Fuel 295 (2021) 120544

Contents lists available at ScienceDirect

Fuel
journal homepage: www.elsevier.com/locate/fuel

Full Length Article

Alternative fuels co-fired with natural gas in the pre-calciner of a cement

plant: Energy and material flows
Daya R. Nhuchhen *, Song P. Sit , David B. Layzell
Canadian Energy Systems Analysis Research (CESAR) Initiative, University of Calgary, Calgary, AB, Canada

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

Keywords: Cement-making is an energy-intensive industrial process that contributes 8% of the global CO2 emissions. This
Natural gas-fired cement plant study develops a thermal energy flow model (TEF) for 4200 tonnes of clinker per day, a natural gas-fired cement
Alternative fuels plant in which 50% of the pre-calciner energy requirements can be supplied by alternative fuels (AF) including
Co-firing
biomass, plastics, etc. The TEF shows that the lower heating value (LHV dry), oxygen content (dry basis) and
Thermal energy flow
moisture content of the AF, as well as the flue gas oxygen concentration [O2] needed for complete combustion,
Energy demand
GHG emissions affect the thermal energy intensity (TEI) and total air demand (AD). Compared to the reference system fueled by
Regression natural gas (NG) that burns completely at 1% [O2] in flue gas, all AFs require more total air and thermal energy
demand, especially when the solid AFs need a higher [O2] in the flue gas (typically 3%) to ensure complete
combustion. Due to the higher carbon (C) intensity, co-firing AFs could increase the total CO2 emissions by 1% to
18%. For the wood dust with 100% biogenic C, 50% NG replacement in pre-calciner could avoid 55.5 and 43.1
kgCO2/t clinker at 1% and 3% [O2] in flue gas respectively, an equivalent of 7.5% and 5.8% decrease in total
GHG emissions relative to the reference case. Results of the TEF model from 24 diverse AFs were used to generate
regressions that link fuel properties and flue gas oxygen requirement with TEI, making it possible to compare
highly diverse AFs for their likely performance in the clinker-making.

1. Introduction Keeping organic materials out of landfill sites is also known to eliminate
the emission of methane, a potent GHG [10].
The global cement sector accounts for about 8% of anthropogenic Tsiliyannis [11] provided a detailed thermal model describing mass
greenhouse gas (GHG) emissions [1] and product demand is expected to and energy flow in a cement plant where AFs were co-fired with coal and
grow by 12–23% by 2050 [2]. About 90% of the GHG emissions [3] are analyzed the impact of AF on clinker productions and GHG emissions.
associated with the production of clinker (clk) which makes up about Rahman et al. [12] extended this work using ASPEN Plus process
65% of cement by weight [2]. modeling. Ariyaratne et al. [6] and Tsiliyannis [13] also used thermal
Global average CO2 emissions are 831 kgCO2/t clk [2], about 560 energy models to quantify the benefits of oxygen enrichment in a cement
kgCO2/t clk of which comes from the conversion of calcium carbonate to plant where AFs were co-fired with coals.
calcium oxide, known as process emissions (PE). The rest of the GHG Natural gas (NG) can also be used for clinker production as a low
footprint (~271 kgCO2/t clk) is associated with energy emissions (EE) to carbon alternative (carbon intensity of 56 kgCO2/GJLHV) [4] to coal or
drive the thermal process of the clinker-making process. pet-coke. One benefit of NG is that complete combustion can be obtained
In most cement plants, coal or pet-coke provides the primary fuel and with relatively low excess air supply (i.e. flue gas oxygen concentration
has a carbon intensity of about 97 kgCO2 per GJLHV [4], resulting in the of about 1% by molar basis), while solid fuels like coal or pet-coke often
EE of about 281 to 611 kg CO2/t clk depending upon the thermal energy need more excess air, resulting in a higher flue gas oxygen concentration
demand of the facility (range from 2.9 to 6.3 GJ/t clk) [5]. Some coal or (2% to 7%) to achieve complete combustion [6,13–15]. Akhtar and co-
pet-coke fired cement plants [6–8] incorporate waste-derived alterna­ workers [14] provided a computational analysis that compared the
tive fuels (AF) to reduce costs [9] and eliminate waste streams. If the AF combustion behavior of coal and NG in the kiln and pre-calciner of a
replacing coal or pet-coke are bio-based, GHGs also can be reduced. cement plant while Atsonios et al. [16] reported on how different fuels

* Corresponding author.
E-mail addresses: [email protected], [email protected] (D.R. Nhuchhen).

https://doi.org/10.1016/j.fuel.2021.120544
Received 2 November 2020; Received in revised form 21 February 2021; Accepted 22 February 2021
Available online 19 March 2021
0016-2361/© 2021 Elsevier Ltd. All rights reserved.
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

including NG firing in a cement plant would impact the technology used The findings of the TEF model are also used to generate the regres­
for carbon capture process for geological storage. sion equation that relates the AFs fuel properties and the required flue
To the knowledge of the authors, no studies have been published on gas oxygen concentration with the TEI requirements for an NG-fired
the use of AFs in natural gas-fired cement plant. This study is focused on cement plant.
addressing this knowledge gap.
In the province of Alberta in Canada, NG is available at a low price so 2. Methodology
it is used by one cement plant (LafargeHolcim Inc, Exshaw, Alberta) to
achieve a CO2 emissions intensity for clinker production of 738 kgCO2/t 2.1. Properties of alternative fuels
clk. The plant must report on its GHG emissions under the province’s
Technology Innovation and Emissions Reduction (TIER) regulations Details on the lower heating values (LHV) and elemental composi­
[17]. The regulation allocates an emissions intensity benchmark of 776 tion of NG and 24 different AFs obtained from the published literature
kgCO2/t clk including both process and energy emissions. A facility [20] are arranged in the order of their carbon intensity (Table 1). It also
exceeding the benchmark will pay $30/tCO2e rising to $50/tCO2e in shows biogenic carbon content published in the literature [21–23].
2022 and to $170/tCO2e in 2030 [18]. Alternatively, a facility per­ Three bolded AFs: High-density polythene (HDPE), railway ties 2 (RT2)
forming below the benchmark can earn emissions reduction offset and wood dust (WD) with 0%, 74%, and 100% biogenic carbon [22],
credits. Despite the low emission intensity of their NG-fired cement respectively were selected to examine the details of the energy and
plant, LafargeHolcim is interested in exploring the potential for AFs to material flows in this study.
reduce both the fuel cost and the CO2 emissions intensity of clinker
production [19]. 2.2. Process flow diagram
In this study, a thermal energy flow (TEF) model describes the
operation of a natural gas-fired cement plant without and with 24 A simplified process flow diagram for the 4200 tonnes clinker per
different waste-derived AFs providing 50% of the energy requirements day cement plant that is the focus of this study is shown in Fig. 1. It
for the pre-calciner. The study explores how the embedded elemental includes a 4-stage raw meal preheater, an inline pre-calciner, rotary kiln,
oxygen content (O), lower heating value (LHV) and moisture content and clinker cooler. NG and/or AFs in the pre-calciner and only NG in the
(MC) of the AFs impacted their thermal energy intensity (TEI, GJLHV/t rotary kiln are supplied to drive the calcining and sintering reactions.
clk), air demand (AD, Nm3/t clk) and CO2 emissions intensity (kgCO2/t Raw materials (RM) with 3% moisture (as received basis) are milled in a
clk). Since the studied AFs are solid fuels and may require more excess vertical rotating mill and dried with the help of the flue gas and exhaust
air to ensure complete combustion, the TEF model also assesses the ef­ vent air before being fed into the preheater tower. The key plant-specific
fect of flue gas oxygen concentrations ([O2]) on TEI, AD and GHG operational temperatures and other parameters for the cement plant are
emissions of the clinker production process. shown in Fig. 1 and the shaded portion indicates the major changes that

Table 1
Characteristics of solid alternative fuels [20].
Alternative fuels LHV (MJ/kg - Dry Elemental analysis (%Wt. - Dry basis) Carbon intensity (kg Biogenic C (% of
basis) C H O N Ar S Cl Ash CO2/GJLHV) total C)

Natural gas (NG)a 47.57 74.15 24.58 0.01 1.26 0.00 0.00 0.00 0.00 57 0
Rubber chips (RC) 35.94 65.84 10.86 5.77 0.78 0.00 1.53 0.82 14.40 67 0
High-density polythene 43.29 86.10 13.00 0.90 0.00 0.00 0.00 0.00 0.00 73 0
(HDPE)
TV back plate (TVBP) 37.73 82.20 7.21 2.96 0.99 0.00 0.03 5.39 1.22 80 0
Electronic plastic waste 1 30.73 67.40 6.84 7.66 2.94 0.00 0.11 4.44 10.60 80 0
(EPW1)
Carpet grains (CG) 20.34 45.09 5.78 30.06 2.78 0.00 0.11 0.00 16.20 81 0
Electronic plastic waste 2 25.13 56.60 5.97 8.24 2.81 0.00 0.10 7.38 18.90 83 0
(EPW2)
Spent coffee (SC) 25.29 57.00 7.60 32.90 2.10 0.00 0.10 0.00 0.30 83 100
Asphalt shingles (AS) 13.48 31.64 3.55 20.81 0.17 0.00 1.43 0.00 42.40 86 20b
Waste tire (WT) 32.89 82.79 6.99 7.67 0.67 0.00 1.56 0.28 0.06 92 0
Soybean oil cake (SOC) 15.38 38.80 5.74 41.86 6.64 0.00 0.00 0.00 6.96 92 100
Brewing industries waste 20.23 51.10 6.87 32.51 4.66 0.00 0.44 0.02 4.40 93 100
(BIW)
Coffee husk (CH) 18.80 49.40 6.10 41.12 0.81 0.00 0.07 0.00 2.50 96 100
Shuttering wood (SW) 17.14 45.46 5.51 37.86 1.79 0.00 0.03 0.00 9.36 97 100
Wood dust (WD) 16.06 43.14 4.84 32.34 0.66 0.00 0.51 0.13 18.37 98 100
Bone meal (BM) 18.28 49.16 7.04 16.73 10.21 0.00 0.63 0.97 15.26 99 100
Railway ties 1 (RT1) 19.73 53.50 5.50 39.78 0.30 0.00 0.01 0.01 0.90 99 74b
Painted wood (PW) 18.40 49.90 5.43 42.02 0.46 0.00 0.04 0.07 2.08 99 52c
Telephone poles (TP) 20.33 55.66 6.02 34.51 0.44 0.00 0.56 0.40 2.41 100 53c
Railway ties 2 (RT2) 20.37 56.11 6.06 35.83 0.47 0.00 0.12 0.04 1.37 101 74b
Lignin from wood (LW) 21.94 62.17 5.89 31.11 0.16 0.00 0.06 0.00 0.62 104 100
Cardboard (CB) 15.39 44.52 5.68 41.18 0.10 0.00 0.12 0.00 8.40 106 100d
Empty fruit brunches palm 15.56 45.53 5.46 43.39 0.46 0.00 0.04 0.00 5.12 107 100
oil (PO)
Textile (T) 14.03 41.11 6.96 50.90 0.01 0.00 0.84 0.00 0.20 107 100b
Plywood (P) 15.89 49.53 6.17 41.98 0.58 0.00 0.03 0.00 1.80 114 54c
a
Composition of NG: Methane-96.9%; Ethane-2.3%; Propane-0.06%; Iso-Butane-0.004%; Normal Butane-0.004%; Traces amount of Hexane; Nitrogen 0.7%; Carbon
dioxide-0.005%, and lower heating value: 35 MJ/Nm3 (Provided by the industry partner of this study).
b
[22];
c
[21];
d
[23].

