Journal of Cleaner Production 248 (2020) 119189

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Two-phase expander refrigeration cycles with ethaneenitrogen: A

cost-efficient alternative LNG processes for offshore applications
Muhammad Abdul Qyyum a, Kinza Qadeer a, Ashfaq Ahmad b, Faisal Ahmed c,
Moonyong Lee a, *
a
School of Chemical Engineering, Yeungnam University, Gyeongsan, 712-749, Republic of Korea
b
Department of Computer Science, COMSATS University Islamabad, Lahore Campus, Pakistan
c
Department of Chemical Engineering, COMSATS University Islamabad (CUI), Lahore Campus, Defense Road, Off Raiwind Road, Lahore, Pakistan

a r t i c l e i n f o a b s t r a c t

Article history: Nitrogen expander-based natural gas (NG) liquefaction processes are considered to be the most feasible
Received 29 April 2019 and economic practices for offshore applications. The nitrogen single expander process is simple owing
Received in revised form to its single-phase operation, and it employs high occupational safety and environment-friendly re-
27 October 2019
frigerants. However, this process has low energy efficiency. Recent advancements in expansion devices
Accepted 4 November 2019
Available online 8 November 2019
show the possibility of handling two-phase refrigerant flow in an economical and simple manner.
Accordingly, this paper proposes two-phase expansion using an innovative binary mixed refrigerant
Handling Editor: Giorgio Besagni (MR) composed of ethane and nitrogen (C2N). Furthermore, a propane-precooling refrigeration cycle is
also implemented and evaluated to analyze greater potential benefits of ethaneenitrogen refrigerant
Keywords: with two-phase expansion. To assess the technical and commercial feasibility of the proposed lique-
Floating liquefied natural gas faction process, energy, exergy, and economic (3E) analysis is performed. Using the C2N two-phase
Offshore expander LNG process, 47.83% energy can be saved with 55.25% exergy destruction minimization and
Propane-precooling 24.12% total annualized costs (TAC) savings as compared to previously published nitrogen single
Ethaneenitrogen
expander process. Whereas, the C3-precooled C2N process gives higher energy savings i.e., 52.45% but
Economic analysis
low TAC savings i.e., 1.6% as compared to nitrogen single expander LNG process. Considering TAC savings,
Exergy destruction
the C2N process (without propane-precooling) can be a promising candidate for offshore applications.
© 2019 Elsevier Ltd. All rights reserved.

1. Introduction acknowledged as one of the most suitable fuels among the various
types of known fossil fuels used to produce energy owing mainly to
The Industrial Revolution is responsible for a dramatic increase its low carbon emissions and its ability to meet strict environ-
in air pollution. Since then, rising levels of air pollutants such as mental regulations. Shell has predicted that the demand for NG will
COx, SOx, and particulates in the atmosphere have promoted increase 60% from 2010 to 2030 (Press Release December 2012.,
climate change. In particular, large amounts of CO2, which leads to 2012); NG currently accounts for 23% of the total global energy
global warming, are emitted primarily from coal and oil-based consumption; this figure is expected to reach 26% by 2040
power-generation industries (Qyyum et al., 2019a). During the (ExxonMobil, 2014; Kuwahara et al., 2000).
past 100 years, the CO2 concentration levels in the atmosphere have However, NG reservoirs are usually found in offshore and
increased rapidly, reaching approximately 404.34 ppm by the end remote areas, from which the product must be transported for
of 2017 (Scripps Institution of Ocenography, 2017). It is estimated trade in the global market; therefore, the locations of natural gas
that by the end of this century, the global temperature will rise reserves play a vital role in the economics of transportation and
2e5
C if CO2 emissions are not controlled (Ghaedi et al., 2017; storage (Strantzali et al., 2019). Natural gas can be either be trans-
Pachauri and Reisinger, 2007). Considering the present energy ported through pipelines or liquefied and transported through
scenario along with environmental challenges, natural gas (NG) is cargo ships from remote areas. The transportation of natural gas in
liquid form is considered to be more safe, feasible, and economical
over long distances, particularly those >3500 km (Dutta et al., 2018;
* Corresponding author. Horvath et al., 2018; Mehrpooya and Ansarinasab, 2015;
E-mail address: mynlee@yu.ac.kr (M. Lee).

https://doi.org/10.1016/j.jclepro.2019.119189
0959-6526/© 2019 Elsevier Ltd. All rights reserved.
2 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189

