Available online at www.sciencedirect.

com

ScienceDirect
Availableonline
Available onlineatatwww.sciencedirect.com
www.sciencedirect.com
Energy Procedia 00 (2017) 000–000

ScienceDirect
ScienceDirect
www.elsevier.com/locate/procedia

Energy
EnergyProcedia
Procedia136 (2017) 000–000
00 (2017) 336–341
www.elsevier.com/locate/procedia

4th International Conference on Energy and Environment Research, ICEER 2017, 17-20 July
2017, Porto, Portugal

Optimization-based approach
The 15th International for design
Symposium andHeating
on District integration of carbon
and Cooling
dioxide separation processes using membrane technology
Assessing the feasibility of using the heat demand-outdoor
temperature function
Minsoofor a Sunghoon
Kim, long-term district
Kim, heat* demand forecast
Jiyong Kim
Department of Energy & Chemical Engineering, Incheon National University, 119 Academy-ro, Yeonsu-gu, Incheon 407-772, Republic of Korea
I. Andrića,b,c*, A. Pinaa, P. Ferrãoa, J. Fournierb., B. Lacarrièrec, O. Le Correc
a
IN+ Center for Innovation, Technology and Policy Research - Instituto Superior Técnico, Av. Rovisco Pais 1, 1049-001 Lisbon, Portugal
b
Abstract Veolia Recherche & Innovation, 291 Avenue Dreyfous Daniel, 78520 Limay, France
c
Département Systèmes Énergétiques et Environnement - IMT Atlantique, 4 rue Alfred Kastler, 44300 Nantes, France
The purpose of this paper is to examine carbon dioxide (CO2) separation process using membrane technologies. In achieving this
goal, we develop an optimization model to design and integrate the CO 2 separation using multi-stage membrane technologies by
maximizing the total profit calculated with the total capital and operating costs, and the expected revenue from the CH 4 selling.
Abstract
Using the proposed model, we are able to determine the optimal design and operation strategies of the individual membrane and
compressor to meet the design specification (i.e., purity and recovery rate) in the scenarios of different CO2 market prices.
District heating networks are commonly addressed in the literature as one of the most effective solutions for decreasing the
greenhouse gas emissions from the building sector. These systems require high investments which are returned through the heat
© 2017 The Authors. Published by Elsevier Ltd.
sales. Due to the changed climate conditions and building renovation policies, heat demand in the future could decrease,
Peer-review under responsibility of the scientific committee of the 4th International Conference on Energy and Environment
prolonging the investment return period.
Research.
The main scope of this paper is to assess the feasibility of using the heat demand – outdoor temperature function for heat demand
forecast. The district of Alvalade, located in Lisbon (Portugal), was used as a case study. The district is consisted of 665
Keywords: CO2 separation; coal bed methane; membrane process; optimization; process synthesis
buildings that vary in both construction period and typology. Three weather scenarios (low, medium, high) and three district
renovation scenarios were developed (shallow, intermediate, deep). To estimate the error, obtained heat demand values were
1.compared with results from a dynamic heat demand model, previously developed and validated by the authors.
Introduction
The results showed that when only weather change is considered, the margin of error could be acceptable for some applications
(the error in annual demand was lower than 20% for all weather scenarios considered). However, after introducing renovation
The depletion
scenarios, the errorof conventional
value increased upresources, in particular
to 59.5% (depending fossil
on the fuelsand
weather such as coalscenarios
renovation and petroleum, leads
combination to the
considered).
development of unconventional energies to meet a significant increase of energy demand [1]. Coal
The value of slope coefficient increased on average within the range of 3.8% up to 8% per decade, that corresponds to the bed methane
(CBM)
decreasegas is anumber
in the kind ofof new promising
heating hours of alternative energy
22-139h during resources
the heating due(depending
season to its huge on potential; its total
the combination reserve and
of weather is
estimated
renovationtoscenarios
be 36%considered).
of the totalOnnatural gas hand,
the other reserve [2]. CBM
function gas increased
intercept is usuallyforproduced
7.8-12.7% byper
pressure
decade depletion
(dependinginonthe
the
coal reservoir.
coupled However,
scenarios). this conventional
The values suggested couldpressure
be useddepletion
to modifymethod is proper
the function only tofor
parameters produce a limited
the scenarios amount of
considered, and
improve
gas the accuracy
in place, of heat 20-60%
which ranges demand estimations.
[3]. Thus, many studies have revealed that injecting carbon dioxide CO2 into a

© 2017 The Authors. Published by Elsevier Ltd.

Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and
Cooling.
* Corresponding author. Tel.: +82-32-835-8875; fax: +82-32-835-0797.
E-mail address: jykim77@incheon.ac.kr (J. Kim)
Keywords: Heat demand; Forecast; Climate change
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 4th International Conference on Energy and Environment
Research.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the Scientific Committee of The 15th International Symposium on District Heating and Cooling.
1876-6102 © 2017 The Authors. Published by Elsevier Ltd.
Peer-review under responsibility of the scientific committee of the 4th International Conference on Energy and Environment
Research.
10.1016/j.egypro.2017.10.284
 Minsoo Kim et al. / Energy Procedia 136 (2017) 336–341 337
Jiyong Kim et al. / Energy Procedia 00 (2017) 000–000

coal seam can significantly improve the CBM production since the CO 2 has a stronger affinity with coals than other
components in coal bed. The enhanced coalbed methane (ECBM) production through CO2 injection into unminable
coal seams is to be an efficient method to improve the productivity of CBM.
CBM gas from the CO2 injected ECBM consists primarily of methane (CH4) and CO2, and remained components
such as water and nitrogen (N2) [4]. However, the amount of CO2 contents in CBM gas varies from 30% to 90%
depending on conditions of the coal seam and operations [5]. Thus, before transporting the CBM gas as a final
energy form to customers (e.g., market or terminal), a low-quality CBM gas must be purified to meet the
specification of the pipeline transportation, for example, the purity of CH 4 component in the final stream should be
improved at least up to 98% by eliminating the other main component, CO2 [6].
Compared with conventional methods to separate CO2, such as cryogenic distillation and amine-based wet
scrubbing, membrane-based separation is a competitive technology since it offers many advantages of i) relative
effective energy consumption per the unit of CO2 separation, ii) simple operation and maintenance, and iii)
relatively easy scale-up compared to other separation processes [7,8]. In general, a single stage membrane process is
not economical to separate CO2 in a high purity with competent recovery rate [9]. Accordingly, a multi stage
membrane process, which connects two more membrane processes in series, is widely necessary to improve both the
separation performance and economics [10,11]. However, the multi stage membrane separation process needs also
to be optimized to ensure high energy efficiency and separation efficiency. In most cases, the optimization is
performed not by any heuristic methods but by mathematical models due to high complexity in the interaction
between membrane modules and with recycles [12,13].
Therefore, this study aims to design and analyze CO2 separation system using the multiple membrane process for
the CO2 injected ECBM application. In achieving this goal, we develop an optimization model using a nonlinear
programming (NLP) technique. The optimization approach is applied to CO2/CH4 separation in the CO2-ECBM
production process, which maximizes the total expected profit. With the proposed model, we are able to determine
the optimal design and operation conditions such as required membrane area and utility consumption to meet the
design specification. Finally, we perform extended analyses to investigate the impacts of CO2 content in the feed on
the economics and design of the membrane-based separation process.

2. Membrane-based CO2 separation system

2.1. Membrane process

Typically a membrane module consists of porous layers that allows a particular molecule to permeate faster than
the other molecules. Molecules with lower size (or, higher diffusivity and solubility) in the membrane module can
permeate faster. Thus, CO2 with lower size can permeate faster than CH4 having higher size. The residual stream
from a membrane is called the retentate (CH4 rich), while CO2 rich stream is called the permeate. In this study, only
one membrane process configuration is considered since it is proven that this configuration has the most economic
process layout under wide ranges of feed conditions [14,15]. In this study, it is assumed no pressure drop, isothermal
operation, ideal gas behaviour, and steady state operation. The proposed membrane process configuration is shown
in Fig. 1.
Membrane 1 Membrane 2
Retentate

Compressor 1 Compressor 2
Permeate

Compressor 3

Fig. 1. Two-stage membrane process configuration with a recycle stream.