2
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

Fig. 1. Process layout of clinker-making

technology of a cement plant. Note: AF -
Alternative fuel; CA - Conveying air; CCA -
Clinker cooler air; EVA - Exhaust vent air;
FG - Flue gas; LA - Leak air; M - Materials;
MC - Moisture content; NG - Natural gas;
PA - Primary air; QA - Quenching air; RM -
Raw materials; SA - Secondary air; TA -
Tertiary air. All shown numbers are plant-
specific data. The italic numbers are input
variables (plant-specific data points) and
the underlined numbers are calculated
values. (Solid line – Solid flow; Dash line –
Gas (Flue gas – Dark brown and Air - Light
blue) flow; Dot line – Fuel flow). (For
interpretation of the references to colour in
this figure legend, the reader is referred to
the web version of this article.)

this study investigated for the AF co-firing. In all co-firing cases, all 2.3.2. Fuel combustion
temperatures except the temperature of the flue gas at the exit of the As shown in Fig. 1, fuels are burned using secondary and primary air
preheater (underlined values in Fig. 1) were assumed to be similar to in the rotary kiln and using tertiary air in the pre-calciner to support the
that of the reference case cement plant with 100% NG combustion. thermal energy needed in the clinker-making process. The rotary kiln
brings in enough air to ensure that 6.5% [O2] in the kiln flue gas passes
on to the pre-calciner, and the pre-calciner brings in enough air to ensure
2.3. Thermal energy flow model complete combustion of fuel, leaving the flue gas [O2] 1% or higher. To
estimate the fuel and air demands and to predict the amount of flue gas
A thermal energy flow (TEF) model was built for the cement plant and its composition, the stoichiometric combustion analysis and an
depicted in Fig. 1. The process flow diagram including all materials actual combustion analysis in the rotary kiln and pre-calciner were
flowing in and out from different components used for modeling the established. They are summarized in Appendix B. The developed TEF
clinker-making process is shown in Figure S1 in the supplementary file. model allows to assess AF co-firing with NG both in the rotary kiln and
pre-calciner, but this study has examined only the case of AF co-firing in
2.3.1. Raw material and chemical reactions in the pre-calciner and kiln the pre-calciner.
Raw materials were supplied by mixing sandstone, black shale, red
shale, mill scale, and limestone so that the mixture could provide 2.3.3. Mass and energy balances
enough oxides of calcium (CaO), aluminium (Al2O3), silicon (SiO2) and Fig. 2 explains the mass and energy balances in the pre-calciner and
iron (Fe2O3) necessary to form clinker. The plant-specific mineral rotary kiln that were established to estimate the fuel demand in the
compositions of respective sources of the raw materials and the clinker clinker-making process. The derived expressions are presented in Ap­
are presented in Table S1. As this study aims to examine the effect of pendix C. As can be seen from Fig. 2, the amount of fuel (NG or NG and
using AFs in the pre-calciner, the composition of clinker, hence the AF) and the flue gas flowrate are interrelated, therefore an iterative
proportion of raw materials mix, was kept constant. Premixed raw ma­ approach was used to determine the fuel demand needed to achieve the
terials are milled and dried before it is fed into the pre-heater tower. set process temperature of the clinker-making process. A detailed mass
Typically, the pre-calciner requires a temperature range of 700–950 ◦ C and energy balances of all the remaining components of the clinker-
[24] and 1450–2000 ◦ C in the rotary kiln [11,13,25,26] to establish making process is presented in Section 5 of the supplementary file.
calcining and sintering reactions, respectively.
This study assumed no chemical decompositions reactions in the 2.3.4. Specific heat capacity of gases and solids
preheater tower [27] and incorporated 6 chemical reactions in the pre- Since the clinker-making process involves a wide range of tempera­
calciner and rotary kiln as presented in Table A.1 in Appendix A. The ture, the sensible heat of flue gas and solid materials entering in and
enthalpies of the calcining and sintering reactions were estimated using leaving out from each component was estimated by using the
the standard heat of formation of the molecules in the chemical re­ temperature-dependent expressions for the specific heat capacity. As
actions. The standard heat of the formation of different molecules is described in Appendix D, the temperature-dependent Shomate
provided in Table S2.

3
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

Fig. 2. Steady state flow balance in pre-calciner and kiln: A) Carbon ( Source
Destination Con/out ); B) Oxygen ( Destination Oon/out ); and C) Energy ( Destination Qon/out ), where fuel
Form Source Form Source Form

supply is depicted as red, air supply as blue, flue gases as brown, solid materials as black, and heat losses as purple. (For interpretation of the references to colour in
this figure legend, the reader is referred to the web version of this article.)

equation [28] was used to determine the specific heat capacity of gases • The changes in electrical power demand that is needed to handle the
in the flue gas. To estimate the specific heat capacities of the major additional air demand and flue gas flow are important in the cost
components of raw materials and clinkers such as SiO2, Al2O3, Fe2O3, analysis and life cycle GHG emissions, but it does not affect the
CaCO3, MgCO3, CaO, MgO, tricalcium silicate (Alite-C3S), dicalcium overall thermal process of clinker-making. Therefore, the estimation
silicate (Belite-C2S), tricalcium aluminate (C3A) and tetracalcium alu­ of electrical power demand was not part of the TEF model.
minoferrite (C4AF), the temperature-dependent functions were used • Materials flows were steady-state and gases were ideal.
[29–31]. The temperature-dependent correlations of specific heat ca­ • In the reference case, 60% of GJ supplied was in the pre-calciner and
pacity of solid components are presented in Table S4 in the supple­ the rest in the rotary kiln.
mentary file. • This study, because it examined the case when 50% of GJ demand in
the pre-calciner was to be substituted by AFs, used θAF− PC = 0.5.
2.3.5. Case study and model assumptions • Any air leakages except in the kiln (3% of the air supply) were
The developed TEF model represented a layout of a dry clinker- neglected.
making process plant of capacity 4200 tonnes per day. The plant has a • The solid collection efficiency of cyclone separators (Cy: II to Cy: V,
4-stage cyclone preheater system followed by a long cylindrical vertical Cy: EVA) was 95% except for Cy: I = 98% (refer Figure S1).
pre-calciner that completes the majority of the calcining reactions. The • While the dust leaving kilns were neglected, the dust leaving the
calcined raw materials move along with the flue gas from the pre- preheater tower was estimated using the efficiency of the cyclone
calciner through a bottom cyclone separator (Cy: V, see Figure S1) separators.
and then enter into the rotary kiln. The plant uses natural gas as a • Clinker dust leaving exhaust vent air cyclone was 1% of clinker
baseline fuel in the pre-calciner and rotary kiln. In addition, the production.
following assumptions were made: • The temperature of the flue gas and the entrained solids leaving
together from the pre-calciner and the cyclone separators are in
• Since AF co-firing could impact other aspects such as environmental thermal equilibrium.
performance (air pollutants level), impact on traffic due to AF • Since natural gas-fired cement plants tend to produce 26%-54%
transport and health issues of burning AF, their studies [32] have more NOx than a coal-fired cement plant [33], the facility studied
been carried out separately and therefore were not considered here. here was equipped with a Low NOx Calciner Burner and selective
In addition, due to the design flexibility of the cement plant, the non-catalytic reduction [34]. Therefore, this study excluded a
impact of AF co-firing on the clinker production rate and operational detailed analysis of NOx formation.
issues were also excluded from this study. • The ash (non-combustible material) content in AFs was incorporated
in the calculation of the energy and material flows in the TEF model,