Yuan et al. (2014) also used CO2 as a precooling refrigerant to

Nomenclature and abbreviations enhance the N2 expander process. They optimized the process to
achieve a liquefaction rate of 77% with a unit energy consumption
Ep Equipment purchase cost, $ of 9.90 kWh/kmol. Shah et al. (2009) used non-dominated sorting
k1, k2, k3 Constants genetic algorithm (NSGA-II) to optimize the total shaft work, capital
A Capacity parameter cost, and annualized cost for propane-precooled dual N2 expander
CBM Bare module cost, $ LNG process. Khan et al. (2014) enhanced the performance of the N2
FBM Bare module factor expander process using the CO2 precooling cycle for offshore ap-
FLNG Floating liquefied natural gas plications. They optimized the process using knowledge-inspired
MR Mixed refrigerant approach and reduced the compression power by 15.8%. Ding
LNG Liquefied natural gas et al. (2016) developed a model of the N2 expander process in
NG Natural gas Aspen Hysys® simulation environment and optimized the model
TDCC Temperatureeapproach temperature composite using the genetic algorithm (GA). In addition, they also studied the
curves effect of propane precooling corresponding to overall energy sav-
THCC Temperatureeheat flow composite curves ings. Most recently (Qyyum et al., 2018d), introduced two-phase
TAC Total annualized costs, $ single expander refrigeration cycle with propane and nitrogen
NBP Normal boiling point,
C (C3N) as a binary mixed refrigerant. They optimized the process
MCD Modified coordinate descent using the particle swam optimization technique to reduce the
JT JouleeThomson specific energy requirements.
MITA Minimum internal temperature approach,
C It can be deduced from the literature that the major issue
associated with the N2 expander LNG processes is the low energy
efficiency (high exergy destruction), which ultimately leads to high
total annualized costs (TAC), mainly due to the high operating costs.
Mehrpooya and Ghorbani, 2018). However, NG liquefaction is an The high exergy destruction is attributed mainly due to the large
energy- and cost-intensive process; the liquefaction sections alone temperature difference between the cooling temperature of the N2
take up approximately 40%e50% of the total expenditure of the refrigerant used in the refrigeration cycle and the feed NG, which
liquefied natural gas (LNG) value chain (Lim et al., 2013; Mehrpooya leads to rapidly liquefaction and subcooling of NG to result in en-
and Ansarinasab, 2015; Vatani et al., 2014). Liquefaction and energy tropy generation (Qyyum and Lee, 2018). It is believed that the
requirement costs depend on various parameters such as the cooling of the NG before going to liquefaction and subcooling can
liquefaction technology used (Qyyum et al., 2018c) and the envi- be reduced the overall entropy generation, which ultimately will
ronmental conditions at the plant site (Qyyum et al., 2018b). improve the overall energy efficiency. As reported previously (Gao
The most widely used liquefaction technologies are nitrogen et al., 2010; He and Ju, 2014; Mortazavi et al., 2012; Shah et al.,
(N2) expander-based LNG and mixed refrigerant (MR)-based pro- 2009), exergy destruction can be reduced by modifying the N2
cesses, the latter of which is more economically feasible owing to expander process through the involvement a precooling refrigera-
its lower energy consumption and therefore its lower operating tion cycle (e.g., C3/CO2/R410a precooling cycle) or through the
costs (Yin et al., 2008). However, the major drawbacks of the MR addition of hydrocarbons (e.g., methane or propane) in the main
process are its high degree of complexity, high capital investment, refrigeration cycle. Previous research has shown that methane
and involvement of flammable hydrocarbons as a refrigerant, addition in the main refrigeration cycle (N2eCH4) increases the
making them less attractive for offshore applications (Qyyum et al., performance of the N2 expander process and C3-precooling cycle
2018c). On the contrary, the N2 expander-based LNG process en- helped to further enhance the performance of the N2eCH4 cycle
sures high occupational safety, simple and user-friendly operability, when a single-phase expander was used (Ding et al., 2016). How-
and easy availability and portability. Further, its low capital in- ever, recent innovations and improvements in expander technol-
vestment makes it the best choice for offshore operations. How- ogy facilitate an efficient handling of two-phase mixture.
ever, high energy consumption is the main limitation associated Furthermore, heat transfer characteristics of two-phase mixed hy-
with nitrogen expander-based liquefaction technologies (Qyyum drocarbon refrigerants flow boiling in shell and tube heat ex-
et al., 2019b). changers (Hu et al., 2018) as well as LNG heat exchangers (Hu et al.,
Therefore, many researchers have proposed different ap- 2019) have also been investigated. Therefore, the high boiling hy-
proaches using either refrigeration cycle retrofitting and optimi- drocarbon such as propane can be added to the main refrigeration
zation or only optimization to minimize the energy consumption of cycle of nitrogen (C3N) as reported by (Qyyum et al., 2018d). They
nitrogen expander liquefaction processes. For example (Cao et al., retrofitted a two-phase expander with a single-phase (gas)
2006), introduced a binary mixed refrigerant N2eCH4, however, expander to handle a single mixed two-phase refrigerant, thus
they used single-phase expander to produce LNG with lower power enhancing the process efficiency. Nevertheless, C3-precooling cycle
consumption than conventional N2 expander process. Recently is not a suitable option to integrate with C3N two-phase expander
(Haider et al., 2019), also used N2eCH4 expander liquefaction pro- process, mainly due to the temperature distribution range of the
cess to produce liquefied biomethane followed by biogas upgrad- precooling refrigerant (propane) that is already available with ni-
ing. Gao et al., (2010) employed the propane pre-cooling trogen in the main refrigeration loop. Therefore, new binary MRs
refrigeration cycle to improve the energy efficiency of the N2 are required to realize the benefits of the propane precooling
expander process for the liquefaction of coalbed methane. He and refrigeration cycle integrated with two-phase expander LNG pro-
Ju (2014) reduced the operating costs of an N2 expander process cess. Ethane, with a normal boiling point (NBP) of 89.0
C, can be
by using R410a and propane for precooling, which reduced the one of the possible candidates to make combination with nitrogen
energy consumption by 20% and 22.74%, respectively, compared in two-phase expander single loop.
with the conventional nitrogen expander process. (Aspelund et al., Hence, in this study, we propose a binary MR of
2007) improved the operational performance and reduced the ethaneenitrogen that can work with newly developed cryogenic
overall operational cost of the CO2 precooled N2 expander process two-phase expander. To achieve the maximum benefits of the
using the Extended Pinch Analysis and Design (ExPAnD) technique. proposed ethaneenitrogen refrigerant, a propane precooling
 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189 3

refrigeration cycle is also integrated with ethaneenitrogen two- The optimal parameters (e.g., temperature, pressure, and flow rate)
phase single expander liquefaction process. The proposed pro- for each stream are used to explain the process description of the
cesses are optimized using the modified coordinate descent (MCD) proposed processes.
optimization technique to reduce the overall energy consumption. In the proposed C2N process, the refrigerant (C2N) mixture
To evaluate the technical and commercial feasibility of the pro- (stream 1) is compressed in a four-stage compression cycle with
posed processes, a detailed exergy and economic analysis is also interstage cooling in the pressure range of 4.78 bare38.39 bar
performed. Further, the performance of the proposed optimized (stream 9), which is found by the optimization algorithm. The
processes are compared with previously published nitrogen interstage coolers help to maintain the temperature of the propane
expander-based liquefaction processes. This article is organized as and C2N mixture at 30
C. A compression ratio of 1:3 is maintained
follows. Section 2 describes the main motivation and theory about in the compressors to favor reversibility. This is because any irre-
the proposed ethaneenitrogen refrigeration cycles and propane versibility generated in the process causes an overall increase in
precooling-assisted ethaneenitrogen refrigeration cycles for FLNG exergy destruction (Tsatsaronis and Morosuk, 2012). The streams of
projects. Section 2.1 provides the detailed process description of feed NG (stream A) and C2N (stream 10) are then passed through
the proposed liquefaction processes. Section 2.2 includes the feed the LNG exchanger (LNG-100); the two-phase expander expands
NG composition, conditions, and simulation assumptions. Section stream 10 through a high-pressure two-phase C2N to 4.88 bar
2.3 consists of process optimization. Section 3 includes the process (stream 11). This produces sufficient cooling for LNG-100 by pro-
analysis in terms of energy, exergy, and economic. Finally, the moting heat exchange between the warm NG and C2N streams and
conclusions drawn from this study are presented in Section 4. provides adequate cooling to liquefy the NG and to partially liquefy
the C2N refrigerant. The exiting stream 12 from the heat exchangers
2. Propane precooling-assisted ethaneenitrogen is in the form of superheated vapors that are recycled to complete
refrigeration cycles the refrigeration cycle. Similarly, in the second proposed process,
the feed NG and refrigerant (C2N) are cooled to about 16
C in a C3-
For efficient NG liquefaction, it is important to follow the ther- precooling cycle. This is in addition to a C2N refrigeration cycle that
modynamic pattern of gradual cooling, liquefaction, and subcool- condenses and subcools NG.
ing. A high temperature gradient between the selected refrigerant In the precooling refrigeration cycle, propane (stream 1-1) is
and feed gas gives a rapid liquefaction and subcooling of the feed compressed to an optimum pressure of 11.05 bar (stream 1e6)
gas, resulting exergy destruction. This exergy destruction is the using three compression stages occurring in the interstage cooling
main reason for the low performance of the N2 expander process system. Similarly, the C2N mixture (stream 1) is compressed in a
when compared with the MR liquefaction process. For example, the four-stage compression cycle that includes interstage cooling from
expander cycle in the N2 expander liquefaction process uses N2, a pressure of 4.78 bare38.39 bar (stream 9). The interstage coolers
which has a boiling point of 195
C (1 kJ/mol specific rerfrigera- maintain the temperatures of the propane and C2N mixture at
tion effect against 246.0 bar pressure), in the main refrigeration 30
C. The compressed propane is expanded through a
cycle, and feed NG is normally introduced at 30
C (Khan et al., JouleeThomson (JT) valve to reduce its pressure to 2.48 bar and
2015a, 2014). In this case, a large difference is present between temperature to 19.48
C before it is introduced into the LNG
the temperatures of the introduced feed gas and the boiling point of exchanger, LNG-101, where it exchanges heat with the feed NG and
the refrigerant in the refrigeration cycle. Therefore, the process compressed C2N and precools them both to 16.48
C.
undergoes liquefaction and subcooling rapidy by exchanging the The precooled streams of feed NG (stream B) and C2N (stream
only sensible heat of nitrogen with feed NG, which results in high 11) are then passed through the LNG-100 exchanger; the high-
exergy destruction. To avoid this exergy destruction, (Qyyum et al., pressure two-phase stream of C2N is expanded in the two-phase
2018d) added an optimal flow rate of propane in the nitrogen expander to 4.88 bar to produce sufficient cooling for the LNG-
refrigeration cycle to incorporate the precooling phenomenon. 100. This occurs through the heat exchange from the warm NG
However, when propane was mixed with nitrogen, this mixture and C2N streams, thus providing adequate cooling to liquefy the NG
was appeared as two-phase, thus, two-phase expander was intro- and to partially liquefy the C2N refrigerant. The exiting streams
duced. In the current investigation, ethane is mixed with nitrogen (stream 1-1) and (stream 13) from the heat exchangers (LNG-101
(C2N) to form a single MR for the main refrigeration cycle. The and LNG-100) are in the form of superheated vapors that are
boiling point of ethane is significantly lower than that of propane recycled to complete the refrigeration cycle. Stream C exits LNG-
(i.e., 89.0
C) and it gives 1 kJ/mol specific refrigeration effect with 100 as a subcooled liquid at a pressure of 50 bar. For economically
significant lower pressure i.e., 16.5 bar. Therefore, ethane is feasible as well as safe storage and transportation of LNG this high
preferred for the main loop, whereas propane is used inside the pressure is not suitable. Therefore, stream C is passed through LNG
precooling loop owing to its strong performance as a precooling turbine (Turb-1) to obtain LNG at 158.5
C and a pressure which is
refrigerant (Khan et al., 2015a; Qyyum et al., 2018d). The reason for slightly higher than the atmospheric pressure (1.2 bar was fixed in
using propane inside the precooling refrigeration cycle can be this study). LNG is produced at a liquefaction rate of 92% with 8%
attributed to its effective precooling properties. If used inside the end flash gas. The produced LNG is then stored in storage tanks.
main loop, propane provides a precooling effect inside the loop.
However, it cannot enhance the cycle efficiency because its distri- 2.2. Process simulation
bution range differs significantly from that of nitrogen; thus, C3-
precooling for the C3N cycle would not have a significant effect. The proposed LNG processes were simulated using Aspen
Therefore, ethane was mixed with nitrogen rather than propane. Hysys® version 10 software. The PengeRobinson equation of state
was chosen along with the LeeeKesler equation for rigorous cal-
2.1. Description of the proposed liquefaction process culations of thermodynamic properties (Qadeer et al., 2018). It
should be noted that the LeeeKesler model is the precise standard
The proposed C2N two-phase expander and C3-precooled C2N used to calculate the enthalpy of gases at high pressures (Li et al.,
two-phase expander liquefaction processes are illustrated in Fig. 1 2012; Yuan et al., 2015). The essential process feed conditions are
and Fig. 2, respectively. To understand the process description, shown in Table 1. Additionally, for process simulation, the following
various streams are labeled as “stream n” (n ¼ 1, 2, 3 …) in figures. presumptions were incorporated:
4 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189

