Energy Consumption and Greenhouse Gas Emissions During Ferromolybdenum Production

Abstract

Molybdenum is mainly used as an alloy material in the iron and steel industry and typically in the form of ferromolybdenum (FeMo). The current study aims to evaluate the energy consumption and greenhouse gas emissions (GHG) of four ferromolybdenum production cases using inventory inputs from a process model based on mass and energy conservations. The total energy required for producing 1 tonne of FeMo can vary between 29.1 GJ/t FeMo and 188.6 GJ/t FeMo. Furthermore, the corresponding GHG emissions differ from 3.16 tCO2-eq/t FeMo to 14.79 tCO2-eq/t FeMo. The main variances are from the mining and beneficiation stages. The differences in these stages come from the beneficiation degree (ore grade) and the mine type (i.e., co-product from copper mining). Furthermore, the mine type has a larger impact on the total energy consumption and GHG emissions than the beneficiation degree. More specifically, FeMo produced as co-product from copper mining has a lower environmental impact measured as the energy consumption and GHG emission among all the four cases. The inventory, consumed energy or associated GHG emission is independent on the initial ore grade and mine type in the downstream production stages such as roasting and smelting. Also, transport has the least impact on the energy consumption and GHG emission among all production stages.

Full text

Vol.:(0123456789)
1 3
Journal of Sustainable Metallurgy
https://doi.org/10.1007/s40831-019-00260-8
RESEARCH ARTICLE
Energy Consumption andGreenhouse Gas Emissions During
Ferromolybdenum Production
WenjingWei1,2 · PeterB.Samuelsson1· AndersTilliander1· RutgerGyllenram1,2· PärG.Jönsson1
© The Author(s) 2020
Abstract
Molybdenum is mainly used as an alloy material in the iron and steel industry and typically in the form of ferromolybdenum
(FeMo). The current study aims to evaluate the energy consumption and greenhouse gas emissions (GHG) of four ferromo-
lybdenum production cases using inventory inputs from a process model based on mass and energy conservations. The total
energy required for producing 1tonne of FeMo can vary between 29.1GJ/t FeMo and 188.6GJ/t FeMo. Furthermore, the
corresponding GHG emissions differ from 3.16 tCO2-eq/t FeMo to 14.79 tCO2-eq/t FeMo. The main variances are from the
mining and beneficiation stages. The differences in these stages come from the beneficiation degree (ore grade) and the mine
type (i.e., co-product from copper mining). Furthermore, the mine type has a larger impact on the total energy consumption
and GHG emissions than the beneficiation degree. More specifically, FeMo produced as co-product from copper mining has
a lower environmental impact measured as the energy consumption and GHG emission among all the four cases. The inven-
tory, consumed energy or associated GHG emission is independent on the initial ore grade and mine type in the downstream
production stages such as roasting and smelting. Also, transport has the least impact on the energy consumption and GHG
emission among all production stages.
Keywords Ferromolybdenum· Energy consumption· Greenhouse gas emission· Material balance· Energy balance
Introduction
Molybdenum is widely used as an alloy material in the iron
and steel industry, and in particular in the stainless steel
industry, in the form of ferromolybdenum (FeMo). Alloying
with molybdenum contributes to a better corrosion resist-
ance of iron and steel products. With increasing global atten-
tion to the climate change issue, it is assumed that molybde-
num producers are facing a need for lowering the product’s
environmental impacts, such as the greenhouse gas (GHG)
emission [1, 2] to improve their sustainable profile in the
alloy market.
Cradle-to-gate life cycle assessment (LCA) [3] is a sys-
tematic tool developed for assessing the environmental
aspects associated with a product from resource extrac-
tion to the factory gate. Life cycle inventory (LCI) [4] is
one of the execution steps in the LCA to account for the
energy and resource flow within a defined boundary. To
date, limited LCA and LCI studies have been published
for molybdenum. The International Molybdenum Associa-
tion (IMOA) completed a life cycle inventory [1] for three
molybdenum-containing metallurgical products (roasted
molybdenite concentrates, ferromolybdenum, and techni-
cal molybdic oxide briquette). IMOA’s LCI study regards
each production stage as a ‘black box’ with input and output
datasets from several production sites, which cover 30% (if
not specified, the default value is given as weight percent)
of the world’s total molybdenum production. Therefore,
the consequent resource usage and environmental impacts
results are reported as an industrial average level. For down-
stream customers like stainless steel manufacturers, lack of
the plant-specific data on the molybdenum product results in
The contributing editor for this article was Veena Sahajwalla.
Electronic supplementary material The online version of this
article (https ://doi.org/10.1007/s4083 1-019-00260 -8) contains
supplementary material, which is available to authorized users.
* Wenjing Wei
wenjingw@kth.se
1 Department ofMaterials Science andEngineering, Royal
Institute ofTechnology, Brinellvägen 23, 10044Stockholm,
Sweden
2 Kobolde & Partners AB, Stockholm, Sweden

