=GUDYNR.

UDYDQMD0LORã%RJDWDM(GLWRUV3URFHHGLQJVRIWKHWK(XURSHDQ6\PSRVLXPRQ

&RPSXWHU$LGHG3URFHVV(QJLQHHULQJ±(6&$3(
-XQHWKWK3RUWRURå6ORYHQLD‹(OVHYLHU%9$OOULJKWVUHVHUYHG
http://dx.doi.org/10.1016/B978-0-444-63428-3.50206-X

Natural gas liquids (NGL) recovery in the liquefied

natural gas production
Mengyu Wang, Ali Abbas*
School of Chemical and Biomolecular Engineering, The University of Sydney, NSW
2006, Australia. *ali.abbas@sydney.edu.au
Abstract
This paper focuses on the energy benefits of two configurations of natural gas liquids
(NGL) recovery and LNG liquefaction under various feed gas and operating conditions.
Simulation results show that specific power consumption is affected by the process
configurations and feed conditions. An integrated NGL recovery within natural gas
(NG) liquefaction unit is found to be the most energy efficient configuration while a
front-end NGL recovery unit is the least. As methane concentration of feed gas
decreases, an integrated NGL recovery within the liquefaction unit has an increase of
0.74% in energy consumption while a decrease of 0.18% of the same is observed for a
front-end NGL recovery unit. When the methane concentration in natural gas is greater
than 95%, both processes have almost the same specific OPEX.
Keywords: LNG, NGL recovery, modelling, sensitivity analysis, feed gas conditions
1. Introduction
The NG liquefaction process is the heart of the LNG value chain however it suffers
from being a very energy intensive process. Lim et al. (2012) reviewed the current NG
liquefaction technologies and optimization studies for performance improvement. The
efficient design and operation of liquefaction plants also depend on the variations in
upstream gas conditions (feed compositions, pressure and temperature), downstream
conditions as well as process conditions (Mokhatab and Poe, 2012). Raw natural gas at
the wellhead commonly contains methane as the primary gas but may also contain small
amounts of heavier hydrocarbons, and other contaminants such as water, acid gases
(CO2 and H2S) and mercury. Many licensed technologies can be independently selected
to remove these contaminants prior to liquefaction in order to avoid blockages and
damage to process equipment and to meet pipeline specifications (Kidnay and Parrish,
2006).
In addition, the heavier of hydrocarbons may be required to be recovered as natural gas
liquids (NGL). The NGL recovery has ability to lower the heating value of the LNG
product and to minimize the impact of gas composition variation on process operation
over time. Getu et al. (2013) compared the economic performance of different NGL
recovery schemes under a range of feed compositions known as lean and rich feeds.
Khan et al. (2014) represented the benefits of three proposed integrated schemes of
NGL recovery with NG liquefaction. Park et al. (2014) evaluated a proposed
configuration of NGL recovery process and nine patented schemes for offshore
applications. There are various patented configurations and methods to extract the NGL
from a natural gas stream. However, few studies have addressed and compared the
performance of different NGL recovery and NG liquefaction process under various feed
and operating conditions. There are two configurations of NGL recovery and NG
1208 M. Wang and A. Abbas

liquefaction: a front-end NGL recovery unit in the upstream of the LNG value chain and
a NGL recovery integrated within liquefaction unit.
The aim of this study is to understand the benefits of a front-end NGL recovery and a
NGL recovery integrated within liquefaction unit under different feed gas conditions
through model-based sensitivity analysis. The aforementioned two configurations use a
propane precooled mixed refrigerant process (C3MR) and a patented NGL recovery
studied by Mak and Graham (2013) as the basis.

2. Process description
Figure 1 shows the flow diagram of a front-end NGL recovery and C3MR process (a
front-end NGL/C3MR). A portion of the vapor is condensed against the column
overhead stream and the remaining is expanded through a turbo-expander before
entering the scrub column to remove the heavier hydrocarbons. The column overhead
stream is recompressed and then cooled through a refrigeration section.
C3MR consists of a precooling cycle and a subcooling cycle. The treated natural gas
and MR is partially cooled to around -35°C in a propane precooling cycle. The
precooled MR is then separated into gaseous stream and liquid stream to provide the
cooling for natural gas in main cryogenic heat exchanger (MCHE) to around -161°C at
atmospheric pressure. After that, MR is completely vaporized from the exit of MCHE
and then compressed by a series of compressors to its initial inlet conditions.
Figure 2 show the flow diagram of an integrated NGL recovery within C3MR process
(an integrated NGL/C3MR). It is an alternative configuration of Figure 1 which NGL
recovery is located at an intermediate location. Feed gas is precooled in a propane
refrigerant cycle and then on to NGL recovery. The column overhead stream is liquefied
and subcooled in the MCHE to produce LNG.

S301
E301 E302 E303 HX204 HX205 HX206

C301 C302 C303 LNG101

V204 V205
V206 V301
T203 T204
T200 V203
V202 S302
Front-end NGL V201 T202
MCHE

T201

LNG102
T401 HX401
Natural Gas HX201 HX202 HX203
V401
V302
C403

E201 LNG103
MIX203
MIX202
R401 MIX201
S401
Flare
C401 C402 V303
S101
C203 C201 V101
C202
LNG
NGL product

Figure 1 A front-end NGL recovery and C3MR process.