2.2. CO2-ECBM process

Fig.2 shows a schematic diagram of the overall CO2-ECBM production process. During CO2 injection, CH4 in
the coal seam is desorbed by the injected CO2 due to a higher affinity of CO2 with coals compared to CH4. After the
338 Minsoo Kim et al. / Energy Procedia 136 (2017) 336–341
Jiyong Kim et al. / Energy Procedia 00 (2017) 000–000

adsorption/desorption process, the displaced CH4 with CO2 flows to the production well as a free gas phase,
meanwhile the injected CO2 would be sequestrated in the coal. It is assumed that the injected CO2 is obtained from
the recycled CO2 from the separation process and external grid. Table 1 summarizes the initial feed, product gases
conditions and requirements, and technical data of the hollow-fiber membrane [16-18].
Production well Injection well

CH4 product CO2 recycle CO2 supply

CO2,CH4 CO2

Unminable coal seam CO2,CH4

Fig. 2. Schematic diagram of the CO2-ECBM production process

Table 1. Base conditions for optimization of the membrane-based CO2 separation.

Feed gas flow rate 936 Nm3/h
Feed gas pressure 1 bar
Product gas pressure 10 bar
CO2 injection pressure 25 bar
Initial feed gas composition CH4 75 %
Product gas composition CH4 97 %
CO2 permeance 60 GPU
CO2/CH4 selectivity 20
CO2/CH4 displacement ratio 2

3. Optimization model

To achieve the goal of this study, we develop a new optimization model which is extended based upon the
previous works [12]. The optimization model is proposed using a non-linear programming (NLP) technique and
implemented in GAMS environment with Baron Solver. The proposed model is mathematically formulated as
follows:
Maximize {Total expected profit}
x, y
Subject to

Design spec ification constraints 

Mass & concentration balance 
 
h( x, y )  0 Separation process design 
g ( x, y )  0
 
Membrane area&utility consumption 
Cost correlations 
 Non-negative constraints 
 
n m
x  R , y  Y  {0,1}

The objective of this model is to maximize the total expected profit calculated with the total expenses including
capital and operating costs, and the expected revenue cost, subject to the set of equality ( h( x, y)  0 ) and inequality
( g ( x, y)  0 ) constraints. In this study, the operational variables ( x  R n ) indicate decisions including the flow rate
and composition of individual stream (product, permeate, and recycle), whereas the strategic decision variables
( y Y  {0,1}m ) capture the investment decisions such as the required membrane area and compression power.
Table 2 shows the parameters used to estimate costs of the membrane-based CO2 separation [15-17].
 Minsoo Kim et al. / Energy Procedia 136 (2017) 336–341 339
Jiyong Kim et al. / Energy Procedia 00 (2017) 000–000

Table 2. Economic parameters used for optimization of membrane-based CO2 separation.

Membrane module cost 50 $/m2
Membrane replacement cost 10 $/m2
Membrane replacement rate 20 %/year
Compressor cost 500 $/kW
Electricity cost 0.06 $/kWh
Membrane life time 5 year
Compressor life time 25 year
Interest rate 8%
CH4 sale price 5 $/mmbtu

4. Results

We implement the proposed model in the General Algebraic Modeling System (GAMS) environment and use the
non-leaner programming (NLP) of Baron Solver.

4.1. Optimal system configurations

(a) $10/ton of CO2

1,103 Nm3/hr 1,103 Nm3/hr 769 Nm3/hr 602 Nm3/hr
936 Nm3/hr 1 bar 49.4 bar 49.4 bar 10 bar
1 bar CH4 0.74 CH4 0.74 CH4 0.91 289.9 m2 CH4 0.97
325.7 m2
CH4 0.75
334 Nm3/hr Major result:
212.9 kW 25 bar Profit: 57,158 ($/year)
CH4 0.39 Recovery: 83.1%
167 Nm3/hr
1 bar 334 Nm3/hr
CH4 0.67 1 bar 48.4 kW
CH4 0.35
(b) $20/ton of CO2
1,083 Nm3/hr 1083 Nm3/hr 726 Nm3/hr 579 Nm3/hr
936 Nm3/hr 1 bar 35.7 bar 35.7 bar 10 bar
1 bar CH4 0.74 CH4 0.74 531.3 m2 CH4 0.91 CH4 0.97
363.9 m2
CH4 0.75 Major result:
357 Nm3/hr Profit: -189,093 ($/year)
183.2 kW 25 bar Recovery: 79.9%
CH4 0.39
147 Nm3/hr
1 bar 357 Nm3/hr
CH4 0.69 49.9 bar 51.7 kW
CH4 0.39
(c) $30/ton of CO2
1,062 Nm3/hr 1,062 Nm3/hr 661 Nm3/hr 535 Nm3/hr
936 Nm3/hr 1 bar 55.1 bar 55.1 bar 10 bar
1 bar CH4 0.74 CH4 0.74 457.6 m2 CH4 0.91 193.9 m2 CH4 0.97
CH4 0.75 Major result:
401 Nm3/hr Profit: -434,563 ($/year)
213.8 kW 25 bar Recovery: 74.2%
CH4 0.46
126 Nm3/hr
1 bar 401 Nm3/hr
CH4 0.68 3.5 bar 29.9 kW
CH4 0.46

Fig. 3. Optimal process design and operation strategy for different CO2 purchase price.