Fig. 1. Process flow diagram (PFD) of the proposed C2N two-phase expander-based LNG process.

Fig. 2. PFD of the proposed propane precooled C2N two-phase expander-based LNG process.

 The pressure drop (DP) in all inter-stage coolers was set at  The isentropic efficiencies of the compressor, two-phase
0.25 bar. expander, and LNG turbine were selected as 75%, 80%, and
90%, respectively.
 The temperature of cooling medium (water/air) for inter-stage
cooling was taken to be 20
C.
 The pressure of the end flash gas was set at 1.2 bar.
Table 1  The minimum internal temperature approach (MITA) was
Feed composition, conditions, and basis for process simulation. selected as 3.0
C in the main cryogenic heat exchanger.
Feed composition Mole (%)

Methane 91.28
All assumptions were taken from the recent research (Abdul
Ethane 5.40 Qyyum et al., 2018; Khan et al., 2015b; Qyyum et al., 2018d) of ni-
Propane 2.10 trogen expander-based LNG processes.
i-Butane 0.50
n-Butane 0.50
i-Pentane 0.01
2.3. Process optimization
n-Pentane 0.01
Nitrogen 0.20
After-cooler outlet temperature (
C) 30.00 The non-optimal execution of design variables contribute to-
ward significant exergy destruction, which results in low energy
Natural gas feed conditions Value
efficiency of the process. Suitable optimization techniques can
Temperature (
C) 30.00
result in large amounts of energy savings for a given LNG process.
Pressure (bar) 50.00
Flow rate (kg/h) 1.00 Alteration by adding or removing new devices in existing LNG
Pressure drop (DP) across the main LNG cryogenic heat exchanger processes can also change the optimal operation parameters of a
“Stream-9” to “Stream-1000 1.0 bar (hot stream) given liquefaction process.
“Stream-11” to “Stream-1200 0.1 bar (cold stream) The key design parameters influencing process efficiency such
00
“Stream-15” to “Stream-16 1.0 bar (hot stream)
as the refrigerant flow rate in refrigeration cycles, refrigerant
 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189 5

evaporation and condensation pressures in both cycles, and pre- 3.1. Energy analysis
cooling and subcooling temperatures need to be optimized to
obtain the optimum design variables. For the design optimization Energy analysis was conducted with the optimal design vari-
of the proposed LNG processes, the design variables were chosen ables for the proposed C2N and C3-precooled C2N two-phase
based on a deep knowledge-inspired based procedure as followed expander processes, and the results were compared with those of
by (Khan et al., 2013; Pham et al., 2017) and are shown in Table 2 previously published nitrogen single, dual, and C3N two-phase
with the corresponding upper and lower limits for the proposed expander processes (Table 3).
processes. According to Table 3, the proposed C2N and C3-precooled C2N
The optimization objective was to minimize the total power processes yielded 33.30% and 15.61% lower mass flow rates,
input of the compression cycles (C2N and C3) while constraining the respectively, in the refrigeration cycle. In the case of N2 only, the
MITA value to 3
C in the LNG-100 and LNG-101 heat exchangers, mass flow rate in the single expander process was 8.25 kg*h1,
respectively. The objective function is represented by Eq. (1): whereas those of the proposed C2N and C3-precooled C2N processes
were 1.515 kg*h1 and 1.502 kg*h1, respectively. The evaporation