4
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

but the impact of mineral compositions of the ash on the clinker 2.4. Generalized regression equations
quality was not included.
• The developed model assumed there will not be unburn carbon if AF Using the results of the energy demands of clinker-making for 24 AFs
is co-fired in the rotary kiln, but there may be unburned char in the and NG predicted by the TEF model (Results of reference case and case
pre-calciner. This study, however, assumed that AF burned 1), regression equations were developed for thermal energy intensity
completely without leaving any unburned char in the pre-calciner (TEI, MJ/t clk) using lower heating value (LHV) and embedded
(m9 = 0 in Eq.(B.3); CUB− C− K− i = 0 in Eq.(B.2)). elemental oxygen fraction (O) of fuel.
Since a linear regression equation did not provide a good fit to the
The TEF model results for the reference case was validated using the TEF modeled data, resulting in a higher mean absolute error (MAE) than
plant-specific data for the process temperatures, and materials and gases the non-linear curve fit, a second-order curve fit was developed as shown
flow into and out of the kiln, pre-calciner and clinker cooler. The model in Eq. (1):
assumptions and projections were within ±5% of the plant-specific data
TEI MC=0,[O2 ]=1% = c1 + c2 × (LHV) + c3 × (O) + c4 × (LHV)
which are not presented here for reasons of confidentiality. The vali­
dated TEF model was then used to predict the changes in the air, energy, × (O) + c5 ×(LHV)2 + c6 × (O)2 (1)
and emissions intensities when AFs were co-fired to replace 50% of GJ
demand in the pre-calciner. Table 2 presents all plant-specific input where, TEIMC=0,[O2 ]=1% is the thermal energy intensity when an AF was
variables and calculated parameters for the reference case and other co-fired at the dried condition (Moisture content of 0%) with 1% [O2] in
examined cases (1 to 4): the flue gas at the pre-calciner exit. The values of constants c1 to c6
were determined by minimizing the sum of square error method using
• Reference case: NG as a fuel keeping 1% [O2] in the flue gas at the the Solver tool in Microsoft Excel in Office 365, version 2019.
exit of pre-calciner – [NG-1] Similarly, to predict the changes in the TEI with the moisture content
• Case 1 (To examine the effect of AF co-firing without incorporating (MC) (>0%) in AFs and the [O2] in flue gas (>1%) at the exit of the pre-
the effect of [O2] in the flue gas): AFs (dry basis) replaced 50% of pre- calciner, more TEF model results generated in cases 3 and 4 were used to
calciner GJ demand keeping 1% [O2] in the flue gas at the exit of the devise both linear (ignoring constant terms c8 , and c10 ) and non-linear
pre-calciner – [NG/AF-1]. equations of MC and [O2] as shown in Eq. (1a). The regression equa­
• Case 2 (To examine the effect of AF co-firing by incorporating the tion with a lower MAE value was then selected:
effect of [O2] in the flue gas): AFs (dry basis) replaced 50% of pre-
TEI = TEIMC=0,[O2 ]=1% + {c7 + c8 × (MC) } × (MC) + {c9 + c10 × ([O2 ]− 1)}
calciner GJ demand keeping 3% [O2] in the flue gas at the exit of
pre-calciner – [NG/AF-3] × ([O2 ]− 1)
• Case 3 (To examine the effect of the moisture content in AFs): (1a)
Keeping 1% [O2] in the flue gas at the exit of the pre-calciner, 50% of
pre-calciner GJ demand was substituted by AFs with the moisture 3. Result and discussions
content (MC) of 10%; 15%, and 20%.
• Case 4 (To examine the effect of the [O2] in the flue gas): 50% of GJ 3.1. Reference case (NG-1)
demand in pre-calciner was substituted by AFs (dry basis) by
changing [O2] in the flue gas at the exit of pre-calciner to 3%, 5% and The new cement plant (Lafargeholcim Exshaw plant retrofitted in
6% [O2]. Repeat Case 3 at each of these new oxygen concentrations 2018) in this study has several features that enhance thermal efficiency,
in the flue gas. including multi-stage preheating, a separate pre-calciner and kiln, and
the ability to use natural gas making it possible to achieve complete
The CO concentration in the flue gas was assumed to be 0.02% (mole combustion with a flue gas [O2] as low as 1%. Under these conditions,
basis) for all cases based on the pre-calciner plant design [34]. This is a the thermal energy flow (TEF) model was developed to assess the per­
reasonable assumption for a plant consuming natural gas with 1% [O2] formance of its clinker-making process. Table 3 presents a summary of
in the flue gas (Reference Case), or for a co-fired plant where AFs have a thermal energy and material flows predicted by the TEF model. The TEF
very small particle size (Cases 1 and 3). With larger particle sizes for the model results for the reference case presented in Table 3 and the flue gas
AFs, a higher flue gas oxygen requirement would be needed to ensure composition summarized in Table S5 (at the exit of the rotary kiln and
complete combustion and a low CO concentration in the flue gas (Cases pre-calciner) was in close agreement with the plant-specific data. A
2 and 4). summary of air supply and excess air in the rotary kiln, the pre-calciner,
and the overall clinker-making process is also presented in Table S6 of
the supplementary file.
The clinker-making process requires 1.58 kg of raw materials (dry)
Table 2 for each kg of clinker (clk), resulting in process emissions of 556 kgCO2/t
Reference Case information and summary of examined variables.
clk (Table 3, R24), a value that is similar to previous reports [35–37].
Parameters Reference Examined The TEF model estimates thermal energy intensity (TEI) of 3293
case range
MJLHV/t clk (Table 3, R1) of which 60% is supplied to the pre-calciner
Input variable and the remaining to the rotary kiln. This TEI is lower than the
[O2] in flue gas at kiln exit, % mole 6.5 Kept constant average energy intensity of 4500 MJLHV/t clk for other cement plants in
[O2] in flue gas at pre-calciner exit, % mole 1.0 1.0 to 6.0
[CO] in flue gas, % mole 0.02 Kept constant
Canada [38]. Of the total thermal energy supplied, 53% (1737 MJ/t clk)
Share of AF in pre-calciner, % GJ basis 100 50 supports the calcining and sintering reactions listed in Table A.1, a value
Share of AF in kiln, % GJ basis 0.0 0.0 that is similar to those reported in other studies [26,39–42]. The flue gas
Moisture in alternative fuels, % wt. 0.0 0.0 to 20.0 losses (FGL) of 17.7% (Table 3, R26) and exhaust vent air losses (EVAL)
LHV and elemental compositions of AF, dry N/A See Table 1
of 7.9% (Table 3, R27) are also within the reported literature values
basis
Calculated parameter [42].
Air demand, Nm3/t clk • TBD • TBD The TEF model also estimates that primary air (PA) supply of 34
Thermal energy intensity, MJ/t clk • TBD • TBD Nm3/t clk (Table 3, R10) is supplemented by secondary air (SA) at 512
Fuel-related combustion emissions, tCO2/t • TBD • TBD Nm3/t clk (Table 3, R11) reflecting the need to support combustion and
clk
maintain 6.5% [O2] in flue gas at the kiln exit. This excess oxygen enters

5
 D.R. Nhuchhen et al.
Table 3
Results of the TEF model: Co-firing AF in pre-calciner keeping 1% [O2] in flue gas at the exit of pre-calciner (Case: 1).
Row Results Units Ref. AF at 50% of pre-calciner’s energy demand with 50% NG
[NG:1] RC HDPE TVBP EPW1 CG EPW2 SC AS WT SOC BIW CH SW WD BM RT1 PW TP RT2 LW CB PO T P

R1 A. Thermal Energy Intensity (TEI) GJLHV/t clk 3.293 3.277 3.272 3.264 3.275 3.302 3.286 3.291 3.336 3.274 3.331 3.310 3.312 3.319 3.323 3.326 3.304 3.310 3.305 3.305 3.299 3.334 3.330 3.355 3.335
R2 NG in Kiln 1.311 1.314 1.311 1.311 1.313 1.316 1.316 1.311 1.334 1.311 1.314 1.312 1.312 1.315 1.319 1.317 1.311 1.312 1.312 1.311 1.311 1.315 1.313 1.311 1.312
R3 NG in PC 1.982 0.982 0.981 0.976 0.981 0.992 0.985 0.990 1.001 0.981 1.008 0.998 1.000 1.002 1.002 1.004 0.996 0.999 0.997 0.997 0.994 1.009 1.008 1.022 1.011
R4 AF in PC 0.000 0.982 0.981 0.976 0.981 0.992 0.985 0.990 1.001 0.981 1.008 0.998 1.000 1.002 1.002 1.004 0.996 0.999 0.997 0.997 0.994 1.009 1.008 1.022 1.011
R5 Total energy demand GJLHV/hr 576 573 573 571 573 578 575 576 584 573 583 579 579 581 582 582 578 579 578 578 577 583 583 587 584
R6 NG in Kiln tonne of fuel/t clk 4.84 4.85 4.84 4.84 4.85 4.86 4.86 4.84 4.92 4.84 4.85 4.84 4.84 4.85 4.87 4.86 4.84 4.84 4.84 4.84 4.84 4.85 4.85 4.84 4.84
R7 NG in PC 7.32 3.63 3.62 3.60 3.62 3.66 3.64 3.66 3.70 3.62 3.72 3.69 3.69 3.70 3.70 3.71 3.68 3.69 3.68 3.68 3.67 3.73 3.72 3.77 3.73
R8 AF in PC 0.00 4.80 3.98 4.54 5.61 8.57 6.88 6.88 13.04 5.24 11.51 8.67 9.34 10.26 10.96 9.65 8.86 9.53 8.61 8.59 7.95 11.52 11.38 12.78 11.17
3
R9 B. Air Demand (AD) Nm /t clk 1591 1691 1649 1686 1677 1721 1666 1679 1740 1609 1675 1618 1648 1652 1651 1510 1649 1656 1609 1609 1590 1601 1611 1559 1531
R10 Primary air 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34
R11 Secondary air 512 513 512 512 513 514 514 512 521 512 513 513 512 514 515 515 512 512 512 512 512 514 513 512 512
R12 Leak air kiln 16 16 16 16 16 16 16 16 17 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16 16
6

R13 Tertiary air 430 386 403 385 392 375 401 388 385 422 394 419 404 406 411 476 403 400 423 422 431 430 424 446 460
R14 Exhaust vent air 614 758 700 754 738 798 717 745 800 641 733 651 696 698 691 485 699 709 639 641 612 622 640 566 525
R15 Total combustion air (TCA) 993 950 965 948 956 940 966 951 957 984 958 983 968 970 977 1041 966 963 986 985 994 995 987 1009 1023
R16 C. Emissions Intensity (EI) kgCO2/t clk 744 753 758 765 766 769 769 769 776 777 782 780 784 786 787 788 787 787 788 789 791 796 797 799 804
R17 EI excluding Bio. C 744 753 758 765 766 769 769 687 758 777 689 688 688 688 689 689 714 736 735 714 688 689 689 689 742
R18 Energy emissions include Bio. C 188 197 203 209 210 213 213 213 220 222 226 225 229 230 231 232 231 231 232 233 235 240 241 243 248
R19 AF emissions-PC 0 66 71 78 79 81 81 82 86 91 93 92 96 97 99 99 99 99 100 101 103 107 108 110 116
R20 Fossil C 0 66 71 78 79 81 81 0 69 91 0 0 0 0 0 0 26 48 47 26 0 0 0 0 53
R21 Biogenic C (Bio.C) 0 0 0 0 0 0 0 82 17 0 93 92 96 97 99 99 73 52 53 74 103 107 108 110 62
R22 NG emissions-PC 113 56 56 56 56 57 56 57 57 56 58 57 57 57 57 58 57 57 57 57 57 58 58 59 58
R23 NG emissions-K 75 75 75 75 75 75 75 75 76 75 75 75 75 75 75 75 75 75 75 75 75 75 75 75 75
R24 Process emissions 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556
R25 D. Waste Heat GJ/t clk 0.841 0.826 0.822 0.815 0.824 0.850 0.834 0.841 0.877 0.823 0.878 0.857 0.860 0.866 0.868 0.869 0.852 0.858 0.853 0.853 0.846 0.879 0.876 0.900 0.880
R26 Flue gas loss (FGL) 0.581 0.505 0.527 0.496 0.512 0.513 0.531 0.526 0.539 0.552 0.568 0.581 0.565 0.570 0.576 0.664 0.557 0.558 0.582 0.582 0.588 0.616 0.606 0.660 0.658
R27 Exhaust vent air loss (EVAL) 0.260 0.321 0.296 0.319 0.312 0.337 0.303 0.315 0.338 0.271 0.310 0.275 0.294 0.295 0.292 0.205 0.295 0.300 0.270 0.271 0.259 0.263 0.271 0.239 0.222