Based on previous results by Huang et al [18], we knew that CO2 purchase price was illustrated to be the key
point for determining economics of the CO2-ECBM production process. In this section, we focus on comparing
different CO2 purchase price on the optimal design and operation conditions along with the profit and recovery rate.
Fig. 3 shows the optimal design together with the operation conditions for three different CO2 purchase price ($10,
20, and 30/ton of CO2). For assumed CO2 purchase prices, the first stage membrane carries out a significant
separation of CO2 and CH4, whereas considerable amounts of CH 4 loss through the permeate stream. The pressure of
the retentate stream from the second membrane stage is decompressed to 10 bar through the pressure valve.
340 Minsoo Kim et al. / Energy Procedia 136 (2017) 336–341
Jiyong Kim et al. / Energy Procedia 00 (2017) 000–000

In Fig. 3(a) and 3(b), it is shown that all permeate pressure is 1 bar in order to improve the separation efficiency
although the required compressor power of the second compressor is increased. However, when CO2 purchase price
is 30$, the permeate pressure from the first stage membrane is 3.5 bar. Accordingly, the increased pressure on the
permeate side of the first stage membrane reduces the compressor power consumption needed to achieve the target
pressure in the CO2 injection well (25 bar). The profit and CH4 recovery rate decrease with increasing CO 2 purchase
price. This means that despite the loss of CH4, it is more economical to provide CH4 to the CO2 injection well in
order to reduce the amount of CO2 supplied from the external grid.

4.2. Sensitivity of feed condition

1,000 350
(a) (b)
Total membrane

Total compressor
800 280

power (kW)
area (m2)

600 210
400 140
200 70
0 0
10 15 20 25 30 35 40 10 15 20 25 30 35 40
CO2 contents in the feed (%) CO2 contents in the feed (%)
100 400
CH4 recovery rate

(c) (d)
80
Profit ($103/year)
200
60
0
(%)

40 10 15 20 25 30 35 40
-200
20
0 -400
10 15 20 25 30 35 40 -600
CO2 contents in the feed (%) CO2 contents in the feed (%)
$10/ton $20/ton $30/ton

Fig. 4. Effect of feed composition on (a) total membrane area; (b) total compressor power; (c) recovery rate and; (d) profit.

The feed condition, in particular mole ratio between CO 2 and CH4, is one of the critical factor to determine the
system configuration, energy consumption, thereby capital and operating costs. Therefore we extensively investigate
the effect of the feed conditions on technical and economical performances.
Fig. 4 shows the effect of CO2 contents in the feed stream on the total membrane area, total compressor power,
and CH4 recovery rate along with profit in the different scenarios of the CO2 purchase price. As expected, it is
observed that the required membrane area and compressor power increase according with the increase of CO 2
contents in the feed stream to satisfy the desired product purity (97%  ). As a result, the profit by CH4 sales is
expected to be decreased, as shown in Fig. 4(d). It is also observed that an increase in CO2 contents in the feed
stream leads to the lower recovery rate. This is because that the benefits from savings in the membrane and
compressor costs are larger than the revenue obtained from the higher recovery rate.

5. Conclusions

In this study, an optimization model for the design and analysis of multiple membrane separation system were
proposed and applied to the CO2/CH4 separation in the CO2-ECBM production process. The proposed model using
NLP was developed to determine the optimal design and operation conditions of each membrane module, which
maximizes the total expected profit subject to various constraints. Furthermore, we considered three cases, which
differed in terms of CO2 purchase price and investigated the effect of an internal factor associated with the CO2
separation process such as the uncertainty of CO2 contents in the feed. The higher CO2 purchase price caused the
decrease in the recovery rate for increasing the amount injected into the CO2 injection well. It was also shown that
the increased CO2 contents in the feed decreases the recovery rate and profit due to the low separation efficiency
which required the more membrane area and compressor power. While this study proposed the optimal design and
operation of the membrane-based CO2 separation process in the CO2-ECBM, future research may focus on the
 Minsoo Kim et al. / Energy Procedia 136 (2017) 336–341 341
Jiyong Kim et al. / Energy Procedia 00 (2017) 000–000

development of advanced optimization model considering a long time period to plan a long-term investment strategy
for the sustainable development of the CO2-ECBM production process.