X
n _ 
W j pressure of the refrigerant was also lower for the proposed C2N and
Minf ðXÞ ¼ Min: (1) C3-precooled C2N processes, at 85.77 bar and 38.14 bar, respec-
j¼1
m_ LNG
tively. This advantage regarding reduced energy consumption is not
possible in conventional processes. In particular, the reduction in
which is subject to the conditions
the amount of N2 and evaporation pressure of the refrigerant
controls the capital cost as well as the operating costs as it affects
DTmin ðXÞ  3 (2)
the compressors’ power consumption. The compression ratio in the
N2 single expander process was 2.12, whereas the compression
and
ratios in the proposed C2N and C3-precooled C2N processes were
1.857 and 1.680, respectively. A low compression ratio is crucial for
Xlb < X < Xub (3)
all refrigeration cycles because it increases the refrigerant perfor-
where X is a vector of decision variables given as X ¼ mance in NG liquefaction.
  Previously reported N2 expander processes have shown low
P1 ; P2 ; TP ; C3 P1 ; C3 P2 ; m_ C2; m_ C3 ; m_ N2 .
The constrained objective function and design variables un- energy efficiency with a 92% liquefaction rate, whereas the pro-
dergo mutual non-linear interaction, which makes it difficult to posed processes exhibit a lower power demand at the same
utilize commonly available simulation software for optimization. liquefaction rate. Table 4 compares the specific power consumption
Consequently, the process model in Aspen Hysys® should be linked values of the N2 expander-based LNG production processes against
with an externally available optimization technique. The MCD al- the proposed C2N and C3-precooled C2N two-phase expander-
gorithm (Park et al., 2015; Qyyum et al., 2018a) was employed to based processes while neglecting their different feed conditions but
optimize the proposed processes. Microsoft Visual Studio was uti- keeping a constant liquefaction rate of 92% respectively. The
lized to develop MCD algorithm based optimizer and was then comprehensive specific power consumptions of the proposed C2N
linked to Aspen Hysys®. The MCD optimization method has been and C3-precooled C2N processes were 79.72% and 81.52%, when
used successfully to optimize highly non-linear and complex sys- compared to the conventional N2 single expander-based LNG pro-
tems (Long et al., 2016; Qyyum et al., 2018a). Therefore, it is suitable cess demonstrated by (Du et al., 2010). When compared to the N2
for interactive and highly non-linear optimization problems such as single expander-based LNG process developed by (Austbø and
LNG processes. Fig. 3 presents the working mechanism of the MCD Gundersen, 2015), the proposed processes lowered the energy
algorithm; details of the MCD optimization strategy can be found in consumption by 51.16% and 55.49%, respectively. Although, the
previous research (Park et al., 2015). energy savings were relatively less when compared to the dual
expander process (Khan et al., 2013), it still resulted in major re-
ductions i.e. 20.83% and 27.85% respectively. A comparison with the
C3N two-phase expander process proposed in previous research
3. Energy, exergy, and economic analysis (Qyyum et al., 2018d) showed that the proposed processes yielded
energy savings of 2.46% and 11.11%, respectively. In these cases, the
This section provides a detailed analysis of the proposed lique- power consumption depended specifically on the NG feed
faction processes with respect to energy, exergy, and economic composition, liquefaction rate, and operating conditions.
evaluation (3Es). Furthermore, the optimal pressures and temper-
atures for all streams of the proposed C2N two-phase expander
process and the C3-precooled C2N two-phase expander process are 3.2. Exergy analysis
included in Supporting Information as Table S1 and Table S2,
respectively. Exergy analysis is used to identify the irreversibilities within

Table 2
Key decision variables with upper and lower limits for the proposed processes.

Decision Variables Lower Limit Upper Limit

Refrigerant high pressure, P1 (bar) (stream-9) 25.00 87.00

Refrigerant low pressure, P2 (bar) (stream-1) 3.0 10.00
Nitrogen mass flow rate, m _ N2 (kg/h) 1.50 10.0
Ethane mass flow rate, m _ C2 (kg/h) 0.50 4.50
Precooling refrigerant high pressure, C3P1 (bar) (stream-1-6) 8.00 40.0
Precooling refrigerant low pressure, C3P2 (bar) (stream-1-1) 2.00 15.00
Precooling Temperature, TP (
C) 60.0 20.0
Propane flow rate, m _ C3 (kg/h) 3.00 10.0
6 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189

Fig. 3. MCD optimization algorithm.

each system component. By performin an exergy analysis, we can

find the location, magnitude and causes of process inefficiencies. Ex ¼ ðh  ho Þ  To ðs  so Þ: (4)
This technique introduces single unit operations, which represent a In this study, exergy analysis was performed for each equipment
large amount of lost work; it thus helps to identify valuable infor- associated with previous published nitrogen expander-based and
mation for process improvement from the design and operations proposed LNG processes. The equations used to calculate exergy
perspectives. Generally, the exergy of any stream constitutes of destruction were sourced from previous studies (Venkatarathnam
physical, chemical and mechanical; whereby the mechanical and Timmerhaus, 2008) and are given below as Eq. (5) to (8). For
exergy can be neglected due to its small involvement (Qyyum et al., equations (5)e(8), the values for Exin and Exout were taken directly
2018a). Moreover, due to the absence of any chemical reaction in from the stream properties calculated by Aspen Hysys v10. The
the process, the chemical exergy is also taken as zero (Qyyum et al., Aspen Hysys v10 estimate the exergy of each stream assuming
2018a; Tsatsaronis and Morosuk, 2012). The physical exergy can standard conditions as; T0 ¼ 25.0
C and P0 ¼ 1.013 bar.
then be expressed as: The equation for exergy destruction calculation in the
compressor is given in Eq. (5).
 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189 7

Table 3
Energy analysis with the optimal design variables for the proposed C2N and C3-precooled C2N two-phase expander processes.

Parameters N2 single expander N2 dual expander (Khan C3N two-phase expander Proposed C2N Proposed C3-
(Khan et al., 2015a) et al., 2015a) (Qyyum et al., 2018d) process precooled C2N process

Mass flow rate of nitrogen, m_ N2 (kg/h) 8.257 13.660 3.160 1.515 1.502
Mass flow rate of ethane, m_ C2 (kg/h) e e e 3.992 2.973
Mass flow rate of propane, m_ C3 (kg/h) e e 2.810 e 2.493
Total refrigerant flow rate (kg/h) 8.257 13.660 5.970 5.507 6.968
Refrigerant evaporating pressure, P2 (bar) 100.0 100.0 80.0 85.77 38.14
Refrigerant condensation pressure, P1 (bar) 6.0 14.0 4.80 7.205 4.783
Precooling refrigerant high pressure, C3P1 (bar) e e e e 11.05
Precooling refrigerant low pressure, C3P2 (bar) e e e e 2.487
Precooling Temperature, TP (
C) e e e e 16.48
MITA (
C) 3.0 3.0 3.0 3.0 3.0
LNG liquid fraction (by mole) 0.92 0.92 0.92 0.92 0.92
Pressure ratio 2.12 1.464a 2.021 1.857 1.680
1.826b
Refrigerant specific compression power (kW/kg) 0.9357 0.5010 0.4734 0.4139 0.2913
Specific compression power recovered through 0.1908 0.1472 0.07139 0.02214 0.02055
two-phase expander (kW/kg)
Specific compression power recovered through JT-Valve JT-Valve 0.003105 0.003105 0.003089
the LNG turbine (kW/kg)
Net specific compression power requirement 0.7449 0.5010 0.3989 0.3886 0.3542
(kW/kg)
a
Low-pressure loop.
b
High-pressure loop.

Table 4
Energy efficiency comparison between previously reported N2-expander processes and the proposed C2N and C3-precooled C2N two-phase expander-based processes.