Journal of Sustainable Metallurgy
1 3
difficulties when selecting a raw material supplier for their
production processes, as well as to make life cycle assess-
ment for their products.
Production inventory data from a specific plant or mine
is usually not accessible for external parties. Therefore,
a process model based on mass and energy balances was
employed to resolve the issue and provides an alternative to
estimate the plant inventory data. In comparison to the previ-
ously mentioned ‘black box’ model [1], the process model
can be used as the reference source of LCI. Such process
modeling efforts have been reported for several pyrometal-
lurgical processes such as the metallothermic smelting of
ferromolybdenum [5]; production of MoO2 from molyb-
denite using the looping-sulfide oxidation process [6]; iron-
making in a blast furnace [7]; steelmaking in an electric arc
furnace [8–11]; lime production in a kiln [12]; ferronickel
alloy processing in rotary kiln-electric furnace [13] and
MgO/Mg(OH)2 production [14]. The previously mentioned
material and energy balance models mainly evaluates the
energy consumption, material recovery, and GHG emission.
The focus of the present study is to determine the energy
consumption and GHG emission of ferromolybdenum pro-
duced by pyrometallurgical processes and to estimate the
contribution to the GHG emissions from the various process
stages. Beside developing a base for estimating the LCI of
different production routes, the results will also create an
understanding on which GHG emission can be influenced
through technical and organizational development, change
of technology, etc. In doing this, a process calculation model
was used to estimate the plant-specific process data and to
compare the FeMo product with respect to the differences
in beneficiation degree (the ratio of concentrate grade to
raw ore grade), mine type, and required processing stages.
Production Process ofFerromolybdenum
Mining andBeneciation
Molybdenum exists as different minerals, such as molyb-
denite (MoS2), wulfenite (PbMoO4), and powellite
(CaMoO4). At present, the main commercial source is
molybdenite which commonly is found and extracted in
China, the United States, and Chile [15]. Molybdenum min-
erals are mined in both open pit and underground mines. The
molybdenum ore grade from primary mine is in general low,
containing approximately 0.1 to 0.3% Mo. The molybdenum
ore derived as a co-product from copper mine has an even
lower grade, which contains 0.01 to 0.05% Mo [16].
The mining stage can be further divided into two parts:
extraction and material handling. The first part extraction
aims to expose the ore from the overburden and it involves
several activities such as drilling, blasting, digging, and
ventilation. The extraction process may require fuel and
electricity running equipment for instance drills, shovels,
motors, pumps, haul trucks, etc. The second part material
handling includes the transport of the extracted ore to the
ore processing site and the transport of the waste from the
mine to the disposal site. Material handling part may require
diesel fuel powered mobile equipment (haul truck, loader,
dozer) and electric equipment (load-haul-dumper, hoist, con-
veyors, and pump).
The beneficiation stage is arranged in different flowsheets
including crushing, grinding, and floatation. Molybdenite is
released from gangue and reduced into coarse particles by
using crushers. Through the grinding process, the ore mate-
rial is further reduced into fine particles by a grinder or a
mill. In the following floatation step, the ground ore powder
is mixed with certain surface-active chemicals to make the
valuable metallic minerals rise with the air bubbles to the
water surface and thus separates ore from the gangue. The
typical concentrate after the beneficiation process results in
a product containing 45–60% Mo [17], while the rest is made
up of impurities such as silica, iron, and copper.
Molybdenum is extracted not only from primary min-
ing, but also as a co-product from copper mining. About
50% of the world’s molybdenum production is based on a
co-product from copper mining [18]. During the copper ben-
eficiation process, liquid chemicals are added to aid float-
ing molybdenite, which is then separated from the copper
containing minerals. This flotation step is repeated several
times, where the molybdenite subsequently goes into a sepa-
rate floatation cycle for further concentration. Sometimes, an
acid leaching step is required to remove the impurities, such
as copper, to raise the purity of molybdenite concentrate.
Roasting
Roasting is performed in a multiple hearth, vertical furnace
with the purpose to remove sulfur from the mineral. The
roasting furnace is typically designed with 8 levels and run-
ning at an operating temperature not higher than 750°C
[17]. The furnace temperature should be strictly controlled
to minimize molybdenum losses due to sublimation. The
injected air flows upwards from the bottom of the roaster,
while the concentrate enters the top of the roaster in a coun-
ter current flow. Initially, oil and moisture are removed from
the concentrate. Then, the main roasting reactions occur in
the subsequent hearths at a temperature of 600–650°C [16].
As the roasting reaction is exothermic (Table1-Eq.2), the
reaction heat can provide a sufficient heat to reach the work-
ing temperature in the furnace, and excess heat may require
large amounts of cooling air to maintain a proper furnace
temperature [16, 17].
After a completed roasting process, the molybdenum
concentrate (MoS2) is converted into a roasted molybdenite

Journal of Sustainable Metallurgy
1 3
concentrate (RMC), or technical grade molybdenum oxide,
consisting more than 50% Mo, where 90% molybdenum is
present in the form of MoO3 and the rest is in the form of
MoO2 [17]. Most of the sulfur in the raw ore is converted
into SO2 gas which can be utilized to produce sulfuric acid
as a co-product.
Smelting
In the production of FeMo, the RMC is processed in an
electric furnace at high temperatures, between 1760 and
2100°C depending on the smelting time [17]. The RMC
is mixed with iron oxide and slag formers such as lime and
fluorspar. Thereafter, it is melted and subsequently reduced
by reduction agents such as carbon, silicon, aluminum in a
metallothermic process. However, most of the FeMo is pro-
duced by using aluminum and silicon as the reducing agents,
due to the requirement of low carbon (C < 0.5%) in the final
product [17]. In some case, carbon is used as the reduction
agent to lower the production cost, which results in a prod-
uct containing maximum 2% carbon [16]. The molybdenum
content in the FeMo may vary in the range of 55 to 70% [16]
and may contain impurities such as Si, Cu, P, S, and Pb. The
slag consists of 60–70% SiO2, 9–13% Al2O3, 7–11% FeO,
and 6–9% CaO [17]. The chemical reactions that occur in the
electric furnace are given in Eqs.3–6 in Table1.
Methodology
Analysis ofEnergy Consumption andGreenhouse
Gas Emissions
In order to evaluate ferromolybdenum’s energy consump-
tion and greenhouse gas emissions, the difference in ben-
eficiation degrees (the ratio of concentrate grade to raw ore
grade), mine types, and processing stages are studied. A
model based on mass and energy balance is used to provide
the inventory analysis for the impact evaluation. Figure1
illustrates the boundary of the model system for a FeMo
production. It starts from extraction of molybdenum ore,
through ore processing and roasting to the manufacturing
of FeMo. The model is also taking into consideration of
production of auxiliary materials lime, steel scrap, iron ore
pellets, FeSi, and aluminum. Thus, it provides a cradle-to-
gate assessment of the primary FeMo production through-
out the whole supply chain system. One tonne of FeMo is
chosen as the calculation base in the process model. FeMo
is assumed to contain 60% Mo, which represents a standard
Table 1 Chemical reactions during ferromolybdenum production
Table1 is based on data from reference [19]
Typical reactions in a roaster