Fuel 295 (2021) 120544

D.R. Nhuchhen et al. Fuel 295 (2021) 120544

the pre-calciner and is supplemented by tertiary air (TA) at 430 Nm3/t 3.2. Alternative fuel scenarios: [NG/AF-1] and [NG/AF-3]
clk (Table 3, R13), resulting in a total combustion air requirement of
993 Nm3/t clk (Table 3, R15), including about 16 Nm3/t clk of leakage The TEF model was used to explore the use of solid AFs to replace
into the kiln system, and 1% [O2] in the flue gas. In addition, there is an 50% of the fuel energy input to the pre-calciner where natural gas
exhaust vent air demand of 614 Nm3/t clk (Table 3, R14) to cool the provided the balance of pre-calciner energy and all of the kiln energy
clinker, resulting in a thermal energy loss of 7.9% (260 MJ/t clk) of the demands. While 24 AFs were studied and compared to the natural gas
total TEI. reference case , three of the AFs (Bolded rows in Table 1) were selected
The CO2 emission intensity depends mainly on direct emissions due for their diversity in biogenic carbon content for a more detailed
to fuel combustion and limestone decompositions, and indirect emission assessment. They were:
from electricity consumption. As this study focuses mainly on the ther­
mal energy requirement, details of indirect CO2 emissions due to elec­ • HDPE: High-density polythene (no biogenic carbon)
tricity consumption are not included here. Henceforth, all emissions in • WD: Wood dust (100% biogenic carbon)
this study will be direct emissions. The combustion of NG in the kiln and • RT2: Railway ties (74% biogenic carbon)
pre-calciner produces 75 kgCO2/t clk (Table 3, R23) and 113 kgCO2/t
clk (Table 3, R22), respectively. Therefore, the total CO2 emissions in­ Typically, complete combustion of solid fuels is not possible without
tensity including process emissions is 744 kgCO2/t clk (Table 3, R16). providing excess air supply that ultimately results in flue gas oxygen

Fig. 3. Effects of AF co-firing at 1% and 3% [O2] in

flue gas on energy flows (I-In & O-Out), air demand
and CO2 emissions. Note that both natural gas (NG)
only cases had a flue gas [O2] of 1%. (Note: AF-F:
Fossil emissions from AF; AF-PC: AF energy input
in PC; Bio: Biogenic emissions; CA: Conveying air;
Dissociation: Dissociation energy input to break­
down limestone; EE: Energy Emissions; EVA:
Exhaust vent air; Flue: Flue gas loss; HDPE: Plastic
Waste; LA: Leak air; NG-K: NG input in kiln and its
emissions; NG-PC: NG input in PC and its emissions;
PA: Primary air; PE: Process Emissions; Other: Other
losses that includes exhaust vent air loss, radiation
and unaccounted losses; RT2: Railway Ties 2; SA:
Secondary air; TA: Tertiary air; WD: Wood dust;
represents the net GHG emissions; GWPBio = 0).

7
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

concentrations of 2% to 7% or more [15]. To explore the importance of with 74% biogenic carbon is projected to reduce GHGs by 1.6% (12
oxygen concentration in the flue gas [O2] on the energy and material kgCO2/t clk) (Fig. 3F). Therefore, for AFs lacking biogenic carbon, the
flows through the clinker-making process, scenarios assuming both 1% cement plant operator would face a trade-off between the cost of AFs and
and 3% in the flue gas were carried out with the TEF model and their the cost of the excess CO2 emissions with respect to the emission
summary for the selected AFs are shown in Fig. 3. benchmark.
1% [O2] in Flue Gas (Case 1). The flows of energy, air, and CO2 for The GHG values presented here (Fig. 3E and F) only consider the
24 AFs and reference natural gas fuels are presented in Table 3. direct CO2 emission for which the cement plant operator is responsible
Compared to the reference case, the TEI of the clinker-making process for the carbon price regulation in Canada. In the reference cement plant,
decreased by 0.6% for HDPE and increased by 0.9% for WD and 0.4% for these include process emissions of 556 kg CO2e/t clk and fuel combusted
RT2 when the respective AFs were co-fired (Fig. 3A). The decrease in TEI related emissions of 188 kg CO2e/t clk. Life cycle emissions would be
with HDPE (assuming complete combustion of the HDPE can be ach­ higher and include:
ieved when there is only 1% [O2] in the flue gas) is due to a reduction in
the total combustion air (TCA) supply from 993 Nm3/t clk to 965 Nm3/t • Upstream emissions associated with the production and pipeline
clk (Table 3, R15), thereby avoiding the need to heat the air to process distribution of NG are estimated to be 18% more compared to its
temperatures. The decreased in TCA supply could be explained, in part, carbon intensity [43]. Therefore, for the reference plant, including
by the higher level of embedded elemental oxygen content in the HDPE life cycle emissions would increase energy emissions by about 34
compared to that in NG (0.9% vs 0%, Table 1). CO2e/t clk.
Similar results for TCA are also observed for woody biomass fuels • For the cement plants using AFs, additional emissions may be asso­
such as WD and RT2, which have significantly higher embedded ciated with the processing and delivery of AFs, but not with their
elemental oxygen content in the range of 32–36% (Table 1). However, production processes since they are all from waste streams. On the
the TEI for WD and RT2 is higher than the reference case since TEI not other hand, there may be emission reductions resulting from the
only depends on embedded elemental oxygen content in the AFs but also diversion of biomass from landfill sites where methane can be
on its lower heating value (LHV) and moisture content. If the LHV of the released due to anaerobic digestion [10].
AF was low, more fuel (Table 3, R7 and R8) would have to be com­ • Indirect CO2 emissions associated with electricity use (about 90
busted to supplement the increase in sensible heat demand and heat of kwh/t clk for the reference clinker-making process) would add 61
evaporation of the moisture in the AF. For example, the WD with low kgCO2e/t clk, given the emission intensity of the grid in Alberta,
LHV of 16.1 MJLHV/kg resulted in higher TEI compared to the RT2 with Canada of 680 kg CO2/MWh [44].
LHV of 20.4 MJLHV/kg (Table 3, R1). • Due to the increase in air demand and the flue gas volume for plants
Since the tertiary air (TA) demand supplying to the pre-calciner is using AF, electricity demand is expected to increase by about 14
drawn from the clinker cooler exhaust, a reduction in that air demand kWh/t clk (16%) over the 90 kWh/t clk in the reference case. This
for three selected AFs results in an increase in the exhaust vent air by would result in an additional 10 kgCO2e/t clk of indirect emissions.
14%, 12% and 4% for HDPE, WD, and RT2, respectively. This, therefore,
increases the respective total air demand (AD) by 3.6%, 3.8%, and 1.2% The Effect of AF on Flue Gas Composition and Flow. Compared to
(Table 3, R9) as shown in Fig. 3C. the reference case plant with 1% [O2], the AFs cases at 3% [O2] in the
Due to the higher carbon intensity of AFs compared to NG (Table 1), flue gas show an increase in the flue gas volumetric flow (15 to 29%),
all AFs co-firing have higher total emissions intensity (kgCO2/t clk), and thermal flue gas loss (49 to 97%) (Table S7). This is primarily due to
ranging from 1.2 to 8.1% (Table 3, R16). However, not all carbon in the the need for excess air when AFs are combusted.
AFs are from fossil fuel sources, therefore the total GHG emission in­ Even though CO2 emissions per tonne clinker are higher when AFs
tensity using an AF like WD, resulted in a 7.5% reduction (56 kgCO2/t are combusted, the increased volumetric flow of the flue gas results in
clk) compared to the NG-based reference plant (Table 3, R17, Fig. 3E). the concentration of CO2 being reduced from about 27% CO2 in the
By comparison, HDPE (0% biogenic C) increases the total GHG emission reference case to 23 to 25% CO2 in the AF cases. However, the thermal
intensity by 1.9% (Fig. 3E). flue gas loss from CO2 increases from 0.20 GJ/t clk in the reference case
As noted previously, complete combustion of AFs can rarely be to 0.29 to 0.36 GJ/t clk in the AF cases.
achieved with a flue gas [O2] of only 1%. Therefore, the effect of
increasing flue gas [O2] to 3% was modeled and compared to the NG- 3.3. Regression for thermal energy intensity (TEI)
based reference plant having only 1% [O2] in the flue gas.
3% [O2] in Flue Gas for AF only (Case 3). To ensure complete The results of the TEF model (Tables 3 and 4) identified four major
combustion of the AFs, 3% flue gas [O2] is assumed, made possible by factors that determine the TEI of the clinker-making process, including
supplying additional air from the conveying air stream to the pre- the embedded elemental oxygen content (O) and lower heating value
calciner (Table 4, R14). This results in more thermal heat loss from (LHV) of AFs, as well as required [O2] in flue gas, and the moisture
the flue gas stream as shown in Fig. 3B. On average, the shift to 3% [O2] content (MC) of AFs. To examine the combined effect of these parame­
in the AF co-firing cases increases the TEI by 443 MJ/t clk (13% more) ters on the TEI and provide a quick assessment of the performance of
compared to the NG-based reference cement plant with 1% [O2] in the clinker-making under AF co-firing cases, a regression equation was
flue gas (Table 4, R1). developed here. The TEF model results in Tables 3 and 4 along with the
To co-fire AFs and maintain 3% [O2] in the flue gas, the clinker- results obtained for cases 3 and 4 were used to generate the regression
making process requires on average of 334 Nm3/t clk (21%) more air, equation making it possible to calculate TEI for a wide range of AFs
compared to the NG-based reference cement plant with 1% [O2] in the without needing to run the full TEF model for the cement plant:
flue gas (Table 4, R9; Fig. 3D).
The requirement to maintain 3% [O2] in the flue gas of plants using TEI = 3488.5 − 11.5 × (LHV) − 182.6 × (O) + 3.2 × (LHV) × (O) + 15.4
AFs, also increases total CO2 emission intensity by on average 69 kgCO2/ × 10− 2 ×(LHV)2 + 232.0 × (O)2 + 3.1 × (MC) + {42.4
t clk (9.3% more) compared to the NG-based reference cement plant
× ([O2 ] − 1) + 131.2} × ([O2 ] − 1)
with 1% [O2] in the flue gas (Table 4, R17). In the case if a carbon­
–neutral AF such as WD, the total GHG intensity was reduced by 5.8% (2)
(43 kgCO2/t clk) compared to the NG-based reference cement plant with As shown in Fig. 4, this regression equation provides a good fit to the
1% O2 in the flue gas. However, non-biogenic waste-derived fuels like TEF-modeled TEI for the 24 AFs with known values for embedded
HDPE could increase GHGs by 5.5% (41 kgCO2/t clk) (Fig. 3F). The RT2 elemental oxygen content (O) and LHV. The mean absolute error (MAE)