LNG Processes Liquefaction Specific energy Relative energy savings w.r.t Relative energy savings w.r.t proposed Ref.
(%) consumption (kWh/kmol) proposed C2N process (%) C3eC2N process (%)

N2 single expander 92.00 34.2 79.72 81.52 (Du et al., 2010)

N2 single expander 92.00 14.20 51.16 55.49 (Austbø and
Gundersen, 2015)
N2 single expander 92.00 13.1 47.06 51.75 (Khan et al., 2015a)
N2 double expander 92.00 8.76 20.83 27.85 (Khan et al., 2015a)
C3N two-phase expander 92.00 7.11 2.46 11.11 (Qyyum et al.,
2018d)
Proposed C2N two-phase 92.00 6.935 e 8.85
expander
Proposed C3 precool C2N two- 92.00 6.320 e e
phase expander

As shown in the figure, approximately 75% exergy destruction

_
Exdestruction ¼ ðmÞðEx _
in  Exout Þ  w (5) occurred in the compression section of the N2 single expander
process, whereas that in the compression section of all other pro-
The equation for exergy destruction calculation in the two-
cesses was below 40%. In the inter-stage coolers and expander, the
phase expander and LNG turbine is given in Eq. (6).
N2 dual expander and C3N single expander processes exhibited
_ _ major exergy destruction of more than 20%, whereas the expander
Exdestruction ¼ ðmÞðExin  Exout Þ þ w (6)
section in the N2 dual expander process also exhibited exergy
The equation for exergy destruction calculation in the after- destruction of more than 20%. With respect to the expander section,
coolers is given in Eq. (7). the second-highest exergy destruction was observed in the C3N
single expander process, whereas other processes exhibited less
_
Exdestruction ¼ ðmÞðExin  Exout Þ (7) than 20% exergy destruction. However, in the LNG exchanger sec-
tion, the highest exergy destruction occurred in the proposed
The equation for exergy destruction calculation in the LNG heat
processes, whereas the commercial processes exhibited low exergy
exchanger is given in Eq. (8).
destruction. Similarly, Fig. 5 displays the overall exergy destruction
X X occurring in various processes. The conventional N2 single
Exdestruction ¼ _
ðmÞExin  _
ðmÞExout (8) expander process (Khan et al., 2015a) shows a total exergy
destruction of 2298.52 kW, whereas the proposed C2N two-phase
The exergy destruction analysis were performed for the pro-
expander and C3-precooled C2N two-phase expander-based LNG
posed liquefaction processes in comparison with previous pub-
processes exhibited total exergy destruction values of 1028.52 kW
lished N2 single expander (Khan et al., 2015a), N2 dual expander
and 946.23 kW with 55.25% and 58.83% individual decrease in
(Khan et al., 2015a), and C3N single expander (Qyyum et al., 2018d).
exergy destruction, respectively. These values suggest that energy
Fig. 4 presents the percentage of exergy destruction rate associated
efficiency can be improved in the proposed and commercially
with each equipment of the proposed liquefaction in comparison
available processes either by optimization or by retrofitting the
with the exergy destruction rate (%) associated with the conven-
existing processes.
tional liquefaction processes.
8 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189

the refrigeration cost. Hence, the THCC composite curves play an

important role in the energy efficiency analysis of any liquefaction
process. The high energy consumption in the N2 single expander
process is attributed to the large temperature difference (>3
C)
between 149
C and 30
C in the composite curves as shown in
Fig. 6(a). Comparatively, the C2N process depicts a smaller gap
between the hot and cold composite curves between 116
C
and 67
C representing a lower degree of irreversibility corre-
sponding to the beneficial addition of ethane as a refrigerant.
However, since the peak approach temperature for C2N is
comparatively higher than the N2 expander based process TDCC, it
presents opportunity for further energy enhancement. For this, the
C3-precooled C2N TDCC shows smaller gap along with a peak
approach temperature lower than both Fig. 6(a) and (b). Therefore,
Fig. 6(c) represents a case with lowest entropy generation and
highest energy savings. From a thermodynamic perspective, the
liquefaction process is referred to as an economical process as long
as the THCC curves are kept at the lowest gap possible. Compara-
tively, the approach temperature (TDCC) must be minimal
throughout the cryogenic heat exchanger. The value of MITA must
be within 1
Ce3
C for effective and modest heat transfer (Hasan
et al., 2009). In this context, the value of MITA was kept at 3
C,
Fig. 4. Percentage of exergy destruction rate associated with each equipment of the and the value of the approach temperature in the heat exchanger
proposed and the conventional liquefaction processes.
for the N2 expander process met the desired values at the inlets of
the heat exchanger. Fig. 6(a) illustrates the changes in the MITA
values of this process; the highest MITA value was 50
C. Compar-
atively, in the proposed C2N and C3-precooled C2N processes, as
presented in Fig. 6(c) and (e), respectively, the approach tempera-
ture was satisfied at both ends of the exchanger and for a wide
range of temperatures inside the exchanger, at 70
C to 130
C.
This phenomenon is attributed to the presence of ethane in the
refrigeration cycle. Furthermore, the TDCC curves of the proposed
liquefaction processes in comparison with conventional N2 single
expander process are shown in Fig. 7. Accordingly, for C3-precooled
C2N process, the height of TDCC throughout the heat exchanger’s
length is lower than that of N2 single expander process and the
proposed C2N process. The significant lowering (especially in the
encircled region of Fig. 7) in heights of TDCC curves proves the
overall high energy-efficiency of the proposed liquefaction pro-
cesses as compared to N2 single expander LNG process.
In this study, the gap between the hot and cold composite
curves of the proposed C2N and C3-precooled C2N processes
(Fig. 6(d) and (f)) was reduced by MCD optimization, resulting in an
overall reduction in power consumption of up to 47.83% and 52.45%,
respectively. Moreover, the THCC composite curves of the C2N
system, as depicted in Fig. 6(d), indicate that the process still con-
tains gaps, which implies a further potential for energy savings by
applying highly efficient and rigorous optimization methods.
Fig. 5. Overall exergy destruction observed in the conventional and proposed lique- Fig. 8 shows the THCC and TDCC curves of the precooling
faction processes.
exchanger in the propane refrigeration cycle and the main cryo-
genic heat exchanger. The TDCC curves in Fig. 8(a) show that in the
precooling exchanger, DT is large between the hot and cold com-
3.3. Energy and exergy analysis through composite curves posite curves when compared with the composite curves of the
main LNG exchanger in Fig. 8(c). The gap between the THCC hot and
Fig. 6 compares the conventional N2 expander and proposed cold composite lines in Fig. 8(b) and (d) indicate that the process
processes in terms of composite curve analysis. Fig. 6(a), (b), and (c) has further potential for energy savings not only in the precooling
show the temperature difference between composite curves exchanger but also in the main cryogenic exchanger.
(TDCC) in the main LNG exchanger for the N2 single expander
process and the proposed C2N and C3-precooled C2N processes. 3.4. Economic evaluation
Fig. 6(d), (e), and (f) demonstrate the temperatureeheat flow
composite curves (THCC) in the main LNG exchanger of the N2 The module costing technique was incorporated for economic
single expander and the proposed C2N and C3-precooled C2N evaluation of the proposed configurations for NG liquefaction. This
processes. technique was broadly incorporated for the preliminary cost esti-
The gap between the THCC curves is a measure of the entropy mation of chemical plants (Nagy, 2016; Turton et al., 2008). On the
generation in the cryogenic heat exchanger, which in turn increases basis of this approach, the equipment purchase cost was calculated
 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189 9

Fig. 6. (a), (b), and (c) TDCC composite curves and (d), (e), and (f) THCC composite curves of previously reported N2 single expander processes and the proposed C2N and C3-
precooled C2N processes.

as shown in Eq. (10) (Nagy, 2016; Turton et al., 2008).