H2O(l)=H2O(g),ΔH298=43.39 kJ/mol
Equation1

MoS2+3.5O2=MoO3+2SO2,ΔH298 =−1086.08 kJ/mol
Typical reactions in an electric furnace Equation2

MoO3+2Al =Mo +2Al2O3,ΔH298 =−928.4 kJ/mol
Equation3

2MoO3+3Si =2Mo +3SiO2,ΔH298 =−1242.2 kJ/mol
Equation4

2Fe2O3+Si =4FeO +SiO2,ΔH298 =−348 kJ/mol
Equation5

2FeO +Si =2Fe +SiO2,ΔH298 =−367.01 kJ/mol
Equation6
Fig. 1 Model boundary of producing ferromolybdenum (Color figure online)

Journal of Sustainable Metallurgy
1 3
and commercial FeMo quality [17]. Auxiliary raw materi-
als such as lime and reduction agents are assumed to be
produced at a relatively close distance so that the influence
of their transport can be omitted. Furthermore, the electric-
ity distribution loss, reuse of SO2 in the acid plant, and use
of water and explosives in the mining process are excluded
in this study.
Description oftheProcess Model
A process model was developed together with the underly-
ing assumptions to obtain the inventory data for different
production stages. The process model follows the principle
of mass and energy conservations under steady-state condi-
tions. It is assumed that all input gases, fuels, raw materi-
als are charged at an ambient temperature (25°C), which
means that no sensible heat is added into the system. Also,
the gases in the model are following ideal gas behavior.
To simplify the modeling of the roasting process, it is
firstly assumed that all molybdenum is present in the form of
MoO3 in the RMC which means no MoO2 is formed during
process. Meanwhile, all sulfur is converted into a SO2 gas
after roasting. Secondly, the moisture present in the ore is
assumed not reacting with the metal and all vaporized into
the flue gas. Besides, the remaining components in the ore
such as Cu, Pb, P, Ca and Si remain in the outgoing roasted
concentrate. In addition, the discharged RMC and the flue
gas leave the furnace at 650°C. Also, it is assumed that no
sublimation of MoO3 occurs at this temperature. Thirdly,
assuming there is no extra heating source in the roasting pro-
cess as the matter of fact that the roasting reaction theoreti-
cally can provide a significant and sufficient heat for process.
Furthermore, in the present model, it is taking into account
a part of air as cooling air for maintaining the temperature
of the furnace and outgoing products at 650°C. At the
end, the enthalpy of mixing different chemical compounds
(MoO3, Cu, P etc.) for the RMC is considered as zero, thus
the enthalpy of RMC is determined by Eqs.7 and 8. This
applies to the flue gas, the slag, and the FeMo as well.
ΔHi
is the change of enthalpy for the compound i.
Ci
is
the specific heat capacity of the compound i at a constant
pressure. Thermodynamic data are from [19].
mi
is the mass
of the compound i.
ΔT
is the temperature difference between
the reference temperature (25°C) and the outlet temperature.
ΔHmixture
is the enthalpy change of the liquid, gas, and solid
mixture. It refers to the energy required to heat up RMC, flue
gas, FeMo, and slag from reference temperature to outgoing
(7)
ΔHi=CimiΔT
(8)
Δ
Hmixture =
∑
xiΔH
i
temperature in the present work.
xi
is the molar fraction of
the compound i.
It is assumed that the smelter operates as an electric arc
furnace with a 58% energy efficiency [20]. Other than RMC,
the other charged materials in the smelter include two iron-
based materials (iron ore and steel scrap), reduction agents
(FeSi and Al) and lime. The fluorspar is sometimes charged
to reduce the viscosity of the slag. In the present model, it
is assumed that the amount of fluorspar is small enough so
that it can be neglected [5]. To facilitate the modeling of the
smelting process, a charging ratio between the roasted con-
centrates and iron ore, aluminum, lime (650:309:58:80) [21]
is assumed. Addition of other raw materials such as steel
scrap and FeSi will be calculated through the process model.
Then, Fe2O3 in iron ore is considered as partly reduced; 70%
reduced to Fe and 30% reduced to FeO. This distribution
ratio can keep the concentration of FeO in the slag around
10% which is a reasonable level according to the published
FeMo slag analysis [5]. Moreover, the slag will be recycled
back to the process if the molybdenum is more than 0.3%
[17], so in this study the concentration of molybdenum in
the slag is assumed to be 0.3%. In addition, the model does
not take the dust loss into account. Finally, it is assumed
that the tapped molten FeMo and the slag are maintained at
the same temperature of 2000°C for energy calculation [5].
The following equations in Table2 may express mass
and energy conservation in a roaster and an electric smelter:
The greenhouse gas emission is estimated through Eq.21.
EMi is the emission source i, EFi is the emission factor
(EF) of pollutant i, Q is the amount of emission source such
as energy, fuel, and ingoing materials.
Results
In this study, four cases representing different molybdenum
concentrates were evaluated based on a developed process
model. The following results are presented as (1) inventory
data for FeMo production; a comparison of cases A, B, and
C with respect to (2) effect of beneficiation degree on energy
consumption and GHG emission; a comparison of cases C
and D with respect to (3) effect of mine type on energy con-
sumption and GHG emission; and (4) effect of process stage
on energy consumption and GHG emission.
Inventory Data forFeMo Production
The energy consumption and greenhouse gas emissions
during the manufacturing of FeMo are dependent on a cou-
ple of factors. FeMo alloys produced from four concentrate
(21)
EMi=EFi×Q