Natural gas liquids (NGL) recovery in the liquefied natural gas production 1209

S301
E301 E302 E303 HX204 HX205 HX206

C301 C302 C303 LNG101

CB301
V204 V205
V206 V301
T203 T204
T200 V203 NGL Extraction
V202 S302
V201

MCHE
T201 T202

T401 HX401
LNG102
Natural Gas
V401
HX201 HX202 HX203 C403
V302

P-162
R401
S401 LNG103
C401 C402
E201 MIX203 Flare
MIX202
MIX201 V303
S101
V101
LNG
C203 C202 C201
NGL product

Figure 2 An integration of NGL recovery within C3MR process.

3. Main assumptions and performance equations

The NGL/C3MR processes are simulated under the same feed gas conditions and
assumptions. Aspen HYSYS® was used for the selected configurations and assesse their
performance.
3.1. Assumptions
1) The feed natural gas is dry and sweet prior to entering the liquefaction process.
2) Natural gas is composed of 96.92% methane, 2.94% ethane, 0.06% propane, 0.01%
n-butane and 0.07% nitrogen.
3) It is fed at T = 25°C and P = 5000kPa to produce 3.0MTPA of LNG.
4) The flowrate variation of feed gas is within the range of 20% of its design value.
5) The gas splitting ratio to the column bottom stream (T401 to C401) is in a range of
0.4 and 0.65.
6) The overall heat transfer coefficient and area (UA) of MCHE is fixed at design value
of 16.39MW/°C.
7) The prices of electricity and cooling water are $10.99/GJ and $0.40/GJ respectively
(Turton et al., 2009). The price of feed natural gas is $2/mmBTU.
8) The operating conditions used for simulating NGL recovery are listed in Table 1.
3.2. Performance equations
The total power consumption and variable operating cost of the NGL/C3MR process are
calculated using the following equations. Total shaft work is represented by Eq. (1) in
which Wi is the shaft work used at compressor Ci.
Table 1. Column operating conditions.
Operating conditions Value Unit
Total number of trays 40
Column top stream pressure 2689 kPa
Column bottom stream pressure 2758 kPa
1210 M. Wang and A. Abbas

Table 2. Feed natural gas composition (in unit of mole fraction).

Feed composition case Methane Ethane Propane n-butane Nitrogen
1
Base case 96.92 2.94 0.06 0.01 0.07
NG1 96.70 3.15 0.06 0.01 0.07
NG2 96.31 3.53 0.07 0.01 0.08
NG3 94.80 4.96 0.10 0.02 0.12
NG4 91.63 7.99 0.16 0.03 0.19
NG5 84.55 14.75 0.30 0.05 0.35
1
Refer to Wang et al. (2013)
n
WTotal ¦W
i 1
i (1)

Operating expenditure (OPEX) includes electricity, water and feed natural gas. It is
represented as Eq. (2).
n m
PElec ¦Wi  PWater ¦ Q j  PNG FNG
i 1 j 1 (2)
OPEX
FLNG
where, PElec is electricity price, PWater is cooling water price, PNG is the price of feed
natural gas.
4. Results and discussion
Three scenarios are carried out to explain the differences between two configurations
under different feed (feed flowrate and feed composition) and operating conditions (gas
splitting ratio to the column bottom stream). For comparison, they are simulated using
the same feed gas conditions, refrigerant mixture, operating conditions, and LNG
production. For the process configurations, they use the same configuration basis. The
main difference is the location of NGL recovery within NG liquefaction plant.
4.1. Scenario 1: Effect of different feed gas compositions
Figure 3 shows the effect of the supplying natural gas with different compositions on
specific shaft work and specific OPEX. It can be seen in Table 2 that six different
compositions of natural gas are conducted. The mole fraction of methane in natural gas
is adjusted from its base value of 96.92% to 84.55% with 10% decrements of
component mole flow of methane while keeping the mole flow of remaining
components constant. For a given amount of feed gas, the decrease of 12.37 mol% in
methane concentration results in the reduction of specific shaft work by 0.175% for a
front-end NGL/C3MR, in contrast with 0.74% increase for an integrated NGL/C3MR.
However, a front-end NGL/C3MR obtained 38.3% savings in specific OPEX and
21.4% for an integrated NGL/C3MR. When the methane concentration in natural gas is
greater than 95%, both processes have the almost same specific OPEX. In summary,
both processes will be more preferable to operate at processing natural gas with high
methane content. By contrast, an integrated NGL/C3MR will be more preferable to gas
with low methane content.
Natural gas liquids (NGL) recovery in the liquefied natural gas production 1211

2460 200

Specific shaftwork (MW/t-LNG)

190

Specific OPEX ($/t-LNG)

2450
180

2440
170

160
2430

150
2420
Integrated NGL/C3MR 140
Front-end NGL/C3MR Integrated NGL/C3MR
Front-end NGL/C3MR
2410 130
0.80 0.85 0.90 0.95 1.00 0.80 0.85 0.90 0.95 1.00