CBM ¼ Ep FBM (10)

The total capital investment is given by Eq. (11) (Turton et al.,
2008).

X
n
TCI ¼ 1:18 CBM;i (11)
i

The grassroots cost for establishing a plant from scratch is given

by Eq. (12) (Turton et al., 2008).

X
n
GRC ¼ TCI þ 0:5  CBM;i (12)
i

In this context, the total cost of the compressors, two-phase

expanders, cryogenic heat exchangers, intercoolers, and liquids
Fig. 7. Comparison of the TDCC curves of the proposed and conventional processes. turbines are estimated by using the relations of Turton given above
(Turton et al., 2008). To find the commercial feasibility of the pro-
posed processes, the LNG production capacity was set at 6480 kg/h.
using Eq. (9): The capacity factors of the compressors and expanders can be ob-
  tained from Aspen Hysys®. It should be note that Aspen Hysys®
log10 Ep ¼ k1 þ k2 log10 A þ k3 ðlog10 AÞ2 (9) does not provide the area A of the exchanger, although it provides
the overall heat transfer coefficient UA. Therefore, to find the area
where Ep is the equipment purchase cost, and A is the capacity A, a value of U ¼ 3600 W/m2K was adopted from the literature
parameter, i.e., fluid power (kW) for the compressors, expanders, (Coulson et al., 1985; Kakac et al., 2002; Luyben and Chien, 2011;
and turbines, area (m2) for heat exchangers and inter-coolers. In Turton et al., 2008), which is the mean value of 1200e6000 W/m2K,
addition, k1, k2, and k3 are capacity-based constants specific for as used by (Qyyum and Lee, 2018). The inter-cooler area A is
each equipment piece and are given in Table 5 (Turton et al., 2008). required to estimate costs according to the method prescribed by
The capacity constants for the two-phase expander are assumed to Luyben and Chien (2011). As discussed in the previous section, the
be the same as that of the single expander because of the un- LNG processes are energy intensive owing to the high power con-
availability of cost constants in open literature for this relatively sumption in the compression section. Therefore, the operating cost
new technology. of the compressors is affected by electricity charges. Electricity
After calculating Ep, all other direct and indirect costs related to consumption is calculated by using the total power required by the
the equipment are calculated by using the bare module factor FBM compressors while ignoring the power produced by the expander.
10 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189

Fig. 8. TDCC and THCC curves of the (a) and (c) C3-precooling exchanger and (b) and (d) C2N main cryogenic heat exchanger.

Table 5
Equipment purchase cost constants.

Equipment Type Equipment Description K1 K2 K3

Compressor Centrifugal, axial, reciprocating 2.2897 1.3604 0.1027

Turbine Axial gas turbine 2.7051 1.4398 0.1776
Turbine Liquid expander 2.2476 1.4965 0.1618
Heat exchanger Flat plate 4.6656 0.1557 0.1547
Heat exchanger Air cooler 4.0336 0.2341 0.0497

The cost of electricity, USD 16.8/GJ (Turton et al., 2008), was taken investment along with a five-year payback period to calculate the
from the literature to calculate the operating cost (Eq. (13)). annualized cost. The total annualized cost was estimated by Eq.
(14), as described in previous research (Luyben and Chien, 2011).

 The results of economic analysis of the proposed processes are
$ given in Table 6.
OC ¼ Cost of electricity  ðSpecific compression powerÞ
kW:yr

(13) Capital cost
TAC ¼ þ Operating cost (14)
Payback period
The plant maintenance cost was fixed at 2% of the total capital

Table 6
Economic analysis of the proposed C2N and C3-precooled C2N two-phase expander liquefaction processes.

Cost N2 single expander (Khan N2 dual expander (Khan C3N two-phase expander (Qyyum C2N two-phase C3-precooled C2N two-
et al., 2015a) et al., 2015a) et al., 2018d) expander phase expander

Total Equipment Purchase 4.76 4.84 4.62 4.46 6.35

Cost (106 $)
Total Base Module Cost (106 19.69 19.99 18.98 18.25 25.90
$)
Total Capital Investment 23.23 23.59 22.40 21.53 30.56
(106 $)
Grass root Cost (106 $) 33.08 33.59 31.90 30.66 43.51
Total Operating cost (106 $) 2.81 2.24 1.54 1.35 1.23
Maintenance cost (106 $) 0.46 0.47 0.44 0.43 0.61
Total Annualized Cost (106 7.46 6.96 6.02 5.66 7.34
$)
Relative operating cost _ 20.28 45.19 51.95 56.22
savings (%)
Relative TAC cost savings _ 6.70 19.30 24.12 1.60
(%)
 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189 11

The results of the economic analysis revealed that as the process appeared to influence the work reported in this paper.
complexity increased, the capital cost required also increased.
Furthermore, as the energy efficiency of a process increases, its Acknowledgements
capital cost and the overall operating cost reduce. The TAC is
affected by the total capital investment and operating costs. The N2 This work was supported by the Basic Science Research Program
single expander process exhibited the highest operating and capital through the National Research Foundation of Korea (NRF) funded
costs among all single-cycle processes, with a TAC of US $7.46 by the Ministry of Education (2018R1A2B6001566) and by the
million. Moreover, the proposed C2N process has a TAC of US $5.66 Priority Research Centers Program through the National Research
million owing to its low capital investment and operating costs. Foundation of Korea (NRF) funded by the Ministry of Education
Although the relative operating cost savings show that the pro- (2014R1A6A1031189).
posed C3-precooling process results in 56.22% savings when
compared with the conventional N2 single expander process, the Appendix A. Supplementary data
relative TAC savings show that the proposed C2N process is the
most economical process. For this reason, apart from applying C3- Supplementary data to this article can be found online at
precooling and increasing the overall TAC, the proposed C2N pro- https://doi.org/10.1016/j.jclepro.2019.119189.
cess is the most economical and feasible process.