Journal of Sustainable Metallurgy
1 3
through pyrometallurgy production processes are selected
as case studies for examining the influence of different
factors such as process stage, beneficiation degree, etc.
on the energy consumption and GHG emissions. The four
cases are denoted as A, B, C, and D. In Table3, the basic
information of the four cases is shown.
Based on Table3, the following assumptions were made
in carrying out the energy consumption and greenhouse gas
emission calculations for producing FeMo:
(1) The energy consumption for case A during mining and
beneficiation is estimated through each equipment’s
yearly working hours, engine power, load factor, yearly
production, etc. [see details in electronic supplemen-
tary material (ESM), Tables1 and 2]. The energy con-
sumption data during the mining and beneficiation pro-
cess are not accessible for cases B, C, and D due to the
lack of information of plant data. A previous study has
indicated that the degraded ore can significantly affect
the energy consumption and greenhouse gas emission,
since additional energy must be consumed during
Table 2 Mass and energy conservation in the roaster and the electric smelter
Reactions in a roaster

MR,in =MR,out
Equation9

MR,in =MR,ORE +MR,AIR
Equation10
MR,out =MR,RMC +MR,GAS
Equation11

ER,in =ER,out
Equation12

ER,in =ER,EX
Equation13

ER,out =ER,RMC +ER,EN +ER,GAS
Equation14
Reactions in an electric furnace

ME,in =ME,out
Equation15

ME,in =ME,RMC +ME,IRON +ME,SCRAP +ME,AL +ME,SI +ME,LIME
Equation16

ME,out =ME,FM +ME,SLAG
Equation17

EE,in =EE,out
(18) Equation18

EE,in =EE,EL +EE,EX
Equation19

EE,out =EE,FM +EE,SLAG +EE,LOSS
Equation20
Mi, in i = R or E; R represents the roaster and E stands for the electric arc
furnace. Mi, in represents the total mass input in the roaster or in the
electric arc furnace
Mi, out i = R or E; R represents the roaster and E stands for the electric arc
furnace. Mi, out represents the total mass output in the roaster or in
the electric arc furnace
Ei, in i = R or E; R represents the roaster and E stands for the electric arc
furnace. Ei, in represents the total energy input in the roaster or in the
electric arc furnace
Ei, out i = R or E; R represents the roaster and E stands for the electric arc
furnace. Ei, out represents the total energy output in the roaster or in
the electric arc furnace
MR, jj = ORE, AIR, RMC, GAS; MR, j represents the mass of raw ore
(ORE), air (AIR), roasted molybdenum concentrate (RMC), and
flue gas (GAS) in the roaster, respectively
ER, jj = EX, RMC, EN, GAS; ER, j represents the chemical energy released
or required from exothermic reaction (EX), roasted molybdenum
concentrate (RMC), endothermic reaction (EN), and flue gas (FG)
in the roaster, respectively
ME, jj = RMC, IRON, SCRAP, AL, SI, LIME, FM, SLAG; ME, j represents
the mass of roasted molybdenum concentrate (RMC), iron ore
(IRON), scrap (SCRAP), aluminum (AL), ferrosilicon (SI), lime
(LIME), ferromolybdenum (FM), and slag (SLAG) in the electric
arc furnace, respectively
EE, jj = EL, EX, FM, SLAG, LOSS; EE, j represents the electricity (EL),
the energy released or required from exothermic reaction (EX),
ferromolybdenum (FM), the slag (SLAG), and energy loss in the
electric arc furnace, respectively

