi.org/10.1007/BF02473032. Web of Conferences 2022;361:06001. https://doi.org/10.1051/matecconf/
[94] Uchikawa H, Ogawa K, Uchida S. Influence of character of clinker on the early 202236106001.
hydration process and rheological property of cement paste. Cement Concr Res
1985;15:561–72. https://doi.org/10.1016/0008-8846(85)90053-5.

377