8
 D.R. Nhuchhen et al.
Table 4
Results of the TEF model: Co-firing AF in pre-calciner keeping 3% [O2] in flue gas at the exit of pre-calciner (Case: 2).
Row Results Unit Ref. [NG:3] AF at 50% of pre-calciner’s energy demand with 50% NG
[NG:1] RC HDPE TVBP EPW1 CG EPW2 SC AS WT SOC BIW CH SW WD BM RT1 PW TP RT2 LW CB PO T P

R1 A. Thermal Energy Intensity GJLHV/t clk 3.293 3.735 3.669 3.678 3.650 3.671 3.693 3.693 3.691 3.746 3.697 3.752 3.746 3.734 3.745 3.755 3.830 3.722 3.727 3.742 3.742 3.742 3.791 3.779 3.839 3.825
(TEI)
R2 NG in Kiln 1.311 1.311 1.314 1.311 1.311 1.314 1.318 1.317 1.311 1.338 1.311 1.315 1.313 1.312 1.316 1.321 1.318 1.311 1.312 1.312 1.311 1.311 1.316 1.314 1.311 1.312
R3 NG in PC 1.982 2.424 1.177 1.184 1.169 1.179 1.188 1.188 1.190 1.204 1.193 1.219 1.216 1.211 1.214 1.217 1.256 1.205 1.208 1.215 1.215 1.215 1.237 1.232 1.264 1.256
R4 AF in PC 0.000 0.000 1.177 1.184 1.169 1.179 1.188 1.188 1.190 1.204 1.193 1.219 1.216 1.211 1.214 1.217 1.256 1.205 1.208 1.215 1.215 1.215 1.237 1.232 1.264 1.256
R5 Total energy demand GJLHV/hr 576 654 642 644 639 642 646 646 646 655 647 657 655 653 655 657 670 651 652 655 655 655 663 661 672 669
R6 NG in Kiln tonne of fuel/t 4.838 4.838 4.850 4.838 4.839 4.849 4.863 4.862 4.838 4.939 4.838 4.852 4.845 4.842 4.855 4.875 4.866 4.839 4.841 4.842 4.840 4.839 4.856 4.849 4.838 4.842
R7 NG in PC clk 7.32 8.96 4.35 4.37 4.32 4.35 4.38 4.39 4.39 4.44 4.40 4.50 4.49 4.47 4.48 4.49 4.64 4.45 4.46 4.49 4.48 4.48 4.57 4.55 4.66 4.64
R8 AF in PC 0.00 0.00 5.76 4.80 5.44 6.74 10.25 8.30 8.27 15.68 6.37 13.92 10.56 11.31 12.44 13.31 12.07 10.72 11.52 10.49 10.48 9.72 14.12 13.91 15.81 13.88
R9 B. Air Demand (AD) Nm3/t clk 1591 1890 1959 1925 1950 1948 1986 1944 1949 2021 1897 1955 1911 1930 1937 1942 1849 1929 1936 1902 1903 1888 1905 1909 1876 1856
R10 Primary air 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34 34
R11 Secondary air 512 512 513 512 512 513 515 515 512 523 512 514 513 513 514 516 515 512 512 512 512 512 514 513 512 512
R12 Leak air kiln 16 16 16 16 16 16 16 16 16 17 16 16 16 16 16 17 16 16 16 16 16 16 16 16 16 16
9

R13 Tertiary air 430 430 386 403 385 392 375 401 388 385 422 394 419 404 406 411 476 403 400 423 422 431 430 424 446 460
R14 Conveying air 0 298 266 277 264 270 263 276 270 270 288 279 292 282 284 287 335 280 279 293 293 298 302 297 317 324
R15 Exhaust vent air 614 614 759 700 754 739 800 719 745 809 641 734 652 697 700 694 488 699 710 640 641 612 624 641 566 525
R16 Total combustion air (TCA) 993 1292 1216 1242 1213 1226 1203 1242 1221 1229 1273 1237 1275 1250 1254 1265 1377 1247 1243 1279 1278 1292 1297 1285 1326 1347
R17 C. Emissions Intensity (EI) kgCO2/t clk 744 769 777 785 791 793 796 797 797 805 809 813 813 817 819 821 827 820 820 822 823 826 833 834 839 846
R18 EI excluding Bio. C 744 769 777 785 791 793 796 797 699 784 809 701 701 700 701 701 703 731 758 758 732 700 702 701 703 769
R19 Energy emissions include Bio. C 188 214 222 229 235 237 240 241 241 249 253 258 257 261 263 265 271 264 264 266 267 271 277 278 283 290
R20 AF emissions-PC 0 0 79 86 93 95 96 98 98 104 110 113 113 117 118 120 124 120 120 122 123 126 131 132 136 144
R21 Fossil C 0 0 79 86 93 95 96 98 0 83 110 0 0 0 0 0 0 31 58 57 32 0 0 0 0 66
R22 Biogenic C (Bio.C) 0 0 0 0 0 0 0 0 98 21 0 113 113 117 118 120 124 89 62 65 91 126 131 132 136 77
R23 NG emissions-PC 113 139 67 68 67 68 68 68 68 69 68 70 70 69 70 70 72 69 69 70 70 70 71 71 72 72
R24 NG emissions-K 75 75 75 75 75 75 75 75 75 76 75 75 75 75 75 75 75 75 75 75 75 75 75 75 75 75
R25 Process emissions 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556 556
R26 D. Waste Heat GJ/t clk 0.841 1.266 1.203 1.212 1.183 1.204 1.225 1.224 1.224 1.269 1.229 1.282 1.276 1.265 1.274 1.282 1.354 1.253 1.259 1.273 1.272 1.276 1.318 1.307 1.365 1.351
R27 Flue gas loss (FGL) 0.581 1.006 0.882 0.916 0.864 0.892 0.887 0.920 0.910 0.927 0.958 0.971 1.000 0.971 0.979 0.989 1.147 0.958 0.959 1.002 1.001 1.013 1.054 1.036 1.126 1.129
R28 Exhaust vent air loss (EVAL) 0.260 0.260 0.321 0.296 0.319 0.312 0.338 0.304 0.315 0.342 0.271 0.310 0.276 0.295 0.296 0.293 0.207 0.296 0.300 0.270 0.271 0.263 0.264 0.271 0.239 0.222

Fuel 295 (2021) 120544

D.R. Nhuchhen et al. Fuel 295 (2021) 120544

Fig. 4. A curve describing the effect of LHV and

embedded elemental oxygen fraction of alternative
fuel on the modeled specific thermal energy demand
in MJ/t clk (MC in AFs = 0%, [O2] in pre-calciner
exit = 1%; Other waste (blue circle) includes
textile, asphalt shingles and carpets; Local materials
(red star) include locally available waste-drived
alternative fuels); Wood waste (black circle); Plas­
tic/Rubber (white circle); NG (yellow circle)). (For
interpretation of the references to colour in this
figure legend, the reader is referred to the web
version of this article.)

between the TEF modeled TEI and the non-linear curve fit TEI (Eq. (2)) Validation using Local AF Data. To test the regression equation,
is about 5 MJ/t clk, which is less than the MAE of 10 MJ/t clk for the embedded elemental oxygen content (O) and LHV values for ten locally
linear regression equation (Refer to Section S9 and Table S8 of the available waste-derived AFs are both analyzed by the TEF model to
supplementary file). To put these numbers in perspective, the range of generate values for TEI (red stars in Fig. 4), which were then compared
modeled TEI values with various AFs are within 91 MJ/t clk with the 0% with the TEI value generated by the curve fit (Eq. (2)) at 0% MC and 1%
MC and 1% [O2] in the flue gas data presented in Fig. 4. [O2] in the flue gas. The regression equation predicts the TEI for these
The inclusion of non-linear terms of LHV and O also helps to justify local fuels within the MAE of 4.4 MJ/t clk (Table S9). This supports the
the decrease in the TEI for some AFs with high LHV and slightly more premise that Eq. (2) can provide a reasonable estimate for the TEI for all
embedded elemental oxygen content than that in NG (Refer to the dis­ wide range of AFs.
cussion in Section 3.2). For other AFs having a lower LHV, the TEI of Effect of Moisture Content (MC). Most AFs are not completely dry
clinker-making increases thereby forming a concave upward curvature so when they are combusted, the MC of AF would be evaporated,
in Fig. 4. It is noted here that this analysis has assumed the same com­ consuming a portion of the supplied fuel energy thereby increasing the
bustion efficiency for AF and NG at the same level of [O2] in the flue gas. TEI required for clinker production. The energy requirement to process