References
4. Conclusions
Abdul Qyyum, M., Qadeer, K., Lee, M., 2018. Closed-loop self-cooling recuperative
A new binary MR, i.e., ethaneenitrogen, was applied in a two- N2 expander cycle for the energy efficient and ecological natural gas liquefac-
phase expander process to reduce the overall operating costs of tion process. ACS Sustain. Chem. Eng. 6, 5021e5033. https://doi.org/10.1021/
acssuschemeng.7b04679.
nitrogen expander-based liquefaction processes for LNG-FPSO Aspelund, A., Berstad, D.O., Gundersen, T., 2007. An Extended Pinch Analysis and
projects. A C3-precooling refrigeration cycle was also integrated in Design procedure utilizing pressure based exergy for subambient cooling. Appl.
the two-phase expander process to further reduce the operating Therm. Eng. 27, 2633e2649. https://doi.org/10.1016/
j.applthermaleng.2007.04.017.
costs at the expense of an increase in capital costs. The MCD al- Austbø, B., Gundersen, T., 2015. Optimization of a single expander LNG process.
gorithm was applied to realize the maximum benefits from the Energy Procedia 64, 63e72. https://doi.org/10.1016/j.egypro.2015.01.009.
proposed improvements. The proposed enhancements significantly Cao, W., Lu, X., Lin, W., Gu, A., 2006. Parameter comparison of two small-scale
natural gas liquefaction processes in skid-mounted packages. Appl. Therm.
reduced the energy consumption for LNG production, which ulti-
Eng. 26, 898e904. https://doi.org/10.1016/j.applthermaleng.2005.09.014.
mately led to reductions in the operating cost and the total annu- Coulson, J.M., Richardson, J.E., Sinnott, R.K., 1985. Heat transfer equipment. Chem.
alized cost. The following conclusions were drawn from the current Eng. 6, 511e587.
study. Ding, H., Sun, H., He, M., 2016. Optimisation of expansion liquefaction processes
using mixed refrigerant N2eCH4. Appl. Therm. Eng. 93, 1053e1060. https://
doi.org/10.1016/j.applthermaleng.2015.10.004.
 When compared with the N2 single expander process, the pro- Du, H.P., Huang, Y.D., Li, H.Y., Ying, Q.S., Fan, Q.H., Jia, L.X., 2010. Numerical simu-
posed C2N-based and C3-precooled C2N processes can produce lation and optimization of small-scale LNG plant for skid mounted. In: Power
and Energy Engineering Conference (APPEEC), 2010 Asia-Pacific. IEEE, pp. 1e4.
LNG with significant energy savings (79.72% and 81.52%), Dutta, A., Karimi, I.A., Farooq, S., 2018. Economic feasibility of power generation by
depending on the feed conditions, composition, and design recovering cold energy during LNG (liquefied natural gas) regasification. ACS
parameters. Sustain. Chem. Eng. 6, 10687e10695. https://doi.org/10.1021/
acssuschemeng.8b02020.
 The relative TAC savings were 24.12% for the proposed C2N ExxonMobil, 2014. The outlook for energy: a view to 2040, energy outlook. .
process and 1.60% for the proposed C3-precooled C2N process. Gao, T., Lin, W., Gu, A., Gu, M., 2010. Coalbed methane liquefaction adopting a ni-
Hence, it is concluded that the proposed C2N process is the most trogen expansion process with propane pre-cooling. Appl. Energy 87,
2142e2147. https://doi.org/10.1016/j.apenergy.2009.12.010.
optimal, in terms of economic efficiency, because its TAC savings
Ghaedi, H., Ayoub, M., Sufian, S., Murshid, G., Farrukh, S., Shariff, A.M., 2017.
were much higher than those realized by the propane- Investigation of Various Process Parameters on the Solubility of Carbon Dioxide
precooled C2N two-phase expander and the single and dual in Phosphonium-Based Deep Eutectic Solvents and Their Aqueous Mixtures:
Experimental and Modeling.
nitrogen-based and C3N two-phase expander liquefaction
Haider, J., Qyyum, M.A., Kazmi, B., Zahoor, M., Lee, M., 2019. Simulation study of
processes. biomethane liquefaction followed by biogas upgrading using an imidazolium-
 The exergy analysis revealed that the process performance can based cationic ionic liquid. J. Clean. Prod. 231, 953e962. https://doi.org/
be further enhanced by using a rigorous optimization technique 10.1016/j.jclepro.2019.05.252.
Hasan, M.M.F., Karimi, I.A., Alfadala, H.E., Grootjans, H., 2009. Operational modeling
or by improving the refrigeration cycle. of multistream heat exchangers with phase changes. AIChE J. 55, 150e171.
https://doi.org/10.1002/aic.11682.
The major limitation of the proposed liquefaction processes is He, T.B., Ju, Y.L., 2014. Performance improvement of nitrogen expansion liquefaction
process for small-scale LNG plant. Cryogenics (Guildf) 61, 111e119. https://
the stability and reliability of the two-phase expander. Therefore, it doi.org/10.1016/j.cryogenics.2013.09.004.
is mandatory to evaluate the robustness of the proposed processes. Horvath, S., Fasihi, M., Breyer, C., 2018. Techno-economic analysis of a decarbonized
Furthermore, detailed thermodynamic evaluation in terms of shipping sector: technology suggestions for a fleet in 2030 and 2040. Energy
Convers. Manag. 164, 230e241. https://doi.org/10.1016/
advanced exergy analysis can also be performed to evaluate the J.ENCONMAN.2018.02.098.
real-potential improvements to make the feasible process for Hu, H., Ding, C., Ding, G., Chen, J., Mi, X., Yu, S., 2019. Heat transfer characteristics of
commercialization. The proposed LNG processes can be further two-phase mixed hydrocarbon refrigerants flow boiling in shell side of LNG
spiral wound heat exchanger. Int. J. Heat Mass Transf. 131, 611e622. https://
improved by investigating and employing other precooling re- doi.org/10.1016/j.ijheatmasstransfer.2018.11.106.
frigerants (such as CO2, HFO-1234yf, and NH3) or introducing in- Hu, H., Yang, G., Ding, G., Chen, J., Yang, W., Hu, S., 2018. Heat transfer characteristics
termediate cooling step in order to reduce the overall entropy of mixed hydrocarbon refrigerant flow condensation in shell side of helically
baffled shell-and-tube heat exchanger. Appl. Therm. Eng. 133, 785e796. https://
generation.
doi.org/10.1016/j.applthermaleng.2018.01.083.
Kakac, S., Liu, H., Pramuanjaroenkij, A., 2002. Heat Exchangers: Selection, Rating,
Declaration of competing interest and Thermal Design. CRC press.
Khan, M.S., Lee, S., Getu, M., Lee, M., 2015a. Knowledge inspired investigation of
selected parameters on energy consumption in nitrogen single and dual
The authors declare that they have no known competing expander processes of natural gas liquefaction. J. Nat. Gas Sci. Eng. 23, 324e337.
financial interests or personal relationships that could have https://doi.org/10.1016/j.jngse.2015.02.008.
12 M.A. Qyyum et al. / Journal of Cleaner Production 248 (2020) 119189