Journal of Sustainable Metallurgy
1 3
the mining and beneficiation processes to remove the
unwanted materials in the ore [28]. Here, it is assumed
that the energy consumption during the mining and
beneficiation stages proportional to the beneficiation
degrees are listed in Table3. The energy consumption
from the mining and beneficiation stages for cases B,
C, and D can be derived based on this assumption.
(2) For case D, molybdenum was extracted as a co-product
from a copper mine. The energy consumption during
the copper mining and beneficiation stages was esti-
mated based on reported data [29]. Here, the energy
consumption and environmental impact were allocated
to the co-product molybdenum based on mass value
rather than economic value, because economic value is
highly influenced by the volatility of the metal market.
The average production mass ratio of molybdenum to
copper during 5years is 3:100 [30], which is used as
the allocation factor in this study. Also, molybdenum
is separated during the copper’s beneficiation process.
This means that the energy consumption during the
mining and beneficiation will include two parts. One
part is from the allocation of copper mining and ben-
eficiation while the other part is from molybdenite’s
further floatation cycle, which usually accounts for 4%
of the total energy consumption in mining and ore pro-
cessing [31].
(3) Transport distances between each processing site
were estimated based on public information [24, 32,
33]. Moreover, the energy intensities for the different
transport modes were calculated by using the average
freight energy intensity values from the International
Energy Agency [34]. It is assumed that only diesel was
consumed during the transportation by rail and by ship.
Also, the conversion factor of energy consumption dur-
ing transportation is taken from a public source (see
ESM, Table3). Distances shorter than 50km have a
relatively low impact and were therefore not considered
and marked as “N.C.” in Table4.
(4) Cases A, C, and D use the emission factors from elec-
tricity production in the USA, which is mainly based
on coal (69%) and natural gas (29%) combustion [35].
The emission factors of electricity together with other
sources used in the presented study are provided in
ESM-Table3. Emission of CH4 and the N2O emis-
sions can be converted to CO2 equivalents (CO2-eq)
based on the global warming potential (GWP) values
Table 3 Information of four
molybdenum concentrates [15,
18, 22–27]
a Beneficiation degree = concentrate grade: the ore grade
Case Mo annual pro-
duction (tonne)
Share of the
world’s produc-
tion
Main production
and mine location
Ore grade Concen-
trate grade
Ben-
eficiation
degreea
A 11,900 4.2 Mo mine (USA) 0.03 58 1933
B 16,270 5.8 Mo mine (China) 0.115 51 443
C 23,130 8.2 Mo mine (USA) 0.21 53 252
D 11,500 4.1 Cu mine (USA) 0. 057 55 965
Table 4 Inventory data for one tonne FeMo (60%Mo) production
Underlined values are calculated outputs from the process model
Process Case A (Mo mine) Case B (Mo mine) Case C (Mo mine) Case D (Cu mine)
Mining and benefi-
ciation
Diesel (MJ) 152,304 39,760 21,762 55
Electricity (MJ) 10,506 2743 1501 3536
Transport 1 Rail transport
3500km
N.C Rail transport
1450km
Rail transport
1600km
Roasting Air (kg) 9033 8612 8388 8540
Transport 2 N.C N.C Sea transport
7000km
N.C
Smelting Iron scrap (kg) 114 92 111 114
Iron ore (kg) 437 488 450 437
Lime (kg) 113 113 113 113
Aluminum (kg) 82 92 85 82
FeSi75 (kg) 376 379 377 376
Electricity (MJ) 3630 4389 3855 3635

Journal of Sustainable Metallurgy
1 3
from the Intergovernmental Panel on Climate Change
fifth assessment report [36], where GWP100(CH4) = 28,
GWP100(N2O) = 265.
(5) It is assumed that all cases use the same quality aux-
iliary raw materials in their processes, including lime,
iron ore, scrap, FeSi, and aluminum. In other words,
the chemical analysis, energy requirement, and emis-
sion factor when producing these raw materials are
assumed to be the same in all cases (see ESM-Tables3
and 4). Other than the concentration of molybdenum,
the chemical compositions of other components in the
raw concentrate are described in online source ESM-
Table4. In the raw concentrate, the molar ratio of sulfur
to molybdenum in the raw ore is assumed to be 2 in the
ore concentrate, as the main components in the ore is
MoS2. Also, minor components of elements such as Cu,
Pb, P, CaO, and SiO2 are taken from literature data [17,
25, 37]. The rest is balanced with moisture.
According to the previously described conditions and the
calculation results from a process model, an inventory table
of producing one metric tonne FeMo (60%Mo) is shown in
Table4. The inventory marked with an underline represents
the calculation outputs from the process model while the
other inventory data are generated from Table2 and assump-
tions. In addition, it should be mentioned again that case D
represents a co-product from copper mine. An example of
case A’s calculation result is expressed in the online source
(see ESM-Tables5 and 6).
Eects ofOre’s Beneciation Degree onEnergy
andGHG Emission
The stage-by-stage energy consumption and greenhouse gas
emission results from model calculation of producing FeMo
(60%Mo) are illustrated in Fig.2.
As shown in Table3, the ore’s beneficiation degree (the
ratio of concentrate grade: raw ore grade) in case A (1933) is
higher than in case B (443) and case C (252). The decrease
in beneficiation degree will result in a reduction of the over-
all energy consumption from 188.6GJ/t FeMo to 50.2GJ/t
FeMo, as well as a decrease in the associated GHG emis-
sion from 14.79tCO2-eq/t FeMo to 4.69tCO2-eq/t FeMo.
The main influential stage is the mining and beneficiation
stages, since a higher beneficiation degree requires a higher
additional energy to remove the unwanted gangue material.
This consequently results in a higher associated GHG emis-
sion value, as 93% of the energy consumed during the min-
ing and ore processing stages is from fossil fuel diesel (see
ESM-Tables1 and 2).
The concentrate grade for cases A, B, and C is in the
range of 51–58%, while the ore grade is between 0.03 and
0.21%. The beneficiation degree, in accordance with the
definition (the ratio of concentrate grade:raw ore grade) is
therefore essentially determined by the initial ore grade.
When the ore grade contains 0.03%Mo (case A), the energy
output during mining and beneficiation stages accounts for
about 86% of the overall energy consumption (188.6GJ/t
FeMo) and 81% of the total GHG emission (14.79tCO2-eq/t
FeMo). Furthermore, for an ore grade of 0.21%Mo (case
C) these stages account for 46% of the overall energy con-
sumption (50.2GJ/t FeMo) and 37% of total gas emission
(4.69tCO2-eq/t FeMo). This agrees with the results from
Norgate [28], who reported that when the metal ore grade
is below 1%, the effect of additional energy and associated
greenhouse gas emissions due to deteriorated ore will be
significant. The downstream molybdenum roasting and
smelting processes are not affected by the ore grade, since
the generated concentrate grade during the ore processing
has a relatively constant value as is independent on the raw
ore grade.
Eects ofMine Type ontheEnergy andGHG
Emissions
Around 50% of the global molybdenum production is from
co-products from copper mining [18]. In the following, a
comparison is made between case D produced as a co-prod-
uct from a copper mine and case C where molybdenum is
produced from primary molybdenum mine. As shown in
Fig. 2 Stage-by-stage results for producing one tonne of FeMo a
energy consumption, b GHG emission (Color figure online)