10
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

the MC increases linearly with the moisture level as shown in Fig. 5, increase the TEI by 62 MJ/t clk of which 31 MJ/t clk (50% GJ basis in
therefore a simple linear curve fit (as it has similar MAE as that of the pre-calciner) would have to be supplied from AFs. This means if wood
non-linear curve fit) was selected in Eq. (2). Note that every 1% increase dust with 20% moisture content is used, there will be a 3% increase in
in the MC in an AF increases the TEI by about 3.1 MJ/t clk. Therefore, in the amount (tonne/t clk) of wood dust supply (Table 3, R4 and R8)
a typical AF with an average 10% MC [45], the TEI increases by 31 MJ/t relative to the dry wood dust supply. Similarly, increasing [O2] in flue
clk, or approximately a 1% increase in the TEI (3293 MJ/t clk) for the gas from 1% to 3% increases the TEI by 432 MJ/t clinker of which 216
reference clinker-making plant. MJ/t clinker is from AFs. This means that the wood dust supply would be
Effect of [O2] in Flue Gas. As noted previously solid AFs are un­ increased by 22% (Table 3, R4 and R8). Therefore, the regression
likely to be fully combusted when the [O2] in flue gas is only 1%, so the equation could help the cement plant operator to design the AF
TEF model also assesses the increase in TEI under conditions where the conveying system and drying system that take into consideration the
flue gas [O2] is greater than 1%. The energy required to maintain the range of moisture, and to estimate the incremental clinker production
same process temperature as the NG-based reference plant increases cost.
non-linearly with the [O2] in the flue gas as illustrated in Fig. 6. To In addition to this, the cement plant operator may also need to trade-
incorporate this non-linear trend with [O2], the non-linear curve fit with off the energy required for grinding AFs and the excess air necessary to
a lower MAE than the linear curve fit was selected in Eq. (2). Increasing ensure complete combustion of larger particles of AFs. For instance, the
[O2], from 1% to 2% in flue gas, increases the TEI by about 174 MJ/t clk specific grinding electricity, though it also depends on many other pa­
(5.3% of the TEI of the reference clinker-making plant) while an increase rameters such as type of grinder and type of AFs, increases from 58 to
from 2% to 3% in flue gas increases by about 258 MJ/t clk (7.8% of the 672 kWh/tonne of dry wood when the particle size reduces from 1000 to
TEI of the reference clinker-making plant). 100 µm [48]. Therefore, having the developed regression equation
Application of the Regression Equation. Given the potential in­ would also help the cement plant operator to identify a cheaper option
crease in thermal energy intensity (TEI), if AFs needed to burn at higher between the grinding energy and thermal energy cost, without running
[O2] in flue gas, there could be several challenges in the clinker-making the complex thermal energy flow model.
plant. For example, higher combustion air demand (Table 4, R16) in the
pre-calciner compared to the reference case could affect the solid 4. Conclusions
retention time. To address this technical challenge, the cement plant
operator would have to find different measures such as an increase in The thermal energy flow (TEF) model developed here, for a cement
turbulence inside the pre-calciner [46] or feeding air at multiple stages plant with a 4-stage cyclone preheater, a vertical cylindrical pre-calciner
[47]. As the pre-calciner unit of the clinker-making plant studied here is and a rotary kiln, presents how the properties of alternative fuels (AFs)
new and tall enough to burn AFs completely (personal communication and oxygen concentration in flue gas affect the TEI, AD and CO2 emis­
with the cement plant operator), this study has excluded the impact of sions when AFs with higher carbon intensities are introduced to replace
change in airflow on the solid retention time. 50% of pre-calciner GJ demand in a natural gas-fired cement plant.
However, changes in the TEI due to AF co-firing could increase the While the TEF model can assess the changes in energy and air flows for
production cost. The regression equation presented here could easily all kind of solid fuels co-firing with NG in the pre-calciner, it is difficult,
predict the changes for a different type of AFs and various [O2] in the time-consuming, and costly to run the TEF model for any new AF.
flue gas, thereby facilitating the cement plant operator in decision- Therefore, a regression equation was also devised to determine the TEI
making. of the cement plant.
For example, the use of AFs with the moisture content of 20% could This study concludes that:

Fig. 5. The calculated effect of moisture content (0%, 10%, 20%) on thermal energy intensity for a range of values for alternative fuel’s LHV and embedded
elemental oxygen fraction in dry basis (db) ([O2] at the exit of pre-calciner = 1%).

11
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

Fig. 6. The calculated effect of [O2] in flue gas at the exit of pre-calciner (1%, 2%, 3%) on thermal energy intensity for a range of values for alternative fuel’s LHV
and embedded elemental oxygen fraction in dry basis (db) MC in AF = 0%).

• Replacing natural gas by AFs without increasing the [O2] in the flue would GHG emissions be reduced. The lower (or even negative) cost for
gas has little impact on thermal energy intensity (TEI) and total air AFs relative to natural gas may provide a justification for their use
demand (AD). However, to ensure complete combustion of solid depending on the carbon price, the biogenic carbon content in AFs, fuel
fuels this option may not be available to the plant operator. costs, electricity cost, and the emission benchmark for the cement plant.
• Replacing natural gas by AFs and raising the [O2] in flue gas from 1%
to 3% increases the average AD by 334 Nm3/t clk and TEI by 443 CRediT authorship contribution statement
MJ/t clk, which are increases of 21% and 13%, respectively relative
to the NG-based reference plant with 1% [O2] in the flue gas. Daya R. Nhuchhen: Conceptualization, Modeling, Software. Song
• While moisture content of AFs is linearly coupled to increases in TEI P. Sit: Validation. David B. Layzell: Conceptualization.
(3.1 MJ/t clk for every 1% rise in AF moisture content), the increase
in the flue gas [O2] required to achieve complete combustion has a
proportionately greater impact on TEI at higher flue gas [O2] Declaration of Competing Interest
concentrations.
• Due to higher carbon intensity (kgCO2/GJ) of AFs compared to the The authors declare that they have no known competing financial
reference case fuel, the total CO2 emissions intensity increases in the interests or personal relationships that could have appeared to influence
range of 1–8% and 4–14% when AFs are co-fired at 1% and 3% [O2] the work reported in this paper.
in the flue gas, respectively.
• The use of carbon–neutral AFs such as wood dust at 3% [O2] in the Acknowledgements
flue gas could avoid GHG emissions by up to 43 kgCO2/t clk,
equivalent to an emission reduction of 65,919 tCO2 per year. The authors thank the LafargeHolcim Team at the Exshaw Cement
• The TEI regression equation validated using local AFs could quickly Plant for their support in providing technical information and Mr.
assess the performance of AF co-firing and then facilitate the cement Rustam Punja from Geocycle Canada Inc. for data on the elemental
plant operators to estimate the incremental cost of the clinker composition of local waste materials in Alberta, Canada. We appreciate
production. the valuable insights and advice provided by Mr. Sam Fujimoto (Process
Environmental Consultant), and Dr. Darko Matovic (Professor, Queen
This study shows that in a natural gas-fired cement plant, the use of University, Kingston). The funding for this work was provided to DBL
AFs is likely to increase thermal energy intensity, air demand, electricity from Alberta Innovates, Emissions Reduction Alberta, Natural Science
requirement and total CO2 emissions. Only if the AFs are biogenic, and Engineering Research Council of Canada and LafargeHolcim Inc.

Appendix A:. Major reactions in clinker-making process

This section provides the major decomposition and mineral formation reactions in the pre-calciner and rotary kiln of the cement plant and a
summary of the standard enthalpy of respective reactions.

12
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

Table A1
Major decomposition and mineral formation reactions and standard enthalpy of reaction 25 ◦ C and atmospheric pressure.
SN Reactions Location Standard enthalpy of reaction (Δh), kJ/mole

R1 CaCO3 →CaO + CO2 Pre-calciner + Kiln 179.17

R2 MgCO3 →MgO + CO2 Pre-calciner + Kiln 100.69
R3 2CaO + SiO2 →2(CaO).SiO2 orC2 S Pre-calciner + Kiln − 127.46
R4 4CaO + Al2 O3 + Fe2 O3 →4(CaO).Al2 O3 .Fe2 O3 orC4 AF Kiln − 40.42
R5 3CaO + Al2 O3 →3(CaO).Al2 O3 orC3 A Kiln 19.46
R6 CaO + 2(CaO).SiO2 →3(CaO).SiO2 orC3 S Kiln 11.92

Appendix B:. Combustion analysis

B.1 Stoichiometric fuel combustion

Generic stoichiometric combustion equation for NG/AF co-firing can be expressed as:
{ }
XNG n1 CH 4 + n2 C2 H6 + n3 C3 H8 + n4 (C4 H10 )iso + n5 (C4 H10 )n + n6 C6 H12 + n7 N2 + n8 CO2 + n9 H2 S + WAF (w1 C + w2 H2 + w3 O2 + w4 N2 + w5 Ar + w6 S
( )
aN aAr aH O (B.1)
+ w7 Cl2 + w8 H2 O) + λs− NG/AF O2 + 2 N2 + Ar + 2 H2 O = m1 CO2 + m2 H2 O + m3 N2 + m4 Ar + m5 SO2 + m6 Cl2
a O2 a O2 aO2

where, ni is the molar fraction of the various molecules in NG (kmole per kmole of NG), wi is the molar fraction of the elemental and moisture
compositions in AF (kmole per kg of AF), mi is the molar flow rate of combustion products (kmole per unit time), XNG is the molar flow rate of NG
(kmole per unit time), WAF is the mass flow rate of AF (kg per unit time), and λs− NG/AF is the amount of oxygen required for the stoichiometric
combustion, which is supplied through the air with compositions:
1kmoleair = aN2 N2 + aO2 O2 + aAr Ar + aH2 O (H2 O) (B.1a)

where, ai is the molar compositions of air (kmole per kmole of air). While the oxygen demand could be estimated by balancing oxygen atoms, the
molar flow rate of flue gas at stoichiometric combustion condition was determined by adding combustion products of each fuel and other gases entered
in the system via air. The mathematical analysis is presented in the supplementary Section 4.1. The simplified stoichiometric molar oxygen demand
and flue gas flow rate can be expressed as:
λs− NG/AF = XNG O2O2 /NG + WAF O2 O2 /AF (B.1b)

where, O2O2 /NG and O2O2 /AF are the stoichiometric oxygen demand for NG and AF, respectively and can be estimated as:
O2O2 /NG = (2n1 + 3.5n2 + 5n3 + 6.5n4 + 6.5n5 + 9n6 + 1.5n9 ) (B.1c)

O2O2 /AF = (w1 + 0.5w2 − w3 + w6 ) (B.1d)

The molar flow rate of flue gas from the stoichiometric combustion of natural gas and alternative fuel (mt− s− NG/AF ) can be estimated as:
mt− s− NG/AF = XNG CNG + WAF CAF + λs− NG/AF Cair (B.1e)

where, CNG , CAF and Cair are the constants associated with the properties of NG, AF and air, respectively and can be estimated as:
CNG = (3n1 + 5n2 + 7n3 + 9n4 + 9n5 + 12n6 + n7 + n8 + 2n9 ) (B.1f)

CAF = (w1 + w2 + w4 + w5 + w6 + w7 + w8 ) (B.1g)