Acknowledgements

This work was supported by the Technology Innovation Program (10067793, Engineering education system of
integrated design by case based plant process and safety) funded by the Ministry of Trade, Industry & Energy (MI,
Korea).

References

[1] Sayyafzadeh M, Keshavarz A, Alias ARM, Dong KA, Manser M. (2015) “Investigation of varying-composition gas injection for coalbed
methane recovery enhancement: a simulation-based study.” J Nat Gas Sci Eng 27 (2015):1205–12.
[2] Bing L, Weiji S, Qingxin Q, Hongyan L. (2012) “Technical evaluation system of co-extraction of coal and gas.” Int J Min Sci Technol 22
(2012):891-4.
[3] Wang GX, Wei XR, Wang K, Massarotto P, Rudolph V. (2010) “Sorption-induced swelling/shrinkage and permeability of coal under
stressed adsorption/desorption conditions.” Int J Coal Geol 83 (2010):46-54.
[4] Kim S, Ko D, Row SW, Kim J. (2016) “Techno-economic evaluation of hybrid systems of pressure swing adsorption and membrane
processes for coalbed methane separation.” Chem Eng Res Des 115 (2016):230-40.
[5] Wu Y, Liu J, Chen Z, Elsworth D, Pone D. (2011) “A dual poroelastic model for CO2-enhanced coalbed methane recovery.” Int J Coal
Geol 86 (2011):177-89.
[6] Abdulrahman RK, Sebastine IM. (2013) “Natural gas sweetening process simulation and optimization: A case study of Khurmala field in
Iraqi Kurdistan region.” J Nat Gas Sci Eng 14 (2013):116-20.
[7] Anjum MW, Bueken B, De Vos D, Vankelecom IF. (2016) “MIL-125(Ti) based mixed matrix membranes for CO2 separation from CH4 and
N2.” J Membr Sci 502 (2016):21-8.
[8] Dong L, Chen M, Li J, Shi D, Dong W, Li X, Bai Y. (2016) “Metal-organic framework-graphene oxide composites: A facile method to
highly improve the CO2 separation performance of mixed matrix membranes.” J Membr Sci 520 (2016):801-11.
[9] Datta AK, Sen PK. (2006) “Optimization of membrane unit for removing carbon dioxide from natural gas.” J Membr Sci 283 (2006):291-
300.
[10] Gabrielli P, Gazzani M, Mazzotti M. (2017) “On the optimal design of membrane-based gas separation processes.” J Membr Sci 526
(2017):118-30.
[11] Scholz M, Alders M, Lohaus T, Wessling M. (2015) “Structural optimization of membrane-based biogas upgrading processes.” J Membr
Sci 474 (2015):1-10.
[12] Jung HJ, Han SH, Lee YM, Yeo YK. (2011) “Modeling and simulation of hollow fiber CO2 separation modules.” Korean J Chem Eng 28
(2011):1497-504.
[13] Hao J, Rice PA, Stern SA. (2008) “Upgrading low-quality natural gas with H2S- and CO2-selectve polymer membranes. Part II. Process
design, economics, and sensitivity study of membrane stages with recycle streams.” J Membr Sci 320 (2008):108-22.
[14] Sayyafzadeh M, Keshavarz A. (2016) “Optimisation of gas mixture injection for enhanced coalbed methane recovery using a parallel
genetic algorithm.” J Nat Gas Sci Eng 33 (2016):942-53.
[15] Zhang Y, Sunarso J, Liu S, Wang R. (2013) “Current status and development of membranes for CO2/CH4 separation: A review.” Int J
Greenhouse Gas Control 12 (2013):84-107.
[16] Roussanaly S, Anantharaman R, Lindqvist K, Zhai H, Rubin E. (2016) “Membrane properties required for post-combustion CO2 capture at
coal-fired power plants.” J Membr Sci 511 (2016):250-64.
[17] Shao P, Dal-Cin MM, Guiver MD, Kumar A. 2013 “Simulation of membrane-based CO2 capture in a coal-fired power plant.” J Membr Sci
427 (2013):451-9.
[18] Huang Y, Zheng Q, Fan N, Aminian K. (2014) “Optimal scheduling for enhanced coal bed methane production through CO 2 injection.”
Appl Energy 113 (2014):1475-83.