Journal of Sustainable Metallurgy
1 3
Table3, the beneficiation degree in case D (965) is higher
than in case C (252). Although case D has a higher ben-
eficiation degree, the total energy consumption and GHG
emissions of producing one tonne FeMo are smaller than for
case C, which are 29.1GJ/t FeMo and 3.16tCO2-eq/t FeMo,
respectively. This is due to the allocation of energy based
on the mass ratio of copper and molybdenum in case D,
which consequently leads to a reduction of the GHG emis-
sion as well. As described earlier, the production mass ratio
of molybdenum to copper, the allocation factor 3:100 was
used in the study. It indicates that the mine type has a more
influential impact on the total energy and GHG emission
than the beneficiation degree has.
Eect ofProduction Stage ontheEnergy andGHG
Emission
As the analysis results for the four cases suggested, the
energy consumption during the mining and beneficiation
stages varies from 3.6GJ/t FeMo to 162.8GJ/t FeMo, or
between 12 and 86% of the overall energy consumption.
The associated GHG emissions for these stages range from
0.46tCO2-eq/t FeMo to 12.06tCO2-eq/t FeMo and the emis-
sion contribution varies from 14 to 81%. The differences
between cases derive from the beneficiation degree and the
allocation of co-products in the copper mine.
Apart from the mining and beneficiation stages, the roast-
ing stage is free from fuel consumption and GHG emissions,
as the model considers that the roasting is processed under
ideal conditions. In this case, oxidation can sufficiently pro-
vide the process heat and no extra heat source is required.
In reality, a small amount of fuel is required at the begin-
ning of the roasting to start the process and at the end of the
roasting process to make a final temperature adjustment of
the outgoing product. Thus, the fuel consumption is highly
dependent on the operational conditions. Besides, the inven-
tory consumption in the roasting stage, such as injection air,
is more dependent on the concentrate grade than the raw
ore grade and mine type, since the concentrate grade is at a
relatively constant level between 51 and 58%. In the roaster,
more than 90% of the input energy or oxidation heat goes
into the flue gas because large amounts of air are injected
to oxidize and cool the molybdenum ore material, as shown
in ESM-Table6.
Similarly, the smelting process is dependent on the con-
centrate grade. A large amount of chemical reaction heat
will be released during smelting stage (see ESM, Table6).
This means that less external energy is required for heating
the smelter. In the smelting process, more than 83% of the
energy and more than 73% of the GHG emissions origi-
nates from the upstream processes producing raw materials
such as FeSi, Al, and lime. The rest is from the extra energy
source electricity used by the furnace.
The transport stage consumes < 3% of the total energy
for all cases. Consequently, < 6% of the total GHG emis-
sion comes from the fuel consumption during transporting
processes, which has the least environmental impact of all
stages for producing one tonne of FeMo.
Discussion
The mine types and variation of ore qualities are of impor-
tance when making life cycle inventory studies for produc-
ing ferromolybdenum. It is usually resource consuming to
conduct the standard LCI study for each individual plant.
Thus, the developed process model is feasible to be used
by steel producers to estimate the energy consumption and
greenhouse gas emission for FeMo products from different
suppliers with a limited input data.
It is inevitable that the higher-grade reserves are extracted
first and that the high-grade ores are depleted over time. As
the ore grade deteriorates, the energy consumption and GHG
emissions for primary FeMo productions will obviously
increase. However, one approach to reduce the GHG emis-
sions during FeMo production can be a change of energy
sources. Today, the mining and beneficiation stages are
largely fossil fuel based. In the present study, case A’s diesel
use made up of 93% of the overall energy consumption dur-
ing the mining and beneficiation stages. During the carbon
emission factor of the diesel, fuel consumption in case A for
a stationary and a mobile combustion are 70.33gCO2-eq/MJ
and 70.44gCO2-eq/MJ, respectively. If the diesel is replaced
by nuclear electricity in case A, which has a much lower
carbon emission factor 1.667gCO2-eq/MJ, it will reduce
the emission in the mining and beneficiation stages from
12.06tCO2-eq/t FeMo to 1.58tCO2-eq/t FeMo. The result-
ing GHG reduction is therefore as large as 87%.
Additionally, the energy consumption and GHG emis-
sion of FeMo from the present study are shown in Table5,
together with LCA results of other common ferroalloys
based on average process data [38]. It can be observed
that both energy and GHG emission show distinct varia-
tions among different alloys. The purpose of Table5 is not
to compare the ferroalloys per se, but rather to underline
Table 5 Energy consumption and GHG emission of different ferroal-
loys from reference [38] together with results for FeMo from the pre-
sent study
FeMn FeNi FeSi FeCr FeMo
Ore grade (%) – 1.3 – 25.5 0.03–0.21
Alloy grade (%) 77 30 76 53 60
Energy (GJ/t ferroalloy) 48 325 90 77 29.1–188.6
GHG (tCO2-eq/t ferroalloy) 1.77 13.9 3.44 3.04 3.16–14.79