( )
aH2 O aN2 aAr
Cair = + + (B.1h)
aO2 aO2 aO2

B.2 Combustion in Kiln

Combustion of a fuel occurs at higher air supply than the stoichiometric air demand since the kiln passes the flue gas with 6.5% [O2] onto the pre-
calciner (Fig. 1). Therefore, the combustion equation in the rotary kiln, which could burn NG or AF along with the supplied air, can be expressed as:
{ }
XNG− K n1 CH 4 + n2 C2 H6 + n3 C3 H8 + n4 (C4 H10 )iso + n5 (C4 H10 )n + n6 C6 H12 + n7 N2 + n8 CO2 + n9 H2 S + WAF− K (w1 C + w2 H2 + w3 O2 + w4 N2 + w5 Ar + w6 S
( )
aN aAr aH O
+ w7 Cl2 + w8 H2 O) + CUB− C− K− i + λa− NG/AF− K (1 + Llk− K ) O2 + 2 N2 + Ar + 2 H2 O + ncal− K CO2 + λs− NG/AF Oenrich− K O2 = m1 CO2
a O2 a O2 aO2
+ m2 H2 O + m3 N2 + m4 Ar + m5 SO2 + m6 Cl2 + m7 CO + m8 O2
(B.2)

13
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

CUB− C− K− i is the amount of unburned solid carbon (kmole per unit time) entering in the kiln along with the calcined raw meal collected by the the
bottom cyclone separator (Cy: V, see Figure S1). Llk− K is the fraction of oxygen entering into the kiln in the form of a leakage air and is assumed to be
proportional to the actual oxygen demand. ncal− K is the amount of CO2 (kmole per unit time) released from the calcination reaction in the kiln. Oenrich− K
is the term introduced in the model to examine oxygen-enriched combustion in the kiln, which is expressed as a fraction of stoichiometric oxygen
requirement for fuel combustion. However, this study excludes cases of unburned solid carbon and oxygen enrichment. The mathematical analysis is
presented in the supplementary Section 4.2. The simplified expression for the actual oxygen demand in the rotary kiln (λa− NG/AF− K ) can be expressed
as:
{ }
1 ( )
λa− NG/AF− K = XNG− K O2 O2 /NG + WAF− K O2 O2 /AF (1 − Oenrich− K ) + CUB− C− K− i − 0.5COK mt− a− NG/AF− K + O2K mt− a− NG/AF− K (B.2a)
1 + Llk− K

where, COK and O2K are concentrations of CO and O2 in the flue gas leaving rotary kiln, respectively and are considered as input model parameters.
The actual molar flow rate of the total flue gas (mt− a− NG/AF− K ) from the rotary kiln can be evaluated as:
1 { }
mt− a− NG/AF− K = XNG− K CNG + WAF− K CAF + CUB− C− K− i + λa− NG/AF− K (1 + Llk− K )Cair + ncal− K (B.2b)
(1 − O2K )

B.3 Combustion in pre-calciner

The actual combustion equation in pre-calciner, which could burn NG or AF along with air supplied and the kiln flue gas, can be expressed as:
{ }
XNG− PC n1 CH 4 + n2 C2 H6 + n3 C3 H8 + n4 (C4 H10 )iso + n5 (C4 H10 )n + n6 C6 H12 + n7 N2 + n8 CO2 + n9 H2 S + WAF− PC (w1 C + w2 H2 + w3 O2 + w4 N2 + w5 Ar + w6 S
( )
aN aAr aH O
+ w7 Cl2 + w8 H2 O) + λa− NG/AF− PC (1 + Llk− PC ) O2 + 2 N2 + Ar + 2 H2 O + nfgk− CO2 CO2 + nfgk− H2 O H2 O + nfgk− CO CO + nfgk− O2 O2 + nfgk− N2 N2
aO2 aO2 aO2
+ nfgk− Ar Ar + nfgk− SO2 SO2 + nfgk− Cl2 Cl2 + ncal− PC CO2 + λs− NG/AF Oenrich− PC O2 = m1 CO2 + m2 H2 O + m3 N2 + m4 Ar + m5 SO2 + m6 Cl2 + m7 CO + m8 O2
+ m9 C
(B.3)

where, nfgk− i is the molar flow rate of gases species in flue gas from the kiln. Llk− PC is the fraction of oxygen entering into the pre-calciner in the form of
a leakage air. ncal− PC is the amount of CO2 (kmole per unit time) released from the calcination reaction in the pre-calciner. Oenrich− PC is the term
introduced in the model to examine oxygen-enriched combustion in the pre-calciner. The mathematical analysis is presented in the supplementary
Section 4.3. The simplified expression for the actual oxygen demand in the pre-calciner (λa− NG/AF− PC ) can be expressed as:
{ }
1 ( )
λa− NG/AF− PC = XNG− PC O2O2 /NG + WAF− PC O2O2 /AF (1 − Oenrich− PC ) + 0.5nfgk− CO − 0.5COPC mt− a− NG/AF− PC − m9 + O2 PC mt− a− NG/AF− PC − nfgk− O2
1 + Llk− PC
(B.3a)

where, COPC and O2PC are concentrations of CO and O2 in the flue gas leaving the pre-calciner, respectively and are considered as the model input
parameters. The actual molar flow rate of the total flue gas leaving the pre-calciner (mt− a− NG/AF− PC ) can be estimated as:
1 { }
mt− a− NG/AF− PC = XNG− PC CNG +WAF− PC CAF +λa− NG/AF− PC (1+Llk− PC )Cair +nfgk− CO2 +nfgk− CO +nfgk− H2 O +nfgk− N2 +nfgk− Ar +nfgk− SO2 +nfgk− Cl2 +ncal− PC − m9
1− O2 PC
(B.3b)

Appendix C:. Expressions for fuel flow rate in the kiln and pre-calciner

C.1 Rotary Kiln

The energy balance in the rotary kiln would help to determine the thermal energy intensity (QF− K ) and corresponding fuel (F) demands
(mNG− K or WAF− K ). Assuming there is no alternative fuel combustion, the total mass flow rate of NG in the rotary kiln (mNG− K ) can be determined as:
QF− K
mNG− K = ( ) (C.1)
Cp− NG− K TNG− K− i − Tref + LHV NG βNG− K

where, LHV NG is the lower heating value of NG, βNG− K is the carbon conversion factor of NG in the kiln, Cp− NG− K is the specific heat capacity of NG at
kiln’s operating condition, TNG− K− i is the temperature of NG at the inlet the kiln (assumed 25 ◦ C in this study), and Tref is the reference temperature
condition.

C.2 Pre-calciner

In the pre-calciner, the thermal energy demand (QF− PC ) is supported by combustion of AF and NG, so we can determine the mass flow rate of AF
(WAF− PC ) and NG (mNG− PC ) as:

14
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

θAF− PC QF− PC
WAF− PC = ( ) ( ) (C.2a)
(1 − Mw− AF )Cp− AF− PC TAF− PC− i − Tref + Mw− AF hAF− PC− i − href + LHV AF βAF− PC

(1 − θAF− PC )QF− PC
mNG− PC = ( ) (C.2b)
Cp− NG− PC TNG− PC− i − Tref + LHV NG βNG− PC

where, θAF− PC is the fraction of thermal energy demand in the pre-calciner supported by the combustion of alternative fuels. βNG− PC and βAF− PC are the
carbon conversion factor of NG and AF in the pre-calciner, respectively. TNG− PC− i and TAF− PC− i are the temperature of NG and AF, respectively at the
inlet of the pre-calciner (assumed 25 ◦ C in this study). Mw− AF is the amount of moisture in AF, hAF− PC− i is the mass enthalpy of moisture at the inlet
temperature and pressure conditions and href is the enthalpy of moisture at the reference conditions.

Appendix D:. Shomate equation

The temperature-dependent Shomate equation [28] can be expressed as:

Cp = A + Bt + Ct2 + Dt3 + E/t2 (D.1)

where, Cp is the specific heat capacity (J/mol K), t is the temperature (K)/1000, and A to E are the constants [29] specific to each gas molecules and
are presented in Table S3 in the supplementary file.

Appendix E. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.fuel.2021.120544.

References [16] Atsonios K, Grammelis P, Antiohos SK, Nikolopoulos N, Kakaras E. Integration of