Mole fraction of methane Mole fraction of methane

Figure 3 The effect of feed gas composition variation on specific shaftwork and specific OPEX.
4.2. Scenario 2: Effect of feed gas flowrate variation
It is found in Figure 4 that overall power consumption decreases at fixed UA, as natural
gas flowrate decreases from 3.990×105kg/h to 3.129×105kg/h. The integrated
NGL/C3MR requires 1.26% less specific power consumption than a front-end
NGL/C3MR. However, specific OPEX increases from $138.5/t-LNG to $166.0/t-LNG
for an integrated NGL/C3MR and from $138.8/t-LNG to $166.4/t-LNG for a front-end
NGL/C3MR since natural gas flowrate decreases at fixed UA. It is noted that it is
slightly less than specific OPEX of a front-end NGL. The decline in natural gas flowrate
with a constant composition requires 19.9% less in specific OPEX for both processes.
4.3. Scenario 3: Effect of splitting feed gas flow
Figure 5 shows the effect of gas splitting ratio to the column bottom stream (T401 to
C401) on specific shaft work and specific OPEX. This variation has a trivial effect on
specific power consumption and specific OPEX for both configurations. The specific
power remains constant for a front-end NGL/C3MR and an integrated NGL/C3MR
when gas flow ratio increases. It is noted that an integrated NGL/C3MR is more
efficient in energy consumption which is 1.25% less than a front-end NGL/C3MR to
produce the same amount of LNG.

5. Conclusion
This study concludes that the variations in feed gas composition and flowrate have more
refrigeration effect than gas splitting ratio for a NGL/C3MR plant (i.e. with a constant
UA). With reduction in feed gas composition over time, the specific shaft work of a
front-end NGL/C3MR declines while specific shaft work of an integrated NGL/C3MR
increases. However, the two processes are favorable to remove the heavier
hydrocarbons from natural gas at high methane feed content.
1212 M. Wang and A. Abbas

2460 170

Specific shaftwork (MW/t-LNG)

165

Specific OPEX ($/t-LNG)

2450
160

2440
155

Integrated NGL/C3MR 150

2430 Front-end NGL/C3MR
145
2420
140 Integrated NGL/C3MR
Front-end NGL/C3MR
2410 135
3.0e+5 3.2e+5 3.4e+5 3.6e+5 3.8e+5 4.0e+5 3.0e+5 3.2e+5 3.4e+5 3.6e+5 3.8e+5 4.0e+5

Feed gas flowrate (kg/h) Feed gas flowrate (kg/h)

Figure 4 Effect of feed gas flowrate variation on specific shaftwork and specific OPEX.
2460 140.0
Specific shaftwork (MW/t-LNG)

Specific OPEX ($/t-LNG)

2450 Integrated NGL/C3MR
139.5
Front-end NGL/C3MR

2440
Integrated NGL/C3MR
Front-end NGL/C3MR 139.0
2430

138.5
2420

2410 138.0
0.40 0.45 0.50 0.55 0.60 0.65 0.40 0.45 0.50 0.55 0.60 0.65

Flow ratio Flow ratio

Figure 5 Effect of gas splitting ratio variation on specific shaftwork and specific OPEX.
References
GETU, M., MAHADZIR, S., LONG, N. V. D. & LEE, M. 2013. Techno-economic analysis of
potential natural gas liquid (NGL) recovery processes under variations of feed
compositions. Chemical Engineering Research and Design, 91, 1272-1283.
KHAN, M. S., CHANIAGO, Y. D., GETU, M. & LEE, M. 2014. Energy saving opportunities in
integrated NGL/LNG schemes exploiting: Thermal-coupling common-utilities and
process knowledge. Chemical Eng. & Processing: Process Intensification, 82, 54-64.
KIDNAY, A. J. & PARRISH, W. R. 2006. Fundamentals of Natural Gas Processing, CRC Press.
LIM, W., CHOI, K. & MOON, I. 2012. Current Status and Perspectives of Liquefied Natural Gas
(LNG) Plant Design. Industrial & Engineering Chemistry Research, 52, 3065-3088.
MAK, J. & GRAHAM, C. 2013. Configurations and methods of integrated ngl recovery and lng
liquefaction. Google Patents.
MOKHATAB, S. & POE, W. A. 2012. Handbook of Natural Gas Transmission and Processing,
Elsevier Science.
PARK, J. H., KHAN, M. S., ANDIKA, R., GETU, M., BAHADORI, A. & LEE, M. 2014.
Techno-economic evaluation of a novel NGL recovery scheme with nine patented
schemes for offshore applications. Journal of Natural Gas Science and Engineering.
TURTON, R., BAILIE, R. C. & WHITING, W. B. 2009. Analysis, synthesis, and design of
chemical processes, Prentice Hall.
WANG, M., KHALILPOUR, R. & ABBAS, A. 2013. Operation optimization of propane
precooled mixed refrigerant processes. J. of Natural Gas Science and Engineering, 15,
93-105.