Khan, M.S., Lee, S., Getu, M., Lee, M., 2015b. Knowledge inspired investigation of Qyyum, M.A., Ali, W., Long, N.V.D., Khan, M.S., Lee, M., 2018a. Energy efficiency
selected parameters on energy consumption in nitrogen single and dual enhancement of a single mixed refrigerant LNG process using a novel hydraulic
expander processes of natural gas liquefaction. J. Nat. Gas Sci. Eng. 23, 324e337. turbine. Energy 144. https://doi.org/10.1016/j.energy.2017.12.084.
https://doi.org/10.1016/j.jngse.2015.02.008. Qyyum, M.A., Chaniago, Y.D., Ali, W., Qadeer, K., Lee, M., 2019a. Coal to clean energy:
Khan, M.S., Lee, S., Hasan, M., Lee, M., 2014. Process knowledge based opportunistic energy-efficient single-loop mixed-refrigerant-based schemes for the lique-
optimization of the N2eCO2 expander cycle for the economic development of faction of synthetic natural gas. J. Clean. Prod. 211, 574e589. https://doi.org/
stranded offshore fields. J. Nat. Gas Sci. Eng. 18, 263e273. https://doi.org/ 10.1016/j.jclepro.2018.11.233.
10.1016/j.jngse.2014.03.004. Qyyum, M.A., Lee, M., 2018. Hydrofluoroolefin-based novel mixed refrigerant for
Khan, M.S., Lee, S., Rangaiah, G.P., Lee, M., 2013. Knowledge based decision making energy efficient and ecological LNG production. Energy 157, 483e492. https://
method for the selection of mixed refrigerant systems for energy efficient LNG doi.org/10.1016/j.energy.2018.05.173.
processes. Appl. Energy 111, 1018e1031. https://doi.org/10.1016/ Qyyum, M.A., Minh, L.Q., Ali, W., Hussain, A., Bahadori, A., Lee, M., 2018b. Feasibility
j.apenergy.2013.06.010. study of environmental relative humidity through the thermodynamic effects
Kuwahara, N., Bajay, S., Castro, L., 2000. Liquefied natural gas supply optimisation. on the performance of natural gas liquefaction process. Appl. Therm. Eng. 128
Energy Convers. Manag. 41, 153e161. https://doi.org/10.1016/S0196-8904(99) https://doi.org/10.1016/j.applthermaleng.2017.08.090.
00105-3. Qyyum, M.A., Qadeer, K., Lee, M., 2018c. Comprehensive review of the design
Li, C., Jia, W., Wu, X., 2012. Application of Lee-Kesler equation of state to calculating optimization of natural gas liquefaction processes: current status and per-
compressibility factors of high pressure condensate gas. Energy Procedia 14, spectives. Ind. Eng. Chem. Res. 57, 5819e5844. https://doi.org/10.1021/
115e120. https://doi.org/10.1016/j.egypro.2011.12.904. acs.iecr.7b03630.
Lim, W., Choi, K., Moon, I., 2013. Current status and perspectives of liquefied natural Qyyum, M.A., Qadeer, K., Lee, S., Lee, M., 2018d. Innovative propane-nitrogen two-
gas (LNG) plant design. Ind. Eng. Chem. Res. 52, 3065e3088. https://doi.org/ phase expander refrigeration cycle for energy-efficient and low-global warming
10.1021/ie302877g. potential LNG production. Appl. Therm. Eng. 139 https://doi.org/10.1016/
Long, N.V.D., Minh, L.Q., Pham, T.N., Bahadori, A., Lee, M., 2016. Novel retrofit de- j.applthermaleng.2018.04.105.
signs using a modified coordinate descent methodology for improving energy Qyyum, M.A., Qadeer, K., Minh, L.Q., Haider, J., Lee, M., 2019b. Nitrogen self-
efficiency of natural gas liquid fractionation process. J. Nat. Gas Sci. Eng. 33, recuperation expansion-based process for offshore coproduction of liquefied
458e468. https://doi.org/10.1016/j.jngse.2016.05.038. natural gas, liquefied petroleum gas, and pentane plus. Appl. Energy 235,
Luyben, W.L., Chien, I.-L., 2011. Design and Control of Distillation Systems for 247e257. https://doi.org/10.1016/j.apenergy.2018.10.127.
Separating Azeotropes. John Wiley & Sons. Scripps Institution of Ocenography, 2017. 2017. Concentration of CO2.
Mehrpooya, M., Ansarinasab, H., 2015. Exergoeconomic evaluation of single mixed Shah, N.M., Hoadley, A.F.A., Rangaiah, G.P., 2009. Inherent safety analysis of a pro-
refrigerant natural gas liquefaction processes. Energy Convers. Manag. 99, pane precooled gas-phase liquified natural gas process. Ind. Eng. Chem. Res. 48,
400e413. https://doi.org/10.1016/J.ENCONMAN.2015.04.038. 4917e4927. https://doi.org/10.1021/ie8015939.
Mehrpooya, M., Ghorbani, B., 2018. Introducing a hybrid oxy-fuel power generation Strantzali, E., Aravossis, K., Livanos, G.A., Nikoloudis, C., 2019. A decision support
and natural gas/carbon dioxide liquefaction process with thermodynamic and approach for evaluating liquefied natural gas supply options: implementation
economic analysis. J. Clean. Prod. 204, 1016e1033. https://doi.org/10.1016/ on Greek case study. J. Clean. Prod. 222, 414e423. https://doi.org/10.1016/
j.jclepro.2018.09.007. j.jclepro.2019.03.031.
Mortazavi, A., Somers, C., Hwang, Y., Radermacher, R., Rodgers, P., Al-Hashimi, S., Tsatsaronis, G., Morosuk, T., 2012. Advanced thermodynamic (exergetic) analysis.
2012. Performance enhancement of propane pre-cooled mixed refrigerant LNG J. Phys. Conf. Ser. 395 https://doi.org/10.1088/1742-6596/395/1/012160.
plant. Appl. Energy 93, 125e131. https://doi.org/10.1016/j.apenergy.2011.05.009. Turton, R., Bailie, R.C., Whiting, W.B., Shaeiwitz, J.A., 2008. Analysis, Synthesis and
Nagy, M., 2016. Techno-economic Analysis of LNG Production Alternatives. Design of Chemical Processes. Pearson Education.
Pachauri, R.K., Reisinger, A., 2007. Synthesis report. Fifth Assess. Rep. Intergov. Panel Vatani, A., Mehrpooya, M., Palizdar, A., 2014. Advanced exergetic analysis of five
Clim. Chang. 151e165. natural gas liquefaction processes. Energy Convers. Manag. 78, 720e737.
Park, J.H., Khan, M.S., Lee, M., 2015. Modified coordinate descent methodology for https://doi.org/10.1016/j.enconman.2013.11.050.
solving process design optimization problems: application to natural gas plant. Venkatarathnam, G., Timmerhaus, K.D., 2008. Cryogenic Mixed Refrigerant
J. Nat. Gas Sci. Eng. 27 (Part 1), 32e41. https://doi.org/10.1016/ Processes.
j.jngse.2014.10.014. Yin, Q.S., Li, H.Y., Fan, Q.H., Jia, L.X., 2008. Economic analysis of mixed-refrigerant
Pham, T.N., Long, N.V.D., Lee, S., Lee, M., 2017. Enhancement of single mixed cycle and nitrogen expander cycle in small scale natural gas liquefier. In: AIP
refrigerant natural gas liquefaction process through process knowledge Conference Proceedings. AIP, pp. 1159e1165.
inspired optimization and modification. Appl. Therm. Eng. 110, 1230e1239. Yuan, Z., Cui, M., Song, R., Xie, Y., 2015. Evaluation of prediction models for the
https://doi.org/10.1016/j.applthermaleng.2016.09.043. physical parameters in natural gas liquefaction processes. J. Nat. Gas Sci. Eng. 27
Press Release December 2012., T., 2012. Shell and the Technip Samsung Consortium (Part 2), 876e886. https://doi.org/10.1016/j.jngse.2015.09.042.
Sign Agreement to Strengthen Floating LNG Collaboration. Yuan, Z., Cui, M., Xie, Y., Li, C., 2014. Design and analysis of a small-scale natural gas
Qadeer, K., Qyyum, M.A., Lee, M., 2018. Krill-herd-based investigation for energy liquefaction process adopting single nitrogen expansion with carbon dioxide
saving opportunities in offshore liquefied natural gas processes. Ind. Eng. Chem. pre-cooling. Appl. Therm. Eng. 64, 139e146. https://doi.org/10.1016/
Res. 57, 14162e14172. https://doi.org/10.1021/acs.iecr.8b02616. j.applthermaleng.2013.12.011.