calcium looping technology in existing cement plant for CO2 capture: Process
modeling and technical considerations. Fuel 2015;153:210–23. https://doi.org/
[1] Andrew RM. Global CO2 emissions from cement production. Earth Syst Sci Data
10.1016/j.fuel.2015.02.084.
2018;10(1):195–217. https://doi.org/10.5194/essd-10-195-2018.
[17] Government of Alberta. Technology Innovation and Emissions Reduction
[2] IEA. Technology Roadmap - Low-Carbon Transition in the Cement Industry.
Regulation. Alberta Queen’s Printer, 2019. Retrieved online at http://www.qp.
International Energy Agency, 2018. Retrieved online at https://www.iea.org/
alberta.ca/1266.cfm?page=2019_133.cfm&leg_type=Regs&isbncln.
publications/freepublications/publication/
[18] Government of Canada. A Healthy Environment and A Healthy Economy: Canada’s
TechnologyRoadmapLowCarbonTransitionintheCementIndustry.pdf.
Strengthened Climate Plan to Create Jobs and Support People, Communities and
[3] de Pee A, Pinner D, Reolofsen O, Somers K, Eveline S, Witteveen, M.
the Planet. Environment and Climate Change Canada, 2020. Retrieved online at
Decarbonization of Industrial Sectors: The Next Frontier. McKinsey & Company,
https://www.canada.ca/content/dam/eccc/documents/pdf/climate-change/
2018. Retrieved online at https://www.mckinsey.com/industries/oil-and-gas/our-
climate-plan/healthy_environment_healthy_economy_plan.pdf.
insights/decarbonization-of-industrial-sectors-the-next-frontier.
[19] LafargeHolcim. The 2030 Plan - Building for Tomorrow. LafargeHolcim, 2016.
[4] Pardo N, Moya JA, Mercier A. Prospective on the energy efficiency and CO2
Retrieved online at https://www.lafargeholcim.com/2030-plan.
emissions in the EU cement industry. Energy 2011;36(5):3244–54. https://doi.org/
[20] ECN. Phyllis2 - Database for the physico-chemical composition of (treated)
10.1016/j.energy.2011.03.016.
lignocellulosic biomass, micro- and macroalgae, various feedstocks for biogas
[5] Naqi A, Jang J. Recent progress in green cement technology utilizing low-carbon
production and biochar. Energy Research Center of the Netherlands, 2019.
emission fuels and raw materials: A review. Sustainability 2019;11:537–55.
Retrieved online at https://phyllis.nl/.
https://doi.org/10.3390/su11020537.
[21] Bergman R, Puettmann M, Taylor A, Skog KE. The carbon impacts of wood
[6] Ariyaratne WKH, Melaaen MC, Tokheim LA. The effect of alternative fuel
products. Forest Product Journals 2014; 64: 220–231. 10.13073/FPJ-D-14-00047.
combustion in the cement kiln main burner on production capacity and
[22] Gorski J, Israel B. Life Cycle Assessment of Low-Carbon Fuels: Lafarge Exshaw
improvement with oxygen enrichment. Int J Mater Metall Eng, 2013;7(4):185–91.
Cement Plant Kiln 6-Greenhouse Gas Emissions Life Cycle Assessment (version A).
Retrieved online at https://publications.waset.org/13206/the-effect-of-
The Pembina Institute, 2018. Retrieved online at https://lafargeexshaw.ca/wp-
alternative-fuel-combustion-in-the-cement-kiln-main-burner-on-production-
content/uploads/2019/01/2018-Lafarge-LCA-Final-Report-GHGs-rA.pdf.
capacity-and-improvement-with-oxygen-enrichment.
[23] Larsen AW, Fuglsang K, Pedersen NH, Fellner J, Rechberger H, Astrup T. Biogenic
[7] Bourtsalas AC, Zhang J, Castaldi MJ, Themelis NJ. Use of non-recycled plastics and
carbon in combustible waste: Waste composition, variability and measurement
paper as alternative fuel in cement production. J Cleaner Prod 2018;181:8–16.
uncertainty. Waste Management & Research: The Journal for Sustainable Circular
https://doi.org/10.1016/j.jclepro.2018.01.214.
Economy 2013;31(10):56–66. https://doi.org/10.1177/0734242X13502387.
[8] Ngadi Z, Lahlaouti ML. Impact of using alternative fuels on cement rotary kilns:
[24] Benhelal E, Rafiei A. Overview of process modeling software: Utilizing alternative
Experimental study and modeling. Procedia Eng 2017;181:777–84. https://doi.
fuels in cement plant for air pollution reduction. Energy Science and Technology
org/10.1016/j.proeng.2017.02.465.
2012; 4(1): 10-18. Retrieved online at http://citeseerx.ist.psu.edu/viewdoc/
[9] Brunner PH, Rechberger H. Waste to energy – key element for sustainable waste
download?doi=10.1.1.826.4946&rep=rep1&type=pdf.
management. Waste Manage 2015;37:3–12. https://doi.org/10.1016/j.
[25] Gao T, Shen L, Shen M, Liu L, Chen F. Analysis of material flow and consumption in
wasman.2014.02.003.
cement production process. J Cleaner Prod 2016;112(1):553–65. https://doi.org/
[10] Cherubini F, Bargigli S, Ulgiati S. Life cycle assessment (LCA) of waste management
10.1016/j.jclepro.2015.08.054.
strategies: Landfilling, sorting plant and incineration. Energy 2009;34(12):
[26] Kabir G, Abubakar AI, El-Nafaty UA. Energy audit and conservation opportunities
2116–23. https://doi.org/10.1016/j.energy.2008.08.023.
for pyroprocessing unit of a typical dry process cement plant. Energy 2010;35(3):
[11] Tsiliyannis CA. Alternative fuels in cement manufacturing: Modeling for process
1237–43. https://doi.org/10.1016/j.energy.2009.11.003.
optimization under direct and compound operation. Fuel 2012;99:20–39. https://
[27] Politecnico di Milano. Design and Performance of CEMCAO Cement Plant Without
doi.org/10.1016/j.fuel.2012.03.036.
CO2 Capture. CEMCAP, 2016. Retrieved online at https://zenodo.org/record/
[12] Rahman A, Rasul M, Khan MMK, Sharma S. Assessment of energy performance and
1001664#.XFTPj1xKiUk.
emission control using alternative fuels in cement industry through a process
[28] Linstrom PJ. A Guide to the NIST Chemistry WebBook. National Institute of
model. Energies 2017; 10: 1996-1-1996-17. 10.3390/en10121996.
Standards and Technology, 2016. https://webbook.nist.gov/chemistry/guide/.
[13] Tsiliyannis CA. Cement manufacturing using alternative fuels: Enhanced
[29] Chase MW. NIST-JANAF Thermochemical Tables, Fourth Edition. Journal of
productivity and environmental compliance via oxygen enrichment. Energy 2016;
Physical and Chemical Reference Data, Monograph 1998; 9: 1–1951.
113:1202–18. https://doi.org/10.1016/j.energy.2016.07.082.
[30] Matschei T, Lothenbach B, Glasser FP. Thermodynamic properties of Portland
[14] Akhtar S, Ervin E, Raza S, Abbas T. From coal to natural gas: Its impact on kiln
cement hydrates in the system CaO–Al2O3–SiO2–CaSO4–CaCO3–H2O. Cem Concr
production, clinker quality and emissions. IEEE Trans Industry Appl 2016;52(2):
Res 2007;37(10):1379–410. https://doi.org/10.1016/j.cemconres.2007.06.002.
1913–24. https://doi.org/10.1109/TIA.2015.2504554.
[31] Green RH, Perry DW. Perry’s Chemical Engineers’ Handbook. eighth ed. USA:
[15] Price L, Sinton J, Worrell E, Phylipsen D, Xiulian H, Ji L. Energy use and carbon
McGraw-Hill Companies Inc.; 2008.
dioxide emissions from steel production in China. Energy 2002;27(5):429–46.
[32] Lafarge Exshaw. Reports | Lafarge Exshaw: Lafarge Low Carbon Fuel Projects.
https://doi.org/10.1016/S0360-5442(01)00095-0.
LafargeHolcim Inc, 2019. Retrieved online at https://lafargeexshaw.ca/reports/.

15
D.R. Nhuchhen et al. Fuel 295 (2021) 120544

[33] CCME. National Emission Guidelines for Cement Kilns. Canadian Council of [41] Kolip A, Savas AF. Energy and exergy analyses of a parallel flow, four-stage cyclone
Ministers of the Environment, Winnipeg, Canada, 1998. Retrieved online at precalciner type cement plant. Int J Phys Sci 2010;5(7):1147–63. Retrieved online
https://www.ccme.ca/files/Resources/air/emissions/pn_1284_e.pdf. at https://academicjournals.org/journal/IJPS/article-full-text-pdf/
[34] KHD. Burning Technology: PYROCLON Calciner. KHD, 2012. Retrieved online at FB76B7828727.
https://www.khd.com/tl_files/downloads/products/burning-technology/kiln- [42] Summerbell DL, Barlow CY, Cullen JM. Potential reduction of carbon emissions by
system/calcinersystems/khd_burningtechnology_pyroclon.pdf. performance improvement: A cement industry case study. J Cleaner Prod 2016;
[35] De Lena E, Spinelli M, Gatti M, Scaccabarozzi R, Campanari S, Consonni S, et al. 135:1327–39. https://doi.org/10.1016/j.jclepro.2016.06.155.
Techno-economic analysis of calcium looping processes for low CO2 emission [43] Nie Y, Zhang S, Liu RE, Roda-Stuart DJ, Ravikumar AP, Bradley A, et al.
cement plants. Int J Greenhouse Gas Control 2019;82:244–60. https://doi.org/ Greenhouse-gas emissions of Canadian liquefied natural gas for use in China:
10.1016/j.ijggc.2019.01.005. Comparison and synthesis of three independent life cycle assessments. J Cleaner
[36] Gibbs MJ, Soyka P, Conneely D. Good Practice Guidance and Uncertainty Prod 2020;258:120701–12. https://doi.org/10.1016/j.jclepro.2020.120701.
Managment in National Greenhouse Gas Inventories: CO2 Emissions from Cement [44] ECCC. National Inventory Report 1990 –2018: Greenhouse Gas Sources and Sinks
Production. Intergovernmental Panel on Climate Change 2001. Retrieved online at in Canada. Environment and Climate Change Canada, Canada, 2020. Retrieved
https://www.ipcc-nggip.iges.or.jp/public/gp/bgp/3_1_Cement_Production.pdf. online at http://www.publications.gc.ca/site/eng/9.506002/publication.html.
[37] Worrell E, Price L, Martin N, Hendriks C, Meida LO. Carbon dioxide emissions from [45] Mokrzycki E, Uliasz- Bocheńczyk A. Alternative fuels for the cement industry. Appl
the global cement industry. Annu Rev Energy Env 2001;26(1):303–29. https://doi. Energy 2003;74:95–100. https://doi.org/10.1016/S0306-2619(02)00135-6.
org/10.1146/annurev.energy.26.1.303. [46] Fande MA, Joshi MPV, Gabhane MS, Sahu MMK. CFD investigation of combustion
[38] NRC. Canadian Cement Industry Energy Benchmarking: Summary Report. Natural in 660MW tangential fired boiler. Int J Eng Res Tech 2014;3(1):892-908. Retrieved
Resources Canada, Canadian Industry Program for Energy Conservation (CIPEC), online at https://www.ijert.org/research/cfd-investigation-of-combustion-in-
2009. Retrieved online at https://www.nrcan.gc.ca/sites/www.nrcan.gc.ca/files/ 660mw-tangential-fired-boiler-IJERTV3IS10356.pdf.
oee/pdf/Publications/industrial/cement-eng.pdf. [47] Nosek R, Holubčík M, Papučík S. Emission controls using different temperatures of
[39] Farag LM. Energy and exergy analysis of Egyptian cement kiln plant with complete combustion air. The Scientific World Journal 2014; 2014: Article ID 487549, 6
kiln gas diversion through bypass. Int J Adv Appl Sci 2012;1(2):35–44. Retrieved pages, 2014. https:/doi.org/10.1155/2014/487549.
online at https://www.semanticscholar.org/paper/Energy-and-Exergy-Analyses- [48] Wang J, Gao J, Brandt KL, Wolcott MP. Energy consumption of two-stage fine
of-Egyptian-Cement-Kiln-Farag/ac02be9b37f84cc7229f48270fe99b19396cf546. grinding of Douglas-fir wood. J Wood Sci 2018;64(4):338–46. https://doi.org/
[40] Khurana S, Banerjee R, Gaitonde U. Energy balance and cogeneration for a cement 10.1007/s10086-018-1712-1.
plant. Appl Therm Eng 2002;22(5):485–94. https://doi.org/10.1016/S1359-4311
(01)00128-4.

16