Journal of Sustainable Metallurgy
1 3
the feasibility to provide indicative results when lacking
actual production data. In other words, the methodologies
employed in the present study could be used in the analysis
of energy consumption and GHG emissions from other fer-
roalloys, such as FeMn, FeNi, FeSi, and FeCr in Table5,
to shed some light on the influence of different mining and
processing routes.
Conclusion
In this paper, a case study was carried out to evaluate the
energy consumption and greenhouse gas emissions during
pyrometallurgical production of FeMo. Four cases were
selected to assess the effect of the beneficiation degree,
mine type, and production stage on the final commercial
ferromolybdenum product’s energy consumption and GHG
emission. The most important findings from this study may
be summarized as follows:
• Among cases A, B, and C, the total energy consump-
tion of producing one tonne of FeMo (60%Mo) varies
from 29.1GJ/t FeMo to 188.6GJ/t FeMo, while the over-
all GHG emission varies from 3.16tCO2-eq/t FeMo to
14.79tCO2-eq/t FeMo. The main variance comes from
the mining and beneficiation stages. During these stages,
the consumed energy varies between 3.6GJ/t FeMo and
162.8GJ/t FeMo, while the associated GHG emission
varies between 0.46tCO2-eq/t FeMo and 12.06tCO2-
eq/t FeMo. The fluctuations are affected by the benefi-
ciation degree and the mine type, i.e., the allocation of
co-products from copper mining.
• The effect of mine type on the energy consumption and
the GHG emissions of FeMo’s production is more influ-
ential than the beneficiation degree. Among all the four
cases, the co-product in Case D produces FeMo with the
lowest energy consumption and least GHG emission.
• The initial ore grades and mine types have very little
influence on the downstream roasting and smelting pro-
cesses, since these processes are dependent on the con-
centrate grade. This, in turn, is relatively constant for all
the cases.
• The transport stage accounts for less than 3% of the over-
all energy consumption for all cases. Less than 6% of
the total GHG emission comes from the fuel combus-
tion during the transporting process, which has the least
impact of all stages in producing one tonne of FeMo.
Acknowledgements Open access funding provided by Royal Institute
of Technology.
Open Access This article is licensed under a Creative Commons Attri-
bution 4.0 International License, which permits use, sharing, adapta-
tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
included in the article’s Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
need to obtain permission directly from the copyright holder. To view a
copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.
References
1. Four Elements Consulting, LLC (2018) Life cycle inventory
of molybdenum products for metallurgical applications, update
study-summary report. International Molybdenum Association
2. Kennecott Utah Copper (2006) Kennecott Utah Copper molybde-
num oxide environmental profile-life cycle assessment. Kennecott
Utah Copper
3. Castro-Molinare J, Korre A, Durucan A (2014) Sustainability
analysis of copper extraction and processing using life cycle
analysis methods: a case study in the North of Chile. Comput
Aided Chem Eng 33:1861–1866
4. Muthu SS (2014) Chapter6—Estimating the overall environ-
mental impact of textile processing: life cycle assessment (LCA)
of textile products. In: Assessing the environmental impact of
textiles and the clothing supply chain. Woodhead Publishing,
Amsterdam, pp 105–131
5. Swinbourne DR, Arnout S (2018) Thermodynamic model of met-
allothermic smelting of ferromolybdenum. Miner Process Extr
Metall 128(3):1–12
6. Lessard JD, Shekhter LN, Gribbin DG, Mchugh LF (2013) Ther-
modynamic analysis of looping sulfide oxidation production of
MoO2 from molybdenite for energy capture and generation. JOM
65:1566
7. Tsalapatis J (2000) Application of an online heat and mass balance
model to an ironmaking blast furnace in School of Geoscience,
Minerals and Civil Engineering. University of South Australia
8. Eckelman MJ (2010) Facility-level energy and greenhouse gas
life-cycle assessment of the global nickel industry. Resour Con-
serv Recycl 54(4):256–266
9. Qi Z, Gao C, Na H, Ye Z (2018) Using forest area for carbon
footprint analysis of typical steel enterprises in China. Resour
Conserv Recycl 132:352–360
10. Camdali U, Tunc M (2016) Calculation of chemical reaction
energy in an electric arc furnace and ladle furnace system. Met-
allurgist 60(7):669–675
11. Gyllenram R, Jönsson PG, Ekerot S, Persson F, Ternstedt P
(2011) Taking in house and upstream CO2 emission into account
in charge optimization for scrap based steelmaking, METEC
InSteelCon 2011 proceedings, Düsseldorf, Germany
12. Gutiérrez AS, Martínez JBC, Vandecasteele C (2013) Energy
and exergy assessments of a lime shaft kiln. Appl Therm Eng
51(1):273–280
13. Liu P etal (2016) Material and energy flows in rotary kiln-electric
furnace smelting of ferronickel alloy with energy saving. Appl
Therm Eng 109:542–559
14. Luong VT etal (2018) A comparison of carbon footprints of mag-
nesium oxide and magnesium hydroxide produced from conven-
tional processes. J Clean Prod 202:1035–1044
15. United States Geological Survey (2017). 2014 Minerals Year-
book-Molybdenum. United States Geological Survey Web. https

Journal of Sustainable Metallurgy
1 3
://s3-us-west-2.amazo naws.com/prd-wret/asset s/palla dium/
produ ction /miner al-pubs/molyb denum /myb1-2014-molyb .pdf.
Accessed 18 Sept 2019
16. Jha MC (2001) Extractive metallurgy of molybdenum. In: Mishra
B (ed) Review of extraction, processing, properties & applications
of reactive metals. The Minerals, Metals & Materials Society, pp
73–82
17. Gasik M (2013) Chapter12 - Technology of Molybdenum Fer-
roalloys. In: Handbook of Ferroalloys, Butterworth-Heinemann:
Oxford pp 387–396.
18. Fthenakis V, Wang W, Kim HC (2009) Life cycle inventory analy-
sis of the production of metals used in photovoltaics. Renew Sus-
tain Energy Rev 13(3):493–517
19. Barin I, Knacke O (1973) Thermochemical properties of inorganic
substances. Springer, Berlin
20. Kirschen HPM (2002) Thermodynamic analysis of EAF electrical
energy demand, 7th European electric steelmaking conference,
Venice
21. Northcott L (1956) Molybdenum. Butterworth, Metalllurgy of the
rarer metals
22. United States Geological Survey (2017). 2014 minerals yearbook-
China. United States Geological Survey Web. https ://s3-us-west-2.
amazo naws.com/prd-wret/asset s/palla dium/produ ction /miner al-
pubs/count ry/2014/myb3-2014-ch.pdf. Accessed 18 Sept 2019
23. Thompson Creek Mine (2013) Air permit application to convert
tier II operating permit to a permit to construct. Thompson Creek
Mining Company
24. Cappa JA etal (2006) Colorado Mineral and Energy Industry
activities, 2005. Colorado geological survey web. https ://color
adoge ologi calsu rvey.org/wp-conte nt/uploa ds/2013/08/MMF20
05.pdf. Accessed 18 Sept 2019
25. Nair KU et al (1987) Chlorination of commercial molyb-
denite concentrate in a fluidized bed reactor. Metall Trans B
18(2):445–449
26. China Molybdenum Co., Ltd (2018) Semi-annual report for
2018. China Molybdenum Co.,Ltd Web. https ://www.china moly.
com/06inv est/doc_a/2018/60399 3_2018_zzy.pdf. Accessed 18
Sept 2019
27. Krahulec K (2018) Production history of the Bingham mining dis-
trict. Salt Lake County, Utah - an update. https ://doi.org/10.13140
/RG.2.2.28618 .00966
28. Norgate T, Jahanshahi S (2006) Energy and greenhouse gas impli-
cations of deteriorating quality ore reserves. 5th Australian confer-
ence on life cycle assessment
29. Norgate T, Rankin W (2000) Life cycle assessment of copper and
nickel production, International Congress on Mineral Processing
and Extractive Metallurgy. Melbourne, The Australasian Institute
of Mining and Metallurgy
30. Benavides PT, Dai Q, Sullivan J, Kelly J, Dunn JB (2015) Mate-
rial and energy flows associated with select metals in GREET2:
molybdenum, platinum, zinc, nickel, silicon. Argonne national
laboratory web. https ://greet .es.anl.gov/publi catio n-mo-pt-zn-ni-
si. Accessed 18 Sept 2019
31. Lelinski D, Govender D, Dabrowski B, Traczyk F (2011) Effective
use of energy in the floatation process, 6th Southern African base
metals conference 2011. The Sourthern African Institute of Min-
ing and Metallurgy Web. https ://www.saimm .co.za/Confe rence s/
BM201 1/137-Mulli gan.pdf. Accessed 18 Sept 2019
32. Schulz KJ etal (2017) Critical mineral resources of the United
States: economic and environmental geology and prospects for
future supply. United States Geological Survey Web. https ://pubs.
usgs.gov/pp/1802/pp180 2_entir ebook .pdf. Accessed 18 Sept 2019
33. U.S. Environmental Protection Agency (1992) Mine site visit:
Cyprus Thompson Creek. U.S. Environmental Protection Agency
Web. https ://archi ve.epa.gov/epawa ste/nonha z/indus trial /speci al/
web/pdf/phosm ol2.pdf. Accessed 18 Sept 2019
34. International Energy Agency (2017) Average freight energy inten-
sity and activity in 2015, International Energy Agency Web. https
://www.iea.or g/n ewsr oom/energ ysnap shots /avera ge-freig ht-energ
y-inten sity-and-activ ity.html. Accessed 18 Sept 2019
35. U.S. Energy Information Administration(2019) How much of
U.S. carbon dioxide emissions are associated with electricity
generation? https ://www.eia.gov/tools /faqs/faq.php?id=77&t=11.
Accessed 18 Sept 2019
36. IPCC (2013) Climate change 2013: the physical science basis.
Contribution of working group I to the fifth assessment report of
the intergovernmental panel on climate change, IPCC web. https
://www.ipcc.ch/repor t/ar5/wg1/. Accessed 18 Sept 2019
37. China Molybdenum Co., Ltd. (2019) Molybdenum concentrate.
China Molybdenum Co., Ltd. Web. https ://www.china moly.com/
en/03pro ducts /detai l_mujin gkuan g.htm. Accessed 18 Sept 2019
38. Haque N, Norgate T (2013) Estimation of greenhouse gas emis-
sions from ferroalloy production using life cycle assessment with
particular reference to Australia. J Clean Prod 39:220–230
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.